| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Molecular Diagnostics and Genetics |
1 Division of Human Genetics, Children’s University Hospital, Inselspital, Bern, Switzerland.
aAddress correspondence to this author at: Head of Division of Human Genetics, Children’s University Hospital, Inselspital, 3010 Bern, Switzerland. Fax: 41-31-632-94-84; e-mail: sabina.gallati{at}insel.ch.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResults and DiscussionReferences |
|---|
Methods: For relative quantitative real-time PCR detection of large deletions in the CFTR gene, we designed DNA-specific primers for each exon of the gene and primers for a reference gene (ß2-microglobulin). For PCR we used a LightCycler system (Roche) and calculated the gene-dosage ratio of CFTR to ß2-microglobulin. We tested the method by screening all 27 exons in 3 healthy individuals and 2 patients with only 1 pathogen sequence variant. We then performed specific deletion screenings in 10 CF patients with known large deletions and a blinded analysis in which we screened 24 individuals for large deletions by testing 8 of 27 exons.
Results: None of the ratios for control samples were false positive (for deletions or duplications); moreover, for all samples from patients with known large deletions, the calculated ratios for deleted exons were close to 0.5. In addition, the results from the blinded analysis demonstrated that our method can also be used for the screening of single individuals.
Conclusions: The LightCycler assay allows reliable and rapid screening for large deletions in the CFTR gene and detects the copy number of all 27 exons.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResults and DiscussionReferences |
|---|
1 in 20 to 1 in 22. CF is primarily caused by pathogen sequence variants in the CF transmembrane conductance regulator (CFTR)2
gene, which encodes a protein produced in the apical membrane of exocrine epithelial cells. In addition to the most common variation, 
F508, more than 1400 different pathogen sequence variants (http://www.genet.sickkids.on.ca/cftr) have been reported to the Cystic Fibrosis Genetic Analysis Consortium. Symptomatic CF is associated with at least 1 pathogen sequence variant (mainly point variants) on each CFTR allele. However, a nonnegligible portion of patients with typical or CF-like symptoms have only 1 detectable variation. The 2nd pathogen sequence variant is attributable to several possible genetic mechanisms, for example, large deletions (spanning multiple or single exons), unknown splice variants, combinations of multiple sequence variants, modifier genes, and epigenetic factors. In patients with CF, the frequency of large deletions in the CFTR gene is estimated to be 1%–3% (1). In well-defined study cohorts [e.g., patients with 1 unidentified CFTR allele and typical CF clinical findings (e.g., pancreas insufficiency, lung disease, and pathologic sweat test results)], the reported detection rates of large deletions are 15%–25% (1)(2)(3)(4), but the actual rates may be even higher (5). Data obtained from a large study (6) that determined the frequencies of the CFTRdele2,3 (21 kb), which is by far the most frequent large deletion in the CFTR gene, revealed that the aforementioned frequency is underestimated, especially in CF patients from Eastern Europe. The number of studies in this field has been increasing since the first systematic examination of rearrangements in the CFTR gene (2). In the past, large deletions of the CFTR gene were identified by coincidence, either by noting a uniparental inheritance pattern (7) or through failure of PCR amplification (8). It is conceivable that the screening for large deletions will be integrated into CF total screening, especially in patients with only 1 known pathogen sequence variant who have CF-like symptoms.
The number of studies dealing with the detection of deletions and duplications by real-time PCR is increasing (9). Some of the most frequently analyzed genes are described in studies (10)(11)(12)(13)(14)(15) showing that real-time PCR is a reliable and efficient way to detect large deletions and duplications.
In addition to real-time PCR, several other techniques are used to detect large deletions and duplications in genes. Multiplex ligation-dependent probe amplification (MLPA), first described by Schouten et al. (16), is a widely used method. The CFTR MLPA reagent set (MRC Holland) allows the detection of the copy number of all 27 exons in a single reaction.
We aimed at developing a reliable and simple real-time PCR assay for detection of all known types of large deletions in the CFTR gene and assess advantages and disadvantages of deletion-detection methods on the basis of relative quantitative real-time PCR and MLPA.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResults and DiscussionReferences |
|---|
.
|
Deletion screening of the entire CFTR gene in patient and control samples.
We used DNA samples from 3 healthy control individuals and from 2 CF patients with only 1 known variation, 1 with a W1282X and 1 with a 5T sequence variant in intron 8. Both patients carried multiple homozygous sequence variants and therefore were also likely to carry large heterozygous deletions in the CFTR gene.
Deletion screening in CF patients with known large deletions.
For this screening we used 9 DNA samples obtained by 2 French research groups from CF patients with known large deletions and 8 DNA samples obtained from CF patients in our group with newly identified large deletions.
Blinded analysis.
We conducted blinded analysis on DNA samples chosen from a pool containing 61 DNA samples (1st pool) from patients with classic or atypical CF symptoms in whom only 1 pathogen sequence variant in the CFTR gene could be detected by the aforementioned screening method and on 10 DNA samples (2nd pool) from CF patients with known large deletions. We chose a total of 24 samples for the deletion screening, 18–21 from the 1st pool and 3–6 from the 2nd pool. We used the CFTR MLPA Kit from MRC Holland to simultaneously screen all samples from the 1st pool for large deletions. To correctly determine the deletion size, we screened for all necessary exons in all samples containing a deletion.
Controls.
Control DNA (calibrator) samples from individuals in whom no pathogen CFTR sequence variants were identified by our variation screening were included in every experiment. One control was used for comprehensive deletion screening and for confirmation experiments, and another was used only for the blinded analysis.
All study participants gave their full informed consent for research use of their blood samples. This study was approved by a local ethics commission.
dna extraction
All DNA specimens were obtained by DNA extraction from peripheral blood cells. For the participants from our laboratory the DNA extraction was performed with the QIAamp DNA Blood Maxi Kit (Qiagen) according to the supplier’s protocol. We measured the concentration of once-diluted DNA samples by spectrophotometry (Eppendorf BioPhotometer; absorbance at 260 nm) and then diluted each sample to the desired final concentration [5 mg/L for patients and controls (calibrator) and 10 mg/L for the control samples used for standard curve dilutions].
primer design
By use of the OLIGO 6.0 software, we designed primer pairs for each exon of the CFTR gene and 2 fragments of the reference gene (ß2-microglobulin, exon/intron 2). Amplification products were all DNA-specific, included at least 1 primer (forward or reverse) that laid in an intron or covered an intron/exon boundary and met the following criteria: amplification product at 250 bp, melting temperature 60 °C, and no stable dimer formation (in particular, no 3'-terminal dimer formation) (see Table 1 in the Data Supplement that accompanies the online version of this article at http://www.clinchem.org/content/vol52/issue11).
relative quantitative real-time pcr
PCRs were set up in duplicate in a total volume of 20 µL per capillary. One reaction mixture contained 10 µL of SYBR® Premix Ex TaqTM Premix from TaKaRa, including a Hot start Taq Polymerase (TaKaRa Ex TaqTM HS), a dNTP mixture, MgCl2 (2 mmol/L) and SYBR green, 2 µL of forward and reverse primer each (concentration = 5 µmol/L), 2 µL H2O, and 4 µL DNA [40, 20, 10, and 5 ng total DNA for the standard curves and 20 ng DNA for the patients/control samples (calibrator)]. All experiments were performed on LightCycler instruments. For all amplification reactions, the same cycling program was used (see Table 2 in the online Data Supplement).
To verify the specificity of the amplifications, we loaded all PCR products on 2% agarose gels and sequenced them with an ABI 3100 (Applied BioSystems) sequencing system.
We used LightCycler data analysis software (version 3.5) for data evaluation. Ratios of potentially deleted to nondeleted exons were calculated with 2 different methods on the basis of the same mathematical model.
Calculation by the relative standard curve method.
We generated standard curves for both the test exon and the exon of the reference gene. The slope and y intercept of the generated standard curve allowed the calculation (Eq. 1) of the concentration and the total amount of DNA, respectively, for the sample and the calibrator (control). This calculation is done automatically by the LightCycler data analysis software. To achieve best reproducibility, we used the 2nd derivative method for the calculation of the concentrations.
The concentrations (the total amounts of DNA) were taken directly from the quantification screen of the LightCycler data analysis software and inserted into Eq. 1 :
 | (1) |
Alternatively, the ratio [or dosage quotient (DQ)] can be calculated by the mathematical model presented by Pfaffl (17) and Livak and Schmittgen (18).
Calculation by the comparative CT (CT) method.
The use of this method does not require standard curve data. The CT method is a simplification of the relative standard curve method insofar as the efficiencies for the reference and target gene are assumed to be 2 (18).
We performed all experiments except for the blinded analysis by applying the relative standard curve method. We used both the relative standard curve method and the CT method to calculate the ratios of potentially deleted exons, but for clarity, we chose only the results from the relative standard curve method for the creation of the figures.
screening strategies
In single individuals, screenings for large deletions (see Fig. 1
in the online Data Supplement) can be accomplished by 2 different screening strategies.
|
Screening strategy 1.
Examining 8 of 27 exons (exons 1, 2, 4, 14a, 17b, 19, 21, and 22) allows assessment of whether a person carries a large deletion but does not permit determination of the size of the deletion, except for deletions in which only 1 exon is lacking (exons 1, 4, 14a, 17b, 19, and 21). The advantage of this method is that it can be accomplished with only 36 reactions (4 reactions per each CFTR exon or B2-microglobulin amplification product) and is therefore a cost-efficient alternative to MLPA screening. The choice of the 8 tested exons was made on the basis of data obtained from the CF database (http://www.genet.sickkids.on.ca/cftr).
Screening strategy 2.
Testing of 16 exons (exons 1, 2, 3, 4, 6a, 7, 8, 14a, 14b, 17a, 17b, 18, 19, 21, 22, and 24) allows determination of the extent of the deletion. Confirmation of the endpoints of a large deletion can be done by testing the 2 adjacent exons.
Both screening strategies can be easily adapted to newly identified large deletions by increasing the number of tested exons. We tested the applicability of these screening strategies by performing a blinded analysis in which 24 individuals were screened for large deletions according to screening strategy 1.
Finally, we compared the ratios obtained by the relative standard curve method with those obtained by the CT method.
|
Results and Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResults and DiscussionReferences |
|---|
and 2
, respectively (also see Fig. 2 in the online Data Supplement).
|
For the blinded analysis, we detected large deletions in 5 of the 24 samples, 4 samples from positive controls and 1 from a CF patient with only 1 identified pathogen sequence variant (Table 2
). To confirm our findings, we used MLPA to simultaneously screen all 24 samples for large deletions. Calculated ratios from all experiments except the blinded analysis are summarized in Fig. 3
. In summary, none of the results for the control samples were false-positive (deletions or duplications) ratios; moreover, calculated ratios of all samples from patients with known large deletions were accurate and confirmed the previous findings. Our comparison of the calculated ratios obtained by the relative standard curve method with those obtained by the CT method showed no significant differences between the 2 calculation methods (Fig. 4
).
|
|
|
Our results show that this assay allows reliable detection of large heterozygous deletions within the CFTR gene. Compared with existing methods of deletion screening of the complete coding sequence of the CFTR gene on the basis of semiquantitative fluorescent multiplex PCR assays [quantitative multiplex-PCR of short fluorescent fragments (2), quantitative fluorescent multiplex-PCR (3), and semiquantitative fluorescent-PCR (5), our assay enables quantitative detection of the copy number from all 27 exons. The principles of these methods are described in detail elsewhere (19)(20)(21)(22). The 3 studies of existing methods of deletion screening (2)(3)(5) did not present detailed statistical data of the calculated ratios (DQs), so a comparison of the differences in performance is limited. Our results for complete screening of 5 individuals (Fig. 1
) and the blinded analysis (Table 2
) demonstrate the accurate detection capability of our quantitative real-time PCR assay: mean ratios were 0.96–1.07 for the 5 samples from 5 healthy individuals and 0.95–1.02 for the 24 individual samples used in the blinded analysis. SDs varied from 0.02 to 0.10 (SD of sample) and from 0.04 to 0.09 (SD of exon), respectively (Table 2
). Thus, the risk of false positive ratios is lower with our method than the method of Hantash et al. (5), decreasing the identification of erroneous deletions or duplications. This risk is not that relevant for duplications, however, because they are extremely rare. The lower performance of the semiquantitative fluorescent-PCR assay can be easily explained. The PCR efficiencies for each fragment are different and decrease with increasing length of the PCR product, and single-tube multiplex PCR leads to additional variation (primer concurrence) in PCR efficiencies. In MPLA, this problem has been solved by use of a universal primer for the amplification of all fragments. Another disadvantage of semiquantitative multiplex-PCR assays is that the amplification of the PCR products must be stopped at the exponential and not at the plateau phase. Because of the different PCR efficiency of each fragment, it is possible that some fragments are still in the exponential phase at the same time that other fragments have already reached the plateau phase. Both research groups overcame this problem by choosing a rather low cycle number.
In our study (Table 2
) DQ values were 0.81–1.21 (mean, 0.99) for control samples and 0.41–0.56 (mean, 0.49) for heterozygous deletions, but the DQ values reported by Niel et al. (3) were 0.75–1.31 (control samples) and 0.38–0.64 (deletion samples). Interestingly, the methods used by Niel et al. (3) and Hantash et al. (5) were essentially the same, yet the differences in the accuracy of the DQ values are large. A possible explanation is that Niel et al. (3) performed the multiplex PCR in 3 different tubes and Hantash et al. (5) used a single tube. Primer competition seems to have been a more important problem in the latter case, leading to a greater variation of calculated DQ values.
The lack of statistical data of calculated DQs in the work from Audrezet et al. (2) prevents a comparison with our assay.
real-time pcr as an alternative screening method to mlpa
Both real-time PCR and MLPA detection methods are reliable and have considerable advantages, as well as some disadvantages (also see Table 3 in the online Data Supplement). The decision whether to use real-time PCR or MLPA depends primarily on the number of patients to be screened. With a duration of 50 min per run, our assay allows the screening of 6–8 persons per day (on one 32-well LightCycler instrument). With MLPA, a maximum of 96 persons can be tested in 3 days. Therefore, the MLPA technique is useful for large CF centers and research laboratories where 10 to 100 patient samples per day are tested. For clinical laboratories where at most 6–8 patient samples are screened per day, our method may be preferable to MPLA. In addition, there are 3 situations for which relative quantification by real-time PCR is the method of choice:
influence of nonspecific products on the CT value and calculated ratios
With the use of intercalating dyes such as SYBR green, fluorescence signals could be caused by nonspecific double-stranded products (e.g., primer dimers) and lead to inaccurate CT values. We were confronted with this problem in our study. Analyzing the melting curves from PCR products of exon 4, 5, 6a, 11, and 18, we observed the presence of nonspecific products in the form of an additional peak or a hump preceding the main peak. The proportion of the fluorescence signal was for all unspecific products lower than 1/5th of the specific product. Moreover, primer dimers, as well as unspecific amplification products, could be excluded. In a final step, we used sequencing (ABI 3100, Applied BioSystems) to confirm the specificity of each PCR product. Thus, we considered the additional peaks found in the melting curves of the aforementioned amplification products to be of no importance, all the more so because the calculated ratios for these exons did not differ from those of the other exons.
use of pcr agents
SYBR Premix Ex TaqTM premix from TaKaRa can be used with different real time PCR instruments, such as ABI PRISM 7000/7700/7900HT, Applied BioSystems 7300/7500 Real Time PCR System (Applied BioSystems), LightCycler (Roche Diagnostics), and Smart Cycler (Cepheid). Alternatives to the TaKaRa Premix include 3 premixes from Roche Diagnostics that can be used for amplification: LightCycler FastStart DNA MasterPLUS SYBR Green I (which already includes MgCl2), LightCycler FastStart DNA Master SYBR Green I (plus additional MgCl2), or LightCycler DNA Master SYBR Green I.
Comparing the TaKaRa and the LightCycler FastStart DNA Master SYBR Green I premixes, we made 2 important observations. First, the efficiencies with both premixes are high (E 
0.9) and almost identical. Second, the TaKaRa premix is much more susceptible to alteration by thaw-freeze cycles but less expensive than the premix from Roche.
relative standard curve method vs CT method
Generation of new standard curves for each experiment is a time-consuming procedure that can be overcome by importing standard curve data with the LightCycler software. The CT method has been shown to be a reliable and cost-effective alternative to the relative standard curve method (15)(23). Because the differences between the 2 calculation methods were negligible for our assay (Fig. 4
), the relative standard curve method can be replaced by the cheaper and faster CT method without loss accuracy, as clearly confirmed by our blinded analysis.
In conclusion, our assay is a rapid and reliable screening method for (a) detection of large deletions, especially in cases in which specific testing is demanded, such as carrier identification, and screening of individual patients, especially in prenatal diagnosis, and (b) confirmation of large deletions previously identified by MLPA.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 Nonstandard abbreviations: CF, cystic fibrosis; MLPA, multiplex ligation-dependent probe amplification; CT, comparative CT method; DQ, dosage quotient.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResults and DiscussionReferences |
|---|
CT method. Methods 2001;25:402-408.[CrossRef][Web of Science][Medline]
[Order article via Infotrieve]The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  F. M. Hantash, A. Rebuyon, M. Peng, J. B. Redman, W. Sun, and C. M. Strom Apparent Homozygosity of a Novel Frame Shift Mutation in the CFTR Gene Because of a Large Deletion J. Mol. Diagn., May 1, 2009; 11(3): 253 - 256. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Lyon and C. T. Wittwer LightCycler Technology in Molecular Diagnostics J. Mol. Diagn., March 1, 2009; 11(2): 93 - 101. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Schrijver, K. Rappahahn, L. Pique, M. Kharrazi, and L.-J. Wong Multiplex Ligation-Dependent Probe Amplification Identification of Whole Exon and Single Nucleotide Deletions in the CFTR Gene of Hispanic Individuals with Cystic Fibrosis J. Mol. Diagn., July 1, 2008; 10(4): 368 - 375. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. P. Vaughn, E. Lyon, and W. S. Samowitz Confirmation of Single Exon Deletions in MLH1 and MSH2 Using Quantitative Polymerase Chain Reaction J. Mol. Diagn., July 1, 2008; 10(4): 355 - 360. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Castellsague, S. Gonzalez, M. Nadal, O. Campos, E. Guino, M. Urioste, I. Blanco, T. Frebourg, and G. Capella Detection of APC Gene Deletions Using Quantitative Multiplex PCR of Short Fluorescent Fragments Clin. Chem., July 1, 2008; 54(7): 1132 - 1140. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Montgomery, C. T. Wittwer, J. O. Kent, and L. Zhou Scanning the Cystic Fibrosis Transmembrane Conductance Regulator Gene Using High-Resolution DNA Melting Analysis Clin. Chem., November 1, 2007; 53(11): 1891 - 1898. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. von Kanel, F. Adolf, M. Schneider, J. Sanz, and S. Gallati Sample Number and Denaturation Time Are Crucial for the Accuracy of Capillary-Based LightCyclers Clin. Chem., July 1, 2007; 53(7): 1392 - 1394. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
