Most often molecular mimicry has been demonstrated
Most often molecular mimicry has been demonstrated for structures present on pathogens and hosts [60,61]. Predominantly virus infections have been implicated in the initiation and/or propagation of AIH . The most compelling association has been found for HCV, since LKM-1 PFK-158 have been also found in HCV-infected patients and antibodies binding to the HCV proteins NS3 and NS5a cross-react to a specific conformational epitope on the major AIH-2 autoantigen CYP2D6 . Further, LKM-1 antibodies specific for another CYP2D6 epitope cross-react with epitopes on proteins of HCV as well as human cytomegalovirus (HHV-5) . Further sequence homologies have been found between CYP2D6 and a variety of human pathogens including herpes simplex virus 1 (HSV-1) , Legionella pneumophila, Influenza A virus (H1N1), and Kaposi's sarcoma associated herpes virus (HHV-8) . However, similar to many other autoimmune diseases with associations with pathogen infections, there is yet no definite proof for pathogens to be directly involved in the initiation and/or acceleration of AIH. For more information about a possible role of pathogen infection on AIH, I would like to refer to a more detailed review article .
Relative immune tolerance in the liver Compared to other autoimmune diseases, such as autoimmune thyroiditis, multiple sclerosis (MS), or type 1 diabetes (T1D), the frequency of autoimmune diseases affecting the liver is rather low. Considering that the liver is the main organ responsible for detoxification and drug-metabolism and is therefore prone to cellular damage, such a low frequency of autoimmune liver diseases might come as a surprise. First, cellular damage may lead to local inflammation and activation of resident immune competent cells, such as Kupffer cells (KC) or liver sinusoidal endothelial cells (LSEC) and subsequently the attraction of infiltrating lymphocytes that may cause specific or unspecific damage to the liver parenchyma. Second, the metabolism of some drugs and xenobiotics results in the formation of reactive metabolites with the potential to covalently attach to macromolecules, such as proteins, lipids, or even DNA [67,68]. Such modified self-components may act as neo-antigens causing an aggressive immune response, as extensively demonstrated for the anesthetic agent halothane that can cause severe halothane hepatitis in susceptible individuals . Third, the liver is target of many pathogens, including Hepatitis-, Coxsackie- and Herpes simplex viruses. Such infections directly damage hepatocytes and other liver-resident cells and cause a strong local inflammation. Thus, pathogen infections of the liver might act as triggering factor for subsequent autoimmune reactivity . As protection from autoimmunity many mechanisms have evolved to regulate the immune balance in the liver. With the help of several animal models such protective mechanisms, some of which are unique to the liver, have been identified. For example, LSEC cross-present antigenic liver peptides resulting in T cell inactivation [70,71] and induction of antigen-specific regulatory T cells  and thereby maintain hepatic tolerance to liver autoantigens. Further, hepatic stellate cells (HSC) have been shown to induce T cell apoptosis . The tolerogenic environment of the liver has been impressively demonstrated by Lüth et al. . They found that an ectopic expression of myelin basic protein (MBP, a major autoantigen in MS) in the liver, but not in the skin, resulted in the protection from neuroinflammatory disease in a mouse model. MBP-expression in the liver caused the generation of MBP-specific regulatory T cells, which inhibited the proliferation of MBP-specific aggressive T cells. These regulatory T cells were sufficient to protect from MS-like disease, since adoptive transfer into mice not expressing MBP in the liver also protected from disease . In summary, the liver microenvironment seems to be a milieu that regulates peripheral tolerance and prevents excessive damage of the liver by autoimmune dysregulations. Thus, in order to generate a model for any autoimmune-mediated liver disease one has to overcome such tolerogenic mechanisms.