Fax +41 61 306 12 34E-Mail karger@karger.chwww.karger.com

Fetal Diagn Ther 2005;20:275–280 DOI: 10.1159/000085085

Non-Invasive Fetal RHD Exon 7 and Exon 10 Genotyping Using Real-Time PCR Testing of Fetal DNA in Maternal Plasma

Ilona Hromadnikova a Lenka Vechetova a Klara Vesela a Blanka Benesova c Jindrich Doucha b Eduard Kulovany b Radovan Vlk b

a 2nd Clinic of Paediatrics, b

Clinic of Obstetrics and Gynecology, and c Blood Bank, 2nd Medical Faculty,

Charles University, University Hospital Motol, Prague , Czech Republic

These fi ndings may refl ect that DC w – paternally inherited haplotype probably possesses no RHD exon 10. In an-other case no cord blood sample has been available for additional studies. The specifi city of both RHD exon 7 and 10 systems approached 100% since no RhD-positive signals were detected in women currently pregnant with RhD-negative foetus (n = 8). Using real-time PCR and DNA isolated from maternal plasma, we easily differen-tiated pregnant woman whose RBCs had a weak D phe-notype (n = 4) from truly RhD-negative patients since the threshold cycle (C T ) for RHD exon 10 or 7 amplicons reached nearly the same value like C T for control � -globin gene amplicons detecting the total DNA present in ma-ternal plasma. However in these cases foetal RhD status cannot be determined. Conclusion: Prediction of foetal RhD status from maternal plasma is highly accurate and enables implementation into clinical routine. We sug-gest that safe non-invasive prenatal foetal RHD genotyp-ing using maternal plasma should involve the amplifi ca-tion of at least two RHD-specifi c products.

Copyright © 2005 S. Karger AG, Basel

Introduction

Rhesus (Rh) blood group system is the most polymor-phic of the human blood groups, consisting of at least 45 independent antigens and, next to ABO, is the most clin-

Key Words Foetal DNA � Exon 7 � Exon 10 � Maternal plasma � Real-time polymerase chain reaction � RHD gene

Abstract Objective: In this prospective study, we assessed the fea-sibility of foetal RHD genotyping by analysis of DNA ex-tracted from plasma samples of Rhesus (Rh) D-negative pregnant women using real-time PCR and primers and probes targeted toward exon 7 and 10 of RHD gene. Methods: We analysed 24 RhD-negative pregnant wom-an and 4 patients with weak D phenotypes at a gesta-tional age ranging from 11th to 38th week of gestation and correlated the results with serological analysis of cord blood after the delivery. Results: Non-invasive pre-natal foetal RHD exon 7 genotyping analyses of maternal plasma samples was in complete concordance with the serological analysis of cord blood in all 24 RhD-negative pregnant women delivering 12 RhD-positive and 12 RhD-negative newborns. RHD exon-10-specifi c PCR ampli-cons were not detected in 2 out of 12 studied plasma samples from women bearing RhD-positive foetus, de-spite the positive amplifi cation in RHD exon 7 region ob-served in all cases. In 1 case red cell serology of cord blood revealed that the mother had D–C–E–c+e+ C w – and the infant D+C–E–c+e+ C w + phenotypes. RhD exon 10 real-time PCR analysis of cord blood was also negative.

Received: January 15, 2004 Accepted after revision: April 7, 2004

Dr. Ilona Hromadnikova 2nd Clinic of Paediatrics, 2nd Medical Faculty, University Hospital Motol V Uvalu 84, CS–15018 Prague 5 (Czech Republic) Tel. +420 2 24432258, Fax +420 2 24432218 E-Mail ilona.hromadnikova@lfmotol.cuni.cz

© 2005 S. Karger AG, Basel 1015–3837/05/0204–0275$22.00/0

Accessible online at: www.karger.com/fdt

Hromadnikova/Vechetova/Vesela/Benesova/Doucha/Kulovany/Vlk

Fetal Diagn Ther 2005;20:275–280 276

ically signifi cant in transfusion medicine. Haemolytic disease of newborn in 50% of cases is caused by maternal anti-D (IgG) antibody crossing the placenta, binding to foetal red blood cells followed by their destruction caus-ing anaemia. Alloimmunisation due to anti-K and anti-c is in clinical importance next to anti-D [1, 2] . It is now possible to obtain foetally derived DNA using non-inva-sive procedures. Foetal RHD status can be determined in RhD-negative women as most D-negative phenotypes re-sult from a complete depletion of the RhD gene [3] . Cur-rent experimental non-invasive methods using free extra-cellular DNA extracted from maternal peripheral blood (serum or plasma) seems to be a promising alternative for at least foetal gender and RhD status determination. A novel approach uses the real-time quantitative PCR (RQ-PCR) to detect and quantify foetal DNA levels in mater-nal peripheral blood [3, 4, 5–12] .

In this prospective study, we assessed the feasibility of foetal RHD genotyping by analysis of DNA extracted from plasma samples from RhD-negative pregnant wom-en using real-time PCR and primers and probes targeted toward exon 7 and 10 of RHD gene.

Materials and Methods

Blood samples from Caucasoid blood donors both negative (n = 10) and positive (n = 10) on serological testing for RhD were used to establish the RHD exon 7 and 10 real-time PCR assays. There was a complete concordance between the results of RHD real-time PCR genotyping and the serological results.

DNA Extraction from Plasma Samples Five millilitre of maternal peripheral blood from pregnant

women was collected into EDTA-containing tubes and processed within a few hours (maximally 24 h). In details, blood samples were centrifuged fi rst at 1,200 g (protocol 1, P1) [5, 13] and 3,000 g (pro-tocol 2, P2) for 10 min [8, 9] , then plasma samples were re-centri-fuged again and the supernatants were collected and stored at

–80 ° C until further processing. DNA was extracted from 400 � l plasma using QIAamp DNA Blood Mini Kit (Qiagen, Hilden, Ger-many) according to the manufacturer’s instructions. To minimise the risk of contamination, DNA was isolated in laminar airfl ow and aerosol-resistant tips were used. DNA was eluted in 50 � l Buffer AE and 5.0 � l were used as a template for the RHD PCR reaction and 2.5 � l of DNA for the � -globin PCR reaction.

Real-Time PCR Analysis The real-time PCR analysis was performed using ABI PRISM

7700 Sequence Detection System (Applied Biosystem, Branchburg, N.J., USA). The RHD exon 10 TaqMan system consisted of for-ward primers, 5´-CCT CTC ACT GTT GCC TGC ATT-3´; reverse primer 5´-AGT GCC TGC GCG AAC ATT-3´ and a probe 5´-(FAM) TAC GTG AGA AAC GCT CAT GAC AGC AAA GTC T (TAMRA)-3´ [8] . The RHD exon 7 TaqMan system consisted of

forward primer 5´-CTC CAT CAT GGG CTA CAA-3´; reverse primer, 5´-CCG GCT CCG ACG GTA TC-3´ and a probe 5´-(FAM) AGC AGC ACA ATG TAG ATG ATC TCT CCA (TAMRA)-3´ [6] . Sequence data were obtained from the GeneBank Sequence database (RHD gene).

The � -globin TaqMan system consisting of two primers � -glo-bin-354 (forward), 5´-GTG CAC CTG ACT CCT GAG GAG A-3´; � -globin-455 (reverse), 5´-CCT TGA TAC CAA CCT GCC CAG-3´ and a dual-labelled fl uorescent TaqMan probe � -globin-402T, 5´-(FAM) AAG GTG AAC GTG GAT GAA GTT GGT GG (TAM-RA)-3´, served as a control and differentiated between a true nega-tive from a false negative result coming from defi cient DNA extrac-tion process or PCR inhibitors in the eluted DNA [9, 13] . Amplicons for � -globin control gene were detected in all analysed samples.

TaqMan amplifi cation reactions were set up in a reaction vol-ume of 25 � l using the TaqMan Universal PCR Master Mix (Ap-plied Biosystems). Primers and probes were optimised to determine the minimum primer and probe concentrations that gave the max-imum Rn. The RHD exon 10 and � -globin probes were used at concentrations of 100 n M and RHD exon 7 probe at concentration of 200 n M . The PCR primers were used at concentrations of 200 n M and 300 n M . DNA amplifi cations were carried out in 8-well reaction optical tubes/stripes (Applied Biosystem). The Taq-Man PCR conditions were used as described in TaqMan guidelines using 50 cycles of 95 ° C for 15 s and 60 ° C for 1 min with 2-min preincubation at 50 ° C required for optimal AmpErase UNG activ-ity and 10-min preincubation at 95 ° C required for activation of AmpliTaq Gold DNA polymerase. Each sample was analysed in at least 5 replicate settings. Some RhD-negative cases (UPN 4, 8, 9, 772, 775, 788 and 789) were re-tested several times from several vials of maternal plasma samples to make sure that foetal DNA was not really present in maternal circulation. A patient’s specimen was considered positive if one or more individual replicates were posi-tive (threshold cycle C T ! 40).

Results

We analysed 39 RhD-negative pregnant woman (34 non-sensitised and 5 sensitised: 1 ! anti-D, 2 ! anti-D and anti-C, 2 ! non-specifi c antibodies in enzymatic as-say) and 5 patients with weak D phenotypes including 4 pregnant women and 1 blood donor. Serological analysis of cord blood after the delivery has been available until now in 24 RhD-negative pregnant woman (21 non-sensi-tised and 3 sensitised: 1 ! anti-D, 2 ! non-specifi c an-tibodies in enzymatic assay) at a gestational age ranging from 11th to 38th week of gestation. Local Ethics Com-mittee approval and informed consent was obtained for all patients in the study.

Non-invasive prenatal foetal RHD exon 7 genotyping analyses of maternal plasma samples was in complete concordance with the serological analysis of cord blood in all 29 blood samples collected from 24 RhD-negative pregnant women. Twenty-four RhD-negative pregnant

Fetal RHD Exon 7 and Exon 10 Genotyping in Maternal Plasma

Fetal Diagn Ther 2005;20:275–280 277

Table 1. Prenatal diagnosis of foetal RhD status by analysis of DNA extracted from maternal plasma in RhD-negative pregnant women

Number UPN Weeks ofgestation

RHD exon 10CT

RHD exon 7CT

�-GlobinCT

RhDcord blood

Alloantibodies

1 1 33 2/8+P2 7/7+P1 + + Non-specifi c Ab in enzymatic assay38.00 36.20 32.40

2 2 27 0/8– 0/7– + – anti-D 50.00 50.00 32.50

3 4 27 0/22– 0/7– + – No50.00 50.00 32.10

4 6 11 0/8– 0/7– + – No50.00 50.00 32.50

5 7 34 2/8+P2 7/7+P1 + + No38.50 36.30 33.40

6 8 37 0/16– 0/7– + – No50.00 50.00 29.90

7 9 21 0/28– 0/7– + – Non-specifi c Ab in enzymatic assay50.00 50.00 32.79

8 479 A 20 0/7– 6/7+ + + No50.00 36.50 33.50

9 479 B 33 0/7– 5/7+ + + No50.00 36.40 32.10

10 504 33 3/7+P1 7/7+P2 + + No36.50 37.63 32.90

11 574 A 16 6/7+ 6/7+ + + No37.30 36.50 32.80

12 574 B 33 6/7+ 7/7+ + + No35.10 35.20 31.60

13 608 A 16 6/7+ 7/7+ + + No36.24 37.70 31.00

14 608 B 33 7/7+ 5/7+ + + No33.40 34.90 31.40

15 714 33 4/7+ 6/7+ + + No35.50 37.20 32.00

16 743 27 7/7+ 7/7+ + + CVS for risk of X-linked haemophilia,partobulin application in 19th week35.50 37.20 30.30

17 772 38 0/28– 0/7– + – No50.00 50.00 31.95

18 775 23 0/21– 0/7– + – No50.00 50.00 33.74

19 780 25 7/7+ 7/7+ + + No36.35 38.20 33.71

20 787 36 0/21– 7/7+ + + No50.00 35.10 32.19

21 788 A 23 0/28– 0/7– + – No50.00 50.00 32.44

22 788 B 32 0/5– 0/5– + – No50.00 50.00 32.14

23 789 A 25 0/21– 0/7– + – No50.00 50.00 32.51

24 789 B 33 0/5– 0/5– + – No50.00 50.00 33.41

25 795 33 7/7+ 7/7+ + + No35.00 36.30 29.50

26 808 27 0/7– 0/7– + – No50.00 50.00 32.06

27 822 32 0/5– 0/5– + – No50.00 50.00 32.88

28 827 37 5/5+ 5/5+ + + No33.49 35.15 31.68

29 844 37 0/5– 0/5– + – No50.00 50.00 32.42

UPN = Unique patient number.UPN-A and UPN-B means that they are samples from one patient collected at various times during pregnancy.There are a number of positive amplifi cation wells/number of total replicates with the median threshold cycle of RHD exon 7, RHD exon 10 and

control �-globin amplicons.

Hromadnikova/Vechetova/Vesela/Benesova/Doucha/Kulovany/Vlk

Fetal Diagn Ther 2005;20:275–280 278

women delivered 12 RhD-positive and 12 RhD-negative newborns (none of the newborns had weak D pheno-type).

RHD exon 10-specifi c PCR amplicons were detected in 12 out of 15 studied plasma samples from 12 women bearing RhD-positive foetus. In 1 case (P479) the re-anal-ysis made in late pregnancy (33rd week) confi rmed previ-ous negative RHD exon 10 result obtained in 20th week of pregnancy, despite the positive amplifi cation in RHD exon 7 region observed in both maternal plasma samples. In case of P479 RHD exon 10 analysis of cord blood (sam-pling after the delivery) was also negative. Red cell serol-ogy of cord blood revealed that the mother had D–C–E–c+e+ C w – and the infant D+C–E–c+e+ C w + phenotypes. Negative results in RHD exon 10 amplifi cation in case of the patient P479 bearing RhD positive foetus may refl ect that DC w – paternally inherited haplotype probably pos-sesses no RHD exon 10 [1] . In case of P787 no cord blood sample has been available for additional serological stud-ies and for our real-time PCR analysis. The mother re-fused to test the infant to check its Rh phenotype.

The specifi city of both RHD exon 7 and 10 systems approached 100% since no RhD-positive signals were de-tected in 12 women currently pregnant with RhD-nega-tive foetus ( table 1 , fi g. 1 ).

RHD exon 10 assay differs from RHD exon 7 assay in several samples by having a higher or a lower C T value. This small discrepancy not exceeding 2 C T values is caused probably by using DNA extracted from various plasma preparation protocols: protocol 1 (1,200 g ) [7, 15] and protocol 2 (3,000 g ) for 10 min [8, 9] . If we made RHD exon 10 and 7 PCR analysis on the same DNA ex-tracted from one plasma vial, C T were equal for both ex-ons. Equal C T for RHD exon 10 and 7 was observed in case of testing DNA extracted from blood samples of RhD-positive Caucasoid blood donors used to establish both real-time PCR assays. All the pregnant women with weak D phenotype were also tested using the same DNA from one plasma preparation protocol ( table 2 ).

That is a common practice to classify pregnant women whose RBCs weakly expressed D antigens as D negative. They receive D-negative RBC products and prophylactic Rh immunoglobulin preventing alloimmunisation. How-ever weakly expressed D antigens are unlikely to be im-munogenic and do make anti-D alloantibodies rarely [1] . Using real-time PCR and DNA isolated from maternal plasma, pregnant women whose RBCs had a weak D phe-notype have been easily differentiated from truly RhD-negative patients since the C T for RHD exon 10 or 7 am-plicons reached nearly the same value like C T for control

� -globin gene amplicons detecting the total DNA present in maternal plasma ( table 2 , fi g. 2 ). However in these cases foetal RHD status cannot be determined.

Discussion

Prediction of foetal Rhesus D status from maternal plasma is highly accurate and enables implementation into clinical routine [4–6, 8, 12, 14] . We confi rmed this fi nding also, however, our study cohort involved predom-inantly pregnant women in second and third trimester of pregnancy.

To avoid false negative results the assay performing foetal RHD genotyping using maternal plasma samples should involve the amplifi cation of at least two RhD-spe-cifi c products. We suggest that safe non-invasive prenatal foetal RHD genotyping using maternal plasma should

Fig. 1. Non-invasive foetal RHD genotyping by analysis of DNA extracted from maternal plasma using real-time PCR. Threshold cycle (C T ) of foetal-derived RHD exon 10 and 7 amplicons is de-layed against C T for control � -globin gene amplicons detecting the total DNA extracted from maternal plasma. Fig. 2. Determination of weak D phenotype in mother by analysis of DNA extracted from maternal plasma using real-time PCR. Pregnant woman whose RBCs have a weak D phenotype can be easily differentiated from truly RhD-negative patients since the threshold cycle (C T ) for RHD exon 10 or 7 amplicons reach nearly the same value like C T for control � -globin gene amplicons to detect the total DNA extracted from maternal plasma.

Table 2. Determination of weak D phenotype in mother by analy-sis of DNA extracted from maternal plasma

No. UPN Weeks ofpregnancy

RHD exon 10CT

RHD exon 7CT

�-GlobinCT

1 3-A 25 5/5+33.00

5/5+33.00 33.20

2 3-B 27 5/5+30.80

5/5+31.41 31.31

3 818 12 5/5+30.20

5/5+30.10 30.00

4 834 38 3/3+28.90

2/2+29.30 31.04

5 760 Blooddonor

5/5+29.30

2/2+30.00 30.00

Fetal RHD Exon 7 and Exon 10 Genotyping in Maternal Plasma

Fetal Diagn Ther 2005;20:275–280 279

1

2

Hromadnikova/Vechetova/Vesela/Benesova/Doucha/Kulovany/Vlk

Fetal Diagn Ther 2005;20:275–280 280

follow the recommended procedure for RHD genotyping of partial D and weak D [15] . Checking one of the RhD-specifi c nucleotides in RHD exon 7 and at least one sec-ond region, exon 4, intron 4 [4] or exon 10 of RHD gene. RHD genotyping based solely on RHD-specifi c sequenc-es in the 3´ UTR of RHD exon 10 was not considered a safe procedure [15] .

Foetal RHD genotyping may allow the identifi cation of foetuses at risk of haemolytic disease of newborn [1, 2] . When D-positive foetuses are identifi ed and anti-D alloantibodies have already been present then anti-D al-loantibody titre variations are important to be carefully monitored during the pregnancy in order to perform foe-tal blood sampling in time to determine the level of foetal haemolysis and start early treatment of sensitised RhD-negative pregnancies (ultraviolet phototherapy, exchange or intrauterine transfusion if required). Invasive proce-dures are used as a last option in monitoring haemolytic disease of newborn, because they may cause further leak-age of foetal RBCs into the maternal circulation. On the other hand the detection of D-negative foetuses in the current pregnancy will exclude the need to perform inva-sive procedures despite the presence of anti-D alloanti-bodies that may be present in maternal circulation from previous pregnancies for a variety of reasons, including nonadministration of Rh immunoglobulin, unrecognised

miscarriage, leakage of foetal RBCs into the maternal cir-culation in late pregnancy, etc.

However, we applied non-invasive prenatal RHD ge-notyping assay in the routine testing of all RhD-negative women involving both sensitised pregnancies as well as those nonsensitised, which is the majority. In case of non-sensitised RhD-negative pregnant women carrying RhD-positive foetus, RhD immune globulin may be adminis-tered precautionary to minimise the alloimmunisation due to foetomaternal transfusion occurring in 3rd trimes-ter of pregnancy, however in RhD-negative foetuses it can be avoided.

Acknowledgements

This project was supported by 2nd Medical Faculty, Charles University in Prague, No. VZ 111300003 as well as MSM 0021620806. This project was also supported by MZO 00064203 and MSM 0021620806. We thank Dr. Martin Pisacka from Insti-tute of Haematology and Blood Transfusion for his helpful advice.

The authors greatly appreciate the assistance of Prof. Dr. Lukas Rob, Dr. Ivana Spalova, Dr. Tomas Binder, Dr. Miluse Mrstinova, Dr. Karel Hynek, Dr. Milada Brandejska, Dr. Blanka Vavrinkova, Dr. Leos Teslik, Dr. Lenka Oborna, Dr. Dita Holdsvendova, Dr. Ladislav Krofta, Dr. Pavla Vavrinkova, Dr. Radka Vetesnikova, Dr. Helena Souckova and Dr. Pavla Vavrincova.

References

1 Avent ND, Reid ME: The Rh blood group sys-tem: A review. Blood 2000; 95: 375–387.

2 Avent ND: Antenatal genotyping of the blood groups of the fetus. Vox Sang 1998; 74: 365–374.

3 Colin Y, Cherif-Zahar B, Le Van Kim C, Raynal V, Van Huffel V, Cartron JP: Genetic basis of the RhD-positive and RhD-negative blood group polymorphism as determined by Southern analysis. Blood 1991; 78: 2747–2752.

4 Faas BH, Beuling EA, Christiaens GC, von dem Borne AE, van der Achoot CE: Detection of fetal RhD-specifi c sequences in maternal plasma. Lancet 1998; 352: 1196.

5 Hahn S, Zhong XY, Burk MR, Troeger C, Holzgreve W: Multiplex and realtime quanti-tative PCR on fetal DNA in maternal plasma: A comparison with fetal cells isolated from ma-ternal blood. Ann NY Acad Sci 2000; 906: 148–152.

6 Legler TJ, Lynen R, Maas JH, Pindur G, Ku-lenkampff D, Suren A, Osmers R, Kohler M: Prediction of fetal RhD and Rh CcEe pheno-type from maternal plasma with real-time polymerase chain reaction. Transfus Apheresis Sci 2002; 27: 217–223.

7 Lo YMD, Corbetta N, Chamberlain PF, Sar-gent JL: Presence of fetal DNA in maternal plasma and serum. Lancet 1997; 350: 485–487.

8 Lo YMD, Hjelm NM, Fidler C, Sargent IL, Murphy MF, Chamberlain PF, Poon PM, Red-man CW, Wainscoat JS: Prenatal diagnosis of fetal RhD status by molecular analysis of ma-ternal plasma. N Engl J Med 1998; 339: 1734–1738.

9 Lo YMD, Tein MS, Lau TK, Haines CJ, Leung TN, Poon PM, Wainscoat JS, Johnson PJ, Chang AM, Hjelm NM: Quantitative analysis of fetal DNA in maternal plasma and serum: Implications for noninvasive prenatal diagno-sis. Am J Hum Genet 1998; 62: 768–775.

10 Poon LL, Lo YM: Circulating fetal DNA in maternal plasma. Clin Chim Acta 2001; 313: 151–155.

11 Zhong XY, Hahn S, Holzgreve W: Prenatal identifi cation of fetal genetic traits. Lancet 2001; 357: 310–311.

12 Zhong XY, Holzgreve W, Hahn S: Detection of fetal rhesus D and sex using fetal DNA from maternal plasma by multiplex PCR. Brit J Ob-stet Gynaecol 2000; 107: 766–769.

13 Hromadnikova I, Houbova B, Hridelova D, Voslarova S, Kofer J, Komrska V, Habart D: Replicate real-time PCR testing of DNA in ma-ternal plasma increases the sensitivity of non-invasive fetal sex determination. Prenat Diagn 2003; 23: 235–238.

14 Cotorruelo C, Biondi C, Garcia Borras S, Di Monaco R, Racca A: Early detection of RhD status in pregnancies at risk of hemolytic dis-ease of the newborn. Clin Exp Med 2002; 2: 77–81.

15 Flegel WA, Wagner FF: Molecular biology of partial D and weak D: Implications for blood bank practise. Clin Lab 2002; 48: 53–59.