Massscale highthroughput multiplex ... - Wiley Online Library

HEMAPHERESIS Mass-scale high-throughput multiplex polymerase chain reaction for human platelet antigen single-nucleotide polymorphisms screening of apheresis platelet donors _3082

2028..2033

Nadine Shehata, Gregory A. Denomme, Barbara Hannach, Nancy Banning, and John Freedman

BACKGROUND: Treatment with human platelet antigen (HPA)-matched platelets (PLTs) is the optimal therapy for bleeding secondary to neonatal alloimmune thrombocytopenia. Recent advances in high-throughput DNA-based blood group and PLT antigen genotyping have made it possible to screen plateletpheresis donors for potential HPA-matched PLT transfusion. STUDY DESIGN AND METHODS: This prospective study evaluated genomic DNA from plateletpheresis donors for single-nucleotide polymorphisms (SNPs) associated with HPA-1, -2, -3, -4, -5, and -15 to determine whether high-throughput multiplex genomic DNA PCR and oligonucleotide extension technology can be used for mass-scale PLT antigen genotyping. Genotyping using SNP technology was confirmed using sequence-specific polymerase chain reaction (SSP-PCR). RESULTS: Of the 748 donors screened, 277 were found to be negative for antigens implicated in alloimmune thrombocytopenia. In addition, two donors were homozygous for HPA-1b/b and -2b/b, six donors for HPA-1b/b and -3b/b, one for HPA-2b/b and -3b/b, one for HPA-1b/b and -5b/b, 10 for HPA-1b/b and -15 b/b, four for HPA-5b/b and -15b/b, and one for HPA-2b/b and -15b/b. Retesting using SSP-PCR was conducted for 60 donors. Discrepant results occurred between SNP and SSP-PCR in less than 20% of samples for HPA-1b/1b/HPA-3b/3b, HPA-5b/5b, and HPA-15b/b. DISCUSSION: High-throughput multiplex PCR SNP and confirmatory molecular genotyping are useful for massscale screening of apheresis PLT donors to provide antigen-negative genotypes. Refinements to mass-scale multiplex analysis technology would reduce further the confirmatory testing needed.

2028 TRANSFUSION

Volume 51, September 2011

H

uman platelet (PLT)-specific antigens have been recognized since 1959 with the discovery of the human PLT antigen (HPA)-1 (PlA1) system.1 The amino acid changes leading to PLT antigenicity in humans are all due to singlenucleotide polymorphisms (SNPs) in the genes for one of seven polymorphic proteins expressed on PLTs.2 The PLT antigen nomenclature is unique in the sense that some of the PLT antigen “systems” reside on the same moiety. For example, HPA-1 and HPA-4 systems are both expressed on the glycoprotein IIIa and are the result of two different amino acid substitutions.2 The antigens resemble “systems” because their alloimmunization profiles are independent of each other. Historically, PLT antigens have been identified with the use of specific well-characterized antibodies. These “reagents” are derived from alloimmunized maternal and transfusion recipient plasma and or sera that have been ABBREVIATIONS: HPA = human platelet antigen; SBE = singlebase extension; SNP(s) = single-nucleotide polymorphism(s); SSP-PCR = sequence-specific polymerase chain reaction. From the Department of Medicine, University of Toronto, the Li Ka Shing Knowledge Institute in the Keenan Research Centre of St. Michael’s Hospital, and Canadian Blood Services, Toronto, Ontario, Canada; and the Immunohematology Reference Laboratory, Blood Center of Wisconsin, Milwaukee, Wisconsin. Address reprint requests to: Nadine Shehata, MD, Queen Wing, 2-065c, St. Michael’s Hospital, 30 Bond Street, Toronto, Ontario, Canada, M5B 1W8; e-mail: [email protected]. on.ca. Supported by Grants XT00060 and XT00061 Research & Development, Canadian Blood Services, Ottawa, Ontario, Canada. Canadian Blood Services as a funding agency did not have any role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, and approval of the manuscript. Received for publication August 17, 2010; revision received January 7, 2011, and accepted January 7, 2011. doi: 10.1111/j.1537-2995.2011.03082.x TRANSFUSION 2011;51:2028-2033.

MASS-SCALE GENOTYPING OF PLATELETPHERESIS DONORS

have donated 21 to 40 times. In total, the Toronto Centre collects approximately 8000 apheresis PLT units per year. Given the published HPA frequencies for Caucasians,8,9 we anticipated that 15 HPA-1a–negative donors would be identified. We anticipated that sufficient numbers of plateletpheresis donors would be identified for each of the other HPA systems as summarized for the expected frequencies in Table 1. We did not anticipate that any homozygous HPA-4a donors would be identified, because our plateletpheresis donors are primarily Caucasian, although empirical data on Canadian donors has not been collected. HPA genotyping was performed using a commercially available genotyping system (GenomeLab SNPstream, Beckman Coulter, Brea, CA) high-throughput multiplex PCR SNP technology (Fig. 1), which has been demonstrated to have a greater than 98% success rate for red blood cell (RBC) blood antigen genotyping.10,11 In this study, multiplex PCR primers were designed to flank the six HPA SNPs; primers were designed to amplify both the sense and the antisense strand for each SNP. After PCR, the amplified fragments were made single stranded by exonuclease digestion, and then single-base extension (SBE) “probes” were annealed. SBE probes were a hybrid oligonucleotide; one portion annealed to the PCRamplified target immediately proximal to the SNP of interest, the “tag” portion (indicated in Fig. 1), was designed to immobilize the annealed complex to a microchip for laser activation and fluorescence. One or both fluorescent nucleotides were incorporated and the HPA genotype was inferred by fluorescence emission. Six DNA samples with HPA-1, -2, -3, -4, -5, and -15 genotype previously

extensively tested for their anti-PLT antigen specificity. However, PLT antigen phenotyping relies on the availability of these antibodies1 and it may not be possible to detect these antigens even in experienced laboratories.3 Furthermore, assays employed, for example, flow cytometry and monoclonal antibody immobilization of PLT antigens, were complex and time-consuming and sometimes difficult to interpret. In the mid-1990s, advances in knowledge of the nucleotide changes leading to the expression of PLT antigens and in DNA technology allowed easy DNA-based genotyping for PLT antigens.4,5 Genotypings were performed mainly by manual polymerase chain reaction (PCR) restriction length polymorphism methods,1,6 but more recently, high-throughput multiplex genomic DNA PCR and oligonucleotide extension technology has provided the opportunity to perform mass-scale genotyping.7 The purpose of this study was to determine whether the technology can be used to screen plateletpheresis donors for the clinically significant human PLT antigens, HPA-1 through -5 and -15, to provide antigen-negative donors for transfusion to alloimmunized patients. Since these transfusions may often be required on short notice, it would be useful to have previously identified suitable donors available.

MATERIALS AND METHODS This was a prospective study of 750 plateletpheresis donors at Canadian Blood Services’ Toronto Centre. The Centre has 2400 plateletpheresis donors, including 1500 who donate frequently. Approximately 23% of plateletpheresis donors have donated two to five times, and 19%

TABLE 1. Summary of mass-scale high-throughput HPA SNP screening of apheresis PLT donors* HPA 1as dbSNP# Major frequency nucleotide Minor frequency nucleotide Homozygous (major) Heterozygous Homozygous (minor) Genotype frequencies (%) Homozygous “a” (major) Heterozygous Homozygous “b” (minor) Allele frequency (% observed) “a” (major) allele “b” (minor) allele Allele frequency (% expected) “a” (major) allele “b” (minor) allele

1s

rs5918 T C 498 516 170 185 17 16

2as

2s

rs6065 C T 594 611 90 94 10 5

3s rs5911 T G 308 321 88

4as 4s rs5917 G A 672 672 0 0 0 0

5as 5s rs10471371 G A 532 531 108 124 9 7

15as 15s rs10455097 C A 184 184 350 352 149 153

73 25 2.5

72 26 2.2

86 13 1.4

86 13 0.6

43 45 12

100 0 0

100 0 0

82 17 1.4

80 19 1.1

27 51 22

27 51 22

85 15

85 15

92 8

93 7

65 35

100 0

100 0

90 10

90 10

53 47

52 48

85 15

83 17

70 30

100 0

93 8

60 40

* HPA genotypes were determined from antisense (as) and sense (s) analysis for each SNP-associated HPA genotype. The numbers (N) of homozygous HPA “a” (major) and HPA “b” (minor) and heterozygous genotypes were enumerated from the high-throughput data. The observed allele frequencies were calculated from the genotype numbers. The observed antisense and sense allele frequencies were compared to the expected published frequencies for Caucasians. Five of 10 homozygous HPA-2b/b (minor) genotypes showed a significant discordance between the antisense and sense strand SNP analysis (1.4% vs. 0.6%). Only HPA-3 sense data were evaluated due to poor PCR amplification of the HPA3as design.


TRANSFUSION 2029

SHEHATA ET AL.

G 3’ PCR 5’ amplification

3’ C

PCR

SNP

5’ Exonuclease digestion

3’ ssDNA target

C Hybrid ‘capture’

Tag 3’

5’

C

Probe annealing

Labeled 5’ Nucleotide 5’ extension

3’ C

Fig. 1. Schematic diagram of high-throughput multiplex PCR SNP detection using oligonucleotide extension technology.

determined by manual sequence-specific PCR (SSP-PCR) were used as controls, but not all genotypes were represented. Confirmatory testing was performed by SSP-PCR (either in-house established assay or GTI-ASP genotyping kit, Madison, WI) on a subsequent donation when highthroughput sense and antisense SNP results indicated a homozygous minor HPA genotype. Amplification failures and discordant results between sense and antisense assays were not further analyzed. We genotyped for the SNPs associated with HPA-1, -2, -3, -4, -5, and -15. Appendix S1 (available as supporting information in the online version of this paper) outlines the HPA systems, the nucleotide polymorphisms, the PCR primers, and the SBE probes required for the analyses.10 The multiplex PCR primers and SBE probes were designed using the computer-assisted software provided by Beckman Coulter, Inc. (at http://www.autoprimer.com). The minor homozygous genotypes identified by highthroughput SNP were confirmed by standard methods, that is, SSP-PCR. The study protocol was approved by the Canadian Blood Services Research Ethics Board.

RESULTS Table 1 illustrates the genotype frequencies for clinically relevant PLT antigens among 748 plateletpheresis donors. Overall, 277 donors were found to be homozygous for a minor frequency allele by SNP analysis. In addition, several donors were noted to be homozygous for multiple minor frequency alleles. Amplification failures ranged from 4.8% (HPA-3s) to 0.9% (HPA-5s). HPA-3as was not included in the analysis due to poor amplification noted during in the design stage of the multiplex assay development. Sixty (22%) donors who typed homozygous for minor frequency allele were tested on a subsequent donation by manual SSP-PCR HPA genotyping to confirm the SNPderived genotype of these donors. Table 2 compares the 2030 TRANSFUSION


TABLE 2. Correlation of homozygous minor genotype results using high-throughput multiplex SNP screening and confirmatory SSP-PCR HPA system HPA-1b/b HPA-2b/b HPA-3b/b HPA-5b/b HPA-15b/b

Number of genotypes identified by SNP technology 17 10 23 1 9

Confirmed results identified by SSP-PCR* 15 (88) 5 (50) 22 (96) 1 (100) 8 (89)

* Data are reported as number (%).

genotyping results between high-throughput SNP technology and SSP-PCR confirmation for the donors. Overall, 83% of the SNP-derived genotypes were confirmed. Fifteen of 17 (88%) HPA-1b/b donors identified by highthroughput analysis were confirmed with manual SSPPCR genotyping, and 35 of 43 (81%) donors identified by high-throughput SNP technology were confirmed to be homozygous for more than one low-frequency allele (Table 2). However, a much lower confirmatory rate was observed for the HPA-2b/b SNP-derived genotype. This higher failure rate was attributed to the high discordance noted between results from the antisense and sense SNP probes, with the sense probe often the root cause of the incorrect genotype assignment. Two of 17 HPA-1b/b donors identified by high-throughput analysis were discrepant with manual SSP-PCR genotyping (Table 2). HPA2b/b discordance was high due to the discordance noted with the antisense and sense SNP analysis with the sense probe often the root cause of the incorrect genotype assignment. Among the donors homozygous for more than one minor frequency antigen by the SNP assay confirmatory testing revealed that two HPA-1b/b donors were also homozygous for HPA-2b/b, six donors were HPA-3b/b, one donor was HPA-5b/b, and 10 donors were


HPA-15b/b. Four HPA-5b/b donors were HPA-15b/b, one HPA-2b/b donor was HPA-3b/b, and one HPA-2b/b donor was HPA-15 b/b.

DISCUSSION Alloimmunization to PLT antigens can result in three known clinical conditions: fetal and neonatal alloimmune thrombocytopenia, posttransfusion purpura, and PLT transfusion refractoriness. Maternal-fetal PLT antigen incompatibility, PLT transfusions, and occasionally RBC transfusions can lead to alloimmunization to PLT antigens. The frequency of PLT specific alloantibodies in pregnancy is approximately 1 in 500 to 1 in 5000 live births.12-17 Alloimmunization as a result of PLT transfusion occurs in 2 to 17% of transfusion recipients18-24 and posttransfusion purpura is a rare entity.25 The most frequent alloantibody specificities leading to neonatal alloimmune thrombocytopenia are anti-HPA-1a and anti-HPA-5b in Caucasian patients12,26 and anti-HPA-5b and anti-HPA-4b in Japanese patients.27 Among patients with refractoriness to PLT transfusion, anti-HPA-1a anti-HPA-5b have been implicated,1,18,19 whereas among patients with posttransfusion purpura, anti-HPA-1a, HPA-3a, and HPA-3b1 are the most common.1,20 HPA-3, HPA-15, and HPA-2 account for the remainder of the clinically important PLT antigens. Patients with neonatal alloimmune thrombocytopenia require antigen-matched PLT transfusion whereas patients with posttransfusion purpura and PLT transfusion refractoriness may benefit from antigenmatched PLT transfusion.3,17,23,28-34 Establishing a blood donor repository for HPAmatched PLTs is beneficial because it enables timely and adequate PLT support for patients requiring these products. Phenotyping blood donors for HPAs can be limited by the availability of antibodies and genotyping using DNA techniques such as manual PCR is time-consuming. High-throughput DNA-based PLT genotyping has provided the opportunity to perform mass-scale screening of blood donors, particularly useful when large numbers of donors need to be screened for rare antigen types. For example, the frequency of neonatal alloimmune thrombocytopenia is reported at 1 in 1000 to 1 in 2000 live births.35 Because there are approximately 450,000 live births in Canada,36 an estimated 225 to 450 births may be complicated by neonatal alloimmune thrombocytopenia per year and one to two HPA PLT products should be available daily for potentially affected neonates. Assuming that 2% of the population is negative for HPA-1a, then approximately 11,000 to 22,500 donors would need to be screened to maintain an active registry by the Canadian Blood Services, which maintains the HPA inventory for all of Canada (excluding Quebec) based on forecasted needs. The use of this technology would enable mass screening of this large number of donors. In addition, this technology is not

costly. Direct costs including DNA extraction, consumables, and staff time but excluding capital equipment investments total $7500 Canadian dollars ($20/sample) for 384 samples (including controls) compared to approximately $150 for each genotyping by manual SSP-PCR. The process requires only one extra blood sample, and results are available the next day. We screened 748 apheresis donors using a multiplex high-throughput SNP genotyping platform to identify 60 potential donors with a minor HPA genotype or combination of minor genotypes. The mass-scale genotyping platform can accommodate up to 384 DNA samples (378 test samples and six controls) per run, therefore providing sufficient throughput to test all accrued apheresis donor samples in two assays. The HPA genotypes of 51 of 60 donors were confirmed correct using a manual confirmatory genotyping method. For our testing algorithm, high-throughput multiplex PCR SNP provided an inexpensive screen of apheresis donors for minor homozygous HPA genotypes, and confirmatory molecular genotyping confirmed these antigen genotypes. Further refinements to mass-scale multiplex analysis technology would optimize the confirmatory testing needed to provide antigen-negative apheresis PLTs for alloimmunized patients.

ACKNOWLEDGMENTS The authors thank Ms Susan Farkas, Dr Ines Bonacossa, Ms Zofia Salomon de Friedberg, and the apheresis clerks and nurses at Canadian Blood Services Toronto Centre for their assistance.

CONFLICT OF INTEREST The authors declare that they have no conflicts of interest relevant to the manuscript submitted to TRANSFUSION.

REFERENCES 1. Kroll H, Kiefel V, Santoso S. Clinical aspects and typing of platelet alloantigens. Vox Sang 1998;74(Suppl 2):345-54. 2. Santoso S. Human platelet alloantigens. Transfus Apher Sci 2003;28:227-36. 3. Goldman M, Filion M, Proulx C, Chartrand P, Decary F. Neonatal alloimmune thrombocytopenia. Transfus Med Rev 1994;8:123-31. 4. Porretti L, Lazzaroni S, Rebulla P, Poli F, Scalamogna M, Sirchia G. Use of a rapid method for genotyping human platelet antigen systems in neonatal alloimmune thrombocytopenia. Haematologica 1997;82:600-1. 5. Ficko T, Galvani V, Rupreht R, Dovc T, Rozman P. Real-time PCR genotyping of human platelet alloantigens HPA-1, HPA-2, HPA-3, HPA-5 is superior to the standard PCR-SSP method. Transfus Med 2004;14:425-32. Volume 51, September 2011

TRANSFUSION 2031

SHEHATA ET AL.

6. Hurd CM, Cavanagh G, Schuh A, Ouwehand WH, Metcalfe P. Genotyping for platelet-specific antigens: techniques for the detection of single nucleotide polymorphisms. Vox

genotyping: from patient to high-throughput donor screening. Vox Sang 2009;97:198-206. 8. Denomme GA, Rios M, Reid ME. Molecular protocols in transfusion medicine. New York: Academic Press; 2000. 9. European Bioinformatics Institute. The immuno polymorphism database. 2010. [cited 2010 Nov]. Available from: URL: http://www.ebi.ac.uk/ipd/hpa/freqs_1.html 10. Denomme GA, Oene MV. High-throughput multiplex single-nucleotide polymorphism analysis for red cell and platelet antigen genotypes. Transfusion 2005;45: 660-6. 11. Palacajornsuk P, Halter C, Isakova V, Tarnawski M, Farmar J, Reid ME, Chaudhuri A. Detection of blood group genes using multiplex SNaPshot method. Transfusion 2009;49: 740-9. 12. Mueller-Eckhardt C, Kiefel V, Grubert A, Kroll H, Weisheit M, Schmidt S, Mueller-Eckhardt G, Santoso S. 348 cases of suspected neonatal alloimmune thrombocytopenia. Lancet 1989;1:363-6. 13. Dreyfus M, Kaplan C, Verdy E, Schlegel N, Durand-Zaleski I, Tchernia G; the Immune Thrombocytopenia Working Group. Frequency of immune thrombocytopenia in newborns: a prospective study. Blood 1997;89:4402-6. 14. Burrows RF, Kelton JG. Fetal thrombocytopenia and its relation to maternal thrombocytopenia. N Engl J Med 1993;329:1463-6. 15. Blanchette VS, Chen L, de Friedberg ZS, Hogan VA, Trudel E, Décary F. Alloimmunization to the PlA1 platelet antigen: results of a prospective study. Br J Haematol 1990;74:20915. 16. Panzer S, Auerbach L, Cechova E, Fischer G, Holensteiner

18. 19.

20.

21.

A, Kitl EM, Mayr WR, Putz M, Wagenbichler P, Walchshofer S. Maternal alloimmunization against fetal platelet antigens; a prospective study. Br J Haematol 1995;90: 655-60. Doughty HA, Murphy MF, Metcalfe P, Waters AH. Antenatal screening for fetal alloimmune thrombocytopenia; the results of a pilot study. Br J Haematol 1995;90:321-5. Kiefel V, König C, Kroll H, Santoso S. Platelet alloantibodies in transfused patients. Transfusion 2001;41:766-70. Sanz C, Freire C, Alcorta I, Ordinas A, Pereira A. Plateletspecific antibodies in HLA-immunized patients receiving chronic platelet support. Transfusion 2001;41:762-5. McFarland JG. Posttransfusion purpura. In: Popovsky MA, editor. Transfusion reactions. 3rd ed. Bethesda (MD): AABB Press; 2007. p. 275-99. Taaning E, Simonsen AC, Hjelms E, Svejgaard A, Morling N. Platelet alloimmunization after transfusion. A prospective study in 117 heart surgery patients. Vox Sang 1997;72:23841.

2032 TRANSFUSION

ity of platelet-specific alloantibodies in HLA-immunized haematologic-oncologic patients. Transf Med 1996;6: 111-4.

Sang 2002;83:1-12. 7. Veldhuisen B, van der Schoot CE, de Haas M. Blood group

17.

22. Schnaidt M, Northoff H, Wernet D. Frequency and specific-


23. Novotny VMJ, van Doorn R, Witvliet MD, Class FHJ, Brand A. Occurrence of allogeneic HLA and non-HLA antibodies after transfusion of prestorage filtered platelets and red blood cells: a prospective study. Blood 1995;87:1736-41. 24. Godeau B, Fromont P, Seror T, Duedari N, Bierling P. Platelet alloimmunization after multiple transfusions: a prospective study of 50 patients. Br J Haematol 1991;81: 395-400. 25. Mueller-Eckhardt C. Post-transfusion purpura. Br J Haematol 1986;64:419-24. 26. Davoren A, McParland P, Barnes CA, Murphy WG. Neonatal alloimmune thrombocytopenia in the Irish population: a discrepancy between observed and expected cases. J Clin Pathol 2002;55:289-92. 27. Ohto H, Miura S, Ariga H, Ishii T, Fujimori K, Morita S; the Collaborative Study Group. The natural history of maternal immunization against foetal platelet alloantigens. Transf Med 2004;14:399-408. 28. Letsky EA, Greaves M. Guidelines on the investigation and management of thrombocytopenia in pregnancy and neonatal alloimmune thrombocytopenia. Br J Haematol 1996; 95:21-6. 29. Murphy MF, Verjee S, Greaves M. Inadequacies in the postnatal management of fetomaternal alloimmune thrombocytopenia (FMAIT). Br J Haematol 1999;105:123-6. 30. Bussel J, Kaplan C, McFarland J; the Working Party on Neonatal Immune Thrombocytopenia of the Neonatal Hemostasis Subcommittee of the Scientific and Standardization Committee of the ISTH. Recommendations for the evaluation and treatment of neonatal autoimmune and alloimmune thrombocytopenia. Thromb Haemost 1991;65: 631-4. 31. Rayment R, Birchall J, Yarranton H, Hewertson J, Allen D, Murphy MF, Roberts DJ. Neonatal alloimmune thrombocytopenia. Br Med J 2003;327:331-2. 32. Pappalardo PA, Secord AR, Quitevis P, Haimowitz MD, Goldfinger D. Platelet transfusion refractoriness associated with HPA-1a (PlA1) alloantibody without coexistent HLA antibodies successfully treated with antigen-negative platelet transfusions. Transfusion 2001;41:984-7. 33. Allen DL, Samol J, Benjamin S, Verjee S, Tusold A, Murphy MF. Survey of the use and clinical effectiveness of HPA-1a/ 5b-negative platelet concentrates in proven or suspected platelet alloimmunization. Transfus Med 2004;14:409-17. 34. Win N, Peterkin MA, Watson WH. The therapeutic value of HPA-1a-negative platelet transfusion in post-transfusion purpura complicated by life-threatening hemorrhage. Vox Sang 1995;69:138-9. 35. Williamson LM, Hackett G, Rennie J, Palmer CR, Maciver C, Hadfield R, Hughes D, Jobson S, Ouwehand WH. The natural history of fetomaternal alloimmunization to the


platelet-specific antigen HPA-1a (PlA1, Zwa) as determined by antenatal screening. Blood 1998;92:2280-7. 36. Statistics Canada. Pregnancy outcomes by province or territory of residence. 2010. [cited 2010 Nov]. Available from: URL: http://www40.statcan.ca/l01/cst01/hlth64a-eng.htm

SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article:

Appendix S1. HPA systems, nucleotide polymorphisms, PCR primers, and single base extension probes use to perform mass-scale high-throughput multiplex PCR screening for HPA genotypes. Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.


TRANSFUSION 2033