Erna Möller

Contact-Induced Cytotoxicity by Lymphoid Cells Containing Foreign Isoantigens

In tissue culture, immune lymph node cells containing foreign histocompatibility antigens of the H-2 type exert marked cytotoxic effects on tumor cells incompatible with the H-2 antigen. An equally pronounced effect is obtained when normal allogeneic and semi-isologous lymphoid cells of F1 hybrids are caused to aggregate around the target tumor cells by treating the cultures with either heat-inactivated rabbit antiserum to mouse cells or phytohemagglutinin. Isologous lymph node cells have no effect. Thus, aggregation of lymphoid cells and target cells is a necessary but insufficient requirement for cytotoxicity in vitro; in addition, close contact must be established between histoincompatible cells.

Hyperacute rejections of two consecutive renal allografts and early loss of the third transplant caused by non-HLA antibodies specific for endothelial cells

The immunological rejection of HLA-identical kidney transplants indicates that non-HLA antigens may also be targets for transplant rejection. Interest in the possible role of endothelial specific antigens has grown steadily over the years. Most of the studies published, regarding the association of such antibodies with rejection, have demonstrated the reactivity of endothelial antibodies also with monocytes and keratinocytes, but not with lymphocytes. Such antibodies escape detection in conventional crossmatch tests. In this paper, we present a case report of a 10-year-old girl, whose two consecutive kidney allografts, (one living and one cadaveric donor) were hyperacutely rejected in spite of the fact that she had neither been alloimmunized, nor had any HLA-specific antibodies. Endothelial cell specific antibodies were detected in vivo and in vitro after transplantation only 11 days apart, which were considered to be responsible for rejection. The third cadaveric kidney was lost within 1 week post-transplant. Immunopathological investigation of the three rejected grafts revealed deposition of IgM in the endothelium of arteries and in some glomeruli. No deposition of IgG antibodies was found. Antibodies from this patient did not react with lymphocytes, monocytes or keratinocytes. Patient serum had IgM antibodies that were specifically reactive with cultured endothelial cells, demonstrated by binding in vitro and by complement-dependent cytotoxicity of IL-beta stimulated endothelial cells. No HLA antibodies were found following the first two transplantations, but were demonstrated 1 week after the third transplantation, at the time of an acute irreversible rejection. Western blots of proteins solubilized from endothelial cell membranes, indicated that the antibodies reacted with a 97-110 kD protein. Endothelial cell antigen preparations were made from several different umbilical cord veins. Some primary cell cultures, but not all, reacted with the patients serum. Therefore, we suggest that the target determinant might be polymorphic. These findings imply that the non-HLA endothelial cell specific molecules may function as target(s) for hyperacute antibody-mediated destruction of kidney allografts.

HLA-DR-DQ haplotypes defined by restriction fragment analysis: Correlation to serology

Through the analysis of RFLP (restriction fragment length polymorphism) of the HLA-DR beta, -DQ alpha, and -DQ beta genes from 70 serologically well-characterized individuals, we have established unique HLA-DR-DQ RFLP haplotypes correlating to all of the DR1-w14 specificities. The RFLP of DR beta, DQ alpha, and DQ beta genes is very high using the restriction enzyme TaqI and 21 DR-DQ RFLP haplotypes were defined with this restriction enzyme. Our analysis confirms the strong linkage disequilibrium between alleles in the DR and DQ loci. DR beta RFLP indicates a common ancestor for the DR alleles within either of the supertypic DRw52 and DRw53 specificities. The DQ beta gene shows a high degree of RFLP, and the RFLP alleles partly reflect the serologic DQw1-w3 specificities. The results presented here also demonstrate the heterogeneity of DRw6 (DRw13 and DRw14) associated haplotypes, and the DRw13 related Dw18 and Dw19 specificities were found to have distinct DR-DQ haplotypes. The DQw1 positive haplotypes DR1, 2, w10, w13, and w14 are related with regard to DQ alpha and DQ beta RFLPs and the DRw52 positive haplotypes DR3, w11, and w12, as well as the DRw53 positive haplotypes DR4, 7, and w9, are related with regard to DR beta and DQ alpha RFLPs. These findings indicate that polymorphic sequences around the DQ alpha gene are associated with DR beta and DQ beta polymorphism, which suggests a location of the DQ alpha gene between DR beta and DQ beta.

Studies on antigen-binding cells: I. The origin of reactive cells☆

Abstract The origin of rosette-forming cells of mice immunized with sheep red blood cells has been investigated. By the use of antigenic markers on thymus cells (H-2, θ) it was shown that a considerable proportion of RFC formed in the spleens in irradiated animals reconstituted with thymus and bonemarrow cells, as well as in intact normal and immunized animals were of thymus origin. The possible limitations of the test systems and the significance of the results have been discussed.

CYTOMEGALOVIRUS-INDUCED CD13-SPECIFIC AUTOIMMUNITY-A POSSIBLE CAUSE OF CHRONIC GRAFT-VERSUS-HOST DISEASE1

Cytomegalovirus (CMV) infection has been suggested to be associated with certain autoimmune phenomena as well as with the development of chronic graft-versus-host disease (cGVHD) following allogeneic bone marrow transplantation. Earlier we found that the CMV-associated host protein CD13 is immunogenic during CMV infection in allogeneic bone marrow transplant patients, resulting in production of CD13-specific antibodies (7). Here we found a close correlation between CD13-specific immunity and later development of cGVHD in 26 of 33 patients who could be evaluated for this disease. Of seven patients with CMV disease, six developed extensive cGVHD, all of whom had CD13 specific antibodies (P=0.0002). All 14 patients who were CD13-immune later developed either limited or extensive cGVHD (P=0.0008). Antibodies in sera from the CD13-immune patients suffering from cGVHD recognized normal structures in cryosectioned skin biopsies from control individuals, producing a staining pattern similar to that of CD13-specific monoclonal antibodies. The antibody binding could be specifically blocked by preincubation of the skin sections with a mixture of monoclonal antibodies against CD13, and was also abolished after preabsorption of sera to mouse cells expressing human CD13. No other common autoantibodies were identified in more than single patients. Furthermore, in vivo binding of IgM antibodies in a CD13-like fashion was preferentially demonstrated in skin and oral mucosa biopsies from the CD13-immune patients suffering from cGVHD. Thus, we suggest that CMV-induced CD13-specific autoimmunity contribute to tissue damage in chronic graft-versus-host reactions.

Cytokines in rheumatoid arthritis.

Joints with rheumatoid arthritis are a site for chronic inflammation involving T cells, B cells, macrophages and dendritic cells. When these cells interact cytokines are likely to be produced. The presence of different cytokines in the synovial fluid of patients with rheumatoid arthritis has been studied and the macrophage derived cytokines such as IL-1, IL-6, TNF-alpha, TGF-beta and PDGF have usually been detected in large quantities, whereas T cell produced cytokines (IL-2, IL-4, IFN-gamma) are absent or present in small quantities. IL-1, IL-6 and TNF-alpha have several functions which suggest that they participate in the chronic disease process of rheumatoid arthritis, such as increasing production of eicosanoid, collagenase and prostaglandin E2. Many synovial B cells are activated and produce large amounts of immunoglobulins. We searched for a B cell stimulatory activity in rheumatoid synovial fluid and found a B cell differentiation and helper activity. Cytokines in the joints of patients with rheumatoid arthritis seem central for the propagation of the disease process. Specific intervention in cytokine production or in its effects might help to relieve symptoms in rheumatoid patients.

Xenograft rejection of porcine islet-like cell clusters in normal and natural killer cell-depleted mice

Fetal porcine islet-like cell clusters (ICC) were transplanted under the renal capsule of normoglycemic normal or athymic (nu/nu) C57BL/6 mice. Control animals were implanted with allogeneic minced kidney tissue from C57BL/Ks mice. The animals were killed 6 or 14 days after transplantation and the grafts were processed for flow cytometric analyses or immunohistochemistry. Xenograft destruction was evident in normal mice on day 6 after transplantation. The majority of infiltrating cells were macrophage-like cells expressing the F4/80 antigen. Lymphocytes expressing the CD3 antigen were in minority and mainly located in the peripheral parts of the ICC xenograft. The frequency and distribution of CD4+ cells were found to resemble those of the CD3+ cells. A large number of infiltrating cells, including several macrophage-like cells, expressed the Thy 1.2 antigen. Flow cytometry of infiltrating cells in the ICC xenograft revealed that approximately half of the cells expressing the F4/80 antigen also expressed Thy 1.2 and/or CD4. No cells were found expressing both the F4/80 and CD8 antigens. Both the F4/80 single-positive and the F4/80, CD4 double-positive cells were found to be larger and more granular than the CD4 single-positive cells. No co-expression of CD4 or Thy 1.2 with the F4/80 antigen was detected on cells infiltrating allogeneic tissue grafts. Moreover, a relative large number of cells (approximately 15%) in the xenograft expressed the NK 1.1 antigen as determined by flow cytometry. The role of natural killer (NK) cells in islet xenograft rejection was further evaluated in mice depleted of NK cells, using intraperitoneal injections of the monoclonal antibody NK 1.1. The simultaneous inoculation and subsequent growth of the NK cell-sensitive beta 2-microglobulin-deficient mutant, C4.4-25-, lymphoma cell line EL-4 served as an in vivo control of NK cell depletion. However, all NK cell-depleted mice rejected the ICC xenograft. In contrast, athymic mice permanently accepted the porcine ICC xenograft but, readily rejected the NK cell-sensitive lymphoma cell line. Taken together, ICC xenograft rejection in mice seems to be T cell dependent, as evidenced in the nude mice model, while the main effector cell appears to be a macrophage with a unique phenotype.

The occurrence of cytotoxic and non-complement-fixing antibodies in the crossmatch serum of patients with early acute rejection episodes

In an earlier study we found a strong correlation between the presence of donor-reactive HLA-specific antibodies in the crossmatch serum and early acute rejection episodes. Our experience was also that some of these antibodies were not cytotoxic and could therefore not be detected using the microcytotoxicity assays. In the present study, 11 patients from the earlier study who had weakly positive B cell reactive cytotoxic antibodies of the IgG class were further characterized. In addition, 14 new patients were selected who experienced early acute rejections but had a completely negative donor-B cell cytotoxicity crossmatch. A group of 12 controls without immunological complications was added, as well as 5 patients with early graft losses due to nonimmunological causes. Using the flow cytometric crossmatch test we confirmed the presence of HLA-specific antibodies in all 11 patients from the earlier study. In addition, positive flow cytometric crossmatches shown to be caused by HLA antibodies were observed in 11 of the 14 patients with acute rejections and negative cytotoxic crossmatch. One of 17 control patients had antibodies that were not HLA-reactive. IgA antibodies as well as IgG subclass determinations were performed in all positive sera. A substantial proportion of patients had HLA-specific antibodies of non-complement-binding classes (IgG2, IgG4, IgA) often of higher titers than IgGl and IgG3. The subclass distribution pattern was heterogeneous and often included several subclasses. We conclude that non—complement-fixing antibodies can also contribute to the risk for development of early acute rejections after necrokidney transplantation. Immunological mechanisms for these findings are discussed.

The Role of Adherent Cells in B and T Lymphocyte Activation

Macrophages have been found to be of importance in several difierent immunological reactions. Thus, macrophages are required for induction of a primary immune response to heterologous red blood cells in vitro (Mosier & Coppleson 1968), and for a proliferative response induced in vitro to alloantigens (Alter & Bach 1970, Rode & Gordon 1970, Twormey et al. 1970), as well as for the developement of cytotoxic cells (Wagner et al. 1972, McDonald et al. 1973). Fxirthermore, macrophage and factors released by macrophages can induce proliferation in spleen cells and cause production of polyclonal immunoglobuiins (Lemke & Opitz 1976, Moller et al. 1976, Opitz et al. 1976). In all these systems, however, fetal calf serum together with 2-mercaptoethanol (2-ME) can substitute for macrophages, indicating a nonspecific function of adherent cells (Chen & Hirsh 1972, Bevan et al. 1974, Lemke & Opitz 1976). In this paper, we report on some effects of adherent cells on B and T cell activation induced by polyclonal cell activators, and we discuss the possible mechanisms behind the synergistic effect observed between B and T cell mitogens and adherent cells. In the experiments shown, Fi hybrids of the inbred mouse strains A/Sn and C57B1 or B10.5M were used, if not otherwise indicated. Cells were cultured senimfree in microcultures (usually triplicate cultures with 5 X 10* cells/0.2 ml of medium) and incorporation of ^H-thymidine was measured on day 2.

Limited specificity of xenoantibodies in diabetic patients transplanted with fetal porcine islet cell clusters. Main antibody reactivity against α‐linked galactose‐containing epitopes

Abstract: The immunological specificity of antibodies formed as a result of xenotransplantation of fetal porcine islet‐like cell clusters to diabetic patients was characterized. High titer increases were recorded against porcine cells, solubilized membrane fractions of porcine cells, and purified MHC class I antigens. However, titer increases were also noted against ssDNA and dsDNA and against pig thyroglobulin but not against actin, myoglobin, or haptenated BSA. Antibody titers against tetanus toxoid were unaffected. The reactivity against porcine RBC could be completely blocked by absorption with pig thyroglobulin. Since pig thyroglobulin contains the galαl‐3gal antigen, the reactivity against RBC was most probably mainly due to antibodies against this oligosaccharide epitope. The reactivity against porcine mononuclear cells was only partially absorbed by pig thyroglobulin, indicating a heterogeneity of the clonal response. These conclusions were substantiated by data showing that the antigenic determinants on pig thyroglobulin were completely destroyed by treatment with α‐ but not with β‐galactosidase. Further studies showed that immune reactivity against pig RBCs, platelets, islet‐cells, endothelial cells and pig MHC class I molecules, caused by xenoimmunization, was almost completely blocked by either absorption of antibodies on a column of Sepharose beads coated with galα1‐3gal or by pretreatment of the antigen fractions with α‐galactosidase. Western blot experiments revealed that both the natural and xenoimmune antibodies reacted with a large number of different glycoproteins. There was no difference in heterogeneity of the response when comparisons were made between pre‐and posttransplantation sera, nor was there any difference of patterns caused by IgM or IgG antibodies. Absorption studies revealed that this epitope was present in a large number of different glycoproteins, a conclusion verified by staining with the eluate from a thyroglobulin‐immunosorbent column or with the eluate from the galα1‐3gal‐coupled Sepharose column. Our findings demonstrate that the xenoimmune response is mainly directed against oligosaccharide residues and that there is a limited clonal heterogeneity of antibodies in xenografted patients. There was no direct evidence of reactivity against any porcine proteins nor of specific immunization against porcine MHC peptides. The data form a basis for a future understanding of means to cope with and prevent the rejection of xenogeneic islet cells in man. We also suggest, based on the specificity found in xenoimmunized patients sera, that a major part of naturally occurring antibodies might be specifically reactive with carbohydrate antigens.