Real-time quantitative PCR is a sensitive and accurate method for gene expression studies (1). The detection chemistries of all real-time PCR procedures are based on one of two principles for monitoring amplification products: binding to double-stranded DNA or hybridization to single-stranded DNA. Small molecules bind to double-stranded DNA either as intercalators or as minor groove binders, e.g., ethidium bromide (2), Hoechst 33258 (3), or SYBR® Green I (4). Several approaches using target-specific hybridization to single-stranded DNA have been introduced, including Molecular Beacons (5), Scorpions (6)(7), the TaqMan or hydrolysis/5′-nuclease assay (8)(9), the AEGIS probe system (10), labeled primers (11)(12), and light-up probes (13). In contrast to binding of dyes to double-stranded DNA, these methods are suitable for multiplexing approaches because they use differentially labeled fluorescent dyes. However, as they require a unique probe or modified primer for each target, currently used hybridization-based methods for real-time quantitative PCR have high reagent costs and require large developmental efforts.
Here we present a real-time PCR assay that uses universal hybridization-based probe sets suitable for any target. Because the assay uses tailed locus-specific nonmodified amplification primers, PCR products can be monitored via common reporters (cr) hybridizing to the common tails. The general principle of combining tailed PCR primers with universal probes has been introduced for other genetic applications, such as single-nucleotide polymorphism genotyping (14)(15) and in situ amplification (16), but these methods have not been applied to quantitative gene expression studies. Our system, which is similar to the method developed by Whitcombe et al. (15), leads to a more flexible and low-cost setup than conventional hybridization-based approaches. Using differentially labeled universal reporting reagents, we have developed a multiplex setup for simultaneous analysis of target gene and internal control (housekeeping gene), which is an accepted method for normalizing sample-to-sample variation. We then compared the cr-real-time PCR assay with SYBR Green I assays with respect to robustness and sensitivity. As a proof of principle, we applied the new assay to a study of three previously published, differentially expressed candidate genes for congenital heart defects (17).
The principle of the cr-real-time PCR assay is as follows: Target DNA is subjected to PCR in the presence of four oligonucleotides; one low-concentration, tailed, locus-specific primer; one locus-specific primer without a tail; one common primer annealing to a common sequence stretch of the tail; and one universal TaqMan probe corresponding to another common part of the tail (Fig. 1⇓ and Table 1⇓ ). The use of locus-specific primers that span at least one intron of the genomic sequence minimizes problems associated with DNA contamination. During the first amplification cycle, the tailed locus-specific primer initiates the polymerase reaction, leading to the synthesis of a fragment with the tail at the 5′ end. In the second cycle, the complement of the tail is synthesized. Starting from cycle three, the common amplification primer primes synthesis on the fragment including the tail sequence. From this step on, the TaqMan probe anneals to the products resulting from the amplification with the common primer and the locus-specific primer without a tail. The TaqMan probe bound to the tailed target amplicon is hydrolyzed by the 5′-nuclease activity of the DNA polymerase, leading to a physical separation of the reporter and quencher dye and release of fluorescence emission (8).
Introduction of further common tails and corresponding, differentially labeled TaqMan probes opens up the possibility for multiplexing (Fig. 1⇓ ), although the degree of multiplexing is limited by the number of different dyes and the restrictions of currently available instruments (1). The universal tails and TaqMan probes can be easily transferred to other targets in combination with two locus-specific primer sequences. In cases in which locus-specific primers are already designed, ligation to their corresponding common tails by use of the ligation-based synthesis method (18) could be envisaged. Furthermore, the amplification of primer-dimers or pseudogenes can be tested by a melting analysis using SYBR Green I instead of the TaqMan probe for amplicon detection.
Because the cr-real-time PCR assay requires three interacting amplification primers per target gene (Fig. 1⇓ ), the optimum ratio had to be adjusted. For separate analysis, the 20-μL reactions contained 1× TaqMan Universal PCR Master Mix (Applied Biosystems), 0.3 μM appropriate TaqMan probe, 0.4 μM each of the reverse amplification primer (R1/2-primer) and common forward primer (L-primer), 0.004–0.2 μM tailed forward primer (L1/2-primer), and various amounts of template. In the multiplexed analysis, the 20-μL assay mixture contained 1× TaqMan Universal PCR Master Mix (Applied Biosystems), 0.3 μM each of both universal TaqMan probes, 0.4 μM each of both locus-specific reverse primers (R-primers), 0.04–0.2 μM both tailed forward primers (L1/2-primer), 0.4–1.0 μM the common forward primer (L-primer), and various amounts of plasmid of cDNA templates. Cr-real-time PCRs were performed and measured on an ABI Prism 7900HT system. The thermocycling protocol consisted of an initial denaturation at 95 °C for 10 min, followed by 45 cycles of 95 °C for 15 s and 60 °C for 1 min. Subsequently, a dissociation curve was generated in the range of 60–95 °C.
A concentration of 0.016 μM of the tailed locus-specific primer was sufficient as this primer is required only during the initial steps. No signal improvement was achieved with higher concentrations, whereas amplifications were less efficient with lower concentrations. The concentrations of the other two amplification primers were optimally kept at 0.4 μM each. Transferring these assay conditions for separate analyses to multiplexing approaches revealed preferential amplification of one target gene over the other. Given the competitive nature of multiplexed reactions, an increase in concentration of the universal primer necessary for all targets (L-primer) to 0.8 μM appeared crucial. When we used these conditions, the same amplification results were obtained independently from the uniplex or multiplex background of the reaction (see the Data Supplement that accompanies the online version of this Technical Brief at http://www.clinchem.org/content/vol50/issue9/).
The same assay conditions could be transferred to all loci of interest without any further optimization effort (see the online Data Supplement). This successful application of the same reaction conditions demonstrated the robustness and flexibility of the cr-real-time PCR assay. Because each reaction was performed in triplicate, showing only marginal variations (see the online Data Supplement), the cr-real-time PCR demonstrated high reproducibility and accuracy, which were in the same range as the results obtained here (see the online Data Supplement) and the results reported previously (19)(20) for the SYBR Green I assay.
Two different setups were performed using SYBR Green I for real-time detection of amplification products. The two-primer setup contained 1× SYBR Green PCR Master Mix (Applied Biosystems), 0.4 μM forward and reverse primer (L1/2-primer without common tail and R1/2-primer), and various amounts of plasmid or cDNA templates. To simulate the amplification conditions of the cr-real-time PCRs, the three-primer setup was performed with an amplification mixture containing 1× SYBR Green PCR Master Mix, 0.4 μM each reverse primer (R1/2-primer) and common forward primer (L-primer), 0.016 μM tailed forward primer (L1/2-primer), and various amounts of template. All 20-μL amplification reactions were carried out and measured on an ABI Prism 7900HT system (Applied Biosystems), using the same thermal profile as described for the cr-real-time PCR assay.
To evaluate the sensitivity and dynamic range of the cr-real-time PCR assay, we prepared two serial dilutions and subjected various amounts to cr-real-time PCRs as well as to two- and three-primer SYBR Green I assays. A dilution series of total cDNA ranging from 1 (corresponding to 100 ng of reverse-transcribed RNA) to 1:50 000 (corresponding to 2 pg of reverse-transcribed RNA) was used to quantitatively target the housekeeping gene B2M. The three assays were highly linear over the examined range (see the online Data Supplement). Likewise, in a 10-fold dilution series of a FLJ10350 cDNA clone ranging from 109 to 10 copies per reaction, assayed by all three assays, real-time quantification was linear, although it showed more deviation among the triplicates performed on low amounts of template (see the online Data Supplement).
We analyzed expression of three genes that had been found, by SYBR Green I analyses, to be differentially expressed in patients with congenital heart defects (17). The results confirmed the previously published data, with FLJ10350 and TNNI1 being significantly up-regulated and PIPPIN being significantly down-regulated (see the online Data Supplement). Throughout our assays, we saw no amplification of the no-template controls (see the online Data Supplement).
For normalization of the target genes analyzed in the course of this study, the housekeeping gene B2M was simultaneously assayed with the genes of interest. The obvious sample-to-sample variations (see the online Data Supplement) stress the importance of effective systems for normalization, as achieved with the multiplexed cr-real-time PCR assay.
In summary, we have described a single-step method for real-time PCR that is sensitive, robust, and requires minimal optimization effort. Because the system uses nonmodified, tailed amplification primers and universal reporting reagents, it is characterized by a flexible and low-cost format. The use of differentially labeled reporting reagents enables multiplexing approaches for monitoring of more than one target per well, e.g., both a candidate and housekeeping gene. Therefore, the cr-real-time PCR assay appears suitable for the broad spectrum of real-time PCR applications.
Scheme of the multiplexed cr-real-time PCR assay.
Each gene of interest is amplified with a set of three primers: one tailed locus-specific forward oligonucleotide (L1/2-primer), one locus-specific reverse primer (R1/2-primer), and one common primer (L-primer) annealing to a common sequence stretch of the tails. Because the sequence tags of the tails vary, amplification products of different cDNAs are specifically detected by use of TaqMan probes corresponding to the appropriate tags. As illustrated, the design of exon-junction-spanning primers appears advisable, although it is not mandatory for the assay. Open arrows indicate common sequences and primers; gray arrows indicate locus-specific sequences and primers; patterned arrows indicate TaqMan probes and their corresponding sequence stretches in the tailed oligonucleotides; the heads of the arrows indicate the 5′ ends of sequences, and the tails of the arrows indicate the 3′ ends of sequences. FAM, 6-carboxyfluorescein; TET, tetrachloro-6-carboxyfluorescein.
Oligonucleotides used for cr-real-time PCR assays.1
Acknowledgments
This work was funded by EU FP6 Grant 503155 (Moltools). We are grateful to Martin Lange for providing the FLJ10350 cDNA clone and Bogac Kaynak for providing the cDNA samples. We thank Maike Tribbels and Christina Grimm for critical reading of the manuscript.
- © 2004 The American Association for Clinical Chemistry
