Antibody responses to xenogenic antigens--a study in the mouse-to-rat system.

Antibodies play a crucial role in the rejection of an organ that has been transplanted between different animal species, i.e. xenotransplantation. In previous work, we have induced a state of humoral tolerance where mouse-to-rat heart grafts continued to beat under ciclosporine A monotherapy. Initially, a combined treatment with ciclosporine A and 15-deoxyspergualin was given. This state of tolerance could not be reproduced when the vascularised heart graft was replaced with a free tissue graft or xenogeneic blood transfusions. To gain further insight into the humoral response against mouse antigens, we studied the antibody production in naive rats and rats challenged with heart transplants, heart cells, mononuclear cells (MNC) and erythrocytes from mice. Rats not challenged with any mouse cells or organs had a moderate amount of antibodies targeted against mouse MNC as well as rosette-forming cells in the spleen targeted against mouse erythrocytes. A challenge with either mouse MNC or erythrocytes lead to immunisation with antibodies of both IgM and IgG subtype directed against both MNC and erythrocytes. Antibody titres against mouse erythrocytes in animals challenged with MNC were not detectable until day 7, whereas antibody titres against mouse MNC in animals challenged with erythrocytes were detected on day 1. Immunisation with mouse erythrocytes raised the titre of rosette-forming cells in the spleen compared with naive rats (P < 0.05). Our data indicate that different xenogeneic antigens in the mouse-to-rat system are shared between heart cells, MNC and erythrocytes; however, the immunisation patterns differ regarding the time when antibodies are first detected.

Pretransplant xenogeneic blood transfusions reduce the humoral response in a mouse-to-rat heart transplantation model.

A mouse heart transplanted to a rat is rejected promptly 3 days after transplantation, independent of whether cyclosporin A (CyA) is used as an immunosuppressant or not. Adding a short course of deoxyspergualin (DSG) initially, in addition to continuous CyA treatment, results in long-term graft survival and permits retransplantation during CyA monotherapy. In this paper, we have explored the possibility of substituting the initial heart transplant with blood transfusions. Lymphocyte-enriched blood transfusions combined with CyA and an initial course of DSG proved to lower or eliminate the haemagglutinating antibody titre normally seen in acute vascular xenorejection. The therapy, however, did not prolong the mean survival of the cardiac xenograft, but the same treatment protocol could result in either hyperacute rejection or prolonged survival of up to 11 days. In conclusion, this and earlier studies propose that a humoral unresponsiveness can be induced if the recipient vascular circulation is exposed to a xenoantigen in a mouse-to-rat combination.

Potent antitumor effects of CD154 transduced tumor cells in experimental bladder cancer.

PURPOSE: Current intravesical immunotherapy for bladder cancer with bacillus Calmette-Guerin instillations is standard treatment for patients with high risk superficial tumors but relapses are common. We evaluated the tumor vaccine concept in murine bladder cancer by comparing tumor cell transduction with genes coding for the immunostimulatory molecules CD154, interleukin (IL)-12 and CD80 to design a novel vaccination strategy. MATERIALS AND METHODS: Adenoviral vectors were used to transduce murine bladder cancer MB-49 cells with genes coding for CD154, IL-12 and CD80. Parental or transduced MB-49 cells were injected subcutaneously into syngeneic mice. The effects of transgene expression on tumorigenicity and the generation of protective immunological memory against challenge with parental tumor were studied. RESULTS: All 76 animals injected with parental MB-49 cells had tumors within 8 to 12 days. Tumor cell expression of CD154 combined with IL-12 completely inhibited tumor outgrowth with all 21 mice tumor-free and CD154 transduction alone was almost as effective with 33 of 35 tumor-free. IL-12 production by tumor cells delayed tumor outgrowth and 4 of 10 mice remained tumor-free. Over expression of CD80 had no effect on tumorigenicity. CD154 expressing tumors were rapidly infiltrated with large numbers of CD4+ and CD8+ T cells. Mice vaccinated 4 times with adenoviral CD154 transduced MB-49 cells were completely protected against challenge with parental tumor. Co-injection of CD154 modified cells with parental MB-49 cells retarded tumor growth. CONCLUSIONS: Our experimental results suggest that the potent antitumor effects of CD154 gene transduction should be considered for immunostimulatory gene therapy for bladder cancer.

Expression of complement regulatory proteins on islets of Langerhans: a comparison between human islets and islets isolated from normal and hDAF transgenic pigs.

BACKGROUND: The expression of regulators of complement activity (RCAs) on islet cells may be of great importance for protecting them against complement-mediated lysis in the immediate posttransplant period after intraportal islet transplantation. We examined porcine and human islet cells for expression of RCA. We also examined to what extent human decay accelerating factor (hDAF) is expressed on adult and fetal islet cells isolated from hDAF transgenic (TG) pigs having a high transgene expression on endothelial cells. Moreover, the susceptibility of the various types of cells to lysis in human serum and blood was investigated. METHODS: Adult human islets (n=5), normal adult and fetal porcine islets (n=9 and n=8, respectively), and islets from adult and fetal hDAF TG pigs (n=5 and n=6, respectively) were examined. With islet single-cell suspensions and flow cytometry, adult human islet cells were examined for expression of hDAF (CD55), hCD59, and human membrane cofactor protein (hMCP; CD46), while porcine islet cells were examined for expression of pCD59 and pMCP. Islet cells from hDAF TG pigs were also examined for hDAF expression. Porcine peripheral blood lymphocytes, normal and hDAF TG porcine endothelial cell lines, a human endothelial cell line, and the human cell line U937 served as controls. Islet cytotoxicity was assayed after incubation of the islet cells with fresh human serum. Furthermore, adult islets from normal control pigs and hDAF TG pigs were exposed to fresh human blood in vitro for 60 min, and the inflammatory reaction elicited was compared between the different types of islets. RESULTS: All human islet cell preparations expressed hCD59, two of five expressed hMCP, but none expressed hDAF. Porcine islet cells expressed both pCD59 and pMCP. Normal adult porcine islet cells exposed to fresh human serum resulted in 74+/-5.4% cell lysis (mean+/-SEM, n=16). In comparison, only 1.3+/-2.8% (n=20, P<0.001) of human islet cells were lysed in the human serum. One islet cell preparation from an hDAF TG pig expressed small amounts of hDAF. This preparation from hDAF TG pigs bound significantly less C3c than did normal control islets (mean fluorescence ratio 16+/-2.2 and 58+/-4.3, respectively; P=0.046) and were partially protected from cell lysis in fresh human serum (47+/-10% and 78+/-18% cell lysis, respectively; P=0.046). The other four preparations from hDAF TG pigs were negative for hDAF and were equally susceptible to lysis as normal control islets. All fetal pancreatic islet cells from hDAF TG pigs analyzed were negative for hDAF expression. When exposed to fresh human blood in vitro, adult and fetal islets from hDAF TG pigs elicited equally strong inflammatory changes as did the normal control islets. The inflammatory changes were characterized by activation of the complement and coagulation systems, resulting in islet damage with "dumping" of insulin into the blood. CONCLUSIONS: Porcine and human islet cells express species-restricted complement regulatory proteins, with the human islet cells expressing mainly hCD59. A low expression of hDAF was detected on islet cells from one of five hDAF TG pigs. These islet cells displayed reduced islet cell cytotoxicity in fresh human serum. We conclude that protection from complement-mediated lysis will be important in the context of intraportal pig-to-human islet transplantation, and expression of a human RCA on islet cells should be beneficial in this context.

Concordant xenotransplantation--non-vascularized pancreatic islets are more difficult to regraft than the vascularized heart.

We have previously demonstrated that it is possible to perform retransplantation of a xenogeneic heart (mouse-to-rat) using cyclosporine A as monotherapy, provided that the first heart is transplanted under a short course of deoxyspergualin (DSG). If DSG is omitted, the first heart is rejected within four days and the second heart succumbs to hyperacute rejection within minutes. A mouse heart as first graft does not protect a consecutive pancreatic islet graft, although the heart continues to function after rejection of the cellular graft. One explanation for this discrepancy may be the fact that cellular grafts, as pancreatic islets, lack an endothelial lining. We have, therefore, further investigated possible differences between vascularized and non-vascularized xenografts regarding their capacity to induce unresponsiveness. The use of pancreatic islets as primary graft neither accelerated nor decelerated the speed of rejection of the vascularized heart used as secondary graft. Furthermore, hemagglutinating and cytotoxic antibody titres responded in the same manner as in naive rats transplanted with a mouse heart. Retransplantation with pancreatic islets also resulted in complete rejection of both the primary and secondary grafts. Thus, the lack of unresponsiveness cannot simply be explained by differences, between the pancreatic and cardiac tissues, in antigen expression. In addition, intraperitoneal transplantation of mouse heart cells as primary graft resulted in rejection of a secondary cardiac graft after three days. However, it cannot be totally excluded that the time of antigen exposure had an impact on these results. In conclusion, our previous and present studies suggest that the presence of an intact vascular bed, both in the first and second graft, is necessary to create a state of unresponsiveness. Because the pancreatic islets lack an endothelial lining, they do not benefit from an unresponsiveness of the immune system. Neither are they able to induce such an unresponsiveness.

A comparison of fetal and adult porcine islets with regard to Gal alpha (1,3)Gal expression and the role of human immunoglobulins and complement in islet cell cytotoxicity.

BACKGROUND: It is still debated whether fetal or adult porcine islets should be the preferred choice for future clinical islet xenotransplantation. Each type of islet preparation has advantages and disadvantages compared with the other. Here we present a direct comparison between fetal and adult porcine islets with regard to Gal alpha(1,3)Gal expression, immunoglobulin and complement binding, and cytotoxicity after exposure to fresh human serum. METHOD: Islet single cell suspensions were prepared from adult and fetal islets by trypsin digestion. Fluorescein isothiocyanate-conjugated Bandeiraea simplicifolia isolectin B4 (BS-IB4) and affinity-purified chicken anti-Gal alpha(1,3)Gal antibody was used to detect Gal alpha(1,3)Gal expression. Immunoglobulin and complement binding to the islet cells and cytotoxicity for islet cells was compared after incubation with fresh and heat-inactivated human sera and with an immune serum from a diabetic patient who received a fetal porcine islet transplant. Furthermore, two pools of human AB sera were depleted of porcine endothelial cell cytotoxic human anti-Gal alpha(1,3)Gal antibodies by absorption and were used to analyze the effect of Gal alpha(1,3)Gal antibody removal on islet cell cytotoxicity. RESULTS: Fetal islet cells readily bound both BS-IB4 and the chicken anti-Gal alpha(1,3)Gal antibody. None of 10 adult porcine islet preparations were stained by BS-IB4. In comparison, IgY anti-Gal Ab binding was detected in two of eight adult islet isolations, whereas the other six preparations showed marginal/no binding. After incubation of fetal islet cells with fresh human serum, C3c binding was strongly positive and IgM binding variable, with occasional binding of IgG and no detectable binding of IgA. Adult islet cells were also strongly positive for C3c but did not bind detectable amounts of IgM, IgG, or IgA. Immune sera from a patient who had received fetal porcine islets showed the presence of induced antibodies that bound to fetal islet cells and to porcine peripheral blood lymphocytes, whereas binding to adult islet cells was barely detectable. Fresh human sera showed a high and similar level of complement-mediated lytic activity for both adult islet cells (78+/-22%) and fetal islet cells (75+/-16%). Cytotoxicity for fetal islet cells and peripheral blood lymphocytes was significantly reduced when the corresponding sera were depleted of anti-Gal antibodies before use (P=0.002 and P=0.003, respectively). In contrast, no difference in cytotoxicity for adult islet cells was detected when anti-Gal-depleted human sera were used. CONCLUSION: Gal alpha(1,3)Gal expression is occasionally detectable on adult porcine islet cells, but not as readily and at a lower level, compared with fetal islet cells. Thus, as porcine fetal islets mature to adult islets, the expression of the Gal alpha(1,3)Gal epitope gradually diminishes. Consequently, cytotoxic anti-Gal alpha(1,3)Gal antibodies in human serum play an important role in the lysis of fetal but not adult porcine islet cells.

Comparative studies of antimitochondrial autoantibodies in sera and bile in primary biliary cirrhosis.

Primary biliary cirrhosis (PBC) is an autoimmune liver disease characterized by destruction of intrahepatic bile ducts. Although the pathogenesis of this disease is still unknown, high titers of antimitochondrial autoantibodies (AMA) have long been recognized in patient sera. However, little is known about the presence of AMA in bile. In this study, we investigated bile and sera from patients with PBC and healthy controls for the presence of AMA and mitochondrial autoantigens. AMA were detected in the bile of 17 of 19 patients (89.4%) with PBC; they were specifically directed against the pyruvate dehydrogenase complex (PDC-E2) in 15 of 19 patients (78.9%), to the branched-chain 2-oxo-acid dehydrogenase complex E2 (BCOADC-E2) in 6 of 19 patients (31.6%), and to the 2-oxoglutarate dehydrogenase complex E2 (OGDC-E2) in 1 of 19 patients (5.3%). In a comparative study of sera from the same patients, anti-PDC-E2 antibodies were found in 19 of 19 patients (100%), anti-BCOADC in 9 of 19 patients (47.3%), and anti-OGDC-E2 in 4 of 19 patients (21.1%) patients. AMA in bile were always found together with antibodies of corresponding specificities in the serum from the same patient. Immunoglobulin (Ig)A AMA were found in the bile of 9 of 19 patients (47.7%) with PBC; they were specifically directed against PDC-E2 in 8 of 19 patients (42.1%) and to BCOADC in 2 of 19 patients (10.5%). Epitope mapping of IgA anti-PDC-E2 antibodies indicated that, like serum autoantibodies, the immunodominant epitope is directed against the inner lipoyl domain of PDC-E2. The prevalence and antigen reactivity of IgA AMA in sera correlated completely with IgA AMA in bile. Autoantibodies against nuclear envelope pore proteins (gp210) were found in 1 of 8 (12.5%) sera of patients with PBC, but not in bile. Furthermore, and of particular interest, we detected the autoantigens, PDC-E2, OGDC-E2, and BCOADC-E2, in the bile of 12 of 19 patients (63.2%), 9 of 19 patients (47.4%), and 9 of 19 patients (47.4%), respectively; PDC-E2 was found in only 1 of 17 (5.9%) disease controls. Although the presence of AMA in bile may merely reflect the presence of these antibodies in sera, the simultaneous detection of mitochondrial autoantigens in bile suggests an increase of mitochondrial autoantigens at inflammatory sites. Such autoantigens, coupled with AMA, may augment the local immune response and disease progression.

Multicolor flow cytometric analysis of the CD45 antigen provides improved lymphoid cell discrimination in bone marrow and tissue biopsies.

Samples from bone marrow or non-hematopoietic tissue such as solid organ biopsies often contain an excess of non-leukocytes exhibiting lymphocyte-like light scatter characteristics, making it sometimes difficult to define satisfactory light scatter lymphocyte gates. To circumvent this, we describe here a multiparametric method of identifying lymphoid cells by expression of the CD45 antigen, in conjunction with light scatter parameters. A 'third color'-conjugated anti-CD45 antibody was included with every FITC/PE double staining, thereby permitting live or list mode analysis gating on CD45 positive cells. The triple-staining technique was applied to (a) human bone marrow, showing that special attention has to be given to the enumeration of B cells, and (b) to liver biopsies, where gating on CD45 fluorescence and orthogonal light scatter was shown to clearly resolve all lymphocyte subsets from debris. All cell types examined in tissue biopsies as well as T and NK cells in bone marrow were best distinguished by gating on bright CD45 expression in conjunction with low orthogonal light scatter, while accurate identification of marrow B cells relied upon including all levels of CD45 intensity. The multicolor gating procedure, aimed mainly at immune-monitoring of non-malignant tissues, is applicable to most kinds of single cell samples, and may prove to be an aid for lymphocyte gating in cases where leukocyte populations are not clearly resolved on a light scatter basis alone.

Primary biliary cirrhosis. High proportions of B cells in blood and liver tissue produce anti-mitochondrial antibodies of several Ig classes.

Anti-mitochondrial Abs (AMA) of the M2 type are recognized as being specific for primary biliary cirrhosis (PBC). We developed a highly specific assay to detect single AMA-producing cells using pyruvate dehydrogenase (PDH), the major Ag for AMA. Total Ab and AMA production of in vivo activated B cells was measured in peripheral blood from 13 PBC patients, 22 patients with other liver diseases, and 15 healthy controls. Anti-PDH-producing cells, PDH spot-forming cells (PDH-SFC), were detected in 12 of 13 PBC patients and represented a very high proportion of circulating Ig-producing lymphocytes (9 +/- 8.6%; mean +/- SD). The autoantibodies were mainly of IgG and IgM Ig classes. The number of PDH-SFC was positively correlated with the numbers of total Ig-SFC within each Ig class. An increased number of total IgM SFC in blood was found in PBC patients compared with controls, and IgM SFC correlated with the elevated serum levels of IgM typical for this disease. Among the lymphocytes extracted from PBC liver tissue (n = 5), we detected PDH-SFC of IgG (range 0 to 33% of total SFC), IgM (0 to 42%), and IgA types (1 to 30%). IgA PDH-SFC were found in five of five livers, but only in two of thirteen blood samples investigated. Taken together, our data reveal that very high numbers of B cells spontaneously produce disease-specific autoantibodies in blood and liver tissue in PBC. Furthermore, the Ig class distribution of produced autoantibody differs between these two compartments, and this may have pathogenetic significance.

Is primary biliary cirrhosis an autoimmune disease?

This review summarizes the experimental and clinical support for an autoimmune origin of primary biliary cirrhosis (PBC). Direct proof is lacking, but indications in favour of an immunologic destructive mechanism include the demonstration of antibodies and T cell clones with specificity for mitochondrial autoantigens, and the lymphocytic infiltration/destruction of small bile ducts similar to that of graft-vs-host disease and rejection. There is a weak association with other autoimmune diseases, but no clear HLA linkage. Spontaneous animal models for PBC are lacking, and immunization of animals with purified autoantigen does not result in typical disease. Anti-mitochondrial antibodies (AMAs) of M2 type are diagnostic of PBC, and are mainly directed against a functional, restricted epitope on the E2 subunit of the pyruvate dehydrogenase complex (PDC). PDC-E2 shows several similarities to other classical autoantigens. The pathogenic role of AMA remains elusive. Recent studies have shown that AMAs detect an antigenic epitope expressed on the luminal surface of biliary epithelium in PBC liver. The initial triggering event might represent a microbial infection (molecular mimicry), or an aberrant surface expression of a true autoepitope.

Blood and liver-infiltrating lymphocytes in primary biliary cirrhosis: increase in activated T and natural killer cells and recruitment of primed memory T cells.

We used two-color and three-color flow cytometric analysis to study phenotypical activation and functional subsets of T and natural killer cells in the blood and liver tissue of patients with primary biliary cirrhosis, other chronic liver diseases and the blood of healthy subjects. The changes in blood lymphocyte phenotype in patients with primary biliary cirrhosis and other chronic liver diseases were similar and comprised elevated relative or absolute numbers of activated human leukocyte antigen-DR + T subset (CD4+ and CD8+) cells and DR+ natural killer-like (CD16+) cells. B cell (CD19+) numbers were normal. In primary biliary cirrhosis a selective reduction in T cells of suppressor-inducer (CD45RA + CD4 + ) type was registered. The human leukocyte antigen-DR expression among CD4+ T cell subsets was investigated further in primary biliary cirrhosis and healthy controls using triple antibody flow cytometric analysis. Phenotypical cell activation was confined to helper T cells of the primed, memory (CD45RO + CD4+) type. The decrease in suppressor-inducer T cells in primary biliary cirrhosis was paralleled by a reciprocal increase in primed memory T cells. Several significant differences were observed when blood and liver-infiltrating cells from primary biliary cirrhosis patients were compared. In the liver tissue, the CD4/CD8 ratio was decreased, the relative activation of T-subset cells and NK cells was further increased, the suppressor-inducer T subset was further depressed and the primed memory T subset was increased. The cytotoxic T-cell subset (CD11b-) dominated within the CD8+ population. In liver tissue from other chronic liver disease subjects, a lower CD4/CD8 ratio was found compared with primary biliary cirrhosis.(ABSTRACT TRUNCATED AT 250 WORDS)

Primary biliary cirrhosis (PBC): characterization of a monoclonal antibody (PBC-MoAb) having specificity identical with disease-associated autoantibodies.

We have raised a monoclonal antibody (PBC-MoAb) directed against mitochondria which resembles patent anti-mitochondrial autoantibodies (AMA) (M2 type) in several respects. The reaction pattern of PBC-MoAb was characterized by western blot experiments, immunoaffinity purification and enzyme inhibition studies. PBC-MoAb reacts specifically with an epitope on the E2 subunit of pyruvate dehydrogenase (dihydrolipoamide acyltransferase) which is essential for enzymatic activity. This was shown as follows: (1) PBC-MoAb, like PBC-AMA, completely inhibited PDH enzyme activity and reacted weakly with OGDH; (2) PBC-MoAb bound strongly to the E2 subunit in western blots, with weaker binding to a doublet of about 56 kDa; and (3) in immunosorbent experiments, PBC-MoAb absorbed most (greater than 95%) of the AMA reactive material found in solubilized mitochondria. The present data together with earlier findings that the majority of PBC patient autoantibodies bind to epitopes defined by the PBC-MoAb, makes this antibody a valuable tool for characterizing the major PBC-associated epitope on PDH-E2 and localizing this epitope in liver tissue.

Single laser flow cytometric detection of lymphocytes binding three antibodies labelled with fluorescein, phycoerythrin and a novel tandem fluorochrome conjugate.

In order to implement three-color surface immunofluorescence on a single laser flow cytometer, we combined a new fluorescent tandem conjugate (TC, R-phycoerythrin plus Texas red, available as DuoCHROME) with fluorescein (FITC)- and phycoerythrin (PE)-labelled monoclonal antibodies for the simultaneous detection of three antigens on individual lymphoid cells. Considerable amounts of fluorescence leaked from the PE component of the TC complex into the PE detector, which necessitated compensation of the PE signal. Single and double stainings of peripheral lymphocytes with FITC-, PE- and/or TC-labelled antibodies revealed that the individual labels all identified populations of similar size, and that subsets of T cells could be properly and accurately detected with TC as one label. Triple staining was used to study phenotypically activated HLA-DR+ cells within functionally important subsets of CD4+ T cells, i.e., CD45R+ CD4+ (naive, 'suppressor inducer' T) and UCHL1+ CD4+ (memory, 'helper' T) cells. In the blood of healthy subjects as well as patients with primary biliary cirrhosis, CD4+ UCHL1+ (i.e., memory) T cells were significantly more activated than CD4+ UCHL1- cells. The reciprocal CD45R+ CD4+ (naive) T subset expressed HLA-DR to a much lesser extent. In patient samples, this difference in CD4+ T subset activation was more pronounced than among normal lymphocytes. The described triple immunofluorescence staining technique facilitates true multicolor analysis of individual cells on a single laser flow cytometer. Such multiparameter investigations may prove to be important for the further understanding of the phenotypic and functional diversity of immunocompetent cells.

Mitochondrial autoantigens in primary biliary cirrhosis. Association of disease-specific determinants with a subunit of complex I (NADH-ubiquinone reductase) of the inner mitochondrial membrane.

Anti-mitochondrial autoantibodies (AMA) from patients with primary biliary cirrhosis (PBC) were analysed for fine specificity by immunoblotting and enzyme-linked immunosorbent assay (ELISA). Inhibition ELISA showed that complex I (NADH-ubiquinone reductase) from beef heart mitochondria completely inhibited the binding of AMA to mitochondrial inner membranes (SMP), indicating that the major mitochondrial antigens are located in complex I. Immunoblot analysis of beef heart SMP, complex I and the iron sulphur (IP) subfraction of complex I revealed several antigens, one of which (75 kDa) reacted with all PBC sera but not with the additional autoimmune sera tested. Resolution of SMP or complex I by two-dimensional electrophoresis yielded in both preparations a polypeptide of 75 kDa with an isoelectric point of 6.4, which reacted with PBC serum and with rabbit antisera raised against the 75,000 subunit of complex I. In immunoblot experiments, the antigenicity of the 75,000 polypeptide in SMP, complex I, and the IP subfraction is increased by prior reduction of the sample with mercaptoethanol. This suggests a similarity to the PBC-specific 'M-2' antigen, which is also sensitive to sulphur reagents. The data indicate that the 75 kDa polypeptide of complex I is a major mitochondrial antigen binding AMA in PBC sera, and allows us to identify the location and probable function of the PBC antigen.

Primary biliary cirrhosis: indication for a single mitochondrial antigenic epitope detected by patient autoantibodies and a novel monoclonal antibody.

A monoclonal antibody specific for the major primary biliary cirrhosis (PBC)-associated mitochondrial antigen (subunit I of NADH-ubiquinone reductase) was produced and used to study the binding sites recognized by anti-mitochondrial autoantibodies (AMA) in PBC sera. Immunization of mice with purified beef heart mitochondrial inner membranes resulted in one monoclonal antibody which reacted with mitochondrial proteins. This antibody (PBC-MoAb), which was of the IgG2b subclass with kappa light chains, exhibited a pattern of immunofluorescence reactivity with rat kidney, human thyroid, and cultured human epithelial cells (Hep-2) similar to that obtained with sera from PBC patients. Similar binding patterns between PBC-MoAb and AMA were also found in western blot analysis using mitochondria as antigen. Both types of antibodies revealed a major antigen of 75 kDa, a minor antigen of 60 kDa, and a third antigen (70 kDa), which was detected only in samples that had not been boiled prior to electrophoresis. Furthermore, optimal binding of the PBC-MoAb and AMA to the 75 and 70 kDa antigens required reduction of the antigen with mercaptoethanol prior to electrophoresis. Competition ELISA experiments were conducted to compare the epitopes recognized by PBC-MoAb and AMA. Of 28 PBC sera tested, 27 inhibited the binding of PBC-MoAb to mitochondrial inner membranes by almost 100% and one serum inhibited binding by 50%, indicating that most PBC sera contain autoantibodies reactive with the same or a closely related antibody binding site as the PBC-MoAb. PBC-MoAb inhibited AMA binding to the inner membrane by more than 80% in 10 sera, 60-80% in 11 sera, and 40-59% in seven sera, with an average inhibition of 71%. Our observations strongly indicate that anti-mitochondrial autoantibody binding sites are restricted to a highly immunogenic epitope on the major PBC-specific antigen (NADH-ubiquinone reductase subunit I), and that the anti-mitochondrial monoclonal antibody obtained has a specificity identical with the human PBC-specific M2 type anti-mitochondrial autoantibody.

Evidence that the major primary biliary cirrhosis-specific mitochondrial autoantigen is a subunit of complex I of the respiratory chain.

Primary biliary cirrhosis (PBC)-specific antigens were purified from beef heart mitochondria by immunoaffinity chromatography. Three major polypeptides (75, 60, and 40 kDa) were detected in the purified antigen fraction both by Coomassie blue staining and by western blot analysis. The 75 kDa antigen was identified as a subunit of Complex I (NADH-ubiquinone reductase) by the following criteria: (1) antibodies against the purified 75 kDa subunit of beef heart Complex I react with the immunoaffinity-purified 75 kDa antigen. (2) the 75 kDa subunit present in isolated Complex I, like that in the immunoaffinity-purified antigen, reacts with PBC sera only after reduction with mercaptoethanol, and (3) the 75 kDa antigen is enriched in isolated Complex I. A relationship between the 75 kDa and the 60 and 40 kDa antigens is suggested, since optimal binding of anti-mitochondrial autoantibodies (AMA) to the latter antigens also requires prior reduction with mercaptoethanol. A fourth major antigen (70 kDa) was also detected by western blot analysis, but only in samples that had not been boiled prior to electrophoresis. This antigen, which is also present in isolated Complex I, resembles the 75, 60, and 40 kDa antigens in its response to mercaptoethanol and its reaction with antibodies against the 75 kDa subunit of Complex I. A scheme is presented which relates all of the PBC antigens to the parent 75 kDa subunit of Complex I, probably as proteolytic products of the latter.