Bidirectional PCR amplification of specific alleles

Abstract

Bi-directional polymerase chain reaction (PCR) amplification of specific alleles (Bi-PASA). Two outer primers (P and Q) and two inner primers (A and B) are used. A and B are each specific for different alleles. In heterozygotes, three segments are amplified: a segment of size AQ resulting from one allele, another segment of size PB resulting from the second allele, and a combined segment of size PQ. In homozygotes, segment PQ and either segments AQ or PB amplify.

Description

BACKGROUND OF THE INVENTION

This is an invention the field of amplification of DNA by polymerase chain reaction (PCR).

PCR is a method that typically utilizes two oligonucleotide primers to amplify a DNA segment >1 million-fold. The two primers anneal to opposing strands of DNA at positions which span a target sequence of interest. A DNA polymerase is used for sequential rounds of template dependent synthesis of the DNA sequence. The PCR method is more fully described in U.S. Pat. No. 4,683,202, issued Jul. 28, 1987, the disclosure of which is incorporated herein.

PCR can be adapted for the rapid detection of single-base changes in genomic DNA by using specifically designed oligonucleotides in a method called PCR amplification of specific alleles (PASA) (Sommer et al. 1989; Sarkar et al. 1990; Wallace et al. U.S. Pat. No. 5,639,611, issued Jun. 17, 1997). This rapid method is also known as allele-specific amplification (ASA), allele-specific PCR, and amplification refractory mutation system (ARMS) (Newton et al. 1989; Nichols et al. 1989; Wu et al. 1989). For this technique an oligonucleotide primer is designed to match one allele perfectly but mismatch the other allele at or near the 3′ end, thereby preferentially amplifying one allele over the other. PASA assays can be developed for assaying virtually all alleles (Sommer et al. 1992). However, each PASA reaction provides information on the presence or absence of only one allele. Two PASA reactions must be performed to determine the zygosity of any sequence change.

To detect zygosity in one PCR reaction, PCR amplification of multiple specific alleles (PAMSA) utilizes three primers in one reaction to generate two allele-specific segments that differ sufficiently in size to be distinguished by agarose gel electrophoresis (Dutton and Sommer 1991). However, problems arise from PAMSA because of differences in length, and hence, amplification efficiency of the allele-specific primers. A similar approach-termed competitive oligonucleotide primary (COP)-utilized primers that mismatched the undesired allele within the middle rather than the end of the oligonucleotide (Gibbs et al. 1989; Ruano and Kidd 1989).

Tetra-primer PCR is a method by which two allele-specific amplifications occur in opposite directions (Ye et al. 1992). Tetra-primer PCR and the method of this invention, which we have named Bi-Directional Polymerase Chain Reaction Amplification of Specific Alleles (Bi-PASA), both rely on allele-specific PCR to amplify two alleles simultaneously and in opposite directions. However, the methods differ in the following ways. (1) In tetra-primer PCR, the allele specificity of the inner primers derive from mismatches in the middle of two complementary primers, whereas, in Bi-PASA, the mismatches are at (or near) the 3′ end of the primers. (2) The inner primers in Bi-PASA have short complementary segments and G+C-rich tails to efficiently switch from template-based amplification to self-amplification and to prevent megapriming. (3) Tetra-primer PCR utilizes two annealing conditions of high and low stringency, whereas Bi-PASA utilizes a constant annealing temperature. (4) The inner primers used in tetra-primer PCR are concentrated 35-fold more than the outer primers, whereas in Bi-PASA the primers are of similar concentration.

SUMMARY OF THE INVENTION

This invention is a method for conducting a bi-directional PCR amplification of specific alleles. DNA which may contain one or both of first and second alleles is subjected to a PCR utilizing an outer pair of primers P and Q and an inner pair of primers A and B.

Q is complementary to the sense strand of both alleles in a region downstream of the sequence difference (mismatch) X which distinguishes the alleles. X may be a substitution, deletion or insertion of one or more base pairs. P is complementary to the anti-sense strand of both alleles in a region upstream of X. The terms “upstream” and “downstream” relate to the direction of transcription.

B has a region at its 3′ end which is complementary to the sense strand of the first allele and A has a region at its 3′ end which is complementary to the antisense strand of the second allele. Each of A and B also has a non-complementary G+C-rich tail at its 5′end. X occurs at or near the 3′ end of each of A and B.

DNA which is heterozygous with respect to the two alleles results in amplification of three overlapping sequences, PQ, PB and AQ. PQ, PB and AQ, respectively, stand for the sequences extending from P to Q inclusive, from P to B inclusive, and from A to Q inclusive. DNA which is homozygous with respect to the first allele results in amplification of two overlapping sequences, PQ and PB. DNA which is homozygous with respect to the second allele results in amplification of two overlapping sequences, PQ and AQ.

Following amplification, the sample can be analyzed to determine whether PB and/or AQ is present along with PQ. Preferably, the number of nucleotides separating P from X is sufficiently different from the number of nucleotides separating Q from X that the segments PB and AQ can be distinguished on an agarose gel. However, this is not necessary when other methods of analysis are used.

Bi-PASA provides a rapid, one-tube method for simultaneously differentiating homozygotes and heterozygotes. It is applicable to detecting small deletions and insertions as well as single base changes. The method is particularly useful for determining zygosity in a sample containing a wild-type allele and/or a single-base change mutant and, in general, for determining the zygosity of common sequence changes in which heterozygotes are likely to be common. By using 3 or more inner primers, it can be used to differentiate 3 or more alleles. Bi-PASA can be used to perform population screening, haplotype analysis, patient screening, and carrier testing. It is rapid, reproducible, inexpensive, non-isotopic, and amenable to automation. The method can be conducted simultaneously on 2 or more loci in different tubes, using a uniform annealing temperature.

BRIEF DESCRIPTION OF THE FIGURES

FIG. 1A is a schematic representation of Bi-PASA. The four primers (P,Q, A and B) are represented by arrows. X represents the sequence change relative to the other allele. Wavy lines on the inner primers A and B represent the non-complementary 5′ tail sequences.

FIG. 1B illustrates the network structure of Bi-PASA. The four boxes represent the template used in the reaction. Arrows indicate the segments that can be produced from each template. The thickness of the arrows approximates the efficiency of amplification. The Bi-PASA reaction can be divided into two parts: template transfer amplification (genomic DNA→PQ, PB, and AQ; PQ→PB and AQ) and self-amplification (PQ→PQ, PB→PB, and AQ→AQ). The amplification efficiency, λ, for a particular segment from a given template is defined as the molecular ratio from cycle n to cycle n+1. [PQ] n+1 , [PB] n+1 and [AQ] n+1 are the PCR product yields of PQ, PB, and AQ, respectively, at the end of the n+1 cycle. [Genomic DNA] represents the yield of genomic DNA. The accumulation of PQ, PB, and AQ DNA products can be indicated as follows: [PQ] n+1 =[PQ] n ×(1+λ n PQ→PQ )+[genomic DNA]×λ n geomic DNA→PQ ; [PB] n+1 +[PB] n ×(1+λ n PB→PB )+½×[genomic DNA]×λ n genomic DNA→PB +½×[PQ]×λ n PQ→PB ; and [AQ] n+1 =[AQ] n ×(1+λ n AQ→AQ)+½×[genomic DNA]×λ n genomic DNA→AQ +½[PQ]×λ n PQ→AQ .

FIG. 2 illustrates the kinetics of Bi-PASA. Identical [α-32P]dATP-labeled Bi-PASA reactions were removed from the PCR thermocycler every three cycles, and the yield for each segment was quantitated with a PhosphorImager. Potentially, 135, 87 and 214 adenine nucleotides can be labeled by [α-32P]dATP in the PB, AQ and PQ segments, respectively. ( FIG. 2A ) All four primers were used for WT/M, WT/WT and M/M samples, indicating that the type of genomic template used can affect the yields. ( FIG. 2B ) The three two-primer combinations for the WT/M sample were performed. A greater yield was observed for the three segments individually as compared to the WT/M sample in FIG. 2A , indicating that interactions among the primers can influence the yield in a Bi-PASA reaction.

FIG. 3 shows the results of Bi-PASA reactions for the FV and COMT genes as described in the Examples, and illustrates interactions among the primers. Two-, three-, and four-primer combinations for the WT/WT (lanes 1-6), M/M (lanes 7-12) and WT/M (lanes 13-18) templates were amplified to observe the efficiency of the primers. (M) Size standard (120 ng of θX174/HaeIII). ( FIG. 3A ) COMT gene Bi-PASA reaction. The inner primers used were A2 and B3 (see Table 1 for primer specifications). The primer concentrations were 0.05 μM for P, Q and B, and 0.1 μM for A. The annealing temperature was 65° C., and 5% DMSO was added to the reaction. ( FIG. 3B ) FV Bi-PASA reaction. The inner primers used were A11 and B11 (see Table 1). The primer concentrations were 0.1 μM for P, B, and A, and 0.05 μM for Q. The annealing temperature was 55° C.

FIG. 4 shows the effects of primer concentrations for the FV Bi-PASA reaction. WT/M genomic DNA was used for all reactions. Lanes 1-6, 7-12, 13-18, and 19-24 show the effects of varying only the P, Q, B11 and A11 primers, respectively. Lanes 25-30 show the effects of varying all four primers at once. The primers used and the 1× concentrations were determined by the optimal conditions used in FIG. 3 B. (M) Size standard (120 ng of θX174/HaeIII).

FIG. 5 shows the results of prospective study described below. Optimized BI-PASA reactions for single-base changes in the factor IX, FV-opp, and D1 genes. (M) Size standard (120 ng of φX174/HaeIII). ( FIG. 5A ) Optimation of the FIX gene Bi-PASA. Bi-PASA reactions shown were performed with a WT/M sample. PQ segment T m =73.6° C. The annealing temperature was set at 55° C. (Lanes 1-4 PCR with each P and Q primer at 0.1, 0.05, 0.025, and 0.0125 μM concentration. Lanes 2 was designated as the minimum optimal concentration for the PQ segment. (Lanes 5-10) Control reactions with P and Q primers at 0.05 μM concentrations and A and B primers at 0.1 μM concentration. Lanes 11-16) Final Bi-PASA reaction for the FIX gene after adjustment of the A primer concentration to 0.05 μM. ( FIG. 5B ) Bi-PASA reaction of FV-opp and D1 genes. In the initial experiments using the FV gene, the WT and M allele-specific primers were in the upstream and downstream directions, respectively. As a test of the robustness of the Bi-PASA reaction, new primers were designed such that the WT and M allele-specific primers were in the downstream and upstream direction, respectively. WT/WT, M/M, and WT/M templates were used for optimized Bi-PASA reactions for the FV-opp and D1 genes. The T m values of the PQ segments of the FV-opp and D1 genes were 74.8° C. and 81.4° C., respectively, and the annealing temperatures were 55° C. and 60° C., respectively. Final primer concentrations for the FV-opp Bi-PASA reaction were 0.1 μM for the P primer, 0.05 μM for the Q primer, 0.1 μM for the B primer, and 0.1 μM for the A primer. Final primer concentrations for the D1 Bi-PASA reaction were 0.1 μM for the P primer, 0.1 μM for the Q primer, 0.1 μM for the B primer, and 0.1 μM for the A primer. (M) Size standard (120 ng φX171/HaeIII). (Lanes 1-4) Bi-PASA reaction for FV-opp; (lanes 5-8) bi-PASA for D1.

DETAILED DESCRIPTION OF THE INVENTION

Principle of Bi-PASA

For Bi-PASA, PCR is performed with four oligonucleotide primers: two outer primers, P and Q; and two allele-specific inner primers, A and B (FIG. 1 A). Preferably, P and Q should anneal at different distances from the sequence change to differentiate the downstream AQ and upstream PB segments on an agarose gel. A and B are each specific to an allele with the mismatch at (or near, i.e, within about 3 bp of) the 3′ end of the primer. A and B primers consist of two parts: a short region (about 10 to 20) that is complementary to one of the alleles and a 5′ non-complementary tail. The tail sequence consists of about 6 to 20 nucleotides, preferably about 10 nucleotides, of high G+C content (about 50-100% G+C, preferably about 60 to 100%) and serves two purposes: (1) The tail acts as a “switch” from inefficient amplification of genomic DNA to efficient amplification of previously amplified template DNA; and (2) the tail prevents “megapriming”, which occurs when a segment (PB or AQ) generated by PCR in an earlier cycle acts as a primer for a larger template (i.e., genomic DNA or PQ) in a subsequent cycle (Sarkar and Sommer 1990, 1992).

Depending on the zygosity, Bi-PASA produces two or three overlapping segments. PQ is always produced and serves as a positive control. PB and AQ are both present in a heterozygote (WT/M), but only PB is produced in homozygous wild-type (WT/WT) and only AQ is produced in homozygous mutant (M/M) samples.

Network Structure of Amplification Process

FIG. 1B depicts the network structure of Bi-PASA amplification for heterozygotes. Two types of amplification occur during Bi-PASA “template transfer amplification” and “self amplification.” Template transfer amplification occurs when a larger template is used to produce a smaller product (genomic DNA→PQ, PB, and AQ; PQ→PB and AQ). Self-amplification occurs when a template is used to reproduce itself (PQ→PQ, PB→PB, and AQ→AQ). The amplification conditions are designed to favor self-amplification, as represented by the thickness of the arrows. The efficiency may be influenced by the cycle number and by the zygosity of the sample.

Because of the relatively short complementary region of the A and B primers that compromises efficient annealing at the annealing temperature used, the PQ template transfer amplification occurs at high efficiency compared to the efficiency of PB or AQ transfer reactions during the first few cycles of Bi-PASA. However, as the reaction progresses, PB and AQ are produced at higher efficiency because of self-amplification. Self-amplification occurs at high efficiency because both the complementary and the 10 bp tail regions are incorporated into the products providing a much longer region of complementarity.

Kinetics of Bi-PASA

To quantitate the accumulation of each segment during a Bi-PASA reaction, identical radioactive Bi-PASA reactions were removed from the thermocycler every three cycles and each segment was quantified by a Phosphorlmager after agarose gel electrophoresis (FIG. 2 ). During cycles 18-30, PQ, PB, and AQ accumulated at remarkably similar rates, although the efficiency of accumulation for PB and AQ is somewhat greater in later cycles. In the samples from homozygous wild type or mutants (WT/WT, M/M), individual yields were up to two-fold greater in comparison to the heterozygote samples, suggesting that there are interactions among either the three segments or primers that inhibit efficiency in the WT/M sample. Thus, the type of genomic template used in the reaction may influence the efficiency of accumulation and the final amount of product produced.

For comparison, the three possible two-primer PCR amplifications were performed on the heterozygote samples. Yields of up to twofold greater were observed for each of the segments as compared to Bi-PASA with four primers. This demonstrates that interactions among primers can influence the yield in Bi-PASA.

EXEMPLIFICATION OF THE INVENTION

Methods

All primers were designed and analyzed with Oligo 4 software (National Biosciences, Inc.). Oligo 4 calculates the melting temperature values of a primer by the nearest neighbor method at 50 mM Kcl and 250 μM DNA. PCR was performed from human genomic DNA isolated from white blood cells. The PCR mixtures contained a total volume of 25 ul:50 mM Kcl, 10 mM Tris-HCl (pH 8.3), 1.5 mM MgCl2, 200 uM of each dNTP, 0.5 units ot Taq DNA polymerase (Boehringer Mannheim), and 100 ng of genomic DNA. The effects of the annealing temperature were explored extensively during the course of this study. Cycling conditions were otherwise constant. The recommended annealing temperature is 20° C. below the T m of segment PQ (see guidelines). Additional PCR cycling conditions were denaturation at 95° C. for 15 sec, annealing for 30 sec, a 1-min ramp time from the annealing temperature to the elongation temperature, and elongation at 72° C. for 2 min. An additional 15 sec of denaturation time was always supplemented during the first cycle of the reaction. Thirty cycles were performed on a Perkin Elmer GeneAmp PCR system 9600. Standard agarose gel electrophoresis with ethidium bromide staining and UV photograhy with Polaroid 667 film was used to visualize all PCR segments.

To quantitate PCR yield, PCR was performed with 5 uCi of [α-32P]dATP per 25 μl reaction (3000 Ci/mmole, Amersham). The reaction was electrophoresced through an agarose gel, dried, and subjected to autoradioagraphy. The PCR yield was quantitated with a PhosphorImager with ImageQuant software (Molecular Dynamics) after a 20 min exposure. The relative PCR yields were quantitated as “random units”, that is, the number of pixels in the PCR band minus the background, indicated as a random unit.

Parameters Affecting Bi-PASA

The parameters important for optimizing Bi-PASA were investigated in detail for common sequence changes in the human procoagulant Factor V (FV) and catechol-O-methyltransferase (COMT) genes. A G-A transition at base pair 266 in exon 10 of the FV gene (Genbank Accession L32764 J05368) is a mutation associated with venous thromboembolism. The polymorphism in the COMT gene is a G-A transition at base pair 1947 in exon 4 (GenBank Accession Z26491). These genes were chosen partly on the basis of the large differences in G+C content (FV, 40% G+C; COMT, 63% G+C).

Table 1 lists primers designed to examine differences in mismatch position, T., of the complementary region and tail composition. In the FV gene, the position of the 3′ mismatch at the −1, −2, or −3 positions did not have a noticeable effect on primer efficiency for the A primers (compare yields using A4, A6, and A7).

The T m and tail composition were both important factors in designing the inner primers. Primers in both the FV and COMT genes were much more efficient at higher T m (see Table 1 for details). The tail sequence was an important factor in primer design. Several of those with a tail of G10 were capable of priming amplification. However, primers with G9C1 or G8C2 tails were much more efficient in the Bi-PASA reactions (e.g., FV primers A8-A10 and B10 and B11).

TABLE 1
Selected Primers for the Initial
Optimazation of Fv and COMT Genes
	Sequence		T m
Name a	Numbers are Seq ID Nos.	WT/M	(° C.)	Tail	Yld b	c
FV gene
P F5(25)-22D	TGCAAATGAAAACAATTTTGAA 2		50.6		+++	−
Q F5(419)-25U	TATCACACTGGTGCTAAAAAGGACT 3		53.3		+++	−
A1 F5(257)-20D	qqqqqqqqqq TGGACAGGC A d 4	A-1 M	14.5	G10	−	−
A2 F5(256)-21D	qqqqqqqqqq CTGGACAGGC A 5	A-1 M	20.0	G10	−	−
A3 F5(255)-22D	qqqqqqqqqq CCTGGACAGGC A 6	A-1 M	29.6	G10	−	−
A4 FS(254)-23D	qqqqqqqqqq CCCTGGACAGGC A 7	A-1 M	37.4	G10	+	−
A5 FS(253)-24D	qqqqqqqqqq TCCCTGGACAGGC A 8	A-1 M	41.0	G10	+	−
A6 FS(255)-23D	qqqqqqqqqq CCTGGACAGGC A A 9	A-2 M	34.0	G10	+	−
A7 FS(256)-23D	qqqqqqqqqq CTGGACAGGC A AG 10	A-3 M	29.9	G10	+	−
A8 FS(254)-25D	qqqqqqqqqq CCCTGGACAGGC A AG 11	A-3 M	43.3	G10	+	−
A9 FS(254)-23D	qqqqqcqqqq CCCTGGACAGGC A 12	A-1 M	37.4	G9C1	++	−
A10 FS(254)-23D	qcqqqcqqqq CCCTGGACAGGC A 13	A-1 M	37.4	G8C2	+++	−
A11 FS(253)-24D	qcqqqcqqqq TCCCTGGACAGGC A 14	A-1 M	41.0	G8C2	+++	−
B1 F5(276)-20U	qqqqqqqqqq TGTATTCCT C 15	C-1 WT	−3.8	G10	−	−
B2 FS(277)-21U	qqqqqqqqqq CTGTATTCCT C 16	C-7 WT	3.3	G10	−	−
B3 FS(278)-22U	qqqqqqqqqq CCTGTATTCCT C 17	C-1 WT	14.5	G10	−	−
B4 FS(279)-23U	qqqqqqqqqq ACCTGTATTCCT C 18	C-1 WT	18.4	G10	−	−
BS FS(280)-24U	qqqqqqqqqq TACCTGTATTCCT C 19	C-1 WT	20.7	G10	−	−
B6 FS(282)-26U	qqqqqqqqqq AATACCTGTATTCCT C 20	C-1 WT	28.0	G10	−	−
B7 FS(284)-28U	qqqqqqqqqq AAAATACCTGTATTCCT C 21	C-1 WT	34.7	G10	−	−
B8 FS(278)-24U	qqqqqqqqqq CCTGTATTCCT C GC 22	C-3 WT	33.2	G10	−	−
B9 FS(279)-25U	qqqqqqqqqq GACCTGTATTCCT C GC 23	C-3 WT	35.6	G10	−	−
B10 FS(280)-26U	qqqqqqqqqq TACCTGTATTCCT C GC 24	C-3 WT	36.7	G10	+	−
B11 FS(280)-26U	qqqqcqqqcq TACCTGTATTCCT C GC 25	C-3 WT	36.7	G8C2	+++	−
B12 FS(284)-28U	qqqqcqqqcq AAAATACCTGTATTCCT C 26	C-1 WT	34.7	G8C2	+++	−
COMT gene
P CDMT(1706)-25D	ATCCAAGTTCCCCTCTCTCCACCTG 27		61.6		+++	−
Q CGMT(2511)-25U	GTTGGGGCTCACCTCCAAGAGAAGC 28		66.7		+++	−
A1 COMT(1934)-24D	qqqqcqqqqc GGATTTCGCTGGC A 29	A-1 M	41.7	G8C2	+/−	−
A2 CDMT(1932)-26D	qqqqcqqqqc GTGGATTTCGCTGGC A 3C	A-1 M	47.5	G8C2	+++	−
A3 CDMT(193D)-28D	qqqqcqqqqc TGGTGGATTTCGCTGGC A 31	A-1 M	55.6	G8C2	+++	−
A4 CCMT(1928)-3QD	qqqqcqqqqc GATGGTGGATTTCGCTGGC A 32	A-1 M	58.1	G8C2	+++	−
B1 CDMT(1960)-24U	qqqqcqqqcq ACCTTGTCCTTCA C 33	C-1 WT	27.C	G8C2	−	−
B2 COMT(1962)-26U	qqqqcqqqcq ACACCTTGTCCTTCA C 34	C-1 WT	34.7	G8C2	+/−	−
E3 CCMT(1964)-28U	qqqqcqqqcq GCACACCTTGTCCTTCA C 35	C-1 WT	45.0	G8C2	+++	−
B4 CDMT(1966)-3DU	qqqqcqqqcq ATGCACACCTTGTCCTTCA C 36	C-1 WT	49.6	G8C2	+++	+
B5 CCMT(1968)-32U	qqqqcqqqcq GCATGCACACCTTGTCCTTCA C 37	C-1 WT	56.8	G8C2	+++	+
a The sequence of the FV gene was from a revised version of GenBank Accession L32764 J05368 with an additional 127 base sequence in intron 10, as shown in Seq. ID 1. As an example for oligo-nucleotide P, F5 = FV, (25) − 22D = 5′ end of the primer beginning at 25 (numbering as shown above), and the length is 22 bases downstream (D) (i.e., in the direction of transcription). The precise sizes and locations of the PCR fragment can be obtained from the informative names. The#
sequence of the COMT gene was from a revised version of GenBank Accession Z26491, in which a 122 base sequence was inserted after base 2084, as shown in Seq ID 48.
b under optimal PCR conditions, the relative yield of the DNA products as indicated as −, no DNA band(s) on the agarose gel stained by ethidium bromide for UV photograph; +/\, very weak DNA bankd; I, weak DNA band; ++, strong DNA band; +++, very strong DNA band.
c specificity - defined as amplification of only the expected allele. (−) No mismatched amplified PCR DNA with homozygote wild-type and mutant genomic DNA template; (+) false-amplified PCR DNA.
d The lowercase letters indicate noncomplementary and the uppercase letters indicate complementary sequences. Also the nucleotide at the mutation site is in boldface type.
e (A-1M) the first base from the 3′ end is a mutated A nucleotide.

Seq. ID 1: Exon 10 of FV gene (bases 1-319) and 127 base intron sequence (bases 320-446 shown in lower case)

1 TTATTTATTA TCATGAAATA ACTTTGCAAA TGAAAACAAT TTTGAATATA

51 TTTTCTTTCA GGCAGGAACA ACACCATGAT AGTTTGATGA ACCCACAGAA

101 AAACCTATAC TTATAAGTGG AACATCTTAG AGTTTGATGA ACCCACAGAA

151 AATGATGCCC AGTGCTTAAC AAGACCATAC TACAGTGACG TGGACATCAT

201 GAGAGACATC GCCTCTGGGC TAATAGGACT ACTTCTAATC TGTAAGAGCA

251 GATCCCTGGA CAGGCGAGGA ATACAGGTAT TTTGTCCTTG AAGTAACCTT

301 TCAGAAATTC TGAGAATTTC ttctggctag aacatgttag gtctcctggc

351 taaataatgg ggcatttcct tcaagagaac agtaattgtc aagtagtcct

401 ttttagcacc agtgtgataa catttattct tttttttttt ttgtct

Seq. ID 48: Bases 1701-2520 of COMT gene, a revised version of Genbank Accession Z26491 in which a 122 base region in intron 4 (bases 2085-2206, shown in lower case) is inserted.

1701 TGGGGATCCA AGTTCCCCTC TCTCCACCTG TGCTCACCTC TCCTCCGTCC

1751 CCAACCCTGC ACAGGCAAGA TCGTGGACGC CGTGATTCAG GAGCACCAGC

1801 CCTCCGTGCT GCTGGAGCTG GGGGCCTACT GTGGCTACTC AGCTGTGCGC

1851 ATGGCCCGCC TGCTGTCACC AGGGGCGAGG CTCATCACCA TCGAGATCAA

1901 CCCCGACTGT GCCGCCATCA CCCAGCGGAT GGTGGATTTC GCTGGCGTGA

2001 GCTGTGGGCA GGGCGGGCAT GCGCACTTTG ATCCTCCCCA CCAGGTGTTC

2051 ACACCACGTT CACTGAAAAC CCACTATCAC CAGGcccctc agtgcttccc

2101 agcctggggc tgaggaaaga cccccccagc agctcagtga gggtctcaca

2151 gctctgggta aactgccaag gtggcaccag gaggggcagg gacagagtgg

2201 ggccttGTCA TCCCAGAACC CTAAAGAAAA CTGATGAATG CTTGTATGGG

2251 TGTGTAAAGA TGGCCTCCTG TCTGTGTGGG CGTGGGCACT GACAGGCGCT

2301 GTTGTATAGG TGTGTAGGGA TGGCCTCCTG TCTGTGAGGA CGTGGGCACT

2351 GACAGGCGCT GTTCCAGGTC ACCCTTGTGG TTGGAGCGTC CCAGGACATC

2401 ATCCCCCAGC TGAAGAAGAA GTATGATGTG GACACACTGG ACATGGTCTT

2451 CCTCGACCAC TGGAAGGACC GGTACCTGCC GGACACGCTT CTCTTGGAGG

2501 TGAGCCCCAA CCAGGATGGC

Interactions Among Primers

FIG. 3 illustrates Bi-PASA reactions for the COMT and FV genes. Various combinations of the primers listed in Table 1 were used to identify the optimal set of inner primers and reaction conditions for these genes (the optimal primers and conditions can be found in the legend to FIG. 3 ). To visualize the interactions among the primers in Bi-PASA, reactions were performed with different sets of the four optimal primers on the three genomic templates. FIG. 3A shows the Bi-PASA reaction for the COMT gene. Interactions among the primers affect both the yield and the specificity of the reactions. In general, the greater the number of primers added to the reaction, the lower the yield is of the individual segments (cf. lanes 2, 4, and 6). The specificity of the inner primers to a particular allele can also be affected because of the competition between the primers. Lane 8 shows a faint false-positive signal for the nonspecific allele for the two-primer combination. However, the nonspecificity disappears when more primers are added to the reaction as in lanes 10 and 12. FIG. 3B shows the Bi-PASA reaction for the FV gene. Specificity was not a problem, but again the yield was affected by the number of primers in the reaction.

Primer concentrations were optimized for the COMT and FV Bi-PASA reactions. The outer and inner primer concentrations were adjusted to the minimum optimal point at which the wild-type and mutant alleles could be amplified specifically and efficiently (see FIG. 3 ). FIG. 4 shows the effects on the FV Bi-PASA reaction when the concentrations of each individual primer and then all four primers were varied. A change in the concentration of a primer can affect all of the segments produced in the reaction, for example, FIG. 4 , lanes 1-6, demonstrates that when the concentratin of the P primer is varied, not only do the yields of the PQ and PB segments vary (which would be expected) but the AQ segment is also affected. This demonstrates again that interactions among the primers are an important factor in a Bi-PASA reaction.

Blinded Analysis

To test the sensitivity and specificity of Bi-PASA under actual screening conditions, a blinded analysis of the COMT gene was performed. For the common COMT polymorphism, 249 samples were screened independently by RFLP analysis and Bi-PASA. The sensitivity and specificity were 100% and 99.6%, respectively; the one false-positive result was attributable to contamination by the neighboring well of a gel.

The sensitivity of Bi-PASA was also tested by diluting the mutant allele into the wild-type allele. Detection of the mutant allele by staining with ethidium bromide was possible with a 20- to 40-fold excess of the wild-type allele (data not shown).

Guidelines for Primer Design

Reactions were designed successfully to detect sequence changes in the FV and COMT genes. For Bi-PASA to be a useful tool, optimization needs to be relatively rapid and simple. On the basis of experience gained by analysis of multiple primers and optimization parameters, guidelines for primer design and a strategy for reaction optimization were developed. Although subsequent experiments support the utility of the guidelines and strategy, the folowing guidelines will liekly evolve with time. Desiging the primers is a critical step in successful Bi-PASA reaction. The T m value of each PCR segment was estimated by the formula of Wetmur (1991): T m product =81.5+16.6 log[K + ]+0.41(%G+%C)−675/length.

1. The T m value of P and Q (outer primers) should be 20-25° C. lower than that of the PQ segment. A good size for the PQ segment is about 300-1000 bp, although the method can be used with much longer PQ segments. For the COMT gene, the T m value of the PQ PCR product was 83.8° C. For the FV gene, the T m value of the PQ PCR product was 74.8° C. The T m value of successful P and Q primers were 61.6° C. and 66.7° C. for COMT gene, and 50.6° C. and 53.3° C. for FV gene, respectively. When P and Q were designed with a lower T m value, the interactions among the four primers were more pronounced, especially for regions of high G+C content (data not shown).

2. The T m values of the complementary region of inner primers A and B should be −35° C. lower than that of the PQ segment. Parameters were optimized in the COMT and FV genes by systematically increasing the length of the complementary regions of the inner primers. It was observed that once the T m value of the inner primers reached a certain point below the T m of the PQ PCR products, the inner primers began to work with high yield and specificity (see Table 1). For the COMT gene, this “'threshold” T m value was 47.5° C. in A2 and 45.O° C. in B3, which were 36.3° C. and 38.8° C. below the T m value of the PQ PCR DNA product, respectively. Beyond that point, T m was 55.6° C. in A3, 58.1° C. in A4, 49.6° C. in B4, and 56.8° C. in B5, all having produced high-yield AQ and BP segments, respectively. However, B4 and B5 also amplified a faint false-positive segment when M/M genomic DNA was used, indicating that as primer length increases, some specificity may be sacrificed. In the FV gene, the threshold T m was 37.4° C. for A4 and 34.7° C. for B7, which is 37.4° C. and 40.1° C. below the T m of the PQ segment, respectively. The A and B inner primers were T m values above this point also amplified with high yield and specificity (see Table 1).

3. The annealing temperature should be 20° C. below the T m value of the PQ segment in these studies. The optimal annealing temperature also was closely linked to the T m value of the PQ PCR products. The annealing temperature was set relatively high to avoid hybridization among the multiple PCR segments. For the analyzed regions, the recommended annealing temperatures happen to be within 2° C. of the annealing temperatures estimated by the formula of Rychlik et al. (1990), T m annealing =0.3×T m P or Q primer +0.7×T m PQ product −14.9.

In the COMT gene, good results could be achieved only within a narrow range of annealing temperatures. When the annealing temperature was set at 55° C. or 60° C., the yield and specificity were poor in comparison to the recommended temperature of 65° C. However, good results were achieved for the FV gene with annealing temperatures ranging from 50° C.-60° C. (data not shown).

Strategy of Optimation of Bi-PASA

1. Design the primers based on the above guidelines.

2. Alter the concentration of P and Q. This seems to have more effect on the yields of all the segments than altering the concentration of the inner primers A and B. Optimize the PQ segment by titrating oligonucleotide concentrations. Use the lowest concentration of primer. necessary to obtain a high yield. For regions of high G+C content, DMSO may be used.

3. Determine the range of oligonucleotide concentrations with robust specification that are compatible with primers P and B and perform a similar analysis with primers A and Q. Then adjust the concentrations of the inner primers such that appropriate amplifications with three primers (PQB and PQA) or four primers (PQBA) occur (see FIGS. 3 and 5 ). If it is not possible to achieve acceptable yield but the specificity is good, increase the length of the complementary regions of the inner primers and reoptimize. If specificity is not optimal, decrease the length of the inner primers or alter the position of the mismatch by 1 bp or more. This should be performed for WT/WT, WT/M, and M/M samples.

4. Once the above conditions have been satisifed, check the sensitivity and specificity of the reaction. Dilute the DNA samples to determine the minimum amount of template necessary to perform the reaction. Dilute the M sample into the WT sample and vice versa to determine the maximum ratio of WT to M that can be detected.

Prospective Test of the Guidelines and Strategy

To test the utility of the guidelines and strategy, three Bi-PASA reactions were designed (Table 2) and tested for (1) the TaqI polymorphism in the factor IX gene, (2) the FV Leiden mutation assayed in a direction opposite to that performed initially, and (3) a polymorphism in the D1 dopamine receptor genes, respectively. All the reactions were successful and rapidly optimized by utilizing the above primer guidelines and optimization strategy (FIG. 5 ).

TABLE 2
List of Primers in the Prospective Study
	Sequence
Name a	Numbers are Seq. ID Nos.	WT/M	T m (° C.)	Tail	Yield b	Specificity c
FIX a
P F9(10545)-24D	TCCCTCTGAAACAAGTTGAAACTG 38		54.3		+++	−
Q E9(11446)-23U	TCATCTGTCAGGTGGATTGCTGT 39		55.9		+++	−
A F9(11095)-30D	ggggcggggc TGACTATATTGATTACATCG 40	G-1 WT	37.3	G8C2	+++	−
B E9(11129)-28U	ggggcgggcg GCTATGTAACATTTTTGA 41	T-3 M	34.8	G8C2	+++	−
FV(opp) c :	Numbers are Seq. ID Nos.
A F5(254)-23D	gcgggcgggg CCCTGGACAGGCC 42	C-1 WT	37.4	G8C2	+++	−
B F5(280)-26U	ggggcgggcg TACCTGTATTCCTTGC 43	C-3 M	33.5	G8C2	+++	−
D1 d :
P D1(33)-24D	GACCTGCAGCAAGGGAGTCAGAAG 44		60.3		+++	−
Q D1(544 )-24U	AGAAGCCAGCAATCTCAGCCACTG 45		60.7		+++	−
A D1(213)-27D	gggggcgcgt CCCCTATTCCCTGCTTG 46	G-1 M	47.5	G6C3T1	+++	−
B D1(299)-28U	ggggcggg CgGACACCCCTCAAGTTCCT 47	T-1 WT	45.4	G8C2	+++	−
a See Table 1 footnote for details on primer sequence, etc.
b The sequence of the FIX gene was numbered on the basis of Yoshitake et al. (1985).
c The P and Q primers for FV (opp) were the same as used in FV (see Table 1).
d The sequence of the D1 gene was submitted to GenBank under accession no. X55760.Gb_Pr.

REFERENCES

Dutton, C. And S. S. Sommer. 1991. Simultaneous detection of multiple single-base alleles at a polymorphic site. BioTechniques 11:700-702

Gibbs, R. A., P. Nguyen, and C. T. Caskey. 1989. Detection of single DNA base differences by competitive oligonucleotide priming. Nucleic Acids Res. 17:2437-2448.

Newton, C. R., A. Graham, L. E. Heptinstall, S. J. Powell, C. Summers, N. Kalsheker, J. G. Smith, and A. F. Markham. 1989. Analysis of any point mutation in DNA. The amplification refactory mutation system (ARMS). Nucleic Acids Res. 17:2503-2516.

Nichols, W. C., J. J. Liepnieks, V. A. McKusick, and M. D. Benson. 1989. Direct sequencing of the gene for Maryland/German familial amyloidotic polyneuropathy type II and genotyping by allele-specific enzymatic amplification. Genomics 5:535-540.

Ruano, G. and K. K. Kidd. 1989. Direct haplotyping of chromosomal segments from multiple heterozygotes via allele-specific PCR amplification. Nucleic Acids Res. 17:8392.

Rychlik, W., W. J. Spencer, and R. E. Rhoads. 1990. Optimization of the annealing temperature of DNA amplification in vitro. Nucleic Acids Res. 18:6409-6412.

Sarkar, G. and S. S. Sommer. 1990. The “megaprimer” method of site-directed mutagenesis. BioTechniques 8:404-407.

Sarkar, G and S. S. Sommer 1992. Double-standard DNA segments can efficiently prime the amplification of human genomic DNA. Nucleic Acids Res. 20:4937-4938.

Sarkar, G., J. Cassady, C. D. K. Bottema, and S. S. Sommer. 1990. Characterization of polymerase chain reaction amplification of specific alleles. Anal. Biochem. 186:64-68.

Sommer, S. S., J. D. Cassady, J. L. Sobell, and C. D. K. Bottema. 1989. A novel method for detecting point mutations or polymorphisms and its application to population screening for carriers of phenylketonuria. Mayo Clin. Proc. 64:1361-1372.

Sommer, S. S. A. R. Groszbach, and C. D. K. Bottema. 1992. PCR amplification of specific alleles (PASA) is a general method for rapidly detecting known single-base changes. BioTechniques 12:82-87.

Wetmur, J. G. 1991. DNA Probes: Application of the principles of nucleic acid hybridization. Crit. Rev. Biochem. Mol. Biol. 26:227-259.

Wu, D. Y., L. Ugozzoli, B. K. Pal, and R. B. Wallace. 1989. Allele-specific enzymatic amplification of β-globin genomic DNA for diagnosis of sickle cell anemia. Proc. Natl. Acid. Sci. 86:2757-2760.

Ye, S., S. Humphries, and F. Green. 1992. Allele specific amplification by tetra-primer PCR. Nucleic Acids Res. 20:1152.

Yositake, S., B. G. Schach, D. C. Foster, E. W. Davie, and K. Kuarchi. 1985. Nucleotide sequence of the gene for human factor IX (anti-hemophiliac factor B). Biochemistry 24:3736-3750.

48 1 446 DNA Homo sapiens 1ttatttatta tcatgaaata actttgcaaa tgaaaacaat tttgaatata ttttctttca 60ggcaggaaca acaccatgat agtttgatga acccacagaa aaacctatac ttataagtgg 120aacatcttag agtttgatga acccacagaa aatgatgccc agtgcttaac aagaccatac 180tacagtgacg tggacatcat gagagacatc gcctctgggc taataggact acttctaatc 240tgtaagagca gatccctgga caggcgagga atacaggtat tttgtccttg aagtaacctt 300tcagaaattc tgagaatttc ttctggctag aacatgttag gtctcctggc taaataatgg 360ggcatttcct tcaagagaac agtaattgtc aagtagtcct ttttagcacc agtgtgataa 420catttattct tttttttttt ttgtct 446 2 22 DNA Homo sapiens 2tgcaaatgaa aacaattttg aa 22 3 25 DNA Homo sapiens 3tatcacactg gtgctaaaaa ggact 25 4 20 DNA Homo sapiens 4gggggggggg tggacaggca 20 5 21 DNA Homo sapiens 5gggggggggg ctggacaggc a 21 6 22 DNA Homo sapiens 6gggggggggg cctggacagg ca 22 7 23 DNA Homo sapiens 7gggggggggg ccctggacag gca 23 8 24 DNA Homo sapiens 8gggggggggg tccctggaca ggca 24 9 23 DNA Homo sapiens 9gggggggggg cctggacagg caa 23 10 23 DNA Homo sapiens 10gggggggggg ctggacaggc aag 23 11 25 DNA Homo sapiens 11gggggggggg ccctggacag gcaag 25 12 23 DNA Homo sapiens 12gggggcgggg ccctggacag gca 23 13 23 DNA Homo sapiens 13gcgggcgggg ccctggacag gca 23 14 24 DNA Homo sapiens 14gcgggcgggg tccctggaca ggca 24 15 20 DNA Homo sapiens 15gggggggggg tgtattcctc 20 16 21 DNA Homo sapiens 16gggggggggg ctgtattcct c 21 17 22 DNA Homo sapiens 17gggggggggg cctgtattcc tc 22 18 23 DNA Homo sapiens 18gggggggggg acctgtattc ctc 23 19 24 DNA Homo sapiens 19gggggggggg tacctgtatt cctc 24 20 26 DNA Homo sapiens 20gggggggggg aatacctgta ttcctc 26 21 28 DNA Homo sapiens 21gggggggggg aaaatacctg tattcctc 28 22 24 DNA Homo sapiens 22gggggggggg cctgtattcc tcgc 24 23 25 DNA Homo sapiens 23gggggggggg acctgtattc ctcgc 25 24 26 DNA Homo sapiens 24gggggggggg tacctgtatt cctcgc 26 25 25 DNA Homo sapiens 25ggggcgggcg taccgtattc ctcgc 25 26 28 DNA Homo sapiens 26ggggcgggcg aaaatacctt gattcctc 28 27 25 DNA Homo sapiens 27atccaagttc ccctctctcc acctg 25 28 25 DNA Homo sapiens 28gttggggctc acctccaaga gaagc 25 29 24 DNA Homo sapiens 29ggggcggggc ggatttcgct ggca 24 30 26 DNA Homo sapiens 30ggggcggggc gtggatttcg ctggca 26 31 28 DNA Homo sapiens 31ggggcggggc tggtggattt cgctggca 28 32 30 DNA Homo sapiens 32ggggcggggc gatggtggat ttcgctggca 30 33 24 DNA Homo sapiens 33ggggcgggcg accttgtcct tcac 24 34 26 DNA Homo sapiens 34ggggcgggcg acaccttgtc cttcac 26 35 28 DNA Homo sapiens 35ggggcgggcg gcacaccttg tccttcac 28 36 30 DNA Homo sapiens 36ggggcgggcg atgcacacct tgtccttcac 30 37 32 DNA Homo sapiens 37ggggcgggcg gcatgcacac cttgtccttc ac 32 38 24 DNA Homo sapiens 38tccctctgaa acaagttgaa actg 24 39 23 DNA Homo sapiens 39tcatctgtca ggtggattgc tgt 23 40 30 DNA Homo sapiens 40ggggcggggc tgactatatt gattacatcg 30 41 28 DNA Homo sapiens 41ggggcgggcg gctatgtaac atttttga 28 42 23 DNA Homo sapiens 42gcgggcgggg ccctggacag gcc 23 43 26 DNA Homo sapiens 43ggggcgggcg tacctgtatt ccttgc 26 44 24 DNA Homo sapiens 44gacctgcagc aagggagtca gaag 24 45 24 DNA Homo sapiens 45agaagccagc aatctcagcc actg 24 46 27 DNA Homo sapiens 46ggcggcgcgt cccctattcc ctgcttg 27 47 28 DNA Homo sapiens 47ggggcgggcg gacacccctc aagttcct 28 48 820 DNA Homo sapiens 48tggggatcca agttcccctc tctccacctg tgctcacctc tcctccgtcc ccaaccctgc 60acaggcaaga tcgtggacgc cgtgattcag gagcaccagc cctccgtgct gctggagctg 120ggggcctact gtggctactc agctgtgcgc atggcccgcc tgctgtcacc aggggcgagg 180ctcatcacca tcgagatcaa ccccgactgt gccgccatca cccagcggat ggtggatttc 240gctggcgtga aggacaaggt gtgcatgcct gacccgttgt cagacctgga aaaagggccg 300gctgtgggca gggcgggcat gcgcactttg atcctcccca ccaggtgttc acaccacgtt 360cactgaaaac ccactatcac caggcccctc agtgcttccc agcctggggc tgaggaaaga 420cccccccagc agctcagtga gggtctcaca gctctgggta aactgccaag gtggcaccag 480gaggggcagg gacagagtgg ggccttgtca tcccagaacc ctaaagaaaa ctgatgaatg 540cttgtatggg tgtgtaaaga tggcctcctg tctgtgtggg cgtgggcact gacaggcgct 600gttgtatagg tgtgtaggga tggcctcctg tctgtgagga cgtgggcact gacaggcgct 660gttccaggtc acccttgtgg ttggagcgtc ccaggacatc atcccccagc tgaagaagaa 720gtatgatgtg gacacactgg acatggtctt cctcgaccac tggaaggacc ggtacctgcc 780ggacacgctt ctcttggagg tgagccccaa ccaggatggc 820

Claims

1. A method for conducting a bi-directional PCR amplification of specific alleles which comprises subjecting DNA which may contain one or both of first and second alleles to a single one tube PCR under a thermal cycling condition utilizing an outer pair of primers P and Q and an inner pair of primers A and B, whereQ is complementary to the sense strand of both alleles in a region downstream of the sequence difference X which distinguishes the alleles and P is complementary to the anti-sense strand of both alleles in a region upstream of X; B has a region at its 3′ end which is complementary to the sense strand of the first allele and A has a region at its 3′ end which is complementary to the antisense strand of the second allele, each of A and B also has a non-complementary G+C-rich tail at its 5′ end, and X occurs at or near the 3′ end of each of A and B, whereby A and B are in two directions with partially overlapped 3′ ends; whereby DNA which is heterozygous with respect to the two alleles results in simultaneous amplification of three overlapping sequences, PQ, PB and AQ, DNA which is homozygous with respect to the first allele results in simultaneous amplification of two overlapping sequences, PQ and PB, and DNA which is homozygous with respect to the second allele results in simultaneous amplification of two overlapping sequences, PQ and AQ, where PQ, PB and AQ, respectively, stand for the sequences extending from P to Q inclusive, from P to B inclusive, and from A to Q inclusive, wherein the method is conducted at a substantially constant annealing temperature in the range of about 50-65° C. and with a ratio of concentrations of the outer and inner primers in the PCR mixture in the range of about 1:1 to 1:6.

2. Method of claim 1 wherein X is a base substitution, deletion or insertion.

3. Method of claim 2 wherein the the complementary regions of both primers A and B are about 10 to 20 nucleotides in length, the G+C-rich tails of both primers A and B are about 6 to 20 nucleotides in length, and the G+C-rich tails contain about 50 to 100% G+C.

4. A method for conducting a bi-directional PCR amplification of specific alleles in DNA which may contain one or both of first and second alleles, one of which is a mutant allele containing a point mutation X and one of which is a wild type allele, which comprises subjecting the DNA to a single one-tube PCR under a thermal cycling condition utilizing an outer pair of primers P and Q and an inner pair of primers A and B, whereQ is complementary to the sense strand of both alleles in a region downstream of X and P is complementary to the anti-sense strand of both alleles in a region upstream of X; B has a region at its 3′ end which is complementary to the sense strand of the first allele and A has a region at its 3′ end which is complementary to the antisense strand of the second allele, each of A and B also has a non-complementary G+C-rich tail at its 5′ end, and X occurs at or near the 3′ end of each of A and B, whereby A and B are in two directions with partially overlapped 3′ ends; whereby DNA which is heterozygous with respect to the two alleles results in simultaneous amplification of three overlapping segments, PQ, PB and AQ, DNA which is homozygous with respect to the first allele results in simultaneous amplification of two overlapping segments, PQ and PB, and DNA which is homozygous with respect to the second allele results in simultaneous amplification of two overlapping segments, PQ and AQ, where PQ, PB and AQ, respectively, stand for the segments extending from P to Q inclusive, from P to B inclusive, and from A to Q inclusive, wherein the method is conducted at a substantially constant annealing temperature in the range of about 50-65° C. and with a ratio of concentrations of the outer and inner primers in the PCR mixture in the range of about 1:1 to 1:6.

5. Method of claim 4 wherein the annealing temperature is about 20-25° C. below the Tm of the segment PQ.

6. Method of claim 5 wherein the Tm values of outer primers P and Q are about 20-25° C. lower than that of the PQ segment.

7. Method of claim 6 wherein the Tm values of the complementary regions of the inner primers A and B are about 35° C. lower than that of the PQ segment.

8. Method of claim 1 wherein 3 or more inner primers are used to differentiate 3 or more alleles.

9. Method of claim 8 wherein 3 or more inner primers are used to differentiate 3 or more alleles.

10. Method of claim 1 wherein a uniform annealing temperature is used to conduct the method on 2 or more loci.

11. Method of claim 8 wherein a uniform annealing temperature is used to conduct the method on 2 or more loci.

12. Method of claim 1 wherein B and A primers have the same or different 5′ tails.

13. Method of claim 1 wherein the tail of primer B or A forms a stem-loop structure within its sequence.