T cell immunity and graft-versus-host disease (GVHD)

Autoimmunity Reviews 5 (2006) 1 – 9 www.elsevier.com/locate/autrev

T cell immunity and graft-versus-host disease (GVHD) Yasunori Ichikia, Christopher L. Bowlusb, Shinji Shimodac, Hiromi Ishibashid, John M. Vierlinge, M. Eric Gershwina,T a

Division of Rheumatology, Allergy and Clinical Immunology, University of California at Davis School of Medicine, TB192, One Shields Avenue, Davis, CA 95616, USA b Division of Gastroenterology, University of California at Davis Medical Center, Sacramento, CA 95817, USA c Medicine and Biosystemic Science, Kyushu University Graduate School of Medical Sciences, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan d Clinical Research Center, Nagasaki Medical Center and Department of Hepatology, Nagasaki University Graduate School of Biomedical Sciences, Kubara 2-1001-1, Omura, Nagasaki 856-8562, Japan e Center for Liver Diseases and Transplantation, Cedars-Sinai Medical Center and David Geffen School of Medicine at UCLA, Los Angeles, CA 90048-6110, USA Received 19 August 2004; accepted 28 February 2005 Available online 21 April 2005

Abstract Graft-versus-host disease (GVHD), induced by the reaction of donor T cells to recipient histoincompatible antigens, is a serious complication of allogeneic bone marrow transplantation (BMT), resulting in considerable morbidity and mortality. In MHC-disparate BMT, donor T cells directly react with major histocompatibility complex (MHC) antigens, while in MHC-matched BMT, T cells react with minor histocompatibility antigens (miHA) presented by shared MHC molecules. Clinically, acute and chronic GVHD can be distinguished on the basis of the time of onset, clinical manifestations and distinct pathobiological mechanisms. Acute GVHD usually occurs within 2 to 6 weeks following allogeneic BMT and primarily affects the skin, liver and the gastrointestinal tract with T cell infiltration of the epithelia of the skin, mucous membranes, bile ducts and gut. In addition, hair follicle cells, airways, bone marrow, and a variety of other tissue systems can be involved. Acute GVHD occurs in up to 50% of allogeneic HLA-matched and 70% of HLA-disparate BMT recipients despite prophylactic immunosuppressive drugs. Chronic GVHD involves a wider range

Abbreviations: aa, amino acids; APC, antigen-presenting cells; BMT, bone marrow transplantation; CTL, cytotoxic T lymphocytes; DC, dendritic cells; GVHD, graft-versus-host disease; Th1 cells, type 1 helper T cells; Th2 cells, type 2 helper T cells; ICAM-1, intercellular adhesion molecule-1; IL, interleukin; IFN-g, interferon-g; LFA-1, lymphocyte function-associated antigen-1; MHC, major histocompatibility complex; MIG, monokine induced by IFN-g, NKT cells, natural killer T cells (NK1.1+ in mice); PBMC, peripheral blood mononuclear cells; TNF-a, tumor necrosis factor-a; VCAM-1, vascular adhesion molecule-1; VLA-4, very late antigen-4; TCR, T cell receptor. T Corresponding author. Tel.: +1 530 752 2884; fax: +1 530 752 4669. E-mail address: [email protected] (M.E. Gershwin). 1568-9972/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.autrev.2005.02.006

2

Y. Ichiki et al. / Autoimmunity Reviews 5 (2006) 1–9

of organs and clinical manifestations include scleroderma, liver failure, immune complex disease, glomerulonephritis, and autoantibody formation. D 2005 Elsevier B.V. All rights reserved. Keywords: Graft-versus-host disease; Autoantibodies; Allogeneic bone marrow transplantation; Major histocompatibility complex

Contents 1. Introduction . . . . . . . . . . . . 2. Immunopathology in acute GVHD 3. Regulatory T cells . . . . . . . . . 4. MiHA and T cell epitopes. . . . . 5. B7-CD28/CTLA4 . . . . . . . . . 6. Antigen-presenting cells (APC) . . 7. Cytokines and chemokines . . . . 8. Cytotoxic effector pathways. . . . 9. Adhesion molecules . . . . . . . . 10. Conclusion . . . . . . . . . . . . Take-home messages . . . . . . . . . . References . . . . . . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

1. Introduction GVHD lesions are mediated by infiltrating CD3+ T cells, principally of the CD8+ subset [1] (Fig. 1). Epithelial cell targets aberrantly express MHC class II molecules and expression of intercellular adhesion molecule (ICAM)-1 and vascular cell adhesion molecule (VCAM)-1, are

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

2 3 3 4 4 4 5 6 7 7 8 8

upregulated in cutaneous and intestinal lesions of GVHD patients, but rarely in hepatic lesions [2]. In contrast, increased expression of biliary ICAM-1 and, to a lesser extent, VCAM-1, are observed in primary biliary cirrhosis (PBC), which is intriguing since the histopathology of hepatic GVHD and PBC are indistinguishable by light and electron microscopy.

BEC

endothelial cell

macrophage

lymphocyte

MIP-1α

MIP-1α

MIP-1α

MIP-1α

CCR5 CD8+ T cell

Fig. 1. CD8+ T cell migration in hepatic GVHD.

Y. Ichiki et al. / Autoimmunity Reviews 5 (2006) 1–9

3

2. Immunopathology in acute GVHD

3. Regulatory T cells

Acute GVHD can be conceptualized as a three-stage process consisting of (1) initiation of tissue damage, (2) activation and proliferation of donor T cells, and (3) the effector phase, resulting in tissue damage [3] (Table 1). Initiation of tissue damage, or conditioning, consists of irradiation and/or chemotherapy which eliminates recipient T cells that could prevent engraftment and causes tissue injury in the liver and intestine. Conditioning promotes secretion of proinflammatory cytokines and growth factors, such as TNF-a, IFN-g, IL-1, and granulocyte-macrophage colony-stimulating factor that can induce the expression of adhesion and MHC molecules on recipient cells, facilitating recognition of alloantigens by donor T cells. In the second phase, antigen presentation to donor T cells results in their activation, proliferation and differentiation. Several lines of evidence attest to the importance of activated donor T cells in GVHD including (1) a dose–response relationship between the number of transplanted T cells and the severity of GVHD in murine models; (2) reduced incidence and severity of GVHD after depletion of donor T cells from the graft or in the recipient; (3) inflammatory infiltrates composed of donor T cells in GVHD lesions of humans and mice. Although depletion of T cells before grafting reduces GVHD, it is now rarely performed because of its deleterious effects, which include an increased rate of graft failure, prolonged need for immunosuppression, and poor graft-versusleukemia responses.

T cells with regulatory functions in GVHD include natural killer T cells (NKT, NK1.1+ in mice) and CD4+CD25+ cells. NKT cells are CD3+ T cells restricted by the non-polymorphic, non-classical MHC class I-like CD1 molecules that secrete high levels of both IFN-g and interleukin (IL)-4. In mice, these cells represent b 1% of peripheral T cells, but comprise ~30% of bone marrow T cells. The addition of CD4+/CD8+ T cells containing NK1.1+ T cells to T-cell-depleted bone marrow results in only minimal signs of GVHD and completely prevents mortality in MHC-mismatched, lethally irradiated recipients. In contrast, grafts depleted of NK1.1+ T cells produce typical GVHD with a 70% mortality within 20 days. Failure of NK1.1+ T cells from IL-4 / donors to protect against GVHD indicated that NK1.1+ T cell regulation of GVHD is IL-4-dependent. Compared to recipients of total body irradiation, the spleens of recipients of only lymphoid irradiation contain an increased proportion of NK1.1+TCRah+ T cells. BMT supplemented with peripheral blood mononuclear cells (PBMC) into lymphoid irradiated recipients resulted in significantly improved survival and reduced quantities of donor T cells in blood, gut, and liver. These results suggest that both donor and recipient regulatory NK1.1+ T cells can protect against GVHD. A potential role for host-derived NKT cells (NK1.1+) in hepatic lesions of chronic, but not acute GVHD, was suggested by evidence of hepatic expansion of this subset of T cells in the (C57BL/ 6  DBA/2) F1 hybrid mouse model of chronic GVHD [4]. In contrast, hepatic NK1.1+ T cells were depleted in a F1 model of acute GVHD. CD4+ T cells that constitutively express the IL-2 receptor a-chain (CD25) play central roles in selftolerance and can regulate T cells mediating GVHD. The presence or addition of CD4+CD25+ T cells can induce antigen-specific tolerance to bone marrow grafts [5] while the absence of CD4+CD25+ T cells in grafts accelerates GVHD. The mechanisms of GVHD inhibition appear to include increased secretion of IL-10 [6], and inhibition of donor T cell expansion associated with expression of IL-2 receptor a-chains [7].

Table 1 Three steps leading to acute GVHD Step role

Factors

Step 1 initiating tissue damage

Conditioning regimen (radiation, chemotherapy) YHost cell damage YInflammatory cytokines YUpregulation of adhesion molecules and MHC molecules APC necessary, via MHC-TCR interaction CD4+ T cells, CD8+ CTL, NK cells TNF-a, Fas/FasL pathway, perforin/granzyme pathway Th1/Th2 cytokines

Step 2 donor T cell activation and proliferation Step 3 effector phase

4

Y. Ichiki et al. / Autoimmunity Reviews 5 (2006) 1–9

4. MiHA and T cell epitopes MiHA are polymorphic protein antigens, which can be presented to donor T cells in MHC class I- or class II-restricted manner [8]. Although estimates of more than 40 miHA differences between two strains of mice have been reported, the number of differences in humans is unknown. Despite an array of possible miHA, allogeneic reactions appear to be restricted to a limited number of immunodominant miHA. This results in preferential activation of T cell receptors (TCRs) expressing a limited number of Vh-associated families [9]. Even a single dominant miHA can induce GVHD, as shown by development of GVHD following donor T cell activation with a single immunodominant CD8+ T cell epitope. In a model of GVHD using grafts of BM and spleen cells from B10.D2 parents into MHC-identical, but miHA-disparate (DBA/2  B10.D2) F1 recipients, restricted Vh repertoire usage implied clonal or oligoclonal expansion, yet differences in the Jh segments associated with the Vh segment strongly suggested that donor and recipients had distinct immunodominant antigens [10]. To date, few human miHA have been characterized molecularly, and most represent gene polymorphisms resulting in immunogenic amino acid differences between the host and the donor proteins [8]. However, homozygous gene deletions can also serve as miHA, as recently shown for human UGT2B17, an autosomal gene in the UDP-glycosyltransferase 2 family [11]. Interestingly, mRNA expression of UGT2B17 was detected primarily in liver, colon, small intestine, and pancreas, which represent target tissues in GVHD. Other evidence suggests that tissue-specific antigen expression may play a role in the preferential targeting of epithelia of skin, liver, and gut in GVHD. Certain Vh families are significantly overrepresented in CD4+ and CD8+ T cells infiltrating the liver and other target organs. Different patterns of Va and/or Vh skewing between PBMC and target organs as well as between the different target organs themselves have been described in both mice [12] and humans [13]. Biased usage of TCR Vh families was characteristic of GVHD lesions, but absent in biopsies without evidence of GVHD. Distinct tissue patterns of miHA expression in healthy mice were altered qualitatively and quantitatively after inductin of GVHD, partic-

ularly the target organs of GVHD [14]. Thus, tissuespecific differences in antigen may account for the restricted targeting of the skin, liver, and gut in GVHD.

5. B7-CD28/CTLA4 T cell activation requires not only engagement of the MHC/peptide complex but also additional costimulatory signals. B7 molecules (B7-1, B7-2, and possibly B7-3) on APCs serve as costimulatory molecules and bind CD28 and CTLA4 on naRve and activated T cells, respectively. B7 binding to CD28 costimulates T cells, but the effect of B7 binding to CTLA-4 includes downregulation of T cell functions. Administration of anti-B7 mAb (anti-B7.1 + antiB7.2) or CTLA4-Ig, a soluble fusion protein that competitively blocks B7 binding to both CD28 and CTLA-4, suppresses GVHD [15], but the extent varies among models. Grafts of CD28 / T cells into either unirradiated H2-incompatible or irradiated MHC class I- or class II-mismatched recipients reduced GVHD severity and mortality. In recipients of CD28 / donor T cells, CTLA4-Ig paradoxically accelerated GVHD mortality, suggesting that beneficial effects of CTLA4-Ig on GVHD resulted from inhibition of B7:CD28 signaling, while B7:CTLA4 signaling may be protective. Anti-CD28 mAbs reduces GVHD, mortality more effectively than CTLA4-Ig. In contrast to the beneficial effect of blocking costimulation mediated by CD28 or B7, anti-CTLA-4 mAbs abrogated the protective regulatory function of CTLA-4, accelerating GVHD mortality in a CD28dependent manner [15]. A later report indicated that anti-CD28 mAbs prevented GVHD by selectively depleting donor T cells rather than inhibiting CD-28 [16].

6. Antigen-presenting cells (APC) Both host APC and cross-presenting donor APC are involved in GVHD. Chimeric mice whose hematopoietic cells were unable to present MHC class I-restricted peptides, but whose target tissues expressed MHC class I, did not develop GVHD after grafting with T-cell-depleted bone marrow and CD8+

Y. Ichiki et al. / Autoimmunity Reviews 5 (2006) 1–9

T cells. Presumably, exogenous antigen processing and presentation by donor APCs was insufficient to activate alloreactive T cells. Conversely, host APCs, specifically host dendritic cells (DC), but not B cells, are both obligatory and sufficient for activation of GVHD effector T cells [17]. Conditioning irradiation followed immediately by BMT resulted in rapid activation and maturation of host DC, which were in close proximity to donor CD8+ T cells expressing activation markers in the spleen. Subsequently, rapid elimination of host DCs from spleen, lymph nodes, blood, liver and skin resulted in undetectable quantities 5 days after BMT. Slower kinetics of elimination of up to 4 weeks has also been observed [18]. In the CD4+ T cell-dependent MHC class IIdisparate B6YB6.MHC II / or B6Ybm12 chimera models, alloantigen presentation by host APCs was sufficient for T cell activation and generation of GVHD lesions, indicating that antigen presentation by epithelial cells was unnecessary. In a CD8+ T-cellmediated model of GVHD in which BM plus T cells from bm1 donors are engrafted in B6Ybm1 chimeras, GVHD mortality was significantly lower in recipients lacking MHC class I alloantigens on target epithelia, even though the severity of tissue damage was not reduced. In the CD4+ T cell-dependent, miHA-disparate C57BL/6YBALB.B model, nonhematopoieticallyderived alloantigens were crucial for the development of GVHD after transplantation of T-cell-depleted bone marrow plus CD4+ T cells [19]. Similarly, CD8+mediated GVHD was significantly reduced in chimeras in which expression of recipient miHA was restricted to hematopoietic cells [20]. A lower precursor frequency of donor T cells specific for miHA compared to those specific for MHC antigens may result in marked differences in quantities of circulating proinflammatory cytokines, which are required for GVHD and likely contribute to the lack of a requirement for nonhematopoietic antigen presentation in the MHC-disparate model [21]. Expression of alloantigen by target cells suggests that antigen-specific processes are involved in both activation and effector phases of GVHD. Indeed, lipoclodronate depletion of hepatic and splenic macrophages and DC in a tissue-specific manner before irradiation and grafting of donor CD8+ T cells (with or

5

without T-cell-depleted bone marrow) markedly reduced donor T cell infiltration of the liver and spleen. Donor T cell proliferation was inhibited in the spleen, but not in the liver. Interestingly, donor T cells within the liver had undergone at least four divisions in both lipo-clodronate-treated animals and controls. The conclusions that splenic APC are necessary for activation and proliferation of donor T cells, whereas hepatic macrophages and DC mediate recruitment of activated allogeneic T cells were confirmed using adoptive transfer of in vivo activated donor CD8+ T cells into lipo-clodronated-treated animals. Lipoclodronate-treated animals exhibited lesser portal inflammation, lower ALT levels and significantly improved survival.

7. Cytokines and chemokines Among cytokines produced by activated T cells, both IL-2 and IFN-g play key roles in GVHD. Indeed, grafting of type 2-polarized T cell populations significantly reduced GVHD severity and mortality, suggesting a pivotal role for type 1 T cells in GVHD. However, grafting of IFN-g-deficient donor cells does produce GVHD with atypical pathology and with accelerated mortality [22]. Absence of IFN-g exhibited dichotomous effects in a model of sublethal irradiation and subsequent hematopoietic failure compared to lethal irradiation and donor–anti-host CTL-mediated solid organ destruction. Improvement was seen in the former, but deterioration in the latter. Conversely, absence of IL-4 production by donor cells protected from GVHD in murine models of chronic GVHD with MHC or miHA mismatches. Treatment of donor mice with G-CSF polarized both donor and host CD4+ T cells towards type 2 cytokine production and reduced GVHD mortality [23]. In signal transducer and activator of transcription (STAT)4 and STAT6 knockout mice, both type 1 and type 2 cytokines play roles in GVHD, but differentially effect specific target organs [24]. STAT4 and STAT6 transcription factors are required for responses to IL-12 and IL-4/IL-13. In STAT4 / mice, Th1 responses are reduced while Th2 responses are enhanced; conversely, the opposite pattern is observed in STAT6 / mice. Recipients of either STAT4 / or STAT6 / grafts exhibited delayed GVHD mortality.

6

Y. Ichiki et al. / Autoimmunity Reviews 5 (2006) 1–9

STAT4 / BMT produced severe hepatic and cutaneous GVHD, but only mild intestinal inflammation. In contrast, STAT6 / BMT recipients had severe intestinal GVHD, mild skin lesions and no hepatic lesions. Thus, type 2 responses are essential for hepatic and cutaneous GVHD in this model, whereas both type 1 cells and type 2 cells mediate intestinal pathology. Oligonucleotide microarrays in a predominantly CD4+ T-cell-mediated model of hepatic GVHD demonstrated upregulation of genes encoding a variety of adhesion molecules and chemokines, including CC chemokines, such as monocytes chemoattractant protein-1 (MCP-1, CCL2), MCP-3 (CCL7), and regulated by activation, normal T cell expressed and secreted (RANTES, CCL5), as well as CXC chemokines, such as monokine induced by IFN-g (MIG, CXCL9), IFN-a-inducible protein (IP-10, CXCL10), and KC (CXCL1). In models of GVHD across MHC class I or class II barriers, increased mRNA and protein expression of MIP-1a was observed in spleen, GI tract, liver, and lung. Increased expression of MIP-1a was not detected in heart or pancreas. Others observed increased early hepatic expression of MIP-1a mRNA expression, but not of MIP-1h or RANTES. Macrophages, lymphocytes and endothelial cells, sinusoidal lining cells in the liver parenchyma and intralobular bile ducts in the portal areas, expressed MIP-1a protein. A comparative analysis of chemokine mRNA expression in various tissues from GVHD mice showed that MIP-1a, MIP-2 and MIG were predominantly expressed in liver, whereas the spleen predominantly expressed MIP-1a and MIP-2 and the skin MIP-2 and MCP-3 [25]. Thus, tissue-specific expression of distinct combinations of chemokines may play a role in the chemoattraction and further activation of allogeneic T cells into GVHD target tissues. Evidence that a lack of MIP-1a on donor T cells results in decreased infiltration of donor CD8+ T cells in lung, liver, and spleen of MHC class I disparate recipients [26] indicates that expression of chemokines by both recipient tissues activated donor T cells is critical for migration of GVHD effector cells to target organs. Expression of CCR5, the receptor for MIP-1a, MIP-1h and RANTES, and CCR5 on CD8+ liverinfiltrating T cells was significantly increased during

the second week of GVHD, which corresponded with the peak of MIP-1a mRNA expression. Treatment of recipient mice with anti-CCR5 Ab markedly reduced infiltration of CCR5+ effector CD8+ T cells in liver and serum ALT levels. In contrast, donor T cells without CCR5 produced accelerated acute GVHD with liver and kidney inflammation in lethally irradiated recipients [27]. These discrepant results are most likely due to use of unconditioned mice in one study [28] and lethally irradiated mice in the other [27]. This interpretation is directly supported by a recent study showing accelerated GVHD in irradiated, but not unirradiated, recipients of CCR5-deficient donor T cells [29]. In a CD8+ T-cell-mediated model of miHA GVHD gut-infiltrating donor CD8+ T cells lacking CXCR3, a receptor for MIG and IP-10, were sevenfold less frequent than CD8+ T cells from wild-type donors [30] due to splenic retention of activated CXCR3 / T cells. Histopathological lesions in the gut and liver were also reduced in recipients of CXCR3 / T cells. However, in a model of CD4+ T-cell-mediated GVHD, GVHD scores and mortality were not significantly affected by grafting CXCR3 / T cells.

8. Cytotoxic effector pathways The cytolytic effector phase of GVHD can be mediated by CD8+ or CD4+ CTL or NK cells. Both major mechanisms of cell-mediated cytotoxicity, Fas (APO-1, CD95) engagement with T cell ligand (FasL) or perforin and granzymes, have been implicated in GVHD pathogenesis. In addition, some cytokines secreted by T cells, particularly TNF-a and IFN-g, are also cytotoxic. The role of Fas–FasL has been underscored by evidence of delayed mortality after treatment with neutralizing anti-FasL antibody [31]; delayed disease onset and mortality in sublethally irradiated MHC-disparate recipients of FasL-deficient (B6.gld) T cells or using cytotoxically double deficient (cdd, perforin- and FasL-deficient) T cells [32]. In contrast, others reported that GVHD severity and mortality were unaltered in lethally irradiated MHCmatched recipients of T cells either lacking functional FasL (B6.gld) or both FasL and perforin compared to recipients of wild-type T cells [33]. Expansion of gld or cdd T cells, particularly the CD4+ subset, was less

Y. Ichiki et al. / Autoimmunity Reviews 5 (2006) 1–9

efficient in sublethally compared to lethally irradiated hosts. Hence, the different conditioning regimens used in these studies may partially explain these discrepant results. In addition, Fas-mediated cytotoxicity may depend on whether the alloreactive immune response is predominantly mediated by CD4+ or CD8+ T cells, since grafting of B6.gld T cells produced the same mortality as wild-type T cells in class I-mismatched recipients, but reduced mortality in class II-mismatched hosts. Interestingly, Fas-deficient (B6.lpr) recipients of MHC-matched C3H.SW T cells showed increased mortality and severity compared to recipients of wild-type T cells. Significantly greater donor T cell expansion was observed in the absence of Fas on host T cells. This, along with elevated serum levels of IFN-g and TNFa and a fourfold higher number of residual host peritoneal macrophages, may have contributed to increased GVHD severity and mortality. In several models, interference with Fas–FasL interactions significantly inhibited hepatic GVHD [34]. Thus, the Fas–FasL cytotoxic pathway appears to be crucial for generation of hepatic pathology. Additional cytotoxic mechanisms also coexist, since the pathological changes in liver and stomach after grafting of FasL and perforin double deficient were similar to those in recipients of wild-type T cells [33]. Blocking Fas–FasL pathway can also inhibit cutaneous GVHD [35], but this has not been a consistent finding [36]. The absence of Fas either on donor or recipient cells did not affect intestinal GVHD, consistent with evidence that TNF-a cytotoxic mechanisms mediate gut injury [37]. Perforin–granzyme-mediated cytotoxicity appears to play a lesser role in GVHD tissue damage. T cells from perforin-deficient donor mice significantly delayed onset and mortality of GVHD. Indeed, mortality was absent in MHC class I-disparate, but not in MHC class II-disparate, recipients of perforindeficient T cells. In another MHC class I-dependent model, lack of granzyme B on donor T cells also provided some protection from mortality. Donor T cells were deficient in both perforin and FasL produced non-lethal GVHD across MHC I and II barriers, but lack of either perforin or FasL alone only delayed mortality. Double deficient donor T cells, however, did not reduce mortality in a MHC-matched model [33].

7

9. Adhesion molecules Data on expression of adhesion molecules in human GVHD are limited. In murine models of GVHD, however, enhanced expression of ICAM-1 on biliary and portal vein endothelial cells and sinusoidal cells is commonly observed [38]. Also, the number of T cells expressing the VCAM-1 ligand, VLA4 is increased. ICAM-1 interaction with its ligand, lymphocyte function-associated antigen (LFA-1), is important for transendothelial T cell migration, adhesion and costimulatory signal during T cell activation and for regulating processes requiring contact between effector and target cell, including cytolysis. Increased expression of ICAM-1 by the target cells in hepatic GVHD lesions reflects its role in liver pathogenesis. Conversely, grafting of ICAM-1-deficient donor T cells resulted in unaltered hepatic and colonic pathology, indicating the importance of ICAM-1 expression by recipient target tissues [39]. As anticipated, treatment of mice during GVHD induction with anti-ICAM-1 mAb was ineffective. Others reported successful inhibition of hepatic lesions using anti-ICAM-1 mAb, but not anti-LFA-1 mAb. Inhibition was optimal using a combination of the two mAbs. Blocking interaction between VCAM and VLA-4 with anti-VLA-4 led to a significant reduction in the incidence and severity of hepatic lesions, but anti-VCAM had no effect. Since anti-ICAM-1 or antiVLA-4 mAbs did not prevent infiltration of T cells into the liver, the inhibition of hepatic lesions resulted from failure of effector and target cell interactions.

10. Conclusion Conditioning T cell activation with cytokine production, adhesion molecule expression and maturation and trafficking of effector T cells all play roles in the pathogenesis of GVHD. Fas-mediated effector cell cytotoxicity is the dominant mechanism producing hepatic and cutaneous GVHD. In contrast, TNF-a plays a more central role in intestinal GVHD. Effector cells mediating cytotoxicity through perforin–granzyme mechanisms are less essential for pathogenesis but can accelerate the process. Detailed understanding of the sequential events and pathogenetic mechanisms

8

Y. Ichiki et al. / Autoimmunity Reviews 5 (2006) 1–9

involved in GVHD should yield important insights to be translated into novel future therapies. [10]

Take-home messages ! Graft-versus-host disease is the most important complication of allogeneic bone marrow transplantation. ! Immune recognition is based upon differences in major and minor histocompatibility complex antigens and clinically significant acute GVHD occurs in up to 50% of allogenic BM transplant recipients. ! Tissue-specific expression of alloantigen, cytokines, chemokines and adhesion molecules contribute to the pattern of clinical pathology. ! MHC expression on host dendritic cells but not on target cells is required for GVHD.

References [1] Dickinson AM, Wang XN, Sviland L, et al. In situ dissection of the graft-versus-host activities of cytotoxic T cells specific for minor histocompatibility antigens. Nat Med 2002;8:410 – 4. [2] Norton J, al-Saffar N, Sloane JP. Adhesion molecule expression in human hepatic graft-versus-host disease. Bone Marrow Transplant 1992;10:153 – 6. [3] Goker H, Haznedaroglu IC, Chao NJ. Acute graft-vs-host disease: pathobiology and management. Exp Hematol 2001;29:259 – 77. [4] Kataoka Y, Iwasaki T, Kuroiwa T, et al. The role of donor T cells for target organ injuries in acute and chronic graft-versushost disease. Immunology 2001;103:310 – 8. [5] Joffre O, Gorsse N, Romagnoli P, Hudrisier D, van Meerwijk JP. Induction of antigen-specific tolerance to bone marrow allografts with CD4+CD25+ T lymphocytes. Blood 2004;103: 4216 – 21. [6] Hoffmann P, Ermann J, Edinger M, Fathman CG, Strober S. Donor-type CD4+CD25+ regulatory T cells suppress lethal acute graft-versus-host disease after allogeneic bone marrow transplantation. J Exp Med 2002;196:389 – 99. [7] Edinger M, Hoffmann P, Ermann J, et al. CD4+CD25+ regulatory T cells preserve graft-versus-tumor activity while inhibiting graft-versus-host disease after bone marrow transplantation. Nat Med 2003;9:1144 – 50. [8] Simpson E, Scott D, James E, et al. Minor H antigens: genes and peptides. Transpl Immunol 2002;10:115 – 23. [9] Jones SC, Friedman TM, Murphy GF, Korngold R. Specific donor Vh-associated CD4+ T-cell responses correlate with

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

severe acute graft-versus-host disease directed to multiple minor histocompatibility antigens. Biol Blood Marrow Transplant 2004;10:91 – 105. Miconnet I, de La Selle V, Bruley-Rosset M. Relative importance of CD4+ and CD8+ T cell repertoires in the development of acute graft-versus-host disease in a murine model of bone marrow transplantation. Bone Marrow Transplant 1998;21:583 – 90. Murata M, Warren EH, Riddell SR. A human minor histocompatibility antigen resulting from differential expression due to a gene deletion. J Exp Med 2003;197:1279 – 89. Friedman TM, Statton D, Jones SC, Berger MA, Murphy GF, Korngold R. Vh spectratype analysis reveals heterogeneity of CD4+ T-cell responses to minor histocompatibility antigens involved in graft-versus-host disease: correlations with epithelial tissue infiltrate. Biol Blood Marrow Transplant 2001; 7:2 – 13. Hirokawa M, Matsutani T, Saitoh H, et al. Distinct TCRAV and TCRBV repertoire and CDR3 sequence of T lymphocytes clonally expanded in blood and GVHD lesions after human allogeneic bone marrow transplantation. Bone Marrow Transplant 2002;30:915 – 23. Bruley Rosset MB, Tieng V, Charron D, Toubert A. Differences in MHC-class I presented minor histocompatibility antigens extracted from normal and graft-versus-host disease (GVHD) mice. Clin Exp Immunol 2003;132:46 – 52. Yu XZ, Bidwell SJ, Martin PJ, Anasetti C. CD28-specific antibody prevents graft-versus-host disease in mice. J Immunol 2000;164:4564 – 8. Yu XZ, Albert MH, Martin PJ, Anasetti C. CD28 ligation induces transplantation tolerance by IFN-g-dependent depletion of T cells that recognize alloantigens. J Clin Invest 2004; 113:1624 – 30. Duffner UA, Maeda Y, Cooke KR, et al. Host dendritic cells alone are sufficient to initiate acute graft-versus-host disease. J Immunol 2004;172:7393 – 8. Merad M, Hoffmann P, Ranheim E, et al. Depletion of host Langerhans cells before transplantation of donor alloreactive T cells prevents skin graft-versus-host disease. Nat Med 2004;10:510 – 7 [Erratum Nat Med 10, 649]. Jones SC, Murphy GF, Friedman TM, Korngold R. Importance of minor histocompatibility antigen expression by nonhematopoietic tissues in a CD4+ T cell-mediated graftversus-host disease model. J Clin Invest 2003;112:1880 – 6. Shlomchik WD, Couzens MS, Tang CB, et al. Prevention of graft versus host disease by inactivation of host antigenpresenting cells. Science 1999;285:412 – 5. Teshima T, Ordemann R, Reddy P, et al. Acute graft-versushost disease does not require alloantigen expression on host epithelium. Nat Med 2002;8:575 – 81. Welniak LA, Blazar BR, Anver MR, Wiltrout RH, Murphy WJ. Opposing roles of interferon-g on CD4+ T cell-mediated graft-versus-host disease: effects of conditioning. Biol Blood Marrow Transplant 2000;6:604 – 12. MacDonald KP, Rowe V, Filippich C, et al. Chronic graftversus-host disease after granulocyte colony-stimulating factor-mobilized allogeneic stem cell transplantation: the role of

Y. Ichiki et al. / Autoimmunity Reviews 5 (2006) 1–9

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

donor T-cell dose and differentiation. Biol Blood Marrow Transplant 2004;10:373 – 85. Nikolic B, Lee S, Bronson RT, Grusby MJ, Sykes M. Th1 and Th2 mediate acute graft-versus-host disease, each with distinct end-organ targets. J Clin Invest 2000;105:1289 – 98. New JY, Li B, Koh WP, et al. T cell infiltration and chemokine expression: relevance to the disease localization in murine graft-versus-host disease. Bone Marrow Transplant 2002; 29:979 – 86. Serody JS, Burkett SE, Panoskaltsis-Mortari A, et al. Tlymphocyte production of macrophage inflammatory protein1alpha is critical to the recruitment of CD8+ T cells to the liver, lung, and spleen during graft-versus-host disease. Blood 2000;96:2973 – 80. Welniak LA, Wang Z, Sun K, et al. An absence of CCR5 on donor cells results in acceleration of acute graft-vs-host disease. Exp Hematol 2004;32:318 – 24. Murai M, Yoneyama H, Harada A, et al. Active participation of CCR5+CD8+ T lymphocytes in the pathogenesis of liver injury in graft-versus-host disease. J Clin Invest 1999;104: 49 – 57. Wysocki CA, Burkett SB, Panoskaltsis-Mortari A, et al. Differential roles for CCR5 expression on donor T Cells during graft-versus-host disease based on pretransplant conditioning. J Immunol 2004;173:845 – 54. Duffner U, Lu B, Hildebrandt GC, et al. Role of CXCR3induced donor T-cell migration in acute GVHD. Exp Hematol 2003;31:897 – 902. Miwa K, Hashimoto H, Yatomi T, Nakamura N, Nagata S, Suda T. Therapeutic effect of an anti-Fas ligand mAb on lethal graft-versus-host disease. Int Immunol 1999;11:925 – 31. Jiang Z, Podack E, Levy RB. Major histocompatibility complex-mismatched allogeneic bone marrow transplantation

[33]

[34]

[35]

[36]

[37]

[38]

[39]

9

using perforin and/or Fas ligand double-defective CD4(+) donor T cells: involvement of cytotoxic function by donor lymphocytes prior to graft-versus-host disease pathogenesis. Blood 2001;98:390 – 7. Marks L, Altman NH, Podack ER, Levy RB. Donor T cells lacking Fas ligand and perforin retain the capacity to induce severe GvHD in minor histocompatibility antigen mismatched bone-marrow transplantation recipients. Transplantation 2004; 77:804 – 12. Ueno Y, Ishii M, Yahagi K, et al. Fas-mediated cholangiopathy in the murine model of graft versus host disease. Hepatology 2000;31:966 – 74. Hattori K, Hirano T, Miyajima H, et al. Differential effects of anti-Fas ligand and anti-tumor necrosis factor a antibodies on acute graft-versus-host disease pathologies. Blood 1998; 91:4051 – 5. van den Brink MR, Moore E, Horndasch KJ, et al. Fasdeficient lpr mice are more susceptible to graft-versus-host disease. J Immunol 2000;164:469 – 80. Schmaltz C, Alpdogan O, Muriglan SJ, et al. Donor T cellderived TNF is required for graft-versus-host disease and graft-versus-tumor activity after bone marrow transplantation. Blood 2003;101:2440 – 5. Ichiba T, Teshima T, Kuick R, et al. Early changes in gene expression profiles of hepatic GVHD uncovered by oligonucleotide microarrays. Blood 2003;102:763 – 71. Panoskaltsis-Mortari A, Hermanson JR, Haddad IY, Wangensteen OD, Blazar BR. Intercellular adhesion molecule-I (ICAM-I, CD54) deficiency segregates the unique pathophysiological requirements for generating idiopathic pneumonia syndrome (IPS) versus graft-versus-host disease following allogeneic murine bone marrow transplantation. Biol Blood Marrow Transplant 2001;7:368 – 77.

LKM1 autoantibodies in chronic hepatitis C infection: A case of molecular mimicry? Anti-liver-kidney microsome type I (LKM1) autoantibodies directed against the cytochrome P450 2 D6 (CYP2D6) are found in 5% of sera from patients chronically infected by hepatitis C virus (HCV). Molecular mimicry between HCV proteins and CYP2D6 has been proposed to explain the emergence of these autoantibodies. In this study by Marceau G. et al. (Hepatology 2005;42:675-82), anti-LKM1 autoantibodies from hepatitis C-infected patients were affinity-purified against immobilized CYP2D6 protein and used to screen a phage display library. Cross-reactivity between CYP2D6 and HCV protein candidates was tested by immunoprecipitation. Nineteen different clones were isolated, and their sequencing resulted in the mapping of a conformational epitope to the region of amino acids 254-288 of CYP2D6. Candidate HCV proteins for molecular mimicry included: core, E2, NS3 and NS5a. Affinity-purified autoantibodies from HCV+/LKM1+ patients, immunoprecipitated NS3, NS5a, or both, and these reactivities were specifically inhibited by immobilized CYP2D6. Cross-reactivity due to molecular mimicry at the B-cell level was shown between the CYP2D6 and the HCV NS3 and NS5a proteins and could explain the presence of anti-LKM1 in patients chronically infected with HCV.