BIOLOGY OF REPRODUCTION 82, 745–750 (2010)Published online before print 14 January 2010.DOI 10.1095/biolreprod.109.082271

MALDI-TOF Mass Array Analysis of RASSF1A and SERPINB5 Methylation Patternsin Human Placenta and Plasma1

Maria Luz Bellido,3,4 Ramin Radpour,3,5 Olav Lapaire,6 Isabelle De Bie,5 Irene Hosli,6 Johannes Bitzer,6

Abdelkrim Hmadcha,4 Xiao Yan Zhong,2,5 and Wolfgang Holzgreve7

Department of Cell Therapy and Regenerative Medicine,4 Andalusian Center for Molecular Biology and RegenerativeMedicine (CABIMER), Sevilla, SpainLaboratory for Prenatal Medicine and Gynecologic Oncology,5 Women’s Hospital/Department of Biomedicine,University of Basel, SwitzerlandDepartment of Obstetrics and Gynecology,6 University Hospital of Basel, Basel, SwitzerlandUniversity Medical Center Freiburg,7 Freiburg, Germany

ABSTRACT

Differences in DNA methylation patterns between placentaand blood cells of pregnant women have been suggested aspotential biomarkers for noninvasive prenatal diagnostic strat-egies, including for common obstetrical complications, such aspreeclampsia. New findings in epigenetic origins of fetal orplacental disorders may improve our ability for optimalmanagement of these conditions. Using a novel high-throughputmass spectrometry on matrix-assisted laser desorption/ioniza-tion time-of-flight (MALDI-TOF) mass array, we compared thequantitative methylation changes of RASSF1 and SERPINB5(also known as MASPIN) genes in placenta and plasma samples.We analyzed the methylation status of a total of 3569 CpGdinucleotides on these two genes in 83 different samples: 50plasma samples (20 from pregnant women and 30 fromnonpregnant women) and 33 placenta tissue samples (25 fromnormal pregnancies and eight from preeclamptic pregnancies).The aim of this study was to assess the utility of epigeneticchanges as biomarkers for noninvasive prenatal diagnosticprocedures. Using a two-way hierarchical cluster analysis,significantly different methylation levels of the RASSF1 genewere found between placenta (normal and preeclamptic) andplasma samples of pregnant women. Although the SERPINB5gene was hypomethylated in placenta DNA more than inplasma DNA, it did not demonstrate significant differencesbetween studied groups. The MALDI-TOF mass spectrometryanalysis of placenta and plasma DNA methylation patterns mayserve as a tool for the study of gender-independent biomarkersin noninvasive prenatal diagnosis.

DNA methylation, MALDI-TOF, mass spectrometry, placenta,preeclampsia, pregnancy

INTRODUCTION

Prenatal diagnosis of fetal aneuploidies or monogenicdisorders currently requires that fetal genetic material beobtained by amniocentesis, chorionic villus sampling, orcordocentesis. These invasive procedures are associated witha significant risk of fetal loss (0.5% and 1% for amniocentesisand chorionic villus sampling, respectively) [1]. For thisreason, there is an urgent need for the development ofdiagnostic procedures that do not put the fetus at risk.

The discovery that cell-free fetal DNA (cffDNA) is presentin the maternal circulation during pregnancy became a focusfor alternative approaches toward the development of nonin-vasive prenatal procedures [2, 3]. cffDNA has been success-fully used for the determination of fetal sex and fetal RhDstatus in maternal plasma [4, 5]. It was also shown that therelative proportion of cffDNA in maternal plasma is increasedin pregnancies with aneuploid conceptuses, as well as in certainobstetrical complications, such as preeclampsia [6, 7].

Two major aspects currently hamper the widespread use ofnoninvasive prenatal diagnostic procedures. First, the limitedamount of cffDNA (3%–6%) in the overwhelming backgroundof maternal cell-free DNA presents a great challenge for thedetection and analysis of circulating fetal genetic material [6].Second, the lack of a fetal universal marker makes theinterpretation of negative results problematic, particularly inthe context of low cffDNA levels. The search for such auniversal marker has becomes particularly important forquantitative and qualitative applications of cffDNA analysis.

Recently, it was discovered that physical and molecularcharacteristics of cffDNA could be used to discriminate or enrichthis specific fraction from the surrounding circulating maternalcell-free DNA [8, 9]. Among those characteristics, epigeneticmodifications between blood and placental DNA were used as amean to specifically detect fetal DNA [8] because the source ofcffDNA is the syncytiotrophoblast [10]. Two genes, SERPINB5(also known as mammary serine protease inhibitor [MASPIN])and RASSF1 (Ras association [RalGDS/AF-6] domain familymember 1) were initially used as markers to detect theseepigenetic changes. Both genes could present different methyl-ation status in placenta and blood cell DNA as an epigenetic fetalDNA marker for noninvasive prenatal diagnosis [11, 12].

Both RASSF1 and SERPINB5 are tumor suppressor genes.RASSF1 is frequently inactivated by promoter hypermethyla-tion in at least 37 tumor types, including lung, breast, prostate,and kidney [13, 14]. The placenta seems to be the onlynonmalignant tissue where RASSF1 hypermethylation has alsobeen demonstrated [15]. The SERPINB5 product protease

1Supported in part by Swiss National Science Foundation grant 320000-119722/1, the Swiss Cancer League, Krebsliga Beider Basel, and Dr. HansAltschueler Stiftung. M.L.B. is a Fund for Health Research (FIS) researchersupported by the Spanish Ministry of Health (the Carlos III Health Institute;grant CM06/00183) and the Foundation Progreso y Salud.2Correspondence: Xiao Yan Zhong, Laboratory for Prenatal Medicineand Gynecologic Oncology, Women’s Hospital/Department of Bio-medicine, University of Basel, Hebelstrasse 20, Room 420, CH 4031Basel, Switzerland. FAX: 41 61 265 9399; e-mail: xzhong@uhbs.ch3These authors contributed equally to this work.

Received: 30 October 2009.First decision: 23 November 2009.Accepted: 5 January 2010.� 2010 by the Society for the Study of Reproduction, Inc.eISSN: 1529-7268 http://www.biolreprod.orgISSN: 0006-3363

745

Dow

nloaded from w

ww

.biolreprod.org.

inhibitor 5 has been shown to have inhibitory actions onmotility, invasion, metastasis, and angiogenesis in humanbreast cancer and prostate cancer [16, 17]. SERPINB5expression is differentially regulated during human placentadevelopment [11]. Very low levels of SERPINB5 are detectedin the first trimester, whereas these levels increase in the earlysecond trimester and remain elevated in the third trimester ofpregnancy. Although the exact role of SERPINB5 in humangestation remains unclear, in vitro data suggest that SERPINB5may regulate trophoblast cell invasion.

Sequenom’s EpiTYPER assay (San Diego, CA) created theopportunity for high-throughput quantitative analysis of DNAmethylation status using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOFMS) and MassCLEAVETM reagent (Sequenom), which isbased on base-specific (C/T) cleavage reactions [18, 19]. Therobustness of this approach for quantifying methylated and

unmethylated DNA has been previously demonstrated by theSequenom group [20] and ourselves [21, 22].

In the present study, we compared RASSF1 and SERPINB5methylation patterns in the placenta from normal andpreeclamptic pregnancies, as well as in the plasma frompregnant and nonpregnant women using the high-throughputMALDI-TOF MS assay. The aim of this study was to assessthe utility of these epigenetic changes as potential biomarkersfor noninvasive prenatal diagnostic procedures.

MATERIALS AND METHODS

Samples

This study was performed in the laboratory of Prenatal Medicine and

Gynecologic Oncology at the University Women’s Hospital and Department of

Biomedicine, University of Basel, Switzerland. Informed consent was obtained

from all participants. In total, 50 blood samples were used in this study: 30 from

nonpregnant women and 20 from pregnant women. Thirty-three placenta

samples were collected: 25 from women with normal pregnancies (median

38 6 4 wk of gestation) and eight from preeclamptic pregnant women (median

34 6 3 wk of gestation). Local ethics institutional review boards had approved

use of these human samples.

Plasma DNA Preparation

Blood samples were collected in ethylenediaminetetraacetic acid tubes and

centrifuged at 1600 3 g (10 min), and plasma was carefully transferred into 2-

ml microtubes. Samples were centrifuged in a microcentrifuge at full speed (10

min), and supernatants were stored at �808C until analysis was performed.

DNA Extraction

DNA extraction was performed from 25–50 mg of placenta tissue using the

High Pure PCR Template Preparation Kit (Roche Diagnostics). Cell-free DNA was

extracted from 500 ll of plasma using the High Pure PCR Template Preparation Kit

as previously published [23], and it was eluted in a final volume of 100 ll.

TABLE 1. Sequences of PCR tagged primers and PCR conditions for in vitro transcription.

Gene Primer* Sequence (50!30) Length Ta

(8C)� Product size (bp)

SERPINB5 tag-FW AGGAAGAGAGTGAGAAATTTGTAGTGTTATTATT 10þ24 57 436T7-RV CAGTAATACGACTCACTATAGGGAGAAGGCTCACCTTACTTACCTAAAATCAC 31þ22

RASSF1 tag-FW AGGAAGAGAGGGGYGGTAAAGTTGTTGA 10þ18 56 346T7-RV CAGTAATACGACTCACTATAGGGAGAAGGCTCRCAATAAAAACCTAAATACA 31þ21

* FW, forward; RV, reverse.� T

a, annealing temperature.

FIG. 1. Primers for in vitro transcription. A) Reverse primer with T7-promoter tag. B) Forward primer with 10-mer tag sequence as balance.

TABLE 2. High-throughput methylation analysis of informative CpG sites per amplicons for two studied genes.*

GeneAmpliconsize (bp)

Total no. of CpGsites in amplicon

No. of analyzed CpGsites in amplicon

No. of analyzed CpG sites per amplicons�

Single sites Composite sites

SERPINB5 436 21 13 3 10RASSF1 346 22 11 5 6

* The in silico digestion was performed for the T-cleavage assay.� The percentage of informative CpG sites (total sites) in the amplicon is divided into single sites (single CpG sites)and composite sites (two or three adjacent CpG sites fall within one fragment, or when fragment masses areoverlapping).

"

FIG. 2. High-throughput analysis of the informative CpG sites for SERPINB5 tumor suppressor gene. A) Amplicon size and place of CpG sites in theamplicon. B) The two-way hierarchical cluster analysis of methylation patterns between different studied groups (red clusters indicate 0% methylated,yellow clusters indicate 100% methylated, color gradient between red and yellow indicates methylation ranging from 0 to 100, and black clusters indicateCpG sites not analyzed). C) Comparison of methylation patterns in nonpregnant and pregnant plasma samples. D) Comparison of methylation patternsbetween placenta (normal and preeclamptic) and pregnant plasma samples.

746 BELLIDO ET AL.

Dow

nloaded from w

ww

.biolreprod.org.

METHYLATION PROFILE OF PLACENTA 747

Dow

nloaded from w

ww

.biolreprod.org.

748 BELLIDO ET AL.

Dow

nloaded from w

ww

.biolreprod.org.

Bisulfite Treatment

To perform bisulfite conversion of the target sequence, the Epitect BisulfiteKit (Qiagen AG, Basel, Switzerland) was used according to the manufacturer’sprotocol.

Primer Design and PCR Tagging for EpiTYPER Assay

We designed primers for the RASSF1 and SERPINB5 genes to cover theregions with the most CpG sites (Table 1). Selected amplicons were mostlylocated in the two genes’ promoter regions, or they started from the promoterand ended in the first exon. The primers were designed using MethPrimer [24].For PCR amplification, a T7-promoter tag was added to the reverse primer, anda 10-mer tag sequence was added to the forward primer to balance the PCRprimer length (Fig. 1). The following PCR conditions were used foramplification of the bisulfite-treated genomic DNA: one cycle, 958C for 10min; 48 cycles, 958C for 20 sec; T

afor 30 sec (Table 1); 728C for 1 min; and

one cycle, 728C for 5 min. After the first round of PCR amplification, 1.5 ll ofthe PCR product was used for reamplification of selected amplicons with thesame primer pairs and conditions.

In Vitro Transcription and T-Cleavage Assay

Unincorporated dinucleotide triphosphates (dNTPs) were removed byshrimp alkaline phosphatase (SAP; Sequenom) treatment. Typically, 2 ll of thePCR product was then directly used as template for the transcription reaction.Twenty units of T7 R&DNA polymerase (Epicentre, Madison, WI) were usedto incorporate thymidine triphosphate in the transcripts. Ribonucleotides anddNTPs were used at concentrations of 1 mmol/L and 2.5 mmol/L, respectively.In the same step, RNase-A (Sequenom) was added to cleave the in vitrotranscripts (T-cleavage assay). Samples were diluted with H

2O to a final

volume of 27 ll. Conditioning of the phosphate backbone was achieved byadding 6 mg of Clean Resin (Sequenom) before performing MALDI-TOF MSanalysis.

Mass Spectrometry

A total of 22 nl of the RNase-A treated product was robotically dispensedonto silicon matrix preloaded chips (SpectroCHIP; Sequenom), the massspectra were collected using a MassARRAY Compact MALDI-TOF(Sequenom), and spectra’s methylation ratios were generated by the EpiTYPERsoftware v1.0 (Sequenom).

Statistical Methods

Data analysis was performed using SPSS software (Statistical SoftwarePackage for Windows, v17). The Shapiro-Wilk test and the Kolmogorov-Smirnov test were used to analyze normality. The two tests showed that ourdata set was not normally distributed (P , 0.0001 by Shapiro-Wilk test and P. 0.0001 by Kolmogorov-Smirnov). Methylation rates in the different samplegroups were compared with the Mann-Whitney U Test. All P values were twosided; P , 0.05 was considered statistically significant.

Using the two-way hierarchical cluster analysis, the most variable CpGfragments for each gene were clustered based on pairwise Euclidean distancesand linkage algorithm for all samples, according to the previously developedmethod [21]. The gene clustering was performed independently for each gene,based on the methylation ratio in the studied groups using a hierarchicalalgorithm. The procedure was performed using the double dendrogram functionof the Gene Expression Statistical System for Microarrays (GESS) version7.1.13 (NCSS, Kaysville, UT).

RESULTS

In this study, we analyzed the methylation patterns of twogenes (RASSF1 and SERPINB5) in 33 placenta tissue samplesand 50 plasma samples. The two analyzed amplicons in theselected genes contained CpG-rich islands (with the number of

CpG sites higher than 20; Table 2). For both RASSF1 andSERPINB5 genes, 43 CpG sites per sample were analyzed(total of 3569 CpG sites in 83 samples; Table 2). For theRASSF1 gene, .50% of the CpG sites in amplicons wereamenable to analysis, whereas for SERPINB5, .62% of CpGsites in amplicons were detected (Table 2). In the RASSF1gene, 43% of CpG sites were found to be at a low degree ofmethylation, whereas 32% of CpG sites were at a high degreeof methylation. In the SERPINB5 gene, about 53% of CpGsites were defined as having a low degree of methylation, withmean methylation levels below 30%, and 17% of CpG siteswere defined as having a high degree of methylation, withmean methylation levels above 80%. Using the two-wayhierarchical cluster analysis, significant differences in themethylation levels of the RASSF1 gene were demonstratedbetween genomic DNA isolated from placenta samples (normaland preeclamptic) and cell-free DNA isolated from pregnantwomen’s plasma (P , 0.05). However, no significantdifference between the methylation pattern of normal andpreeclamptic placenta samples was detected (Fig. 2D). Thehigh-throughput methylation profiling of the SERPINB5 genedid not reveal a significant difference between placenta andplasma samples from pregnant women (P . 0.05; Fig. 3D).Analysis failed to detect significant methylation patternsdifferences for these two genes between plasma samples frompregnant and nonpregnant women (P . 0.05; Figs. 2C and3C). The pairwise analysis of SERPINB5 and RASSF1 genestogether could not show any significant correlation in theanalyzed sample groups (P . 0.05).

DISCUSSION

In this study, we analyzed the methylation status of twotumor suppressor genes in placenta samples from preeclampticand normal pregnancies, as well as in the plasma of pregnantand nonpregnant women. Using hierarchical clustering, wefound that CpG sites in the promoter regions of RASSF1 geneexhibited significantly different methylation degrees betweenthe studied samples (Fig. 2).

Development of sex- and polymorphism-independent fetalDNA markers via epigenetic approaches has recently beenpursued for the purpose of noninvasive prenatal diagnostics[25]. Selected pairs of fetal-maternal-imprinted loci were usedto show the feasibility of this approach for the detection of fetalDNA in maternal plasma.

In one study using nonquantitative bisulfite sequencing,Chim et al. [11] reported the hypomethylation status ofSERPINB5 gene in placenta DNA in comparison with plasmaDNA. In that study, SERPINB5 was used for the follow-up ofclinical conditions associated with decreased cytotrophoblastinvasion, such as preeclampsia and intrauterine growthretardation. After methylation quantification of the SERPINB5gene using MALDI-TOF MS, our results showed hypomethy-lation patterns for placenta DNA (normal and preeclamptic)compared with plasma DNA, but this was not statisticallysignificant (P . 0.05). This kind of discordance between ourresults with previous reports might be due to either using adifferent sample size or different methodological approaches.We also compared the average methylation value of SERPINB5

3

FIG. 3. High-throughput analysis of the informative CpG sites for RASSF1 tumor suppressor gene. A) Amplicon size and place of CpG sites in theamplicon. B) The two-way hierarchical cluster analysis of methylation patterns between different studied groups (red clusters indicate 0% methylated,yellow clusters indicate 100% methylated, color gradient between red and yellow indicates methylation ranging from 0 to 100, and black clusters indicateCpG sites not analyzed). C) Comparison of methylation patterns in nonpregnant and pregnant plasma samples. D) Comparison of methylation patternsbetween placenta (normal and preeclamptic) and pregnant plasma samples. *Significant correlation; Mann-Whitney U test.

METHYLATION PROFILE OF PLACENTA 749

Dow

nloaded from w

ww

.biolreprod.org.

in normal and preeclamptic placenta tissues but could notdetect a significant difference. Significant differences wereonly observed at a few CpG sites (Fig. 3B). The confirmationof the methylation profile of SERPINB5 as a biomarker innormal and preeclamptic placenta would need further evalu-ation in different ethnic groups.

The high-throughput profiling of methylation of RASSF1gene revealed hypermethylated patterns in placenta DNA(normal and preeclamptic) but hypomethylated patterns in cell-free DNA from plasma of pregnant women (Fig. 3D). Thisconfirms the findings of a previously published study [12]. Itwas previously reported that the levels of hypermethylatedRASSF1 progressively increase in maternal plasma in pre-eclamptic pregnancies [26]. Unfortunately, we did not haveaccess to serial plasma samples from patients with thisobstetrical complication, and therefore we could not confirmthis finding with our method.

No significant differences were found in the methylationlevels of RASSF1 and SERPINB5 genes from the plasma ofnonpregnant and pregnant women. This may result from thetechnical limitations inherent to analyzing a low amount ofcffDNA in the presence of excess free maternal DNA [6].

The potential clinical implication of our study is thatepigenetic modifications of RASSF1 show promise as potentialbiomarkers for the follow-up of abnormal placentationdisorders.

ACKNOWLEDGMENTS

We thank Vivian Kiefer for her excellent technical assistance. We areindebted to the patients for their participation in this study.

REFERENCES

1. Hulten MA, Dhanjal S, Pertl B. Rapid and simple prenatal diagnosis ofcommon chromosome disorders: advantages and disadvantages of themolecular methods FISH and QF-PCR. Reproduction 2003; 126:279–297.

2. Lo YM, Corbetta N, Chamberlain PF, Rai V, Sargent IL, Redman CW,Wainscoat JS. Presence of fetal DNA in maternal plasma and serum.Lancet 1997; 350:485–487.

3. Dennis Lo YM, Chiu RW. Prenatal diagnosis: progress through plasmanucleic acids. Nat Rev Genet 2007; 8:71–77.

4. Lo YM, Hjelm NM, Fidler C, Sargent IL, Murphy MF, Chamberlain PF,Poon PM, Redman CW, Wainscoat JS. Prenatal diagnosis of fetal RhDstatus by molecular analysis of maternal plasma. N Engl J Med 1998; 339:1734–1738.

5. Bianchi DW, Avent ND, Costa JM, van der Schoot CE. Noninvasiveprenatal diagnosis of fetal Rhesus D: ready for Prime(r) Time. ObstetGynecol 2005; 106:841–844.

6. Lo YM, Tein MS, Lau TK, Haines CJ, Leung TN, Poon PM, WainscoatJS, Johnson PJ, Chang AM, Hjelm NM. Quantitative analysis of fetalDNA in maternal plasma and serum: implications for noninvasive prenataldiagnosis. Am J Hum Genet 1998; 62:768–775.

7. Hahn S, Chitty LS. Noninvasive prenatal diagnosis: current practice andfuture perspectives. Curr Opin Obstet Gynecol 2008; 20:146–151.

8. Poon LL, Leung TN, Lau TK, Chow KC, Lo YM. Differential DNA

methylation between fetus and mother as a strategy for detecting fetalDNA in maternal plasma. Clin Chem 2002; 48:35–41.

9. Chan KC, Zhang J, Hui AB, Wong N, Lau TK, Leung TN, Lo KW, HuangDW, Lo YM. Size distributions of maternal and fetal DNA in maternalplasma. Clin Chem 2004; 50:88–92.

10. Alberry M, Maddocks D, Jones M, Abdel Hadi M, Abdel-Fattah S, AventN, Soothill PW. Free fetal DNA in maternal plasma in anembryonicpregnancies: confirmation that the origin is the trophoblast. Prenat Diagn2007; 27:415–418.

11. Chim SS, Tong YK, Chiu RW, Lau TK, Leung TN, Chan LY, OudejansCB, Ding C, Lo YM. Detection of the placental epigenetic signature of themaspin gene in maternal plasma. Proc Natl Acad Sci U S A 2005; 102:14753–14758.

12. Chan KC, Ding C, Gerovassili A, Yeung SW, Chiu RW, Leung TN, LauTK, Chim SS, Chung GT, Nicolaides KH, Lo YM. HypermethylatedRASSF1A in maternal plasma: a universal fetal DNA marker thatimproves the reliability of noninvasive prenatal diagnosis. Clin Chem2006; 52:2211–2218.

13. Agathanggelou A, Cooper WN, Latif F. Role of the Ras-associationdomain family 1 tumor suppressor gene in human cancers. Cancer Res2005; 65:3497–3508.

14. Hesson LB, Cooper WN, Latif F. The role of RASSF1A methylation incancer. Dis Markers 2007; 23:73–87.

15. Chiu RW, Poon LL, Lau TK, Leung TN, Wong EM, Lo YM. Effects ofblood-processing protocols on fetal and total DNA quantification inmaternal plasma. Clin Chem 2001; 47:1607–1613.

16. Sternlicht MD, Safarians S, Rivera SP, Barsky SH. Characterizations ofthe extracellular matrix and proteinase inhibitor content of humanmyoepithelial tumors. Lab Invest 1996; 74:781–796.

17. Zhang M, Volpert O, Shi YH, Bouck N. Maspin is an angiogenesisinhibitor. Nat Med 2000; 6:196–199.

18. Ehrich M, Nelson MR, Stanssens P, Zabeau M, Liloglou T, Xinarianos G,Cantor CR, Field JK, van den Boom D. Quantitative high-throughputanalysis of DNA methylation patterns by base-specific cleavage and massspectrometry. Proc Natl Acad Sci U S A 2005; 102:15785–15790.

19. Stanssens P, Zabeau M, Meersseman G, Remes G, Gansemans Y, StormN, Hartmer R, Honisch C, Rodi CP, Bocker S, van den Boom D. High-throughput MALDI-TOF discovery of genomic sequence polymorphisms.Genome Res 2004; 14:126–133.

20. Ehrich M, Turner J, Gibbs P, Lipton L, Giovanneti M, Cantor C, van denBoom D. Cytosine methylation profiling of cancer cell lines. Proc NatlAcad Sci U S A 2008; 105:4844–4849.

21. Radpour R, Haghighi MM, Fan AX, Torbati PM, Hahn S, Holzgreve W,Zhong XY. High-throughput hacking of the methylation patterns in breastcancer by in vitro transcription and thymidine-specific cleavage mass arrayon MALDI-TOF silico-chip. Mol Cancer Res 2008; 6:1702–1709.

22. Radpour R, Kohler C, Haghighi MM, Fan AX, Holzgreve W, Zhong XY.Methylation profiles of 22 candidate genes in breast cancer using high-throughput MALDI-TOF mass array. Oncogene 2009; 28:2969–2978.

23. Zimmermann B, El-Sheikhah A, Nicolaides K, Holzgreve W, Hahn S.Optimized real-time quantitative PCR measurement of male fetal DNA inmaternal plasma. Clin Chem 2005; 51:1598–1604.

24. Li LC, Dahiya R. MethPrimer: designing primers for methylation PCRs.Bioinformatics 2002; 18:1427–1431.

25. Wright CF, Burton H. The use of cell-free fetal nucleic acids in maternalblood for non-invasive prenatal diagnosis. Hum Reprod Update 2009; 15:139–151.

26. Tsui DW, Chan KC, Chim SS, Chan LW, Leung TY, Lau TK, Lo YM,Chiu RW. Quantitative aberrations of hypermethylated RASSF1A genesequences in maternal plasma in pre-eclampsia. Prenat Diagn 2007; 27:1212–1218.

750 BELLIDO ET AL.

Dow

nloaded from w

ww

.biolreprod.org.