| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Department of Chemical Pathology,2 Centre for Emerging Infectious Diseases, and4 Department of Obstetrics & Gynaecology, The Chinese University of Hong Kong, Prince of Wales Hospital, Hong Kong, Special Administrative Region, China;3 Harris Birthright Research Centre for Fetal Medicine, King’s College Hospital, London, United Kingdom
aaddress correspondence to this author at: Department of Chemical Pathology, The Chinese University of Hong Kong, Prince of Wales Hospital, Room 38023, 1/F Clinical Sciences Bldg, 30-32 Ngan Shing St., Shatin, New Territories, Hong Kong Special Administrative Region, China; fax 852-2194-6171, e-mail loym{at}cuhk.edu.hk)
Down syndrome is the most common chromosomal abnormality of live-born babies. Currently, cytogenetic analysis of fetal cells such as full chromosome karyotyping (1) and fluorescence in situ hybridization (2) are the most widely used techniques for prenatal diagnosis of this chromosomal aneuploidy. Recently, however, alternative molecular strategies have been developed. Quantitative fluorescence PCR involves the detection of short tandem repeats on chromosome 21 and is by far the most thoroughly evaluated molecular technique. Additional reported molecular detection methodologies include real-time quantitative PCR (3)(4), multiplex probe ligation assays (5), and paralogous sequence quantification (6). Using melting curve analysis of single-nucleotide polymorphisms (SNPs), Pont-Kingdon et al. (7) have measured the relative allele copy number in fixed cells and amniocyte cell cultures of trisomy 21 samples. Compared with short tandem repeats, SNPs occur much more frequently in the human genome (8) and thus would potentially provide more information on chromosomal abnormalities. In the present study, we investigated the feasibility of measuring the allelic ratios of the chromosome-21 SNPs to identify trisomy 21 in various clinical samples. We detected and quantified the allelic ratios by PCR followed by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) (9). Theoretically, normal euploid samples would give an allelic ratio of 1, whereas trisomy 21 samples would give allelic ratios of ratios of 0.5 or 2.0.
Placenta, amniotic fluid, and chorionic villus sampling (CVS) from normal and trisomic 21 cases were collected from the Department of Obstetrics and Gynecology, Prince of Wales Hospital, Hong Kong, and the King’s College Hospital, London, United Kingdom. Normal or trisomic status was confirmed by karyotyping. Ethics approval was obtained from the corresponding Institutional Review Boards. DNA extraction from the placentas and CVS was performed with a QIAamp DNA Mini Kit (Qiagen) with the "tissue protocol". For the amniotic fluid, DNA was extracted from 800 µL of each sample with the "blood and body fluid protocol"; the final elution volume was 30 µL.
Five highly polymorphic SNPs located within the Down syndrome critical region of chromosome 21 were selected from the Sequenom RealSNPTM Assay Database (www.realsnp.com; see Table 1
in the Data Supplement that accompanies the online version of this Technical Brief at http:www.clinchem.org/content/vol51/issue12). Four of the SNPs are located within the genes Down syndrome critical region gene 2 (DSCR2), PR domain containing 15 (PRDM15), ubiquitin-conjugating enzyme E2G 2 (UBE2G2), and collagen type VI alpha 2 (COL6A2), whereas 1 SNP is located in the locus HS21C101.
|
For each 25-µL PCR reaction, 25 ng of DNA from the placental and CVS samples and 5 µL of DNA from the amniotic fluid samples were amplified. Each reaction contained 1x HotStar Taq PCR buffer with 1.5 mM MgCl2 (Qiagen), an additional 1 mM MgCl2 (Qiagen), 50 µM each of dATP, dGTP, and dCTP, 100 µM dUTP (Applied Biosystems), 200 nM each of the forward and reverse primers (Integrated DNA Technologies; see Table 1
in the online Data Supplement), and 0.1 U of HotStar Taq Polymerase (Qiagen). The PCR reaction was initiated at 95 °C for 15 min, followed by 94 °C for 20 s, 56 °C for 30 s, and 72 °C for 1 min for 45 cycles, and a final incubation at 72 °C for 3 min. PCR products were subjected to shrimp alkaline phosphatase treatment with 0.6 µL of shrimp alkaline phosphatase (Sequenom), 0.34 µL of MassARRAYTM Homogenous MassEXTENDTM (hME) buffer (Sequenom), and 3.06 µL of water. The mixture was incubated at 37 °C for 40 min followed by 85 °C for 5 min. SNP genotyping and allelic ratio determination were performed with hME assays (Sequenom). Briefly, 4 µL of base extension reaction cocktail containing 771 nM extension primer (Integrated DNA Technologies; see Table 1
in the online Data Supplement), 1.15 U of Thermosequenase (Sequenom), and 64 µM each of ddATP, ddCTP, ddTTP, and dGTP (Sequenom) were added to 10 µL of the PCR products. The reaction conditions were 94 °C for 2 min, followed by 94 °C for 5 s, 52 °C for 5 s, and 72 °C for 5 s for 75 cycles. The primer extension assays were designed such that the allele-specific extension products for each SNP demonstrated distinct masses that were readily resolvable by MALDI-TOF MS analysis (see Table 2 in the online Data Supplement). The final base extension product was cleaned up by addition of 12 mg of the Clean Resin (Sequenom) and 24 µL of water. The mixtures were mixed in a rotator for 20 min. After centrifugation at 361g for 5 min, 
10 nL of reaction solution was dispensed onto a SpectroCHIP (Sequenom) by a MassARRAY Nanodispenser S (Sequenom). A MassARRAY Analyzer Compact Mass Spectrometer (Bruker) was used for data acquisition from the SpectroCHIP. Mass spectrometric data were automatically imported into a MassARRAY Typer (Sequenom) database for analysis.
For each heterozygous SNP, the frequency of each allele was calculated. The allele frequencies represented the relative copy number of the 2 alleles. To account for any experimental biases between the 2 alleles, skew corrections were applied. For each SNP assay, the skew correction factors were determined by their allele frequencies from 10 euploid placental DNA samples (see Table 3 in the online Data Supplement). The skew correction factors were calculated by the "Genotype Area report" program of the MassARRAY Typer software. They were calculated such that a frequency ratio of 0.5:0.5 was obtained for the 2 alleles in these normal control samples (10). These correction factors were then used to adjust the allele frequencies in the subsequent experiments by the "Allelotype Correction report" program of the MassARRAY Typer software. The corrected frequencies would more truly reflect the relative allele copy number of the SNP. The allelic ratio was finally calculated by dividing the corrected frequency of the higher-mass allele by the corrected frequency of the lower-mass allele.
This study comprised three parts. In the first part, we tested our strategy by measuring the allelic ratios of the 5 SNPs in the placentas. The placental samples were collected in Hong Kong from 13 women known to be carrying trisomy 21 fetuses after termination of pregnancy (14–23 weeks) and 10 women carrying euploid fetuses after elective cesarean delivery (38–41 weeks). Examples of the mass spectra are shown in Fig. 1
in the online Data Supplement. Systematic allelic biases were observed for all of the SNP assays, as shown by the differences in frequencies between the 2 alleles in the DNA samples from the euploid fetuses, which possessed equal copy numbers for the 2 alleles. The differences in allele frequencies were attributable to allele-specific differences in the PCR, primer extension, and MALDI-TOF MS steps. The steps for correcting these differences are detailed above. The allelic ratios of the trisomic and nontrisomic placental DNA samples are shown in Fig. 1
. For each SNP, the samples could be categorized into 3 nonoverlapping groups: the allelic ratios of the normal samples were close to 1 (ranging from a median of 0.860 for PRDM15 to a median of 1.152 for COL6A2). The trisomic placental samples were segregated into 2 groups. For the heterozygous samples with the higher-mass alleles overrepresented, the median allelic ratios ranged from 1.438 for PRDM15 to 2.016 for DSCR2, whereas for the samples overrepresented with the lower-mass alleles, the median allelic ratios ranged from 0.437 for DSCR2 to 0.545 for HS21C101. These data demonstrate that measurement of the SNP allelic ratios by MS could differentiate the trisomic placental samples from the normal samples with 100% discrimination.
|
We next investigated whether the MS-based strategy could be applied to amniotic fluid specimens. The amniotic fluid samples were collected in Hong Kong from 5 women known to be carrying trisomy 21 fetuses (16–19 weeks) and 23 women carrying euploid fetuses (12–20 weeks). The median ratios of the normal amniotic fluid samples ranged from 0.896 for HS21C101 to 1.071 for UBE2G2, which were consistent with the data obtained from the placental tissue samples. For the trisomy 21 amniotic fluid samples, their allelic ratios clearly deviated from that of the normal samples (see Fig. 2 in the online Data Supplement). All trisomy 21 samples could thus be distinguished from the normal samples by comparing the SNP allelic ratios. Similar to other allelic ratio–based detection methods, maternal cells contamination of the amniotic fluid samples remains a potential problem for interfering with the diagnosis.
We conducted a blinded study on CVS samples from 57 pregnant women (11–13 weeks of gestation), recruited in the United Kingdom, who carried either a trisomy 21 or a euploid fetus. In this experiment, the person performing the MS analysis was blinded with regard to the aneuploidy status of the specimens. To establish a reference interval for the allelic ratios of normal pregnancies, we determined the allelic ratios of 10 additional control CVS samples collected in Hong Kong from women with confirmed euploid pregnancies. The reference interval for each SNP was calculated as the mean ± 2 SD of the control CVS samples (Table 1
). For the 57 blinded CVS samples, 53 (93%) were heterozygous for at least 1 SNP. The allelic ratios of the 53 informative samples were separated into the trisomic and the normal groups. A total of 44 cases were assigned to the normal group because they showed allelic ratios within the reference interval for at least 1 SNP. This sample group was subsequently confirmed to be from normal pregnancies. On the other hand, the allelic ratios of 9 cases fell outside the reference intervals (Table 1
). These 9 cases were subsequently confirmed to be trisomy 21 samples. Thus, these results demonstrated that our present approach of allelic ratio measurement accurately identified trisomy 21 cases, with no false-positive or -negative results, in this cohort. In a clinical laboratory setting, at least 2 informative markers with consistent trisomic results are required to classify a sample as abnormal. With such a requirement, 8 of the 9 trisomic CVS samples were heterozygous for at least 2 SNPs, and such confirmation was therefore possible.
Regarding interassay variation, for each SNP assay, the allelic ratios for all control and blinded CVS samples were determined in 2 independent experiments, starting from the PCR step. Consistent allelic ratios were detected between the duplicate sets of data (Table 1
). We further randomly selected 4–5 normal samples and 1–2 trisomic samples for each assay and performed a third independent experiment. The CVs of the triplicate experiments for each sample were 0.7% to 22.3% (see Table 4 in the online Data Supplement). All of the selected cases could be assigned correctly as either normal or trisomy according to any one of the triplicate data sets.
In this study, we have shown that MS is a simple and reproducible approach to measure the allelic ratio of a SNP in clinical samples such as placenta, amniotic fluid, and CVS. By comparing the allelic ratios of the heterozygous SNPs, we were able to clearly differentiate between the trisomy 21 and karyotypically normal samples. The allelic ratio detection procedure, including the PCR, hME, and MALDI-TOF steps, is highly automatable and suitable for multiplexing up to 12-plex (11). These features would allow simple, fast, and high-throughput screening procedures. In the future, it would be valuable to perform larger-scale studies to confirm our preliminary data and to further resolve the analytical issues such as reference intervals, assay specificity and sensitivity, and experimental and biological variations.
Acknowledgments
This project is supported by the Innovation and Technology Fund (ITS/195/01). We thank SEQUENOM, Inc. (San Diego, CA) for loaning the MassARRAY Compact system.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  K. C.K. Chow, R. W.K. Chiu, N. B.Y. Tsui, C. Ding, T. K. Lau, T. N. Leung, and Y.M. D. Lo Mass Spectrometric Detection of an SNP Panel as an Internal Positive Control for Fetal DNA Analysis in Maternal Plasma Clin. Chem., January 1, 2007; 53(1): 141 - 142. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, heterozygous sample.