Autoimmune hepatitis: 50 years of (slow) progress


  • M. Eric Gershwin,

    Corresponding author
    1. Division of Rheumatology, Allergy and Clinical Immunology, University of California at Davis School of Medicine, Davis, CA
    • Address reprint requests to: M. Eric Gershwin, M.D., Division of Rheumatology, Allergy and Clinical Immunology, University of California at Davis School of Medicine, 451 Health Sciences Dr., Ste. 6510, Davis, CA 95616. E-mail:; fax: 530-752-4669.

    Search for more papers by this author
  • Edward L. Krawitt

    1. Division of Gastroenterology and Hepatology, Geisel College of Medicine at Dartmouth, Hanover, NH
    Search for more papers by this author

  • Potential conflict of interest: Dr. Krawitt consults for Intercept and AbbVie.

  • See Article on Page 1007

  • Funding provided by National Institutes of Health Grant DK39588.


autoimmune hepatitis


forkhead box P3


primary biliary cirrhosis


primary sclerosing cholangitis


T regulatory cells

The first textbook on autoimmune diseases was published 50 years ago in 1963 by Ian R. Mackay and F. Macfarlane Burnet, about a decade after the concept of immunologic tolerance had been established. It was entitled “Autoimmune Diseases: Pathogenesis, Chemistry and Therapy.”[1] Interestingly, it was published as part of a series called “Living Chemistry.” In this seminal work there were 13 chapters, of the approximately 30 known autoimmune diseases, including what was called “Active chronic and lupoid hepatitis,” now known as autoimmune hepatitis (AIH). There are now more than 100 well-defined autoimmune diseases, and within that group the liver, essential to normal immune tolerance, can itself become a victim of an autoimmune attack, in such diseases as primary biliary cirrhosis (PBC), primary sclerosing cholangitis (PSC), AIH, autoimmune sclerosing cholangitis, and conditions with overlapping features within this spectrum, referred to as overlap or variant syndromes.[2, 3]

In 1963 the concept of AIH was primarily an epiphenomenon and the discussion in the textbook revolved around differences between alcoholic hepatitis, systemic lupus, and the rudimentary science of serology. A half century later the serologic findings in autoimmunity and autoimmune diseases include entire catalogs, virtually a kaleidoscope of nonspecific and specific antibodies, directed at epitopes including those in diagnostic kits based on recombinant DNA technology. More important, however, is our appreciation that tolerance requires a well-orchestrated and balanced cellular immune system involving innate and adaptive responses.

How has our understanding of AIH changed in 50 years? First, from a generic perspective, it is clear that a breach in tolerance involves a genetic predisposition, a gender bias, and environmental contributions. The environmental factors, or “triggers,” remain for the most part ill-defined. In many autoimmune diseases there have been rigorous genome-wide association studies, intensive analyses of family pedigrees, and, in some cases, even epigenetic analysis. In AIH there are genetic factors that influence the incidence, clinical expression, severity, and response to treatment, most commonly ascribed to alleles along the major histocompatibility class II complex.[4] However, unlike PBC and PSC, more detailed analysis using genome-wide association studies of adequate size are not completed; thus, there are only very limited data beyond HLA candidate gene studies.[5] One interesting exception has been the occurrence of AIH in children that bear the AIRE gene mutation, but these data and the analogy appear to be more the exceptions than the rule.[6] Studies of twins have proven to be an exceedingly important resource to distinguish the relative role of genetics and environment, allowing quantification of risk based on comparisons of disease concordance between monozygotic and dizygotic twins but the relative infrequency of twinning in AIH has made such analysis problematic and data are limited to isolated case reports.[7, 8]

In this issue of Hepatology,[9] the Vergani group provides significant hope that what has otherwise been slow progress understanding the mechanisms underlying immunoregulatory dysfunction in AIH is accelerating. The Vergani laboratory has moved beyond descriptive immunobiology and has been instrumental in emphasizing that a breach in tolerance occurs when balance in immunoregulatory mechanisms goes astray. Indeed, the very concept of T regulatory cells (Tregs) is a relatively recently recognized phenomenon. Historically, in the 1970s and 1980s evidence emerged that the immune system included a variety of suppressor mechanisms supporting the concept that dysregulation of reactivity against self was crucial to the development of autoimmune diseases. In AIH a study of families measuring nonspecific suppressor cell function suggested that immunoregulatory dysfunction was a familial abnormality, although unlinked to HLA loci.[10]

With the realization that there is an entire repertoire of regulatory mechanisms that involve innate and adaptive responses, the significance of CD4 Tregs as primary mediators of peripheral tolerance in infectious and autoimmune inflammatory reactions became apparent. In fact, Tregs come in multiple “flavors” and reflect the ability to generically regulate immunity. As such, it is difficult to explain why a global defect in T-cell-mediated regulation leads to a selective loss of tolerance and disease-specific pathology such as AIH. Interestingly, recent observations suggest that thymic Tregs suppress responses to foreign antigens while peripheral Tregs suppress those to self-antigens. The data reported by Vergani et al. build on their previous investigations of a decrease in number and function of a subset of Tregs which express the forkhead box P3 (FOXP3) transcription factor, and the relationships of these cells to T helper (Th) 17 cells and the expression of Gal 19,[11-13] and illustrate that immunoregulatory dysfunction in AIH is dependent on multiple defects in CD39pos Tregs. Central to these most recent experiments is the theory that release of adenosine via catalysis of nucleotides by concomitant CD39 and CD73 ectoenzymes in concert with adenosine A2A receptors on T effector cells generates immunosuppressive loops, inhibiting T cell effector function and interleukin (IL)6 expression, thereby affecting Treg generation, as well as promoting transforming growth factor beta (TGF-β) expression.[13] Not only did Vergani's team find a decrease in circulating CD4 posCD25posCD39 pos Tregs but there was a decreased capacity to catalyze adenosine triphosphate (ATP) and adenosine diphosphate (ADP), and a failure to suppress IL17 production by effector cells. By defining the frequency and cellular phenotype, including ectonucleotidase activity, the group demonstrated that the lineage expressed in AIH has a very proinflammatory profile characterized by a Th1 response. More important, there is not only reduction of Tregs and function, but also that during disease this subpopulation can be converted into pathogenic effector cells. Essentially a multiplicity of flaws in CD39pos, Tregs creates a perfect immunological storm that leads to AIH (Fig. 1). The Vergani model brings a new level of sophistication to a disease with no clear murine model and a human disease that continues to defy classification. In the case of murine models, other than the unique AIRE mutation, the best-defined system is mice with medullary epithelial cell depletion secondary to conditional deletion of TRAF6, an E3 ubiquitin protein ligase. These mice develop a very focused inflammatory liver disease with striking similarities to human AIH.[14] Interestingly, peripheral tolerance in this model is normal in spite of a 50% reduction of thymic Treg production. The relationship of these data to the Vergani schema are unclear and may simply reflect the diversity of the events that lead to common organ-specific pathology. On the other hand, data reported by Vergani and colleagues fit very nicely with generic models of tolerance that clearly demonstrate that selective dysregulation of Th17 responses via qualitative and/or quantitative changes in Tregs will amplify disrupted regulatory circuits and promote Th17-mediated inflammation.[15] A logical next step is to define the transcription factors and regulatory network operative in AIH. Obviously, many other questions remain to be addressed, including the genetic, environmental, and hepatic-specific contributions to this subpopulation, whether these defects are inherited or acquired, whether these defects are liver-specific, and especially whether they present a platform for therapy of autoimmune diseases, perhaps crudely via cell infusion or selectively by genetic manipulation.

Figure 1.

Genetic and environmental factors influence the dysregulation of the immune response and contribute to the predisposition for AIH. CD39pos Tregs from AIH patients reflect a decrease in cell number, skewed subset phenotype, and impaired ectonucleotidase activity. This leads to defective IL17 suppression and T-cell proliferation suppression. Lastly, there is an increased rate of conversion of the CD39pos Tregs to IL17, interferon-gamma (IFN-γ)-secreting effector cells.

Thus, the new findings by the Vergani group imply potentially new therapeutic pathways that may be altered to restore tolerance. Indeed, there is no cure for any autoimmune disease and even those agents that are effective in reducing patient symptomatology are successful only because they dampen the secondary inflammatory response. In AIH, a highly specific and selective disease, there appears to be the potential to down-regulate pathogenic effector pathways. One now has the ability to examine in detail the microRNA population, the proteins, and their pathways activated in this CD39pos Treg population. It will only be through a rigorous molecular approach that specific targeting will be possible. Cellular immunology is a smorgasbord of cells, cell populations, stages of activation, and regulation. Just as serology has moved to the use of recombinant autoantigens,[16] designer antibodies, and specific peptides, so cellular immunology needs to incorporate the newer technologies so that the pathogenic events in clinical disease can be defined on the basis of cell receptors and their specific amino acid residues and spatial orientation. We hope that progress in the future will include the burgeoning fields of epigenetics, proteomics, and metabolomics. Such work holds the promise of defining the elusive environmental factors that influence genetic predisposition. We note that the use of microRNA technology holds significant promise to defining molecular signatures which may be important both diagnostically as well as to predict therapeutic responses; such has already been proposed for the diagnosis of acute rejection in kidney transplantation.[17]

There has been enormous progress in our understanding of autoimmunity but most of this progress has been descriptive. The era of molecular autoimmunity is in its infancy as we move to explain the serologic abnormalities in clinical hospital laboratories to instead defining the cellular repertoire and the mechanisms that underlie this function. It is hoped that such study would lead to a better understanding of the earliest events that determine loss of tolerance and, more important, facilitate development of selective therapeutic tools.

  • M. Eric Gershwin, M.D.1 Edward L. Krawitt, M.D.2

  • 1Division of Rheumatology, Allergy and Clinical Immunology, University of California at Davis School of Medicine, Davis, CA

  • 2Division of Gastroenterology and Hepatology, Geisel College of Medicine at Dartmouth, Hanover, NH