- Email: [email protected]
The development of a simple and quick enzyme-linked immunosorbent assay for anti-HPA1a (PLA1) antibodies
THROMBOSIS RESEARCH 69; 395-400,1993 0049-3848/93 $6.00 + .OOPrinted in the USA. Copyright (c) 1993 Pergamon Press Ltd. All rights reserved.
BRIEF COMMUNICATION THE DEVELOPMENT OF A SIMPLE AND QUICK ENZYMFs-LINKED IMMUNOSORBENT ASSAY FOR ANTI-RPAla (PLAl) ANTIBODIES Hagop Bessos, Roe1 Goldschmeding*, Albert von dem Borne*, Anne Atkinson, and William G Murphy** Edinburgh and South East Scotland Blood Transfusion Service, Royal Infirmary, EDINBURGH EH3 9HB; *Central Laboratories of the Red Cross, Amsterdam; and *"University of Edinburgh
(Received 26.10.1992; accepted in revised form 7.12.1992 by Editor M.P. Esnouf) (Received by Executive Editorial Office 7.1 .1993)
HPAl is a polymorphic determinant on the platelet membrane glycoprotein (GP) IIb/IIIa complex situated at the N-terminal region of GPIIIa. Its immunogenic property results in a variety of immune-mediated thrombocytopenias including post-transfusion purpura (PTP) and neonatal alloimmune thrombocytopenia (NAITP) (1). Approximately 72% of the Caucasian population is homozygous for HPAla (PLAl), 2% for HPAlb (PLA2), and 26% are heterozygous (1). Although PTP is a rare condition (approximately 200 reported cases worldwide), NAITP occurs relatively frequently at a rate of 1 in 2000 live births, with about 50% of the cases occurring in first pregnancy (1,2). The early detection of anti-HPAla antibodies may be of crucial importance as prompt diagnosis may help management of patients. There are several methods for the detection of anti-HPAla antibodies including platelet suspension immunofluorescence test (PSIF"f),monoclonal antibody immobilisation of platelet antigens (MAIPA), flow cytometry, and immunosorbent assays (x,4,5,6). Although such methods are well established and, in the case of MAIPA, very sensitive, they are time consuming and not amenable to automation and wide-scale application. We have developed an enzyme-linked immunosorbent assay (ELISA) which is sensitive, specific, cheap, quick and could potentially be used in the large scale screening for anti-HPAla antibodies.
MATERIALS AND METHODS Reagents were of the Analar grade and were purchased from BDH Chemicals Ltd (UK) or Aldrich Chemicals Ltd (UK). Anti-GPIIb and IIb/IIIa monoclonal antibodies (MAb) were provided by the Insitute for Clinical Immunology and Transfusion Medicine, Justus Liebeg University, Giessen. Anti-GPIb and IIIa MAb were purchased from Dakopatts (Denmark), and PAC-1 MAb (antiKEYWORDS: HPAl. Glycoprotein IIb/IIIa. NAITP, ELISA Correspondence to: Dr H Bessos 395
396
ELISA FOR ANTI-HPAl a ANTIBODY
Vol. 69, No. 4
activated GPIIb/IIIa) was purchased from the Cell Centre, Universtiy of Pennsylvania Medical School, USA. HRP-conjugated sheep anti-human IgG and IgM were obtained from the Scottish Antibody Production Unit, while HRP-conjugated goat anti-mouse IgG and IgM were purchased from Bio-Rad (UK). Anti-HPAla and normal sera were obtained from mothers of babies with NAITP and volunteer donors from our centre respectively. The extraction and purification of GPIIb/IIIa was based on the methods of Pytela et al and Kirchhofer et al with few modifications (7,8). Briefly, a unit of genotyped apheresed platelets (obtained from the Cambridge Transfusion Centre) was washed and the platelets extracted with 10 ml of 100 mM octyl glucopyranoside (OG) and 2 mM phenylmethylsulphonyl fluoride UK) in buffer containing 50 mM Tris/HCl and 150 mM NaCl, pH 7.5 :::s"pa*The extract was then centrifuged and the supernatant supplemented with cations (8) and mixed end-over-end with 2 gm of immobilised GRGDSPK (Peptide from Novabiochem. UK: and CNBr activated Sepharose 4B from Pharmacia) at 12 mg peptide/g Sepharose 4B. overnight at 4°C. The beads were then centrifuged in a bench centrifuge (MSE, UK) at 1000 rpm for 5 min at room temperature (RT), and the supernatant removed. The beads were washed by centrifugation until absorbence at 280 nm fell to < 0.05 (Pye Unicam, Philips, UK) and the bound GPIIb/IIIa eluted by mixing the beads thrice with 5 ml of 1 mM MgC12, 1 mM MnC12 in TBS end-over-end for 15 min at RT, with the first mixing buffer containing 1 mg/ml GRGDSP (Novabiochem. UK). supernatants, harvested by The three eluate centrifugation as above, were pooled and centrifuged once more to remove residual bead particles. After recording the absorbence of the eluate, it was dialysed twice at 4'C against 1L of 10 mM Tris/HCl, 150 mM NaCl, pH Finally, the absorbence of the dialysed eluate was noted and it was 7.4.
. .
I 1
IjRWla 1
HPAWlb I
Anti-GP
I23 HPAWla
III a MAb
i
123 HPAlb/lb
3 HPAlah
3 HPAWlb
Amount coated pg/mU sj$”
‘I-
Anti-GPllbMAb
Anti-GPlbMAb
FIG 1 The binding of anti-GPIIIa, IIb, and Ib respectively) to various amounts of coated HPA.
(50,250
and
1000
w/ml
Vol. 69, No. 4
ELISA FOR ANTI-HPAl a ANTIBODY
397
stored in aliquots at -40 C. The immobilised RGD was washed and stored at 4'c for re-use. The protein concentration and purity of the GPIIb/IIIa was assessed by the bicinchoninic assay (pierce, UK) and SDS-PAGE (silver staining) respectively as before (9). Up to 1.4 mg of GPIIb/IIIa could be obtained from one unit of apheresed platelet concentrate with over 90% purity. The ELISA consisted of the following steps: microtitre plates (Greiner, Germany) were coated with 1 fig/ml of one of the following genotyped GPIIb/IIIa (100 Hi/well): HPAlala/3a3a, HPAlblb/3b3b, HPAlala/3b3b, or HPAlblb/3a3a (coating buffer only included also for background) overnight at 37°C (ELISA coating, washing, and colour development buffers, and washing procedures were as in reference 9). The wells were washed and blocked with 5% bovine serum albumin (Boseral; Organon Technika) in coating buffer (300 ul/well) for 1 hour at 37°C. After washing, the plates were incubated with dilutions of human sera or mouse MAbs at 100 pi/well in triplicate for 1 hour at 37°C. The wells were washed again and incubated with HRP-conjugated anti-human IgG or IgM, or antimouse IgG or IgM for 1 hour at 37°C. Finally, the plates were washed and colour developed using tetramethyl benzidine as substrate. Specific absorbance at 450nm was equal to the absorbance in GP coated wells minus the absorbance in wells with coating buffer only. Human test and control samples were screened using serum dilutions of 1 in 5. RESULTS AND DISCUSSION There was a specific and dose dependent binding of coated GPIIb/IIIa to corresponding MAbs (figure 1). Anti-GPIb at relatively high concentrations
I
640
I
320
I
160 Reciprocal
I
80 of Ditution
1
I
I
60
20
10
I G
of Serum FIG. 2
The sensitivity of the HPAla ELISA, using two anti-HPAla
sera.
ELISA FOR ANTI-HPAl
398
a ANTIBODY
Vol. 69, No. 4
0
HPAla/la
ffl HPAlb/lb
GPII b@Ia
??Glycocalicin
cotp@ation r Anti-GPIII a MA b
Anti- activated GP II b/III0’ VAC- IIat 100X concentration of anti-GPIIIa MAC FIG 3
A comparison between anti-GPIIIa and PAC-1 MAb bindings. The binding to glycocalicin (major segment of GPIb) is also shown as control.
0
HPA lala HPA lblb
?? o-5. Specific Absorbance (650 nm)
II-
-.
1
1
2
Normals - controls
Anti - HPAla + controls
2 Patient
FIG 4 A practical example of HPAla ELISA as a diagnostic test.
Vol. 69, No. 4
ELISA FOR ANTI-HPAl a ANTIBODY
399
did not bind to the highest amount of coated GPIIb/IIIa. Specific binding was also observed with anti-GPIIb/IIIa MAb (data not shown). Furthermore, using two human anti-HPAla positive sera (with normal serum as negative COntrd) the assay was found to be highly sensitive, at titres of 1:80 to 1:160 (figure 2). The specificity of the assay was further demonstrated when anti-HPAla antisera bound only to HPAlala/3a3a and HPAlala/3b3b, and not to HPAlblb/3a3a and HPAlb lb/3b3b (data not shown). An assessment of PAC-1 binding to the coated GPIIb/ IIIa suggested that there was little if any activation of GPlIb/IIIa during its extraction and purification (figure 3). Even at relativelv high concentrations, PAC-1 binding to purified GPIIb/IIIa decreased sharply with dilution compared to anti-GPIIIa MAb. However, the presence of residual pentide bound to the activation site of GPIIb/IIIa cannot be excluded. Such a ueptide-GP complex could inhibit PAC-1 binding, but on the other hand may contribute to the preservation of the HPAl antigenic site. Microtiter plates coated with HPAlala/3aSa or HPAlblb/Sb3b overnight, washed, air-dried, and sealed were left at 4°C. Assays carried out weekly thereafter up to 12 weeks using these plates showed consistent and specific binding of two anti-HPAla sera, but not normal serum (data not shown). Further results to date indicate that plates stored up to 6 months perform similarly, and the assessment of coated plates stored at 4'C and -2O'C is continuing. The suitability of the ELISA for the screening of human anti-HPAla sera was assessed using sera from mothers of babies with NAITP. Thirty-one sera positive for anti-HPAla antibodies (ab), two and eight sera negative and weak respectively for anti-HPAla ab, and four sera whose ab status was unknown (all determined by PSIFT and MAIPA) were provided by CLB, Amsterdam, and tested blind in the ELISA. Subsequent analysis of the results showed 29131 agreement of positive sera; 6/8 agreement of weak sera, the other two being negative in the ELISA; the two negative sera were positive in the ELISA; and of the 4 unclear sera, two were weakly positive and two reacted with uncoated wells (similar correlations were observed when coated plates were sent to CLB for ELISA). A further benefit of the ELISA is shown in figure 4. A sixty vear old female patient who had undergone a coronary bypass followed by platelet transfusion suffered a severe thrombocytopenia about a week later. Since she was receiving henarin, it was important to determine the cause of the thrombocytopenia. Using the ELISA we were able to establish within 3.5 hours of receiving her serum sample that it contained strong anti-HPAla antibody. Kekomaki et al recently used a direct ELISA for the localization of platelet autoantigens (10). However, the GPIIb/IIIa was purified bv a different procedure and the value and cost of the ELISA as an anti-HPAla diagnostic test was not assessed. Our results show that the newly developed ELISA is a simple, cheap (cost per test E 0.20), quick, specific and sensitive assay. In exceptional cases where nonspecific reactivity occurs with uncoated wells, higher dilutions of sera may be used or confirmation may be sought using another assay. However, the stability of coated plates, the potential of developing the ELISA for other HPA antibodies, and its amenability for automation should enable the ELISA to be used widely for the screening of anti-platelet antibodies, including large scale population studies in NAITP. Furthermore, the observation of high correlation between PSIFT/MAIPA and the ELISA indicates that the ELISA could be developed as a diagnostic test. Finally, it is possible that the ELISA, due to its enhanced sensitivity and GP accessibility, may detect antibodies missed by other assays.
400
ELISA FOR ANTI-HPAl
a ANTIBODY
Vol. 69, No. 4
ACKNOWLEDGEMENTS We would like to thank: Dr S Santoso (Justus Liebeg University, Giessen) for providing us with anti-GPIIb and IIb/IIIa MAb; Dr W Ouwehand (Cambridge BTS) for providing the genotyped platelet concentrates: The Scottish Antibody Production Unit for providing the HRP-conjugated anti-human IgG and IgM; Dr V Bedian (University of Pennsylvania) for his help in delivering PAC-1; Mrs E Huiskes and Mrs B Bossers (CLB, Amsterdam) for expert technical and administrative assistance; Penelope Carter for her valuable contribution
earlier in the study; and Mrs H Forlow for typing
the paper.
REFERENCES
1.
NEWMAN, P. Platelet GPIIb-IIIa: Molecular alloantigens. Thromb. Haem., 66, 111-118 (1991).
2.
KAPLAN, C., DAFFOS, F.. FORESTIER, F., MOREL, M.C., CHESNEL, N., and TCHERNIA, G. Current trends in neonatal allimmune thrombocytopenia: Diagnosis and therapy. &, platelet immunology: fundamental and clinical aspects. Eds. Kaplan-Gouet, C., Schlegel, N., Salmon, Ch., McGregor, J., Inserm/John Libbey Eurotext Ltd., 206, 267-278 (1991).
3.
BORNE, D.E.G. KR VCN DEM. VER HEUGHT, F.W.A., OOSTERHOF. F., RIESZ, E. A simple VON, BRUTEL DE LA RIVIERE, A., and ENGELFRIET, C.P. immunofluorescence test for the detection of platelet antibodies. Br. J. Haem., 3, 195-207 (1978) .
4.
its applications KIEFEL, V. MAIPA assay and The immunohaematology. Transfusion Med., 2, 181-188 (1992).
5.
FREEDMAN, J., and HORNSTEIN, A. Simple method for differentiating between HLA and platelet specific antibodies by flow cytometry. Am. J. Haem., 3, 314-320 (1991).
6.
BORZINI, P. ELISA procedures for the characterisation of platelet specific antibodies. Evaluation of some variables. Haematologia, 76, 20-27 (1991).
7.
PYTELA. R., PIERSCHBACHER. M.D., GINSBERG, M.H., PLOW, E.F., and RUOSLAHTI, E. Platelet membrane glycoprotein IIb/IIIa: member of a Science, 231, family of Arg-Gly-Asp-specific adhesion receptors. 1559-1561 (1986).
8.
GRZESIAK, J., and KIRCHHOFFER D., GAILIT, J., RUOSLAHTI, E., the binding in Cation-dependent changes PIERSCHBACHER, M.D. specificity of the platelet receptor GP IIb/IIIa. J. Biol. Chem., 265, 18525-18530 (1990).
9.
BESSOS. H., and MURPHY, W.G. A new competitive binding enzyme-linked and platelet immunosorbent assay for glycocalicin in plasma concentrate supernatants. Thromb. Res., ;s, 497-507 (1990).
10.
variations
and
in
KEKOMAKI, R., DAWSON, B., MCFARLAND, J., and KUNICKI, T.J. Localization of human platelet autoantigens to the cysteine-rich tein IIIa. J. Clin. Invest., 88, 847-854 (1991).
region of glycopro-
BRIEF COMMUNICATION THE DEVELOPMENT OF A SIMPLE AND QUICK ENZYMFs-LINKED IMMUNOSORBENT ASSAY FOR ANTI-RPAla (PLAl) ANTIBODIES Hagop Bessos, Roe1 Goldschmeding*, Albert von dem Borne*, Anne Atkinson, and William G Murphy** Edinburgh and South East Scotland Blood Transfusion Service, Royal Infirmary, EDINBURGH EH3 9HB; *Central Laboratories of the Red Cross, Amsterdam; and *"University of Edinburgh
(Received 26.10.1992; accepted in revised form 7.12.1992 by Editor M.P. Esnouf) (Received by Executive Editorial Office 7.1 .1993)
HPAl is a polymorphic determinant on the platelet membrane glycoprotein (GP) IIb/IIIa complex situated at the N-terminal region of GPIIIa. Its immunogenic property results in a variety of immune-mediated thrombocytopenias including post-transfusion purpura (PTP) and neonatal alloimmune thrombocytopenia (NAITP) (1). Approximately 72% of the Caucasian population is homozygous for HPAla (PLAl), 2% for HPAlb (PLA2), and 26% are heterozygous (1). Although PTP is a rare condition (approximately 200 reported cases worldwide), NAITP occurs relatively frequently at a rate of 1 in 2000 live births, with about 50% of the cases occurring in first pregnancy (1,2). The early detection of anti-HPAla antibodies may be of crucial importance as prompt diagnosis may help management of patients. There are several methods for the detection of anti-HPAla antibodies including platelet suspension immunofluorescence test (PSIF"f),monoclonal antibody immobilisation of platelet antigens (MAIPA), flow cytometry, and immunosorbent assays (x,4,5,6). Although such methods are well established and, in the case of MAIPA, very sensitive, they are time consuming and not amenable to automation and wide-scale application. We have developed an enzyme-linked immunosorbent assay (ELISA) which is sensitive, specific, cheap, quick and could potentially be used in the large scale screening for anti-HPAla antibodies.
MATERIALS AND METHODS Reagents were of the Analar grade and were purchased from BDH Chemicals Ltd (UK) or Aldrich Chemicals Ltd (UK). Anti-GPIIb and IIb/IIIa monoclonal antibodies (MAb) were provided by the Insitute for Clinical Immunology and Transfusion Medicine, Justus Liebeg University, Giessen. Anti-GPIb and IIIa MAb were purchased from Dakopatts (Denmark), and PAC-1 MAb (antiKEYWORDS: HPAl. Glycoprotein IIb/IIIa. NAITP, ELISA Correspondence to: Dr H Bessos 395
396
ELISA FOR ANTI-HPAl a ANTIBODY
Vol. 69, No. 4
activated GPIIb/IIIa) was purchased from the Cell Centre, Universtiy of Pennsylvania Medical School, USA. HRP-conjugated sheep anti-human IgG and IgM were obtained from the Scottish Antibody Production Unit, while HRP-conjugated goat anti-mouse IgG and IgM were purchased from Bio-Rad (UK). Anti-HPAla and normal sera were obtained from mothers of babies with NAITP and volunteer donors from our centre respectively. The extraction and purification of GPIIb/IIIa was based on the methods of Pytela et al and Kirchhofer et al with few modifications (7,8). Briefly, a unit of genotyped apheresed platelets (obtained from the Cambridge Transfusion Centre) was washed and the platelets extracted with 10 ml of 100 mM octyl glucopyranoside (OG) and 2 mM phenylmethylsulphonyl fluoride UK) in buffer containing 50 mM Tris/HCl and 150 mM NaCl, pH 7.5 :::s"pa*The extract was then centrifuged and the supernatant supplemented with cations (8) and mixed end-over-end with 2 gm of immobilised GRGDSPK (Peptide from Novabiochem. UK: and CNBr activated Sepharose 4B from Pharmacia) at 12 mg peptide/g Sepharose 4B. overnight at 4°C. The beads were then centrifuged in a bench centrifuge (MSE, UK) at 1000 rpm for 5 min at room temperature (RT), and the supernatant removed. The beads were washed by centrifugation until absorbence at 280 nm fell to < 0.05 (Pye Unicam, Philips, UK) and the bound GPIIb/IIIa eluted by mixing the beads thrice with 5 ml of 1 mM MgC12, 1 mM MnC12 in TBS end-over-end for 15 min at RT, with the first mixing buffer containing 1 mg/ml GRGDSP (Novabiochem. UK). supernatants, harvested by The three eluate centrifugation as above, were pooled and centrifuged once more to remove residual bead particles. After recording the absorbence of the eluate, it was dialysed twice at 4'C against 1L of 10 mM Tris/HCl, 150 mM NaCl, pH Finally, the absorbence of the dialysed eluate was noted and it was 7.4.
. .
I 1
IjRWla 1
HPAWlb I
Anti-GP
I23 HPAWla
III a MAb
i
123 HPAlb/lb
3 HPAlah
3 HPAWlb
Amount coated pg/mU sj$”
‘I-
Anti-GPllbMAb
Anti-GPlbMAb
FIG 1 The binding of anti-GPIIIa, IIb, and Ib respectively) to various amounts of coated HPA.
(50,250
and
1000
w/ml
Vol. 69, No. 4
ELISA FOR ANTI-HPAl a ANTIBODY
397
stored in aliquots at -40 C. The immobilised RGD was washed and stored at 4'c for re-use. The protein concentration and purity of the GPIIb/IIIa was assessed by the bicinchoninic assay (pierce, UK) and SDS-PAGE (silver staining) respectively as before (9). Up to 1.4 mg of GPIIb/IIIa could be obtained from one unit of apheresed platelet concentrate with over 90% purity. The ELISA consisted of the following steps: microtitre plates (Greiner, Germany) were coated with 1 fig/ml of one of the following genotyped GPIIb/IIIa (100 Hi/well): HPAlala/3a3a, HPAlblb/3b3b, HPAlala/3b3b, or HPAlblb/3a3a (coating buffer only included also for background) overnight at 37°C (ELISA coating, washing, and colour development buffers, and washing procedures were as in reference 9). The wells were washed and blocked with 5% bovine serum albumin (Boseral; Organon Technika) in coating buffer (300 ul/well) for 1 hour at 37°C. After washing, the plates were incubated with dilutions of human sera or mouse MAbs at 100 pi/well in triplicate for 1 hour at 37°C. The wells were washed again and incubated with HRP-conjugated anti-human IgG or IgM, or antimouse IgG or IgM for 1 hour at 37°C. Finally, the plates were washed and colour developed using tetramethyl benzidine as substrate. Specific absorbance at 450nm was equal to the absorbance in GP coated wells minus the absorbance in wells with coating buffer only. Human test and control samples were screened using serum dilutions of 1 in 5. RESULTS AND DISCUSSION There was a specific and dose dependent binding of coated GPIIb/IIIa to corresponding MAbs (figure 1). Anti-GPIb at relatively high concentrations
I
640
I
320
I
160 Reciprocal
I
80 of Ditution
1
I
I
60
20
10
I G
of Serum FIG. 2
The sensitivity of the HPAla ELISA, using two anti-HPAla
sera.
ELISA FOR ANTI-HPAl
398
a ANTIBODY
Vol. 69, No. 4
0
HPAla/la
ffl HPAlb/lb
GPII b@Ia
??Glycocalicin
cotp@ation r Anti-GPIII a MA b
Anti- activated GP II b/III0’ VAC- IIat 100X concentration of anti-GPIIIa MAC FIG 3
A comparison between anti-GPIIIa and PAC-1 MAb bindings. The binding to glycocalicin (major segment of GPIb) is also shown as control.
0
HPA lala HPA lblb
?? o-5. Specific Absorbance (650 nm)
II-
-.
1
1
2
Normals - controls
Anti - HPAla + controls
2 Patient
FIG 4 A practical example of HPAla ELISA as a diagnostic test.
Vol. 69, No. 4
ELISA FOR ANTI-HPAl a ANTIBODY
399
did not bind to the highest amount of coated GPIIb/IIIa. Specific binding was also observed with anti-GPIIb/IIIa MAb (data not shown). Furthermore, using two human anti-HPAla positive sera (with normal serum as negative COntrd) the assay was found to be highly sensitive, at titres of 1:80 to 1:160 (figure 2). The specificity of the assay was further demonstrated when anti-HPAla antisera bound only to HPAlala/3a3a and HPAlala/3b3b, and not to HPAlblb/3a3a and HPAlb lb/3b3b (data not shown). An assessment of PAC-1 binding to the coated GPIIb/ IIIa suggested that there was little if any activation of GPlIb/IIIa during its extraction and purification (figure 3). Even at relativelv high concentrations, PAC-1 binding to purified GPIIb/IIIa decreased sharply with dilution compared to anti-GPIIIa MAb. However, the presence of residual pentide bound to the activation site of GPIIb/IIIa cannot be excluded. Such a ueptide-GP complex could inhibit PAC-1 binding, but on the other hand may contribute to the preservation of the HPAl antigenic site. Microtiter plates coated with HPAlala/3aSa or HPAlblb/Sb3b overnight, washed, air-dried, and sealed were left at 4°C. Assays carried out weekly thereafter up to 12 weeks using these plates showed consistent and specific binding of two anti-HPAla sera, but not normal serum (data not shown). Further results to date indicate that plates stored up to 6 months perform similarly, and the assessment of coated plates stored at 4'C and -2O'C is continuing. The suitability of the ELISA for the screening of human anti-HPAla sera was assessed using sera from mothers of babies with NAITP. Thirty-one sera positive for anti-HPAla antibodies (ab), two and eight sera negative and weak respectively for anti-HPAla ab, and four sera whose ab status was unknown (all determined by PSIFT and MAIPA) were provided by CLB, Amsterdam, and tested blind in the ELISA. Subsequent analysis of the results showed 29131 agreement of positive sera; 6/8 agreement of weak sera, the other two being negative in the ELISA; the two negative sera were positive in the ELISA; and of the 4 unclear sera, two were weakly positive and two reacted with uncoated wells (similar correlations were observed when coated plates were sent to CLB for ELISA). A further benefit of the ELISA is shown in figure 4. A sixty vear old female patient who had undergone a coronary bypass followed by platelet transfusion suffered a severe thrombocytopenia about a week later. Since she was receiving henarin, it was important to determine the cause of the thrombocytopenia. Using the ELISA we were able to establish within 3.5 hours of receiving her serum sample that it contained strong anti-HPAla antibody. Kekomaki et al recently used a direct ELISA for the localization of platelet autoantigens (10). However, the GPIIb/IIIa was purified bv a different procedure and the value and cost of the ELISA as an anti-HPAla diagnostic test was not assessed. Our results show that the newly developed ELISA is a simple, cheap (cost per test E 0.20), quick, specific and sensitive assay. In exceptional cases where nonspecific reactivity occurs with uncoated wells, higher dilutions of sera may be used or confirmation may be sought using another assay. However, the stability of coated plates, the potential of developing the ELISA for other HPA antibodies, and its amenability for automation should enable the ELISA to be used widely for the screening of anti-platelet antibodies, including large scale population studies in NAITP. Furthermore, the observation of high correlation between PSIFT/MAIPA and the ELISA indicates that the ELISA could be developed as a diagnostic test. Finally, it is possible that the ELISA, due to its enhanced sensitivity and GP accessibility, may detect antibodies missed by other assays.
400
ELISA FOR ANTI-HPAl
a ANTIBODY
Vol. 69, No. 4
ACKNOWLEDGEMENTS We would like to thank: Dr S Santoso (Justus Liebeg University, Giessen) for providing us with anti-GPIIb and IIb/IIIa MAb; Dr W Ouwehand (Cambridge BTS) for providing the genotyped platelet concentrates: The Scottish Antibody Production Unit for providing the HRP-conjugated anti-human IgG and IgM; Dr V Bedian (University of Pennsylvania) for his help in delivering PAC-1; Mrs E Huiskes and Mrs B Bossers (CLB, Amsterdam) for expert technical and administrative assistance; Penelope Carter for her valuable contribution
earlier in the study; and Mrs H Forlow for typing
the paper.
REFERENCES
1.
NEWMAN, P. Platelet GPIIb-IIIa: Molecular alloantigens. Thromb. Haem., 66, 111-118 (1991).
2.
KAPLAN, C., DAFFOS, F.. FORESTIER, F., MOREL, M.C., CHESNEL, N., and TCHERNIA, G. Current trends in neonatal allimmune thrombocytopenia: Diagnosis and therapy. &, platelet immunology: fundamental and clinical aspects. Eds. Kaplan-Gouet, C., Schlegel, N., Salmon, Ch., McGregor, J., Inserm/John Libbey Eurotext Ltd., 206, 267-278 (1991).
3.
BORNE, D.E.G. KR VCN DEM. VER HEUGHT, F.W.A., OOSTERHOF. F., RIESZ, E. A simple VON, BRUTEL DE LA RIVIERE, A., and ENGELFRIET, C.P. immunofluorescence test for the detection of platelet antibodies. Br. J. Haem., 3, 195-207 (1978) .
4.
its applications KIEFEL, V. MAIPA assay and The immunohaematology. Transfusion Med., 2, 181-188 (1992).
5.
FREEDMAN, J., and HORNSTEIN, A. Simple method for differentiating between HLA and platelet specific antibodies by flow cytometry. Am. J. Haem., 3, 314-320 (1991).
6.
BORZINI, P. ELISA procedures for the characterisation of platelet specific antibodies. Evaluation of some variables. Haematologia, 76, 20-27 (1991).
7.
PYTELA. R., PIERSCHBACHER. M.D., GINSBERG, M.H., PLOW, E.F., and RUOSLAHTI, E. Platelet membrane glycoprotein IIb/IIIa: member of a Science, 231, family of Arg-Gly-Asp-specific adhesion receptors. 1559-1561 (1986).
8.
GRZESIAK, J., and KIRCHHOFFER D., GAILIT, J., RUOSLAHTI, E., the binding in Cation-dependent changes PIERSCHBACHER, M.D. specificity of the platelet receptor GP IIb/IIIa. J. Biol. Chem., 265, 18525-18530 (1990).
9.
BESSOS. H., and MURPHY, W.G. A new competitive binding enzyme-linked and platelet immunosorbent assay for glycocalicin in plasma concentrate supernatants. Thromb. Res., ;s, 497-507 (1990).
10.
variations
and
in
KEKOMAKI, R., DAWSON, B., MCFARLAND, J., and KUNICKI, T.J. Localization of human platelet autoantigens to the cysteine-rich tein IIIa. J. Clin. Invest., 88, 847-854 (1991).
region of glycopro-