Highly selective nucleic acid amplification primers

Abstract

The present invention relates to highly selective nucleic acid amplification primers, and in particular, the present invention discloses multi-part primers for use in primer-dependent nucleic acid amplification methods. The present invention also discloses primer-dependent nucleic acid amplification reactions, in particular DNA amplification reactions, reaction mixtures and kits for such reactions. The present invention relates to primer-dependent nucleic acid amplification reactions, in particular DNA amplification reactions such as RCR, as well as primers, reaction mixtures and kits for such reactions, and assays using the same.

Description

[0001] This application is a divisional application of the Chinese patent application with application number 201480019974.2, application date February 7, 2014, and invention name “Highly Selective Nucleic Acid Amplification Primer”.

[0002] CROSS-REFERENCE TO RELATED APPLICATIONS

[0003] This application claims priority to U.S. Provisional Application No. 61 / 762,117, filed on February 7, 2013. The contents of that application are incorporated herein by reference in their entirety. Technical Field

[0004] The present invention relates to primer-dependent nucleic acid amplification reactions, in particular DNA amplification reactions (such as PCR), as well as primers, reaction mixtures and kits for use in such reactions and assays using the same. Background Art

[0005] Primer-dependent nucleic acid amplification reactions require "specificity" and the reaction can include the detection of an amplification product ("amplicon"), i.e., the annealing of the primer to the expected position in the nucleic acid chain and the extension of the primer that binds only to the expected target sequence. Conventionally, specificity is achieved by making the primer sufficiently long so that under the conditions of the amplification reaction, primarily during the primer annealing step, the primer reaches only one position in the nucleic acid chain.

[0006] Certain amplification reactions are intended to distinguish between allelic variants, such as single nucleotide polymorphisms (SNPs). One way to do this is to amplify all variants and distinguish between them by allele-specific hybridization probes, such as molecular beacon probes. For such methods, amplification primers are equally complementary to all variants to amplify regions of sequences that vary between alleles, and the probes identify the alleles present in the amplified products. See, for example, Tyagi et al. (1998) Nature Biotechnology 16: 49-53. If the sequence under investigation is an allele, such as a SNP present in a mixture with another allele, such as a wild-type (WT) variant, the use of probes for distinction has a practical detection limit of about 3% (no less than about 30,000 target allele molecules in the presence of 1,000,000 molecules of alternative alleles) due to the tendency of the amplification of the dominant allele to overwhelm the amplification of the rare allele.

[0007] Another approach of distinguishing between allelotrope is to use the primer of selectivity for the sequence being studied.For such method, the sequence complementation of variation between primer and allelotrope, and can detect amplified product by the primer of labelling, DNA binding dye or the probe of labelling (in this case, the common sequence of all allelotrope amplicons of probe detection).Typically, the primer of high specificity has the length of 15-30 Nucleotide.Such conventional primer for an allelotrope has very limited selectivity with respect to another allelotrope.Known that primer shortening can improve its selectivity, but because this improvement is from specific consumption, and because short primer is unlikely to form stable hybridization with its target sequence at typical annealing temperature, primer shortening is limited for the mixture value of analyzing allelotrope.

[0008] Other modifications of primers have been developed to improve their selectivity while retaining their specificity. One such method is ARMS ("amplification retardation mutation system"). ARMS primers have 3'-terminal nucleotides that are complementary to the sequence variant being studied but mismatched with another or other alleles. See Newton et al. (1989) Nucleic Acids Res. 17: 2503-2516; and Ferrie et al. (1992) Am. J. Hum. Genet. 51: 251-262. ARMS relies on the retardation properties of certain DNA polymerases, which tend not to extend primer-target hybrids with such mismatches. ARMS has been shown to be effective for determining heterozygosity (homozygous WT, heterozygous or homozygous mutant (MUT)), but it has a practical detection limit of about 1% for other uses (not less than about 10,000 target allele molecules in the presence of 1,000,000 molecules of alternative alleles).

[0009] Another approach is to make the primer a hairpin to improve its selectivity. See, Tyagi et al. European Patent EP1 185 546 (2008), which discloses the preparation of a hairpin loop complementary to the sequence under investigation but mismatched with another or other alleles; and Hazbón and Alland (2004) J. Clin. Microbiol. 42: 1236-1242, which discloses the preparation of the terminal nucleotide of the 3' arm of the hairpin primer complementary to the sequence variant under investigation but mismatched with another or other alleles, as with ARMS. These modifications also have a practical detection limit of about 1% (no less than about 10,000 target allele molecules in the presence of 1,000,000 molecules of alternative alleles).

[0010] JongYoon Chun and colleagues at Seegene Biosciences in Seoul, South Korea, have designed a class of primers they call "dual-priming oligonucleotides (DPOs)". See Chun et al. (2007) Nucleic Acids Res. 35(6) e40; Kim et al. (2008) J. Virol. Meth. 149:76-84; Horii et al. (2009) Lett. Appl. Microbiol. 49:46-52; WO 2006 / 095981 A1 and WO 2007 / 097582 A1. DPO primers consist of three segments: a long 5' thermophilic segment (e.g., 20-25 nucleotides long), a central separating segment of five deoxyriboinosines, and a 3' priming segment (typically 8-12 nucleotides long) that is complementary to the intended target sequence but mismatches with other target sequences. The target sequence is complementary to all three segments, but due to its shorter length, the Tm of the 3' segment is lower than that of the 5' segment, and the isolated segment has the lowest Tm due to the five deoxyriboinosines. The DPO primers are designed so that amplification occurs only when both the 5' segment and the 3' segment hybridize to the target chain. According to Chun et al. (2007), the isolated segment was selected to be five deoxyriboinosines because 3-4 and 6-8 deoxyriboinosines did not give good results; the 3' segment was positioned so as to provide a GC content of 40-80%, and the length of the 5' segment provided was sufficient to increase its Tm above the annealing temperature to be used in 3'-RACE amplification (page 2 of Nucleic Acids Res. 35 (6) e40). Chun et al. reported that genotyping (homozygous wild type, heterozygous or homozygous mutant) of a SNP (G→A mutation) in CYP2C19 was successfully performed using two pairs of DPO primers. Of the four DPO primers, one has a 12-nucleotide 3' segment that is perfectly complementary to both alleles; one has a 9-nucleotide 3' segment that is perfectly complementary to both alleles; and two have an 8-nucleotide 3' segment with a variable nucleotide located in the middle, i.e., at the fourth nucleotide position from the 3' end. Genotyping is accomplished by gel electrophoresis.

[0011] There are situations where it is desirable to detect very rare first alleles in the presence of very abundant second alleles. This has been referred to as "sensitivity". In other words, the primer must not only be "specific" (go to the correct position in the genome), and must be "selective" (reject wild-type or other abundant sequences similar to the target sequence), and it must be highly selective, i.e. "sensitive" enough to detect very few mutants or other rare first sequences in the presence of abundant wild-type or other abundant second sequences. See paragraph

[0004] of International Patent Application WO 2012 / 112 582 A2 of Makarov and Chupreta.

[0012] To increase sensitivity while maintaining specificity and selectivity, Vladimir Makarov and colleagues at Swift Biosciences (Ann Arbor, Michigan, USA) disclosed a “discontinuous polynucleotide [“primer”] design” (paragraph

[0051] of WO 2012 / 112 582 A2), which has been commercialized as myT TMPrimer. Such a primer can be regarded as a long conventional primer, which consists of two oligonucleotides to establish a 3'priming sequence of eight nucleotides; and a tail complementary to the 5' end of the sequence and to the 3' end of another oligonucleotide is added to form a high temperature stem. Through the stem, the two oligonucleotides are non-covalently bound and form a stable three-way junction when bound to the target sequence. The oligonucleotide with a 3' end of eight nucleotides is called a "primer", and the other oligonucleotide is called a "fixer". The function of the fixer is to provide specificity, that is, to bind the primer to the expected position in the genome. Therefore, it is long, typically about 30 nucleotides long. The function of the tail is to hybridize the two oligonucleotides under amplification conditions, so the tail is also quite long, forming a stem of 20-25 nucleotides long. The function of the 3' region of eight nucleotides is to selectively initiate. This discontinuous hybridization "effectively stabilizes the binding between the [priming] regions of the primer oligonucleotides, even if the region is as small as eight bases, thereby improving the efficiency of PCR." (WO 2012 / 112582 A2). Further improvements are disclosed in Examples 9-11 of WO 2012 / 112582 A2. As in ARMS, a nucleotide that mismatches with the wild-type target forms the 3' terminal nucleotide; a third oligonucleotide, a blocking oligonucleotide ("blocker"), is included in the amplification reaction (its 5' terminal nucleotide overlaps with the 3' terminal nucleotide of the primer and is complementary to the wild-type target); and the 3' terminal nucleotide of the primer is formed by a locked nucleic acid ("LNA"). For the detection of multiple nucleotide polymorphisms in the K-ras and B-raf genes, the invention discloses a detection sensitivity of one mutant in 14,000 wild types (approximately 0.01%).

[0013] There remains a need for a single oligonucleotide primer having the ability to detect and preferably quantify a rare first target sequence (e.g., a mutant target sequence) in the presence of a very large number of a second target sequence that differs from the first target sequence by as little as a single nucleotide, such as a wild-type sequence. Summary of the invention

[0014] The present invention includes multi-part primers for use in primer-dependent nucleic acid amplification methods, particularly including polymerase chain reaction (PCR) methods, which are capable of distinguishing between rare expected targets (e.g., mutant DNA targets) and closely related sequences (e.g., wild-type DNA targets) that differ by a single nucleotide substitution (sometimes referred to as a multiple nucleotide polymorphism, or SNP for short).

[0015] The present invention includes primer-dependent nucleic acid amplification methods, such as PCR methods, which utilize multi-part primers according to the present invention and which are capable of selectively amplifying one or more rare target sequences in a population of abundant closely related sequences. Such expected target sequences may be rare mutant sequences, such as sequences found in malignant cells, other abundant wild-type populations found in normal cells. For methods such as PCR methods utilizing DNA-dependent DNA polymerases, the expected targets and related sequences are DNA sequences present in the sample, or they are cDNA sequences formed from RNA sequences (including mRNA sequences) present in the sample by reverse transcription reactions. Reverse transcription may be performed in the same reaction mixture as the subsequent amplification, or it may be performed separately prior to amplification. Multi-part primers may be used as primers in reverse transcription reactions. The present invention also includes amplification and detection methods, which include detection of amplification products or "amplicons". The following description, including examples, describes multi-part primers in conjunction with PCR amplification reactions initiated by DNA targets. Those skilled in the art will understand how to apply these teachings to multi-part primers in conjunction with other primer-dependent nucleic acid amplification methods.

[0016] The present invention also includes kits containing materials for performing such amplification methods (including such amplification and detection methods).

[0017] In addition, the present invention specifically addresses the primary goal of molecular diagnostics, which is to find sensitive and specific means of detecting extremely rare cancer cells in clinical samples containing very abundant normal cells (by identifying somatic mutations), and being able to quantitatively determine their abundance. Being able to do so has multiple advantages, including:

[0018] 1. The ability to detect the presence and abundance of cancer cells after treatment (such as after a bone marrow transplant in a leukemia patient). Utilizing the present invention will enable clinicians to determine whether the use of (rather toxic) drugs can be stopped. The present invention will enable clinical studies to be conducted to determine the level of minimal remaining disease that can be managed by the body without drug treatment. In addition, patients can be monitored over time after treatment to detect higher levels of manifestations that can then be treated by appropriate means.

[0019] 2. The ability to rapidly detect and quantify rare cancer cells (at levels too low to be seen by a pathologist under a microscope) in biopsy specimens obtained during surgery. Utilizing the present invention will enable surgeons to make rational decisions about the extent of surgery, saving the removal of unaffected tissue.

[0020] 3. The ability to detect key mutations in DNA molecules released into the plasma by the natural destruction of rare circulating tumor cells in the blood. The present invention will enable early detection of tumors, where these cells have acquired the ability to metastasize, providing clinicians with the opportunity to intervene early.

[0021] 4. The ability to monitor patients whose genetic inheritance indicates the ability to develop life-threatening tumors during their lifetime (as in many breast cancers). The present invention will enable regular monitoring to determine if key somatic mutations have occurred, allowing therapeutic intervention to be provided at a very early stage of the disease.

[0022] Other applications of the invention will occur to those skilled in the art.

[0023] "Rare" and "abundant" mean that the ratio of the expected target sequence to closely related sequences is at least 1 / 10. 3 to 1 / 10 7 "Closely related" refers to sequences that differ from the expected target sequence by one, two, or at most a few nucleotides. A mutant target sequence that differs from the wild-type sequence by a single nucleotide at a particular position is often referred to as being or having a multiple nucleotide polymorphism (SNP).

[0024] Methods according to the invention include primer-dependent nucleic acid amplification for at least one expected target sequence (e.g., a mutant DNA target sequence) that may be present infrequently in a sample or reaction mixture containing abundant closely related, unexpected target sequences (e.g., a wild-type DNA target sequence). These methods utilize a reaction mixture containing a multi-part primer according to the invention for each rare target. The three parts of the primer cooperate with each other to produce extremely selective amplification. Figure 1 is a representative schematic diagram of primers according to the present invention. Figure 1Two schematic diagrams are included: the upper schematic diagram shows a multi-part primer 103 relative to its intended target 101 (which may be rare) under hybridization conditions (such as those present during the annealing step of a PCR cycle); and the lower schematic diagram shows the same primer relative to a closely related sequence (referred to herein as an unexpected or mismatched target 102). The expected target 101 and the unexpected target 102 have the same nucleotide sequence, except that the expected target 101 has one or more nucleotides "x" (preferably a single nucleotide) that are different from the corresponding nucleotide in the mismatched target 102, referred to herein as "y". For example, the unexpected target sequence 102 can be a wild-type human DNA sequence, and the expected target sequence 101 can be a mutant cancer cell sequence containing a SNP. The upper schematic diagram depicts the primer 103 hybridizing to the expected target strand 101. In the 5' to 3' direction, the primer includes an anchor sequence 104, a bridge sequence 105, and a foot sequence 106. Primer 103 may optionally include a 5' tail 107 to impart additional functionality. It also optionally includes a blocking group 108. During primer annealing at the beginning of amplification, anchor sequence 104 hybridizes to the intended target 101, as conventionally represented by the short vertical line between the anchor sequence and its binding site (indicating the pairing of complementary nucleotides). Bridge sequence 105 is mismatched (not complementary) to target 101 at sequence 109 (which the inventors call the "insertion sequence") and causes a "bubble" in the duplex structure. Foot sequence 106 hybridizes to the intended target 101 and initiates replication by a DNA polymerase. The schematic diagram below depicts the same primer 103 hybridized to an unintended, mismatched target 102. As described, the mismatched target 102 differs from the intended target 101 by at least one nucleotide change (x to y) in the sequence relative to primer foot 106. As shown, during primer annealing at the start of amplification, anchor sequence 104 hybridizes to an unintended target 102 at the anchor sequence binding site. Again, bridge sequence 105 is mismatched with insert sequence 109. However, foot sequence 106 does not hybridize to target 102, and target 102 is not primed for replication.

[0025] In accordance with Figure 1Under the ideal amplification reaction conditions, the expected target 101 (even if rare) will always be replicated, and the unexpected target 102 (even if abundant) will never be replicated. However, priming is a statistical problem. For example, primers go on target and go off target, fully match and mismatch with a certain frequency. As a result, the perfect target is not always replicated, and the mismatched target is sometimes replicated. Therefore, selective amplification and detection of rare targets depend on both the frequency of the perfect target being replicated and the frequency of the mismatched target being replicated. The multi-part primers used in the present invention have three continuous sequences (anchor sequence, bridge sequence and foot sequence), which cooperate with each other to achieve very high selectivity in the actual amplification reaction (including amplification and detection assay). The anchor sequence is used to hybridize the primer with the target sequence in an effective manner that is no different from the hybridization of conventional primers, and it is the same (or almost the same) in the expected target and the unexpected mismatched target. The bridge sequence and foot sequence described more comprehensively below cooperate to give primer specificity, that is, the specificity of the target exceeding the mismatched target for the expected target. The inventors have discovered that a high degree of selectivity can be achieved if the bridge sequence and the foot sequence cooperate to make the intended target less likely than likely to replicate. In addition, the inventors have enabled the bridge sequence to replicate rapidly (rabidly) and efficiently. The bridge sequence is preferably a DNA sequence. The result achieved is the amplification of the intended target sequence, which is delayed at the beginning, but proceeds normally once it starts; but the amplification of the unexpected, mismatched target sequence is more significantly delayed but proceeds normally once it starts. The increased delay of the mismatched target relative to the matched target is the increase in selectivity achieved by the primers. The reason for achieving such increased selectivity is that the probability of an unexpected target sequence being replicated by a DNA polymerase is at most 1 / 1,000, preferably at most 1 / 10,000 and more preferably at most 1 / 100,000 of the probability of the intended target sequence to be replicated.

[0026] By reference Figure 1, the primer includes an anchor sequence 104, which hybridizes the primer to the binding site in the expected target and closely related target sequence during the primer annealing step (which includes the primer annealing temperature) of the amplification reaction. In this regard, the anchor sequence is the same as a conventional primer and functions like a conventional primer. It can be completely complementary to the target and to closely related sequences, or it can contain one or more mismatched nucleotides. In the amplification reaction in which it is used, it usually has a melting temperature Tm at least equal to or higher than the annealing temperature to enhance hybridization. In most embodiments, the Tm of the anchor sequence is between 3°C and 10°C higher than the primer annealing temperature. To the extent that it is not blocked by the blocking group, all or part of the anchor sequence of the multi-part primer used in the present invention is copied by the DNA polymerase. Because the exponential amplification is carried out rapidly with high, normal PCR efficiency, the type and quantity of non-natural nucleotides, nucleotide mimetics and non-natural internucleotide bonds contained in the copied part are limited to allow rapid and efficient replication by the DNA polymerase. The inventors prefer that the anchor sequence is a DNA sequence.

[0027] Typically, the anchor sequence 104 forms a probe-target hybrid that is 15-40 nucleotides long, preferably 15-30 nucleotides long, and more preferably 20-30 nucleotides long. As described above, shorter anchor sequences must still hybridize to their target sequences during primer annealing, which generally means that their Tm must be at least 50°C (e.g., 66-72°C). It can be completely complementary to the target, or it can contain one or more mismatches; for example, when one is studying a target whose sequence is variable to the anchor sequence, one can choose such an anchor sequence 104 that is a consensus sequence that is not completely complementary to any version of the target but hybridizes to all variants during primer annealing. The inventors tend to form DNA anchor sequences that form anchor sequence / target hybrids that are generally in the range of 15-30 base pairs, as is typical for conventional PCR primers. The inventors demonstrate in the following examples that an anchor sequence is 24 nucleotides long, is DNA, and is completely complementary to the target sequence. The multipart primer does not prime other sequences in the reaction mixture in addition to its target sequence (i.e., the expected target sequence and the unintended, mismatched target sequence). Although conventional primers must be designed to achieve this function, the requirement for the anchor sequence is not critical because the anchor sequence helps distinguish other sequences that are or may be present in the sample.

[0028] By referring to Figure 1, primer includes foot sequence 106, which is complementary to the target sequence expected in the region, and the region includes nucleotides (SNP nucleotides), or includes two nucleotides in some cases, and the nucleotides are different from the target sequence that is unexpected and mismatched. The foot sequence can be completely complementary to the expected target sequence, or it can contain one or even two nucleotides in some cases, and the nucleotides are mismatched with the expected target sequence and the unexpected target sequence. Compared with the target sequence of mismatch, the foot sequence 106 is always more complementary to the expected target sequence by at least one nucleotide. During amplification, the foot sequence is copied. Because exponential amplification is carried out quickly with high and normal PCR efficiency, the type and quantity of rapid and effective replication by DNA polymerase containing non-natural nucleotides, nucleotide analogs and non-natural nucleotide bonds are limited. The inventor tends to the foot sequence to be a DNA sequence. Because the subsequent exponential amplification of the expected amplicon is carried out with high and normal PCR efficiency, the type and quantity of effective replication by DNA polymerase containing non-natural nucleotides, nucleotide analogs and non-natural nucleotide bonds in the foot sequence are limited. In a preferred embodiment, the foot sequence is a DNA sequence that is completely complementary to the intended target sequence and contains a single nucleotide mismatch with a nucleotide in an unintended target sequence.

[0029] The foot sequence 106 forms a hybrid with the intended target sequence, the hybrid being at least 5 nucleotides long, for example, in the range of 5-8 base pairs, preferably in the range of 6-8 base pairs and more preferably no longer than 7 nucleotides in length, for example, in the range of 6-7 base pairs. When the anchor sequence hybridizes with the intended target sequence, there is only one binding site for the foot sequence. A consideration in primer design is that because the foot sequence is shortened, the chance that there may be another possible binding site is increased, especially if the foot sequence is shortened to only 5 nucleotides. As demonstrated by the inventors in the Examples, while a mismatched nucleotide vs. an unintended target can occur at any nucleotide position in the foot sequence 106, the inventors prefer that the mismatched nucleotide is either the 3' terminal nucleotide, as in the ARMS primer (Newton et al. (1989) Nucleic Acids Res. 17:2503-2516; and Ferrie et al. (1992) Am. J. Hum. Genet. 51:251-262), or that there is one nucleotide remaining in the 3' end from the foot sequence, which the inventors sometimes refer to as the "3' penultimate nucleotide".

[0030] By seeing again Figure 1The primer includes a bridge sequence 105 which is selected so that it cannot hybridize with the insert sequence 109 during annealing of the multi-part primer to the target molecule. The bridge sequence or, if it contains a blocking group, its 3' portion is replicated by the DNA polymerase. Because it is desired to rapidly perform exponential amplification of the amplicon with high, normal PCR efficiency, the non-natural nucleotides, nucleotide mimetics, and non-natural internucleotide bonds contained in the replicated portion of the bridge sequence are limited to the types and quantities that allow rapid and efficient replication by the DNA polymerase. Preferably, the bridge sequence is a DNA sequence.

[0031] In the target, the bridge sequence 105 and its opposite insertion sequence 109 form a bubble in the primer / expected target hybrid. The perimeter of the bubble is the length of the bridge sequence 105 plus the length of the insertion sequence 109, plus 4 (a pair of nucleotides from the anchor sequence hybrid and a pair of nucleotides from the foot sequence hybrid). The bridge sequence and the insertion sequence do not have to be equal in length: either sequence can be shorter than the other. In certain embodiments, the length of the insertion sequence can be zero. In a preferred embodiment, it is at least six nucleotides long. In a more preferred embodiment, wherein the sum of the lengths of the bridge sequence and the insertion sequence is at least 24 nucleotides, the inventors tend to have an insertion sequence with a length of at least eight nucleotides, more preferably at least ten nucleotides. The bridge sequence should be at least six nucleotides long. Certain preferred embodiments have a bridge sequence and an insertion sequence of equal length. The perimeter of the bubble can be as short as 16 nucleotides and as long as 52 nucleotides, for example 16-52 nucleotides, 20-52 nucleotides or 28-44 nucleotides.

[0032] As a general consideration in designing multipart primers, increasing the bubble perimeter and shortening the foot sequence increases the delay in amplification of the intended target. The number of PCR cycles required to synthesize a predetermined detectable number of amplicons in a reaction initiated by a specific number of intended target sequences (the threshold cycle, the C of the reaction) is the number of cycles required to synthesize a predetermined detectable number of amplicons in a reaction initiated by a specific number of intended target sequences. T ) can be observed by, for example, observing the inserted dye Green fluorescence intensity (whose intensity reflects the number of amplicons present during each PCR cycle). This provides a method for measuring the difference in the probability that a DNA polymerase will extend a multipart primer / unintended target hybrid relative to the probability that a DNA polymerase will extend a multipart primer / intended target hybrid. Given that amplification proceeds exponentially, 10 cycles of C T The difference means that the probability of extension of the multipart primer / unintended target hybrid is 1 / 1,000 of the probability of extension of the multipart primer / intended target hybrid; C of 13.3 cycles T The difference represents a probability of 1 in 10,000; approximately 16.6 cycles of C T The difference represents a probability of 1 / 100,000; and 20 cycles of CT The difference represents a one in a million chance.

[0033] In an assay according to the invention using a multipart primer, such as by adding to the reaction mixture Green dye obtained from each PCR cycle by measuring the fluorescence intensity ΔC T This is reflected in the higher threshold cycles observed for mismatched target sequences and the higher threshold cycles observed for the same number of (e.g., 10 6 The difference between the lower threshold cycle observed for the expected target sequence of the present invention (100 replicates) should be at least 10 cycles, preferably at least 12 cycles, more preferably at least 14 cycles, even more preferably at least 17 cycles, even more preferably at least 18 cycles and most preferably 20 or more cycles. In an amplification reaction in which the multipartite primers according to the present invention replace the well-designed conventional PCR primers, there is a delay (ΔC ) in the threshold cycle obtained using the expected target sequence. T The amount of the delay depends on how well designed the conventional primers being compared are, but typically the delay is at least two amplification cycles, usually at least three cycles, and sometimes at least eight cycles or even ten cycles compared to a conventional primer consisting only of the anchor sequence of the multipart primer.

[0034] Preferred embodiments of the methods according to the invention include detecting the products resulting from amplification of the rare target sequence. Detection of the amplified products can be performed separately after amplification, for example by gel electrophoresis. In preferred embodiments, detection reagents are included in the amplification reaction mixture, in which case detection can be "real time", i.e., performed at multiple times during the amplification process, or "end point", i.e., performed after the amplification reaction is complete, preferably by homogeneous detection without opening the reaction vessel. Detection reagents include DNA binding dyes, such as Green, double-labeled fluorescent probes that generate signals of amplified products, such as molecular beacon probes, and combinations of fluorescent probes that bind dyes and are stimulated by excitation from dyes. In addition, as described herein, the primers themselves can include fluorescent labels that emit fluorescence only when the primers are incorporated into the amplicon (or alternatively, when the primers are bound to complementary amplicon).

[0035] The present invention includes a reaction mixture for amplifying at least one target sequence. The reaction mixture includes a primer pair for each expected target sequence, and one primer in each primer pair is a multi-part primer as described herein. The reaction mixture also includes a reagent for amplifying the target, including deoxyribonucleoside triphosphates, an amplification buffer, and a DNA polymerase. The preferred reaction mixture for the assay method according to the present invention also includes a detection reagent, i.e., a DNA binding dye, a hybridization probe (or both), or a 5' functional tail of each multi-part primer. If the starting sample contains RNA, the amplification reaction mixture can also include a reverse transcriptase and a primer for a reverse transcription reaction.

[0036] The present invention also includes products for the kit for carrying out the above-mentioned amplification reaction and amplification and detection reaction for one or more expected target sequences. The kit includes oligonucleotides and reagents required for setting up a reaction mixture according to the present invention. The kit for starting samples being RNA may include reagents for reverse transcription.

[0037] The details of one or more embodiments of the invention are set forth in the description below. Other features, objects, and advantages of the invention will be apparent from the description and claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 Schematic representation of a multipart primer useful in the present invention that hybridizes to its intended target sequence and to a different mismatch sequence that is substituted with one or more nucleotides from the intended target sequence.

[0039] Figure 2 is a schematic diagram of amplification cycles, wherein a multipartite primer of the invention is first replicated, and the resulting amplicons are subsequently replicated in the next two cycles.

[0040] Figure 3 is a schematic diagram of a multipart primer according to the present invention showing positions for placement of a blocking group which terminates replication by a DNA polymerase.

[0041] Figure 4 is a schematic diagram of several exemplary optional 5' functional moieties.

[0042] Figure 5 Shown are real-time fluorescence results obtained from a conventional linear primer and either 1,000,000 expected target sequences or 1,000,000 unexpected, mismatched target sequences that differ from each other at a single nucleotide located in the middle of the sequence to which the primer binds.

[0043] Figure 6Shown are real-time fluorescence results obtained from ARMS primers and either 1,000,000 expected target sequences or 1,000,000 unexpected, mismatched target sequences that differ by a single nucleotide, where the "interrogating nucleotide" in the primer (which is complementary to the corresponding nucleotide in the expected target sequence but not complementary to the corresponding nucleotide in the unexpected target sequence) is the 3' terminal nucleotide of the primer; and the figure also shows the results obtained from similar primers, where the interrogating nucleotide is located at the second to last nucleotide at the 3' end of the primer.

[0044] Figure 7 Real-time fluorescence results obtained with a multi-part primer according to the invention in a reaction containing either 1,000,000 molecules of the expected target sequence of the primer or 1,000,000 molecules of the unintended target sequence of the primer (wherein the multi-part primer has a query nucleotide at the penultimate position of the foot sequence) are shown.

[0045] Figure 8 Shown are real-time fluorescence results obtained with a multi-part primer according to the invention in a series of reactions each containing 1,000,000 unintended target sequences and either: 0, 10, 100, 1,000, 10,000, 100,000 or 1,000,000 intended target sequences.

[0046] Fig. 9 Yes Display Figure 8 A plot of the inverse linear relationship between the threshold cycle observed for each reaction shown in versus the logarithm of the number of expected target present in each reaction, and the dashed line in the plot represents the threshold cycle obtained for reactions containing 1,000,000 unexpected target sequences and no expected target sequence.

[0047] Fig.10 Is used to display Figure 8 and Fig. 9 Graph of the results obtained for the same dilution series used in the experiments shown in , using three otherwise identical multipart primers with foot sequences of either 6, 7 or 8 nucleotides in length (with the query nucleotide at the penultimate position in each foot sequence).

[0048] Fig.11 Is used to display Figure 8 , Fig. 9 and Fig.10 Graph of the results obtained for the same dilution series used in the experiment shown in , using three multipartite primers with insert sequences of the same length in the target molecule, whose bridge sequences form bubbles of different circumferences.

[0049] Fig.12A graph showing a series of real-time fluorescence results obtained from: the same multi-part primer according to other aspects of the present invention and either 1,000,000 expected target sequences or 1,000,000 unexpected target sequences (which differ from the expected target sequence by a single nucleotide polymorphism), wherein the query oligonucleotide in the foot sequence of the primer (which is complementary to the corresponding nucleotide in the expected target sequence but not complementary to the corresponding nucleotide in the unexpected target sequence) is located at different positions relative to the 3' end of the primer.

[0050] Fig.13 A graph showing a series of real-time fluorescence results obtained with a multi-part primer according to the invention and either 1,000,000 expected target sequences or 1,000,000 unexpected target sequences that differ by a single nucleotide, wherein the length of the bridge sequence plus the length of the insertion sequence in the target molecule is kept constant (i.e. the perimeter of the bubble is the same), but wherein the symmetry of the bubble formed by the bridge sequence and the insertion sequence in the target molecule (the relative lengths of those sequences) is varied.

[0051] Fig.14 is a graph showing the inverse linear relationship between the observed threshold cycle and the logarithm of the number of V600E mutant human B-raf target sequences in a series of reactions, wherein each reaction contained 1,000,000 wild-type human B-raf target sequences and either: 10, 100, 1,000, 10,000, 100,000 or 1,000,000 V600E mutant human B-raf target sequences. The dashed line represents the threshold cycle obtained from reactions containing DNA from 1,000,000 wild-type human B-raf target sequences and no DNA from V600E mutant human B-raf target sequences.

[0052] Fig.15 is a graph showing the inverse linear relationship between the observed threshold cycle and the logarithm of the number of mutant target sequences present in a series of reactions, wherein each reaction contains 10,000 wild-type target sequences present in genomic DNA isolated from cultured normal human cells and either: 10, 30, 100, 300, 1,000, 3,000 or 10,000 mutant target sequences present in genomic DNA isolated from cultured human cancer cells having the T790M mutation in the EGFR gene. The dotted line represents the threshold cycle obtained from reactions containing 10,000 wild-type target sequences and no DNA from cancer cells.

[0053] Fig.16 Display and Fig. 9Similar experimental results are shown in , except that an Applied Biosystems PRISM 7700 Fluorescence Spectrophotometer Thermal Cycler was used instead of the Bio-Rad IQ5 Fluorescence Spectrophotometer Thermal Cycler to perform the experiments.

[0054] Fig.17 Shown are the real-time fluorescence results obtained by the multi-part primer according to the present invention in the following reactions: the reactions contain either 1,000,000 molecules of the expected target sequence of the primer or 1,000,000 molecules of the unexpected target sequence of the primer (wherein the multi-part primer has a query nucleotide at the penultimate position of the foot sequence) (Figure A), and the real-time fluorescence results of a truncated version of the primer lacking the 3' penultimate and 3' terminal nucleotides (Figure B).

[0055] Fig.18 is a schematic diagram of two multipart primers according to the present invention which can be used in a multiplex reaction directed to two closely related expected target sequences.

[0056] Fig.19 is a schematic diagram of two multipart primers and two molecular beacon probes that can be used in a multiplex reaction against two closely related expected target sequences. DETAILED DESCRIPTION

[0057] The present invention is based, at least in part, on the unique design of multi-part primers for use in primer-dependent amplification reactions. Thus, the present invention discloses the design and characteristics of multi-part primers that exhibit extraordinary selectivity when hybridized to a template present in an original sample. Due to this extraordinary selectivity, the inventors refer to the multi-part primers of the present invention as "super selective" primers.

[0058] Significantly, once synthesis is initiated on the mutant template, the resulting amplicons amplify efficiently and exponentially, and the real-time data provides a routine means of assessing the abundance of the mutant template present in the original sample. The experiments described below demonstrate that the super-selective primers have sufficient discriminatory power to inhibit the synthesis of the wild-type sequence to such an extent that as few as 10 molecules of the mutant sequence can be reliably detected in a sample containing 1,000,000 molecules of the wild-type sequence, even when the only difference between the mutant and the wild-type is a multinucleotide polymorphism.

[0059] 1. Primer-dependent amplification

[0060] The primer-dependent amplification reaction useful in the methods of the invention may be any suitable exponential amplification method, including polymerase chain reaction (PCR), symmetric or asymmetric, ligase chain reaction (LCR), nick translation enzyme amplification reaction (NEAR), strand displacement amplification (SDA), nucleic acid sequence-based amplification (NASBA), transcription-mediated amplification (TMA) and rolling circle cycle amplification (RCA). Preferably, the method utilizes PCR. In an asymmetric PCR amplification method (e.g., asymmetric PCR), one primer (the limiting primer) is present in a limiting amount so that it is exhausted before amplification is completed, after which linear amplification occurs using the remaining primers (excess primers). An asymmetric PCR method useful in the present invention is LATE-PCR (see, e.g., European Patent EP 1,468,114; and Pierce et al. (2005) Proc. Natl. Acad. Sci. USA 102:8609-8614). If an asymmetric amplification method is used, the multipart primer is preferably an excess primer. Preferred methods also include digital PCR (see, e.g., Vogelstein and Kinzler (1999) Proc. Natl. Acad. Sci. USA 98:9236-9241), in which it is desired to detect a large number of amplicons from a single mutant template molecule present in a reaction containing a large number of wild-type molecules.

[0061] If the amplification reaction utilizes an RNA-dependent DNA polymerase (an example is NASBA), then the amplification reaction is isothermal. The inventors refer to repeated rounds of synthesis of amplification products as "cycles", but they are not thermal cycles. For such amplifications, the "expected target sequences" and "unexpected target sequences" primed by the multipart primers according to the invention are RNA sequences present in the original sample and in the amplification reaction mixture, where they are present together with the DNA polymerase and the multipart primers.

[0062] If the amplification reaction utilizes a DNA-dependent DNA polymerase (an example is PCR), then the original sample may contain either a DNA or RNA target. For such amplification, the "expected target sequence" and "unexpected target sequence" primed by the multi-part primers according to the present invention are either DNA sequences present in the original sample or DNA sequences prepared by reverse transcription of RNA sequences present in the original sample. If the multi-part primers are used for reverse transcription, then the "expected target sequence" and "unexpected target sequence" are RNA and cDNA. If separate, external primers are used for reverse transcription, then the "expected target sequence" and "unexpected target sequence" are cDNA. In either case, the "expected target sequence" and "unexpected target sequence" are nucleic acid sequences present in the amplification reaction mixture with a DNA polymerase and a multi-part primer. The primer-dependent amplification reaction comprises repeated thermal cycles of primer annealing, primer extension, and chain denaturation (melting). Primer annealing can be performed at a temperature lower than the primer extension temperature (e.g., three-temperature PCR), or primer annealing and primer extension can be performed at the same temperature (e.g., two-temperature PCR). The overall thermal profile of the reaction may include repetitions of a particular cycle, or the temperature / times may be varied during one or more cycles. For example, once amplification has begun and the priming sequence of a multi-part primer has been extended, a higher annealing temperature appropriate for a longer primer may be used to complete the amplification reaction.

[0063] The assay method according to the present invention includes the detection of amplified target sequences. The method according to the present invention is not limited to a specific detection scheme. Detection can be performed after amplification, such as by gel electrophoresis. Alternatively, in a single test tube, well or other reaction vessel, in the amplification reaction (real time) or when the amplification reaction ends (end point), the reagent present during the amplification is used for in-phase detection. Alternatively, by using a microfluidic device, the amplified product can be moved to a chamber where they are contacted with one or more detection reagents or separation reagents (such as immobilized capture probes). Detection reagents include double-stranded DNA binding dyes, such as SYBR Green, and fluorescent or luminescent labeled hybridization probes that produce signals once hybridized, such as molecular beacon probes or probe or a probe that is cleaved during amplification, such as a 5'-ribozyme Probe.

[0064] 2. Multipart Primers

[0065] As described above, the method of the present invention includes the use of multi-part primers for each rare target sequence. Amplification using multi-part primers is directed to primer 103 and the expected target sequence 101 ( Figure 1 ) is an example of Figure 2In the figure. First, primer 103, shown as a forward primer, anneals to target sequence 101 and is extended by a DNA polymerase using chain 101 as a template to produce an extension product 201. By referring to the middle figure, in the next amplification cycle, chain 202 (which contains primer 103 and extension product 201) becomes a template for a reverse primer (conventional primer 203). Reverse primer 203 anneals and is extended by a DNA polymerase using chain 202 as a template to produce an extension product 204. It will be observed that extension product 204 includes a sequence that is completely complementary to primer 103. Extension product 204 includes such a completely complementary sequence regardless of the sequence of chain 101. That is, if primer 103 has been extended in an earlier cycle (top figure), the resulting chain 202 (middle figure) always includes a complete complement of primer 103. In the next amplification cycle (lower figure), strand 205 (which includes reverse primer 203 and extension product 204) contains the perfect complement of primer 103; and primer 103 binds to strand 205 and is extended by the DNA polymerase to produce extension product 206. Thus, Figure 2 Applicable to mismatched target sequence 102 as well as expected target sequence 101 at any time when the multi-part primers anneal and extend to produce amplicon 202 .

[0066] As shown in the previous paragraph, Figure 2 The entirety of primer 103 is shown to be replicated during the extension of reverse primer 203. This establishes a long priming region for the next cycle, i.e., a sequence complementary to anchor sequence 104, bridge sequence 105, and foot sequence 106. In certain embodiments, it may not be desirable to perform the remainder of the amplification with a priming region of this length. Figure 3 The use of a multi-part primer with a blocking group to shorten the priming region in the later cycles is exemplified. As is well known, blocking groups are used to stop extension of DNA polymerase. For example, the blocking group can be hexaethylene glycol (HEG). In particular, Figure 3 As shown in the top figure, if the bridge sequence 105 is long, then it may be desirable to place the blocking group 108 in the bridge sequence 105. The priming region in the later amplification cycles consists of the nucleotides of the foot sequence 106 plus the nucleotides of the bridge sequence 105 located 3' to the blocking group 108. Alternatively, as Figure 3 As shown in the bottom figure, it may be desirable to place blocking group 108A in anchor sequence 104. In such an embodiment, the priming region in the later amplification cycle includes nucleotides of foot sequence 106 plus nucleotides of bridge sequence 105 plus nucleotides of anchor sequence 104 located 3' to blocking group 108A.

[0067] As described above, the multi-part primers used in the present invention may include a functional part, a 5' tail connected to the anchor sequence 104. The present invention does not limit the functions that such groups can perform or their structures. Several examples of functional parts are exemplified in Figure 4 In each figure, a multi-part primer 103 having an anchor sequence 104 and different functional groups at the 5' end of the anchor sequence is shown. Functional group 401 is simply an oligonucleotide tail that can be used to hybridize with a capture probe or hybridize with a labeled probe. As depicted, tail 401 is not complementary to another sequence in primer 103. Because of the presence of blocking group 108B in the primer containing tail 401, DNA polymerase does not replicate tail 401, and tail 401 is always single-stranded and can be used to bind to a capture probe or a labeled probe, regardless of whether the complementary amplicon is single-stranded or double-stranded. Oligonucleotide 401 can be used as a "zip code" for fixing the resulting amplicon to a specific position on an array of capture probes or to a capture probe connected to different elements of a distributed array. Another functional part includes a biotin group 402 connected to the anchor sequence 104 through a connector 403. Because of the presence of blocking group 108B in the primer, DNA polymerase does not replicate connector 403, and connector 403 is always single-stranded. The biotin group 402 enables the amplicon synthesized from the primer to acquire additional functions. For example, the biotin group allows the amplicon to be strongly captured by streptavidin protein fixed to a solid surface (such as paramagnetic beads) via a linker group. Another functional part is a hairpin oligonucleotide 404 having a stem-loop structure, which comprises a single-stranded loop 405 and a double-stranded stem 406, which is labeled with a quencher 407 (preferably a non-fluorescent quencher such as Dabcyl or Black Hole Quencher 2) and an interactive fluorescent part 408 (preferably a fluorophore). Extension of the reverse primer 203 ( Figure 2 ) will continue through the labeled hairpin 404, the quencher 407 separated from the fluorescent moiety 408, thereby generating a fluorescent signal. See Nazarenko et al. (1997) Nucleic Acids Res. 25: 2516-2521. The labeled hairpin 404 included in the primer 103 results in a fluorescent signal indicating amplification. Another functional part is a molecular beacon probe 409 connected to the anchor sequence 104 via an oligonucleotide sequence 414 and a blocking group 108B. This functional part has The additional function of the primer is to allow the tethered molecular probe to hybridize with the target strand (both the intended target sequence and the mismatched target sequence) downstream from the primer 103 when the replication 201 is generated. The molecular beacon probe 409 comprises a loop 410 and a stem 411 covalently linked to an interactive quencher 412 and a fluorescent moiety 413, so that hybridization of the probe 409 with the extension product 201 opens the stem 411 and generates a fluorescent signal indicative of amplification. Unlike the hairpin 404, the hairpin 409 is not replicated because in this case the primer 103 contains a blocking group 108B. Figure 4 The bottom diagram depicts a variant of the hairpin 404 in which the 5' end sequence of the stem 417 of the molecular beacon is complementary to a portion of the bridge sequence 105 and the loop comprises the anchor sequence 104. As a result, upon hybridization with the complementary amplicon strand, the rigidity of the resulting hybrid separates the interacting quencher 415 and fluorescent moiety 416, thereby generating a fluorescent signal indicative of amplification.

[0068] The multipart primer does not prime other sequences in the reaction mixture other than its target sequence (ie, the expected target sequence and the unintended, mismatched target sequence). The 3' portion of the bridge sequence plus the foot sequence do not together form a sequence that is a primer for such unrelated sequences.

[0069] By reference Figure 1 , the multi-part primers useful in the methods of the present invention function as follows. In the first round of synthesis, for example in the first PCR cycle, which may be after a high temperature denaturation step, the anchor sequence 104 hybridizes with the target sequence (both the expected target sequence 101 and the unexpected, mismatched target sequence 102). The bridge sequence 105 does not hybridize with the target sequence. The foot sequence 106 preferably hybridizes with the expected target sequence 101, but to some extent it also hybridizes with the unexpected, mismatched target sequence 102. Hybrids are formed and separated with a certain frequency. DNA polymerase also binds to the formed hybrids with a certain frequency and initiates primer extension. For the expected target sequence 101, when comparing PCR amplification and the use of the multi-part primers and 10 6 When using a detection assay with SYBR Green detection for 100 copies of the expected target sequence 101 (with or without copies of the unintended target sequence 102) and the same assay using corresponding conventional primers (which are similar to the anchor sequence in the multi-part primer), the combined frequency of hybrid formation and polymerase binding / extension results in inefficient replication of the expected target sequence 101, which the inventors measured as C 20 at least two cycles in the PCR threshold cycle. T For unintended, mismatched target sequences 102, the combined frequency of hybrid formation and polymerase binding / extension results in extremely inefficient replication, which the inventors measure in such an assay as the delay of 10% for each primer sequence with multipart primers and 106 The C obtained by mismatching the target sequence 102 T and using multipart primers and 10 6 The C obtained by T The difference between T The delay of the expected target sequence caused by the multi-part primer is at least two PCR cycles, and can be greater, such as four cycles or even 5-10 cycles. The difference (ΔC T ) is at least ten PCR cycles, preferably more. As amplification proceeds through additional cycles, the expected target sequence will be duplicated, and so will the final mismatched target sequence. The copies synthesized from the two target sequences contain multiple partial primers and are therefore identical.

[0070] After the synthesis of the amplicon on the target nucleic acid molecule present in the sample to be tested before the multi-part primer initial amplification, no matter the initialization occurs in the first cycle or in the subsequent cycle, the resulting amplicon is then rapidly exponentially amplified with normal, high efficiency in subsequent cycles, wherein the multi-part primer is used as a conventional primer for the amplicon. For example, for the replication of the amplicon, the multi-part primer plays a role in the same manner as the conventional PCR primer of 20-50 nucleotides in length. This means that once amplification has begun, not only the foot sequence is used as a primer. One possibility is that the entire (or at least the entire except the functional part located at the 5' of the blocking group, such as 401, 403 and 409) is replicated and used as a primer for the replication of the amplicon.

[0071] In those embodiments in which there is a blocking group in the multi-part primer, the purpose of the blocking group is to prevent the replication of some parts of the 5' end of the primer. Blocking groups are familiar to those skilled in the art. For example, the blocking group can be hexaethylene glycol or an abasic nucleotide lacking a nitrogenous base. The blocking group can be placed at the 5' end of the anchor sequence 104 to prevent the replication of the functional part, such as the placement of the blocking group 108B relative to the functional part 401, 403 or 409; or it can be placed at any position within the anchor sequence 104, such as the placement of the blocking group 108A; or it can be placed in the bridge sequence 105, such as the placement of the blocking group 108; that is, as long as the length of the shortened sequence replicated when the template molecule is an amplicon is sufficient to serve as an effective primer. For illustration, assume that the multi-part primer has a foot sequence of six nucleotides long and people want 35 nucleotides to be replicated. If the bridge sequence 105 is twenty-four nucleotides long, five nucleotides of the anchor sequence 104 must be downstream (i.e., 3') of the blocking group to achieve the desired primer length.

[0072] 3. Nomenclature

[0073] In the examples disclosed below, two nomenclatures are used to refer to the numerous multi-part primers of the present invention.

[0074] In one nomenclature, multipart primers are referred to in such a format, for example, a "24-14-5:1:1" primer refers to an anchor sequence of 24 nucleotides, a bridge sequence of 14 nucleotides, and a foot sequence of seven nucleotides (comprising, from the 5' end of the foot sequence, five nucleotides complementary to both the mutant (MUT) and wild-type (WT) targets, a query nucleotide that is not complementary to the corresponding nucleotide in the WT target but complementary to the corresponding nucleotide in the MUT target, and, finally, a nucleotide complementary to both targets). Because the query nucleotide is located one nucleotide inside the 3' end of the primer, the inventors refer to this nucleotide as being located at the "3'-penultimate position".

[0075] By comparing the bridge sequence with the region of the target sequence located between the binding sequence of the anchor sequence and the binding sequence of the foot sequence (which the inventors refer to as the "insertion sequence"), it can be seen that in some of the following examples the insertion sequence is fourteen nucleotides long, which is the same length as the bridge sequence, while in other examples (such as Example 8), the insertion sequence and the bridge sequence are of different lengths. In order to specify the length of the insertion sequence, the second nomenclature is sometimes used. In this case, the multipart primer "24-18 / 10-5:1:1" means that its 5'-anchor sequence is 24 nucleotides long, its bridge sequence is 18 nucleotides long and it is present opposite to the insertion sequence of 10 nucleotides in the template, and its 3'-foot sequence is 7 nucleotides long and consists of a 5' segment that is fully complementary to both the mutant and the wild-type template, followed by a query sequence that is complementary to the corresponding nucleotides in the mutant template only, followed by a 3' nucleotide that is complementary to the corresponding nucleotides in both the mutant and wild-type templates.

[0076] The sequence of the bridge sequence is selected so that it is not complementary to the insert sequence to prevent the bridge sequence from hybridizing with the insert sequence during primer annealing. Instead of annealing to each other, the bridge sequence and the insert sequence form a single-stranded "bubble" when both the anchor sequence and the foot sequence are hybridized to the template. The inventors sometimes refer to the combination of the bridge sequence and the insert sequence as a bubble. For example, the name 24-14 / 14-5:1:1 can be considered to have a "14 / 14 bubble".

[0077] The "perimeter of the bubble" is defined as the sum of the number of nucleotides in the bridge sequence plus the number of nucleotides in the insert sequence plus the 3' nucleotides of the anchor sequence and its complement plus the 5'-terminal nucleotides of the foot sequence and its complement. Thus, the perimeter of the bubble formed by the binding of the 24-14 / 14-5:1:1 multipartite primer (14 / 14 bubble) to the template molecule is 14+14+2+2, which is equal to 32 nucleotides in length. The following table lists some of the primers used in the following examples, using this second format.

[0078] Exemplary primers used in PCR assays

[0079] Primers Sequence (5' to 3') SEQ ID NO: EGFR L858R Normal forward CTGGTGAAAACACCGCAGCATGTC 27 Conventional reverse GCATGGTATTCTTTCTCTTCCGCA 3 24–14 / 14–5:1:1 <![CDATA[CTGGTGAAAACACCGCAGCATGTC GCACGAGTGAGCCC TGGGC G G]]> 6 24–14 / 14-4:1:1 <![CDATA[TGGTGAAAACACCGCAGCATGTCA CACGAGTGAGCCCC GGGC G G]]> 7 24–14 / 14–5:1:1 <![CDATA[CTGGTGAAAACACCGCAGCATGTC GCACGAGTGAGCCC TGGGC G G]]> 6 24–14 / 14-6:1:1 <![CDATA[ACTGGTGAAAACACCGCAGCATGT TGGAGCTGTGAGC CTTGGGC G G]]> 8 24–14 / 14-6:1:0 <![CDATA[ACTGGTGAAAACACCGCAGCATGT TGCACGAGTGAGCC TTGGGC G ]]> 11 24–14 / 14–5:1:1 <![CDATA[CTGGTGAAAACACCGCAGCATGTC GCACGAGTGAGCCC TGGGC G G]]> 6 24–14 / 14-4:1:2 <![CDATA[TGGTGAAAACACCGCAGCATGTCA CACGAGTGAGCCAC GGGC G GG]]> 12 24–14 / 14-3:1:3 <![CDATA[GGTGAAAACACCGCAGCATGTCAA ACGAGTGAGCCACA GGC G GGC]]> 13 24–14 / 14-2:1:4 <![CDATA[GTGAAAACACCGCAGCATGTCAAG GAAGTGAGCCACAA GC G GGCC]]> 14 24–14 / 14-1:1:5 <![CDATA[TGAAAACACCGCAGCATGTCAAGA CAGACTGACCCAA AC G GGCCA]]> 15 24-10 / 10–5:1:1 <![CDATA[TGAAAACACCGCAGCATGTCAAGA CACTCAGCCC TGGGC G G]]> 10 24–14 / 14–5:1:1 <![CDATA[CTGGTGAAAACACCGCAGCATGTC GCACGAGTGAGCCC TGGGC G G]]> 6 24-18 / 18–5:1:1 <![CDATA[CGTACTGGTGAAAACACCGCAGCA CTGACGACAAGTGAGCCC TGGGC G G]]> 9 24-18 / 10–5:1:1 <![CDATA[TGAAAACACCGCAGCATGTCAAGA CACACGACAAGTGAGCCC TGGGC G G]]> 16 24-16 / 12–5:1:1 <![CDATA[GGTGAAAACACCGCAGCATGTCAA TCCAACAAGTGAGCCC TGGGC G G]]> 17 24–14 / 14–5:1:1 <![CDATA[CTGGTGAAAACACCGCAGCATGTC GCACGAGTGAGCCC TGGGC G G]]> 6 24-12 / 16–5:1:1 <![CDATA[TACTGGTGAAAACACCGCAGCATG GACGACGAGCC CTGGGC G G]]> 18 24-10 / 18–5:1:1 <![CDATA[CGTACTGGTGAAAACACCGCAGCA CTGACGGCCC TGGGC G G]]> 19 B-raf V600E 24–14 / 14–5:1:1 <![CDATA[AGACAACTGTTCAAACTGATGGGA AAACACAATCATC TATTTC T C]]> 20 Conventional reverse ATAGGTGATTTTGGTCTAGC 22

[0080] The bridge sequence within each superselective primer is underlined, and the query nucleotide in its foot sequence is indicated by underlined and bold letters. The primers are arranged in groups reflecting their use in comparative experiments.

[0081] 4. Purpose

[0082] The present invention is not limited to a specific intended target, a specific amplification method or a specific instrument. For comparison purposes, the inventors present several series of experiments in Examples 1-8, which utilize the same intended target, EGFR mutation L858, and a homogeneous PCR assay initiated with plasmid DNA, which utilizes Green detection and using the same thermal cycler, the Bio-Rad IQ5 fluorescence spectrophotometric thermal cycler. The inventors have performed other assays that gave results consistent with those reported in the Examples. Such assays have utilized other intended targets, including human EGFR mutant T790M and human B-raf mutant V600E; have utilized genomic DNA; have included detection with molecular beacon probes; have utilized different PCR parameters; and have utilized a different instrument, the ABI PRISM 7700 fluorescence spectrophotometric thermal cycler.

[0083] Example 1 is a control assay in which a conventional PCR forward primer of 21 nucleotides in length was used to amplify a perfectly matched expected target sequence and also to amplify an unexpected, mismatched target sequence that differed in that a single nucleotide polymorphism was located near the middle of the sequence to which the primer binds (as in the other examples, a conventional PCR reverse primer was also used here). SYBR It enables in-phase detection of double-stranded amplification products, significantly increasing their fluorescence when binding to double-stranded amplicons. As a result, the SYBR The fluorescence intensity provides an accurate indication of the amount of amplicon present. Figure 5 The real-time fluorescence kinetic curve (fluorescence intensity vs. amplification cycle number) presented in 12 Sufficient double-stranded product is generated on the order of 100 amplicon to give a detectable signal above background at a point where approximately 20 PCR cycles have been performed (threshold cycle, abbreviated as "C T ”), which is typically for 10 6 PCR assay initiated by a template. Figure 5 It was also shown that the forward primer was less selective for the intended target relative to the unintended, mismatched target, i.e., at the threshold cycle (C T ) There is only a small thermodynamic difference in the stability of fully complementary hybrids compared to the stability of mismatched hybrids (resulting in a virtually unobservable delay in the appearance of amplicons resulting from the less likely mismatched primer-target hybrids).

[0084] Example 2 describes two additional controls in which the substituted nucleotide in the mismatched target is first placed at the 3' terminal nucleotide of a conventional forward primer, a well-known ARMS technique, and then placed at one nucleotide inside the 3' terminal nucleotide of a conventional forward primer. The inventors sometimes refer to the nucleotide at the position within the primer sequence as a "query nucleotide" that will be relative to the nucleotide in the target where the SNP is present or absent. The real-time kinetic curves for these controls are shown in Figure 6 In the figure, it can be seen that for the expected target, C T Still close to 20 cycles, indicating that for the intended target, the amplification reaction was as efficient as reported in Example 1. However, for the mismatched target, C T In the case where the primer has a query nucleotide at the 3'-end of the foot sequence, Figure 6 A, The delay is approximately 10 cycles (11 cycles), which shows a thousand-fold selectivity in favor of the expected perfectly matched target (2 10 In the case where the query nucleotide is located at the penultimate position at the 3' end of the foot sequence, C TSomewhat smaller, about 8 cycles. By comparing Examples 1 and 2, it is seen that the amplification efficiency of the intended target is not reduced by placing the query sequence at or near the 3' end of the primer, but the intended selectivity relative to the unintended target that differs by a single nucleotide is improved. The inventors understand that the selectivity is limited because, due to keto-enol tautomerism, there is occasionally some base pairing of the mismatched query nucleotide with the non-complementary nucleotide in the target sequence, and therefore some undesired extension does occur, so the probability of generating an amplicon is the product of the probability of the hybrid to be formed multiplied by the probability that the resulting hybrid forms a structure that can be extended.

[0085] Example 3 shows the same experiment with a multi-part primer according to the present invention. The inventors record the primers used here as 24-14-5:1:1. The first number 24 is the nucleotide length of the anchor sequence. The second number 14 is the nucleotide length of the bridge sequence (and in this experiment, as in other experiments as described herein, the insert sequence in the target is the same length as the bridge sequence unless explicitly noted by the inventors). The last three numbers (5:1:1) describe the foot sequence, giving the number of nucleotides 5' of the query nucleotide, followed by the number of query nucleotides (which is 1 for all experiments described herein), and finally the number of nucleotides 3' of the query nucleotide. Therefore, in this case, the foot sequence is a sequence seven nucleotides long with the penultimate query nucleotide. These are prepared using SYBR The results of a real-time assay that measures the amount of amplicons present after each thermal cycle is complete (determined at the end of the chain extension phase of each cycle) using fluorescence intensity is shown in Figure 7 By comparison Figure 5 and 7 It can be seen that in this case the expected perfect match to the target C T It can also be seen that the C T was delayed even more, giving a ΔC of approximately 19 cycles between the expected and unexpected target sequences. T , which is approximately a 500,000-fold difference in selectivity (2 19 is 524,288).

[0086] While not wishing to be bound by any theory, the inventors believe the following to be true:

[0087] A. Even though the foot sequence is tethered to the template by the anchor hybrid, the foot sequence is so short and separated from the anchor hybrid by such a large bubble (which contains the bridge sequence of the primer and the insert sequence in the template), and the annealing temperature is so high for the short foot sequence that at any given moment (under the equilibrium regulation of the annealing phase of the PCR assay), only a very small fraction of the template molecules present in the sample to be tested are hybridized to the foot sequence at any given moment.

[0088] B. Furthermore, the hybrids that do form between the foot and target sequences are relatively weak, so the average time they persist is very short (perhaps a hundred microseconds).

[0089] C. As a result of the reduced probability of a hybrid being present at any given time and the reduced average residence time of the resulting weak hybrids, the probability of forming a stable (extendable) complex between a hybrid (even a fully complementary hybrid) and a DNA polymerase molecule is extremely low.

[0090] D. In PCR assays performed with the preferred multipart primer design this can be seen as a delay of approximately 10 cycles in the appearance of the amplicon made from the fully complementary ("mutant") target (i.e., not the 10-cycle delay seen when conventional linear primers are used). 6 There are about 20 C T ), Ct is about 30. T The increase in the value of 10 thermal cycles indicates that the probability of forming a stable complex between a DNA polymerase molecule and a fully complementary foot hybrid is 1 / 1,000 of the probability when a conventional linear primer is employed under the same reaction conditions.

[0091] E. Under these same PCR conditions, using the same preferred multipart primer design, the C obtained with the mismatched ("wild-type") target T The C T The C value was almost 20 cycles later. Therefore, the C value was almost 20 cycles later than that of the C value obtained when a conventional linear primer was used instead of a multipart primer under the same conditions. T Compared to the CT value, there is a delay of about 30 cycles in the appearance of amplicons from these mismatched targets. Therefore, the probability of forming a stable complex between a DNA polymerase and a hybrid containing a foot sequence that binds to a mismatched foot target sequence is much lower. The increase in the CT value over 30 cycles indicates that the probability of forming a stable complex between a DNA polymerase molecule and a mismatched foot hybrid is 1 / 1,000,000,000 of the probability when a conventional linear primer is used under the same reaction conditions.

[0092] F. This significantly lower probability of forming an extendable complex between an unintended target sequence and a DNA polymerase molecule is a product of the following distinguishing elements: (i) the lower stability of mismatched hybrids (compared to the stability of fully complementary hybrids) significantly reduces the fraction of mismatched hybrids that exist at any given time (compared to the fraction of fully complementary hybrids that can exist at any given time); and (ii) the lower stability of mismatched hybrids results in a shorter average retention time of the hybrids, thereby significantly reducing the ability of DNA polymerase molecules (which undergo constant Brownian motion) to find hybrids (and form stabilized complexes with them).

[0093] Example 4 shows that the assay of Example 3 can easily distinguish between 6 samples with 10 copies of the unintended target sequence and in the presence of 6 Different results were obtained for samples containing ten or more copies of the expected target sequence in the absence of 10 copies of the unintended target sequence. 6 10 copies of the unintended target sequence in the reaction mixture 6 , 10 5 , 10 4 , 10 3 , 10 2 , 10 1 The real-time PCR results obtained with 10 copies of the expected target sequence are shown in Figure 8 and the C determined for these results T Shown in Fig. 9 , where they are plotted against the logarithm of the starting copy number of the expected target sequence. By referring to these graphs, it can be seen that for every thousandth decrease in the concentration of the expected target sequence, the SYBR Fluorescent C T is delayed by approximately 10 cycles and has 10 copies of the expected target sequence plus 10 6 The samples with 10 copies of the unexpected target sequence were compared with samples without the expected target sequence and 10 6 The sample area containing 100,000 copies of the unexpected target sequence is separated; that is, it is possible to detect one mutant sequence in a population of 100,000 copies of the corresponding wild-type sequence. Moreover, the assay is quantitative, with a threshold cycle corresponding to the logarithm of the mutant copy number in the starting reaction mixture.

[0094] These results confirm the following aspects of the use of the selective primers according to the invention:

[0095] A. Once the multipart primers bind to the DNA polymerase during the annealing phase of the PCR assay to form a hybrid, the stabilized hybrid is extended during the elongation phase of the PCR assay, and the resulting amplicons are then amplified with high efficiency (just as if the reaction were performed using classical linear primers). This can be seen from the fact that the amount of mutant template originally present in the sample was reduced by a factor of 1,000 with a delay of about 10 thermal cycles resulting in the appearance of significant amounts of amplicons (e.g., as shown in the results of which are shown in Figure 8 and Fig. 9 In the experiment in , the C of the sample with 100,000 mutant templates T The C value of the sample with 100 mutant templates was about 27. T = 37). If the number of amplicons present effectively doubles with each thermal cycle, then after ten cycles there should be 1,024 times as many amplicons (i.e., 2 10 ). These results confirm that amplicons generated from mutant templates present in the test samples are then efficiently amplified.

[0096] B. Efficient amplification of the amplicon occurs because once the multipartite primer is incorporated into the 5' end of the product amplicon (amplicon "+" strand), the complementary amplicon (amplicon "-" strand) produced in the next cycle of synthesis has a sequence at its 3' end that is completely complementary to the entire sequence of the multipartite primer. As a result, with respect to the amplicon (relative to the original template molecule), the multipartite primers behave as if they were classical linear primers for further amplification of the amplicon.

[0097] C. The highly selective generation of amplicons from fully complementary mutant templates present in the test sample (compared to amplicons generated from mismatched wild-type templates present in the test sample), combined with the efficient amplification of the amplicons by primers once the amplicons are synthesized, enables the resulting real-time data to be used to quantitatively measure the number of mutant template molecules present in the test sample.

[0098] As shown in the C obtained from samples containing different numbers of mutant template molecules T As reflected by the values ​​of , there is an inverse linear relationship between the logarithm of the number of target molecules present in the sample to be tested and the number of thermal cycles used to synthesize a predetermined amount of amplicon (in an exponential amplification reaction, such as a PCR assay). See Kramer & Lizardi (1989) Nature 339:401-402. As shown in the results, for example, Fig. 9 In the experiment, C TThe linearity of the plot of the logarithm of the expected number of (mutant) template molecules present in each sample tested indicates that no significant amount of amplicons are produced from the wild-type template (even though 1,000,000 wild-type template molecules are present in each sample). If there were significant amounts of amplicons produced from the wild-type template, then the C values ​​for samples containing only a small amount of mutant template would be significantly higher than those for samples containing only a small amount of mutant template. T The values ​​would have been fewer (ie, the results would not have formed a straight line because the appearance of the amplicons from the very rare intended target molecules would have been obscured by the appearance of the amplicons from the very rare intended target molecules).

[0099] As reported in Example 5, the inventors investigated the effect of the foot sequence length of the multipart primer on the amplification reaction using the assay of Example 4 with a series of three probes: 24-14-4: 1: 1, 24-14-5: 1: 1 and 24-14-6: 1: 1. The length of the anchor sequence was kept at 24 nucleotides. The length of the bridge sequence was kept at 14 nucleotides, the same single nucleotide difference between the target sequences was kept, and the position of the query nucleotide was kept at the penultimate position from the 3' end of the foot sequence. The length of the foot sequence was varied from 6 nucleotides to 7 nucleotides to 8 nucleotides by varying the number of nucleotides 5' to the query nucleotide position from 4 to 5 to 6. The C sequences obtained were T The values ​​are summarized in Table 1 and plotted against the logarithm of the expected starting copy number of the target sequence. Fig.10 Fit lines 1001 (foot sequence length 6 nucleotides), 1002 (foot sequence length 7 nucleotides) and 1003 (foot sequence length 8 nucleotides) to the data. Fig. 9 As reported, it can be seen that all three primers provide quantitative data. It can also be seen that the fitted lines 1001, 1002 and 1003 are nearly parallel, indicating that for all three foot sequence lengths, the T The same quantitative relationship exists between the logarithm of the starting copy number. Fig.10 It was also shown that shortening the length of the foot sequence delayed the C T , but if Fig.10 As seen in the figure, shortening the length of the foot sequence also gives 6 Up to 10 1 The shorter the sequence length, the less likely it is that amplicons synthesized by abundant unintended target molecules in the test sample will hinder amplicons synthesized by rare expected target molecules present in the same sample.

[0100] As reported in Example 6, the inventors also investigated the effect of the perimeter of the bubble formed by the bridge sequence of the multipartite primer and the insert sequence of the expected and unexpected target sequence on amplification using the assay of Example 4 with a series of three primers: 24-10-5:1:1, 24-14-5:1:1 and 24-18-5:1:1. The inventors kept the length of the anchor sequence at 24 nucleotides; the inventors kept the foot sequence at 5:1:1; and the inventors varied the length of the bridge sequence from 10 to 14 to 18 nucleotides, and for each multipartite primer the sequence of the anchor sequence was chosen so that the insert sequence in the target would be the same length as the bridge sequence in that primer. As a result, the perimeter (expressed in nucleotides) of the bubble formed by each of the three primers when their foot sequence hybridized to the target (including the four nucleotides contributed by the anchor sequence hybrid and the foot sequence hybrid) was 24, 32 and 40, respectively. The C obtained T The values ​​are summarized in Table 2 and plotted against the logarithm of the expected starting copy number of the target sequence. Fig.11 Lines 1101 (bubble perimeter 24 nucleotides), 1102 (bubble perimeter 32 nucleotides), and 1103 (bubble perimeter 40 nucleotides) were fitted to the data. Fig. 9 and Fig.10 As reported, it can be seen that all three primers provide quantitative data. It can also be seen that the fitted lines 1101, 1102 and 1103 are nearly parallel, indicating that for all three bubble perimeters, the C T The same quantitative relationship exists between the logarithm of the starting copy number. Fig.11 It was also shown that increasing the bleb perimeter delayed C T , but if Fig.11 As seen in the figure, increasing the bubble perimeter also gives 6 Up to 10 1 The larger the bubble, the less likely it is that amplicons synthesized by abundant unintended target molecules in the test sample will obstruct amplicons synthesized by rare expected target molecules present in the same sample.

[0101] These experimental observations demonstrate that shorter foot sequence lengths and / or larger bubbles result in considerably less likely hybrid formation and that shorter foot sequence lengths and / or larger bubbles result in increased selectivity against mismatched wild-type templates, which is determined by C T The enhanced linearity of the plot of the logarithm of the number of target molecules vs. expected is demonstrated. To understand why this is the case, the inventors examined the thermodynamics of the formation of hybrids of the sequences under the equilibrium conditions present during the annealing phase of a PCR assay. The following is the inventors' understanding:

[0102] A. In the inventors' PCR assay there is a very high concentration of multipart primers (because there needs to be enough available multipart primers to incorporate the approximately 10 13 amplicons). As a result, under the equilibrium conditions that exist during the annealing phase of these PCR assays, nearly every template molecule rapidly binds to the anchor sequence of the multipart primer. Furthermore, because the anchor sequence is long (e.g., 24 nucleotides), the bond between the anchor sequence and the template molecule is very strong and persists for a long time on average (perhaps measured in minutes). In equilibrium, in a very small fraction of these anchor complexes, the short foot sequence also hybridizes to the template molecule. At any given moment, the concentration of anchor complexes whose foot sequences are not hybridized is "[A]", and the concentration of anchor complexes whose foot sequences are hybridized is "[B]". The classical equilibrium constant ("k") that describes the intrinsic relationship between these two states is:

[0103] k = [B] / [A] Equation 1

[0104] Thermodynamically, the probability of forming a hybrid at equilibrium depends on both the hybridization strength (enthalpy) and the physical relationship that determines the probability that two sequences can interact to form a hybrid (entropy). The equilibrium constant can be determined by the change in enthalpy (there is enthalpy once the anchor complex whose foot sequence is not hybridized is converted to the foot sequence that is hybridized, ΔH) and by the change in entropy (there is entropy once the anchor complex whose foot sequence is not hybridized is converted to the foot sequence that is hybridized, ΔS) according to the following classical formula:

[0105] (ΔH-TΔS) = -RT ln(k) Equation 2

[0106] where R is the thermodynamic gas constant, T is the temperature in degrees Kelvin, and ln(k) is the natural logarithm of the equilibrium constant. Rearrange this equation to obtain an expression for k:

[0107] Equation 3

[0108] where e = 2.71828. For the exact same reaction, the fraction of complexes with hybridized foot sequences (θ) is represented by the equation: θ = [B] / ([A]+[B]). However, as [B] becomes very small (as is the case for reactions with multipart primers), θ approaches 0, and the equation for θ is expressed as follows:

[0109]

[0110] Since the expression of Θ in Equation 4 is almost the same as the expression of k in Equation 1, the inventors can replace k in Equation 3 with Θ to obtain an equation that relates very low abundance primer-template complexes (which have a foot of hybridization (Θ)) to classical thermodynamic parameters (ΔH and ΔS) as follows:

[0111] Equation 5

[0112] For a nucleic acid hybridization reaction occurring under PCR conditions, the quantity (ΔH-TΔS) is a positive value, and therefore, e increases to a negative number, giving a fractional value of θ. The smaller the value of (ΔH-TΔS), the smaller the fraction θ. Furthermore, during the annealing phase of a PCR reaction, T is a constant. Therefore, to understand how θ changes as a result of changes in the multi-part primer design, the inventors need only consider the magnitude of the values ​​of ΔH and ΔS for each primer design to understand the effect of the design when the multi-part primers hybridize to the intended target compared to when they hybridize to an unintended target.

[0113] B. Entropy is a measure of the number of conformationally distinct states that a molecular complex can form. Thus, when the foot sequence of the anchoring complex hybridizes to its target, the number of topologically distinct states that the complex can form changes from a high number to a low number. Thus, once the foot hybrid is formed, the change in entropy (ΔS) is negative.

[0114] C. Enthalpy is a measure of the stability of a molecular complex, expressed as the amount of energy present in a solution containing the complex. Because high temperatures are necessary to dissociate nucleic acid hybrids, thermal energy is added as the complex breaks apart and released once the complex is formed. Therefore, once a hybrid is formed, the change in enthalpy (ΔH) is also negative.

[0115] D. When multi-part primers are used in PCR assays, the fraction of complexes that have hybridized foot sequences (Θ) is well described by Equation 5. In the above experiments (where the length of the foot sequence is varied or the perimeter of the bubble is varied), the only variables are ΔH and ΔS. For the formation of foot hybrids, ΔH and ΔS are negative, and the quantity (ΔH-TΔS), which is called the Gibbs free energy (ΔG), is positive. As a result, the quantity TΔS is more negative than ΔH. With respect to calculating the fraction of complexes that have hybridized foot sequences (Θ), the smaller the magnitude of the negative ΔH, the smaller Θ will be. Similarly, the larger the magnitude of the negative ΔS, the smaller Θ will be.

[0116] E. To determine the effect of different foot sequence lengths on the fraction of complexes with foot hybrids (θ), it is necessary to recognize that, all else being equal, the less negative ΔH is, the shorter the foot hybrid length is. As a result, at any given moment, the shorter the foot hybrid length, the lower the fraction of primer-target complexes with foot hybrids.

[0117] F. Similarly, to determine the effect of different bubble perimeters on the fraction of complexes with foot hybrids (θ), it is necessary to recognize that, all else being equal, the more negative ΔS is, the larger the bubble perimeter is. As a result, at any given moment, the larger the bubble perimeter is, the lower the fraction of primer-target complexes with foot hybrids is.

[0118] G. Given these insights, let us now look at how the design of the foot sequence in a multipart primer can help distinguish between perfectly complementary target sequences (expected target sequences) and mismatched target sequences (unexpected target sequences). For example, the results for which are shown in Figure 8 and Fig. 9 The multipart primers for the experiments in have foot sequences of different lengths ("6:1:1" or "5:1:1" or "4:1:1"). These designations indicate that the total length of each foot sequence is 8 nucleotides, 7 nucleotides, or 6 nucleotides, respectively, with a query nucleotide located at the penultimate position from the 3' end of the primer (which is either complementary to the corresponding nucleotide in the expected target sequence or non-complementary to the corresponding nucleotide in the unintended target sequence).

[0119] H. The reason the inventors positioned the critical nucleotide at the penultimate position is that the inventors believe that when the penultimate base pair cannot form (due to a mismatch), the terminal base pair cannot form (even if the 3' nucleotide of the foot sequence is complementary to the corresponding nucleotide in the target) because isolated base pairs are extremely unlikely to be stable at the annealing temperature of the PCR assay (approximately 60° C.). Therefore, for a given foot sequence, a mismatched hybrid will be two base pairs shorter than a fully complementary hybrid.

[0120] I. Here is what this means (conceptually): To illustrate this, assume that temperature (T) = 1, and assume that the gas constant (R) = 1, because they are constants. Imagine that the ΔH value for the formation of a fully complementary hybrid with a 6:1:1 foot sequence is -16 and the ΔH value for the formation of a shorter mismatched hybrid with a 6:1:1 foot sequence is -12. Also imagine that the ΔS value for both of these hybrids (which is determined by the bubble perimeter) is -20. As a result, the ΔG value for the fully complementary hybrid is 4 (written as 20-16), and the ΔG value for the mismatched hybrid is 8 (written as 20-12). Substituting these values ​​into Equation 5, with the conceptual value of θ for the hybrid formed with the expected target (θ m ) is equal to e -4 , which is a value of 0.0183. In contrast, the conceptual value of θ for hybrids formed with unintended targets (θ w ) is equal to e -8, which is a value of 0.000335. Thus, in this conceptual example, the abundance of fully complementary hybrids is 54.6 times greater than the abundance of mismatched hybrids. Although this calculation illustrates that the use of multipart primers according to the present invention results in a much lower probability of a hybrid being formed with an unintended target being present (at any given moment) than the probability of a hybrid being formed with the intended target being present (at any given moment), and despite the C of the amplicon synthesized with the intended target being less than the C of the amplicon synthesized with the intended target being present, the probability of a hybrid being formed with the intended target being present (at any given moment) is significantly lower than the C of the amplicon synthesized with the intended target being present. T This difference, of course, results in the C of amplicons synthesized from unintended targets. T The greater delay of the conceptual embodiment Θ m and θ w Actual values ​​are different.

[0121] J. Let us now perform the same conceptual calculation for a multipart primer with a 5:1:1 foot sequence. In this case, the ΔH value for forming a fully complementary hybrid with a 5:1:1 foot sequence is -14 and the ΔH value for forming a mismatched hybrid with a 4:1:1 foot sequence is -10; and the resulting ΔG values ​​(for bubbles of the same size, where ΔS = -20) are as follows: the ΔG value for the fully complementary hybrid is 6 (denoted as 20-14 = 6), and the ΔG value for the mismatched hybrid is 10 (denoted as 20-10). Substituting these values ​​into Equation 5, the conceptual value of θ for the hybrid formed with the expected target (θ) is m ) is equal to e -6 , which is a value of 0.00248. In contrast, the conceptual value of θ for hybrids formed with unintended targets (θ w ) is equal to e -10 , which is a value of 0.0000454. Surprisingly, in this conceptual example, the abundance of fully complementary hybrids is also 54.6 times greater than the abundance of mismatched hybrids.

[0122] K. Let us now perform the same conceptual calculation for a multipartite primer with a 4:1:1 foot sequence. In this case, the ΔH value for forming a fully complementary hybrid with a 4:1:1 foot sequence is -12 and the ΔH value for forming a mismatched hybrid with a 4:1:1 foot sequence is -8; and the resulting ΔG values ​​(for bubbles of the same size, where ΔS = -20) are as follows: the ΔG value for the fully complementary hybrid is 8 (denoted as 20-12 = 8), and the ΔG value for the mismatched hybrid is 12 (denoted as 20-8). Substituting these values ​​into Equation 5, the conceptual value of θ for the hybrid formed with the expected target (θ) is m ) is equal to e -8 , which is a value of 0.000335. In contrast, the conceptual value of θ for hybrids formed with unintended targets (θ w ) is equal to e -12, which is a value of 0.00000614. And even more surprisingly, in this conceptual example, the abundance of fully complementary hybrids is also 54.6 times that of mismatched hybrids. Therefore, the inventors conclude that even though shorter feet lead to lower θ values, and even though shorter feet lead to increased C T From a strictly thermodynamic point of view, there is no reason to believe that shorter foot sequences lead to stronger discrimination between intended and unintended target sequences.

[0123] L. Furthermore, even though increasing bubble perimeter also decreases the value of Θ, it is clear that increasing bubble perimeter, while making hybrid formation less likely, does not change the equilibrium ratio of foot hybrids formed from intended targets compared to foot hybrids formed from unintended hybrids.

[0124] M. In terms of classical thermodynamic analysis, it can be seen that for any given multipartite primer (where the fraction of molecular complexes that form foot hybrids is extremely low), the fraction of foot hybrids formed with the expected target (θ m ) and the fraction of hybrids formed with unintended targets (θ w ) is not affected by increasing the perimeter of the bubble (which changes ΔS) nor by decreasing the length of the foot sequence (which changes ΔH), but rather these changes reduce Θ m and θ w The values ​​of both, but do not change the ratio (Θ m / Θ w ), which is the difference in enthalpy (ΔH m -ΔH w ). As a result, from a classical thermodynamic point of view, the only thing that affects the relative abundance of the intended hybrid compared to the unintended hybrid is the difference in their enthalpy, and this difference is a result of the difference in the number of base pairs formed, which is the same no matter how long the base sequence is. m / Θ w ) is as follows:

[0125] Equation 6

[0126] Fig.10 and Fig.11 The experimental results shown in demonstrate that increasing the perimeter of the bubble and reducing the length of the foot sequence significantly improve the selectivity of the multipartite primers according to the invention, i.e., these changes in the multipartite primer design, although reducing the abundance of foot hybrids, significantly improve the discriminatory rate (θ m / Θ w ), because this improvement in discrimination is used by 10 6 The C obtained for each expected target molecule T and use 10 6C obtained for unintended target molecules T C T The difference in values ​​(ΔC T These observations suggest that there are additional (perhaps non-thermodynamic) reasons for the extraordinary selectivity of the multipart primers according to the invention.

[0127] The explanation for the enhanced selectivity that occurs when the multipartite primers according to the invention are designed to reduce the proportion of foot targets present at any time under equilibrium conditions during the annealing phase of a PCR amplification assay cannot lie in the discriminating results of ARMS, since the extent to which DNA polymerase molecules reject hybrids that do not have a base pair that includes the 3' terminal nucleotide of the primer is the same regardless of the abundance of those primers. However, it is apparent from the experimental results that additional discriminating mechanisms enable the extraordinary selectivity to occur when primers are designed to rarely form foot hybrids.

[0128] Although not wishing to be bound by any theory, this is why the inventors believe that the length of the foot sequence and the perimeter of the increased bubble enhance the reason for selectivity. The explanation lies in the unexpected discovery of the inventors: at the relatively high temperature present during the PCR assay annealing phase, very short foot hybrids only exist for a very short time (perhaps measured in tens or hundreds of microseconds) before they dissociate. In addition, the shorter the hybrid and the larger the bubble perimeter, the shorter the average time during the existence of the hybrid. The inventors speculate that the shorter the average duration of a particular type of hybrid, the less likely it is that the DNA polymerase molecule will encounter one of those hybrids and then form a stable complex that can extend the chain with the hybrid. The key point here is that whether the hybrid will form a stable complex with the DNA polymerase molecule is a function of the average duration of the hybrid. The inventors believe that when the foot sequence length of the primer is reduced and the bubble perimeter of the primer is increased, the average duration of the fully complementary hybrid formed by a specific multi-part primer is longer than the average duration of the mismatched (shorter) hybrid formed by the multi-part primer of the same type. Thus, a more stringent multipartite primer design (shorter foot sequence, longer bubble) produces shorter-lived hybrids that are considerably less likely to form stable hybrids with DNA polymerase molecules. As a result, shorter foot hybrids are not only less abundant, they also have a reduced chance of forming stable complexes with DNA polymerase molecules, but this otherwise discriminating property accounts for the extraordinary selectivity of multipartite primers.

[0129] As reported in Example 7, the inventors also investigated the effect of varying the position of the query nucleotide in the foot sequence of the multipartite primers according to the invention. The inventors used a series of six primers: 24-14-6: 1:0, 24-14-5: 1:1, 24-14-4: 1:2, 24-14-3: 1:3, 24-14-2: 1:4 and 24-14-1: 1:5. The inventors kept the length of the anchor sequence, the length of the bridge sequence and the length of the foot sequence (seven nucleotides) and varied only the position of the query nucleotide within the foot sequence. Using 10 6 copies of the desired target (mutant) and 10 6 The real-time fluorescence results obtained for each of these primers are shown in Fig.12 and the calculated C T The values ​​are summarized in Table 3. The results show that as the position of the query nucleotide gets closer to the 3' end of the foot sequence, the discrimination window (ΔC T These results indicate that the preferred positions for the query nucleotide are the 3' end of the foot sequence (enabling ARMS discrimination) and the 3' penultimate nucleotide of the foot sequence (preventing two base pair formations instead of just one base pair formation).

[0130] As reported in Example 8, the inventors also studied the shape of the bubble formed between the bridge sequence of the multi-part primers according to the invention and the insertion sequence in the expected and unexpected target sequence. The inventors varied the "shape of the bubble" by selecting the relative lengths of these two sequences. In carrying out the assay, the inventors used a series of primers with an anchor sequence of 24 nucleotides in length and a 5:1:1 foot sequence. The inventors kept the bubble circumference at 32 nucleotides, but the inventors varied the length of the bridge sequence and the length of the insertion sequence (by varying the anchor sequence so that once it hybridizes to the template molecule, the insertion sequence will have the desired length). In addition to testing multi-part primers that form symmetrical bubbles (i.e., primers with a bridge sequence of 14 nucleotides and an anchor sequence that makes the insertion sequence 14 nucleotides long (14 / 14 bubble)), the inventors also tested multi-part primers that produce asymmetric bubbles, bubbles with relatively longer bridge sequences (18 / 10 bubble and 16 / 12 bubble) and with relatively shorter bridge sequences (12 / 16 bubble and 10 / 18 bubble). Using 10 6 copies of the desired target (mutant) and 10 6 The real-time fluorescence results obtained for each of these primers are shown in Fig.13 and the calculated C TThe values ​​are summarized in Table 4. The results show that the symmetrical 14 / 14 bubble enables a discrimination window (ΔC T ) is the largest, but only slightly so. As a result, the inventors' most preferred bubbles are symmetrical.

[0131] Example 9 reports an experiment using the assay of Example 4 for a different target B-raf mutation V600E (instead of the EGFR mutation L858R) and a 24-14-5:1:1 multipartite primer for this mutation. Fig.14 It is C T vs. the logarithm of the expected starting amount of target template. Fig.14 As seen in the assay, the assay provides 6 WT template samples and in 10 6 In the presence of WT template, there are 10 6 The ΔC of 23.1 cycles between samples of MUT template T , which is even greater than the corresponding ΔC obtained in Example 4 T .

[0132] Example 10 reports another variation, this time exploiting the EGFR mutation T790M and PCR amplification using genomic DNA with up to 10,000 copies of wild-type target template and 24-14-4:1:1 multipartite primers. Fig.15 It is C T vs. the logarithm of the expected starting amount of mutant target template. Fig.15 As seen in the assay, the assay provides 4 WT template samples and in 10 4 In the presence of WT template, there are 10 4 The ΔC of 12.6 cycles between samples of MUT template T .

[0133] Example 11 reports an assay similar to that of EGFR mutation L858R in Example 4, using a different spectrofluorometric thermocycler, ABI PRISM 7700, the same 24-14-5:1:1 multipart primers and plasmid DNA, except this time the template was not digested. Fig.16 It is C T vs. the logarithm of the expected starting amount of target template. Fig.16 As seen in the assay, the assay provides 6 WT template samples and in 10 6 In the presence of WT template, there are 10 6 The ΔC of 16.4 cycles between samples of MUT templateT .

[0134] Fig.17 The results of the experiment described in Example 12 are shown. This experiment was designed to demonstrate the relative contribution of thermodynamic considerations versus enzymatic (ARMS-type) considerations in determining the selectivity of the multipart primers described herein. What the inventors did was to repeat the assay of Example 3 using not only the 24-14-5:1:1 primer but also the truncated 24-14-5:0:0 primer (which omitted the 3' penultimate and terminal nucleotides). Thus, the foot sequence of the latter primer was fully complementary to both the expected and unintended target sequences. Fig.17 Figure A compares the amplification of 1,000,000 expected target sequences and 1,000,000 unintended target sequences using the 24-14-5:1:1 multipartite primer (whose foot / target hybrid is destabilized at the 3' end) to distinguish between these two types of templates, as accomplished by ARMS and thermodynamics. T The values ​​were 23.1 for the expected target sequence (curve 1701) and 40.7 for the unintended target sequence (curve 1702), yielding a ΔC of 17.6 cycles. T .

[0135] Fig.17 Panel B of FIG. 5 compares the amplification of 1,000,000 expected target sequences and 1,000,000 unintended target sequences using the 24-14-5:0:0 multipartite primer (whose foot / target hybrid is not destabilized at the 3' end). T The values ​​were 39.7 for the expected target sequence and 39.4 for the unintended target sequence, yielding a ΔC of -0.3 cycles. T .

[0136] Like the truncated primer 24-14-5:0:0, the multipartite primer 24-14-5:1:1 forms a foot hybrid with the same five nucleotides in the wild-type template (curve 1702) because the query nucleotide of this primer is not complementary to the single nucleotide polymorphism and the resulting mismatched base pair at the penultimate position of the foot sequence prevents the adjacent 3' terminal nucleotide of the foot sequence of this primer from forming a separate base pair. However, there is a difference between the hybrid formed by primer 24-14-5:0:0 and the wild-type template and the hybrid formed by primer 24-14-5:1:1 and the wild-type template, and the difference is that the foot sequence in the hybrid formed by primer 24-14-5:1:1 and the wild-type template has two overhanging nucleotides caused by the 3' penultimate mismatch and is therefore subject to ARMS-type discrimination by the DNA polymerase, while the truncated foot sequence in the hybrid formed by primer 24-14-5:0:0 and the wild-type template does not have any overhanging 3' terminal base pair and is therefore not subject to ARMS-type discrimination by the DNA polymerase. If ARMS-type discrimination plays an important role in selectivity when using the multi-part primer according to the present invention, the inventors would have expected that the C-terminal sequence of the reaction involving primer 24-14-5:0:0 and the wild-type template would be significantly different from that of the hybrid formed by primer 24-14-5:0:0 and the wild-type template. T The C value (curve 1704) would have been higher than the C value for the reaction involving primers 24-14-5:1:1 and wild-type template. T The value (curve 1702) is lower (i.e., less delayed) because ARMS-type discrimination cannot play a role in the reaction involving primer 24-14-5:0:0 and wild-type template, but can play a role in the reaction involving primer 24-14-5:1:1 and wild-type template. These results indicate that when using multi-part primers according to the present invention, the effect of ARMS-type discrimination is absent or significantly reduced (perhaps as a result of the extremely short average duration of the foot hybrids formed by these highly selective nucleic acid amplification primers).

[0137] Assays according to the present invention may include screening assays that look for the presence of any rare target when one of a plurality of possible rare targets may be present. For such assays, multi-part primers are used for each possible rare target, but the detection does not require identification of which target is present. Thus, SYBR Green dye may be used as a detection reagent, as a doubly labeled hybridization probe capable of indiscriminately signaling, as a 5' functional sequence on a primer capable of indiscriminately signaling. Assays using multi-part primers according to the present invention include amplification and detection, which may include simultaneous quantification of two or more rare target sequences in a single reaction tube, reaction well, or other reaction vessel, where identification of which target or targets is required. Amplification and detection in a single reaction tube of two or more rare target sequences that have no sequence homology and are located at different locations in the genome (e.g., simultaneous detection of rare single nucleotide polymorphisms located in different genes) may include specific, uniquely colored hybridization probes for each different expected target sequence, such as molecular beacon probes, A probe or 5' ribozyme that hybridizes to a unique sequence in either strand of the amplification product downstream of the multipart primer This applies not only to free-floating detection probes, but also to tethered probes, such as Figure 4 Alternatively, the multi-part primers for each different target sequence may include a labeled hairpin, such as Figure 4 Hairpin 404 in. By reference Figure 4 , two or more different multi-part primers 103, each specific for a different rare expected target sequence and each labeled with a uniquely colored fluorescent label 408, 413 or 416, can be used to simultaneously identify and quantify each expected target sequence present in an individual sample.

[0138] 5. Multiplex Assays

[0139] A particularly attractive feature of the super selective primers of the present invention is their potential use in multiplexed assays that simultaneously measure the abundance of different rare mutant sequences in the same clinical sample. The results of these assays can provide patient-specific information for tailoring therapy for each individual.

[0140] One interesting multiplex labeling strategy is based on the realization that, since there is no relationship between the bridge sequence and the expected target sequence, the assay designer is free to choose a distinct bridge sequence for each of the different super-selective primers that are simultaneously present in the multiplex assay. Since the entire sequence of each primer becomes an integral part of the amplicon generated when that primer binds to its mutant target, the distinct nucleic acid sequence of the bridge segment can serve as a "sequence number" within that amplicon that identifies the mutant target from which it was generated.

[0141] These identification bridge sequences can be relatively long (e.g., 20 nucleotides long to ensure their uniqueness), and primers can be designed to form short insert sequences in the template accordingly. In order to simultaneously detect and quantify different mutant target sequences present in clinical samples, a set of specific molecular beacon probes (Tyagi et al., (1996) Nat. Biotechnol. 14, 303-308, Tyagi et al., (1998) Nat. Biotechnol., 16, 49-53 and Bonnet et al., (1999) Proc. Natl. Acad. Sci. USA, 96, 6171-6176) can be included in a real-time, gene amplification reaction, each specific for the complement of a different bridge sequence of one of the super selective primers, and each labeled with a different color fluorophore.

[0142] In these reactions, the inventors prefer that the concentration of the super-selective normal primer should be limited and the linear reverse primer should be present in excess, thereby ensuring that the reaction will not be symmetrical and that the molecular beacon can bind substantially all of the target amplicon synthesized in excess without significant competition with the less abundant complementary amplicon (Pierce et al., (2005) Proc. Natl. Acad. Sci. USA, 102, 8609-8614). These multiplex assays can even distinguish between different mutations present in the same codon, because super-selective primers designed to detect specific mutations will distinguish adjacent or alternative mutations in the same way as they distinguish wild-type target sequences.

[0143] Another multi-strategy is shown in Fig.18 , which is a schematic diagram of two multipart primers according to the present invention, which can be used in a multiplex reaction against two closely related expected target sequences.

[0144] As long as there is sequence homology between the expected target sequences of the multiplex assay, unique sequences can be introduced by using a unique bridge sequence for each different expected target sequence. Figure 2As explained, the reverse primer copies the entire forward (multipart) primer into the reverse product strand, so that, in subsequent amplification cycles, the entire multipart primer (anchor sequence, bridge sequence and foot sequence) is complementary to the product resulting from extension of the reverse primer. In a multiplex assay, it is important that only one multipart primer (the "correct" primer thus copied) hybridizes and initiates the reverse product strand. It will therefore be appreciated that the bridge sequence of the "correct" multipart primer must be made sufficiently different to prevent another multipart primer from priming the reverse product strand (so-called "cross-hybridization"). In this way, a specific, uniquely colored hybridization probe (either free floating or tethered to a primer) (which targets the complement of the bridge sequence) will signal the amplification of the only intended target and will not emit false signals by hybridizing to the multipart primer itself. Similarly, only a hairpin tail ( Figure 4 The "correct" multi-part primer (hairpin 405 in) will hybridize to the reverse product strand and emit a signal.

[0145] For distinguishing and quantifying the presence of rare target sequences that are nearly identical (differing from each other by only one or two SNPs) and that are very close to each other in the genome (e.g., the presence of medically significant variants of the human K-ras gene, where different SNPs may be present within codon 12, each specifying the identity of a different amino acid in the encoded protein of the gene)), a method with Fig.18 or Fig.19 Two or more multipart primers with the structures outlined in Fig.18, the top structure 103A shows a multi-part primer whose foot sequence 106A is fully complementary to a specific expected rare target sequence, including the nucleotide corresponding to the complementary nucleotide "g" in the target sequence (the query nucleotide). The lower structure 103B shows a multi-part primer whose foot sequence 106B is fully complementary to a different specific rare expected target sequence, which is a variant of the target of foot sequence 106A and which is located at (or very close to) the location in the genome of the expected target sequence for foot sequence 106A. In foot sequence 106B, nucleotide "h" is the query nucleotide, which is fully complementary to the corresponding nucleotide in the expected target sequence of foot sequence 106B. In order to be able to simultaneously distinguish, or distinguish and quantify the abundance of each of these rare target sequences in the same reaction, primer 103A can be linked to a unique structure 404A that differs in sequence 405A and 406A and fluorophore label 408A from the sequences 405B and 406B and fluorophore label 408B in structure 404B of primer 103B. When two or more multi-part primers (such as primers 103A and 103B) are used to simultaneously distinguish and quantify similar expected rare target sequences at the same (or very similar) position, it is often the case that their respective anchor sequences will be the same or very similar (so that the primers bind close to the expected position where the variant sequence to be distinguished is present). However, because there is no correlation between the bridge sequence of the multi-part primers of the present invention and their expected target sequences, the bridge sequence 105A in primer 103A can be selected so that its nucleotide sequence is different from the bridge sequence 105B in primer 103B. This is how two or more multipartite primers of the invention can be utilized to simultaneously distinguish and quantify rare expected target sequences that are alleles of each other and are located at the same (or very similar) position in the genome:

[0146] Reverse primer 203 ( Figure 2 ) continues through labeled structures 404A and 404B, separating quencher 407 from fluorophore labels 408A and 408B, respectively. As a result, when they are incorporated into amplicons, each of primers 103A and 103B will fluoresce in its unique identification color if their fluorescence intensity is measured in real time at the end of each chain extension cycle (in an amplification reaction in which the amplicon becomes double-stranded, such as in PCR amplification). Alternatively, when their fluorescence intensity is measured at the end of the annealing phase of the amplification reaction, each of primers 103A and 103B will fluoresce in its unique identification color because each primer (103A or 103B) has a fully complementary sequence in those amplicon strands 204 ( Figure 2 ), their quencher groups 407 become separated from their fluorophore labels (408A or 408B), and the synthesis of the amplicon strand 204 is initiated by incorporation of the same primer.

[0147] Fig.19 Depicted are primers and probes for a similar assay using free-floating molecular beacon probes instead of labeled hairpin tails. Fig.19 In the embodiment, multi-part primer 1903A has a foot sequence 1906A, which is completely complementary to a specific first expected rare target sequence, including the query nucleotide "r". Multi-part primer 1903B has a foot sequence 1906B, which is completely complementary to a different specific second rare target sequence, which is a variant of the target of foot sequence 1906A and is located at (or very close to) the position in the genome of the expected target sequence of foot sequence 1906A. In foot sequence 1906B, nucleotide "s" is a query nucleotide. In this embodiment, the query nucleotide "r" is not complementary to the second rare target or wild-type sequence. And the query nucleotide "s" is not complementary to the first rare target sequence or wild-type sequence. In order to be able to distinguish the amplification of two rare target sequences in the same reaction, and to prevent cross-hybridization, the sequence of bridge sequence 1905A is completely different from the sequence of bridge sequence 1905B. Molecular beacon probe 1907A, including loop 1908A, stem 1909A, fluorophore 1910A, and quencher 1911A, has a loop specific for the complement of bridge sequence 1905A. Molecular beacon probe 1907B, including loop 1908B, stem 1909B, fluorophore 1910B, and quencher 1911B, has a loop specific for the complement of bridge sequence 1905B. Fluorophores 1911A and 1911B are different colors. Detection using probes 1907A and 1907B can be real-time or end-point.

[0148] The key feature that enables the simultaneous real-time measurement of different amplicons produced by different rare expected allele target sequences is that the multi-part primers of the present invention can be designed to have completely different sequences in the hairpin tails (e.g., 404A and 404B) of their labels and in their bridge sequences (e.g., 105A and 105B). As a result, the annealing conditions can be adjusted to ensure that each type of primer is only combined with the amplicon initiated by the same type of primer for its synthesis. In addition, if a specific type of primer is to be combined with a non-homologous amplicon, the hairpin that sends the signal at the end of the primer will not be complementary to the sequence at the 3' end of the amplicon, so no fluorescence is emitted. As an alternative to simply utilizing different bridge sequences for each multi-part primer that will be present in the reaction at the same time, different anchor sequences can be utilized by shortening one or sliding along the target. Alternatively, the different lengths of the bridge sequence (e.g., 105A and 105B) will be able to use different anchor sequences (e.g., 104A and 104B) without significantly changing the selectivity of each primer. This reduces the probability of forming mismatched hybrids between primer 103A and an amplicon containing a nonhomologous priming sequence for primer 103B, and reduces the probability of forming mismatched hybrids between primer 103B and an amplicon containing a nonhomologous priming sequence for primer 103B.

[0149] 6. Additional considerations for designing multipart primers

[0150] The design of multi-part primers according to the present invention is simple. The inventors suggest that the design should be targeted to a specific amplification protocol on a specific instrument, since instruments vary specifically in their detection and expression of fluorescence. The appropriate procedure is to select the design (length of the anchor sequence, length of the bridge sequence, and length of the foot sequence, and the query nucleotide located at the 3' terminal nucleotide of the foot sequence or at the second to last nucleotide at the 3' end). Then, by simply changing the length of the bridge sequence and the length of the foot sequence, the primer design can be optimized with a small number of attempts to achieve the desired large ΔC between samples containing the expected target and samples containing the unexpected target. T This involves making the primers ineffective for amplifying the intended target sequence. The design considerations are those discussed above with respect to the Examples. In particular, shortening the foot sequence and increasing the size of the bubble formed by the bridge sequence and the target insert sequence increases the C of the intended target. T delay and increase the ΔC between samples containing the expected target and samples containing the unintended target T .

[0151] There are additional considerations in the design of the multi-part primers of the present invention. The primers must not trigger other sequences that are or may be present in the sample. Conventional computer methods to prevent this are well known and readily available.

[0152] a. Anchor sequence

[0153] The anchor sequence is usually (but not necessarily) completely complementary to the template sequence, and it can usually be located about 14 nucleotides from the 5' end of the foot sequence and can usually be 15-40, 15-30 or 20 to 30 (such as 20 to 24) nucleotides long. Its length is selected so that the melting temperature of its hybrid with the template is in a suitable range, such as 66°C to 72°C in several embodiments.

[0154] If the anchor sequence in a multipartite primer designed to discriminate a particular polymorphism turns out to be not specific enough because its target sequence is present elsewhere in the genome, this problem can be solved by designing a multipartite primer that discriminates the same polymorphism but binds to the complementary target strand.

[0155] b. Bridge sequence

[0156] As for the bridge sequence, the inventors propose to check and, if necessary, eliminate transient hybridization events that may occur if the sequence can form a low Tm hybrid with the target, thereby reducing its effective length. In addition, the effect of the bridge sequence can be modified by adjusting the rigidity of the bridge sequence, since different nucleotide sequences have different rigidities to a certain extent. See Goddard et al. (2000) Phys. Rev. Lett. 85: 2400-2403.

[0157] In one embodiment, the bridge sequence may be about at least 6 (e.g., 7, 8, 9, 10, 11, 12, 13, 14, 15, or 20) nucleotides long. Its nucleotide sequence may be selected to ensure that under annealing conditions: (i) it does not hybridize with the corresponding "insertion sequence" in the template strand (which is located between the foot target sequence and the anchor target sequence); (ii) it does not hybridize with any sequence in the human genome; (iii) it does not form any secondary structure that effectively shortens its length under the assay conditions; and (iv) it does not hybridize with the conventional reverse primer used to initiate the synthesis of the complementary template strand. In addition, if the insertion sequence in the template strand forms a secondary structure that effectively shortens its length under the assay conditions, the length of the bridge sequence may be increased and the length of the insertion sequence may be reduced by a corresponding number of nucleotides (by selecting an anchor target sequence that is closer to the foot target sequence with the same number of nucleotides).

[0158] The recognition that the bridge sequence can be chosen to be relatively short or relatively long, and that the probe designer can select an arbitrary sequence for the bridge segment, opens up more functional possibilities for the design of the super selective primers of the invention.

[0159] For example, if the sequence of the putative insert sequence naturally present in the template is such that it is likely to form a secondary structure under the assay conditions, the primers can be designed so that a relatively small insert sequence is established in the primer-template hybrid, thereby disrupting the formation of secondary structure, and the bridge sequence of the primer can be selected to have a relatively longer length, thereby maintaining the selectivity of the assay (see the results shown in Table 4). In addition, primer function can be finely tuned by selecting the sequence of the bridge sequence that takes into account the differences in the flexibility of the insert sequence and the bridge sequence.

[0160] In addition, the selection of appropriate bridge sequences for super selective primers significantly suppresses the occurrence of false positive amplicons (such as primer dimers). Unlike the design of conventional linear primers (whose sequence is determined by the template to which it binds), any sequence can be used for the bridge segment. The inventors carefully selected the bridge sequence: (i) does not form a secondary structure; (ii) is unrelated to the sequence of the template, the sequence of the genomic DNA, and the sequence of the conventional reverse primer; and (iii) when introduced into the full-length primer, it cannot cause the primer to self-hybridize.

[0161] c. The role of the bubble formed by the bridge sequence and the insertion sequence

[0162] Within the acceptable range described above, the greater the perimeter of the bubble formed by hybridization of the superselective primer to the original template molecule, the greater the inhibition of wild-type amplicon synthesis relative to the inhibition of mutant amplicon synthesis (see, e.g. Fig.11 ). From a thermodynamic point of view, larger bubbles should reduce the equilibrium abundance of both wild-type hybrids and mutant hybrids, but should not change their relative abundance. However, from a kinetic point of view, it is appropriate to consider the force of the bubble that impacts the foot hybrid connected to the target hybrid, because the bubble is subjected to the random Brownian motion of water molecules in the reaction mixture. This establishes a force with the potential to separate the foot hybrid. The greater the perimeter of the bubble, the greater this potential destructive force. In addition, mismatched wild-type hybrids (which are weaker than fully complementary mutant hybrids) are more likely to be separated.

[0163] Therefore, mismatched wild-type hybrids not only exist for a shorter time due to their lower stability, but they are also more easily broken apart by the random forces of the collision bubble. Therefore, the inventors believe that the extraordinary selectivity of the super-selective primers originates from thermodynamic factors that affect the stability of the hybrids and from kinetic factors that affect the average duration of the resulting hybrids.

[0164] d. Foot sequence

[0165] The foot sequence is located at the 3' end of the primer; it is complementary to a region of the template strand where there is at least one nucleotide difference between the expected target sequence and its closely related unintended target sequence, such as a localized single nucleotide polymorphism; and it is typically seven nucleotides long. The "query nucleotide" in the foot sequence can be located at the penultimate position of the 3' end of the foot sequence, or at the 3' end of the foot sequence. The length of the foot sequence can be varied to improve selectivity. The foot sequence can be shorter (six or even five nucleotides long), especially if it has a high GC content. If the query nucleotide would form a G:T base pair with the wild-type template strand, conversely, it is desirable to design the primer so that it binds to the complementary template strand.

[0166] If the foot sequence hybridizes to the target sequence, and if the DNA polymerase is able to form a functional complex with the hybrid before the hybrid is separated, then extension of the foot sequence can be catalyzed by the DNA polymerase to produce an amplicon. It should be understood that short foot sequences (e.g., 6 or 7 nucleotides in length) are generally so short that they are complementary to sequences present at a large number of different positions within the nucleic acid that may be present in the test sample (e.g., in genomic DNA from human cells). However, the foot sequence is so short, and as a result has a melting temperature, Tm, that is so extremely low under conditions used for amplification (e.g., conditions used for PCR assays) that the foot sequence will not form a hybrid with any completely complementary sequence in the test nucleic acid sample unless the anchor sequence of the primer has first hybridized to a position within the test nucleic acid that is only a few nucleotides away from the desired target sequence.

[0167] Once designed in the manner disclosed herein, primer sequences can be checked with the aid of any suitable computer program, such as the OligoAnalyzer computer program (Integrated DNA Technologies, Coralville, IA), to ensure that they are unlikely to form internal hairpin structures or self-dimers under the assay conditions and to ensure that they do not form heterodimers with conventional reverse primers.

[0168] 7. Test kit

[0169] The present invention further includes a kit containing reagents for carrying out the above-mentioned amplification method (including amplification and detection methods). For this purpose, one or more reaction components for the methods disclosed herein can be provided in the form of a kit for detecting target nucleic acids. In such a kit, one or more reaction components of a suitable amount are provided in one or more containers or maintained on a substrate (e.g., by electrostatic interaction or covalent bonding).

[0170] The kits described herein include one or more of the above-mentioned primers. The kits may include one or more containers containing one or more primers of the present invention. The kits may contain a single primer in a single container, multiple containers containing the same primer, a single container containing two or more different primers of the present invention, or multiple containers containing different primers or a mixture of two or more primers. Any combination and arrangement of primers and containers are encompassed by the kits of the present invention.

[0171] The test kit also contains other materials for implementing the above method. In some embodiments, the test kit contains these reagents, some or all of the materials for the method carried out using the primer of the present invention. Therefore, the test kit can include some or all of the reagents using the primer of the present invention to carry out PCR reaction. Some or all of the components of the test kit are provided in the container in the form of being separated from the container containing the primer of the present invention. The example of the other components of the test kit includes but is not limited to one or more different polymerases, specific for control nucleic acid or for one or more primers of target nucleic acid, specific for control nucleic acid or for one or more probes of target nucleic acid, buffer for polymerization reaction (with 1X or concentrated form) and one or more dyes or fluorescent molecules for detecting polymerized products. The test kit can also include one or more of the following components: support, termination reagent, modification reagent or digestion reagent, permeation agent and the device for detecting detection probe.

[0172] The reaction components used in the amplification and / or detection process are provided in various forms. For example, the components (e.g., enzymes, nucleotide triphosphates, probes and / or primers) can be suspended in an aqueous solution or as freeze-dried or lyophilized powders, pellets or beads. In the latter case, when reconstructed, the components form a complete mixture of the components to be measured.

[0173] Test kit or system can be sufficient for carrying out at least one measured amount containing any combination of components described herein, and can further include instructions for using the component recorded in a tangible form. In some applications, the amount of single use that one or more reaction components can be measured in advance is provided in a single (typically disposable) tube or equivalent container. With such an arrangement, the sample waiting for the detection of the presence of target nucleic acid can be added to the single tube and directly amplified. The amount of the component provided in the test kit can be any suitable amount, and can depend on the target market involved in the product. The general guide for determining the appropriate amount can be found in, for example, Joseph Sambrook and David W.Russell, Molecular Cloning: A Laboratory Manual, 3rd edition, Cold Spring Harbor Laboratory Press, 2001 and Frederick M.Ausubel, Current Protocols in Molecular Biology, John Wiley & Sons, 2003.

[0174] The kit of the present invention may contain any number of additional reagents or substances useful for practicing the method of the present invention. Such substances include, but are not limited to, reagents (including buffers) for lysing cells, divalent cation chelators or other reagents for inhibiting unwanted nucleic acids, control DNA for ensuring that the primers, polymerases or other reaction components function properly, DNA fragmentation reagents (including buffers), amplification reaction reagents (including buffers) and washing solutions. The kit of the present invention can be provided at any temperature. For example, a kit for storing a protein component or a complex thereof in a liquid is preferably provided and maintained below 0°C, preferably at or below -20°C, or in other frozen states.

[0175] The container in which the component is provided can be any conventional container that can accommodate the provided form, such as a microcentrifuge tube, an ampoule, a bottle, or an integrated test device, such as a microfluidic device, a cartridge, a lateral flow, or other similar device. The kit can include labeled or unlabeled nucleic acid probes for amplification or detection of target nucleic acids. In some embodiments, the kit can further include instructions for use of the components in any method described herein (e.g., a method using a crude matrix without nucleic acid extraction and / or purification).

[0176] Test kits can also include packaging materials for holding a combination of containers or containers. Typical packaging materials for such test kits and systems include solid substrates (e.g., glass, plastic, paper, foil, microparticles, etc.) that hold reaction components or detection probes (e.g., in vials, microtiter plate wells, microarrays, etc.) in any of a variety of configurations.

[0177] 8. Additional Definitions

[0178] As used herein, the term "target nucleic acid" or "target sequence" refers to a nucleic acid containing a target nucleic acid sequence. The target nucleic acid can be single-stranded or double-stranded, and is often DNA, RNA, a derivative of DNA or RNA, or a combination thereof. "Target nucleic acid sequence", "target sequence" or "target region" refers to a specific sequence that contains all or part of a sequence of a single-stranded nucleic acid. The target sequence can be located within a nucleic acid template, which can be a single-stranded or double-stranded nucleic acid in any form. The template can be a purified or separated nucleic acid, or can be non-purified or non-separated.

[0179] As used herein, the term "amplification" and its variants include any process for producing multiple copies or complements of at least a portion of a polynucleotide, which is typically referred to as a "template". The template polynucleotide may be single-stranded or double-stranded. The amplification of a given template can result in the production of a group of polynucleotide amplification products, collectively referred to as "amplicons". The polynucleotides of the amplicon may be single-stranded or double-stranded, or a mixture of the two. Typically, the template will include a target sequence, and the resulting amplicon will include a polynucleotide having a sequence substantially identical to or substantially complementary to the target sequence. In some embodiments, the polynucleotides of a particular amplicon are substantially identical to each other or substantially complementary to each other; alternatively, in some embodiments, the polynucleotides within a given amplicon may have mutually varying nucleotide sequences. Amplification may be performed in a linear or exponential form, and may involve repeated and continuous replication of a given template to form two or more amplification products. Some typical amplification reactions include continuous and repeated cycles of template-based nucleic acid synthesis, thereby generating at least some portions of the nucleotide sequence containing the template and the formation of multiple sub-polynucleotides that share at least a certain degree of nucleotide sequence identity (or complementarity) with the template. In some embodiments, each instance of nucleic acid synthesis, which may be referred to as a "cycle" of amplification, includes establishing a free 3' end (e.g., by nicking of one strand of dsDNA) to produce a primer and a primer extension step; optionally, an additional denaturation step may also be included, in which the template is partially or completely denatured. In some embodiments, a round of amplification includes a given number of repetitions of a single cycle of amplification. For example, a round of amplification may include 5, 10, 15, 20, 25, 30, 35, 40, 50 or more repetitions of a particular cycle. In an exemplary embodiment, amplification includes any reaction in which a particular polynucleotide template is subjected to two consecutive cycles of nucleic acid synthesis. Synthesis may include template-dependent nucleic acid synthesis.

[0180] The term "primer" or "primer oligonucleotide" is a nucleic acid chain or oligonucleotide that can hybridize with a template nucleic acid and serve as a starting point for nucleic acid synthesis to incorporate extension nucleotides according to the composition of the template nucleic acid. "Extension nucleotides" refer to any nucleotide that can be incorporated into an extension product during amplification, i.e., DNA, RNA, or a derivative of DNA or RNA, which may include a label.

[0181] "Hybridization" or "hybridize" or "annealing" refers to the ability of fully or partially complementary nucleic acid chains to come together in parallel or preferably antiparallel orientation under specific hybridization conditions (e.g., under stringent hybridization conditions) to form a stable double-stranded structure or region (sometimes referred to as a "hybrid"), in which the two component chains are bound by hydrogen bonds. Although hydrogen bonds are typically formed between adenine and thymine or uracil (A and T or U) or cytosine and guanine (C and G), other base pairs can also be formed (e.g., Adams et al., The Biochemistry of the Nucleic Acids, 11th ed., 1992).

[0182] The term "stringent hybridization conditions" or "stringent conditions" refers to conditions where a probe or oligomer specifically hybridizes to its intended target nucleic acid sequence and does not hybridize to another sequence. Stringent conditions can be varied based on well-known factors such as GC content and sequence length, and can be empirically predicted or determined using standard methods known to those of ordinary skill in the art of molecular biology (e.g., Sambrook, J. et al., 1989, Molecular Cloning, A Laboratory Manual, 2nd ed., Ch. 11, pp. 11.47-11.57, (Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY)).

[0183] As disclosed herein, a range of many values ​​is provided. It should be understood that each intermediate value (to one tenth of the unit of the lower limit, unless the context clearly indicates otherwise) between the upper and lower limits of the range is also specifically disclosed. Each smaller range between any of the values ​​or intermediate values ​​in the range and any other values ​​or intermediate values ​​in the range is included in the present invention. The upper and lower limits of these smaller ranges may be independently included or excluded in the range, and any, neither, or both limits are included in each range of the smaller range is also included in the present invention, subject to any specifically excluded restrictions in the range. When the range includes one or both of the limits, the range excluding either or both of these included limits is also included in the present invention.

[0184] The term "about" generally refers to the number shown plus or minus 10%. For example, "about 10%" can mean a range of 9% to 11%, and "about 1" can mean from 0.9-1.1. Other meanings of "about" are obvious from the context, such as rounding, so, for example, "about 1" can also mean from 0.5 to 1.4.

[0185] Example

[0186] Example 1: EGFR mutation L858R and conventional linear primers

[0187] Two PCR amplifications and detection assays were performed using plasmid DNA containing EGFR mutation L858R or plasmid DNA containing the corresponding wild-type sequence (which differed in polynucleotide polymorphisms from each other) as templates. Conventional forward and reverse primers were used to produce a 49-nucleotide long double-stranded amplification product. The forward primer (FP) is a conventional primer that contains a query nucleotide close to the middle of the primer sequence. The reverse primer (RP) is a conventional primer that is fully complementary to both target sequences. The primer sequence and the expected target sequence with a mutant allele (MUT) are as follows:

[0188]

[0189] RP:5'-GCATGGTATTCTTTCTCTTCCGCA-3'(SEQ ID No.3)

[0190] In the forward primer sequence, nucleotides that are complementary to the mutant target template but mismatched with the wild-type template are bold, underlined, and larger. In the mutant target sequence, the binding site of the forward primer is underlined, and the sequence of the reverse primer is underlined. In addition, in the mutant target sequence, nucleotides specific for the mutant are bold, underlined, and larger. The SciTools program for calculating the melting temperature of DNA hybrids from Integrated DNA Technologies (specific parameters: [oligo] = 0.06 μM; [Na + ]=60mM; [Mg 2+ ]=3mM; [dNTPs]=0.25mM), the calculated Tm of the forward primer binding to the mutant allele was 67.5℃, and the calculated Tm of the reverse primer was 64.0℃.

[0191] By inserting the EGFR gene fragment (containing EGFR L858R sudden change or corresponding EGFR wild-type sequence) of 115 base pairs into pGEM-11Zf (+) carrier (Promega) to prepare plasmid. Use restriction endonuclease Mse I (New England Biolabs) to digest mutant and wild-type plasmid DNA. Digestion mixture contains 10 units of Mse I and 4 μ g of mutant or wild-type genomic DNA in 20 μ l volume (it contains the dithiothreitol of 5mM KAc, 2mMTris-Ac (pH 7.9), 1mM MgAc, 1% bovine serum albumin and 100 μ M). Incubate reaction 120min at 37 ℃, then hatch 20min at 65 ℃ with enzyme inactivation.

[0192] The mixture was stirred at 4 °C in a 30 μl volume containing 50 mM KCl, 10 mM TrisHCl (pH 8.0), 3 mM MgCl2, 1.5 units of AmpliTaq Gold DNA polymerase (Life Technologies), 250 μM each of the four deoxyribonucleoside triphosphates (dNTPs), 60 nM of each primer, and 1× Green (Life Technologies) was used for PCR amplification. In this series, the reaction mixture contained 10 6 copies of mutant template (MUT) or 10 6 The amplification was performed in a Bio-Rad IQ5 fluorescence spectrophotometer using 0.2 ml polypropylene PCR tubes (white). The thermal cycle configuration was 95°C for 10 min, followed by 60 cycles of 94°C for 15 sec, 60°C for 15 sec, and 72°C for 20 sec. The DNA was measured at the end of each chain extension phase (72°C). Green fluorescence intensity.

[0193] Real-time fluorescence results (i.e., SYBR Fluorescence intensity) is shown in Figure 5 In which curve 501 contains 10 6 The response of the MUT template and curve 502 are 10 6 The assay instrument automatically calculates the threshold cycle (C T ). These values ​​are 20.0 (curve 501) and 19.7 (curve 502). In the upper left corner of the figure is a schematic diagram of a conventional forward primer (straight line) with a query nucleotide (circle) in the middle.

[0194] Example 2: EGFR mutation L858R and conventional linear primer with 3' terminal query nucleotide

[0195] PCR amplification and detection assays were performed using mutant (MUT) and wild-type (WT) templates as described in Example 1. In this experiment, the forward primer was an "ARMS primer", i.e., a primer that was fully complementary to the mutant template but had a 3' terminal mismatch with the WT template, i.e., had a query nucleotide at the 3' end of the priming sequence. The inventors used the same reverse primer as in Example 1. The primer sequences and the expected target sequence with the mutant allele (MUT) are as follows:

[0196]

[0197] RP:5'-GCATGGTATTCTTTCTCTTCCGCA-3'(SEQ ID No.3)

[0198] In the forward primer sequence, nucleotides that are complementary to the mutant target template but mismatched with the wild-type template are bold, underlined, and larger. In the mutant target sequence, the binding site of the forward primer is underlined, and the sequence of the reverse primer is underlined. In addition, in the mutant target sequence, nucleotides specific for the mutant are bold, underlined, and larger. The SciTools program for calculating the melting temperature of DNA hybrids using Integrated DNA Technologies (specific parameters: [oligo] = 0.06 μM; [Na + ]=60mM; [Mg 2+ ]=3mM; [dNTPs]=0.25mM), the calculated Tm of the forward primer binding to the mutant allele was 60.7℃, and the calculated Tm of the reverse primer was 64.0℃.

[0199] PCR amplification was performed as described in Example 1. Real-time fluorescence results (i.e., SYBR Fluorescence intensity) is shown in Figure 6 In Figure A, curve 601 is based on 10 6 The reaction started with 10 MUT templates, and curve 602 is 6 The assay instrument automatically calculated the cycle threshold (C T ). Those values ​​are 19.4 (curve 601) and 30.4 (curve 602), giving a ΔC for 11 cycles. T In the upper left corner of the figure is a representative schematic of a conventional forward primer (line) with a query nucleotide (circle) located at the 3' end of the primer.

[0200] The above experiment was repeated with a forward primer having the query nucleotide at the penultimate position at its 3' end (the inventors added a G to the 3' end of the primer and removed the C at the 5' end to maintain the primer length). The resulting sequence of the forward primer is:

[0201]

[0202] Using the SciTools program from Integrated DNA Technologies and the same reaction conditions as above, the calculated Tm of the forward primer binding to the mutant allele was 61.9°C.

[0203] Real-time fluorescence results (i.e., SYBR Fluorescence intensity) is shown in Figure 6 In FIG. B, curve 603 is based on 10 6 The reaction started with 10 MUT templates, and curve 604 is 6 The C T The values ​​are 19.1 (curve 603) and 27.8 (curve 604), resulting in a ΔC of 8.8 cycles. T In the upper left corner of the figure is a representative schematic of a conventional forward primer (line) with the query nucleotide (circle) located at the penultimate position at the 3' end of the primer.

[0204] Example 3: EGFR mutation L858R and 24-14-5:1:1 multipart primers (real-time data)

[0205] Two PCR amplification and detection assays were performed using the mutant (MUT) and wild type (WT) templates described in Example 1. In this experiment, the forward primer (FP) was a multipart primer according to the present invention. The inventors used the same reverse primer as in Example 1.

[0206] In the inventors' nomenclature, the multipart primers used in this example refer to such a format, such as a "24-14-5:1:1" primer refers to an anchor sequence of 24 nucleotides in length, a bridge sequence of 14 nucleotides in length, and a foot sequence of seven nucleotides in length (comprising, from the 5' end of the foot sequence, five nucleotides complementary to both the MUT and WT targets, one query nucleotide that is not complementary to the corresponding nucleotide in the WT target but complementary to the corresponding nucleotide in the MUT target, and, finally, one nucleotide complementary to both targets). Because the query nucleotide is located one nucleotide inside the 3' end of the primer, the inventors refer to this nucleotide as being located at the "3'-penultimate position". When the bridge sequence is compared to the region of the target sequence located between the binding sequence of the anchor sequence and the binding sequence of the foot sequence (which the inventors refer to as the "insertion sequence"), it is seen that the insertion sequence in this example is fourteen nucleotides in length, which is the same length as the bridge sequence. The sequence of the bridge sequence is selected so that it is not complementary to the insertion sequence to prevent the bridge sequence from hybridizing with the insertion sequence during primer annealing. Instead of annealing to each other, the bridge sequence and insert sequence form a single-stranded "bubble" when both the anchor sequence and the foot sequence are hybridized to the template. The "perimeter of the bubble" is defined as the sum of the number of nucleotides in the bridge sequence plus the number of nucleotides in the insert sequence plus the 3' nucleotides of the anchor sequence and its complement plus the 5'-terminal nucleotides of the foot sequence and its complement. Thus, the perimeter of the bubble formed by the binding of the multi-part primer in this example to the template molecule used in this example is 14+14+2+2, which is equal to 32 nucleotides in length.

[0207] The primer sequences and expected target sequence with mutant allele (MUT) are as follows:

[0208] Primer 24-14-5:1:1 anchors the bridge foot

[0209]

[0210] RP:5'-GCATGGTATTCTTTCTCTTCCGCA-3'(SEQ ID No.3)

[0211] In the multi-part forward primer, the bridge sequence is underlined, and the query nucleotide in the foot sequence is bold, underlined, and larger. In the mutant target sequence, the anchor sequence for the forward primer and the binding sequence for the foot sequence of the forward primer are underlined, and the sequence of the reverse primer is underlined. In addition, in the mutant target sequence, the nucleotides specific for the mutant are bold, underlined, and larger. The SciTools program for calculating the melting temperature of DNA hybrids by Integrated DNA Technologies (specific parameters: [oligo] = 0.06 μM; [Na +]=60mM; [Mg 2+ ]=3mM; [dNTPs]=0.25mM), the Tm of the anchor sequence binding to the template is 66.9℃, and the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 79.9℃.

[0212] PCR amplification was performed as described in Example 1. Real-time fluorescence results (i.e., SYBR Fluorescence intensity) is shown in Figure 7 In which curve 701 is based on 10 6 The reaction started with 10 MUT templates, and curve 702 is 6 The assay instrument automatically calculates the cycle threshold (C T ). These values ​​are 22.9 (curve 701) and 41.1 (curve 702), giving a ΔC of 18.2 cycles. T In the upper left corner of the figure is a representative schematic of a multipartite primer (bridge sequence as a semicircle) with a query nucleotide (circle) located at the penultimate position at the 3' end of the primer.

[0213] Example 4: EGFR mutation L858R and 24-14-5:1:1 multipart primers (selective amplification)

[0214] A series of PCR amplifications and detection assays were performed using the same multipart primers, reverse primers, expected target (MUT) and unintended target (WT) as described in Example 3. Amplifications were performed as described in Example 3. Real-time fluorescence results (i.e., SYBR Fluorescence intensity) is shown in Figure 8 In which curve 801 is based on 10 6 WT template, and curves 802-807 are reactions in which each reaction contains 10 6 WT templates were added with 10 6 , 10 5 , 10 4 , 10 3 , 10 2 or 10 1 The assay instrument automatically calculates the cycle threshold (C T). Those values ​​are 41.1 (curve 801), 23.3 (curve 802), 26.8 (curve 803), 30.5 (curve 804), 33.8 (curve 805), 37.0 (curve 806), and 39.2 (curve 807). In the upper left corner of the figure is a representative schematic of a multipartite primer (bridge sequence is a semicircle) with the query nucleotide (circle) located at the penultimate position of the 3' end of the primer.

[0215] Fig. 9 is the observed C for each reaction as a function of the logarithm of the number of MUT templates present in the reaction containing the MUT template. T Value (from Figure 8 The line 901 is the linear correlation fitted to the data points. The dashed line 902 marks the 6 C of amplification initiated with WT template and without MUT template T value.

[0216] Example 5: Effect of EGFR mutation L858R and reduction of multipart primer foot sequence length

[0217] The experiment described in Example 4 was repeated using the same 24-14-5:1:1 primer (SEQ. ID No. 6) with a foot sequence of seven nucleotides long; and also using two other multipart primers of the same design, except that the foot sequence of one of the additional primers was one nucleotide longer (24-14-6:1:1) and the foot sequence of the other additional primer was one nucleotide shorter (24-14-4:1:1). In all three cases, the anchor sequence was 24 nucleotides long, the bridge sequence was 14 nucleotides long and the insert sequence of the target was 14 nucleotides long, so that the bubble perimeter was 32 nucleotides in all cases. In addition, in all three cases, the query nucleotide was located at the 3' penultimate position of the foot sequence of the primer. The primer sequences and their expected target sequences (MUT) are as follows:

[0218] Primer 24-14-4:1:1 anchors the bridge foot

[0219]

[0220]

[0221] Primer 24-14-5:1:1 anchors the bridge foot

[0222]

[0223] Primer 24-14-6:1:1 anchors the bridge foot

[0224]

[0225] Reverse primer

[0226] RP:5'-GCATGGTATTCTTTCTCTTCCGCA-3'(SEQ ID No.3)

[0227] In the multi-part forward primer, the bridge sequence is underlined, and the query nucleotide in the foot sequence is bold, underlined, and larger. In the mutant target sequence, the anchor sequence for the forward primer and the binding sequence for the foot sequence of the forward primer are underlined, and the sequence of the reverse primer is underlined. In addition, in the mutant target sequence, the nucleotides specific for the mutant are bold, underlined, and larger. The SciTools program for calculating the melting temperature of DNA hybrids was used (specific parameters [oligo] = 0.06 μM; [Na + ]=60mM; [Mg 2+ ]=3mM; [dNTPs]=0.25mM); the Tm of the 24-14-4:1:1 anchor sequence binding to the template is 68.1℃, and the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 80.3℃; the Tm of the 24-14-5:1:1 anchor sequence binding to the template is 66.9℃, and the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 79.9℃; and the Tm of the 24-14-6:1:1 anchor sequence binding to the template is 68.1℃, and the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 79.4℃.

[0228] For each of the three multi-part primer designs, a series of PCR amplifications and detection assays were performed as described in Example 4, using 10 6 WT templates were added with 10 6 , 10 5 , 10 4 , 10 3 , 10 2 or 10 1 The assay instrument automatically calculates the cycle threshold (C T ). The C calculated from the real-time data for each reaction (not shown) T The value is summed with 10 6 C calculated for reactions initiated with WT template and without MUT template T The values ​​are listed together in Table 1.

[0229] Table 1

[0230]

[0231] Fig.10is the observed C shown as a function of the logarithm of the number of MUT templates present in each reaction. T Line 1001 is the C value fitted to a primer with a six nucleotide long foot sequence (4:1:1). T Line 1002 is a linear correlation of the values ​​of C for a primer with a seven-nucleotide long foot sequence (5:1:1). T and line 1003 is a linear correlation of the values ​​of C fitted to a primer having a foot sequence (6:1:1) of eight nucleotides long. T When using 24-14-6:1:1 primers, the less abundant MUT template samples gave C T The values ​​indicate the presence of a small number of obscuring amplicons generated by the abundant WT template in the sample.

[0232] These results demonstrate that the use of multi-part primers with shorter foot sequences (such as primer 24-14-5:1:1) reduces this problem, and the use of multi-part primers with the shortest foot sequences (such as primer 24-14-4:1:1) essentially eliminates the problem, enabling detection and quantification of as few as 10 expected template molecules in the presence of 1,000,000 unintended template molecules.

[0233] Example 6: Effect of EGFR mutation L858R and increased circumference of multipartite primer bubble

[0234] The experiment described in Example 4 was repeated using the same 24-14-5:1:1 primer (SEQ. ID No. 6) with a bridge sequence of 14 nucleotides in length (which, when hybridized to its template, establishes an insert sequence also 14 nucleotides in length); and also using two other multipart primers of the same design, except that the bridge sequence of one of the additional primers was 18 nucleotides in length (24-18-5:1:1) and the bridge sequence of the other additional primer was 10 nucleotides in length (24-10-5:1:1). In all three cases, the anchor sequence was 24 nucleotides in length, the foot sequence was 5:1:1, and the anchor sequence was chosen so that the insert sequence established when the primer binds to its template was the same length as the bridge sequence of the primer. As a result, the perimeter of the bubble formed by the three multipart primers of this series was 24, 32 and 40 nucleotides in length, respectively. In addition, in all three cases, the query nucleotide was located at the 3' penultimate position of the foot sequence of the primer. The primer sequences as well as the expected target sequence (MUT) are as follows:

[0235] Primer 24-10-5:1:1 anchors the bridge foot

[0236]

[0237] Primer 24-14-5:1:1 anchors the bridge foot

[0238]

[0239] Primer 24-18-5:1:1 anchors the bridge foot

[0240]

[0241] Reverse primer

[0242] RP:5'-GCATGGTATTCTTTCTCTTCCGCA-3'(SEQ ID No.3)

[0243] In the multi-part forward primer, the bridge sequence is underlined, and the query nucleotide in the foot sequence is bold, underlined, and larger. In the mutant target sequence, the anchor sequence for the forward primer and the binding sequence for the foot sequence of the forward primer are underlined, and the sequence of the reverse primer is underlined. In addition, in the mutant target sequence, the nucleotides specific for the mutant are bold, underlined, and larger. The SciTools program for calculating the melting temperature of DNA hybrids was used (specific parameters [oligo] = 0.06 μM; [Na + ]=60mM; [Mg 2+ ]=3mM; [dNTPs]=0.25mM); the Tm of the 24-10-5:1:1 anchor sequence binding to the template is 66.3℃, and the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 78.0℃; the Tm of the 24-14-5:1:1 anchor sequence binding to the template is 66.9℃, and the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 79.9℃; and the Tm of the 24-18-5:1:1 anchor sequence binding to the template is 67.9℃, and the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 79.3℃.

[0244] For each of the three multi-part primer designs, a series of PCR amplifications and detection assays were performed as described in Example 4, using 10 6 WT templates were added with 10 6 , 10 5 , 10 4 , 10 3 , 10 2 or 10 1 The assay instrument automatically calculates the cycle threshold (C T ). The C calculated from the real-time data for each reaction (not shown)T The value is summed with 10 6 C calculated for reactions initiated with WT template and without MUT template T The values ​​are listed together in Table 2.

[0245] Table 2

[0246]

[0247] Fig.11 is the observed C shown as a function of the logarithm of the number of MUT templates present in each reaction. T Line 1101 is the fit to the C value of the primers that form a bubble with a perimeter of 24 nucleotides long. T Line 1102 is a linear correlation of the values; line 1102 is a fit to the C of the primer that forms a bubble with a perimeter of 32 nucleotides long. T and line 1103 is a linear correlation of the values; and line 1104 is a fit to the C of the primer that forms a bubble with a perimeter of 40 nucleotides long. T Similar to what happened with the primers with longer foot sequences, when using the 24-10-5:1:1 primers (which form relatively small bubbles), the less abundant MUT template samples gave C that was present somewhat earlier than predicted. T The values ​​indicate the presence of a small number of obscuring amplicons generated by the abundant WT template in the sample.

[0248] These results demonstrate that the use of multi-part primers that form larger bubbles (such as primer 24-14-5:1:1) reduces this problem, and the use of multi-part primers that form the largest bubbles (such as primer 24-18-5:1:1) essentially eliminates the problem, enabling detection and quantification of as few as 10 expected template molecules in the presence of 1,000,000 unintended template molecules.

[0249] Example 7: Effect of EGFR mutation L858R and changing the position of the query nucleotide within the foot sequence of a multipart primer

[0250] The experiment described in Example 3 was repeated using the same 24-14-5:1:1 primer (SEQ. ID No. 6) including a seven nucleotide long foot sequence (where the query nucleotide is located at the penultimate position of the 3' end of the primer), and also using five additional multipart primers of the same design, except that the position of the query nucleotide of the foot sequence was changed. In all six cases, the anchor sequence was 24 nucleotides long, the bridge sequence was 14 nucleotides long, and the foot sequence was 7 nucleotides long. The primer sequences and the expected target sequence (MUT) are as follows:

[0251] Primer 24-14-6:1:0 anchors the bridge foot

[0252]

[0253] Primer 24-14-5:1:1 anchors the bridge foot

[0254]

[0255] Primer 24-14-4:1:2 anchors the bridge foot

[0256]

[0257]

[0258] Primer 24-14-3:1:3 anchors the bridge foot

[0259]

[0260] Primer 24-14-2:1:4 anchors the bridge foot

[0261]

[0262] Primer 24-14-1:1:5 anchors the bridge foot

[0263]

[0264] Reverse primer

[0265] RP:5'-GCATGGTATTCTTTCTCTTCCGCA-3'(SEQ ID No.3)

[0266] In the multi-part forward primer, the bridge sequence is underlined, and the query nucleotide in the foot sequence is bold, underlined, and larger. In the mutant target sequence, the anchor sequence for the forward primer and the binding sequence for the foot sequence of the forward primer are underlined, and the sequence of the reverse primer is underlined. In addition, in the mutant target sequence, the nucleotides specific for the mutant are bold, underlined, and larger. The SciTools program for calculating the melting temperature of DNA hybrids was used (specific parameters [oligo] = 0.06 μM; [Na + ]=60mM; [Mg 2+]=3mM; [dNTPs]=0.25mM); the Tm of the 24-14-6:1:0 anchor sequence binding to the template is 67.9°C, and the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 79.0°C; the Tm of the 24-14-5:1:1 anchor sequence binding to the template is 66.9°C, and the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 79.9°C; the Tm of the 24-14-4:1:2 anchor sequence binding to the template is 68.1°C, and the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 79.9°C. The Tm of the 24-14-3:1:3 anchor sequence binding to the template is 67.0°C, and the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 78.9°C; the Tm of the 24-14-2:1:4 anchor sequence binding to the template is 65.6°C, and the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 78.2°C; and the Tm of the 24-14-1:1:5 anchor sequence binding to the template is 66.6°C, and the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 78.1°C.

[0267] PCR amplification was performed as described in Example 1. Real-time fluorescence results (i.e., SYBR Fluorescence intensity) is shown in Fig.12 In each of the six figures, the multipart primers used are identified. In each figure, the odd-numbered curves are the ones containing 10 6 The results are obtained for samples with 10 MUT templates, and the even-numbered curves are the ones containing 10 6 Table 3 lists the machine calculated C values ​​for each primer for both targets. T values, and also shows the difference (ΔC T ).

[0268] Table 3

[0269]

[0270] Example 8: Effects of EGFR mutation L858R and altered symmetry of multipartite primer bubble

[0271] The experiment described in Example 3 was repeated using the same 24-14-5:1:1 primer (SEQ. ID No. 6) that forms a symmetrical bubble (which includes a 14-nucleotide long bridge sequence and a 14-nucleotide long insert sequence from the template); and the experiment also used four additional multipartite primers that formed different asymmetric bubbles with the mutant target (SEQ ID No. 2). "Asymmetric bubble" refers to a bubble formed by a bridge sequence and an insert sequence in the template with different lengths. In this experiment, all multipartite primers compared had an anchor sequence of 24 nucleotides in length, a 5:1:1 foot sequence, and a bridge sequence of different lengths (which was 18, 16, 14, 12 or 10 nucleotides in length). For each multipartite primer, the identity of the anchor sequence was chosen so that the sum of the length of the bridge sequence plus the length of the insert sequence (formed by the binding of both the anchor sequence and the foot sequence to the template) was equal to 28. As a result, the perimeter of the bubble formed by each of these five multipartite primers was always the same. The purpose of this experiment was to determine whether the formation of an asymmetric bubble affects the selectivity of the primers. The primer sequences and expected target sequence (MUT) are as follows:

[0272] Primer 24-18 / 10-5:1:1 anchors the bridge foot

[0273]

[0274] Primer 24-16 / 12-5:1:1 anchors the bridge foot

[0275]

[0276] Primer 24-14 / 14-5:1:1 anchors the bridge foot

[0277]

[0278] Primer 24-12 / 16-5:1:1 anchors the bridge foot

[0279]

[0280] Primer 24-10 / 18-5:1:1 anchors the bridge foot

[0281]

[0282] Reverse primer

[0283] RP:5'-GCATGGTATTCTTTCTCTTCCGCA-3'(SEQ ID No.3)

[0284] In the multipart forward primer, the bridge sequence is underlined, and the query nucleotide in the foot sequence is bold, underlined, and larger. In the mutant target sequence, the anchor sequence for the forward primer and the binding sequence for the foot sequence of the forward primer are underlined, and the sequence of the reverse primer is underlined. In addition, in the mutant target sequence, the nucleotides specific for the mutant are bold, underlined, and larger. Technologies' SciTools program for calculating the melting temperature of DNA hybrids (specific parameters [oligo] = 0.06 μM; [Na+] = 60 mM; [Mg2+] = 3 mM; [dNTPs] = 0.25 mM); the Tm of the 24-18 / 10-5:1:1 anchor sequence binding to the template is 66.3°C, and the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 79.1°C; the Tm of the 24-16 / 12-5:1:1 anchor sequence binding to the template is 67.0°C, and the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 79.1°C. The Tm of the complementary amplicon binding is 78.5°C; the Tm of the 24-14 / 14-5:1:1 anchor sequence binding to the template is 66.9°C, and the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 79.9°C; the Tm of the 24-12 / 16-5:1:1 anchor sequence binding to the template is 66.3°C, and the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 79.5°C; and the Tm of the 24-10 / 18-5:1:1 anchor sequence binding to the template is 67.9°C, and the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 79.3°C.

[0285] PCR amplification was performed as described in Example 1. Real-time fluorescence results (i.e., SYBR Fluorescence intensity) is shown in Fig.13 Each of the five graphs in FIG. 1 identifies a bubble that can be formed by the length of the bridge sequence and the length of the target insert sequence (e.g., "18 / 10 bubble" indicates that the forward primer 24- 18 / 10-5:1:1, which can form an insertion sequence of 10 nucleotides with the target). In each figure, the odd-numbered curves are those containing 10 nucleotides at the beginning. 6 The results are obtained for samples with 10 MUT templates, and the even-numbered curves are the ones containing 10 6 Table 4 lists the machine calculated C values ​​for the two targets for each primer. T values, and also shows the difference (ΔC T ).

[0286] Table 4

[0287]

[0288] Example 9: B-raf mutation V600E

[0289] The inventors used the method of Example 4 with multi-part primers according to the present invention targeting B-raf mutation V600E (which is a single nucleotide polymorphism). For comparison purposes, the inventors used a 24-14-5:1:1 design for the primers. The primer sequences and the expected target sequence (MUT) are as follows:

[0290] B-raf primer anchor bridge foot

[0291]

[0292] Reverse primer

[0293] RP:5'-ATAGGTGATTTTGGTCTAGC-3'(SEQ ID No.22)

[0294] In the multi-part forward primer, the bridge sequence is underlined, and the query nucleotide in the foot sequence is bold, underlined and larger. In the mutant target sequence, the binding sequence for the anchor sequence of the forward primer and the binding sequence for the foot sequence of the forward primer are underlined, and the sequence of the reverse primer is underlined. The SciTools program for calculating the melting temperature of DNA hybrids by Integrated DNA Technologies was used (specific parameters: [oligo] = 0.06 μM; [Na + ]=60mM; [Mg 2+ ]=3mM; [dNTPs]=0.25mM), the Tm of the anchor sequence binding to the template is 63.5℃, the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 71.1℃, and the calculated Tm of the reverse primer binding is 56.1℃.

[0295] Plasmids were prepared by inserting synthetic oligonucleotides into pGEM-11Zf(+) vector (Promega), corresponding to 116bp EGFR gene fragments containing B-raf V600E mutation or B-raf wild-type sequence. Mutant and wild-type plasmid DNA were digested with restriction endonuclease Mse I (New England Biolabs). The digestion mixture contained 10 units of Mse I and 4 μg of mutant or wild-type genomic DNA in a 20 μl volume (which contained 5 mM KAc, 2 mM Tris-Ac (pH 7.9), 1 mM MgAc, 1% bovine serum albumin and 100 μM dithiothreitol). The reaction was incubated at 37°C for 120 min, followed by incubation at 65°C for 20 min to inactivate the enzyme.

[0296] The PCR product was prepared in a 1% flask containing 50 mM KCl, 10 mM Tris-HCl (pH 8.0), 3 mM MgCl2, 1.5 units of AmpliTaq Gold DNA polymerase, 250 μM of each deoxyribonucleoside triphosphate (dNTP), 60 nM of each primer, and 1x PCR amplification was performed in a 30 μl volume of 1% ... Green fluorescence intensity.

[0297] Use containing 10 6 WT templates were added with 10 6 , 10 5 , 10 4 , 10 3 , 10 2 or 10 1 A series of PCR amplifications and detection assays were performed using a dilution series of 10 copies of the MUT template. 6 Based on the real-time fluorescence data (not shown), the assay instrument automatically calculated the threshold cycle (C T For the B-rafV600E mutant dilution series, those values ​​were 27.7 (10 6 MUT templates), 31.1(10 5 MUT templates), 34.1(10 4 MUT templates), 37.6(10 3 MUT templates), 43.0(10 2 MUT templates), 46.9(10 1 MUT templates) and 50.8(10 6 WT templates and no MUT templates). Fig.14 is the C observed in each reaction containing MUT template as a function of the logarithm of the number of MUT templates present in the reaction T The line 1401 is the linear correlation fitted to the data points. The dashed line 1402 marks the 6 C of amplification initiated with WT template and without MUT template T value.

[0298] Example 10: EGFR mutation T790M in human genomic DNA

[0299] A series of PCR amplification and detection assays were performed using human genomic DNA containing the EGFR mutation T790M (isolated from cell line H1975, which contains the EGFR T790M mutation) and human genomic DNA containing the corresponding wild-type sequence (isolated from human genomic DNA obtained from Coriell Cell Repositories) (which differs from the EGFR gene in a single nucleotide polymorphism) as templates. The forward primer is a 24-14-4:1:1 multi-part primer according to the present invention. The reverse primer is a conventional linear primer. The primer sequences and the expected target sequence (MUT) are as follows:

[0300] T790M primer anchors the bridge foot

[0301]

[0302] Reverse primer

[0303] RP:5'-GAGGCAGCCGAAGGGCATGAGC-3'(SEQ ID No.25)

[0304] In the multi-part forward primer, the bridge sequence is underlined, and the query nucleotide in the foot sequence is bold, underlined and larger. In the mutant target sequence, the binding sequence for the anchor sequence of the forward primer and the binding sequence for the foot sequence of the forward primer are underlined, and the sequence of the reverse primer is underlined. The SciTools program for calculating the melting temperature of DNA hybrids by Integrated DNA Technologies was used (specific parameters: [oligo] = 0.06 μM; [Na + ]=60mM; [Mg 2+ ]=3mM; [dNTPs]=0.25mM), the Tm of the anchor sequence binding to the template is 72.5℃, the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 73.9℃, and the calculated Tm of the reverse primer binding is 68.2℃.

[0305] Mutant and wild-type human genomic DNA were digested with restriction endonuclease Mse I. The digestion mixture contained 10 units of Mse I and 4 μg of mutant or wild-type genomic DNA in a 20 μl volume (which contained 5 mM KAc, 2 mM Tris-Ac (pH 7.9), 1 mM MgAc, 1% bovine serum albumin and 100 μM dithiothreitol). The reaction was incubated at 37°C for 120 min, followed by incubation at 65°C for 20 min to inactivate the enzyme.

[0306] The PCR product was prepared in a 1% flask containing 50 mM KCl, 10 mM Tris-HCl (pH 8.0), 3 mM MgCl2, 1.0 unit AmpliTaq Gold DNA polymerase, 250 μM of each deoxyribonucleoside triphosphate (dNTP), 60 nM of each primer, and 1x PCR amplification was performed in a 20 μl volume of 1% ... Green fluorescence intensity.

[0307] A series of PCR amplification and detection assays were performed using a dilution series containing 10,000 WT template plus 10,000, 3,000, 1,000, 300, 100, 30 or 10 copies of MUT template. The inventors also included samples containing only 10,000 WT template. Based on the real-time fluorescence data (not shown), the assay instrument automatically calculated the threshold cycle (C T For this T790M dilution series, those values ​​were 29.2 (10,000 MUT templates), 31.1 (3,000 MUT templates), 32.7 (1,000 MUT templates), 35.5 (300 MUT templates), 38.2 (100 MUT templates), 38.8 (30 MUT templates), 40.7 (10 MUT templates), and 42.8 (10,000 WT templates and no MUT template). Fig.15 is the C observed for each reaction containing MUT template as a function of the logarithm of the number of MUT templates present in the reaction T 1501 is the linear correlation fitted to the data points. Dashed line 1502 is the C value of the amplification starting with 10,000 WT template and no MUT template. T value.

[0308] Example 11: Quantification of EGFR mutation L858R in an Applied Biosystems PRISM 7700 Fluorescence Spectrophotometer Thermal Cycler

[0309] Using a different thermal cycler Applied Biosystems PRISM 7700 spectrofluorophotometric thermal cycler, an experiment similar to that reported in Example 4 was performed to amplify and detect the mutation L858R in the EGFR gene. Plasmid DNA containing the EGFR mutation L858R and plasmid DNA containing the corresponding wild-type sequence (which has a single nucleotide polymorphism in the EGFR gene) were used as templates to perform a series of PCR amplification and detection assays. Contrary to the template used in Example 4, in this experiment, the template was not digested with a restriction endonuclease. Amplification was performed using the same multi-part forward primer and conventional reverse primer as in Example 3. The primer sequences and the expected target sequence (MUT) are as follows:

[0310] Primer 24-14-5:1:1 anchors the bridge foot

[0311]

[0312] RP:5'-GCATGGTATTCTTTCTCTTCCGCA-3'(SEQ ID No.3)

[0313] In the multi-part forward primer, the bridge sequence is underlined, and the query nucleotide in the foot sequence is bold, underlined and larger. In the mutant target sequence, the binding sequence for the anchor sequence of the forward primer and the binding sequence for the foot sequence of the forward primer are underlined, and the sequence of the reverse primer is underlined. The SciTools program for calculating the melting temperature of DNA hybrids by Integrated DNA Technologies was used (specific parameters: [oligo] = 0.06 μM; [Na + ]=60mM; [Mg 2+ ]=3mM; [dNTPs]=0.25mM), the Tm of the anchor sequence binding to the template is 66.9°C, the Tm of the entire multi-part primer binding to the resulting complementary amplicon is 79.9°C, and the calculated Tm of the reverse primer binding is 68.2°C.

[0314] The PCR product was prepared in a 1% flask containing 50 mM KCl, 10 mM Tris-HCl (pH 8.0), 3 mM MgCl2, 2.0 units of AmpliTaq Gold DNA polymerase, 250 μM of each deoxyribonucleoside triphosphate (dNTP), 60 nM of each primer, and 1x PCR amplification was performed in a 40 μl volume of 1% dapoxetine green. Amplification was performed on an Applied Biosystems PRISM 7700 spectrofluorometric thermocycler using 0.2 ml polypropylene PCR tubes (transparent). The thermal cycle configuration was 95°C for 10 min, followed by 55 cycles of 94°C for 15 sec, 60°C for 20 sec, and 72°C for 20 sec. The concentration of 1% dapoxetine was measured at the end of each chain extension phase (72°C). Green fluorescence intensity.

[0315] Use containing 10 6 WT templates were added with 10 6 , 10 5 , 10 4 , 10 3 , 10 2 or 10 1 A series of PCR amplifications and detection assays were performed using a dilution series of 10 copies of the MUT template. 6 Based on the real-time fluorescence data (not shown), the assay instrument automatically calculated the threshold cycle (C T ). Those values ​​are 21.2(10 6 MUT templates), 24.9(10 5 MUT templates), 28.3(10 4 MUT templates), 32.2(10 3 MUT templates), 36.0(10 2 MUT templates), 37.6(10 1 MUT templates) and 38.7(10 6 WT templates and no MUT templates). Fig.16 is the observed C for each reaction as a function of the logarithm of the number of MUT templates present in the reaction containing the MUT template. T The line 1601 is the linear correlation fitted to the data points. The dashed line 1602 is the linear correlation at 10 6 C of amplification initiated with WT template and without MUT template T value.

[0316] Example 12: Effect of ARMS discrimination when multipartite primers are used in PCR assays

[0317] In order to investigate the functioning of the multipart primers according to the invention, the inventors repeated the experiments described in Example 3, using not only the 24-14-5:1:1 primer described there, but also the truncated 24-14-5:0:0 primer, which is a primer having the same anchor sequence, the same bridge sequence and the same five 5' nucleotides of the foot sequence. It lacks the last two 3' nucleotides of the foot sequence. Therefore, its foot sequence is completely complementary to both the expected, mutant target and the unexpected, wild-type target. The primer sequences and the expected target sequence (MUT) for the reactions employing each of these two multipart primers are as follows:

[0318] Primer 24-14-5:1:1 anchors the bridge foot

[0319]

[0320] Primer 24-14-5:0:0 anchors the bridge foot

[0321]

[0322] Reverse primer

[0323] RP:5'-GCATGGTATTCTTTCTCTTCCGCA-3'(SEQ ID No.3)

[0324] In the multi-part forward primer, the bridge sequence is underlined, and the query nucleotide in the foot sequence is bold, underlined and larger. In the mutant target sequence, the binding sequence for the anchor sequence of the forward primer and the binding sequence for the foot sequence of the forward primer are underlined, and the sequence of the reverse primer is underlined. The SciTools program for calculating the melting temperature of DNA hybrids by Integrated DNA Technologies was used (specific parameters: [oligo] = 0.06 μM; [Na + ]=60mM; [Mg 2+ ]=3mM; [dNTPs]=0.25mM), the Tm of both the anchor sequence and the primer binding to the template is 66.9℃, the Tm of primer 24-14-5:1:1 binding to the resulting complementary amplicon is 79.9℃, and the Tm of primer 24-14-5:0:0 binding to the resulting complementary amplicon is 79.0℃.

[0325] PCR amplification was performed as described in Example 3. Real-time fluorescence results (i.e., SYBR fluorescence intensity). Fig.17Figure A shows the results obtained for the reaction containing primers 24-14-5:1:1, where curve 1701 is the reaction containing 10 6 The curve 1702 is the reaction of 10 6 WT template reactions; and Fig.17 Figure B shows the results obtained for the reaction containing primers 24-14-5:0:0, where curve 1703 is the reaction containing 10 6 The reaction of the MUT template and curve 1704 is a curve containing 10 6 The assay instrument automatically calculated the cycle threshold (C T ). Primer 24-14-5:1:1 C T The values ​​are 23.1 (curve 1701) and 40.7 (curve 1702), resulting in a ΔC of 17.6 cycles. T ; and primer 24-14-5:0:0 of ​​C T The values ​​are 39.7 (curve 1703) and 39.4 (curve 1704), giving a ΔC of -0.3 cycles. T (Showing that these two reactions give essentially the same results).

[0326] The foregoing examples and descriptions of the preferred embodiments should be considered illustrative and not intended to limit the invention as defined by the claims. It is readily understood that many variations and combinations of the above features may be utilized without departing from the scope of the invention as set forth in the claims. Such variations are not considered to depart from the scope of the invention, and all such variations are included within the scope of the following claims. All references cited herein are incorporated herein by reference in their entirety. Sequence Listing <110> Rutgers, The State University of New Jersey University of New Jersey <120> Highly selective nucleic acid amplification primers <130> 1812224US01 <140> PCT / US14 / 15351 <141> 2014-02-07 <150> 61 / 762,117 <151> 2013-02-07 <160> 26 <170> PatentIn version 3.5 <210> 1 <211> twenty one <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 1 attttgggcg ggccaaactg c 21 <210> 2 <211> 95 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 2 ccttgcatga ccacttttgt ggcgtcgtac agttctagtg tctaaaaccc gcccggtttg 60 acgacccacg ccttctcttt cttatggtac gtctt 95 <210> 3 <211> twenty four <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 3 gcatggtatt ctttctcttc cgca 24 <210> 4 <211> twenty one <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 4 caagatcaca gattttgggc g 21 <210> 5 <211> twenty one <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 5 aagatcacag attttgggcg g 21 <210> 6 <211> 45 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 6 ctggtgaaaa caccgcagca tgtcgcacga gtgagccctg ggcgg 45 <210> 7 <211> 44 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 7 tggtgaaaac accgcagcat gtcacacgag tgagccccgg gcgg 44 <210> 8 <211> 46 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 8 actggtgaaa acaccgcagc atgttggagc tgtgagcctt gggcgg 46 <210> 9 <211> 49 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 9 cgtactggtg aaaacaccgc agcactgacg acaagtgagc cctgggcgg 49 <210> 10 <211> 41 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 10 tgaaaacacc gcagcatgtc aagacactca gccctgggcg g 41 <210> 11 <211> 45 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 11 actggtgaaa acaccgcagc atgttgcacg agtgagcctt gggcg 45 <210> 12 <211> 45 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 12 tggtgaaaac accgcagcat gtcacacgag tgagccacgg gcggg 45 <210> 13 <211> 45 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 13 ggtgaaaaca ccgcagcatg tcaaacgagt gagccacagg cgggc 45 <210> 14 <211> 45 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 14 gtgaaaacac cgcagcatgt caaggaagtg agccacaagc gggcc 45 <210> 15 <211> 45 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 15 tgaaaacacc gcagcatgtc aagacagact gacccaaacg ggcca 45 <210> 16 <211> 49 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 16 tgaaaacacc gcagcatgtc aagacacacg acaagtgagc cctgggcgg 49 <210> 17 <211> 47 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 17 ggtgaaaaca ccgcagcatg tcaatccaac aagtgagccc tgggcgg 47 <210> 18 <211> 43 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 18 tactggtgaa aacaccgcag catggacgac gagccctggg cgg 43 <210> 19 <211> 41 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 19 cgtactggtg aaaacaccgc agcactgacg gccctgggcg g 41 <210> 20 <211> 45 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 20 agacaactgt tcaaactgat gggaaaacac aatcatctat ttctc 45 <210> twenty one <211> 75 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> twenty one ggtctgttga caagtttgac taccctgggt gaggtagctc taaagagaca tcgatctggt 60 tttagtggat aaaaa 75 <210> twenty two <211> 20 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> twenty two ataggtgatt ttggtctagc 20 <210> twenty three <211> 44 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> twenty three gcctgctggg catctgcctc acctaataat ctacaacaat catg 44 <210> twenty four <211> 77 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> twenty four cacggcggac gacccgtaga cggagtggag gtggcacgtc gagtagtacg tcgagtacgg 60 gaagccgacg gaggacc 77 <210> 25 <211> twenty two <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 25 gaggcagccg aagggcatga gc 22 <210> 26 <211> 43 <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 26 ctggtgaaaa caccgcagca tgtcgcacga gtgagccctg ggc 43 <210> 27 <211> twenty four <212> DNA <213> Artificial Sequence <220> <221> <222> <223> Primers <400> 27 ctggtgaaaa caccgcagca tgtc 24

Claims

1. A multi-part nucleic acid amplification primer that distinguishes between a mutant DNA target sequence and a closely related wild-type DNA target sequence, the multi-part nucleic acid amplification primer comprising the following three consecutive DNA sequences in the 5'-3' direction: an anchor sequence capable of forming a hybrid with the mutant DNA target sequence and with the wild-type DNA target sequence; a bridge sequence that is at least 10 nucleotides in length and is not complementary to either the mutant DNA target sequence or the wild-type DNA target sequence; and The foot sequence is 6-8 nucleotides long. wherein the mutant DNA target sequence is completely complementary to the foot sequence, and the wild-type DNA target sequence has a mismatch with the foot sequence in at least one nucleotide in the 3' nucleotide or the 3' penultimate nucleotide of the foot sequence, and in, If the anchor and foot sequences of the multi-part nucleic acid amplification primer hybridize to the mutant DNA target sequence or to the wild-type DNA target sequence, there is an insertion sequence of at least 10 nucleotides in length in the target sequence, which insertion sequence does not hybridize to the bridge sequence of the multi-part nucleic acid amplification primer, and the bridge sequence and the insertion sequence together form a bubble in the hybrid, which bubble has a circumference of 24-40 nucleotides.

2. The multi-part nucleic acid amplification primer of claim 1, comprising a functional portion located 5' to the anchor sequence, said functional portion not hybridizing to either the mutant DNA target sequence or the wild-type DNA target sequence.

3. The multi-part nucleic acid amplification primer of claim 2, wherein the functional part contains a fluorophore and generates a fluorescent signal indicative of amplification.

4. The multi-part nucleic acid amplification primer of claim 1, wherein the foot sequence is 6-7 nucleotides long.

5. The multi-part nucleic acid amplification primer according to claim 4, wherein the anchor sequence is capable of forming a hybrid with a length of 15-30 nucleotides with the mutant DNA target sequence and with the wild-type DNA target sequence.

6. The multi-part nucleic acid amplification primer of claim 5, wherein the bubble has a circumference of 28-40 nucleotides.

7. A multi-part nucleic acid amplification primer according to claim 6, wherein the bridge sequence and the insertion sequence have equal lengths.

8. The multi-part nucleic acid amplification primer of claim 1, wherein the bubble has a circumference of 28-40 nucleotides.

9. Use of a multi-part nucleic acid amplification primer according to any one of claims 1 to 8 in the preparation of a reagent or kit comprising the multi-part nucleic acid amplification primer for use in a method comprising repeatedly cycling a reaction mixture in a primer-dependent amplification reaction having a primer-annealing temperature and detecting the amplified product, wherein the reaction mixture comprises at least one mutant DNA target sequence or a closely related wild-type DNA target sequence or both, a DNA polymerase, other reagents required for amplification, and a primer pair comprising a multi-part nucleic acid amplification primer, in, the probability that a multi-part nucleic acid amplification primer / wild-type DNA target sequence hybrid will be extended during said cycle is at most 1 / 1,000 the probability that a multi-part nucleic acid amplification primer / mutant DNA target sequence hybrid will be extended during said cycle, wherein the method amplifies at least one rare mutant DNA target sequence in a mixture containing a wild-type DNA target sequence that is closely related to each mutant DNA target sequence, and "Rare" means that the ratio of the expected target sequence to closely related sequences is at least 1 / 10 3 to 1 / 10 7 "Closely related" refers to sequences that differ from the intended target sequence by one, two, or at most a few nucleotides.

10. The use according to claim 9, wherein at least one mutant DNA target sequence is a cDNA.

11. The use according to claim 9, wherein during the cycle, the probability that a multi-part nucleic acid amplification primer / wild-type DNA target sequence hybrid will be extended is at most 1 / 10,000 of the probability that a multi-part nucleic acid amplification primer / mutant DNA target sequence hybrid will be extended during the cycle.

12. The use according to claim 9, wherein the reaction mixture comprises a detection reagent for detecting the amplified product derived from each expected target sequence.

13. The use according to claim 12, wherein the amplification reaction is polymerase chain reaction (PCR), and the detection is real-time detection.

14. The use according to claim 13, wherein the detection reagent is selected from double-stranded DNA binding dyes and fluorescently labeled hybridization probes that generate a signal upon hybridization.

15. The use according to claim 13, wherein at least one of the multi-part nucleic acid amplification primers comprises a functional portion located 5' to the anchor sequence that does not hybridize to either the mutant DNA target sequence or its closely related wild-type DNA target sequence, wherein the functional group comprises a detection reagent that comprises a hairpin structure labeled with a fluorophore / quencher and generates a fluorescent signal indicative of amplification.

16. The use according to claim 13, wherein 10 6 The cycle threshold (C T ) than 10 6 The C of the assay started with 10 copies of its closely related wild-type DNA target sequence T At least 14 cycles late.

17. The use according to claim 13, wherein 10 6 The cycle threshold (C T ) than 10 6 The C of the assay started with 10 copies of its closely related wild-type DNA target sequence T At least 18 cycles late.

18. The use according to claim 12, wherein the at least one mutant DNA target sequence is at least two mutant DNA target sequences, and wherein the detection reagent for each mutant DNA target sequence comprises a fluorophore with a unique identification color.

Images