| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Department of Molecular Genetics, Erasme Hospital, Bâtiment C niveau 5, 808 Route de Lennik, 1070 Brussels, Belgium;
aauthor for correspondence: fax 32-2-555-4527, e-mail helhousn{at}ulb.ac.be
The emergence of fluorescence techniques in 96- or 384-well formats has led to high-throughput single-nucleotide polymorphism (SNP) detection methods. Among these, allele discrimination based on real-time PCR technologies has been developed along two main methodologic approaches. In the first approach, a temperature-dependent fluorescent signal is generated by hybridization of a probe at the end of each PCR cycle. The monitoring of fluorescence emission is followed during (1)(2) or at the end of (3)(4) the PCR process. In the latter case, allele detection is achieved by carrying out of a melting curve analysis.
In the second approach, exploiting the classic 5' nuclease assay (TaqMan®) technology, a temperature-independent fluorescence signal is generated during the PCR process by the exonuclease degradation of a hybridized TaqMan probe (5)(6). Except for one method discussed later (3), all of these techniques require the use of two probes for allele differentiation.
An allele differentiation study has stressed the risk of artifacts and subsequent erroneous results when a polymorphism is located within the DNA sequence targeted by TaqMan (7). The same study demonstrated the power of methods relying on determination of melting curves.
Although other interesting approaches have been reported recently, such as DASH (8), iFret (9), and dipole calculation (10), some of them are technically difficult or time-consuming. In addition, the SYBR Green technique (11) would represent the simplest method to detect point mutations, but it seems to be limited by its lack of ability to differentiate alleles (12).
We present here a new methodologic approach that allows high-throughput genotyping of SNPs. Allele differentiation is achieved by the generation of melting curves, with a single dual-labeled probe, at the end of a classic PCR reaction. The reliability of the method is demonstrated for detection of the factor V Leiden (13)(14) and prothrombin 20210A mutations (15). The method is based on the basic principle that melting curve analyses performed at the end of a PCR reaction with a single dual-labeled probe should allow differentiation of SNPs.
To avoid hydrolysis of this probe, as occurs classically during the elongation steps in TaqMan real-time PCR, the melting point (Tm) of the probe was chosen to be 
10 °C below the Tms of the PCR primers. At the end of the PCR reaction, the probe is allowed to hybridize and the mixture is subjected to stepwise increase in temperature, with fluorescence monitored continuously. As in classic TaqMan real-time PCR, generation of the fluorescence signal by the probe is based on the Förster resonance energy transfer (FRET) phenomenon. However, and contrary to what happens in TaqMan real-time PCR, no hydrolysis of the probe by Taq polymerase is involved. Rather, the procedure relies on the decrease in FRET observed when the probe detaches from its target to achieve a random single-stranded conformation. The mean distance between the reporter and the quencher molecules of the dual-labeled probe will become shorter when the probe is released from its hybrid with the target sequence. Because the FRET effect is inversely proportional to the sixth power of this distance (16), a difference in fluorescence emission will be readily detectable between hybridized and melted configurations of the probe.
DNA extractions were performed on 220 µL of whole blood with the QIAamp 96 DNA Blood Biorobot reagent set on the Biorobot 9604 platform (Qiagen). Primers and probes (Table 1
) were designed with the freeware MELTCALC, Ver. 2.0, demo (http://www.meltcalc.de) and purchased from Eurogentec (Belgium).
|
Final concentrations of 100 and 900 nmol/L for the forward and reverse primers, respectively, and 100 nmol/L for the probe were mixed with 12.5 µL of Platinum® Quantitative PCR Supermix-UDG (Invitrogen) and 100 ng of DNA in a final volume of 25 µL.
The PCR conditions were as follows: real-time PCR (1 cycle at 50 °C for 2 min; 1 cycle at 95 °C for 2 min; 40 cycles at 95 °C for 10 s, 65 °C for 30 s, and 72 °C for 20 s), followed by melting curve analysis (1 cycle at 95 °C for 2 min, 20 °C for 1 min, and 70 °C for 1 min, with a total ramping time of 19 min and 59 s between the 20 °C and 70 °C conditions). During the optimization phase, final primer concentrations varied between 100 and 900 nM.
All experiments were performed on a ABI Prism® 7700 Sequence Detection System (Applied Biosystems). Data collected during the PCR and the melting curve steps were analyzed with SDS software, Ver. 1.7. The multicomponent file obtained for the melting curve was analyzed with the "Dissociation Curve" software (Applied Biosystems).
In a first round of experiments, we observed that the best curves for detection of the factor V Leiden mutation were obtained under conditions of primer imbalance (e.g., 100 nM forward primer and 900 nM reverse primer; data not shown). Primer concentration imbalance seems to be important because no peak in the melting curve profiles was observed when equimolar concentrations of both primers were used. However, for the prothrombin 20210A assay, this rule did not apply. We did not observe a direct relationship between the quantity of product obtained at the end of the PCR process (estimated by loading the samples on agarose gel) and the aspect of the melting curve.
To assess the effects of hydrolysis of the probe during PCR on the melting curve profile, we first compared the melting curves obtained from experiments in which probes were added either before or after the PCR run (Fig. 1, A and B
). Because we observed no increase in background when the probe was exposed to the entire PCR procedure, we concluded that, under the conditions of the test, no significant degradation of the probe occurs or, at least, that it does not affect subsequent recording of the melting curve. To explore the effect of the annealing temperature on degradation of the probe, we modified the PCR protocol by decreasing the annealing temperature from 65 to 55 °C. Because the Tms of homozygous mutant DNA for the prothrombin 20210A and factor V Leiden mutations are 61.5 and 60 °C, respectively (Fig. 1, A and B
), this would be expected to favor hydrolysis of the probes, at least when annealing to mutant alleles.
|
As shown in Fig. 1
, in which fluorescence was monitored continuously during the PCR cycles (Fig. 1
, C and D), we observed much higher degradation of the probe for factor V Leiden, with annealing at 55 °C (Fig. 1C
) compared with 65 °C (see the Data Supplement that accompanies the online version of this Technical Brief at http://www.clinchem.org/content/vol49/issue10/), with only marginal influence on the melting curve profile. Indeed, a steady increase in fluorescence intensity was observed at both the denaturation (95 °C) and annealing temperatures throughout the PCR process when annealing was at 55 °C. On the other hand, virtually no increase in fluorescence was observed at 65 or 95 °C for an annealing temperature of 65 °C. This phenomenon was, however, not observed for prothrombin 20210A because an increase in fluorescent signal was observed only at the annealing temperature (Fig. 1D
and online Data Supplement) and not at 95 °C, meaning that there was probably no degradation of the probe, only hybridization to its target. Although logical on a theoretical basis, these data indicate that, depending on the sequence context of the SNP, designing probes with a Tm 10 °C lower than that of the primers is not an absolute prerequisite to obtain interpretable melting curves.
To validate our method, we tested 100 patient samples corresponding to different genotypes (homozygous wild-type, heterozygous, and homozygous mutant) for the prothrombin 20210A and factor V Leiden mutations and compared our results with those obtained with the classic restriction fragment length polymorphism approach used in our laboratory (data not shown). The results were unambiguous, and we observed a 100% correlation between both assays.
Our approach is similar to the one based on the natural quenching effect of the desoxyguanosine nucleotide (3). It uses probes labeled with a single reporter molecule and allows recording of a differential fluorescence signal, depending on whether the probes are hybridized to their target. Whereas our method involves the additional cost associated with the need for dual-labeling of the probes, it has the advantage of being compatible with performance under standard PCR conditions, regardless of the SNP to be assayed. This characteristic makes it feasible to assay simultaneously, on the same 96- or 384-well plate, a variety of SNPs. Interestingly, we noted that our factor V Leiden probe was positioned next to three Gs on the complementary strand. This would mean that the reporter molecule should be partially quenched by those three Gs when the probe anneals to its target (3). It is likely, however, that the quenching effect is not sufficient to prevent all fluorescence emission of the reporter because, as Crockett and Wittwer have reported (3), this effect has a maximum efficacy of 40%.
Finally, although we have noted that the integrity of the probe and the primer ratio imbalance are not absolute prerequisites for the success of the method in individual applications, we have devised a generic protocol that is widely applicable. In addition to factor V Leiden and prothrombin 20210A mutations, for which the results were superimposable with those obtained with the current PCR-restriction fragment length polymorphism method, we have successfully applied it to the detection of a series of other single-base substitutions (data not shown).
In summary, we have developed an inexpensive method that combines the use of a single TaqMan probe and melting curves and can be performed under universal conditions.
Acknowledgments
This work was supported by "Fonds de la Recherche Scientifique Médicale" and "Belgium Loterie Nationale". We thank Muriel N'Guyen and Serge Giraud for technical help and Jean-Luc Vaerman for helpful discussions during this study.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  L. M. Holdt, J. Thiery, J. L. Breslow, and D. Teupser Increased ADAM17 mRNA Expression and Activity Is Associated With Atherosclerosis Resistance in LDL-Receptor Deficient Mice Arterioscler Thromb Vasc Biol, June 1, 2008; 28(6): 1097 - 1103. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Amicarelli, E. Shehi, G. M. Makrigiorgos, and D. Adlerstein FLAG assay as a novel method for real-time signal generation during PCR: application to detection and genotyping of KRAS codon 12 mutations Nucleic Acids Res., October 11, 2007; (2007) gkm809v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Amicarelli, D. Adlerstein, E. Shehi, F. Wang, and G. M. Makrigiorgos Genotype-Specific Signal Generation Based on Digestion of 3-Way DNA Junctions: Application to KRAS Variation Detection Clin. Chem., October 1, 2006; 52(10): 1855 - 1863. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D Franchimont, S Vermeire, H El Housni, M Pierik, K Van Steen, T Gustot, E Quertinmont, M Abramowicz, A Van Gossum, J Deviere, et al. Deficient host-bacteria interactions in inflammatory bowel disease? The toll-like receptor (TLR)-4 Asp299gly polymorphism is associated with Crohn's disease and ulcerative colitis Gut, July 1, 2004; 53(7): 987 - 992. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) or after (
) the PCR; x corresponds to wells in which no DNA were added. To visualize the transitions corresponding to melting of the alleles as positive peaks, the values on the y axis are presented as the negative first derivative of the fluorescence. Panels C and D illustrate the effects of annealing on degradation of the probe. The experiments were performed on homozygous mutant genomic DNA for the factor V Leiden (C) or prothrombin 20210A genes (D). The annealing temperature during the PCR step was 55 °C. Fluorescence signals (y axes; arbitrary units) were collected during all individual PCR steps, along the entire PCR (x axes; time in seconds).