- Email: [email protected]
Large scale blood group genotyping
Transfusion Clinique et Biologique 14 (2007) 10–15 http://france.elsevier.com/direct/TRACLI/
Plenary session
Large scale blood group genotyping Ge´notypage des groupes sanguins a` grande e´chelle Neil D. Avent Centre for Research in Biomedicine, Bristol Genomics Research Institute, Faculty of Applied Sciences, University of the West of England, Bristol, United Kingdom Available online 23 May 2007
Abstract Determination of predicted blood group phenotype by determination of genotype has been performed since the 1990s. This evolved due to the rapid accrual of information surrounding the molecular basis of blood group antigen expression, which started in 1990 with ABO and RH systems and has now resulted in the molecular description of 28 of the 29 blood groups. Blood group genotyping is currently performed mostly for fetal blood group incompatibility and for assessment of multi-transfused patients. Both of these clinical scenarios are either dangerous or technically difficult, respectively to define serologically. With the simultaneous development of mass scale genotyping platforms it has now permitted the application of them to blood group genotype determination. In this paper, I describe some recently published work that has demonstrated that mass scale genotyping approaches are feasible. These approaches may lead to more effective management of blood stocks and patient cross-matching by reducing the dependence on serology during the time critical pre-transfusion phase. It is most probable that large scale studies, perhaps involving many European Union and North American based blood suppliers, may drive the introduction of this technology and convince red cell serologists that this approach may allow their work to be more focussed. # 2007 Elsevier Masson SAS. All rights reserved. Re´sume´ La pre´diction du phe´notype a` partir de la de´termination du ge´notype a e´te´ mise en oeuvre depuis les anne´es 1990. Cette e´volution est due a` l’accumulation rapide d’informations entourant la base mole´culaire de l’expression des groupes sanguins qui a commence´ en 1990 avec les syste`mes ABO et RH, et qui concerne maintenant la description mole´culaire de 28 parmi les 29 syste`mes de groupes. Le ge´notypage des groupes sanguins est couramment pratique´ dans les cas d’incompatibilite´ foetomaternelle et pour le suivi des malades polytransfuse´s. Ces deux situations cliniques sont respectivement, soit dangereuse, soit techniquement difficile a` de´finir se´rologiquement. Le de´veloppement simultane´ de diverses plates-formes de ge´notypage de masse a permis d’envisager des applications a` la de´termination des ge´notypes de groupes sanguins. Dans cette revue, je de´crirai quelques travaux publie´s re´cemment qui de´montrent que le ge´notypage de masse est faisable. Ces approches pourraient conduire a` une meilleure gestion des stocks de sangs et du cross-match des malades en re´duisant la part des techniques se´rologiques pendant la phase critique pre´transfusionnelle. Il est tre`s probable que des e´tudes a` grande e´chelle impliquant de nombreuses structures responsables de la distribution du sang, en Europe et aux E´tats-Unis, pourraient conduire a` l’introduction de cette technologie et a` convaincre les se´rologistes que cette approche leur permettra de mieux focaliser leurs travaux. # 2007 Elsevier Masson SAS. All rights reserved. Keywords: Blood group; Polymorphism; Genotyping; SNP; DNA chips Mots cle´s : Groupes sanguins ; Polymorphisme ; Ge´notype ; SNP ; Puces a` ADN
1. Introduction: molecular background of blood group antigens The molecular determination of 28 of the 29 different blood groups has occupied a significant level of activity for the
E-mail address: [email protected].
transfusion medicine community since the first determination of the molecular basis of blood group antigenicity in 1990 (ABO [1,2] and RH [3,4]) (for reviews see [5–8]). During the 1990s, intensive activity focussed on definition of the molecular basis of the major clinically significant blood group alleles, and also to define the molecular background of such phenomena as partial D [9–11] and weak D [12] phenotypes. This analysis revealed that there is a great deal of heterogeneity in blood
1246-7820/$ – see front matter # 2007 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.tracli.2007.04.011
N.D. Avent / Transfusion Clinique et Biologique 14 (2007) 10–15
group antigen expression, especially within the ABO and RH systems, and further demonstrated that a considerable number of alleles caused group O, weak D and partial D phenotypes [6]. (see dbRBC and Rhesus base). http://www.ncbi.nlm.nih.gov/ projects/mhc/xslcgi.fcgi?cmd=bgmut/home and http://www. uni-ulm.de/fwagner/RH/RB/. It is not the purpose of this review to reiterate the molecular bases underlying blood group antigens (see previous cited reviews for this information and above websites), but it is important here to realise that multiple genetic backgrounds can account for the same blood group phenotype. For example, with O alleles, over 60 are described by dbRBC (see above URL), many of which have a single nucleotide deletion (261delG), which results in a frameshift mutation and as a result encodes a truncated A glycosyltransferase. This is unable to add the relevant nucleoside sugars (N-acetylgalactosamine, A transferase; galactose, B-transferase) to the terminal oligosaccharide chain responsible for ABH antigenicity. Other O mutations that inactivate the AB glycosyltransferase include missense mutations. An analogous situation occurs within the RH system with RhD antigenicity in particular having a diverse genetic background. Multiple RHD alleles account for D-negative phenotypes including a complete RHD gene deletion [13,14], or mutated RHD gene (e.g. in Africans RHDC) [15], and other missense RHD mutations [16], all of which result in the complete absence of the RhD protein in erythrocyte membranes. Heterogeneity in RhD antigenicity is apparent in as much as partial D phenotypes occur where in most cases hybrid RHDRHCE genes have been demonstrated—for example, in the clinically significant DVI phenotype [10,11,17], which results in the expression of a chimeric RhCE-RhD protein lacking discrete D epitopes. Other partial phenotypes are caused by missense mutations that occur within regions of the RHD gene, encoding extracellular domains of the RhD protein. Weak D phenotypes show little qualitative difference in D antigen expression, but significant depression in D antigen site numbers. The RHD mutations responsible for these amino acid exchanges are almost completely missense mutations altering membrane or cytoplasmic domain-localised amino acids [18,19]. For ABO, blood group genotyping is possible, but complicated due to the multiple O alleles. RHD genotyping is significantly more powerful than reliance on a large panel of monoclonal anti-D that would be necessary to define partial and weak D phenotypes serologically. Blood group determination using molecular techniques is currently used predominantly as a supporting role for blood group serology in immunohaematology [20]. In certain circumstances however, the use of serology is technically difficult and molecular-based blood grouping is the only realistic option. In prenatal blood group definition in cases of feto-maternal alloimmunisation to paternally inherited blood group antigens, blood grouping, using the PCR, has been used since 1993 [21,22] and is used for RhD [15,23–33], RhCcEe [15,34], Kell [35–37], Duffy [38] and other blood groups known to cause haemolytic disease of the newborn and foetus (HDNF) [23,35]. Multi-transfused patients also present a technical challenge for serology as the peripheral blood from these individuals will contain red cells from previous
11
transfusions, which may also be coated by antibodies. Thus, blood grouping using DNA-based techniques in such cases alleviates the use of serological techniques [39,40]. However, for mainstream applications, serological definition of blood groups remains the dominant force, but with the dramatic technological developments in the utilisation of high throughput genotyping, this situation may not be stable in the long term. As supportive molecular genotyping becomes the norm within clinical diagnostic laboratories, and more technical staff becoming familiar with the discipline of molecular genotyping the transition from serology to genotyping maybe seamless. Recently (in September 2006) the US food and drugs administration held a workshop to explore the potential impact of high-throughput molecular genotyping technology on blood group determination (http://www.fda.gov/ cber/summaries.htm). It is clear from this meeting and debate during its proceedings, that several mass scale genotyping platforms are being developed in North America and Europe to apply the rich knowledge surrounding the molecular basis of blood groups into clinical application. At present, the clear message is that molecular blood grouping has a fundamental supporting role in immunohaematology, but some workers (including the author) feel that mass scale genotyping may indeed be a possibility within the not too distant future. I shall provide a brief description of the technical platforms described at this workshop, which incidentally were mostly described in the May 2005 edition of Transfusion (volume 45 (5)). 2. Bioarray solutions, human erythrocyte antigen (HEA) array [41] (www.bioarray.com) Bioarray solutions of Warren, NJ, USA have developed a BeadarrayTM platform for genotyping of blood group polymorphisms, marketed as human erythrocyte antigen (HEA) beadchipTM. The system involves the multiplex (MPX) PCR amplification of various blood group active genes (RH, KEL, FY, DO, MNS, JK, LU, DI, LW, CO, SC and HbS). The PCR products are then made single stranded, cleaned up and hybridised to pre-fabricated and color-coded beads, each of 50 different dyes attached to an oligonucleotide probe, which is complementary to a MPX PCR product. Four thousands of these beads were added to an array and comprise sufficient of both alleles and positive and negative controls. Following hybridisation in a specific reaction chamber, extension is performed in order to facilitate the labelling of the product-extension, which will only proceed when there is a perfect match to an allelic SNP. The entire array is then visualised using an automated microscope and image analysis system, and the array information is decodified and the genotypes scored. The beadchip platform is now commercially available by Bioarray solutions inc. 3. GenomeLab SNPstream (Beckmann Coulter–Denomme & Van Oene [42]) Utilisation of a commercially available SNP genotyping platform, GenomeLab SNPstream, has been performed by Denomme and Van Oene [42] to investigate SNPs responsible
12
N.D. Avent / Transfusion Clinique et Biologique 14 (2007) 10–15
for RHD (exons 4 and 9) RHCE (C/c, exon 2 and E, exon 5), K/ k, Kpa/Kpb Jka/Jkb, S/s, Dia/Dib, HPA-1a/HPA-1b and Fya/Fyb/ Fy*0. The system utilises the amplification of the different regions of genomic DNA carrying these SNPs by a MPX PCR that includes tagged primers (12 different tags), a post-PCR cleanup and then extension labelling with fluorescent nucleotide terminators. The labelled reactions are then added to a 384well plate that is arrayed with complementary probes corresponding to each of the tags. Each SNP is then identified by virtue of its position within the well. The preliminary study, despite mixed performance for each SNP, clearly showed that the platform has the potential to be utilized in a high throughput setting. 4. Blood group antigen and platelet genotyping on arrays
fabrication of DNA arrays; fluoro single sequence primer technology (later abandoned during the course of the project to focus on the DNA array work); PCR and genomic DNA extraction standardization; small scale clinical trial using a biobank of genomic DNA samples assembled by the consortium; large scale clinical trial leading to the CE-marking of the outcome of the Bloodgen project – Bloodchip. Bloodchip will eventually be commercially available, to be manufactured by Progenika Biopharma and will be distributed by them and Sanquin reagents. Bloodchip development has been driven by a comprehensive approach to blood group genotyping, and has been designed so that new alleles when discovered can be relatively easily be added to the platform (Figs. 1 and 2). Bloodchip will detect
Two similar approaches were adopted to define blood group antigens on arrays. Bugert et al. [43], printed arrays corresponding to probes designed to bind to Cy3 labelled PCR products encoding HPA-1; HPA-2, HPA-3, HPA-5 and HPA-15 polymorphic SNPs. Differential binding to these allele-specific probes permitted accurate genotyping of these polymorphisms using a blinded cohort of 73 samples. The array methodology was directly compared to a 50 -nuclease Taqman assay. Beiboer et al. [44] developed a spotted array with probes corresponding to HPA-1, HPA-2, HPA-3, HPA-4, HPA-5, HPA15, JK, FY, KEL, MNS, RHCE, RHD and DO alleles. In a blinded study, 94 donors were accurately genotyped for all HPA antigens tested for serologically. Both of these studies, published within the same issue of Transfusion heralded the arrival of DNA array technology into molecular grouping, which has been further expanded in the large-scale Bloodgen project (Section 5). 5. The Bloodgen project (www.bloodgen.com) The Bloodgen project was conceived in 2000 as a collaborative effort to unify the rapid advances in both the knowledge surrounding the molecular basis of blood group antigens and the developments in high throughput genotyping methodology particularly using DNA arrays. A consortium was assembled including laboratories with previous extensive experience of molecular blood grouping from Europe (Bristol, UK; Sanquin (Amsterdam and Rotterdam), The Netherlands; Ulm, Germany; Lund, Sweden; Barcelona, Spain; Prague, Czech Republic). The consortium was joined by a SME with specific experience of DNA microarrays (Progenika Biopharma SA, Derio, Spain) and Biotest, Dreieich, Germany who have extensive knowledge of the transfusion medicine diagnostics market. The project secured EC framework V funding as a demonstration project – indicating the use of tried technology to a specific clinical application – molecular blood grouping on a novel technical platform. The work programme undertaken by Bloodgen was divided into different sub-projects or workpackages, led by different academic groups. These included workpackage:
Fig. 1. Multiplex PCR amplification of ABO and RHD genes prior to array hybridisation to Bloodchip. Figure shows an Agarose gel image of a MPX PCR reaction that amplifies regions of the ABO and RHD genes prior to fragmentation, labelling, denaturation and hybridisation to Bloodchip. The MPX PCR comprises the gene specific primers, and in approximately 20-fold excess hybrid primers comprising the same gene specific primers (30 -end) and MAPH (multiplex amplifiable probe hybridisation) tag sequences at the 50 end of each primer. The MPX also contains MAPH forward and reverse primers, which permit consistent amplification of all PCR products in a single tube format. An additional MPX, which amplifies RHCE, KEL, FY, DI, DO, MNS, CO, JK is performed in addition, and both MPXs are labelled with Cy3 and Cy5 dyes before hybridisation to Bloodchip.
N.D. Avent / Transfusion Clinique et Biologique 14 (2007) 10–15
13
was then expended in the development of multiplex PCRs, which were designed to amplify the various genomic regions of genes encoding blood group polymorphisms, and Bloodchip has had considerable focus on the RH system and 87 RH genotypes can be assigned by the chip and associated proprietary software. In order to do this all 10 RHD exons and selected surrounding introns are amplified by the RHD MPX reaction. The MPX reactions that amplify the genomic regions carrying 116 blood group defining SNPs are amplified in a very uniform fashion by the inclusion of MAPH tag binding sites on the 50 ends of all primers as originally adopted by Beiboer et al. [44]. The MPX products are fragmented, labelled and hybridised to an array composed of probes (40 of each) corresponding to sequences within the amplified SNP. By analysis of the ratios of fluorescence intensity of binding of pairs of probes complementary to each SNP, homo- and heterozygosity of each blood group allele is defined. Bloodchip is currently undergoing CEmarking for RHCE, KEL blood group diagnostics, shortly to be followed by CE-marking for RHD diagnosis. The product will be commercially available in late 2007. 7. Discussion
Fig. 2. Scanned image of Bloodchip. The slide comprises an array of 6272 spots of oligonucleotides comprising positive and negative controls and 40 replicates of each of 116 probes, which bind the MPX products as amplified in Fig. 1. The different fluorescent intensities visible in the figure is a reflection of whether the sample is positive or negative for a particular allelic blood group defining SNP.
multiple alleles (116 in total during the project with more being added) from ABO, RHD, RHCE, KEL, FY, JK, DI, CO, MNS, DO and selected HPA alleles. A key area of focus has been the different RHD alleles that cause partial, weak and D elute phenotypes. 6. Technical approach The various technical workpackages described above developed a standardised approach to DNA extraction as there were various procedures adopted by the consortium. In essence, the agreed protocol was based on the Qiagen Blood mini kit, which was adopted as standard. Considerable technical effort
There is considerable debate amongst the transfusion medicine community that molecular genotyping will eventually replace serology [45]. Whilst this seems inevitable for blood groups where serological reagents are not available – in short supply and are technically demanding – for other situations, serology will undoubtedly still prevail. With ABO typing for example, genotyping is complex, but serology is relatively straightforward and cheap. Indeed, in the Bloodgen project, several new ABO alleles were identified revealing that work is still required to comprehensively catalogue all ABO alleles. Perhaps advances in deglycosylation of red cells using novel specificity glycosidases may be sufficiently advanced in the future to render ABO serology obsolete [46]. In the meantime, serology will predominate as the primary method of phenotyping donors and patients for ABO. This is not necessarily the case for other blood group antigens. All technologies described here have demonstrated a high degree of accuracy and concordance with serological results, in especially when focussing on the Rh system, errors by serology have been found. In the Bloodgen project, some samples defined serologically as D-positive were identified by Bloodchip as partial D (DVII), and a sample serotyped as D-C-c+ was shown to possess a ceAR allele, with associated weak D and C antigen expression (Bloodgen consortium, unpublished results). It is clear therefore that in only a limited cohort of 1000 samples, Rh variants will be found on a large scale if comprehensive Rh genotyping technologies are adopted. These individuals, if patients or pregnant women are at risk of alloimmunisation if transfused with Rh-incompatible blood. Furthermore, units of Rh-incompatible blood may be found within the donor population and eliminated for the purpose of transfusion. All of the platform technologies described have the potential for addition of new alleles, but in particular high capacity arrays
14
N.D. Avent / Transfusion Clinique et Biologique 14 (2007) 10–15
are the only feasible option for inclusion of the large numbers of ABO and RH alleles that are undoubtedly still yet to be discovered in the world’s populations. This is especially relevant as it must be stressed that still only a limited number of predominantly Western population groups have been studied to any extent. 8. Economics: genotyping versus serology Currently, blood group serology is inherently cheaper than genotyping, but this is only because a limited number of blood group alleles are tested for on a routine basis. In theory, for donation testing a single genotyping test (arrays in current form can only be processed once) could be performed and entered onto a centralised database for electronic cross-matching with patients. Whilst it is appreciated that in rare circumstances (for example, haematological malignancy) blood group phenotype may alter, blood groups should not change throughout that individuals blood donation career. With 100+ SNPs tested in one reaction series, the price per test is likely to be about one euro per SNP, but with mass scale use this price may reduce considerably. As multiple blood group genotypes can account for similar blood group phenotypes (e.g. O, D-negative, weak D, D-elute) it is imperative that large-scale analyses of blood group genotypes are performed throughout population groups worldwide. It has been known for a considerable time that certain blood groups predominate amongst various populations – for example, Fy(a-b-) in West Africans, VS and V antigens in Africans, D-elute in Asians [47]. Nevertheless, recent largescale analysis have revealed the presence of clusters of certain blood group alleles in discrete population groups. This would indicate that transfusion policy needs to take into account the genetic variation within the immediate vicinity. This could be benefit in better quality control of transfusion policy of recipients of both D-positive and D-negative units within the blood bank. A recent investigation of 33,864 donors revealed the presence of 54 partial RHD alleles amongst the serologically defined D+ samples [48]. These partial Ds if transfused with D-positive blood could become alloimmunised to the D antigen, a potentially serious issue if the individual is of child bearing age or requiring a transfusion. Using a similar approach, D-negative donations can be quality assured by identification of RHD alleles that have been known to cause immunisation in D-negative recipients (for example DEL and weak D type 4.2 (DAR)) [49]. Clearly the frequency of alloimmunisation due to D will be significantly reduced if such potentially problematic donated units are eliminated from the blood bank. 9. Mass scale genotyping? Mass scale genotyping approaches have demonstrated that a comprehensive approach to RH genotyping reaps dividends in the identification of weakened D and C antigen expression and quality assurance of D-negative donor units. The identification of partial D individuals as patients (both transfusion recipients
or patients) has clear clinical value. Whilst molecular diagnostics for genetic disease gathers pace, it makes good economic sense to include blood groups and HLA type in population screens for such diseases as cystic fibrosis, which is a target for genetic diagnosis following a positive screening test. Furthermore, it may well be worthwhile genotyping all individuals at birth for their blood group status and place this information on databases. This information will be highly useful if these individuals become patients, or eventually decide to become blood donors. Naturally, such information needs to be confidential and may raise ethical debate. However, as there is only limited information concerning the risk of disease associated with inheritance of a particular blood group phenotype, this information is not particularly controversial, but potentially lifesaving to an individual if they become transfusion dependent. It is likely that further projects on the mass scale application could be considered in the very near future in order to genotype a large number of donors within EU-based blood banks. This may give the best indication yet as to the potential success of mass scale genotyping and the identification of clinically significant variant blood group genotypes within that local donor cohort. Such an approach is essential as it will enable the streamlining of provision of compatible blood, and reduce the level of initial alloimmunisation made by vulnerable patient groups (e.g. multi-transfused individuals). Acknowledgements The author wishes to thank the Bloodgen consortium and Dr Tracey Madgett for providing the images contained within Figs. 1 and 2. References [1] Yamamoto F, Clausen H, White T, et al. Molecular genetic basis of the histo-blood group ABO system. Nature 1990;345(6272):229–33. [2] Yamamoto F, Marken J, Tsuji T, et al. Cloning and characterization of DNA complementary to human UDP-GalNAc: Fuc alpha 1——2Gal alpha 1——3GalNAc transferase (histo-blood group A transferase) mRNA. J Biol Chem 1990;265(2):1146–51. [3] Cherif-Zahar B, Bloy C, Le Van Kim C, et al. Molecular cloning and protein structure of a human blood group Rh polypeptide. Proc Natl Acad Sci U S A 1990;87(16):6243–7. [4] Avent ND, Ridgwell K, Tanner MJ, Anstee DJ. cDNA cloning of a 30 kDa erythrocyte membrane protein associated with Rh (Rhesus)-blood-groupantigen expression. Biochem J 1990;271(3):821–5. [5] Cartron JP, Colin Y. Structural and functional diversity of blood group antigens. Transfus Clin Biol 2001;8(3):163–99. [6] Reid ME, Lomas-Francis C. The Blood Group Antigens factsbook, 2nd ed, New York: Academic Press; 2004. [7] Daniels G. The molecular genetics of blood group polymorphism. Transpl Immunol 2005;14(3–4):143–53. [8] Avent ND. Blood groups: Molecular genetic basis Encyclopaedia of the Human Genome: Nature publishing group, 2003: 333–343. [9] Rouillac C, Colin Y, Hughes-Jones NC, et al. Transcript analysis of D category phenotypes predicts hybrid Rh D-CE-D proteins associated with alteration of D epitopes. Blood 1995;85(10):2937–44. [10] Mouro I, Le Van Kim C, Rouillac C, et al. Rearrangements of the blood group RhD gene associated with the DVI category phenotype. Blood 1994;83(4):1129–35.
N.D. Avent / Transfusion Clinique et Biologique 14 (2007) 10–15 [11] Wagner FF, Gassner C, Muller TH, et al. Three molecular structures cause rhesus D category VI phenotypes with distinct immunohematologic features. Blood 1998;91(6):2157–68. [12] Wagner FF, Gassner C, Muller TH, et al. Molecular basis of weak D phenotypes. Blood 1999;93(1):385–93. [13] Colin Y, Cherif-Zahar B, Le Van Kim C, et al. Genetic basis of the RhDpositive and RhD-negative blood group polymorphism as determined by Southern analysis. Blood 1991;78(10):2747–52. [14] Wagner FF, Flegel WA. RHD gene deletion occurred in the Rhesus box. Blood 2000;95(12):3662–8. [15] Singleton BK, Green CA, Avent ND, et al. The presence of an RHD pseudogene containing a 37 base pair duplication and a nonsense mutation in africans with the Rh D-negative blood group phenotype. Blood 2000;95(1):12–8. [16] Wagner FF, Frohmajer A, Flegel WA. RHD positive haplotypes in D negative Europeans. BMC Genet 2001;2(1):10. [17] Avent ND, Liu W, Jones JW, et al. Molecular analysis of Rh transcripts and polypeptides from individuals expressing the DVI variant phenotype: an RHD gene deletion event does not generate All DVIccEe phenotypes. Blood 1997;89(5):1779–86. [18] Avent ND. The rhesus blood group system: insights from recent advances in molecular biology. Transfus Med Rev 1999;13(4):245–66. [19] Avent N.D. New insight in the Rh system: stucture and function. Vox Sang 2007; in press. [20] Westhoff CM. Molecular testing for transfusion medicine. Curr Opin Hematol 2006;13(6):471–5. [21] Bennett PR, Le Van Kim C, Colin Y, et al. Prenatal determination of fetal RhD type by DNA amplification. N Engl J Med 1993;329(9):607–10. [22] Wolter LC, Hyland CA, Saul A, Rhesus. D genotyping using polymerase chain reaction. Blood 1993;82(5):1682–3. [23] Avent ND, Finning KM, Martin PG, Soothill PW. Prenatal determination of fetal blood group status. Vox Sang 2000;78(Suppl 2):155–62. [24] Aubin JT, Le Van Kim C, Mouro I, et al. Specificity and sensitivity of RHD genotyping methods by PCR-based DNA amplification. Br J Haematol 1997;98(2):356–64. [25] Ait Soussan ARD, Bonsel GJ, et al. Prenatal RHD testing of fetus and mother: Decision to administer anti-D prophylaxis. Transfus Med Hemother 2004;31(Suppl 3):50. [26] Finning KM, Martin PG, Soothill PW, Avent ND. Prediction of fetal D status from maternal plasma: introduction of a new noninvasive fetal RHD genotyping service. Transfusion 2002;42(8):1079–85. [27] Finning K, Martin P, Daniels G. A clinical service in the UK to predict fetal Rh (Rhesus) D blood group using free fetal DNA in maternal plasma. Ann N Y Acad Sci 2004;1022:119–23. [28] Legler TJ, Lynen R, Maas JH, et al. Prediction of fetal Rh D and Rh CcEe phenotype from maternal plasma with real-time polymerase chain reaction. Transfus Apher Sci 2002;27(3):217–23. [29] Lo YM, Hjelm NM, Fidler C, et al. Prenatal diagnosis of fetal RhD status by molecular analysis of maternal plasma. N Engl J Med 1998;339(24): 1734–8. [30] Van den Veyver IB, Chong SS, Cota J, et al. Single-cell analysis of the RhD blood type for use in preimplantation diagnosis in the prevention of severe hemolytic disease of the newborn. Am J Obstet Gynecol 1995;172(2 Pt 1):533–40.
15
[31] Van der Schoot CE, Soussan AA, Koelewijn J, et al. Non-invasive antenatal RHD typing. Transfus Clin Biol 2006;13(1–2):53–7. [32] Avent ND, Martin PG, Armstrong-Fisher SS, et al. Evidence of genetic diversity underlying Rh D-, weak D (Du), and partial D phenotypes as determined by multiplex polymerase chain reaction analysis of the RHD gene. Blood 1997;89(7):2568–77. [33] Rouillac-Le Sciellour C, Puillandre P, Gillot R, et al. Large-scale prediagnosis study of fetal RHD genotyping by PCR on plasma DNA from RhD-negative pregnant women. Mol Diagn 2004;8(1):23–31. [34] Le Van Kim C, Mouro I, Brossard Y, et al. PCR-based determination of Rhc and RhE status of fetuses at risk of Rhc and RhE haemolytic disease. Br J Haematol 1994;88(1):193–5. [35] van der Schoot CE, Tax GH, Rijnders RJ, et al. Prenatal typing of Rh and Kell blood group system antigens: the edge of a watershed. Transfus Med Rev 2003;17(1):31–44. [36] Lee S, Bennett PR, Overton T, et al. Prenatal diagnosis of Kell blood group genotypes: KEL1 and KEL2. Am J Obstet Gynecol 1996;175(2): 455–9. [37] Avent ND, Martin PG. Kell typing by allele-specific PCR (ASP). Br J Haematol 1996;93(3):728–30. [38] Olsson ML, Hansson C, Avent ND, et al. A clinically applicable method for determining the three major alleles at the Duffy (FY) blood group locus using polymerase chain reaction with allele-specific primers. Transfusion 1998;38(2):168–73. [39] Castilho L, Rios M, Pellegrino Jr J, et al. Blood group genotyping facilitates transfusion of beta-thalassemia patients. J Clin Lab Anal 2002;16(5):216–20. [40] Legler TJ, Eber SW, Lakomek M, et al. Application of RHD and RHCE genotyping for correct blood group determination in chronically transfused patients. Transfusion 1999;39(8):852–5. [41] Hashmi G, Shariff T, Seul M, et al. A flexible array format for large-scale, rapid blood group DNA typing. Transfusion 2005;45(5):680–8. [42] Denomme GA, Van Oene M. High-throughput multiplex single-nucleotide polymorphism analysis for red cell and platelet antigen genotypes. Transfusion 2005;45(5):660–6. [43] Bugert P, McBride S, Smith G, et al. Microarray-based genotyping for blood groups: comparison of gene array and 50 -nuclease assay techniques with human platelet antigen as a model. Transfusion 2005;45(5):654–9. [44] Beiboer SH, Wieringa-Jelsma T, Maaskant-Van Wijk PA, et al. Rapid genotyping of blood group antigens by multiplex polymerase chain reaction and DNA microarray hybridization. Transfusion 2005;45(5): 667–79. [45] Anstee DJ. Goodbye to agglutination and all that? Transfusion 2005;45(5):652–3. [46] Liu QP, Sulzenbacher G, Yuan H, et al. Bacterial glycosidases for the production of universal red blood cells. Nat Biotechnol 2007. [47] Avent ND. High variability of the RH locus in different ethnic backgrounds. Transfusion 2005;45(3):293–4. [48] Denomme GA, Wagner FF, Fernandes BJ, et al. Partial D, weak D types, and novel RHD alleles among 33,864 multiethnic patients: implications for anti-D alloimmunization and prevention. Transfusion 2005;45(10): 1554–60. [49] Flegel WA. How I manage donors and patients with a weak D phenotype. Curr Opin Hematol 2006;13(6):476–83.
Plenary session
Large scale blood group genotyping Ge´notypage des groupes sanguins a` grande e´chelle Neil D. Avent Centre for Research in Biomedicine, Bristol Genomics Research Institute, Faculty of Applied Sciences, University of the West of England, Bristol, United Kingdom Available online 23 May 2007
Abstract Determination of predicted blood group phenotype by determination of genotype has been performed since the 1990s. This evolved due to the rapid accrual of information surrounding the molecular basis of blood group antigen expression, which started in 1990 with ABO and RH systems and has now resulted in the molecular description of 28 of the 29 blood groups. Blood group genotyping is currently performed mostly for fetal blood group incompatibility and for assessment of multi-transfused patients. Both of these clinical scenarios are either dangerous or technically difficult, respectively to define serologically. With the simultaneous development of mass scale genotyping platforms it has now permitted the application of them to blood group genotype determination. In this paper, I describe some recently published work that has demonstrated that mass scale genotyping approaches are feasible. These approaches may lead to more effective management of blood stocks and patient cross-matching by reducing the dependence on serology during the time critical pre-transfusion phase. It is most probable that large scale studies, perhaps involving many European Union and North American based blood suppliers, may drive the introduction of this technology and convince red cell serologists that this approach may allow their work to be more focussed. # 2007 Elsevier Masson SAS. All rights reserved. Re´sume´ La pre´diction du phe´notype a` partir de la de´termination du ge´notype a e´te´ mise en oeuvre depuis les anne´es 1990. Cette e´volution est due a` l’accumulation rapide d’informations entourant la base mole´culaire de l’expression des groupes sanguins qui a commence´ en 1990 avec les syste`mes ABO et RH, et qui concerne maintenant la description mole´culaire de 28 parmi les 29 syste`mes de groupes. Le ge´notypage des groupes sanguins est couramment pratique´ dans les cas d’incompatibilite´ foetomaternelle et pour le suivi des malades polytransfuse´s. Ces deux situations cliniques sont respectivement, soit dangereuse, soit techniquement difficile a` de´finir se´rologiquement. Le de´veloppement simultane´ de diverses plates-formes de ge´notypage de masse a permis d’envisager des applications a` la de´termination des ge´notypes de groupes sanguins. Dans cette revue, je de´crirai quelques travaux publie´s re´cemment qui de´montrent que le ge´notypage de masse est faisable. Ces approches pourraient conduire a` une meilleure gestion des stocks de sangs et du cross-match des malades en re´duisant la part des techniques se´rologiques pendant la phase critique pre´transfusionnelle. Il est tre`s probable que des e´tudes a` grande e´chelle impliquant de nombreuses structures responsables de la distribution du sang, en Europe et aux E´tats-Unis, pourraient conduire a` l’introduction de cette technologie et a` convaincre les se´rologistes que cette approche leur permettra de mieux focaliser leurs travaux. # 2007 Elsevier Masson SAS. All rights reserved. Keywords: Blood group; Polymorphism; Genotyping; SNP; DNA chips Mots cle´s : Groupes sanguins ; Polymorphisme ; Ge´notype ; SNP ; Puces a` ADN
1. Introduction: molecular background of blood group antigens The molecular determination of 28 of the 29 different blood groups has occupied a significant level of activity for the
E-mail address: [email protected].
transfusion medicine community since the first determination of the molecular basis of blood group antigenicity in 1990 (ABO [1,2] and RH [3,4]) (for reviews see [5–8]). During the 1990s, intensive activity focussed on definition of the molecular basis of the major clinically significant blood group alleles, and also to define the molecular background of such phenomena as partial D [9–11] and weak D [12] phenotypes. This analysis revealed that there is a great deal of heterogeneity in blood
1246-7820/$ – see front matter # 2007 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.tracli.2007.04.011
N.D. Avent / Transfusion Clinique et Biologique 14 (2007) 10–15
group antigen expression, especially within the ABO and RH systems, and further demonstrated that a considerable number of alleles caused group O, weak D and partial D phenotypes [6]. (see dbRBC and Rhesus base). http://www.ncbi.nlm.nih.gov/ projects/mhc/xslcgi.fcgi?cmd=bgmut/home and http://www. uni-ulm.de/fwagner/RH/RB/. It is not the purpose of this review to reiterate the molecular bases underlying blood group antigens (see previous cited reviews for this information and above websites), but it is important here to realise that multiple genetic backgrounds can account for the same blood group phenotype. For example, with O alleles, over 60 are described by dbRBC (see above URL), many of which have a single nucleotide deletion (261delG), which results in a frameshift mutation and as a result encodes a truncated A glycosyltransferase. This is unable to add the relevant nucleoside sugars (N-acetylgalactosamine, A transferase; galactose, B-transferase) to the terminal oligosaccharide chain responsible for ABH antigenicity. Other O mutations that inactivate the AB glycosyltransferase include missense mutations. An analogous situation occurs within the RH system with RhD antigenicity in particular having a diverse genetic background. Multiple RHD alleles account for D-negative phenotypes including a complete RHD gene deletion [13,14], or mutated RHD gene (e.g. in Africans RHDC) [15], and other missense RHD mutations [16], all of which result in the complete absence of the RhD protein in erythrocyte membranes. Heterogeneity in RhD antigenicity is apparent in as much as partial D phenotypes occur where in most cases hybrid RHDRHCE genes have been demonstrated—for example, in the clinically significant DVI phenotype [10,11,17], which results in the expression of a chimeric RhCE-RhD protein lacking discrete D epitopes. Other partial phenotypes are caused by missense mutations that occur within regions of the RHD gene, encoding extracellular domains of the RhD protein. Weak D phenotypes show little qualitative difference in D antigen expression, but significant depression in D antigen site numbers. The RHD mutations responsible for these amino acid exchanges are almost completely missense mutations altering membrane or cytoplasmic domain-localised amino acids [18,19]. For ABO, blood group genotyping is possible, but complicated due to the multiple O alleles. RHD genotyping is significantly more powerful than reliance on a large panel of monoclonal anti-D that would be necessary to define partial and weak D phenotypes serologically. Blood group determination using molecular techniques is currently used predominantly as a supporting role for blood group serology in immunohaematology [20]. In certain circumstances however, the use of serology is technically difficult and molecular-based blood grouping is the only realistic option. In prenatal blood group definition in cases of feto-maternal alloimmunisation to paternally inherited blood group antigens, blood grouping, using the PCR, has been used since 1993 [21,22] and is used for RhD [15,23–33], RhCcEe [15,34], Kell [35–37], Duffy [38] and other blood groups known to cause haemolytic disease of the newborn and foetus (HDNF) [23,35]. Multi-transfused patients also present a technical challenge for serology as the peripheral blood from these individuals will contain red cells from previous
11
transfusions, which may also be coated by antibodies. Thus, blood grouping using DNA-based techniques in such cases alleviates the use of serological techniques [39,40]. However, for mainstream applications, serological definition of blood groups remains the dominant force, but with the dramatic technological developments in the utilisation of high throughput genotyping, this situation may not be stable in the long term. As supportive molecular genotyping becomes the norm within clinical diagnostic laboratories, and more technical staff becoming familiar with the discipline of molecular genotyping the transition from serology to genotyping maybe seamless. Recently (in September 2006) the US food and drugs administration held a workshop to explore the potential impact of high-throughput molecular genotyping technology on blood group determination (http://www.fda.gov/ cber/summaries.htm). It is clear from this meeting and debate during its proceedings, that several mass scale genotyping platforms are being developed in North America and Europe to apply the rich knowledge surrounding the molecular basis of blood groups into clinical application. At present, the clear message is that molecular blood grouping has a fundamental supporting role in immunohaematology, but some workers (including the author) feel that mass scale genotyping may indeed be a possibility within the not too distant future. I shall provide a brief description of the technical platforms described at this workshop, which incidentally were mostly described in the May 2005 edition of Transfusion (volume 45 (5)). 2. Bioarray solutions, human erythrocyte antigen (HEA) array [41] (www.bioarray.com) Bioarray solutions of Warren, NJ, USA have developed a BeadarrayTM platform for genotyping of blood group polymorphisms, marketed as human erythrocyte antigen (HEA) beadchipTM. The system involves the multiplex (MPX) PCR amplification of various blood group active genes (RH, KEL, FY, DO, MNS, JK, LU, DI, LW, CO, SC and HbS). The PCR products are then made single stranded, cleaned up and hybridised to pre-fabricated and color-coded beads, each of 50 different dyes attached to an oligonucleotide probe, which is complementary to a MPX PCR product. Four thousands of these beads were added to an array and comprise sufficient of both alleles and positive and negative controls. Following hybridisation in a specific reaction chamber, extension is performed in order to facilitate the labelling of the product-extension, which will only proceed when there is a perfect match to an allelic SNP. The entire array is then visualised using an automated microscope and image analysis system, and the array information is decodified and the genotypes scored. The beadchip platform is now commercially available by Bioarray solutions inc. 3. GenomeLab SNPstream (Beckmann Coulter–Denomme & Van Oene [42]) Utilisation of a commercially available SNP genotyping platform, GenomeLab SNPstream, has been performed by Denomme and Van Oene [42] to investigate SNPs responsible
12
N.D. Avent / Transfusion Clinique et Biologique 14 (2007) 10–15
for RHD (exons 4 and 9) RHCE (C/c, exon 2 and E, exon 5), K/ k, Kpa/Kpb Jka/Jkb, S/s, Dia/Dib, HPA-1a/HPA-1b and Fya/Fyb/ Fy*0. The system utilises the amplification of the different regions of genomic DNA carrying these SNPs by a MPX PCR that includes tagged primers (12 different tags), a post-PCR cleanup and then extension labelling with fluorescent nucleotide terminators. The labelled reactions are then added to a 384well plate that is arrayed with complementary probes corresponding to each of the tags. Each SNP is then identified by virtue of its position within the well. The preliminary study, despite mixed performance for each SNP, clearly showed that the platform has the potential to be utilized in a high throughput setting. 4. Blood group antigen and platelet genotyping on arrays
fabrication of DNA arrays; fluoro single sequence primer technology (later abandoned during the course of the project to focus on the DNA array work); PCR and genomic DNA extraction standardization; small scale clinical trial using a biobank of genomic DNA samples assembled by the consortium; large scale clinical trial leading to the CE-marking of the outcome of the Bloodgen project – Bloodchip. Bloodchip will eventually be commercially available, to be manufactured by Progenika Biopharma and will be distributed by them and Sanquin reagents. Bloodchip development has been driven by a comprehensive approach to blood group genotyping, and has been designed so that new alleles when discovered can be relatively easily be added to the platform (Figs. 1 and 2). Bloodchip will detect
Two similar approaches were adopted to define blood group antigens on arrays. Bugert et al. [43], printed arrays corresponding to probes designed to bind to Cy3 labelled PCR products encoding HPA-1; HPA-2, HPA-3, HPA-5 and HPA-15 polymorphic SNPs. Differential binding to these allele-specific probes permitted accurate genotyping of these polymorphisms using a blinded cohort of 73 samples. The array methodology was directly compared to a 50 -nuclease Taqman assay. Beiboer et al. [44] developed a spotted array with probes corresponding to HPA-1, HPA-2, HPA-3, HPA-4, HPA-5, HPA15, JK, FY, KEL, MNS, RHCE, RHD and DO alleles. In a blinded study, 94 donors were accurately genotyped for all HPA antigens tested for serologically. Both of these studies, published within the same issue of Transfusion heralded the arrival of DNA array technology into molecular grouping, which has been further expanded in the large-scale Bloodgen project (Section 5). 5. The Bloodgen project (www.bloodgen.com) The Bloodgen project was conceived in 2000 as a collaborative effort to unify the rapid advances in both the knowledge surrounding the molecular basis of blood group antigens and the developments in high throughput genotyping methodology particularly using DNA arrays. A consortium was assembled including laboratories with previous extensive experience of molecular blood grouping from Europe (Bristol, UK; Sanquin (Amsterdam and Rotterdam), The Netherlands; Ulm, Germany; Lund, Sweden; Barcelona, Spain; Prague, Czech Republic). The consortium was joined by a SME with specific experience of DNA microarrays (Progenika Biopharma SA, Derio, Spain) and Biotest, Dreieich, Germany who have extensive knowledge of the transfusion medicine diagnostics market. The project secured EC framework V funding as a demonstration project – indicating the use of tried technology to a specific clinical application – molecular blood grouping on a novel technical platform. The work programme undertaken by Bloodgen was divided into different sub-projects or workpackages, led by different academic groups. These included workpackage:
Fig. 1. Multiplex PCR amplification of ABO and RHD genes prior to array hybridisation to Bloodchip. Figure shows an Agarose gel image of a MPX PCR reaction that amplifies regions of the ABO and RHD genes prior to fragmentation, labelling, denaturation and hybridisation to Bloodchip. The MPX PCR comprises the gene specific primers, and in approximately 20-fold excess hybrid primers comprising the same gene specific primers (30 -end) and MAPH (multiplex amplifiable probe hybridisation) tag sequences at the 50 end of each primer. The MPX also contains MAPH forward and reverse primers, which permit consistent amplification of all PCR products in a single tube format. An additional MPX, which amplifies RHCE, KEL, FY, DI, DO, MNS, CO, JK is performed in addition, and both MPXs are labelled with Cy3 and Cy5 dyes before hybridisation to Bloodchip.
N.D. Avent / Transfusion Clinique et Biologique 14 (2007) 10–15
13
was then expended in the development of multiplex PCRs, which were designed to amplify the various genomic regions of genes encoding blood group polymorphisms, and Bloodchip has had considerable focus on the RH system and 87 RH genotypes can be assigned by the chip and associated proprietary software. In order to do this all 10 RHD exons and selected surrounding introns are amplified by the RHD MPX reaction. The MPX reactions that amplify the genomic regions carrying 116 blood group defining SNPs are amplified in a very uniform fashion by the inclusion of MAPH tag binding sites on the 50 ends of all primers as originally adopted by Beiboer et al. [44]. The MPX products are fragmented, labelled and hybridised to an array composed of probes (40 of each) corresponding to sequences within the amplified SNP. By analysis of the ratios of fluorescence intensity of binding of pairs of probes complementary to each SNP, homo- and heterozygosity of each blood group allele is defined. Bloodchip is currently undergoing CEmarking for RHCE, KEL blood group diagnostics, shortly to be followed by CE-marking for RHD diagnosis. The product will be commercially available in late 2007. 7. Discussion
Fig. 2. Scanned image of Bloodchip. The slide comprises an array of 6272 spots of oligonucleotides comprising positive and negative controls and 40 replicates of each of 116 probes, which bind the MPX products as amplified in Fig. 1. The different fluorescent intensities visible in the figure is a reflection of whether the sample is positive or negative for a particular allelic blood group defining SNP.
multiple alleles (116 in total during the project with more being added) from ABO, RHD, RHCE, KEL, FY, JK, DI, CO, MNS, DO and selected HPA alleles. A key area of focus has been the different RHD alleles that cause partial, weak and D elute phenotypes. 6. Technical approach The various technical workpackages described above developed a standardised approach to DNA extraction as there were various procedures adopted by the consortium. In essence, the agreed protocol was based on the Qiagen Blood mini kit, which was adopted as standard. Considerable technical effort
There is considerable debate amongst the transfusion medicine community that molecular genotyping will eventually replace serology [45]. Whilst this seems inevitable for blood groups where serological reagents are not available – in short supply and are technically demanding – for other situations, serology will undoubtedly still prevail. With ABO typing for example, genotyping is complex, but serology is relatively straightforward and cheap. Indeed, in the Bloodgen project, several new ABO alleles were identified revealing that work is still required to comprehensively catalogue all ABO alleles. Perhaps advances in deglycosylation of red cells using novel specificity glycosidases may be sufficiently advanced in the future to render ABO serology obsolete [46]. In the meantime, serology will predominate as the primary method of phenotyping donors and patients for ABO. This is not necessarily the case for other blood group antigens. All technologies described here have demonstrated a high degree of accuracy and concordance with serological results, in especially when focussing on the Rh system, errors by serology have been found. In the Bloodgen project, some samples defined serologically as D-positive were identified by Bloodchip as partial D (DVII), and a sample serotyped as D-C-c+ was shown to possess a ceAR allele, with associated weak D and C antigen expression (Bloodgen consortium, unpublished results). It is clear therefore that in only a limited cohort of 1000 samples, Rh variants will be found on a large scale if comprehensive Rh genotyping technologies are adopted. These individuals, if patients or pregnant women are at risk of alloimmunisation if transfused with Rh-incompatible blood. Furthermore, units of Rh-incompatible blood may be found within the donor population and eliminated for the purpose of transfusion. All of the platform technologies described have the potential for addition of new alleles, but in particular high capacity arrays
14
N.D. Avent / Transfusion Clinique et Biologique 14 (2007) 10–15
are the only feasible option for inclusion of the large numbers of ABO and RH alleles that are undoubtedly still yet to be discovered in the world’s populations. This is especially relevant as it must be stressed that still only a limited number of predominantly Western population groups have been studied to any extent. 8. Economics: genotyping versus serology Currently, blood group serology is inherently cheaper than genotyping, but this is only because a limited number of blood group alleles are tested for on a routine basis. In theory, for donation testing a single genotyping test (arrays in current form can only be processed once) could be performed and entered onto a centralised database for electronic cross-matching with patients. Whilst it is appreciated that in rare circumstances (for example, haematological malignancy) blood group phenotype may alter, blood groups should not change throughout that individuals blood donation career. With 100+ SNPs tested in one reaction series, the price per test is likely to be about one euro per SNP, but with mass scale use this price may reduce considerably. As multiple blood group genotypes can account for similar blood group phenotypes (e.g. O, D-negative, weak D, D-elute) it is imperative that large-scale analyses of blood group genotypes are performed throughout population groups worldwide. It has been known for a considerable time that certain blood groups predominate amongst various populations – for example, Fy(a-b-) in West Africans, VS and V antigens in Africans, D-elute in Asians [47]. Nevertheless, recent largescale analysis have revealed the presence of clusters of certain blood group alleles in discrete population groups. This would indicate that transfusion policy needs to take into account the genetic variation within the immediate vicinity. This could be benefit in better quality control of transfusion policy of recipients of both D-positive and D-negative units within the blood bank. A recent investigation of 33,864 donors revealed the presence of 54 partial RHD alleles amongst the serologically defined D+ samples [48]. These partial Ds if transfused with D-positive blood could become alloimmunised to the D antigen, a potentially serious issue if the individual is of child bearing age or requiring a transfusion. Using a similar approach, D-negative donations can be quality assured by identification of RHD alleles that have been known to cause immunisation in D-negative recipients (for example DEL and weak D type 4.2 (DAR)) [49]. Clearly the frequency of alloimmunisation due to D will be significantly reduced if such potentially problematic donated units are eliminated from the blood bank. 9. Mass scale genotyping? Mass scale genotyping approaches have demonstrated that a comprehensive approach to RH genotyping reaps dividends in the identification of weakened D and C antigen expression and quality assurance of D-negative donor units. The identification of partial D individuals as patients (both transfusion recipients
or patients) has clear clinical value. Whilst molecular diagnostics for genetic disease gathers pace, it makes good economic sense to include blood groups and HLA type in population screens for such diseases as cystic fibrosis, which is a target for genetic diagnosis following a positive screening test. Furthermore, it may well be worthwhile genotyping all individuals at birth for their blood group status and place this information on databases. This information will be highly useful if these individuals become patients, or eventually decide to become blood donors. Naturally, such information needs to be confidential and may raise ethical debate. However, as there is only limited information concerning the risk of disease associated with inheritance of a particular blood group phenotype, this information is not particularly controversial, but potentially lifesaving to an individual if they become transfusion dependent. It is likely that further projects on the mass scale application could be considered in the very near future in order to genotype a large number of donors within EU-based blood banks. This may give the best indication yet as to the potential success of mass scale genotyping and the identification of clinically significant variant blood group genotypes within that local donor cohort. Such an approach is essential as it will enable the streamlining of provision of compatible blood, and reduce the level of initial alloimmunisation made by vulnerable patient groups (e.g. multi-transfused individuals). Acknowledgements The author wishes to thank the Bloodgen consortium and Dr Tracey Madgett for providing the images contained within Figs. 1 and 2. References [1] Yamamoto F, Clausen H, White T, et al. Molecular genetic basis of the histo-blood group ABO system. Nature 1990;345(6272):229–33. [2] Yamamoto F, Marken J, Tsuji T, et al. Cloning and characterization of DNA complementary to human UDP-GalNAc: Fuc alpha 1——2Gal alpha 1——3GalNAc transferase (histo-blood group A transferase) mRNA. J Biol Chem 1990;265(2):1146–51. [3] Cherif-Zahar B, Bloy C, Le Van Kim C, et al. Molecular cloning and protein structure of a human blood group Rh polypeptide. Proc Natl Acad Sci U S A 1990;87(16):6243–7. [4] Avent ND, Ridgwell K, Tanner MJ, Anstee DJ. cDNA cloning of a 30 kDa erythrocyte membrane protein associated with Rh (Rhesus)-blood-groupantigen expression. Biochem J 1990;271(3):821–5. [5] Cartron JP, Colin Y. Structural and functional diversity of blood group antigens. Transfus Clin Biol 2001;8(3):163–99. [6] Reid ME, Lomas-Francis C. The Blood Group Antigens factsbook, 2nd ed, New York: Academic Press; 2004. [7] Daniels G. The molecular genetics of blood group polymorphism. Transpl Immunol 2005;14(3–4):143–53. [8] Avent ND. Blood groups: Molecular genetic basis Encyclopaedia of the Human Genome: Nature publishing group, 2003: 333–343. [9] Rouillac C, Colin Y, Hughes-Jones NC, et al. Transcript analysis of D category phenotypes predicts hybrid Rh D-CE-D proteins associated with alteration of D epitopes. Blood 1995;85(10):2937–44. [10] Mouro I, Le Van Kim C, Rouillac C, et al. Rearrangements of the blood group RhD gene associated with the DVI category phenotype. Blood 1994;83(4):1129–35.
N.D. Avent / Transfusion Clinique et Biologique 14 (2007) 10–15 [11] Wagner FF, Gassner C, Muller TH, et al. Three molecular structures cause rhesus D category VI phenotypes with distinct immunohematologic features. Blood 1998;91(6):2157–68. [12] Wagner FF, Gassner C, Muller TH, et al. Molecular basis of weak D phenotypes. Blood 1999;93(1):385–93. [13] Colin Y, Cherif-Zahar B, Le Van Kim C, et al. Genetic basis of the RhDpositive and RhD-negative blood group polymorphism as determined by Southern analysis. Blood 1991;78(10):2747–52. [14] Wagner FF, Flegel WA. RHD gene deletion occurred in the Rhesus box. Blood 2000;95(12):3662–8. [15] Singleton BK, Green CA, Avent ND, et al. The presence of an RHD pseudogene containing a 37 base pair duplication and a nonsense mutation in africans with the Rh D-negative blood group phenotype. Blood 2000;95(1):12–8. [16] Wagner FF, Frohmajer A, Flegel WA. RHD positive haplotypes in D negative Europeans. BMC Genet 2001;2(1):10. [17] Avent ND, Liu W, Jones JW, et al. Molecular analysis of Rh transcripts and polypeptides from individuals expressing the DVI variant phenotype: an RHD gene deletion event does not generate All DVIccEe phenotypes. Blood 1997;89(5):1779–86. [18] Avent ND. The rhesus blood group system: insights from recent advances in molecular biology. Transfus Med Rev 1999;13(4):245–66. [19] Avent N.D. New insight in the Rh system: stucture and function. Vox Sang 2007; in press. [20] Westhoff CM. Molecular testing for transfusion medicine. Curr Opin Hematol 2006;13(6):471–5. [21] Bennett PR, Le Van Kim C, Colin Y, et al. Prenatal determination of fetal RhD type by DNA amplification. N Engl J Med 1993;329(9):607–10. [22] Wolter LC, Hyland CA, Saul A, Rhesus. D genotyping using polymerase chain reaction. Blood 1993;82(5):1682–3. [23] Avent ND, Finning KM, Martin PG, Soothill PW. Prenatal determination of fetal blood group status. Vox Sang 2000;78(Suppl 2):155–62. [24] Aubin JT, Le Van Kim C, Mouro I, et al. Specificity and sensitivity of RHD genotyping methods by PCR-based DNA amplification. Br J Haematol 1997;98(2):356–64. [25] Ait Soussan ARD, Bonsel GJ, et al. Prenatal RHD testing of fetus and mother: Decision to administer anti-D prophylaxis. Transfus Med Hemother 2004;31(Suppl 3):50. [26] Finning KM, Martin PG, Soothill PW, Avent ND. Prediction of fetal D status from maternal plasma: introduction of a new noninvasive fetal RHD genotyping service. Transfusion 2002;42(8):1079–85. [27] Finning K, Martin P, Daniels G. A clinical service in the UK to predict fetal Rh (Rhesus) D blood group using free fetal DNA in maternal plasma. Ann N Y Acad Sci 2004;1022:119–23. [28] Legler TJ, Lynen R, Maas JH, et al. Prediction of fetal Rh D and Rh CcEe phenotype from maternal plasma with real-time polymerase chain reaction. Transfus Apher Sci 2002;27(3):217–23. [29] Lo YM, Hjelm NM, Fidler C, et al. Prenatal diagnosis of fetal RhD status by molecular analysis of maternal plasma. N Engl J Med 1998;339(24): 1734–8. [30] Van den Veyver IB, Chong SS, Cota J, et al. Single-cell analysis of the RhD blood type for use in preimplantation diagnosis in the prevention of severe hemolytic disease of the newborn. Am J Obstet Gynecol 1995;172(2 Pt 1):533–40.
15
[31] Van der Schoot CE, Soussan AA, Koelewijn J, et al. Non-invasive antenatal RHD typing. Transfus Clin Biol 2006;13(1–2):53–7. [32] Avent ND, Martin PG, Armstrong-Fisher SS, et al. Evidence of genetic diversity underlying Rh D-, weak D (Du), and partial D phenotypes as determined by multiplex polymerase chain reaction analysis of the RHD gene. Blood 1997;89(7):2568–77. [33] Rouillac-Le Sciellour C, Puillandre P, Gillot R, et al. Large-scale prediagnosis study of fetal RHD genotyping by PCR on plasma DNA from RhD-negative pregnant women. Mol Diagn 2004;8(1):23–31. [34] Le Van Kim C, Mouro I, Brossard Y, et al. PCR-based determination of Rhc and RhE status of fetuses at risk of Rhc and RhE haemolytic disease. Br J Haematol 1994;88(1):193–5. [35] van der Schoot CE, Tax GH, Rijnders RJ, et al. Prenatal typing of Rh and Kell blood group system antigens: the edge of a watershed. Transfus Med Rev 2003;17(1):31–44. [36] Lee S, Bennett PR, Overton T, et al. Prenatal diagnosis of Kell blood group genotypes: KEL1 and KEL2. Am J Obstet Gynecol 1996;175(2): 455–9. [37] Avent ND, Martin PG. Kell typing by allele-specific PCR (ASP). Br J Haematol 1996;93(3):728–30. [38] Olsson ML, Hansson C, Avent ND, et al. A clinically applicable method for determining the three major alleles at the Duffy (FY) blood group locus using polymerase chain reaction with allele-specific primers. Transfusion 1998;38(2):168–73. [39] Castilho L, Rios M, Pellegrino Jr J, et al. Blood group genotyping facilitates transfusion of beta-thalassemia patients. J Clin Lab Anal 2002;16(5):216–20. [40] Legler TJ, Eber SW, Lakomek M, et al. Application of RHD and RHCE genotyping for correct blood group determination in chronically transfused patients. Transfusion 1999;39(8):852–5. [41] Hashmi G, Shariff T, Seul M, et al. A flexible array format for large-scale, rapid blood group DNA typing. Transfusion 2005;45(5):680–8. [42] Denomme GA, Van Oene M. High-throughput multiplex single-nucleotide polymorphism analysis for red cell and platelet antigen genotypes. Transfusion 2005;45(5):660–6. [43] Bugert P, McBride S, Smith G, et al. Microarray-based genotyping for blood groups: comparison of gene array and 50 -nuclease assay techniques with human platelet antigen as a model. Transfusion 2005;45(5):654–9. [44] Beiboer SH, Wieringa-Jelsma T, Maaskant-Van Wijk PA, et al. Rapid genotyping of blood group antigens by multiplex polymerase chain reaction and DNA microarray hybridization. Transfusion 2005;45(5): 667–79. [45] Anstee DJ. Goodbye to agglutination and all that? Transfusion 2005;45(5):652–3. [46] Liu QP, Sulzenbacher G, Yuan H, et al. Bacterial glycosidases for the production of universal red blood cells. Nat Biotechnol 2007. [47] Avent ND. High variability of the RH locus in different ethnic backgrounds. Transfusion 2005;45(3):293–4. [48] Denomme GA, Wagner FF, Fernandes BJ, et al. Partial D, weak D types, and novel RHD alleles among 33,864 multiethnic patients: implications for anti-D alloimmunization and prevention. Transfusion 2005;45(10): 1554–60. [49] Flegel WA. How I manage donors and patients with a weak D phenotype. Curr Opin Hematol 2006;13(6):476–83.