Intrahepatic T cell survival versus death: which one prevails and why?
Article Outline
In this issue of the Journal Dikopoulos et al. present studies in which immunization of normal mice with a variety of antigens results in hepatic accumulation of interferon-γ producing activated CD8+ T cells with in vitro cytolytic function and minimal apoptosis [1].
Since the initial demonstration that T cells accumulate and apoptose in the liver after systemic CD8+ T cell activation, a growing body of literature has demonstrated that the healthy liver has unique immunological features [2]. These have been of great interest in their own right, and for understanding the distinctive clinical properties of the liver in rodents and man. The unique aspects of the liver cover a broad range of immunobiology including extra-thymic T cell maturation, priming of naive T cells by liver specific antigens, and retention of activated T cells by the liver in the presence and absence of antigen [3]. In the paper by Dikopoulos et al., CD8+ T cells were probably activated in lymph nodes and subsequently retained by the liver, confirming previous reports [4]. Their additional findings of CD8+ T cell survival and effector function seem at odds with much of the published literature. In many experimental models of CD8+ T cell activation, hepatic T cell retention has been associated with T cell apoptosis [2], [3]. So what are the differences between Dikopoulos et al.’s. model, and earlier models in which apoptosis was a dominant feature, and what do these differences tell us about the choice between survival and apoptosis of hepatic CD8+ T cells? One issue is precursor frequency of activated CD8+ T cells. In the models with the highest degree of hepatic apoptosis, a large percentage of systemic T cells (up to 80%) were activated, whereas in the current study only 0.1–0.5% of the T cell pool was activated. Even if the level of apoptosis is the same when large and small T cell populations are activated, technically it is more difficult to demonstrate apoptosis in a small T cell population. The clearance of apoptotic cells by phagocytosis is highly efficient, making it difficult to identify apoptotic cellular debri in vivo. Even with the same degree of apoptosis, the total number of apoptotic cells are much higher with activation of 80% of cells, and may transiently overwhelm the ability of the liver to remove them by phagocytosis, allowing their detection. However, the frequency of the activated T cells may be a genuine determining factor in the percentage of cells undergoing apoptosis. This is possible if fratricidal T cell death is occurring inside the liver. CD8+ T cell fratracide has been demonstrated after human T cell lymphotropic virus type 1 infection [5]. Dennert et al. have argued, based on the requirement for CD95-L on lymphoid cells for hepatic T cell apoptosis, that T cell fratricide is a significant mechanism in hepatic T cell apoptosis [6]. Such a system has the advantage of directly linking the degree of apoptosis to the number of activated T cells.
In the study by Dikopoulos et al., intramuscular vaccination probably did not result in significant amounts of antigen reaching the liver, whereas high affinity antigen was present in the liver in models with the greatest amount of apoptosis. The presence of high affinity antigen is a vital determinant of the degree of intrahepatic death (Y. Kuniyasu et al., submitted). Maximal induction of T cell death in the setting of hepatic antigen seems counter-productive in the setting of a hepatotropic viral infection. It is, however, important to note that even in the presence of massive T cell apoptosis, there is evidence of T cell effector function as assessed by elevations in transaminases and hepatocyte apoptosis [7]. T cell apoptosis and effector function are therefore occurring simultaneously.
A third reason for minimal T cell apoptosis in Dikopoulos et al.’s study is the presence of effective T cell priming. In the current study DNA vaccines or CpG based adjuvants were used, whereas in the studies with maximal hepatic apoptosis peptide alone was used [3], [7]. Soluble peptide will result in priming by a wide variety of cells, many without effective co-stimulatory function. This is much more likely to lead to T cell activation induced cell death (AICD). The induction of AICD by ineffective priming may be particularly relevant in the liver where a number of non-professional antigen presenting cells like Kupffer cells, endothelial cells and possibly hepatocytes may present antigens to naive or previously activated T cells. We have previously shown that activated CD8+ T cells are in close contact with Kupffer cells, and they also interact with endothelial cells [2], [8].
In addition to adequate priming of T cells, the number of activated T cells in the liver, and the presence of peptide in the liver, are there additional factors which affect the balance between T cell apoptosis, survival and effector function? The answer is a definite yes but such factors are poorly understood. Limmer et al. have demonstrated that activated T cells in the liver do not cause injury, even with specific antigen on hepatocytes, unless an additional hepatic insult such as irradiation, or infection with listeria monocytogenes is present [9]. Which components of irradiation and listeria infection allow the auoreactive cells to induce injury? Lipopolysaccharide (LPS) is a candidate, but LPS could not reproduce the effect of listeria. This may be because the liver is constantly exposed to LPS, and physiological concentrations of LPS result in interleukin-10 production by KC, which down regulates antigen uptake, class II molecules and the co-stimulatory molecules CD80 and CD86 on sinusoidal endothelial cells. In addition LPS increased the production of the potent immunusuppressive cytokine TGF-β by Kupffer cells, although the consequences of this are unclear. Type I interferons recruit T cells during viral infections and may play an important part in enhancing T cell responses against hepatic antigens [2].
In summary Dikopoulos et al. describe a model which maximizes T cell survival in the liver by efficiently priming a low frequency of T cells in the absence of antigen in the liver. From a therapeutic perspective, these data are very encouraging and shows that in mice it is possible to produce significant intrahepatic T cell responses by conventional immunization. It will be of great interest to test for memory T cell responses from healthy human donor liver tissue against previously immunized antigens such as hepatitis B surface antigen. In most clinically relevant conditions such as chronic viral hepatitis and transplantation, large numbers of activated T cells and specific antigen are present in the liver, and T cell priming may have been sub-optimal, all favoring T cell apoptosis. One of the challenges is to understand the regulatory mechanisms in the liver sufficiently to therapeutically manipulate T cell survival and apoptosis in these clinical situations.
References
- . Specific, functional effector/memory CD8+ T cells are found in the liver post-vaccination. J Hepatol. 2003;39::917
- . Hepatic T cells and liver tolerance. Nat Rev Immunol. 2003;3:51–62
- . Immunology of the healthy liver: old questions and new insights. Gastroenterology. 2001;120:250–260
- . Selective retention of activated CD8+ T cells by the normal liver. J Immunol. 1999;163:3202–3210
- Fratricide among CD8(+) T lymphocytes naturally infected with human T cell lymphotropic virus type I. Immunity. 2000;13:657–664
- . Elimination of virus-specific cytotoxic T cells in the liver. Crit Rev Immunol. 2002;22:1–11
- . Liver damage preferentially results from CD8(+) T cells triggered by high affinity peptide antigens. J Exp Med. 1998;188:1147–1157
- Efficient presentation of exogenous antigen by liver endothelial cells to CD8+ T cells results in antigen-specific T-cell tolerance. Nat Med. 2000;6:1348–1354
- Failure to induce organ-specific autoimmunity by breaking of tolerance: importance of the microenvironment. Eur J Immunol. 1998;28:2395–2406
PII: S0168-8278(03)00525-7
doi:10.1016/j.jhep.2003.10.005
© 2003 European Association for the Study of the Liver. Published by Elsevier Inc. All rights reserved.
