The Liver as an Immunological Organ

Vito Racanelli1 and Barbara Rehermann2

The liver is a unique anatomical and immunological site in which antigen-rich blood from the gastrointestinal tract is pressed through a network of sinusoids and scanned by antigenpresenting cells and lymphocytes. The livers lymphocyte population is selectively enriched in natural killer and natural killer T cells which play critical roles in rst line immune defense against invading pathogens, modulation of liver injury and recruitment of circulating lymphocytes. Circulating lymphocytes come in close contact to antigens displayed by endothelial cells, Kupffer cells and liver resident dendritic cells in the sinusoids. Circulating lymphocytes can also contact hepatocytes directly, because the sinusoidal endothelium is fenestrated and lacks a basement membrane. This unique anatomy of the liver may facilitate direct or indirect priming of lymphocytes, modulate the immune response to hepatotrophic pathogens and contribute to some of the unique immunological properties of this organ, particularly its capacity to induce antigen-specic tolerance. (HEPATOLOGY 2006;43:S54-S62.)

Anatomy of the Liver and Its Immune System

The structural organization of the liver has profound implications for its immune function. The livers blood supply depends on a conventional arterial system from the abdominal aorta that supports predominantly bile ducts and other tissues in the portal tracts, and on two venous systems: a minor system from the arterial plexus within portal tracts (peribiliary plexus) and a major system from the splanchnic organs. About 30% of the total blood passes through the liver every minute1 carrying about 108 peripheral blood lymphocytes in 24 hours.2 Blood enters the hepatic parenchyma mainly via terminal portal vessels, then passes through a network of liver sinusoids, leaving the parenchyma via the central hepatic veins (Fig. 1). Due to the small diameter of the sinusoids, minimal increases in systemic venous pressure and perturbations of sinusoidal ow result in stasis, which lengthens the contact between lymphocytes and antigen-presenting cells
Abbreviations: DC, dendritic cells; LPS, lippolysaccharides; APC, antigen-presenting cells; LSEC, liver sinusoidal endothelial cells; NKT cell; natural killer T cell; HCV, hepatitis C virus; ER, endoplasmic reticulum; TAP, transporter associated with antigen processing. From the 1Department of Internal Medicine and Clinical Oncology, University of Bari Medical School, Bari, Italy; and 2Immunology Section, Liver Diseases Branch, NIDDK, National Institutes of Health, DHHS, Bethesda, MD. Supported by the Intramural Research Program of the NIH, NIDDK. Address reprint requests to: Barbara Rehermann, M.D., Immunology Section, Liver Diseases Branch, NIDDK, National Institutes of Health, DHHS, Bldg. 10, Room 9B16, 10 Center Drive, Bethesda, MD, 20892. E-mail: Rehermann@nih.gov; fax: 301-402-0491. Copyright 2006 by the American Association for the Study of Liver Diseases. Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/hep.21060 Potential conict of interest: Nothing to report. S54

and promotes lymphocyte extravasation. Extravasation is further facilitated by fenestrations in the monolayer of sinusoidal endothelial cells (LSEC)3 that allow lymphocytes to access the space of Disse via cytoplasmic exten sions and to touch the underlying extracellular matrix, stellate cells and hepatocytes. Hepatocytes constitute only about two thirds of the total cell population in the liver. The remaining population of nonparenchymal cells is diverse and includes LSEC, Kupffer cells, biliary cells, stellate (Ito or fat-storing) cells and intrahepatic lymphocytes (Fig. 2). Lymphocytes are found scattered throughout the parenchyma, as well as in the portal tracts. The average human liver contains a population of approximately 1010 lymphocytes, which include conventional and unconventional lymphocyte subpopulations of the innate (NKT and NK cells) and adaptive immune systems (T and B cells), respectively (Fig. 2). Conventional T cells comprise CD8 and CD4 T cells. Both populations display a diverse repertoire of T cells with -chain T cell receptors, that recognize antigens in the context of MHC class I and II molecules, respectively. CD8 T cells typically outnumber CD4 T cells in the liver and the frequency of effector/memory cells is higher than in the blood. Unconventional T cells comprise various cell types that are categorized into two major populations: those that express NK cell markers (also called NK T cells) and those that do not. The population of classical NKT cells arises in the thymus, displays a very restricted T cell receptor repertoire (typically the T cell receptor V 24 and V 11 chains in humans) and recognizes antigens in the context of the MHC class I molecule CD1d. While their natural antigen is not known, the marine sponge antigen alpha-

HEPATOLOGY, Vol. 43, No. 2, Suppl. 1, 2006

RACANELLI AND REHERMANN

S55

Fig. 1. Immune cells in the healthy liver. Liver sinusoids are lined by a fenestrated monolayer of sinusoidal endothelial cells (LSEC). Note that each lymphocyte that passes through the sinusoids is in close contact to the endothelial cells anking the sinusoids, because of the small diameter of the liver sinusoids (approx. 6-15 m)86 and the comparably large diameter of lymphocytes ( 7-12 m).87 The Space of Disse contains stellate cells in a loose extracellular matrix. B, B cell; DC, dendritic cell; HSC, hepatic stellate cell; KC, Kupffer cell; NK, natural killer cell; T, T cell.

galactosyl ceramide is used as a reliable experimental tool to activate all classical NKT cells. Classical NKT cells can be either CD4-positive or CD4/CD8-double negative. By contrast, nonclassical NKT cells encompass TCR and TCR T cells, do not use the T cell receptor V 14 chain and do not express the CD8 chain.4 Classical and nonclassical NKT cells are more abundant in the liver than in other organs and can constitute up to 30% of the intrahepatic lymphocyte population.5 Their migration to and/or expansion in the liver is controlled by NK cells 6 which are present at an unusually high frequency among liver-resident lymphocytes (Fig. 2). NK cells represent a lymphoid population with a potent cytolytic activity

against virus-infected or tumor cells. Their function is regulated by both activating and inhibitory receptors with inhibition as the dominant signal. Unconventional T cells that do not express NK cell markers include the major group of TCR cells (also called T cells). This group represents 15% to 25% of all intrahepatic T cells, thereby rendering the liver one of the richest sources of T cells in the body. T cells possess oligoclonal or invariant TCRs that recognize a limited range of antigens such as stress proteins and nonprotein antigens (for a review see reference 7). The liver also contains several types of resident antigen presenting cells that may capture antigens that either

Fig. 2. Cell composition of the healthy liver. Numbers indicate the estimated frequency of each population relative to the total number of parenchymal and nonparenchymal cells in the liver.

S56

RACANELLI AND REHERMANN

HEPATOLOGY, February 2006

pass through the liver or are released in the form of cellassociated antigen when pathogen-infected hepatocytes die. Resident antigen-presenting cells include Kupffer cells, which are members of the reticuloendothelial system, liver sinusoidal endothelial cells (LSEC), which represent an unusual type of vascular endothelial cells, and dendritic cells (DC). All three types of antigen presenting cells are thought be crucial for the maintenance of tolerance under noninammatory conditions. Hepatocytes may also present antigens to liver-inltrating T cells but they are more commonly regarded as targets for cellular immune responses and their role in the induction of primary immune responses, although suggested,8 is still less clear. Kupffer cells represent the largest group of xed macrophages in the body and account for about 20% of nonparenchymal cells in the liver.9 They are derived from circulating monocytes that arise from bone marrow progenitors.10 Once localized within the liver, Kupffer cells reside within the sinusoidal vascular space, predominantly in the periportal area. In this location, they are perfectly situated to clear endotoxins from the passing blood and to phagocytose debris and microorganisms. Their slow migration along the liver sinusoids causes frequent perturbations and even temporary stasis of the sinusoidal blood ow11 thereby facilitating close contact to passing lymphocytes. Kupffer cells also pass through the space of Disse, make direct contact with hepatocytes and phagocytose apoptotic hepatocytes. LSECs make up most of the nonparenchymal cells in the liver ( 50%). Although they line sinusoids in a similar way as vascular endothelial cells line arteries and central and portal veins,3 their morphology differs considerably and they form a sieve-like, fenestrated endothelium. LSECs express molecules that promote antigen uptake, including the mannose receptor and the scavenger receptor, and molecules that promote antigen presentation, including MHC class I and II and the costimulatory molecules CD40, CD80 and CD86.12,13 Receptor-mediated endocytosis and/or phagocytosis, antigen processing and presentation of LSEC occur with similar efcacy as those of DCs.14 Resident hepatic DC are derived from the bone marrow15 and typically located around the central veins and portal tracts.16 In the healthy liver, DCs are predominantly immature cells,17 prone to capture and process antigens. Because IL-10 and TGF- are constitutively expressed by Kupffer cells and LSECs, and are inducible in stellate cells, the uninfected liver provides a unique cytokine microenviroment that may render resident DCs tolerogenic.17-19 Resting DCs can, for example, inhibit proliferation and cytokine-production of activated, tis-

sue-inltrating lymphocytes through CTLA-4 and PD1.20 Once activated, they downregulate these receptors and instead, increase their capacity to migrate via the Space of Disse to the lymphatics in the portal tracts and ultimately to extrahepatic lymph nodes.21,22

Innate Immune Responses in the Liver

As already underlined, the liver is selectively enriched in macrophages (Kupffer cells), NK cells, and NKT cells which are key components of the innate immune system. Kupffer cells are activated by various bacterial stimuli, including lipopolysaccharides (LPS) and bacterial superantigens. Kupffer cell-derived cytokines play a key role in modulating the differentiation and proliferation of other cells. In response to physiological concentrations of LPS, Kupffer cells readily produce TNF- and IL-10,23 which downregulate receptor-mediated antigen uptake and MHC class II-expression on LSEC and DCs and decrease T cell activation.24 Kupffer cells also produce prostanoids, nitric oxide, and reactive oxygen intermediates which suppress T cell activation.25 Indeed, systemic tolerance to antigens in the portal vein26 depends on Kupffer cells because it is impaired if Kupffer cells are depleted.27 On the other hand, Kupffer cells are also a signicant factor in host resistance to primary and secondary infections. In particular, they are known to initiate and orchestrate an effective rst-phase response. Kupffer cell-derived IL-12 and IL-18, for example, regulate NK cell differentiation and promote the local expansion of cytotoxic NK cell subpopulations that express large amounts of antiviral IFN- .28 Other Kupffer cell-derived cytokines such as IL-1 , IL-6, TNF- and leukotrienes promote the inltration and antimicrobial activity of neutrophils.29 Neutrophils kill extracellular bacteria that are bound to Kupffer cells and hepatocytes via surface phagocytosis, stimulate other innate immune cells via secretion of inammatory cytokines and promote the inux and activation of CD4 and CD8 T cells.29 Hepatic NK cells modulate liver injury by balancing local production of proinammatory (Th1) and anti-inammatory (Th2) cytokines upon activation through their activating and inhibitory receptors. Because inhibition dominates over activation, the threshold for NK cell activation is lowest in the absence of ligands such as MHC class I that bind to inhibitory receptors.30 In the absence of inhibitory signals and in the presence of cytokines such as type I IFN and type-IFN-induced CCL3,31 ligation of activating NK cell receptors results in NK cell activation and target cell lysis. Activation does also include rapid production of IFN- , which stimulates hepatocytes and LSEC32 to secrete the chemokine CXCL9 and thereby recruit T cells to the liver. NKT cells recognize non-pep-

HEPATOLOGY, Vol. 43, No. 2, Suppl. 1, 2006

RACANELLI AND REHERMANN

S57

tide antigen targets such as lipid and glycolipid components of mycobacterial cells walls. Recognition is CD1restricted33,34 and CD1 can be expressed by hepatocytes and by antigen-presenting cells such as macrophages, dendritic cells and B cells. The majority of classical NKT cells are activated by IL-12, which is produced by dendritic cells and Kupffer cells and activation typically results in Fas-mediated cell lysis.35,36 This cytotoxic capacity appears to be crucial for the role of NKT cells in the ConA-induced hepatitis model.37 Because of their capacity to rapidly release large amounts of IFN- and IL-4,33,38 NKT cells are thought to polarize the local and systemic adaptive immune responses to either a proinammatory type I (IFN- , TNF- ) or an antiinammatory type II (IL-4, IL-10, IL-13) prole. NKT cells do also play role in infections of the liver, because NKT- and/or CD1d-decient mice are more susceptible to certain viral39 and bacterial40,41 infections and activation of NKT cells with the synthetic CD1d-ligand -galactosylceramide results in IFN- production and downregulation of HBV replication in a transgenic mouse model.42

T Cell Priming: Where and How?

T cell-mediated protection against liver-trophic viruses and bacteria depends on constant supply of activated effector CD8 T cells to maintain immune responses that control emergence, spread and expansion of the virus. Although many excellent studies have focused on the functional and phenotypic features of pathogen-specic T cells in the blood and the liver, fewer have analyzed the mechanisms by which intrahepatic antigens become available for the induction of T cell-mediated immune responses and the sites where they are presented to CD8 T cells in vivo. Resting, nave CD8 T cells are preferen tially located in secondary lymphoid compartments and require two independent signals to become fully activated. The rst signal is provided by the peptide-MHC I complex through the specic TCR. The second signal (costimulation) is independent of the antigen receptor and is critical to allow full activation and differentiation of CD8 T cells.43 Thus, whereas primed effector CD8 T cells can be activated by any target cell that expresses the cognate antigen in the context of MHC class I, only few, appropriately licensed bone marrow-derived professional antigen-presenting cells have the ability to initiate CD8 T cell responses,44 most likely because they express costimulatory molecules and because they carry antigens from the site of infection into lymphoid organs.43 At this time, no denite evidence is available to answer the question whether direct presentation of antigens can efciently occur in the liver or whether it is conned to the draining lymph nodes. It cannot be excluded that

antigens from liver pathogens expressed in the liver are passively taken up by LSECs and Kupffer cells and then presented to nave T cells that re-circulate through the liver sinusoids or reside in portal tract lymphoid aggregates.44 However, in noninammatory conditions, antigen-presentation by LSEC results in immunological tolerance rather than enhancement of T cell responses.45 For example, LSEC-mediated presentation of antigens to nave CD4 T cells results in secretion of IL4 and IL-10 rather than IL-2 and IFN- in a TCR-transgenic model (Fig. 3).12 The dominance of IL-10 in the liver, which is produced not only by CD4 T cells but also by Kupffer cells, and as recently shown, also by pathogen-specic intrahepatic CD8 T cells,46 alters the expression of chemokine receptors on dendritic cells and downregulates their migration to the draining lymph nodes47 (Fig. 3). If T cells are primed in the presence of IL-10, they are prone to lose cytokine production and effector functions.24 LSEC-mediated presentation of antigens to CD8 T cells does also result in tolerance rather than effector functions.45 CD8 T cells that are cocultured with LSEC exhibit low IL-2 and IFN- production and low cytotoxicity, reduced proliferative capacity and are prone to undergo apoptosis (Fig. 4). This defective effector program can be restored by addition of exogenous IL-2 to the LSEC/CD8 T cell coculture. Under noninammatory conditions and in particular in the absence of IL-2, antigen-presentation by LSEC is therefore thought to contribute to tolerance.12 In the context of inammatory cytokines, however, LSEC downregulate MHC-expression and the tolerogenic effect of LSEC-mediated antigen-presentation decreases. A second important issue is whether those DCs that reside in the hepatic portal tracts can be involved in local antigen presentation, or whether they must rst trafc to the draining lymph nodes. From a traditional point of view, priming of antigen-specic lymphocytes must occur in lymph nodes that drain the sites of infection. Therefore, DCs should pick up antigens from liver-trophic pathogens and trafc to regional lymph nodes where resting, nave lymphocytes are preferentially located. Once activated, antigen-specic lymphocytes then enter the blood stream and home to the liver where they exert their effector functions. From an alternative perspective, however, antigens may be presented in situ by liver-resident bone marrowderived DCs and activate inltrating nave lympho cytes.48,49 Most interestingly, a recent study suggests that activation of nave CD8 T cells within lymph nodes and liver results in qualitatively different effector functions and differential immunopathology.50 In this study, the authors used transgenic mouse models, in which antigen

S58

RACANELLI AND REHERMANN

HEPATOLOGY, February 2006

Fig. 3. Antigen-presentation by liver sinusoidal endothelial cells results in tolerization of CD4 and CD8 T cells. See text for details (adapted from 88).

is expressed either within the liver or within the lymph nodes and demonstrated that adoptively transferred, TCR-transgenic CD8 T cells induce hepatitis when initial T cell activation occurs in the lymph nodes, whereas they do not induce hepatitis and instead exhibit defective cytotoxic function and shortened half-life when primary activation occurs in the liver.50 A third issue is the type of cell that primes nave lym phocytes. When antigen-presenting cells themselves are not infected or when their function is inhibited by pathogens, they cannot process endogenous antigen and direct priming of T cells is absent or inefcient. In this scenario, the only way to initiate a CD8 T cell response is via the cross-priming pathway.51-54 In the crosspriming pathway, an antigen inside one cell is endocytosed by another cell and crosspresented by that cells MHC class I to CD8 T cells. Because it had long been thought that endocytosed antigens could only be presented on MHC class II and thus, only prime CD4 T cells, the terms crosspresentation and crosspriming were chosen to describe the apparent crossing of endocytosed antigens into the MHC class I pathway for priming of CD8 T cells. Crosspriming requires bone-marrow derived antigenpresenting cells, that internalize cell-associated particulate antigens when they phagocytose dying or dead, infected cells.55 Macrophages and, in mice, CD8 DCs are such crosspresenting antigen-presenting cells. The phagocytosed cell can for example be a pathogen-infected hepatocyte (Fig. 4A) or a Kupffer cell that itself has taken up a pathogen (Fig. 4B). After internalizing the respective dying or dead cell, the antigen-presenting cell reprocesses

the internalized antigens and re-routes them to its own MHC class I presentation pathway. How is this achieved? Internalized antigens are initially localized in phagosomes. The phagosomes then fuse with parts of the endoplasmic reticulum (ER) membrane56 and thereby acquire the complete antigen-processing machinery.57-59 Two models have recently been developed to describe the mechanism of antigen reprocessing and presentation on MHC class I. According to the rst model, the phagocytosed antigen is channelled out of the phagosome compartment into the cytosol, where it is processed into short peptides by proteasomes, that are located on the cytosolic side of the phagosome/ER membrane. The resulting peptides are subsequently delivered to MHC class I molecules in either ER or phagosomes via ER-derived transporters associated with antigen processing (TAP).57,58,60-62 The resulting peptide/MHC complexes are then transported through the Golgi apparatus to the cell surface. In contrast, in the alternative TAPindependent model, the phagocytosed antigens are cleaved into peptides directly in the phagosomes by proteases.63-65 Because MHC class I molecules contain a tyrosine-based targeting signal in their cytoplasmic domain that routes them through endolysosomal compartments,66 the generated peptides can be loaded onto MHC class I molecules without an additional processing step in the cytosol. In both models, dying virus-infected cells are not just a passive source of antigen but could play an active role in the quality and specicity of T cell priming through the delivery of pre-processed peptide epitopes. In addition, a

HEPATOLOGY, Vol. 43, No. 2, Suppl. 1, 2006

RACANELLI AND REHERMANN

S59

Fig. 4. Uptake and reprocessing of cell-associated antigens for crosspresentation. CD8 dendritic cells internalize cell-associated particulate antigens when they phagocytose apoptotic cells or cell debris.55 Cell associated antigens are reprocessed and displayed on MHC class I molecules to crossprime na CD8 T cells. The phagocytosed cell can, for example, be a pathogen-infected hepatocyte (A) or a Kupffer cell that itself has ve taken up a pathogen (B).

certain adjuvant effect has been attributed to dying and dead cells,67 which may help to shift the balance from maintenance of immune tolerance to induction of immune responses. The effectiveness of this mechanism depends on the number of infected cells, the viability of these cells, the amount of endogenously expressed antigens and the inammatory environment. Whereas an exuberant inammatory response might bypass some of the costimulatory pathways required for CD8 T cell priming, a limited inammatory response might be too weak to achieve effective presentation of antigens to CD8 T cells. Hepatitis C virus (HCV), for example, causes a clinically inapparent and asymptomatic onset of infection68 and appears to escape from innate and adaptive host immune responses early during the course of infection.69 The clinically asymptomatic onset of infection suggests that liver injury is mild and that few virus-infected hepatocytes are dying. If this is the case, the amount of exogenous antigen that can be acquired and cross-presented is limited. Antigen compartmentalization in hepatocytes rather than antigen-presenting cells, lack of or mild cytotoxicity of the pathogen and poor availability of cell death-associated antigens as crosspriming material may therefore contribute to a prolonged state of CD8 T cell ignorance. This scenario would be consistent with the delayed induction, weakness and poor effector function of HCV-specic T cells that has been observed in the majority of HCVinfected patients.68,70 Information on the extent and impact of cross-presentation of antigens from infected cells

in vivo is therefore likely crucial to understand the overall specicity and quality of CD8 T cell responses to infecting pathogens.

Adaptive Immune Responses in the Liver

B Cell Responses. Not much is known about the function of B cells in the liver. The reasons for this gap in knowledge are the small number of B cells residing in the healthy liver and the experimental difculty in isolating and analyzing those cells. Classically, B cell activation, differentiation, and proliferation occur either in the lymphoid follicles of secondary lymphoid organs, such as regional lymph nodes, tonsils, spleen, and mucosalassociated lymphoid tissues71 or in so-called ectopic germinal centers that are found in nonlymphoid organs namely rheumatoid synovial membrane, thyroid gland, choroid, or lung. The HCV-infected liver may, for example, be considered an ectopic lymphoid organ, in which B cells are stimulated to proliferate and differentiate into antibody-secreting cells within germinal centers of intraportal lymphoid follicles. These intraportal lymphoid follicles display a germinal center-like structure in which activated B cells are surrounded by a follicular DC network. A T cell zone with CD4 T cells and CD8 T cells is detected at the periphery of the nodules.72 The distribution of IgM-, IgD-, and IgG-positive B cells and the expression patterns of Ki-67, CD23, or bcl-2 and bcl-6 gene products in intrahepatic germinal centers resemble those in lymph nodes, suggesting that intrahepatic germi-

S60

RACANELLI AND REHERMANN

HEPATOLOGY, February 2006

nal centers function as functional follicular structures. Molecular analyses of immunoglobulin heavy chain (IgH) gene rearrangements demonstrate that the B cells in these intrahepatic structures are often clonally restricted.73 B cell receptor rearrangement typically occurs in the appropriate reading frame for potentially functional proteins74 and for some patients, the resulting amino acid sequences match, suggesting the presence of common antigens that drive the selection.73 This observation may represent a key feature of host-pathogen interaction, in that the increased ability to bind selected antigens provides certain B cell clones with a growth advantage over those that cannot respond or respond less efciently.75 These ndings corroborate the idea that intrahepatic follicle-like structures are functionally similar to those of lymph nodes with respect to B cell activation, expansion and maturation. T Cell Responses. The immune response to most liver-trophic pathogens is kinetically associated with a strong and durable CD4 and CD8 T cell response. Because peptides from interacellular pathogens are predominantly presented on MHC class I, because all nucleated cells express MHC class I molecules and because CD8 T cells recognize peptides of 8-11 amino acids in the context of MHC class I molecules, effector functions of intrahepatic CD8 T cells have received special attention. These effector mechanisms include the production of cytokines, such as IFN- and TNF- , and cytolytic mechanisms.76 In particular, CD8 T cells exert their cytolytic activity by releasing granule contents such as perforin and granzyme and by triggering Fas-mediated apoptosis.77 Several recent studies provide new insights into the kinetics and effector functions of CD8 T cell responses in the liver. Adoptive transfer of HBV-specic CD8 T cells into either transgenic mice that bear replication-competent copies of HBV in their hepatocytes or into nontransgenic littermates results in rapid recruitment of the transferred CD8 T cells into the liver.78 Consistent with previous reports,79,80 this nding conrms that activated CD8 T cells are recruited to and/or trapped in the liver irrespective of their antigen-specicity. Only upon recognition of their cognate antigen, however, do these CD8 T cells undergo rapid proliferation.78 Proliferation presumably occurs directly in the liver in this scenario, as increased numbers of antigen-specic T cells are not detectable in draining lymph nodes during the early days after adoptive transfer. Most importantly, IFN- production of antigen-specic T cells appears early and then declines, whereas granzyme B expression and cytotoxicity are relatively delayed but sustained. This sequential activation and downregulation of CD8 T cell effector functions (proliferation, IFN- production, cytotoxicity) is

reminiscent of the sequence of early, cytokine-mediated suppression of HBV replication and late appearance of liver injury, which have been described in chimpanzees with acute hepatitis B.81 Downregulation of IFN- production coincides with the upregulation of PD-1 on CD8 T cells, a receptor that is known to modulate TCR signaling82 resulting in inhibition of cytokine production and proliferation in the mouse model of acute hepatitis B. One of the PD-1 ligands, PD-L1 is constitutively expressed on LSEC and Kupffer cells which suggests that these liver-resident antigen-presenting cells actively downregulate specic effector functions of activated liver-inltrating T cells. Finally, acute liver injury becomes most evident at the time of nonspecic, chemokine-mediated amplication of the intrahepatic inltrate.83,84 A key step in this nonspecic amplication of the inammatory inltrate is the recruitment of neutrophils. Neutrophils are recruited by adhesion molecules, cytokines and chemokines, many of which are produced by activated Kupffer cells.29 Neutrophils themselves then secrete soluble factors that contribute the recruitment of additional antigen-nonspecic mononuclear cells. Importantly, this amplication of the inammatory inltrate can be reduced by inhibition of neutrophil-derived matrix metalloproteinases, by neutralization of chemokines and by inactivation of macrophages.85 Whereas noncytolytic downregulation of HBV replication by HBV-specic CD8 T cells is not affected, liver injury can be prevented by these measures.84,85

Concluding Remarks
As our understanding of the mechanisms of antigen capture, presentation, and recognition in the liver increases, the biological mechanisms of the so called tolerogenic liver effect are gradually uncovered and the role of the liver as a major organ of the innate and adaptive immune system is increasingly recognized. Questions that remain to be answered are whether and how the liver effect can be manipulated to modify the natural history of important viral, parasitic, autoimmune, and malignant liver diseases.

References
1. Sheth K, Bankey P. The liver as an immune organ. Curr Opin Crit Care 2001;7:99-104. 2. Wick MJ, Leithauser F, Reimann J. The hepatic immune system. Crit Rev Immunol 2002;22:47-103. 3. Wisse E. An electron microscopic study of the fenestrated endothelial lining of rat liver sinusoids. J Ultrastruct Res 1970;31:125-150. 4. Emoto M, Kaufmann SH. Liver NKT cells: an account of heterogeneity. Trends Immunol 2003;24:364-369. 5. Exley M, Koziel M. To be or not to be NKT: Natural killer T cells in the liver. HEPATOLOGY 2004;40:1033-1040.

HEPATOLOGY, Vol. 43, No. 2, Suppl. 1, 2006

RACANELLI AND REHERMANN

S61

6. Emoto M, Miyamoto M, Namba K, Schmits R, Van Rooijen N, Kita E, et al. Participation of leukocyte function-associated antigen-1 and NK cells in the homing of thymic CD8 NKT cells to the liver. Eur J Immunol 2000;30:3049-3056. 7. Welsh RM, Lin MY, Lohman BL, Varga SM, Zarozinski CC, Selin LK. Alpha beta and gamma delta T-cell networks and their roles in natural resistance to viral infections. Immunol Rev 1997;159:79-93. 8. Bertolino P, Trescol-Biemont MC, Rabourdin-Combe C. Hepatocytes induce functional activation of naive CD8 T lymphocytes but fail to promote survival. Eur J Immunol 1998;28:221-236. 9. Mackay IR. Hepatoimmunology: a perspective. Immunol Cell Biol 2002; 80:36-44. 10. Gale RP, Sparkes RS, Golde DW. Bone marrow origin of hepatic macrophages (Kupffer cells) in humans. Science 1978;201:937-938. 11. MacPhee PJ, Schmidt EE, Groom AC. Intermittence of blood ow in liver sinusoids, studied by high- resolution in vivo microscopy. Am J Physiol 1995;269:G692-G698. 12. Knolle PA, Schmitt E, Jin S, Germann T, Duchmann R, Hegenbarth S, et al. Induction of cytokine production in naive CD4( ) T cells by antigenpresenting murine liver sinusoidal endothelial cells but failure to induce differentiation toward Th1 cells. Gastroenterology 1999;116:1428-1440. 13. Lohse AW, Knolle PA, Bilo K, Uhrig A, Waldmann C, Ibe M, et al. Antigen-presenting function and B7 expression of murine sinusoidal endothelial cells and Kupffer cells. Gastroenterology 1996;110:1175-1181. 14. Steffan AM, Gendrault JL, McCuskey RS, McCuskey PA, Kirn A. Phagocytosis, an unrecognized property of murine endothelial liver cells. HEPATOLOGY 1986;6:830-836. 15. Valdivia LA, Demetris AJ, Langer AM, Celli S, Fung JJ, Starzl TE. Dendritic cell replacement in long-surviving liver and cardiac xenografts. Transplantation 1993;56:482-484. 16. Prickett TC, McKenzie JL, Hart DN. Characterization of interstitial dendritic cells in human liver. Transplantation 1988;46:754-761. 17. Lau AH, Thomson AW. Dendritic cells and immune regulation in the liver. Gut 2003;52:307-314. 18. Khanna A, Morelli AE, Zhong C, Takayama T, Lu L, Thomson AW. Effects of liver-derived dendritic cell progenitors on Th1- and Th2-like cytokine responses in vitro and in vivo. J Immunol 2000;164:1346-1354. 19. Buelens C, Verhasselt V, De Groote D, Thielemans K, Goldman M, Willems F. Human dendritic cell responses to lipopolysaccharide and CD40 ligation are differentially regulated by interleukin-10. Eur J Immunol 1997;27:1848-1852. 20. Probst HC, McCoy K, Okazaki T, Honjo T, van den Broek M. Resting dendritic cells induce peripheral CD8 T cell tolerance through PD-1 and CTLA-4. Nat Immunol 2005;6:280-286. 21. Matsuno K, Ezaki T, Kudo S, Uehara Y. A life stage of particle-laden rat dendritic cells in vivo: their terminal division, active phagocytosis, and translocation from the liver to the draining lymph. J Exp Med 1996;183: 1865-1878. 22. Kudo S, Matsuno K, Ezaki T, Ogawa M. A novel migration pathway for rat dendritic cells from the blood: hepatic sinusoids-lymph translocation. J Exp Med 1997;185:777-784. 23. Knolle P, Schlaak J, Uhrig A, Kempf P, Meyer zum Buschenfelde KH, Gerken G. Human Kupffer cells secrete IL-10 in response to lipopolysaccharide (LPS) challenge. J Hepatol 1995;22:226-229. 24. Groux H, Bigler M, de Vries JE, Roncarolo MG. Interleukin-10 induces a long-term antigen-specic anergic state in human CD4 T cells. J Exp Med 1996;184:19-29. 25. Roland CR, Walp L, Stack RM, Flye MW. Outcome of Kupffer cell antigen presentation to a cloned murine Th1 lymphocyte depends on the inducibility of nitric oxide synthase by IFN-gamma. J Immunol 1994;153: 5453-5464. 26. Yu S, Nakafusa Y, Flye MW. Portal vein administration of donor cells promotes peripheral allospecic hyporesponsiveness and graft tolerance. Surgery 1994;116:229-234. 27. Roland CR, Mangino MJ, Duffy BF, Flye MW. Lymphocyte suppression by Kupffer cells prevents portal venous tolerance induction: a study of

28.

29. 30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

macrophage function after intravenous gadolinium. Transplantation 1993;55:1151-1158. Lauwerys BR, Garot N, Renauld JC, Houssiau FA. Cytokine production and killer activity of NK/T-NK cells derived with IL-2, IL-15, or the combination of IL-12 and IL-18. J Immunol 2000;165:1847-1853. Gregory SH, Wing EJ. Neutrophil-Kupffer-cell interaction in host defenses to systemic infections. Immunol Today 1998;19:507-510. Moretta L, Ciccone E, Mingari MC, Biassoni R, Moretta A. Human natural killer cells: origin, clonality, specicity, receptors. Adv Immunol 1994;55:341-358. Salazar-Mather TP, Orange JS, Biron CA. Early murine cytomegalovirus (MCMV) infection induces liver natural killer (NK) cell inammation and protection through macrophage inammatory protein 1alpha (MIP-1alpha)-dependent pathways. J Exp Med 1998;187:1-14. Itoh Y, Morita A, Nishioji K, Fujii H, Nakamura H, Kirishima T, et al. Time course prole and cell-type-specic production of monokine induced by interferon-gamma in Concanavalin A-induced hepatic injury in mice: comparative study with interferon-inducible protein-10. Scand J Gastroenterol 2001;36:1344-1351. Bendelac A, Lantz O, Quimby ME, Yewdell JW, Bennink JR, Brutkiewicz RR. CD1 recognition by mouse NK1 T lymphocytes. Science 1995; 268:863-865. Carnaud C, Lee D, Donnars O, Park SH, Beavis A, Koezuka Y, et al. Cutting edge: Cross-talk between cells of the innate immune system: NKT cells rapidly activate NK cells. J Immunol 1999;163:4647-4650. Bendelac A, Rivera MN, Park SH, Roark JH. Mouse CD1-specic NK1 T cells: development, specicity, and function. Annu Rev Immunol 1997; 15:535-562. Kumagai K, Takeda K, Hashimoto W, Seki S, Ogasawara K, Anzai R, et al. Interleukin-12 as an inducer of cytotoxic effectors in anti-tumor immunity. Int Rev Immunol 1997;14:229-256. Toyabe S, Seki S, Iiai T, Takeda K, Shirai K, Watanabe H, et al. Requirement of IL-4 and liver NK1 T cells for concanavalin A-induced hepatic injury in mice. J Immunol 1997;159:1537-1542. Exley M, Porcelli S, Furman M, Garcia J, Balk S. CD161 (NKR-P1A) costimulation of CD1d-dependent activation of human T cells expressing invariant V alpha 24 J alpha Q T cell receptor alpha chains. J Exp Med 1998;188:867-876. Grubor-Bauk B, Simmons A, Mayrhofer G, Speck PG. Impaired clearance of herpes simplex virus type 1 from mice lacking CD1d or NKT cells expressing the semivariant V alpha 14-J alpha 281 TCR. J Immunol 2003; 170:1430-1434. Behar SM, Dascher CC, Grusby MJ, Wang CR, Brenner MB. Susceptibility of mice decient in CD1D or TAP1 to infection with Mycobacterium tuberculosis. J Exp Med 1999;189:1973-1980. Kumar H, Belperron A, Barthold SW, Bockenstedt LK. Cutting edge: CD1d deciency impairs murine host defense against the spirochete, Borrelia burgdorferi. J Immunol 2000;165:4797-4801. Kakimi K, Guidotti LG, Koezuka Y, Chisari FV. Natural killer T cell activation inhibits hepatitis B virus replication In vivo. J Exp Med 2000; 192:921-930. Gonzalo JA, Delaney T, Corcoran J, Goodearl A, Gutierrez-Ramos JC, Coyle AJ. Cutting edge: the related molecules CD28 and inducible costimulator deliver both unique and complementary signals required for optimal T cell activation. J Immunol 2001;166:1-5. Bertolino P, McCaughan G, Bowen DG. Role of primary intrahepatic T cell activation in the liver tolerance effect. Immunol Cell Biol 2002;80:8492. Limmer A, Ohl J, Kurts C, Ljunggren HG, Reiss Y, Groettrup M, et al. Efcient presentation of exogenous antigen by liver endothelial cells to CD8 T cells results in antigen-specic T-cell tolerance. Nat Med 2000; 6:1348-1354. Accapezzato D, Francavilla V, Paroli M, Casciaro M, Chircu LV, Cividini A, et al. Hepatic expansion of a virus-specic regulatory CD8( ) T cell population in chronic hepatitis C virus infection. J Clin Invest 2004;113: 963-972.

S62

RACANELLI AND REHERMANN

HEPATOLOGY, February 2006

47. Takayama T, Morelli AE, Onai N, Hirao M, Matsushima K, Tahara H, et al. Mammalian and viral IL-10 enhance C-C chemokine receptor 5 but down-regulate C-C chemokine receptor 7 expression by myeloid dendritic cells: impact on chemotactic responses and in vivo homing ability. J Immunol 2001;166:7136-7143. 48. Bertolino P, Bowen DG, McCaughan GW, Fazekas De St Groth B. Antigen-specic primary activation of CD8( ) T cells within the liver. J Immunol 2001;166:5430-5438. 49. Bowen DG, Warren A, Davis T, Hoffmann MW, McCaughan GW, Fazekas de St Groth B, et al. Cytokine-dependent bystander hepatitis due to intrahepatic murine CD8 T-cell activation by bone marrow-derived cells. Gastroenterology 2002;123:1252-1264. 50. Bowen DG, Zen M, Holz L, Davis T, McCaughan GW, Bertolino P. The site of primary T cell activation is a determinant of the balance between intrahepatic tolerance and immunity. J Clin Invest 2004;114:701-712. 51. Bevan MJ. Cross-priming for a secondary cytotoxic response to minor H antigens with H-2 congenic cells which do not cross-react in the cytotoxic assay. J Exp Med 1976;143:1283-1288. 52. Bevan MJ. Minor H antigens introduced on H-2 different stimulating cells cross-react at the cytotoxic T cell level during in vivo priming. J Immunol 1976;117:2233-2238. 53. Heath WR, Carbone FR. Cross-presentation, dendritic cells, tolerance and immunity. Annu Rev Immunol 2001;19:47-64. 54. Yewdell JW, Norbury CC, Bennink JR. Mechanisms of exogenous antigen presentation by MHC class I molecules in vitro and in vivo: implications for generating CD8 T cell responses to infectious agents, tumors, transplants, and vaccines. Adv Immunol 1999;73:1-77. 55. den Haan JM, Lehar SM, Bevan MJ. CD8( ) but not CD8( ) dendritic cells cross-prime cytotoxic T cells in vivo. J Exp Med 2000;192:1685-1696. 56. Gagnon E, Duclos S, Rondeau C, Chevet E, Cameron PH, Steele-Mortimer O, et al. Endoplasmic reticulum-mediated phagocytosis is a mechanism of entry into macrophages. Cell 2002;110:119-131. 57. Guermonprez P, Saveanu L, Kleijmeer M, Davoust J, Van Endert P, Amigorena S. ER-phagosome fusion denes an MHC class I cross-presentation compartment in dendritic cells. Nature 2003;425:397-402. 58. Houde M, Bertholet S, Gagnon E, Brunet S, Goyette G, Laplante A, et al. Phagosomes are competent organelles for antigen cross-presentation. Nature 2003;425:402-406. 59. Ackerman AL, Kyritsis C, Tampe R, Cresswell P. Early phagosomes in dendritic cells form a cellular compartment sufcient for cross presentation of exogenous antigens. Proc Natl Acad Sci U S A 2003;100:12889-12894. 60. Kovacsovics-Bankowski M, Rock KL. A phagosome-to-cytosol pathway for exogenous antigens presented on MHC class I molecules. Science 1995; 267:243-246. 61. Huang AY, Bruce AT, Pardoll DM, Levitsky HI. In vivo cross-priming of MHC class I-restricted antigens requires the TAP transporter. Immunity 1996;4:349-355. 62. Rodriguez A, Regnault A, Kleijmeer M, Ricciardi-Castagnoli P, Amigorena S. Selective transport of internalized antigens to the cytosol for MHC class I presentation in dendritic cells. Nat Cell Biol 1999;1:362-368. 63. Pfeifer JD, Wick MJ, Roberts RL, Findlay K, Normark SJ, Harding CV. Phagocytic processing of bacterial antigens for class I MHC presentation to T cells. Nature 1993;361:359-362. 64. Rock KL. A new foreign policy: MHC class I molecules monitor the outside world. Immunol Today 1996;17:131-137. 65. Albert ML. Death-defying immunity: do apoptotic cells inuence antigen processing and presentation? Nat Rev Immunol 2004;4:223-321. 66. Lizee G, Basha G, Tiong J, Julien JP, Tian M, Biron KE, Jefferies WA. Control of dendritic cell cross-presentation by the major histocompatibility complex class I cytoplasmic domain. Nat Immunol 2003;4:1065-1073.

67. Shi Y, Evans JE, Rock KL. Molecular identication of a danger signal that alerts the immune system to dying cells. Nature 2003;425:516-521. 68. Heller T, Rehermann B. Acute hepatitis C: a multifaceted disease. Semin Liver Dis 2005;25:7-17. 69. Racanelli V, Rehermann B. Hepatitis C virus infection: when silence is deception. Trends Immunol 2003;24:456-464. 70. Thimme R, Bukh J, Spangenberg HC, Wieland S, Pemberton J, Steiger C, et al. Viral and immunological determinants of hepatitis C virus clearance, persistence, and disease. Proc Natl Acad Sci U S A 2002;99:15661-15668. 71. Burrows PD, Cooper MD. B cell development and differentiation. Curr Opin Immunol 1997;9:239-244. 72. Murakami J, Shimizu Y, Kashii Y, Kato T, Minemura M, Okada K, et al. Functional B-cell response in intrahepatic lymphoid follicles in chronic hepatitis C. HEPATOLOGY 1999;30:143-150. 73. Racanelli V, Sansonno D, Piccoli C, DAmore FP, Tucci FA, Dammacco F. Molecular characterization of B cell clonal expansions in the liver of chronically hepatitis C virus-infected patients. J Immunol 2001;167:2129. 74. De Re V, De Vita S, Gasparotto D, Marzotto A, Carbone A, Ferraccioli G, et al. Salivary gland B cell lymphoproliferative disorders in Sjogrens syndrome present a restricted use of antigen receptor gene segments similar to those used by hepatitis C virus-associated non-Hodgkinss lymphomas. Eur J Immunol 2002;32:903-910. 75. Coronella JA, Spier C, Welch M, Trevor KT, Stopeck AT, Villar H, et al. Antigen-driven oligoclonal expansion of tumor-inltrating B cells in inltrating ductal carcinoma of the breast. J Immunol 2002;169:1829-1836. 76. Guidotti LG, Chisari FV. To kill or to cure: options in host defense against viral infection. Curr Opin Immunol 1996;8:478-483. 77. Berke G. The CTLs kiss of death. Cell 1995;81:9-12. 78. Isogawa M, Furuichi Y, Chisari FV. Oscillating CD8( ) T cell effector functions after antigen recognition in the liver. Immunity 2005;23:53-63. 79. Huang L, Soldevila G, Leeker M, Flavell R, Crispe IN. The liver eliminates T cells undergoing antigen-triggered apoptosis in vivo. Immunity 1994;1: 741-749. 80. Mehal WZ, Juedes AE, Crispe IN. Selective retention of activated CD8 T cells by the normal liver. J Immunol 1999;163:3202-3210. 81. Guidotti LG, Rochford R, Chung J, Shapiro M, Purcell R, Chisari FV. Viral clearance without destruction of infected cells during acute HBV infection. Science 1999;284:825-829. 82. Latchman Y, Wood CR, Chernova T, Chaudhary D, Borde M, Chernova I, et al. PD-L2 is a second ligand for PD-1 and inhibits T cell activation. Nat Immunol 2001;2:261-268. 83. Ando K, Moriyama T, Guidotti LG, Wirth S, Schreiber RD, Schlicht HJ, et al. Mechanisms of class I restricted immunopathology. A transgenic mouse model of fulminant hepatitis. J Exp Med 1993;178:1541-1554. 84. Kakimi K, Lane TE, Wieland S, Asensio VC, Campbell IL, Chisari FV, et al. Blocking chemokine responsive to gamma-2/interferon (IFN)-gamma inducible protein and monokine induced by IFN-gamma activity in vivo reduces the pathogenetic but not the antiviral potential of hepatitis B virus-specic cytotoxic T lymphocytes. J Exp Med 2001;194:1755-1766. 85. Sitia G, Isogawa M, Iannacone M, Campbell IL, Chisari FV, Guidotti LG. MMPs are required for recruitment of antigen-nonspecic mononuclear cells into the liver by CTLs. J Clin Invest 2004;113:1158-1167. 86. Kuehnel W. Color Atlas of Cytology, Histology and Microscopic Anatomy. Thieme, 2003. 87. Fawcett D, Jensh R. Bloom & Fawcetts Concise Histology. Arnold, 2002. 88. Crispe IN. Hepatic T cells and liver tolerance. Nat Rev Immunol 2003;3: 51-62.