Incidence and Specificity of HLA-DP Antibodies in Pregnancy Sera Gertrud Mueller-Eckhardt, Volker Kiefel, Annette Tlusty, Jiirgen Scholten, Andrea Schmidt, and Christian Mueller-Eckhardt

ABSTRACT: A total of 630 human pregnancy sera were investigated for HLA-DP antibodies (ab) by monoclonal antibody-specific immobilization of leukocyte antigens (MAILA) using monoclonal antibody (mAB) B7/21.2 and selected B-lymphoblastoid cell lines from the Tenth International Histocompatibility Workshop reference panel. DP-specific abs were detected in 86 of 330 sera (26.1%) of a retrospective series selectedfor positive reactions in lymphocytotoxicity screening, and in 29 of 300 unselected sera (9.7%) of a prospective series. Approximately 80% of DP-reactive sera were also positive in lymphocytotoxicity test for class I and~or class II ab. On the other hand, sera containing lymphocytotoxic ab for both class I and class II ab revealed the highest incidence of DP ab (35%). Out of 115 DP-reactive sera, 28 clearly presented one or more DP specificities. Absorption/elution studies revealed complex patterns of reactivity in some sera which were similar to those of known mABs and possibly reflect supertypic DP specificities and~or serological crossreactions. Serological DP typing of individuals by the MAILA technique appears promising.

ABBREVIATIONS BSA bovine serum albumin B-LCL B-lymphoblastoid cell line EBV Epstein-Barr virus ELISA enzyme-linked immunosorbent assay LCT lymphocytotoxicity test mAB monoclonal antibody MAILA monoclonal antibodyspecific immobilization of leukocyte antigens

MAIPA PFA PBL PHA 10th IHWS

monoclonal antibodyspecific immobilization of platelet antigens paraformaldehyde peripheral blood lymphocytes phythemagglutinin Tenth International Histocompatibility Workshop

INTRODUCTION Recently, a new enzyme immunoassay has been developed to detect antibodies against antigens that occur on different membrane glycoproteins o f platelets [monoclonal antibody-specific immobilization o f platelet antigens] (MAIPA) [ 1] or lymphocytes [monoclonal antibody-specific immobilization of leukocyte antigens (MAILA)] [2] by immobilization with specific monoclonal antibodies (mAB). From the Institute for Clinical lmmunologyand Transfusion Medicine,Justus Liebig University G iessen, FederalRepublic of Germany. Address reprint requeststo Dr. Gertrud Mueller-Eckhardt, Institute for Clinical lmmunology and Transfusion Medicine,Justus Liebig University, Langhansstrasse 7, D-6300 Giessen, FRG. ReceivedSeptember27, 1989; acceptedApril 20, 1990.

166 0198-8859/90/$3.50

HumanImmunology29, 166-174 (1990) © AmericanSocietyfor Histocompatibilityand Immunogenetics,1990

HLA-DP Antibodies in Pregnancy Sera

167

This technique has proven to be of special value in identifying HLA-DP-specific antibodies in human sera even if these sera contain additional antibodies against antigens of other major histocompatibility complex loci [2]. The potential availability of DP-specific antisera opens the possibility of serological DP typing, which would facilitate the study of DP-related clinical problems such as transplantation and disease associations. The present study was undertaken to evaluate the incidence and specificity of DP antibodies stimulated by pregnancy, and the possibility of obtaining serological typing reagents to be used in the MAILA technique and eventually for lymphocytotoxicity tests (LCT). MATERIALS AND METHODS Human sera. A total of 630 sera obtained from women shortly after parturition

were investigated. Of these, 451 sera originated from our own center while 179 were kindly provided by the following colleagues or companies: E. Albert, Mfinchen, FRG (n = 8); Behring AG, Marburg, FRG (n -- 20); J.Bertrams, Essen, FRG (n = 7); Biotest AG, Dreieich, FRG (n = 22); Fresenius AG, Oberursel, FRG (n = 14); E. Du Toit, Capetown, South Africa (n = 37); R. Hancock, Bristol, UK (n = 6); S. Leverenz, Berlin, GDR (n -- 6); G. M. Th. Schreuder, Leiden, The Netherlands (n = 3); F. Schunter, Rothenburg, FRG (n = 22); H. Waltz, Berlin, GDR (n -- 14); R. Wangerin, Bremen, FRG (n = 9); and E. Westphal, Kiel, FRG (n -- 11). All sera had been pretested against selected panels of peripheral T and B lymphocytes for lymphocytotoxic HLA class I and class II antibodies (ab), respectively. DP screening by MAILA against selected B-lymphoblastoid cell lines (BLCL) was performed in two series: a retrospective and a prospective one. The retrospective series of 330 sera, selected for the presence of lymphocytotoxic HLA class I and/or class II ab, included all sera sent by other laboratories. In the prospective series, 300 unselected sera from our own center were investigated, irrespective of the results of LCT. Monoclonalantibodies (mAB). For immobilization of DP molecules, mAB B7/21.2

[3], purchased from Becton Dickinson Inc., Mountain View, CA, was used. Detection of additional DR antibodies was performed using mAB L243 (DRspecific [4], Becton Dickinson). Cells. For screening of human sera by MAILA, Epstein-Barr virus (EBV)-transformed B-LCL from the Tenth International Histocompatibility Workshop (10th IHWS) reference panel [5] were employed as target cells. Cells were selected to represent all available DP antigens (DPwl-DPw5) according to the 10th IHWS assignment. Peripheral blood lymphocytes (PBL) from three DPw6-positive individuals, kindly provided by Dr. N. Odum, Copenhagen, were EBV transformed in our laboratory. After thawing, the cells were cultured for 2 to 3 weeks, then were harvested and fixed with paraformaldehyde (PFA) as described previously [2]. PFA-fixed cells could be stored at 4°C up to several weeks without loss of reactivity. The M A I L A assay. During the retrospective study, screening against B-LCL was performed as described in detail [2]. Briefly, 2 × 106 PFA-fixed cells were incubated simultaneously with 10 to 80/~l of human serum and 10 to 40 ~1 of the diluted mAB (B7/21.2:0.625 lzg purified antibodies/ml; L243:20 izg purified immunoglobulin/ml) for 30 min at 37°C. Cells were then washed, solubilized in

168

G. Mueller-Eckhardt et al. TRIS buffered saline containing 0.5% Nonidet P40, and were afterwards treated in an ultrasound waterbath. After removal of insoluble material by centrifugation, the diluted cell lysate was transferred in duplicate aliquots to microtiter trays coated with goat anti-mouse IgG (Fc) (Dianova, Hamburg, FRG). After incubation for 90 min at 4°C and washing, alkaline phosphatase-labeled goat anti-human IgG (Fc) (Dianova, diluted 1 : 2500) was added. Incubation (120 rain, 4°C) and washing was then followed by addition of substrate solution. Color development was stopped by adding 3 N NaOH. Results (AE) were expressed as the mean extinction values (405 nm) of duplicates minus the mean of at least two blanks containing TRIS wash buffer instead of the cell lysate. For the subsequent prospective study a variation of the original test procedure was introduced using horseradish peroxidase-conjugated goat anti-human IgG (Fc) (Dianova, 1 : 10000) and ortho-phenyldiamine-dihydrochloride as substrate. As this variation proved to be considerably more sensitive, only 0.5 x 106 cells and 10 to 20/~l serum were needed to give clearly discriminating AE values. With each set of MAILA tests one serum with broad DP reactivity was included as positive control, while normal sera from healthy male individuals served as negative controls. AE values surmounting the 95th percentile of negative controls were considered positive. These values were 0.109 for the retrospective series (n = 47) and 0.381 for the prospective series (n = 192).

Absorption and elution of human antibodies from pregnancy sera [6]. Absorption of DP or DR specificity from 200-/zl serum was performed twice using 4 × 106 BLCL of the appropriate HLA type; i.e., either positive for the DR specificity but negative for DP (to remove DR antibodies from sera), or vice versa (to elute DP antibodies from the cells). After incubation for 60 min at room temperature and for 10 min at 0°C, cells were washed three times, pelleted, and the supernatant (absorbate) was collected and stored. For antibody elution, cells were resuspended with elution buffer containing 150 mmol/1 acetic acid and 1.5% bovine serum albumin (BSA) titrated to pH 2.8 with HCI. After incubation on ice for 2 min and centrifugation, the supernatant (eluate) was separated, and pH was adjusted to 7.2 with 2.5 M TRIS solution. Eluates were retested in the MAILA assay against selected B-LCL to assess their final specificity. RESULTS Of 330 pregnancy sera of the retrospective series tested by MAILA against selected B-LCL (panel size 9 to 26 cells), a total of 86 (26.1%) were found to contain DP-specific antibodies. The prevalence of DP ab in the prospective series was 29 of 300 (9.7%). From the prospective series the incidence ofDP ab detected by MAIEA could be related to serum reactions in LCT. Of 29 DP positive sera, 23 (79.1%) were also positive in LCT. As shown in Table 1, the lowest incidence (6.4%) of DP ab appeared with sera lacking cytotoxic HLA ab, while the highest incidence (34.5%) was found in those containing cytotoxic class I as well as class II ab. F~om a total of 115 DP-positive sera only 28 (24.3%) revealed clear specificity for one or more DP determinants (Table 2). Fourteen sera were positive for all DP determinants tested, while for 73 sera the specificity remained unclear. While DPw3 was the most frequent specificity ofmonospecific sera, characteristic reactivity patterns of combined specificities were most frequently seen for DPw2/ w3 and DPwl/w3/w5/(w6). Table 3 presents an example of the qualitative reproducibility of DP-specific screening results.

169

TABLE 1

Detection of DP ab by MAILA in unselected pregnancy sera (prospective series) with respect to the presence or absence of lymphocytotoxic HLA class I and/or class II ab HLA ab detected by LCT n (%)

Class I/class II -/-/+ +/+/+ Total

TABLE 2

Serum Id. BIO 011931 BIO 011712 BG 1045 VR 70311.6 MUE 37133 HB 1040 MUE 892120 MUE 892364 MUE 26168 BIO 009817 BOT 44 MUE 37998 MUE 24070 MUE 49694 MUE 59847 MUE 23102 MUE 38589 MUE 30007 MUE 20686 H 1808 MUE 893831 MUE 892057 MUE 893050 MUE 24102 MUE 67192 MUE 38589 B 65897 B 54579

220 23 28 29 300

DP ab detected by MAILA n (%)

(73.3) (7.7) (9.3) (9.6) (100)

14/220 (6.4) 2/23 (8.7) 3/28 (10.7) 10/29 (34,5) 29/300 (9.7)

List of 28 DP-reactive pregnancy sera revealing clear specificity by MAILA DP specificity 1 2 3 3 3 3 4 4 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 1, 1, 1, 1, 1, 1, 1,

4 5 5 3 3 3 3 3 3 3 4 5 6 3, 3, 3, 3, 3, 3, 3,

5, 6 5, 6 5a 5a 5a 5~ 5a

a Not tested with DPw6 cells.

Panel size

X2

r-value

20 11 12 9 17 12 19 19 23 15 20 22 22 20 24 17 22 16 26 10 19 19 20 21 16 22 21 25

6.5546 2.3956 12.0000 5.7600 13.2468 8.4000 19.0000 19.0000 19.2214 11.4286 10.9091 15.0857 18.3333 20.0000 16.6201 17.0000 18.1587 12.4444 17.5500 6.4286 16.0330 12.0577 15.9184 16.8000 11.0769 18.1587 7.2894 21.0938

0.5725 0.4667 1.0000 0.8000 0.8827 0.8367 1.0000 1.0000 0.9142 0.8729 0.7385 0.8281 0.9129 1.0000 0.8322 1.0000 0.9085 0.8819 0.8216 0.8018 0.8898 0.7966 0.8921 0.8944 0.8320 0.9085 0.5892 0.9186

170

G. Mueller-Eckhardt et al.

TABLE 3 B-LCL Id.

Example of qualitative reproducibility of the MAILA a D P specificity

AE 1st testing

Result ( + / - )

AE 2nd testing

Result ( + / - )

VAVY RSH WT47 EJ32B WT51

1 1 2 2 2

1.141 1.044 1.051 0.133 0.141

+ + + -

0.962 0.779 0.591 0.253 nt b

+ + + nt b

B M 15 LOO81785 JESTHOM HOR KT17 T7527 PF04015 DO2089125 WT8 DEM KAS011 HID 2701 1603 1736

3 3 4 4 5 5 1,4 2,4 2,4 3,4 3,4 2,5 1,6 4,6JET 3,6JET

0.909 1.291 0.169 0.151 1.128 1.100 0.945 0.180 0.195 0.799 0.719 0.784 0.860 0.849 0.940

+ + + + + + + + + + +

1.086 1.126 0.158 0.195 0.879 1.089 0.622 0.170 0.186 0,887 0.848 0.736 0.948 1.046 1.101

+ + + + + + + + + + +

Serum MUE 893050 (DPwl, w3, w5, w6; r = 0.892l) was tested twice against a panel of 20 B-LCL with a time difference of 6 mon. Prospective series, 95th percentile of AE for negative controls: 0.381. b nt, not tested.

Selected sera were further assessed by MAILA for the presence of additional DR antibodies [2] which would interfere with their use for DP typing in LCT. Some sera revealing these antibodies were analyzed by absorption/elution with selected B-LCL. By this procedure, separation of DP from DR-specific antibodies was successful in some cases but not in others, and characteristic cross-absorption patterns were observed for DP specificities. Typical examples are depicted in Fig. 1. While anti-DR3 could be clearly separated from DP, DPw2 as well as DPw4 specificities were eluted together, regardless of whether absorption/elution was performed with DPw2 or DPw4 cells (Fig. 1A). One of three sera with DPw2/ DPw3 and additional DR specificity is depicted in Fig. lB. In these cases, absorption with either DPw2 or DPw3 removed DPw2 and DPw3 as well as DR5 activity, and all three specificities were recovered in the eluates. DISCUSSION Shortly after the detection of a third multialldic gene locus (SB) within the HLAD region by cellular techniques [7,8], later named DP [9], it became evident that in principle these antigens can also be serologically determined [10,11]. Several investigators were able to generate DP-specific mAbs [3,10,12-19], and a human antiserum was raised by planned immunization [20.]. During the 10th IHWS [21], these and other serological reagents [22] were investigated; however, they were not suitable for reliable DP typing by enzyme-linked immunosorbent assay (ELISA) technigues or lymphocytotoxicity [18,23,24]. There is growing evidence that DP antigens play a significant role in immunoregulation, transplantation, and disease associations [25-32], which implicates the need for serological reagents to be used in routine techniques.

HLA-DP Antibodies in Pregnancy Sera

171

Serum 20686 (DP2,4; DR3)

Eluate DP2 (WJR)

Eluate DP4 (HOM2)

Eluate DR3 (VAVY)

2.0 _

~E

A

I

1.0

DP2 WJR

2.0 _

~E-

B

1.0

I ;;__

DP4 DR3 PROM2 VAVY

DP2 WJR

Serum 23102 (DP2,3; DR5 +)

DP4 DR3 HOM2 VAVY

DP2 WJR

DP4 DR3 HOM2 VAVY

Eluate DP2 0N3"51)

1111 I1.,

DP2 DP3,4 DR5 DR6 DR7 w'p31 KAS BM16 HOR MOLl

DP2 DP3 DR,5 DR6 DR7 W'F51 COX BM18 HOR MOU

DP2 DP4 DR3 MOLl HOM2 VAW Eluate DP3 (KAS)

DP2 DP3 DR5 DR6 DR7 W~I KAS JBUSHWT47 IBW9

FIGURE 1 MAILA analysis of complex specificity patterns of DP and DR ab in pregnancy sera after absorption/elution with selected B-LCL. (A) Eluat~s of serum 20686 (DP2,4; DR3) after absorption with WJR (DP2; DR2), HOM2 (DP4; DR1), and VAVY (DP3; DR3). (B) Eluates of serum 23102 (DP2,3; DR5 +) after absorption with WT51 (DP2; DR4) and KAS011 (DP3,4; DR2). ~-DP-specific immobilization (Mab B7/21.2); m-DR-specific immobilization (Mab L243). Presently, there is little information on the incidence of D P immunization during pregnancy. Furthermore, to our knowledge, no systematic studies have been performed to describe serological cross-reactions among different DP-locus products as well as between D P antigens and those of other class II loci. In the present study we were able to show by the MAILA technique that about 10% o f unselected pregnancy sera and 26% of sera containing cytotoxic class I and/or class II antibodies exhibit D P reactivity. This is a considerably higher rate of D P antibodies than was found in an earlier study of comparable size using a radioimmune assay [22]. Some variation of the strength of reactions in different or repeated tests was observed, which is possibly due to variable antigen expression at different stages of cell cultures. However, qualitative reproducibility of the MAILA concerning positive or negative results was excellent (Table 3). The majority o f sera with clear correlation to D P antigens showed characteristic reaction patterns indicating more than one DP specificity; i.e., DPw2/w3, D P w l / w 3 / w5, or DPw2/w4. Similarly complex reactivities were demonstrated by some mAbs [ 14,16,18,19,24 ]. The combined specificities DPw 1/w3/w5 and DPw2/w4 have recently been shown to be included in supertypic D P antibodies corresponding to the allelic DP~-chain amino acid sequences DEAV and G G P M [ 17,18,22]. With respect to our absorption/elution results it seems likely that at least part of the sera with complex reactivity also contain antibodies against supertypic D P

172

G. Mueller-Eckhardt et al. determinants such as W T and B U T [18]. The observation that in some experiments additional D R reactivity could not be separated from D P antibodies points to the existence o f serological cross-reactions across gene products of DP and D R loci. O f special interest seem to be the combined specificities D P w 2 / D P w 3 / D R 5 apparent in three sera (one example is depicted in Fig. 1), as the same pattern is expressed by the mAb ILR-1 [14]. Several other mAbs have been described to bind to D P as well as D R molecules [12-14,16]. Extensive studies are required to fully define the relation between "cross-reacting" antibodies and supertypic specificities. While the rate o f D P ab detected by MAILA is highest for sera containing lymphocytotoxic ab to both HLA class I and class II determinants, there is no marked difference for sera with either class I or class II ab (Table 1). Together with the observation that D P reactivity is also found in sera completely lacking cytotoxic abs, it can be concluded that D P ab are not at all reactive or not sufficiently reactive in LCT. This might explain the disappointing experience of several groups [10,20,22,24] including our own, that the use of LCT for D P typing with absorbed sera or eluates failed to yield conclusive results. In light of the above considerations we feel that the MAILA assay is a valuable tool for detection of D P antibodies in human sera, and potentially also for serological D P typing. As the prominent advantage o f this technique, serum absorption for elimination of class I and other class II ab can be avoided. First attempts employing pregnancy sera described here and phythemagglutinin-stimulated T lymphocytes as typing cells appeared promising. More monospecific ab have to be recruited for complete and reliable serological D P typing. ACKNOWLEDGMENTS

We thank the above-mentioned colleagues and companies for their contribution of pregnancy sera, Dr. N. Odum for providing DPw6 cells, and Ms. K. Conrath for extensive LCT screening. The expert secretarial assistance of Ms. T. Ille is gratefully acknowledged. This work was supported by Deutsche Forschungsgemeinschaft (DFG Mu 725/1-3).

REFERENCES 1. Kiefel V, Santoso S, Weisheit M, Mueller-Eckhardt C: Monoclonal antibody-specific immobilization of platelet antigens (MAIPA): A new tool for the identification of platelet-reactive antibodies. Blood 70:1722, 1987. 2. Mueller-Eckhardt G, Kiefel V, Schmidt A, Tlusty A, Santoso S, Mueller-Eckhardt C: Discrimination of antibodies against antigens of different MHC loci in human sera by monoclonal antibody-specific immobilization of leukocyte antigens. Hum Immunol 25:125, 1989. 3. Watson AJ, Demars R, Trowbridge IS, Bach FH: Detection of a novel human class II HLA-antigen. Nature 304:358, 1983. 4. Lampson LA, Levy R: Two populations of Ia-like molecules ona human B cell line. J Immunol 125:393, 1980. 5. Mickelson E, Reinsmoen N, Robbins FM, Hartzman R, Odum N, Svejgaard A, et al: HLA-DW and HLA-DP typing of the reference panel of B-lymphoblastoid cell lines. In Dupont B (ed): Immunobiology of HLA, vol 1. New York, Springer-Verlag, 1989, p. 38. 6. Hotchkiss AJ, Leissinger CA, Smith ME, Jordan JV, Kautz CA, Shulman NR: Evaluation by quantitative acid elution and radioimmunoassay of multiple classes of immuno-

HLA-DP Antibodies in Pregnancy Sera

173

globulins and serum albumin associated with platelets in idiopathic thrombocytopenic purpura. Blood 67:1126, 1986. 7. Shaw S, Johnson AH, Shearer GM: Evidence for a new segregant series of B cell antigens that are encoded in the HLA-D region and that stimulate secondary allogeneic proliferative and cytotoxic responses. J Exp Med 152:565, 1980. 8. Shaw S, Kavathas P, Pollack MS, Charmot D, Mawas C: Family studies define a new histocompatibility locus, SB, between HLA-DR and GLO. Nature 293:745, 1981. 9. Bodmer WF: The HLA system, 1984. In Albert ED, Baur MP, Mayr WR (eds)Histocompatibility Testing 1984. Berlin, Springer-Verlag, 1984, p. 11. 10. Shaw S, Mars de R, Schlossman SF, Smith PL, Lampson LA, Nadler LM: Serologic identification of the human secondary B cell antigens. J Exp Med 156:731, 1982. 11. van Leeuwen A, Termijtelen A, Shaw S, van Rood JJ: The recognition ofa polymorphic monocyte antigen in HLA. Nature 298:565, 1982. 12. Ziegler A, Heinig J, Miiller C, Gfitz H, Thinnes FP, Uchanska-Ziegler B, Wernet P: Analysis by sequential immunoprecipitations of the specificities of the monoclonal antibodies TIJ22, 34, 35, 36, 37, 39, 43, 58 and YD1/63.HLK directed against human HLA class II antigens. Immunobiology 171:77, 1986. 13. Nadler LM, Stashenko P, Hardy R, Tomaselli KJ, Yunis EJ, Schlossman SF, Pesando JM: Monoclonal antibody identifies a new IA-like (P29,34) polymorphic system linked to the HLA-D/DR region. Nature 290:591, 1981. 14~ Shaw S, Ziegler A, de Mars R: Specificity of monoclonal antibodies directed against human and murine class II histocompatibility antigens as analyzed by binding to HLAdeletion mutant cell lines. Hum Immunol 12:191, 198% 15. Heyes J, Austin P, Bodmer J, Bodmer W, et al: Monoclonal antibodies to HLA-DPtransfected mouse L cells. Immunology 83:3417, 1986. 16. Howard DR, Eaves AC, Takei F: Monoclonal antibodies to HLA-DP, DQ, and DR determinants: Functional effects on the activation and proliferation of normal and EBV-transformed B cells. Exp Hematol 14:887, 1986. 17. Kolstad A, Hannestad K: A supertypic HLA-DP specificity defined by two human-human hybridoma antibodies (TrB50; TrE11). Hum Immunol 25:247, 1989. 18. Colonna M, Tanigaki N, Tosi R, Ferrara GB: Serological detection and molecular localization of allelic HLA-DP supertypic epitopes. Eur J Immunol 19:433, 1989. 19. Viken HD, Gaudernack G, Thorsby E: Characterization of a monoclonal antibody recognizing a polymorphic epitope mainly on HLA-DPw2 and DPw4 molecules. Tissue Antigens 34:250, 1989. 20. Johnson AH, Thorsby E, Nakatsuji T, Fang T, Moen T, Hartzman RJ: Recognition of an HLA-DPwl specific alloantiserum raised by planned immunization. Hum Immun 17:21, 1986. 21. Dupont B: Immunobiology of HLA. New York, Springer-Verlag, 1989. 22. Tanigaki N, Tosi R, Sorrentino R, Ferrara GB, Strominger JL: Detection of HLA-DP serological ailodeterminants by the use of radioiodinated DP molecules. EurJ Immunol 17:743, 1987. 23. Mueller-Eckhardt G, Burth A: Preliminary local experience with HLA-DP serology. 10th IHW Newsletter 3:6, 1987. 24. Johnson AH, Park MS, Bodmer JG, Kennedy L, Marsh S, Ferrara GB, Bratlie A, et al: Antigen Society no. 32 Report (DP). In Dupont B (ed): Immunobiology of HLA, vol. 1. New York, Springer-Verlag, 1989, p 286.

174

G. Mueller-Eckhardt et al.

25. Eckels DD, Lake P, Lamb JR, Johnson AH, Shaw S, Woody JN, Hartzman RJ: SBrestricted presentation of influenza and herpes simplex virus antigens to human Tlymphocyte clones. Nature 301:716, 1983. 26. Pawelec G, Ehninger G, Schmidt H, Wernet P: HLA-DP matching and graft-versushost disease in allogeneic bone marrow transplantation. Transplantation 42:558, 1986. 27. Sanchez-Perez N, Shaw S: HLA-DP: Current status. In Solheim BG, et al (eds): HLA class II antigens. Berlin, Springer-Verlag, 1986, p 83. 28. Odum N, Platz P, Morling N, Jacobsen N, Jakobsen BK, Ryder LP, Svejgaard A: Increased frequency of HLA-DPw3 in severe aplastic anemia (AA). Tissue Antigens 29:184, 1987. 29. Pawelec G, Ehninger G, M~iller C, Blaurock M, Schneider EM, Wernet P: Human leukocyte antigen-DP in leukemia. Cancer 61:475, 1988. 30. Celis E, Karr RW: Presentation of an immunodominant T-cell epitope of hepatitis B surface antigen by the HLA-DPw4 molecule. J Virol 63:747, 1989. 31. Odum N, Petersen C, Jakobsen B, Jacobsen N, Morling N, Lamm L, Svejgaard A: HLA-DP and acute graft versus host disease. In Dupont B (ed): Immunobiology of HLA. New York, Springer-Verlag, 1989, p 517. 32. Bugawan TL, Angelini G, Larrick J, Auricchio S, Ferrara GB, Erlich HA: A combination of a particular HLA-DP 3 allele and an HLA-DQ heterodimer confers susceptibility to coeliac disease. Nature 339:470, 1989.