HLA Class II Differentiates Between Thyroid and Polyglandular Autoimmunity.

The HLA class II genes are susceptibility genes for autoimmune endocrine diseases; however, scarce data are available pertaining to the determinants of genetic susceptibility to polyglandular autoimmunity (PGA). A total of 300 consecutive and unselected patients with either PGA or monoglandular autoimmune thyroid disease (AITD) and 100 healthy control subjects were genotyped for the HLA class II DRB1, -DQA1, and -DQB1 alleles. Compared to patients with AITD and controls, the HLA-DRB1*03 (pc =0.001), *04 (pc<0.001), -DQA1*03 (pc<0.001), and -DQB1*02 (pc =0.001) alleles were increased in patients with PGA. When dividing patients with Hashimoto's thyroiditis (HT) into those with PGA (PGA-HT) vs. those with HT as monoglandular disease, significant differences for the DRB1*03 (pc=0.001) and DQA1*03 (pc=0.001) alleles were observed. In contrast, the DQB1*02 allele was more prevalent in PGA patients with Graves' disease (PGA-GD) vs. those with monoglandular GD (pc=0.002). The HLA-DRB1*15 (pc =0.001), -DQA1*01 (pc =0.001), -DQB1*05 (pc =0.002) and -DQB1*06 (pc =0.002) alleles were significantly less present in PGA compared to monoglandular AITD and controls, thus indicating protective alleles. The HLA class II alleles differentiate between mono- and polyglandular autoimmunity in patients with autoimmune thyroid disease.

HLA class II haplotypes differentiate between the adult autoimmune polyglandular syndrome types II and III.

BACKGROUND: Genetics of the adult autoimmune polyglandular syndrome (APS) is poorly understood. AIM: The aim of this study was to gain further insight into the genetics of the adult APS types. SITE: The study was conducted at a university referral center. METHODS: The human leukocyte antigen (HLA) class II alleles, haplotypes, and genotypes were determined in a large cohort of patients with APS, autoimmune thyroid disease (AITD), and type 1 diabetes and in healthy controls by the consistent application of high-resolution typing at a four-digit level. RESULTS: Comparison of the allele and haplotype frequencies significantly discriminated patients with APS vs AITD and controls. The HLA class II alleles DRB1*03:01 *04:01, DQA1*03:01, *05:01, DQB1*02:01, and *03:02 were observed more frequently (P<.001) in APS than in AITD and controls, whereas the alleles DRB1*15:01, DQB1*03:01, and *06:02 were underrepresented in APS vs AITD (Pc<.001) and controls (Pc<.01), respectively. The DRB1*03:01-DQA1*05:01-DQB1*02:01 (DR3-DQ2) and DRB1*04:01-DQA1*03:01:DQB1*03:02 (DRB1*04:01-DQ8) haplotypes were overrepresented in APS (Pc<.001). Combination of both haplotypes to a genotype was highly prevalent in APS vs AITD and controls (Pc<.001). Dividing the APS collective into those with Addison's disease (APS type II) and those without Addison's disease but including type 1 diabetes and AITD (APS type III) demonstrated DR3-DQ2/DRB1*04:01-DQ8 as a susceptibility genotype in APS III (Pc<.001), whereas the DR3-DQ2/DRB1*04:04-DQ8 genotype correlated with APS II (Pc<.001). The haplotypes DRB1*11:01-DQA1*05:05-DQB1*03:01 and DRB1*15:01-DQA1*01:02-DQB1*06:02 are protective in APS III but not in type II (Pc<.01). CONCLUSIONS: HLA class II haplotypes differentiate between the adult APS types II and III. Susceptible haplotypes favor the development of polyglandular autoimmunity in patients with AITD.

TRALI-new challenges for histocompatibility and immunogenetics in transfusion medicine.

Antibodies against human leukocyte antigens (HLAs) have long been associated with transfusion-related acute lung injury (TRALI). In contrast to febrile transfusion reactions and refractoriness to platelet transfusions in immunized patients, the causative antibodies in TRALI are present in the transfused blood component, i.e. they are formed by the blood donor and not by the recipient. Consequently, blood components with high plasma volume are particularly associated with TRALI. In addition to antibodies against HLAs, antibodies directed against human neutrophil antigens (HNAs) present in the plasma of predominantly multiparous female blood donors can induce severe TRALI reactions. Especially, antibodies to HLA class II and HNA-3a antigens can induce severe or even fatal ALI in critically ill patients. Over the last decade, the clinical importance of TRALI as major cause for severe transfusion-related morbidities has led to the establishment of new guidelines aimed at preventing this condition, including routine testing for HLA and -HNA antibodies for plasma donors with a history of allogeneic sensitization. This, in turn, poses new challenges for close collaboration between blood transfusion centers and histocompatibility and immunogenetics laboratories, for sensitive and specific detection of the relevant antibodies.

Evidence for gene recombination in FCGR3 gene variants.

BACKGROUND AND OBJECTIVES: The genes encoding the Fcgamma receptors (FcgammaR) IIIa and IIIb (FCGR3A and FCGR3B) are clustered on chromosome 1 band q23-24 and exhibit allelic polymorphism. We investigated the molecular basis of additional new FCGR3 genomic variation. MATERIALS AND METHODS: A segment shared by FCGR3A and FCGR3B containing the polymorphic nucleotide positions 141, 147, 227, 266, and 277 in exon 3 was cloned and sequenced from genomic DNA of 30 donors and 3 bacterial artificial chromosome (BAC) clones. A mixture consisting of isolated FCGR3B*2- and FCGR3A- plasmids was cloned and sequenced as well. Additionally, nucleotide databases were screened for clones with variant FCGR3 sequences. RESULTS: A total of 12 FCGR3 variants defined by the polymorphic positions were detected in whole blood genomic DNA from 23 of 24 FCGR3B*2 and/or FCGR3B*3 positive donors, the DNA from two of three BAC clones and in the DNA mixture of isolated FCGR3B*2- and FCGR3A- plasmids. CONCLUSION: Nucleotide exchanges of the variants were non-random and resulted from two alternative nucleotides present in one of the polymorphic position of the basic FCGR3 forms. Polymerase chain reaction (PCR) artefacts cannot be excluded as origin of new variants, but there is strong evidence that at least two variants are the result of a somatic recombination.

The FCGR2A--Arg131 variant is no major mortality factor in the elderly--evidence from a German centenarian study.

The functional single nucleotide polymorphism rs1801274 in the FCGR2A gene (His131Arg) influences the efficiency of hIgG2 binding, the main isotype produced in response to encapsulated bacteria like Streptococcus pneumoniae and Haemophilus influenzae. In contrast to the receptor with the His131 allele, FcgammaRIIa-Arg131 binds hIgG2 poorly and carriers of this variant have been shown to be much more susceptible to succumb to bacterial pneumonia or meningitis. As bacteraemic pneumonia is one of the leading causes of death in elderly individuals, we hypothesized that the Arg131 variant could be a major mortality factor in the old. We analysed the FCGR2A-His131Arg polymorphism in a group of 408 German centenarians and two samples of younger Germans aged 60-75 and 18-49 years, respectively. No statistically significant differences were observed between the three age groups, neither at the allele nor at the genotype level. Apparently, the ability to reach old age is largely unaffected by the genetically determined efficacy of the FCGR2A-based immune response. However, the severely reduced ability of FCGR2A-131Arg carriers to eliminate encapsulated bacteria must apparently be compensated by an alternative mechanism, possibly involving other genetic survival factors.

Post-transfusion purpura in a patient with HPA-1a and GPIa/IIa antibodies.

Post-transfusion purpura is a rare bleeding disorder characterized by severe and sudden thrombocytopenia within 3-12 days after blood transfusion. Typically, preformed antibodies directed against human platelet antigens, especially HPA-1a, are associated with the clinical symptoms. A 46-year-old female presenting to the hospital with acute progressive kidney insufficiency and anaemia received two units of packed red blood cells (RBC) within 2 days. On day 7, platelet count felt from 414 to 189 x 10(9) L(-1) and 1 day later dropped to 4 x 10(9) L(-1). Four platelet concentrates were applied without success. After serological confirmation of an HPA-1a antibody, the patient was treated with intravenous gamma immunoglobulin (ivIgG), and the platelet count increased to normal values on day 17. In addition to the persisting HPA-1a alloantibody, an antibody reactive with GPIa/IIa of HPA-5a- and HPA-5b-positive platelets was detected during the acute phase of thrombocytopenia. After complete remission, the patient was transfused with four units of packed RBC from HPA-1a-negative donors, and platelet counts remained normal.

Fcgamma receptor-mediated immune phagocytosis depends on the class of FcgammaR and on the immunoglobulin-coated target cell.

BACKGROUND AND OBJECTIVES: Three human Fcgamma receptors (FcgammaR) are known to mediate immune phagocytosis. A variety of different phagocytic assays have been described, but their comparability is complicated by the use of different effector cells and different antibody-coated target cells. The aim of this study was to determine the influence of these variable components on the FcgammaR-mediated phagocytosis. MATERIALS AND METHODS: We sensitized human red blood cells (RBC) with polyclonal human anti-D (huaD), or with human monoclonal anti-D of the isotypes IgG1 (huIgG1) or IgG3 (huIgG3). Sheep RBC coated with rabbit immunoglobulin (RBC-RAS) were also used. Monocytes or polymorphonuclear neutrophils (PMN) were incubated with different FcgammaR-specific antibodies or their F(ab')2 fragments to determine the contribution of the different FcgammaRs on these effector cells in the phagocytic process of different antibody-coated target cells. RESULTS: huaD-RBC and huIgG1-RBC were preferentially ingested via the FcgammaRI on monocytes and, to a minor extent, also by the FcgammaRII. PMN ingested these target cells only after induction of the FcgammaRI by interferon-gamma (IFN-gamma). huIgG3-RBC extensively formed rosettes with monocytes but were seldom ingested. RAS-RBC phagocytosis was induced primarily via the FcgammaRI on monocytes and was mediated by the FcgammaRII on PMN. CONCLUSION: When performing phagocytosis assays with different effector and target cells, one has to take into account that phagocytosis is mediated by different FcgammaR, making comparability of these assays more difficult.

HLA-antibody testing: the immune phagocytosis inhibition test is superior to the PRA-STAT and NIH lymphocytotoxic test with respect to specificity.

We compared the specificity and sensitivity of four different methods for the detection of antibodies specific for HLA antigens. The NIH version of the complement-dependent cytotoxic test (CDC) was used as the gold standard to which we compared two Fcgamma receptor (FcgammaR)-dependent immune phagocytosis inhibition tests (IPI) and one commercial enzyme-labelled immunosorbent assay (ELISA) with soluble HLA class I-antigen preparations bound to the plate (PRA-STAT). Both IPI tests are based on the fact that HLA-antibodies specifically bind to antigens on the monocyte surface via their Fab portion, and in so doing block a neighbouring FcgammaR with their Fc region. This blockade prevents phagocytosis of IgG-coated red blood cells (RBCs), which can be measured either microscopically (IPIm) or photometrically (IPIp). The four assays were used in blind tests on 20 human alloantisera or monoclonal antibodies with known HLA-antigen reactivities. Additionally, two monoclonal antibodies and one human serum were titrated to elucidate the sensitivity of each test. After all tests were completed, the identities of the samples were disclosed. Both IPI methods detected and identified all clinically relevant HLA class I and class II specific antibodies. In contrast, the CDC was not able to detect noncytotoxic HLA-antibodies and HLA class II specific antibodies; however, it detected clinically insignificant IgM lymphocytotoxins. The PRA-STAT assay enabled identification of all cytotoxic and noncytotoxic IgG antibodies with specificity for HLA-class I antigens. With respect to sensitivity, the CDC and the IPI methods were superior to the PRA-STAT. These facts demonstrate the advantage of IPI methods in the detection of clinically relevant HLA-antibodies.

NA1/NA2 antisera inhibit FcgammaRI- but not FcgammaRII- mediated phagocytosis.

BACKGROUND AND OBJECTIVES: Alloantibodies against the granulocyte-specific NA antigens play an important role in alloimmune neonatal neutropenia. As the NA system is located on the FcgammaRIIIb, the influence of NA-specific antibodies on granulocyte function is of special interest. MATERIALS AND METHODS: We tested alloantisera specific for NA1 and NA2 for their ability to influence FcgammaR-mediated phagocytosis of polymorphonuclear neutrophils by use of different FcgammaR-specific targets. Red blood cells coated with human IgG anti-D served as specific targets for FcgammaRI-mediated phagocytosis while mouse IgG1 anti-glycophorin A was used to coat red blood cells (RBCs) to obtain FcgammaRII specific targets. To test for a hypothetical induction of phagocytosis by FcgammaRIIIb we used D-- RBCs coated with human monoclonal anti-D as target cells for unprimed neutrophils. RESULTS: Granulocyte phagocytosis was directly induced by FcgammaRI and FcgammaRII but not by FcgammaRIIIb. NA1 alloantisera significantly inhibited FcgammaRI-mediated phagocytosis of IFN-gamma-stimulated neutrophils if the corresponding antigen was expressed. Conversely, NA2 alloantisera inhibited FcgammaRI-mediated phagocytosis in NA2-positive individuals. There was no effect of NA1- and NA2-specific alloantibodies on FcgammaRII-mediated phagocytosis. CONCLUSION: NA-specific alloantisera inhibit the FcgammaRI-induced phagocytosis in primed neutrophils, but they do not significantly inhibit their FcgammaRIIa-specific phagocytosis of mIgG1-coated RBCs.

The NA"null" phenotype of a young man is caused by an Fc gammaRIIIB gene deficiency while the products of the neighboring Fc gammaRIIA and Fc gammaRIIIA genes are present.

A 27-year-old man with an allergy to house dust mites was found to lack the Fc gammaRIIIb on his neutrophils. Cell surface marker and PCR techniques were used to investigate possible reasons for this deficiency. Agglutination and immunofluorescence assays using the man's neutrophils together with NA1- and NA2-specific antibodies were negative, and there was no reaction with the Fc gammaRIII-specific mAb 3G8. Indirect immunofluorescence demonstrated the presence of the CD24 molecule, which, like the Fc gammaRIIIb, is anchored to the cell membrane by glycosylphosphatidylinositol. Thus a lack of the Fc gammaRIIIb cell membrane anchor was excluded. PCR analysis confirmed the absence of the NA1 and NA2 alleles. The individual was therefore typed as NA"null". The products of those genes located together with the Fc gammaRIIIB gene within a complex on chromosome 1 (q23-24) were examined. Fc gammaRII was demonstrated on monocytes and B cells with the use of Fc gammaRII-specific monoclonal antibodies. About 5% of the individual's peripheral blood monocytes were positive with the 3G8 antibody, indicating the presence of Fc gammaRIIIa. From these data we concluded that the Fc gammaRIIIb deficiency on the neutrophil cell surface of this individual is due to a lack of the Fc gammaRIIIB gene while excluding a lack of the Fc gammaRIIA and the Fc gammaRIIIA genes.

Rapid typing of the human Fc gamma receptor IIA polymorphism by polymerase chain reaction amplification with allele-specific primers.

BACKGROUND: The human Fc gamma receptor IIa (Fc gamma RIIa) is expressed in two polymorphic forms, Fc gamma RIIa-H131 and Fc gamma RIIa-R131, that differ by the replacement of histidine by arginine at position 131. This replacement is caused by a single-nucleotide exchange of A-->G. The resulting receptor forms differ in their binding to human IgG2 and mouse IgG1, which may lead to a different immunologic defense to bacterial polysaccharides and encapsulated bacteria. STUDY DESIGN AND METHODS: A rapid and easy polymerase chain reaction(PCR) method of genotyping the Fc gamma RIIa was developed. Allele-specific primers discriminate between the Fc gamma RIIa-H131 and the Fc gamma RIIa-R131 forms of the receptor. The results were compared with those obtained by another DNA-based genotyping method, in which PCR-amplified DNA was hybridized with allele-specific oligonucleotides, and with a functional phagocytosis assay using mouse IgG1-coated red cells as target antigens. RESULTS: The genotypes deduced from the PCR with allele-specific primers were in complete accordance with those obtained by the data from the hybridization of PCR-amplified DNA with allele-specific oligonucleotides. Furthermore, the Fc gamma RIIa genotypes of 28 individuals in all cases corresponded to the functional phenotypes. CONCLUSION: The use of PCR with allele-specific primers provides a rapid and easily performed method for the determination the Fc gamma RIIa polymorphism.

Inhibition of monocyte and polymorphonuclear granulocyte immune phagocytosis by monoclonal antibodies specific for Fc gamma RI, II and III.

We tested two Fc gamma receptor I (Fc gamma RI); six Fc gamma RII; and six Fc gamma RIII-specific monoclonal antibodies (mAb) for their capacity to inhibit monocyte and polymorphonuclear granulocyte (PMN) immune phagocytosis which is mediated by Fc gamma R. We used human red blood cells (rbc) coated with hIgG1 or mIgG1 as Fc gamma RI- and Fc gamma RII-specific target cells, respectively. The Fc gamma RI-specific mAbs 22.2 and 32.2 did not inhibit Fc gamma RI- or Fc gamma RII-specific monocyte immune phagocytosis. The Fc gamma RII-specific mAbs IV.3, CIKM5, FLI8.2, FLI 8.26, 2E1, and 41H16 inhibited Fc gamma RII-specific monocyte immune phagocytosis in all Fc gamma RIIa high-responder (HR) individuals but did not inhibit Fc gamma RI-specific phagocytosis. Using PMN, FL18.2 and 2E1 only partially inhibited phagocytosis in HR individuals, but the Fc gamma RIII-specific mAbs 3G8, DJ130c. MFM-154. B88-9 and MG38 completely inhibited Fc gamma RII-specific phagocytosis if the corresponding antigen was available on the cell surface. In these cases phagocytosis inhibition may be explained by cross-linking of Fc gamma RII and Fc gamma RIII via one antibody molecule, with the Fab portion binding to Fc gamma RIII and the Fc portion binding to Fc gamma RII.

Linkage disequilibria between Duffy blood groups, Fc gamma IIa and Fc gamma IIIb allotypes.

Fc gamma receptors (Fc gamma R) on white blood cells and the Duffy blood group antigens on red blood cells are coded for on the long arm of chromosome 1. They appear in different allotypic forms: the high responder/low responder (HR/LR) forms of Fc gamma RIIa, the alloantigens NA1 and NA2 of Fc gamma RIIIb and the Duffy blood group antigens Fya and Fyb. The aim of this study was to analyze possible linkage disequilibria between these allotypic immunomodulatory receptors, and thus provide evidence for the existence of a hypothetical gene complex. The Duffy phenotype was determined by the indirect antiglobulin test, NA1/NA2 phenotypes by the granulocyte agglutination test and the HR/LR polymorphism by an immunophagocytosis assay. Two haplotypes were found to be in linkage disequilibrium. For the haplotype NA2, Fyb we calculated a delta-value of -0.07 and for the haplotype NA1, HR we obtained a delta-value of -0.02.

'Classic' anti-neutrophil cytoplasmic autoantibodies (cANCA), 'Wegener's autoantigen' and their immunopathogenic role in Wegener's granulomatosis.

Wegener's autoantigen (WA), a 29 kD multifunctional protein, is the principal target antigen of autoantibodies associated with Wegener's granulomatosis (WG). WA was first identified as proteinase 3 (PR3), which is now known to be identical with myeloblastin and AGP7. Like other lysosomal proteins, WA/PR3 displays enzymatic activity, differentiation factor activity for myeloid precursor cells, and antimicrobial functions. Neutrophilic polymorphonuclear leukocytes (PMN) and a subpopulation of monocytes contain high levels of WA/PR3 in their myeloperoxidase-positive granules. The autoantibodies from WG sera produce a finely granular, centrally accentuated fluorescence pattern on PMN and monocytes and have been designated 'classic' pattern antineutrophil cytoplasmic autoantibodies (cANCA). However, PMN/monocyte activation (in vitro/ex vivo) is associated with the translocation of WA/PR3 on the cytoplasm membrane. WA/PR3 is accessible to the WG-associated autoantibody: cANCA stimulate cytokine-preactivated PMN to produce oxygen radicals and to degranulate. Furthermore, cANCA interfere with the biological functions of WA/PR3 (e.g. inhibition of elastinolytic activity). Hence, cANCA represents not only the best seromarker for WG so far available, but several lines of evidence indicate that the autoantibodies against WA/PR3 play a major role in the pathogenesis of this enigmatic disease.

Rat pancreatic islet pretreatment with anti-MHC class II monoclonal antibodies and culture: in vitro MLIC test response does not predict islet allograft survival.

Antigen presenting cells (APC) expressing MHC class II antigens have been attributed with stimulatory capacity for initiating islet allograft rejection (direct pathway). Therefore, we evaluated the effect of pretreating isolated islets with different monoclonal antibodies against MHC class II antigens and complement, with and without culture at 22 degrees C or 37 degrees C, on MHC class II antigen expression, on the allogeneic proliferative response in the mixed lymphocyte islet culture (MLIC) and on islet allograft survival in adult rats. Experiments were performed in two different strain combinations incompatible for MHC class II antigens and either incompatible or compatible for MHC class I antigens, in order to elucidate further the impact of class I antigens on islet allograft rejection. In terms of class II antigen suppression, pretreatment with anti-MHC class II antibodies together with complement and a 5-day (37 degrees C) culture period proved most effective. After this procedure 92.7% of the islets of LEW rats and 91.1% of the islets of LEW.1WR2 rats were negative for MHC class II antigens, as demonstrated by indirect immunofluorescence. Transfer of successfully pretreated islets to a MLIC in vitro test system provoked a significantly reduced allogeneic T-cell proliferative response in the case of additional MHC class I disparity (ratio 1.3 vs 4.7) and a response as low as that of a syngeneic setting when stimulator islets and allogeneic responder lymphocytes shared MHC class I antigens (ratio 1.0 vs 1.6).(ABSTRACT TRUNCATED AT 250 WORDS)

ANCA in systemic vasculitides, collagen vascular diseases, rheumatic disorders and inflammatory bowel diseases.

It was the aim of this study to reevaluate the diagnostic significance and clinical implication of ANCA after testing sera from 13,606 patients for the presence of ANCA. Our data confirm the high specificity (97%) and sensitivity (80%) of cANCA for Wegener's granulomatosis. pANCA were found in renal vasculitides (60%), collagen vascular diseases (SLE 20%, Sjögren's syndrome 26%, polymyositis 16%) and rheumatic disorders (Felty's syndrome 50%, rheumatoid arthritis 20%). A third fluorescence pattern in sera of patients with inflammatory bowel disease (ulcerative colitis 28/72, Crohn's disease 6/84), here called xANCA, was seen. Target antigens of granule proteins from PMN and monocytes (proteinase 3, myeloperoxidase, elastase, cathepsin G, lactoferrin, lysozyme) were identified.

Autoantibodies directed against lysozyme: a new target antigen for anti-neutrophil cytoplasmic antibodies (ANCA).

ANCA-positive sera from 1138 patients and ANCA-negative sera from 90 patients were screened for autoantibodies directed against lysozyme (LZ) by ELISA. Sera from 120 patients did react with LZ. 99 sera bound to LZ only, whereas 56 sera bound to further granule proteins, especially cathepsin G and lactoferrin. In the routine ANCA screening, most of the anti-LZ-positive sera showed a pANCA fluorescence. In total, 8% of 674 pANCA-positive sera did react with LZ. Clinically, anti-LZ antibodies were associated inflammatory rheumatologic, -renal and -bowel diseases.

Pretreatment of rat pancreatic islets with MHC I-A monoclonal antibody: in-vitro and in-vivo effects on islet antigenicity.

Pancreatic islet allografts are rejected very rapidly due to their high immunogenicity. Several attempts have been made to reduce the immunogenicity prior to transplantation. We treated islets with MHC Ia (class II) antigen monoclonal antibody and complement in order to lyse Ia antigen bearing cells within the islets. This treatment caused a distinct reduction of Ia antigen positive cells, which was demonstrated by indirect immunofluorescence tests. A total depletion of the Ia antigens, however, could not be achieved. As a second in-vitro test to evaluate the effect of the antibody treatment a mixed lymphocyte islet culture (MLIC) was performed. The partial Ia antigen reduction provoked a significantly reduced but not completely suppressed allogenic T-cell response. Antibody treatment did neither impair the morphological integrity nor the function of the islets as could be demonstrated by syngenic transplantation of islets pretreated by antibodies in the same way. After allotransplantation of antibody-pretreated islets across a major histocompatibility barrier into non immunosuppressed rats, however, there was no significant prolongation of the graft survival. Thus, a partial reduction of the number of the Ia antigen positive cells within the islets is not sufficient for the prevention of allograft rejection.

Islet transplantation in experimental diabetes of the rat. XIII. Cryopreservation reduces MHC class II but not class I antigens of rat pancreatic islets.

Pretreatment of islet allografts prior to transplantation may reduce islet immunogenicity and prolong graft acceptance. We have studied the MHC antigen reducing effect of cryopreservation onto rat pancreatic islets performing indirect immunofluorescence tests and peroxidase-anti-peroxidase staining (PAP). Three different freezing programs were used. Program A: 0.5 degrees C/min to -35 degrees C and 1 degree C/min from -35 to -100 degrees C. Program B: 2 degrees C/min to -35 degrees C and 6 degrees C/min from -35 to -100 degrees C. Program C: 0.25 degrees C/min to -40 degrees C. Cryopreservation clearly reduced the number of class II antigen positive cells per islet in all cases. Program A was most effective with 45.5% of class II antigen negative islets compared to 6.4% of class II antigen negative fresh islets as shown by indirect immunofluorescence. The class II antigen reducing effect of cryopreservation proved to be permanent and not only temporary. Reduced class II antigen expression of cryopreserved islets could not be reestablished by incubation of the islets with rat IFN. A combination of cryopreservation followed by a 10 day culture period proved to be most effective with 85.6% of class II antigen negative islets. In contrast, we could not show any effect of cryopreservation on class I antigen expression. Viability of the cryopreserved rat islets was shown in-vitro by glucose stimulated insulin secretion.