Potential conflict of interest: Nothing to report.
T cells in the liver: There is life beyond the graveyard†
Article first published online: 30 MAY 2007
Copyright © 2007 American Association for the Study of Liver Diseases
Volume 45, Issue 6, pages 1580–1582, June 2007
How to Cite
Bertolino, P., Bowen, D. G., Benseler, V. (2007), T cells in the liver: There is life beyond the graveyard. Hepatology, 45: 1580–1582. doi: 10.1002/hep.21786
- Issue published online: 30 MAY 2007
- Article first published online: 30 MAY 2007
Keating R, Yue W, Rutigliano JA, So J, Olivas E, Thomas PG, Doherty P. Virus specific CD8+ T cells in the liver: armed and ready to kill. J Immunol 2007;178:2737-2745. (Reproduced by permission)
Influenza A virus infection of C57BL/6 mice is a well-characterized model for studying CD8+ T cell-mediated immunity. Analysis of primary and secondary responses showed that the liver is highly enriched for CD8+ T cells specific for the immunodominant H2DbNP366-374 (DbNP366) epitope. Functional analysis established that these liver-derived virus-specific CD8+ T cells are fully competent cytotoxic effectors and IFN-gamma secretors. In addition, flow cytometric analysis of early apoptotic cells showed that these influenza-specific CD8+ T cells from liver are as viable as those in the spleen, bronchoalveolar lavage, mediastinal lymph nodes, or lung. Moreover, cytokine profiles of the influenza-specific CD8+ T cells recovered from different sites were consistent with the bronchoalveolar lavage, rather than liver population, being the most susceptible to activation-induced cell death. Importantly, adoptively transferred influenza virus-specific CD8+ T cells from the liver survived and were readily recalled after virus challenge. Together, these results show clearly that the liver is not a “graveyard” for influenza virus-specific CD8+ T cells.
For a long time, the liver was thought to play a passive role during an immune response to extrahepatic pathogens. This view shifted in 1994 when Nick Crispe and his colleagues suggested that this organ might be a disposal site for CD8+ T cells activated in peripheral lymphoid tissues (lymph nodes and spleen).1 This hypothesis, known as the graveyard model, a term reminiscent of African elephants going to die in a defined spot in the jungle, was based on observations made with transgenic mouse models: when T cell receptor (TCR) transgenic mice were injected with the antigenic peptide triggering TCR activation, T cells were activated and massively accumulated within the liver, in which a high proportion of them underwent apoptosis.1 The presence of high numbers of apoptotic lymphocytes in the liver was also observed in many experimental and clinical settings, and this increased the credibility of the hypothesis.2 The graveyard model has received major attention from immunologists, because it might explain some of the unique tolerogenic properties of the liver in transplantation and viral hepatitis. However, from its conception, this model did not adequately explain the liver tolerance effect. Even if the liver was able to dispose of T cells at the end of their life or, as proposed later, to actively kill effector T cells, recognizing their antigen in the liver,3 it did not explain why effector and memory T cells generated during viral infections are sometimes capable of clearing infection.
Recent data appear to presage the death of the graveyard model as a mechanism to explain liver-induced tolerance. The first set of evidence comes from the developers of the hypothesis themselves, who recently demonstrated the presence of full effector T cells in a transplanted mouse liver and concluded that their model needs revision.4 This recent article by Keating et al.5 reinforces this view. By following the immune response specific for an immunodominant influenza epitope in experimentally infected mice using major histocompatibility complex class I tetrameric complexes, the authors of this article demonstrated that the liver contains a low but significant number of influenza-specific T cells following primary and secondary immune responses. Most importantly, they show that these cells are not undergoing apoptosis but rather are fully functional effector cells able to secrete high levels of interferon-γ (IFN-γ), interleukin-2, and tumor necrosis factor α (TNF-α) and capable of killing target cells expressing the relevant epitope. Interestingly, influenza-specific T cells isolated from the liver are less sensitive to peptide stimulation than those isolated from bronchoalveolar lavage, and this indicates that they exhibit reduced functional avidity, a concept that is consistent with a tolerogenic role for the liver. Because low-avidity T cells are less susceptible to activation-induced cell death, the authors suggested that the liver might actually preserve antigen-specific effector T cells and act as a reservoir of early response cells for rapid mobilization. Their low functional avidity would protect or limit damage induced by these T cells to other organs and to the liver itself. The presence of low-avidity T cells in the liver, however, might have profound implications for hepatitis B virus or hepatitis C virus (HCV) infections. HCV-specific CD8+ T cells detected in patients developing persistent infection generally display impaired function characterized by reduced IFN-γ production, proliferation, and cytotoxic T lymphocyte activity,6 an anergic phenotype also characteristic of low-avidity T cells.7 Low-avidity cells might be efficient enough to induce liver damage but be ineffective in clearing the virus.8 This might contribute, in particular, to the pathology of chronic HCV infections. Future studies investigating the avidity of HCV-specific T cells are required to address the validity of this hypothesis.
Another important aspect of Keating et al.'s article is the confirmation that virus-specific T cells may be retained in the liver even when the virus is not known to infect this organ.9–11 This retention might reflect sluggish transit through this organ due to slow intrasinusoidal blood flow, rather than cells taking up permanent intrahepatic residence.12 What is the result of the retention of effector T cells in the liver? Several studies have demonstrated that this retention may not always be innocuous and that these T cells might cause liver damage even when hepatocytes do not express the antigen recognized by the T cells13–15 (Fig. 1). Although the influenza virus is not hepatotropic, it is possible that influenza-specific T cells are activated within the liver by hepatic cells cross-presenting viral proteins. Indeed, hepatic cells such as liver sinusoidal endothelial cells16 and, more recently, hepatic stellate cells17 have been shown to be able to cross-present antigens. Such intra-hepatic presentation of antigens derived from a nonhepatotropic agent could lead to the activation of pathogen-specific T cells within the liver, with a resultant bystander hepatitis mediated through the release of cytokines such as IFN-γ and TNF-α.15 This phenomenon would appear to have clinical relevance as it has been reported during influenza infection when the virus was not detectable within the liver.18 More significantly, it might explain the relatively common occurrence of hepatitis in individuals infected with the Epstein-Barr virus, which is not thought to infect hepatocytes.19
In conclusion, this article suggests that there are at least 2 outcomes from intrahepatic T cell activation that may damage the liver rather than result in the graveyard death of cells and tolerance. First, the liver may preferentially induce low-avidity T cells that allow the persistence of viral infections resulting in antigen-specific chronic hepatitis. Second, T cell activation may result in a non-antigen-specific hepatitis through cytokine release, a not uncommon event in non-hepatotropic viral infections. Thus, these studies add new dimensions to understanding the role of the liver in the balance between tolerance and immunity during immune responses.