The mechanism by which immune tolerance of lipoylated subunits of the pyruvate dehydrogenase complex (PDC) is broken in patients with primary biliary cirrhosis remains the subject of vigorous research. The importance of this issue, and the difficulties inherent in addressing it in humans, has led to the development of a range of animal model systems which variably reproduce immunological and/or histological features of the human disease.1–3 A recent article in this journal from the Gershwin group provides data supporting a pathogenetic role for proteins which have been labeled with a lipoic acid–like xenobiotic compound.4 Specifically, the appearance of both PDC-specific autoantibodies and mild portal tract inflammation with bile-duct damage was induced in C57BL/6 mice by intraperitoneal sensitization with 2-octynoic acid–coupled bovine serum albumin in complete Freund's adjuvant (CFA) containing a high concentration (10 mg/mL) of Mycobacterium tuberculosis.
In several earlier reports, our group has described the use of a similar route and “boosted” CFA formulation to sensitize a range of mouse strains (including C57BL/6 and SJL/J) with components of PDC.1, 5–7 This universally results in the formation of autoantibodies. However, inflammatory portal tract lesions are only observed reproducibly in SJL/J mice, with a smaller response observed in the portal tract of some C57BL/6 animals. In our experience, the capacity to break T cell tolerance to self-PDC, critical for the induction of liver histological change, is similarly restricted to SJL/J6 and C57BL/6 mice (Fig. 1). Importantly, our group's established observation that control sensitization fails to induce significant portal tract inflammation in any of the mouse strains we tested is entirely consistent with the data presented by Wakabayashi et al.4
In 2002, Gershwin's group published two similar reports which were wholly critical of the animal model developed and used by our group.8, 9 Specifically, these reports claimed that intraperitoneal sensitization with boosted CFA (or even incomplete Freund's adjuvant) led to antigen nonspecific portal tract inflammation, with the inference that mouse sensitization approaches of this type are invalid for modeling primary biliary cirrhosis. Our group disagreed strongly with these findings and has successfully continued to use the intraperitoneal route with boosted CFA6, 10 or specific toll-like receptor ligands7 to break self-tolerance of PDC in SJL/J mice, leading to the development of specific portal tract inflammatory changes.1, 6, 7, 10
We are now pleased to note that Gershwin's group has successfully extended our intraperitoneal sensitization procedure, and thereby validated our broad approach, by using intraperitoneal antigen in boosted CFA to produce mild but specific portal tract inflammation in C57/BL6 mice.4 Their finding of no significant portal tract inflammation in control animals sensitized with boosted CFA and unmodified bovine serum albumin supports our view that the portal tract changes seen in animals induced to break immune tolerance to self-PDC using this sensitization approach are indeed a disease-specific phenomenon, supporting the validity of both our and their postulated murine models. Our experience using this system to sensitize with biotin-modified proteins10 and our studies comparing responses in different mouse strains lead us to suggest that sensitization of SJL/J mice with octynoic acid–modified proteins in boosted CFA might induce an even greater specific inflammatory response within the portal tract.