2, TATAA Biocenter, Gothenburg
3, Lund University, Lund, Sweden
Currently, in real-time PCR, one often has to choose between using a sequence-specific probe and a nonspecific double-stranded DNA (dsDNA) binding dye for the detection of amplified DNA products. The sequence-specific probe has the advantage that it only detects the targeted product, while the nonspecific dye has the advantage that melting curve analysis can be performed after completed amplification, which reveals what kind of products have been formed. Here we present a new strategy based on combining a sequence-specific probe and a nonspecific dye, BOXTO, in the same reaction, to take the advantage of both chemistries. We show that BOXTO can be used together with both TaqMan® probes and locked nucleic acid (LNA) probes without interfering with the PCR. The probe signal reflect formation of target product, while melting curve analysis of the BOXTO signal reveals primer-dimer formation and the presence of any other anomalous products.
During recent years, the use of realtime PCR to quantify nucleic acids has grown extensively. It is currently the most sensitive method to determine the amount of specific DNA sequences in complex biological samples and has been applied in such different areas as cancer diagnostics (1,2), food and veterinary diagnostics (3,4), and basic genomic research (5). Detection can be performed by using sequence-specific probes such as TaqMan® probes (6), molecular beacons (7), hybridization probes (8), locked nucleic acid (LNA) probes (www.biotechniques.com/article.asp?id = 91200423), and LightUp® probes (9). Alternatively, free dyes that bind to double-stranded DNA (dsDNA) such as SYBR® Green I (10) and BEBO (11) can be used. When using sequence-specific probes, only the targeted product is detected, while all dsDNA contributes to the fluorescence signal of DNA binding dyes. In addition to the correct PCR product, this includes primer-dimers and other nonspecific amplification products. Primer-dimer formation is a major problem when quantifying low amounts of DNA due to competition between the two amplification reactions, which often results in a reduced sensitivity of the assay. DNA binding dyes reflect formation of both the correct product and the primer-dimers and may overestimate the target DNA concentration. When sequence-specific probes are used, primer-dimer products are not seen, but they may still form and interfere with the amplification reaction of the targeted product. An advantage of DNA binding dyes is that melting curve analysis can be performed after the amplification in a closed-tube format to identify the PCR products that have been formed (12). This is not possible when only using sequence-specific probes. In melting curve analysis, the temperature is gradually increased while monitoring the fluorescence. When the temperature is reached that melts the dsDNA, the dye is released, and its fluorescence decreases. The melting temperature depends on the sequence of the DNA, but most importantly, on the length of the dsDNA product. Hence, different PCR products may be distinguished by melting curve analysis.
The most common fluorophore used in probes is FAM, which is a derivative of fluorescein (13). FAM has an excitation maximum at 493 nm and an emission maximum at 525 nm. SYBR Green I, which is the most popular DNA binding dye for real-time PCR, has almost identical maxima, 497 and 521 nm, respectively, and the FAM and SYBR Green I spectra overlap extensively. Their responses are measured in the same detection channel or using the same filter settings in the real-time PCR instruments (14), making it impossible to distinguish the signals from these dyes. A new dsDNA binding dye, BOXTO, was recently developed (15). BOXTO is an asymmetrical cyanine dye that is similar to SYBR Green I (16), but with an excitation maximum at 515 nm and an emission maximum at 552 nm. These spectral properties are distinct from FAM, and BOXTO can be used in the same reaction as FAM without interfering spectroscopically with it by using appropriate detection channels or filters.
In this work, we demonstrate the novel strategy of combining a nonspecific dsDNA binding dye (BOXTO) with either a TaqMan or an LNA probe in the same reaction. The sequence-specific probe is used to quantify target DNA molecules, while the nonspecific BOXTO is used for melting curve analysis to reveal formation of nonspecific PCR products. The whole process is performed in a homogenous solution in closed reaction containers. The approach is demonstrated on three PCR systems with varying amounts of primer-dimers and erroneous PCR products. One assay was designed for the amyloid β (A4) precursor protein gene (amyloid assay), one for the Escherichia coli β-glucoronidase gene (gusA assay), and one for the nanog gene (nanog assay).Materials and Methods Primers, Probes, and Template
Primer sequences for the amyloid assay were 5′-ATGGGTTGGCCGCT-TCTTTG-3′ and 5′-GCTTGTT-GAGACAGCAAAACCTC-3′, which produce a 133-bp product. The TaqMan probe sequence was 5′-CCTGCGCCTTGCTCCTTT-GGTTCG-3′ and was labeled with FAM at the 5′ end and Black Hole Quencher™ 1 (Biosearch Techniologies, Novato CA, USA) at the 3′ end. Primer sequences for the gusA and nanog assays were 5′-CTCTTTGATGTGCT-GTGCCT-3′ (gusA forward) and 5′-ACATATCCAGCCATG-CACAC-3′ (gusA reverse), which produce a 223-bp product. The nanog assay was run as previously described (17). Sequence matching LNA probes (number 51 for gusA and number 69 for nanog) from the Human ProbeLi-brary™ (Exiqon Vedbaek, Denmark) were used. The probes were labeled with fluorescein and a nonfluorescent quencher according to the manufacturer. All measurements for amyloid and gusA were performed using purified PCR products as a template, while cDNA from human stem cells was used for the nanog assay, as described elsewhere (17).