Heteropolymeric triplex-based genomic assay to detect pathogens or single-nucleotide polymorphisms in human genomic samples

Abstract

Human genomic samples are complex and are considered difficult to assay directly without denaturation or PCR amplification. We report the use of a base-specific heteropolymeric triplex, formed by native duplex genomic target and an oligonucleotide third strand probe, to assay for low copy pathogen genomes present in a sample also containing human genomic duplex DNA, or to assay human genomic duplex DNA for Single Nucleotide Polymorphisms (SNP), without PCR amplification. Wild-type and mutant probes are used to identify triplexes containing FVL G1691A, MTHFR C677T and CFTR mutations. The specific triplex structure forms rapidly at room temperature in solution and may be detected without a separation step. YOYO-1, a fluorescent bis-intercalator, promotes and signals the formation of the specific triplex. Genomic duplexes may be assayed homogeneously with single base pair resolution. The specific triple-stranded structures of the assay may approximate homologous recombination intermediates, which various models suggest may form in either the major or minor groove of the duplex. The bases of the stable duplex target are rendered specifically reactive to the bases of the probe because of the activity of intercalated YOYO-1, which is known to decondense duplex locally 1.3 fold. This may approximate the local decondensation effected by recombination proteins such as RecA in vivo. Our assay, while involving triplex formation, is sui generis, as it is not homopurine sequence-dependent, as are "canonical triplexes". Rather, the base pair-specific heteropolymeric triplex of the assay is conformation-dependent. The highly sensitive diagnostic assay we present allows for the direct detection of base sequence in genomic duplex samples, including those containing human genomic duplex DNA, thereby bypassing the inherent problems and cost associated with conventional PCR based diagnostic assays.

Figure 1. Detection of Bacillus globigii genomic dsDNA at low copy number.

100 copies of BG genomic dsDNA (0.46 pg) were reacted with 3.2 pmoles of 25-mer ssDNA probe in the presence of 0.5×TBE, 40 mM TMA-Cl, and 300 nM YOYO-1 in either the presence or absence of 2 ng of human genomic dsDNA (approximately 302 copies). Fluorescent emissions of the reaction mixtures (80 µl) were monitored with the Genexus® Analyzer at a setting of 32% PMT after 5, 15, 25, 35, 45, 55 and 65 minutes of incubation at RT. Intensity of triplex-associated fluorescence is plotted as a function of incubation time for the BG triplexes formed in the presence of either 300 nM YOYO-1 (1) or 300 nM YOYO-1 and human genomic dsDNA (2).

Figure 2. Assay of drosophila genomic dsDNA for the ts1 mutation in the erg gene.

Fifty ng of either wild-type drosophila genomic dsDNA (wt gDNA) or mutant drosophila genomic dsDNA (ts1 gDNA) were reacted at RT for 5 min with 4 pmoles of wild-type probe 2 in the presence of 0.5×TBE and 500 nM YOYO-1. Fluorescent emissions of the reaction mixtures (100 µl) were monitored with the Genexus® Analyzer at a setting of 30% PMT after 5, 15, 30, 45, 60, 75 and 90 minutes of incubation. The intensity of fluorescence is plotted as a function of incubation time for each sample analyzed. The samples consist of YOYO-1 only with no DNA present (1), wt gDNA with no added ssDNA probe (2), ts1 gDNA with no added ssDNA probe (3), ssDNA probe 2 (4), perfectly matched triplex (5) and 1 bp A–A mismatched triplex (6), as indicated.

Figure 3. Assay of CFTR and FVL mutations in human genomic dsDNA.

Human genomic dsDNA was extracted from blood (A, B and D) or saliva (C) as described in the text. (A) 500 pg of wild-type human genomic dsDNA (approximately 75 copies) was reacted at RT with 3.2 pmoles of either wild-type or mutant ssDNA probe in the presence of 0.5×TBE and 600 nM YOYO-1. (B–D) 1 ng, 2 ng or 200 pg of wild-type human genomic dsDNA (approximately 151 copies, 302 copies or 30 copies, respectively) was reacted at RT with 3.2 pmoles of either wild-type or mutant ssDNA probe in the presence of 0.5×TBE and 500 nM YOYO-1. (A–D) Reaction mixtures (80 µl) were irradiated as described in the text and analyzed for fluorescent emission. The intensity of triplex-associated fluorescence is plotted as a function of incubation time for each sample analyzed. The samples consist of perfectly matched triplex (1) and mismatched triplex (2) as indicated for CFTR delta F508 (A), CFTR 3659delC (B), CFTR 2789+5G→A (C) and FVL G1691A (D).

Figure 4. Assay of MTHFR C677T in homozygous and heterozygous human genomic dsDNA samples.

Human genomic dsDNA that was either wild-type homozygous, mutant heterozygous or mutant homozygous with respect to MTHFR C677T, was extracted from blood as described in the text. One ng of human genomic dsDNA (approximately 151 copies) was reacted at RT with 3.2 pmoles of either wild-type or mutant ssDNA probe in the presence of 0.5×TBE and 500 nM YOYO-1. Reaction mixtures (80 µl) were irradiated as described in the text and analyzed for fluorescent emission. The intensity of triplex-associated fluorescence is plotted as a function of incubation time for each sample analyzed. The samples consist of perfectly matched triplexes (1 and 4), mutant heterozygous triplexes (3) and mismatched triplexes (2 and 5) as indicated for MTHFR C677T.

Figure 5. Waxing and waning of triplex-associated fluorescent emissions can enhance a homogeneous in solution assay.

(A, B) Human genomic dsDNA was extracted from blood as described in the text. Two ng of wild-type human genomic dsDNA (approximately 302 copies) was reacted at RT with 3.2 pmoles of either wild-type or mutant ssDNA probe in the presence of 0.5×TBE, 40 mM TMA-Cl and 500 nM YOYO-1. (A) YOYO-1 was added last to the reaction mixtures. (B) Human genomic dsDNA was added last to the reaction mixtures. (A, B) Reaction mixtures (80 µl) were irradiated as described in the text and analyzed for fluorescent emission. The intensity of triplex-associated fluorescence is plotted as a function of incubation time for each sample analyzed. The samples consist of perfectly matched triplex (1) and mismatched triplex (2) as indicated for CFTR 3849+10kbC→T.