Molecular basis of antimitochondrial antibody production: specific cleavage patterns of PDC-E2 by apoptosis related proteases

HEPATOLOGYVoI. 34, No. 4, Pt. 2, 2 0 0 1

AASLD ABSTRACTS

367A

779

780

MOLECULAR BASIS OF ANTIM1TOCHONDRIAL ANTIBODY PRODUCTION: SPECIFIC CLEAVAGE PATTERNS OF PDC-E2 BY APOPTOSIS RELATED PROTEASES. Shuji Matsumura, University of California at Davis, Davis, CA; J u d y Van de Water, Hiroto Kita, University of California, Davis, CA; Ross L Coppel, M o n a s h University, Melbourne Australia; Takao Tsuji, Kazuhide Yamamoto, O k a y a m a University Medical School, O k a y a m a Japan; Aftab A Ansari, E m o r y University School of Medicine, Atlanta, GA; M E Gershwin, University of California, Davis, CA

UTILITY OF HEPATIC 31 P MRS IN ASSESSING THE SERVERITY OF LIVER DISEASE IN A RAT MODEL OF BILIARY CIRRHOSIS. Ian Corbin, J i m Peeling, M a n n a Zhang, Gerald Y Minuk, University of Manitoba, Winnipeg, MB Canada

One of the clear and unique findings in the bfliary epithelium of patients with primary biliary cirrhosis (PBC) is an apical accumulation of immunoreactive material that reacts with a number of mAbs and affinity purified antisera against the major mitochondrial autoantigens, including PDC-E2, OGDC-E2 and BCOADC-E2. This finding occurs in target tissue of the disease and provides a possible reason why immune damage is focused in these cells in PBC. Moreover, uhrastructural characterization of the intrahepatic small bile ducts reveals reactivity in the microvilli and the sublnminal ectoplasm and secretory material in the biliary lumen in PBC. Studies of both synthesis of the mitoehondrial antigens as well as a mean reactivity in bile, supports the proposition that the apically located material is derived from bile duct ceils, but there is no precedent in the literature for over-production of subunits of mitochondria under either normaI or pathologic circumstance. We hypothesize that the apically staining material is derived from host proteins and therefore involves proteolytic mechanisms. In fact, there is evidence for apoptosis occurring in bile duct cells in PBC. Several proteases may be involved in this turnover and catabolism, par ticularly the caspases and granzyme B. The role of caspases are to disable critical homeostatic and repair processes, as well as to cleave key structural components resulting in the systematic disassembling of the apoptotic cell. We therefore defined the specific cleavage pattern of the E2 component of the pyruvate dehydrogenase complex (PDC-E2), the immunodominant mitochondrial autoantigen in PBC, to assess the role of apoptosis as a continuous source of autoantigen. Recombinant full length PDC-E2 with His-Tag at its C-terminus site was digested with caspase-3, -6, -8 or granzyme B under optimal conditions. Immunoblotting of the digested samples was performed with our extensive panel of monoclonal anti-PDC-E2 antibodies as well as sera from patienls with PBC, in addition to control reagents. Our panel of mAbs are particularly useful as they react with multiple sites on PDC-E2 and allow us to define digested fragments. Interestingly, following granzyme B digestion, PDC-E2 lost reactivity, suggesting that the immunodominant epitope at the inner lipoyl domain was destroyed. Since this site is the major epitope for both B cells and T cells, such degradation has clinical significance. In contrast, caspases were less efficient in digesting PDC-E2 and several resulting fragments retained immunoreactivity. Further, we confirmed the presence of activated caspase-3 in marginal hepatocytes and bile ducts of the liver of patients with PBC. These results suggest that following apoptosis, caspases will produce immunogenic fragments which will contribute to antigenicity and the production of anti-mitochondrial antibodies. Such reactivity will be consistent with the generation of an autointmune response against an intracellular antigen that evades catabolism during apoptosis.

Liver biopsies remain the gold standard whereby assessments of the extent of liver disease are obtained. However, in clinical settings s u c h as extrahepatic obstructive jaundice, liver biopsies are contraindicated. In vivo phosphorns-31 magnetic resonance spectroscopy (31P MRS) is a non-invasive method whereby energy levels of tissue m a y be d o c u m e n t e d in situ. The aim of the present study was to determine in a rat model of chronic liver disease ( c o m m o n bile duct ligation) whether hepatic 31 P MR spectra reflect the biochemical and histologic extent of liver disease. Methods. After a n over night fast, baseline MRS examinations were performed on adult male Sprague-Dawley rats (n = 6). One week later, animals were subjected to ligation and excision of the c o m m o n bile duct. Serial MRS examinations were then performed on days 11, 21 a n d 47 post-surgery. Blood samples were collected at the indicated times to monitor s e r u m aspartate aminotransferase (AST) a n d albumin levels. Following day 47 measurements, animals were sacrificed a n d histology was performed on collected liver samples. 31P MRS, liver histology and bIood w o r k was also performed on an additional g r o u p of aged-matched rats ( n = 4 ) not subject to surgery. Quantitiative 31P MRS was performed on a 7 Tesla 21em bore magnet (Bruker Biospec MSLX 7/21). Animals were anesthetized and placed prone on a 31P/1H doubly tunable double ring surface coil (15/40 m m diameter) operating at 1 2 1 . 5 / 3 0 0 MHz. Localized 31P liver spectra were acquired using two-dimensional chemical shift imaging with previously described acquisition parameters. Results. Over the 7 week course of c o m m o n bile duct ligation/ excision a progressive decrease in the concentration hepatic ATP was observed, with significantly lower levels ( P < 0 . 0 1 a n d P < 0 . 0 0 5 ) occurring at days 27 a n d 47 respectively compared to baseline and controls. The reduction in hepatic ATP correlated with elevated AST (r=-0.52, P < 0 . 0 1 ) a n d decreased albumin levels ( r = 0 . 6 4 , P < 0 . 0 0 1 ) . Histologically, the livers from bile duct ligated rats revealed extensive bile d u c m a l proliferation, hepatocyte necrosis a n d fibrosis in keeping with biliary cirrhosis. The reduced hepatocyte area ratio in the livers of bile duct ligated rats strongly correlated with diminished hepatic ATP content ( r = 0 . 7 9 , P < 0 . 0 0 5 ) . Conclusion. The results of this study indicate that during progression of biliary disease to cirrhosis there is a progressive reduction in hepatic ATP. In addition, changes in ATP levels correlate with changes in standard parameters of hepatic injury, dysfunction and histology. Finally, hepatic 31P MRS provides a non-invasive means of d o c u m e n t i n g the severity of liver disease in an animal model of biliary cirrhosis. This study was supported by grants from the Health Sciences Centre Research F o u n d a t i o n a n d the Canadian Liver Foundation.

781

782

NO ROLE FOR AUTOREACTIVE ANTIBODY IN THE BREAKDOWN OF T-CELL TOLERANCE TO PYRUVATE DEHYDROGENASE COMPLEX, THE AUTOANTIGEN IN PRIMARY BILIARY CIRRHOSIS. A m a n d a J Robe, Jeremy M Palmer, J o h n A Kirby, David E Jones, Centre for Liver Research, University of Newcastle, Newcastle-upon-Tyne u k

PRESENCE OF SLA/LP-AUTOANTIBODIES IN PATIENTS W I T H PRIMARY BILIARY CIRRHOSIS AS A MARKER OF SECONDARY AUTOIMMUNE HEPATITIS (OVERLAP SYNDROME). Stephan Kanzler, Sevim Bozkurt, J o h a n n e s Herkel, Peter R Galle, University of Maillz, Mainz Germany; Hans P Dienes, University of Cologne, Cologne Germany; Ansgar W Lohse, University of Mainz, Mainz G e r m a n y

The autoimmune liver disease primary biliary cirrhosis (PBC) is characterised by breakdown in both B-cell and T-cell immune tolerance to the highly conserved self-antigen pyruvate dehydrogenasecomplex (PDC). It remains unclear how tolerance breaks down to such a highly-conserved self-antigen. Addressing how and why this break in tolerance occurs remains critical for our understanding of the pathogenesis of PBC. We have recently showal that when mice are co-sensitised with self (mouse) and non-self (bovine) PDC they undergo a rapid (within 5 weeks) breakdown of both B-ceU and T-cell tolerance to PDC of the type seen in PBC patients. Mice sensi0sed with self-PDC alone show breakdown of neither B-cell nor T-cell tolerance over 5 weeks, whilst animals sensitised with non-self PDC show a breakdown in B-cell but not T-cell tolerance. We have postulated that the mechanism for tolerance breakdown following self- and non-self-PDC co-sensitisation is as follows. Exposure to non-self-PDC in the context of adjuvant induces a T-cell response to non-conserved epitopes within non-self-PDC. These T-cells assist in the induction of an antibody response to non-self-PDC which is cross-reactive with self-PDC. Clear evidence exists for both these processes occurring in this model. It is our hypothesis that crossreactive anti-PDC antibody subsequently promotes breakdown of T-cell tolerance to selfPDC. There are two potential mechanisms. The first is that the cross-reactive antibody binds with self-PDC forming immune complexes which are taken up and processed by activated antigen presenting cells (APCs). The second potential mechanism is through cross-reactive antibody in its role as the B-cell antigen receptor promoting uptake of self-PDC by activated B-cells which themselves act as APCs priming autoreactive T-cells. In the current study we set out to address the first of these potential mechanisms by studying the capacity of cross-reactive antibody alone to replace the need for non-self-PDC in this model of breakdown of T-cell tolerance to PDC. Groups of female SJL mice of 10-12 weeks were sensitised with self (routine (m))PDC and non self (bovine (b))PDC (n=6), with mPDC and affinity purified murine anti-PDC IgG (n=8), with mPDC and control murine IgG (n=6) and anti-PDC IgG alone (n=6). Splenic T-cell proliferative responses to mPDC were asessed at 5 weeks post-sensitisation. As previously demonstrated the self/non-selfPDC co-sensitised animals showed a breakdown of T-cell tolerance to self-PDC (day 4 + mean I'ncorporated cpm 3532 +_ 3028 v control wells 2046 _1691; p<0.005). Animals co-sensitised with mPDC and murine IgG anti-PDC had significant residual serum titres of IgG anti-PDC (>1 in i06). These mice did not, however, show a breakdown in T-cell tolerance to self-PDC (mean incorporated cpm 1064 -+ 1069 v 963 -+ 869; p=ns). Similarly, no breakdown in T-cell tolerance to self-PDC was seen in the mPDC and control IgG and anti-PDC only sensitised animals. Conclusion. Anti-self-PDC antibody cannot replace the need for sensitisation with non-seIf in the self/non-self co-sensitisation pathway of breakdown of T-cell tolerance to self-PDC. This suggests that binding of immune complexes of self-PDC and anti-self-PDC antibody to APCs, and subsequent processing and presentation of self-PDC T-cell epitopes,is not the mechanism of breakdown of T-cell tolerance to self-PDC in this co-sensitisation model. The ability of anti-self reactive B-cells, as opposed to anti-self specific antibody, to promote tolerance breakdown is currently being studied.

Background: The aim of this study was to evaluate w h e t h e r the presence of SLA/LP-autoantibodies in PBC-patients gives evidence for a secondary AIH, also called AfH/PBC-overlap-syndrome. Patients a n d methods: O u t of 233 consecutive patients with PBC w h o h a d been followed between October 1980 a n d April 2000, we evaluated the data of anti-SLA/LP-positive patients a n d compared them to patients with an anti-SLA/LP-negative AIH/PBC-overlapsyndrome as well as to patients with a classical course of M H and PBC. Results: In total we could identify nine PBC-patients with anti-SLA/LP-antibodies (six women/three men) or 3.9% of the study population. Anti-SLA/LP-positive PBC-patients were slightly y o u n g e r at diagnosis in c o m p a r i s o n to anti-SLA/LPnegative PBC-patients (49.9 vs. 53.2 years). Transaminases a n d g-globulins were significantly higher in anti-SLA/LP-posifive PBC-patients in comparison to anti-SLA/LP-negative PBC-pafients (mean: 235 vs. 55 IU/I a n d 27.6 vs. 1 9 5 g/l). Anti-SLA/LP-positive patients h a d significantly more frequent an HLAtype that is characteristic for AIH (B8; DtL3; DR4). l m m u n o s u p p r e s s i v e therapy reduced inflammatory activity a n d cholestasis significantly. Relapses were frequent after reduction or discontinuation of immunosuppressive therapy. Conclusions: The presence of SLA/LP-autoantibodies in PBC-patients has a high specificity for a secondary AIH (AIH/PBC-overlap-syndrome). These patients have a g o o d response to immunosuppressive therapy. The autoantibody profile a n d immunogenetics m a y help in future to identify PBC-patients that benefit most from immunosuppressive therapy.