transplantation immunology: what the clinician needs to know … · 2017-05-06 · transplantation...

TI

DC

TrtmitoiagscaTmtppvlaitgta

Lcsttpple

GASTROENTEROLOGY 2008;134:1789–1801

ransplantation Immunology: What the Clinician Needs to Know formmunotherapy

Hugo R. Rosen

ivision of Gastroenterology & Hepatology, Liver Transplantation, Hepatitis C Center, Department of Medicine, University of Colorado Health Sciences Center, Aurora,
olorado, and Integrated Program in Immunology, National Jewish Hospital, Denver, Colorado
gfadttscwnbHtct

sruHstbdtTpiw

cgcr

he liver is unique among transplanted organs withespect to its interaction with the host immune sys-em. There is evidence, both anecdotal and docu-

ented, that some liver recipients who cease takingmmunosuppressive drugs maintain allograft func-ion, suggesting robust tolerance is in place. More-ver, recipients of human liver allografts require less

mmunosuppression than do other organ recipients,nd liver transplants confer protection on other or-an grafts from the same donor. Hence, the liverhows features of immune privilege. Still, the liveran display destructive immunologic processes suchs rejection in approximately one quarter of patients.he understanding of the cellular and molecularechanisms operant in tolerance vs allograft rejec-

ion is important for developing new agents to im-rove long-term outcome, minimize infectious com-lications (including recurrence of hepatotropiciruses), and deliver immunosuppression withoutong-term toxicity. This review describes the uniquespects of the hepatic immune response, the pathwaysnvolved in T-cell activation and alloantigen recogni-ion, effector cells and pathways mediating liver allo-raft rejection, the role of regulatory T cells, andargets of current and future immunosuppressivegents.

iver transplantation is the definitive treatment ofchoice for patients with end-stage liver disease. Re-

ipients of human liver allografts require less immuno-uppression than do other organ recipients, and liverransplants confer protection on other organ grafts fromhe same donor.1 Nonetheless, a significant subset ofatients does develop allograft rejection. For the mostart, early rejection episodes do not negatively affect

ong-term graft survival.2,3 A number of animal mod-
ls4 – 6 show spontaneous, lifetime tolerance to liver allo-
rafts without therapeutic manipulation; however, therequency and extent to which drug weaning can beccomplished in human liver recipients has yet to beefined.7 It is widely accepted that after organ transplan-ation, hematopoietic donor cells are cotransferred intohe recipient and may survive for prolonged periods. Theize of the liver and its high content in hematopoieticells (both resident and circulating) may in part explainhy liver transplantation is associated with a significantumber of donor hematopoietic cells in the peripherallood of recipients and permissiveness with regards toLA incompatibility.8 Donor leukocytes could mediate

olerance by functioning as immature antigen-presentingells (APCs) or could act as surrogate targets of rejection,hus protecting the graft.8,9

Hematopoietic microchimerism is defined by the per-istence of less than 1% circulating donor cells in aecipient (macrochimerism denotes �1%). A recent studysing nested polymerase chain reaction of donor-specificLA DR allelic analysis showed that all liver recipients

howed microchimerism within the first 3 months afterransplantation.8 However, in accord with prior studiesy Hisanaga et al10 and Devlin et al,11 microchimerismid not correlate with freedom from rejection or abilityo tolerate staged immunosuppressive drug withdrawal.aken together, these data are consistent with the notionut forth by Starzl and Zinkernagel12 that microchimer-

sm is a necessary prerequisite for, but is not synonymousith, tolerance. Clearly, mechanisms mediating graft tol-

Abbreviations used in this paper: APC, antigen-presenting cell; CNI,alcineurin inhibitor; CTLA-4, cytotoxic T lymphocyte–associated anti-en 4; DC, dendritic cell; IL, interleukin; MHC, major histocompatibilityomplex; NK, natural killer; PD-1, programmed death-1; TCR, T-celleceptor; TNF, tumor necrosis factor; Treg, regulatory T cell.

© 2008 by the AGA Institute0016-5085/08/$34.00

doi:10.1053/j.gastro.2008.02.062

eb

ivaeaa

itnlmlndpmsciecc

ncvttrccli(NmusiasTpmcabmo

aaphccaCMtuert

stcliiTtsmcddtpics

dtttdmalmCepcfbIlit

1790 HUGO R. ROSEN GASTROENTEROLOGY Vol. 134, No. 6

rance are more complex than simply induction of anergyy transferred donor cells.

This review covers the unique aspects of the hepaticmmune response; the pathways involved in T-cell acti-ation and expansion; basic mechanisms of non–self-lloantigen recognition, effector cells, and pathways op-rant in allograft rejection; the role of regulatory T cells;nd targets of current and future immunosuppressivegents.

The Liver Is a Specialized ImmuneOrganThe liver is among the most interesting and potent

mmunologic organs, rivaled only by what are consideredhe true immune organs, such as thymus and lymphodes.13 The liver is exposed continuously to a diverse and

arge antigenic load, including pathogens, toxins, and tu-or cells, as well as dietary and commensal proteins.14 The

iver must be actively immunocompetent and simulta-eously control inappropriate inflammatory responses toietary and other harmless antigens encountered in theortal circulation, thus being able to selectively induce im-unity or tolerance to antigens.15 Pathogen invasion is

ensed through various pattern-recognition receptors in-luding Toll-like receptors (TLRs), which are expressed onmmune cells as well as on some parenchymal cells.16 Forxample, Toll-like receptor 4, the receptor for lipopolysac-haride, is expressed on immune cells, as well as hepato-ytes, endothelial cells, and stellate cells.

Lymphocytes are broadly divided into B cells, T cells, andatural killer (NK) cells. Although B and T cells possesslonotypic antigen receptors that confer specificity for di-erse antigenic structures, NK cells possess multiple recep-ors that can detect conserved antigens or danger signalshat mediate major histocompatibility complex (MHC)-un-estricted cytolysis of susceptible tumor and virus-infectedells.17,18 Numerous groups have documented that the liverontains an unusually high percentage of unconventionalymphoid cells that rarely are present in the blood, includ-ng NK cells, NK T cells expressing the �� T-cell receptorTCR), and the invariant V�24J�Q TCR.19,20 For example,

K (CD3�CD16�CD56�) cells represent 5%–15% of theononuclear cell population, but in the liver can comprise

p to 45% of the lymphocyte population.21 It has beenuggested that these cells deliver a death signal to circulat-ng recipient-derived T cells that migrate through the liverfter transplantation, contributing to tolerance.22 Moreover,ignificantly higher proportions of hepatic T cells expressCR �� than matched peripheral blood T cells.23 The he-atic �� TCRpos population contains more than twice asany CD8pos cells and 4 times as many double-negative

ells (T cells that express neither CD8 nor CD4 coreceptors)s the corresponding population in matched peripherallood.23 The liver also hosts the largest population of tissueacrophages (Kupffer cells) in the body and different types

f dendritic cells (DCs).16 c

DCs are professional APCs that are present in virtuallyll organs and provide a critical link between innate anddaptive immunity. The ability of DCs to process andresent various types of antigen (Ag) is unmatched in theuman body.24 In particular, DCs can display Ag pro-essed exogenously on MHC class I (cross-presentation orross-priming)25 or complete MHC-peptide complexescquired externally from dead cells (cross-dressing) toD8� T cells.26,27 Immature DCs express low levels ofHC class II adhesion and costimulatory molecules, but

heir expression is up-regulated dramatically during mat-ration in response to inflammatory stimuli.28 There isxpanding evidence that liver-derived DCs can down-egulate immune responses, hence inducing and main-aining peripheral T-cell tolerance.29

Intrinsic differences between liver DCs and other tis-ue-resident DCs have been noted and may have impor-ant biologic implications. For example, human mono-ytes differentiated into DCs when cocultured with rativer epithelial cells or liver-conditioned media secretenterleukin (IL)-10 but not IL-12p70, thus skewing themmune response towards T helper type 2 rather thanh1.30 Thus, the hepatic microenvironment may support

he generation and survival of regulatory T cells (de-cribed later).27 Furthermore, it has been suggested that

onitoring of DC subsets and their activation status orytokine production may provide a useful tool for pre-icting allograft outcome, including successful with-rawal of immunosuppression.31 Independent of theype or extent of immunosuppressive therapy, circulatinglasmacytoid DCs are increased relative to myeloid DCs

n clinically tolerant pediatric liver transplant recipientsompared with those on maintenance immunosuppres-ion.27,32

The context in which antigen is presented to T cellsetermines whether the responding T cell is activated orolerized; critical variables include the nature of the APC,he presence or absence of costimulatory molecules, andhe cytokine microenvironment.33,34 Recent provocativeata indicate that Ito (stellate) cells, known primarily forediating hepatic fibrogenesis, are APCs that can present

ntigen as efficiently as DCs.35 Liver sinusoidal endothe-ial cells have a unique immune phenotype, expressing

arkers typical of cells of myeloid lineage (CD1, CD4,D11c) even though data suggest that these cells differ-

ntiate from hepatocyte progenitors.36 CD4� T cellsrimed by antigen-presenting liver sinusoidal endothelialells or hepatocytes (which lack costimulatory molecules)ail to differentiate toward effector T helper type 1 cellsut, instead, express high levels of immune-suppressiveL-10.37 Moreover, antigen presentation to CD8 T cells byiver sinusoidal endothelial cells in vivo leads to theirnability to respond to specific antigen on restimula-ion.38 Thus, liver sinusoidal endothelial cells—the only
ells to have direct contact with immune cells passing

tl

aibratAonmrCmiaosssBDclpafCTApCwn

vatlFbpcvPidailc

tmc

Tli4tmiatctcmfmeTclaeCaKclsncta

oomalrmTpwdebtpg

May 2008 TRANSPLANTATION IMMUNOLOGY 1791

hrough the liver—may play an important role in estab-ishing tolerance after liver transplantation.14

Costimulatory Pathways andTransplantationBecause T cells are essential for allograft rejection

nd many immunosuppressive drugs target T-cell signal-ng, it is important to understand the multistep processesy which they become activated. T lymphocytes musteceive 2 distinct coordinated signals to achieve optimalctivation and expansion. The first signal is delivered byhe TCR after recognition of peptide–MHC complexes onPCs and the second signal is provided by the interactionf costimulatory molecules on the T cells and their cog-ate ligands on APCs.39 Figure 1A shows this 2-signalodel, including the wide array of heretofore described

eceptor–ligand pairs.40 Although the CD28/B7 andD40/CD154 costimulatory pathways have garneredost of the attention in transplantation, there is emerg-

ng evidence for the importance of other molecules inlloimmune responses,14 including critical negative sec-nd signals that down-regulate or terminate T-cell re-ponses (Figure 1B). CD28 (the prototypical T-cell co-timulatory molecule) is expressed constitutively on theurface of T cells; the ligands for CD28 (B7-1 [CD80] and7-2 [CD86]) are found on a variety of APCs includingCs, B cells, and macrophages. In naive T cells, CD28

ostimulation enhances cell-cycle entry, potently stimu-ating expression of IL-2 and induction of anti-apoptoticroteins.41 In contrast, cytotoxic T lymphocyte–associ-ted antigen 4 (CTLA-4), which has approximately 20-old higher affinity for both B7-1 and B7-2 than doesD28, is up-regulated after T-cell activation and inhibits-cell responses, regulating peripheral T-cell tolerance.42

recent analysis of CTLA-4 single-nucleotide polymor-hisms in 483 liver transplant recipients showed that theTLA-4 �49A/�6230G haplotype, which is associatedith reduced soluble CTLA-4 production, is a co-domi-ant risk factor for acute rejection.43

Programmed death-1 (PD-1) also is induced after acti-ation of T cells, shares sequence homology with CTLA-4,nd contains an immunoreceptor tyrosine-based inhibi-ory motif in the cytoplasmic tail, characteristic of cellu-ar receptors that deliver a negative signal.14 As shown inigure 1B, when PD-1 is engaged by its ligands or CTLA-4inds to B7 and displaces CD28, the activation andhosphorylation of SHP-2 (SRC homology 2– domain-ontaining protein tyrosine phosphatase 2) that deacti-ates downstream signal transducers recruited to theD-1 tail results.44 SHP-2–mediated dephosphorylation

nactivates multiple signaling molecules, resulting inown-regulation of cytokine messenger RNA synthesisnd inhibition of T-cell proliferation.44 A recent prelim-nary analysis of liver perfusates collected from donorivers before allograft implantation indicated a signifi-
antly higher expression of PD-1 on CD8� T cells from e
he liver relative to peripheral blood and suggested thisay be an important mechanism imparting tolerogene-

ity.45

Several members of the tumor necrosis factor (TNF)/NF receptor families also can provide alternate costimu-

atory signals to T lymphocytes. These receptor/ligand pairsnclude CD40/CD154, 4-1BB/4-1BBL (CD137), OX-40/OX-0L (CD134/CD134L), and CD30-CD30L (CD153).14 Ofhese, CD40/CD154 is the best characterized in alloim-

une responses. CD40 is expressed constitutively on APCsncluding B cells, monocytes, macrophages, and DCs, butlso can be expressed on nonimmune cells including endo-helial cells, mast cells, platelets, and epithelial cells. Inontrast, CD154 is expressed on CD4 T cells after activa-ion, and to a lesser extent on NK cells, B cells, and CD8 Tells. Unlike the CD28/B7 costimulatory pathway that pri-arily has been defined in the context of effects on T-cell

unction, CD40/CD154 ligation enhances APC function aseasured by up-regulation of class II, CD80, and CD86

xpression and production of cytokines including IL-12.14,46

he sum effect of these changes to the APC is to signifi-antly augment B- and T-cell responses to alloantigen. Mo-ecular and immunohistochemical analyses of CD80, CD86,nd CD154 expression in biopsy specimens of liver recipi-nts showed an association between increased expression ofD86 and CD154 (but not CD80) in the graft during severecute cellular rejection.47 CD154 also was detected onupffer cells and sinusoidal macrophages in livers during

hronic rejection, but not in stable liver allografts or normalivers.48 Thus, the ultimate fate of cellular immune re-ponses is determined by the balance between positive andegative signals delivered by costimulatory molecules to Tells; undoubtedly, an expanding list of molecules will con-inue to be studied as possible therapeutic targets to preventnd treat allograft rejection.

Basic Aspects of T-Cell Recognition ofAlloantigenThe term allorecognition refers to T-cell recognition

f genetically encoded polymorphisms between membersf the same species.49 The primary targets are the MHColecules present on donor cells. The recognition of

llograft MHC antigen is the primary event ultimatelyeading to graft rejection. Recipient T lymphocytes canecognize donor alloantigen through 2 distinct but not

utually exclusive pathways. In the direct pathway, hostlymphocytes recognize native MHC molecules ex-

ressed on graft-associated APCs. In the indirect path-ay, host T lymphocytes recognize donor alloantigen-erived peptides in the context of self MHC moleculesxpressed on recipient APCs (Figure 2).50 Evidence foroth pathways exists in liver transplantation. It is likelyhat the direct pathways predominate early posttrans-lant and is a major factor in acute rejection becauseraft-derived APCs expressing donor alloantigen rapidly
gress from the graft and enter secondary lymphoid tis-

FercCssbmi


igure 1. The 2-signal model for positive and negative costimulation of T-cell activation. (A) T-cell activation requires 2 signals: signal 1, TCRngagement with MHC–peptide complex, and signal 2, ligation of costimulatory molecules on T cells with their respective ligands on APC. T cells thateceive both signal 1 and positive costimulation show proliferation, secretion of cytokines, and differentiation into effector cells. The prominentostimulatory molecules delivering positive costimulation signals are CD28 and CD40L. Under certain circumstances, ICOS, CD134, CD30, 4-1BB,D27, and CD70 also can deliver positive T-cell costimulation. Some costimulatory molecules, such as CTLA-4 and PD-1, can lead to negative T-cellignaling, resulting in decreased cell proliferation and cytokine production and cellular anergy. Reprinted from Gao et al.40 (B) Negative regulatoryignaling of T cells. PD-1 and CTLA-4 are negative regulators of TCR signaling. Engagement of PD-1 by its ligands (PD ligand-1 or PD ligand-2) orinding of CTLA-4 to B7 (displacing CD28) is followed by the recruitment of the phosphatase SHP-2 to the immunoreceptor tyrosine-based inhibitoryotifs in the cytoplasmic tails of PD-1 and CTLA-4. Then SHP-2 dephosphorylates the CD3� chains and other signaling pathways, which results in

nhibition of TCR signaling and downstream events associated with T-cell activation. Reprinted with permission from Mak and Saunders.44

sTactic

dMdomi

FrdtiefhR


ue where they can encounter allospecific T cells.51 Hostcells primed through the direct pathway have the

bility to engage the allograft directly and, thus, to effi-iently mediate effector functions. Further, the propor-ion of host T cells that can respond to native alloantigens substantially greater than the proportion of host T

igure 2. Allorecognition pathways and graft rejection. (A) Direct anecognize intact major histocompatibility molecules on donor APCs. Inerived peptides in the context of self MHC molecules expressed on recihat is shed from damaged graft tissue, or perhaps in the case of the livern particular DCs. (B) Interactions among endothelial cells, T cells, anndothelial cells to the graft tissue are transformed to become highly effior maturation. The DCs and intragraft macrophages present donor peand, are activated by donor endothelial cells and either can directly killeprinted with permission from Briscoe and Sayegh.50

ells that can respond through the indirect pathway to i

onor-derived peptides presented in the context of selfHC.52 The indirect pathway probably emanates from

onor alloantigen that is shed from damaged graft tissue,r perhaps in the case of the liver, from soluble MHColecules, that are picked up and presented by self APCs,

n particular DCs.14 Therefore, the direct pathway may be

irect pathways of allorecognition. In the direct pathway (left), T cellsdirect pathway (right), T cells recognize processed donor alloantigen-APCs. The indirect pathway probably emanates from donor alloantigensoluble MHC molecules that are picked up and presented by self APC,ipient APCs in allograft rejection. Recipient monocytes recruited byantigen-presenting DCs that recirculate to peripheral lymphoid organss via the indirect pathway to CD4� T cells. CD8� T cells, on the otherthelial cells or traverse the endothelium and kill parenchymal graft cells.

d indthe inpient, fromd reccientptideendo

mportant in initiating the classic form of acute rejection

bibvAtgtiltopfi

lotmacwtacefnmcHgvHcsHlb

chK(sMrnTlatrtA(

llihTrpp

tiadoarteaiodloa

mtilpriildlv

pmacicptfilrac


ecause donor-derived passenger leukocytes have a lim-ted lifespan, the indirect pathway may predominate latery sustaining a response fueled by epitope spreading as aariety of allopeptides are presented successively by selfPC.14 Accordingly, T cells with specificity for allopep-

ides have been detected readily in liver recipients under-oing chronic rejection.53 At the same time, it is impor-ant to recognize that the indirect pathway may benvolved in immune regulation because T cells with al-opeptide specificity were shown to have regulatory func-ion through inhibition of interferon � production inther organ recipients.52 The contribution of the indirectathway to hepatic allograft tolerance recently was con-rmed.54

In addition to graft rejection and tolerance, these al-orecognition pathways likely are important in recurrencef viral or autoimmune liver diseases. In this regard, liverransplantation is performed with no regard to specific

atching of donor-recipient HLA alleles; this may serves a barrier to the development of protective (ie, antiviral)ell-mediated immunity directed against infected cellsithin the allograft. CD8� T cells are the primary effec-

or lymphocytes for provision of protective immunitygainst intracellular pathogen infection of parenchymalells. Recognition of the infected allograft could occurither via recipient-derived T cells or via those derivedrom the donor. For the recipient-derived T cells, recog-ition could occur either through use of shared HLAolecules, or could occur through the expansion of re-

ipient-derived T cells that are uniquely restricted by theLA molecules of the donor liver. A study from our

roup55 has shown the generation of new hepatitis Cirus (HCV)-specific T cells that are restricted by donorLA alleles, yet derived from the recipient’s original T-

ell pool (Figure 3). For the purposes of this study, weelected HLA A2-negative recipients of HLA A2� grafts;LA-A2 was selected as the restricting allele because a

arge number of HCV HLA-A2 binding peptides haveeen described as targets of HCV-specific CTLs.

As shown in Figure 3A, recipient HCV-specific CTLlones cultured with HLA A2� lymphoid cell lines thatad been pulsed with cognate peptide (NS31406 –1415,LVALGINAV) but not irrelevant HCV core peptide

core35– 44, YLLPRRGPRL) elicited a strong immune re-ponse by interferon � enzyme-linked immunospot.

oreover, when these recipient CTLs were cultured withecipient-derived (syngeneic) lymphoid cell lines and cog-ate peptide, there was no appreciable immune response.hese data suggest that these HCV-specific CTLs circu-

ate within the recipient and only display functionalntiviral activity (ie, secrete interferon � and show cyto-oxicity; Figure 3B), when they encounter allograft-de-ived HLA molecules and viral peptide. Furthermore,hese CTLs can recognize endogenously processed viralg expressed by the HLA molecule of the donor graft

Figure 3C). It is likely that the microenvironment (eg, l

ymphoid aggregates in the portal tracts or secondaryymphoid tissues) in which this immune response wasnitiated contained DCs that engulfed HCV-infectedepatocytes. These results underscore the plasticity of theCR, as well as the need to assess both allograft- and

ecipient-restricted CTLs (and thus direct and indirectathways) to develop a comprehensive understanding ofrotective immunity to HCV after liver transplantation.

Effector Pathways of Graft InjuryAcute cellular rejection occurred historically in 50%

o 75% of liver allograft recipients, although recent advancesn immunosuppression have yielded lower rates (ie, �30%ccording to Scientific Registry of Transplant Recipientsata; http://www.ustransplant.org/glossary). The majorityf episodes occur within 90 days of transplant surgery,56

nd the majority of cases respond to high-dose corticoste-oids. Three main types of allograft rejection are described:he extremely rare hyperacute rejection characterized byndothelial injury and mediated by deposition of antibodynd fibrin but no lymphocytic infiltration or bile ductnjury; acute rejection characterized by the diagnostic triadf portal inflammation, bile duct damage, and venular en-othelial inflammation; chronic rejection characterized by

oss of small bile ducts (ie, ductopenic rejection) and anbliterative vasculopathy affecting large- and medium-sizedrteries and the portal microvasculature.56,57

For both acute and chronic rejection, the targets ofature effector T cells are donor-derived bile duct epi-

helial cells and vascular endothelium, whereas directnvolvement of hepatocytes is uncommon.57 Cytotoxicymphocytes kill cholangiocytes via either perforin-de-endent pathways or activation of members of the TNFeceptor superfamily (particularly Fas)57 (Figure 4). Thenteraction between CTL and target cell is facilitated byntegrins, notably LFA-1, which help form an immuno-ogic synapse containing the secretory domain.57 In ad-ition, as shown in Figure 4B, other integrins such as very

ate antigen-4 may enhance adhesion and provide sur-ival signals for effector cells.57

Recent data57 indicate that bile ducts constitutively ex-ress the chemokine CCL19, which attracts DCs and pro-otes their localization at the biliary epithelium where they

re ideally situated to respond to Ag. Although humanholangiocytes lack CD80 and CD86, minimizing their abil-ty to activate naive T cells,58 the majority of T cells thatome into contact with activated cholangiocytes are alreadyrimed cells, and therefore, cholangiocytes may promotehe local proliferation and survival of such cells.57 Given thending that the majority of lymphocytes infiltrating human

iver allografts in rejection express functional chemokineeceptors CXCR3, CXCR4, and CCR559 indicates that ther-peutic strategies in the future may aim to prevent lympho-yte infiltration of liver allografts by inhibiting chemokine
igand/receptor-mediated pathways.
http://www.ustransplant.org/glossary

FrdgcnllwedLtG(tctmcpaeggsacCH(VcdorcrcfereltaaR


igure 3. Recipient-derived T cells thatecognize HCV peptides in the context ofonor HLA molecules (HLA A2). Taken to-ether, these results suggest that T cells cir-ulate in liver transplant recipients and areot activated until they encounter donor al-

eles containing HCV peptides. (A) Enzyme-inked immunospot assay was performed

ith 1000 T cells, lymphoid cell lines (LCLs)xpressing A2 allele (top row), A3 allele (mid-le row), or syngeneic (recipient-derived)CLs (bottom row) co-cultured with no pep-ide, cognate peptide (NS31406–1415, KLVAL-INAV), or irrelevant HCV core peptide

core35–44, YLLPRRGPRL). (B) Cytotoxic ac-ivity of HCV-specific tetramer�CD8� Tells. T-cell clones were assayed for pep-ide-specific cytotoxicity using the fluoro-etric assessment of T lymphocyte Ag-spe-

ific lysis assay, which determines theercentage of labeled target cells survivingfter a 5-hour incubation with effector cells;ffector:target ratio is 50:1 with 1 � 106 tar-et cells. The target cells are selected byating on the PKH-26high population andhow the reduction in carboxyfluorescein di-cetate succinimidyl ester (CFSE) fluores-ence after incubation with NS31406-specificTL, cognate peptide (KLVALGINAV), andLA-A2 LCLs (a), compared with no peptide

b), irrelevant peptide NS52594–2602 (ALYDV-TKL) (c), and no effector cells (d). The per-entage killing of syngeneic LCLs was notifferent after incubation in the presence (e)r absence (f) of cognate peptide. FITC, fluo-escein isothiocyanate. (C) NS31406-specificlones (HCV1–HCV4, represented by bars)ecognize endogenously processed Ag. T2ells were pulsed with HCV1406–1415 peptideor 2 hours. Cell lines were transduced withither empty retroviral vector or with retrovi-al vector containing the HCV minigene thatncodes NS31406–1415 peptide. Cytomega-

ovirus-specific T cells were used as a con-rol. Interferon � secretion (mean � SD) wasssessed by enzyme-linked immunosorbentssay. Reprinted with permission from
osen et al.55

prw

iic

Ftirmaoietap m Ad


The perforin pathway is mediated by granzymes anderforin stored in specialized secretory lysosomes that areeleased into the immunologic synapse after engagement

igure 4. Cholangiocyte apoptosis is mediated by CTLs via activationo target cells by the release of chemokines, secreting specialized granuln the target cell. Inflamed cholangiocytes express increased MHC clecognition and cell–cell adhesion. On CTL activation, perforin- and granembrane fusion occurs and granzymes are delivered into the target

poptosis; granzyme B activates Bcl2-interacting domain and caspasef Fas and other TNF receptors (TNFRs). Fas ligand is contained in the se

n the immunologic synapse. Cross-linking of Fas with FasL leads to caxpress TNF ligands, implicating them as potential contributors to bilehe expression of FasL via nuclear factor �B-dependent pathways, resund LFA-1 are critical for the formation of the immunologic synapse, otrovide survival signals for effector cells. Reprinted with permission fro

ith a target cell.60 Granzymes cleave specific substrates, w

ncluding caspases and Bcl2-interacting domain, result-ng in the induction of apoptosis.57 There is strong cir-umstantial evidence that the granzyme/perforin path-

F receptors or the granzyme/perforin pathway. (A) CTLs are attractedt contain proteinases and activate the caspase cascade and apoptosisand intercellular cell adhesion molecules (ICAM-1), promoting antigene- containing secretory granules are mobilized to the cell surface whereGranzyme A makes nicks in single-stranded DNA, resulting in directpendent apoptosis. (B) CTLs also kill cholangiocytes by the activationry granules and with other members of the TNF family is concentrated-dependent apoptosis. A wide range of cells (including macrophages)njury. CD40 does not directly activate caspases but instead increasesn autocrine or juxtacrine apoptosis via Fas activation. Although ICAM-1tegrins such as very late antigen-4 (VLA-4) may enhance adhesion andams and Afford.57

of TNes thaass Izymcell.

-3–decreto

spaseduct ilting iher in

ay is activated in human allograft rejection.57,61 Cross-

lnFicncpiccOrtcthda

smsbC1pcTcAaCwded

mmie(arfFarsfftw

copdt

mmprdf

io(oipicmtpsmmm1TfiMiptItbteidaCto

atott


inking of Fas (ubiquitously expressed on lymphoid andonlymphoid tissue including the liver) with trimerizedasL or agonist antibodies leads to formation of the death-

nducing signaling complex in target cells, activation ofaspase 8, and propagation of a death signal that culmi-ates in apoptosis.57 The Fas/FasL pathway has been shownonvincingly to play critical roles in a variety of hepaticathologies,62,63 and there is evidence that this pathway also

s active during liver allograft rejection.64,65 A wide range ofells (including macrophages) express TNF ligands, impli-ating them as potential contributors to bile duct injury.57

ther TNF receptors likely involved in acute and chronicejection (and thus potential targets for pharmacologic in-ervention) are shown in Figure 4B. CD40 ligation onholangiocytes results in up-regulation of FasL; thus, al-hough CD40 does not directly activate caspases, it en-ances the expression of FasL via nuclear factor �B–depen-ent pathways, resulting in autocrine or juxtacrinepoptosis via Fas activation.57

Regulatory T Cells and TransplantationThe balance of effector (proinflammatory) re-

ponses and regulatory responses may ultimately deter-ine whether rejection or tolerance occurs. Several sub-

ets of CD4� T cells with suppressive properties haveeen described, including naturally occurringD4�CD25� regulatory T cells (Treg), T regulatory type, and T helper type 3 cells.66 CD4�CD25� Treg sup-ress the response of conventional T cells via a cellontact-dependent manner, whereas T helper type 3 and

regulatory type 1 cells produce immunosuppressiveytokines (eg, IL-10 and transforming growth factor-�).lthough immunoregulatory activity specific for donorlloantigens is enriched in the CD4� T-cell populations,D8� T cells, double-negative (CD4-CD8-) T cells, asell as NK T cells also have shown regulatory activities inifferent situations.67 Treg cells also may mediate theirffects by modifying the functions of other T cells, eitherirectly or indirectly through APCs.67

CD25 (the �-subunit of the IL-2 receptor) is used as aarker for Tregs, but is by no means a definitive or specificarker. Further, CD4�CD25high Tregs have been character-

zed by the constitutive expression of CD62L, intracellularxpression of CTLA-4, forkhead transcription factor 3FoxP3, the most specific marker for Tregs), lymphocytectivation gene-3 (LAG-3), and, most recently, the down-egulation of IL-7 receptor (CD127) expression.68–71 Tregrequency in the peripheral blood72 and intrahepaticoxP3� cells73 are reduced in liver allograft recipients withcute rejection, and T regulatory type 1 cells are increased inecipients developing spontaneous graft tolerance. A recenttudy showed that donor CD4�CD25� Tregs originatingrom the liver graft are able to suppress responder T cellsrom both recipient and donor.74 These findings indicatehat Treg can suppress across an MHC barrier, congruent
ith previous observations that Tregs suppress Ag nonspe- i
ifically once activated through their TCR. Thus, chimerismf donor Treg may contribute to suppression of the directathway alloresponse, which, as described earlier, is theominant Ag presentation mechanism early after liverransplantation.74

How Immunosuppressive Agents WorkThis section builds on the understanding of basic

echanisms outlined earlier to explain how current im-unosuppressive agents work (Table 1) and to describe

otential new targets for immune manipulation. Theeader is referred to excellent reviews that additionallyescribe important drug– drug interactions and side ef-ects of these medications.75–77

In general, immunosuppression can be attained by block-ng lymphocyte response pathways, depleting lymphocytes,r diverting lymphocyte traffic.76 Calcineurin inhibitorsCNIs), cyclosporine and tacrolimus, remain the backbonef immune suppression in liver transplantation. Cyclospor-

ne engages the immunophilin cyclophilin, forming a com-lex that then engages calcineurin.76 This complex in turn

nhibits transcription of several genes, including IL-2, criti-al to T-cell activation. Tacrolimus engages a different im-unophilin, FK506-binding protein 12, to create a complex

hat also inhibits calcineurin but with greater molecularotency.76 Thus, both CNIs interfere with signal 1 T-cellignal transduction (Figures 1 and 5). Rapamycin (siroli-

us) is a macrolide antibiotic structurally related to tacroli-us but with very different mechanisms of action. Rapa-ycin binds to the immunophilin, FK506-binding protein

2, but rather than inhibiting cytokine gene transcription incells, it blocks signals transduced from a variety of growth

actor receptors to the nucleus by acting on phosphatidylnositol kinases called mammalian targets of RAPA.77,78

oreover, it also inactivates 70-kD-S6 kinase, which resultsn selective inhibition of the synthesis of new ribosomalroteins, prolonging cell-cycle progression from the G1 tohe S phase.77 Thus, CNIs and sirolimus differentially affectL-2: CNIs inhibit IL-2 production whereas sirolimus blockshe IL-2 signaling pathway at a later stage after receptorinding.67 Interestingly, these differences are manifested byhe fact that CNIs (which inhibit IL-2 secretion essential forffector cells and Treg development) prevent activation-nduced cell death of effector cells, whereas sirolimusoes not have the same effect.67 Moreover, CNIs aressociated with depletion in the peripheral blood ofD4�CD25�FoxP3� Tregs after renal transplanta-

ion, whereas sirolimus is associated with preservationf Tregs,79 in keeping with in vitro data.80

Corticosteroids are the most frequently used non-CNIgents. Their effects are primarily transcriptionalhrough DNA binding and protein–protein interactionsf the steroid-receptor complex, which target transcrip-ion factors such as activator protein 1 and nuclear fac-or-�B,76 abrogating expression of multiple cytokines,
ncluding IL-1, IL-2, IL-3, and IL-6.75 In addition, corti-

clttmsomp6a

ocec

raatmmisfooAmse

T

S

P

Aa lishe


osteroids suppress eicosanoid production, down-regu-ate adhesion molecules, and increase the expression ofransforming growth factor-�. Antimetabolites used inransplantation include azathioprine, mycophenolate

ofetil, and mycophenolic acid. All 3 antagonize purineynthesis, which blocks differentiation and proliferationf T and B lymphocytes.75 Mycophenolate mofetil andycophenolic acid inhibit synthesis of guanosine mono-

hosphate nucleotides, whereas azathioprine converts-mercaptopurine to tissue inhibitor of metalloprotein-se, which is converted to thioguanine nucleotides.76

Antibodies used in transplantation target a wide varietyf different pathways. Antithymocyte globulin is a poly-lonal antibody preparation that targets multiple differentpitopes on T cells (CD2, CD3, CD4, CD8, CD28) and NK

able 1. Mechanisms of Action of Immunosuppressive Agent

Drug

mall-molecule immunosuppressive agentsCyclosporine (CyA) Binds to cyclophilin; com

1 T-cell signal transduInhibits IL-2 synthesis a

Tacrolimus (FK506) Binds to FKBP12; compT-cell signal transduct

Inhibits IL-2 synthesis aSiroliums (rapamycin) Binds to FKBP12 (but no

proliferationRelative to CyA and FK5Blocks signals transduc

phosphatidyl inositol kInactivates 70 kilodalton

ribosomal proteinsPreserves Tregs in vitroInhibits B-cell stimulatio

Corticosteroids Target transcription factexpression of multiple

Suppress eicosanoid proDown-regulate adhesionIncrease the expression

Mycophenolate mofetil (MMF) andmycophenolic acid (MPA)

Inhibits synthesis of guaproliferation of T and

Azathioprine (AZA) Converts 6-mercaptopurthoguanine nucleotidepurine synthesis

Inhibits proliferation androtein immunosuppressive drugsAntithymocyte globulin Polyclonal antibody prep

CD4, CD8, CD28) andMuromonab-CD3 (OKT3) Binds to the CD3 antige

1), ultimately leadingBasiliximab (chimeric) and daclizumab

(humanized)IL-2–receptor antibodies

receptor �-chain (CD2proliferation

Alemtuzumab (campath-IH) Humanized, recombinanmonocytes, and macr

Rituximab Chimeric monoclonal anBelatacept (LEA29Y)a CTLA-4 fusion protein im

and signal 2 transduc

dapted from Halloran.76

Demonstrated efficacy in renal and islet cell transplantation (no pub

ells (CD16).75 Muromonab-CD3 (OKT) is a murine-de- r

ived antibody that exerts its activity by binding to the CD3ntigen on the surface of T lymphocytes, inactivating thedjacent TCR (signal 1). Two currently licensed IL-2–recep-or antibodies, basiliximab (chimeric) and daclizumab (hu-

anized), bind to the IL-2 receptor �-chain, achieving im-unosuppression by competitive antagonism of IL-2–

nduced T-cell proliferation (blocking signal 3).75 Althoughhort-term treatment with CD25-specific antibodies is ef-ective at reducing the incidence of allograft rejection, the-retically, it also might compromise the subsequent devel-pment of immunoregulation to donor alloantigens.67

lemtuzumab or campath-1H is a humanized, recombinantonoclonal antibody that targets antigen CD52, a cell-

urface glycoprotein, expressed on the majority of periph-ral blood lymphocytes, thymocytes, monocytes, and mac-

ed in Liver Transplantation

Mechanisms of action

inhibits calcineurin phosphatase and T-cell activation (inhibit signal)leasehibits calcineurin phosphatase and T-cell activation (inhibit signal 1

leaselophilin); complex inhibits target of rapamycin and IL-2 driven T-cell

locks IL-2 signalling pathway at a later stage after receptor bindingm a variety of growth factor receptors to the nucleus by acting on

es called mammalian targets of RAPA (mTOR)kinase, which results in selective inhibition of the synthesis of new

n vivocreasing antibody productionuch as activator protein 1 and nuclear factor-kB,55 abrogatingkines, including IL-1, IL-2, IL-3, and IL-6ionculesFe monophosphate nucleotides, blocks purine synthesis, preventing

lstissue inhibitor of metalloproteinase, which is converted to

erfering with DNA synthesis; thioguanine derivatives may inhibit

rentiation of B and T cells

n that targets multiple different epitopes on T cells (CD2, CD3,cells (CD16), functionally altering and depleting different cell typesthe surface of T lymphocytes, inactivating the adjacent TCR (signalell depletioniliximab (chimeric) and daclizumab (humanized), bind to the IL-2igen), depleting activated T cells and inhibiting IL-2–induced T-cell

oclonal antibody that targets antigen CD52 on T, B, NK cells,es, causing lysis and depletion

y that binds to CD20 on B cells, mediating lysisoglobulin that binds to B7 on T cells, preventing CD28 signaling

d data in liver transplantation).

s Us

plexctionnd relex inion)nd ret cyc

06, bed froinasS6

and in, deors scytoductmoleof TGnosinB celine tos, int

diffe

aratioNK

n onto T-c, bas5 ant

t monophagtibodmuntion

ophages (Figure 5).77 Campath-1H produces profound

dmlicpd

ipfinAhats

mun

tgprrmnt

F(ciefp�tL(t1tl


epletion of circulating lymphocytes for approximately 1onth, and the rationale for its initial use is to facilitate

ower doses of maintenance immunosuppression.77 Ritux-mab is a chimeric monoclonal antibody against CD20 on Bells; its use in liver transplantation has been limited toatients who develop posttransplant lymphoproliferativeisorders81 and in recipients of ABO-incompatible grafts.82

A new generation of biological agents that block thenteraction between CD80/86 and CD28 costimulatoryathways (signal 2) includes belatacept (LEA29Y), a CTLA-4usion protein immunoglobulin, which has shown efficacyn recipients of kidney76,83 and islet cell allografts,84 but hasot yet been studied in liver recipients. Blockade of T-cell/PC costimulatory pathways mediated via CD40–CD154as been accomplished using the anti-CD154 monoclonalntibody, but studies have been halted because of reports ofhromboembolic phenomena, perhaps related to the expres-ion of CD154 on activated platelets.77

In summary, based on our understanding of cellular andolecular mechanisms, the immune strategies currently

sed in liver transplantation target a number of mecha-

igure 5. Immunosuppressive drugs and sites of action. As describedB7-1) and CD86 (B7-2) on the APCs engage CD28 on the T cell to palcium–calcineurin pathway, the mitogen-activated protein kinase pat

n turn, activate transcription factors nuclear factor of activated T cellxpression of CD154 (which further activates APCs), IL-2 receptor � curther inhibiting transcription of several genes (including IL-2, critical to Trotein 1 and NF-�B, abrogating expression of multiple cytokines, inc-chain, achieving immunosuppression by competitive antagonism of IL

hrough the phosphoinositide-3-kinase (PI-3K) pathway and the moleymphocytes require synthesis of purine and pyrimidine nucleotides fo

MMF), and mycophenolic acid (MPA) antagonize purine synthesis, whiherapies likely will interrupt signaling of multiple cytokine receptors such-phosphate (S-1-P) receptors profoundly alters lymphocyte traffic, drivio the graft. The first agent using this novel mechanism (FTY720) waymphocyte trafficking. Reprinted with permission from Halloran.76

isms (eg, signal 1 and 2 transduction, lymphocyte deple-

ion). The imminent future likely will witness an evenreater expansion in our armamentarium of immunosup-ressive agents than in the past decade. Our limitationsemain very similar to those of the past: long-term toxicityemains a concern and we have very few biomarkers to

easure the induction of tolerance85 and fine-tune immu-osuppression based on an individual patient’s characteris-ics.

References

1. Rasmussen A, Davies HF, Jamison NV, et al. Combined trans-plantation of liver and kidney from the same donor protects thekidney from rejection and improves kidney graft survival. Trans-plantation 1995;59:919–921.

2. http://www.medpagetoday.com/Surgery/Transplantation/tb/8085.United Network for Organ Sharing. Available at: http://www.unos.org.

3. Charlton M, Seaberg E. Impact of immunosuppression and acuterejection on recurrence of hepatitis C: results of the NationalInstitute of Diabetes and Digestive and Kidney Diseases LiverTransplantation Database. Liver Transpl Surg 1999;5:S107–S114.

4. Calne RY, Sells RA, Pena JR, et al. Induction of immunological

e text, antigen triggers TCRs (signal 1) and synapse formation; CD80e signal 2. These signals activate signal-transduction pathways—the, and the protein kinase C–nuclear factor-�B (NF-�B) pathway—which,T), activating protein 1 (AP-1), and NF-�B, respectively. The result isCD25), and IL-2. Both cyclosporine and tacrolimus inhibit calcineurin,activation). Corticosteroids target transcription factors such as activator

IL-1, IL-2, IL-3, and IL-6. IL-2 receptor antibodies bind to the IL-2Ruced T-cell proliferation (blocking signal 3). IL-2 delivers growth signals-target-of-rapamycin (mTOR) pathway, which initiates the cell cycle.lication, and antimetabolites azathioprine (AZA), mycophenolate mofetilcks differentiation and proliferation of T and B lymphocytes. Futurenus kinase 3 (JAK3) inhibitors. Engagement of lymphocyte sphingosine-ells into lymphoid tissues, and prevents them from leaving and homingontinued because of toxicity, but future approaches likely will target

in throvid

hways (NFA

hain (-cellluding-2-indcularr repch bloas Ja

ng T cs disc

tolerance by porcine liver allografts. Nature 1969;223:472–474.

http://www.medpagetoday.cont/Surgery/Transplantation/tb/8085

http://www.unos.org

http://www.unos.org

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

22

2

2

2

3

3

3

3

3

3

3

3

3

3

4

4

4

4

4

4

4

4

4

4

5

5


5. Murase N, Starzl TE, Tanabe M, et al. Variable chimerism, graftversus host disease, and tolerance after different kinds of celland whole organ transplantation from Lewis to Brown-Norwayrats. Transplantation 1995;60:158–171.

6. Qian S, Demetris AJ, Murase N, et al. Murine liver allografttransplantation: tolerance and donor cell chimerism. Hepatology1994;19:916–924.

7. Starzl TE, Lakkis FG. The unfinished legacy of liver transplanta-tion: emphasis on immunology. Hepatology 2006;43:S151–S163.

8. Bettens F, Tiercy JM, Campanile N, et al. Microchimerism afterliver transplantation: absence of rejection without abrogation ofanti-donor cytotoxic T-lymphocyte-mediated alloreactivity. LiverTranspl 2005;11:290–297.

9. Wood KJ. Passenger leukocytes and microchimerism: what role intolerance induction? Transplantation 2003;75:17S–20S.

0. Hisanaga M, Hundrieser J, Böker K, et al. Development, stability,and clinical correlations of allogeneic microchimerism after solidorgan transplantation. Transplantation 1996;61:40–45.

1. Devlin J, Doherty D, Thomson L, et al. Defining the outcome ofimmunosuppression withdrawal after liver transplantation. Hepa-tology 1998;27:926–933.

2. Starzl TE, Zinkernagel RM. Transplantation tolerance from a his-torical perspective. Nat Rev Immunol 2001;3:233–239.

3. Crispe IN. Hepatic T cells and liver tolerance. Nat Rev Immunol2003;3:51–62.

4. Martinez OM, Rosen HR. Basic concepts in transplant immunol-ogy. Liver Transplant 2005;11:370–381.

5. Calne RY. Immunological tolerance—the liver effect. ImmunolRev 2000;174:280–282.

6. Szabo G, Mandrekar P, Dolganiuc A. Innate immune responseand hepatic inflammation. Semin Liver Dis 2007;27:339–350.

7. Doherty DG, O’Farrelly C. Lymphoid repertoires in healthy liver. In:Gershwin ME, Vierling JM, Manns MP, eds. Liver immunology.Philadelphia: Hanley & Belfus, 2003:31–46.

8. Lanier LL. Natural killer cell receptor signaling. Curr Opin Immunol2003;15:308–314.

9. Golden-Mason L, O’Farrelly C. Having it all? Stem cells, haema-topoiesis and lymphopoiesis in adult human liver. Immunol CellBiol 2002;80:45–51.

0. Norris S, Collins C, Doherty DG, et al. Resident human hepaticlymphocytes are phenotypically different from circulating lympho-cytes. J Hepatol 1998;28:84–90.

1. Golden-Mason L, Rosen HR. Natural killer cells: primary target forhepatitis C virus immune evasion strategies? Liver Transpl 2006;12:363–372.

2. Lilly LB, Ding J, Levy GA. The immunology of hepatic allograftrejection. In: Maddrey WC, Schiff ER, Sorrell M, eds. Transplan-tation of the liver. Philadelphia: Lippincott Williams & Wilkins,2001:211–228.

3. Golden-Mason L, Douek DC, Koup RA, et al. Adult human livercontains CD8� T cells with naive phenotype, but is not a site forconventional alpha beta T cell development. J Immunol 2004;172:5980–5985.

4. Wertheimer AW, Bakke A, Rosen HR. Direct enumeration andfunctional assessment of circulating dendritic cells in patientswith liver disease. Hepatology 2004;40:335–345.

5. Bevan MJ. Cross-priming. Nat Immunol 2006;7:363–365.6. Dolan BP, Gibbs KD Jr, Ostrand-Rosenberger S. Dendritic cells

cross-dressed with peptide MHC class I complexes prime CD8�

T cells. J Immunol 2006;177:6018–6024.7. Sumpter TL, Abe M, Tokita D, et al. Dendritic cells, the liver, and

transplantation. Hepatology 2007;46:2021–2031.8. Jonuleit H, Schmitt E, Steinbrink K, et al. Dendritic cells as a tool

to induce anergic and regulatory T cells. Trends Immunol 2001;
22:394–400.
9. Banchereau J, Steinman RM. Dendritic cells and the control ofimmunity. Nature 1998;392:245–252.

0. Cabillic F, Rougier N, Basset C, et al. Hepatic environment elicitsmonocyte differentiation into a dendritic cell subset directing Th2response. J Hepatol 2006;44:552–559.

1. Mazariegos GV, Zahorchak AF, Reyes J, et al. Dendritic cellsubset ratio in peripheral blood correlates with successful with-drawal of immunosuppression in liver transplant patients. Am JTransplant 2003;3:689–696.

2. Mazariegos GV, Zahorchak AF, Reyes, et al. Dendritic cell subsetratio in weaning and non-tolerant liver recipients is not affectedby extent of immunosuppression. Am J Transplant 2005;5:314–322.

3. O’Farrelly C. Immunoregulation in the liver and its extrahepaticrelevance. J Pediatr Gastroenterol Nutr 2004;39(Suppl 3):S727–S728.

4. Benseler V, McCaughan GW, Schlitt HJ, et al. The liver: a specialcase in transplant tolerance. Semin Liver Dis 2007;27:194–213.

5. Winau F, Hegasy G, Weiskirchen R, et al. Ito cells are liver-resident antigen-presenting cells for activating T cell responses.Immunity 2007;26:117–129.

6. Knolle PA, Limmer A. Role and function of liver sinusoidal endo-thelial cells. In: Gershwin ME, Vierling JM, Manns MP, eds. Liverimmunology. Philadelphia: Hanley & Belfus, 2003:59–73.

7. Katz SC, Pillarisetty VG, Bleier JI, et al. Liver sinusoidal endothe-lial cells are insufficient to activate T cells. J Immunol 2004;173:230–235.

8. Limmer A, Ohl H, Wingender G, et al. Cross-presentation of oralantigens by liver sinusoidal endothelial cells leads to CD8 T celltolerance. Eur J Immonol 2005;35:2970–2981.

9. Frauwirth KA, Thompson CB. Activation and inhibition of lympho-cytes by costimulation. J Clin Invest 2002;109:295–299.

0. Gao W, Demirci G, Li XC. Negative T cell costimulation and islettolerance. Diabetes Metab Res Rev 2003;19:179–185.

1. Sun ZW, Qiu YH, Shi YJ, et al. Time courses of B7 family mole-cules expressed on activated T-cells and their biological signifi-cance. Cell Immunol 2005;236:146–153.

2. Greenwald RJ, Freeman GJ, Sharpe AH. The B7 family revisited.Annu Rev Immunol 2005;23:515–548.

3. Tapirdamaz O, Pravica V, Metselaar HJ, et al. Polymorphisms inthe T cell regulatory gene cytotoxic T lymphocyte antigen 4 influ-ence the rate of acute rejection after liver transplantation. Gut2006;55:863–868.

4. Mak TW, Saunders ME. T cell activation. In: The immune re-sponse: basic and clinical principles. Burlington, MA: Elsevier,2006:373–402.

5. Wang P, Jassem W, Longhi MS, et al. Abundance of PD-1 andPD-L1 expressing CD8 T-cells in the liver may be key to hepatictolerogenecity. Hepatology 2007;46:A581.

6. Quezada SA, Jarvinen LZ, Lind EF, et al. CD40/CD154 interac-tions at the interface of tolerance and immunity. Annu Rev Im-munol 2004;22:307–328.

7. Bartlett AS, McCall JL, Ameratunga R, et al. Analysis of intragraftgene and protein expression of the costimulatory molecules,CD80, CD86 and CD154, in orthotopic liver transplant recipients.Am J Transplant 2003;3:1363–1368.

8. Gaweco AS, Wiesner RH, Yong S, et al. CD40L (CD154) expres-sion in human liver allografts during chronic ductopenic rejection.Liver Transpl Surg 1999;5:1–7.

9. Rogers NJ, Lechler RI. Allorecognition. Am J Transplant 2001;1:97–102.

0. Briscoe DM, Sayegh MH. A rendezvous before rejection: where doT cells meet transplant antigens? Nat Med 2002;8:220–222.

1. Demetris AJ, Qian S, Sun H, et al. Early events in liver allograft
rejection: delineation of sites of simultaneous intragraft and

5

5

5

5

5

5

5

5

6

6

6

6

6

6

6

6

6

6

7

7

7

7

7

7

7

7

7

7

8

8

8

8

8

8

PB7U


recipient lymphoid tissue sensitization. Am J Pathol 1991;138:609–619.

2. Illigens BM, Yamada A, Fedoseyeva EV, et al. The relative contri-bution of direct and indirect antigen recognition pathways to thealloresponse and graft rejection depends upon the nature of thetransplant. Hum Immunol 2002;63:912–925.

3. Renna-Molajoni E, Cinti P, Evangelista B, et al. Role of theindirect recognition pathway in the development of chronic liverallograft rejection. Transplant Proc 1998;30:2140–2141.

4. Toyokawa H, Nakao A, Bailey RJ, et al. Relative contribution ofdirect and indirect allorecognition in developing tolerance afterliver transplantation. Liver Transpl 2008;14:346–357.

5. Rosen HR, Hinrichs DJ, Leistikow RL, et al. Selection of HCV-specific CD8� T cells by donor HLA alleles following liver trans-plantation. J Immunol 2004;173:5355–5359.

6. Wiesner RH, Ludwig J, Krom RA, et al. Hepatic allograft rejection:new developments in terminology, diagnosis, prevention, andtreatment. Mayo Clin Proc 1993;68:69–79.

7. Adams DH, Afford SC. Effector mechanisms of nonsuppurativedestructive cholangitis in graft-versus-host disease and allograftrejection. Semin Liver Dis 2005;25:281–297.

8. Leon MP, Bassendine MF, Gibbs P, et al. Immunogenicity ofbiliary epithelium: study of the adhesive interaction with lympho-cytes. Gastroenterology 1997;112:968–977.

9. Goddard S, Williams A, Morland C. Differential expression ofchemokines and chemokine receptors shapes the inflammatoryresponse in rejecting human liver transplants. Transplantation2001;72:1957–1967.

0. Bossi G, Griffiths GM. CTL secretory lysosomes: biogenesis andsecretion of a harmful organelle. Semin Immunol 2005;17:87–94.

1. Kuijf ML, Kwekkeboom J, Kuijpers MA. Granzyme expression infine-needle aspirates from liver allografts is increased duringacute rejection. Liver Transpl 2002;8:952–956.

2. Akazawa Y, Gores GJ. Death receptor-mediated liver injury. SeminLiver Dis 2007;27:327–338.

3. Mengshol JA, Golden-Mason L, Rosen HR. Mechanisms of dis-ease: HCV-induced liver injury. Nat Clin Pract Gastroenterol Hepa-tol 2007;4:622–634.

4. Tannapel A, Kohlhaw K, Ebelt J, et al. Apoptosis and the expres-sion of Fas and Fas ligand (FasL) antigen in rejection and rein-fection in liver allograft specimens. Transplant 1999;15:1079–1083.

5. Ogura Y, Martinez OM, Villanueva JC, et al. Apoptosis and allo-graft rejection in the absence of CD8 T cells. Transplant 2001;71:1827–1834.

6. Stassen M, Schmitt E, Jonuleit H. Human CD(4�)CD(25�) reg-ulatory T cells and infectious tolerance. Transplantation 2004;15:S23–S25.

7. Wood KJ, Sakaguchi S. Regulatory T cells in transplantationtolerance. Nat Rev Immuol 2003;3:199–210.

8. Viglietta V, Baecher-Allan C, Weiner HL, et al. Loss of functionalsuppression by CD4�CD25� regulatory T cells in patients withmultiple sclerosis. J Exp Med 2004;199:971–979.

9. Macon-Lemaitre L, Triebel F. The negative regulatory function ofthe lymphocyte-activation gene-3 co-receptor (CD223) on human
T cells. Immunology 2005;115:170–178.
0. Liu W, Putnam AL, Xu-Yu Z, et al. CD127 expression inverselycorrelates with FoxP3 and suppressive function of human CD4�T reg cells. J Exp Med 2006;203:1701–1711.

1. Kang SM, Tang Q, Bluestone JA. CD4�CD25� regulatory T cellsin transplantation: progress, challenges and prospects. Am JTransplant 2007;7:1457–1463.

2. Demirkiran A, Kok A, Kwekkeboom J, et al. Low circulating regu-latory T-cell levels after acute rejection in liver transplantation.Liver Transpl 2006;12:277–284.

3. Demirkiran A, Baan CC, Kok A, et al. Intrahepatic detection ofFOXP3 gene expression after liver transplantation using mini-mally invasive aspiration biopsy. Transplantation 2007;83:819–823.

4. Demirkiran A, Bosma BM, Kok A, et al. Allosuppressive donorCD4�CD25� regulatory T cells detach from the graft and circu-late in recipients after liver transplantation. J Immunol 2007;178:6066–6072.

5. Post DJ, Douglas DD, Mulligan DC. Immunosuppression in livertransplantation. Liver Transpl 2005;11:1307–1314.

6. Halloran PF. Immunosuppressive drugs for kidney transplanta-tion [erratum in: N Engl J Med 2005;352:1056]. N Engl J Med2004;351:2715–2729.

7. Fung J, Kelly D, Kadry Z, et al. Immunosuppression in livertransplantation: beyond calcineurin inhibitors. Liver Transpl2005;11:267–280.

8. Neuhaus P, Klupp J, Langrehr JM. mTOR inhibitors: an overview.Liver Transpl 2001;6:473–484.

9. Segundo DS, Ruiz JC, Izquierdo M, et al. Calcineurin inhibitors,but not rapamycin, reduce percentages of CD4�CD25�FOXP3�regulatory T cells in renal transplant recipients. Transplantation2006;82:550–557.

0. Strauss L, Whiteside TL, Knights A, et al. Selective survival ofnaturally occurring human CD4�CD25�Foxp3� regulatory Tcells cultured with rapamycin. J Immunol 2007;178:320–329.

1. Vincenti F, Larsen C, Durrbach A, et al. Costimulation blockadewith belatacept in renal transplantation. N Engl J Med 2005;353:770–781.

2. Trappe R, Riess H, Babel N, et al. Salvage chemotherapy forrefractory and relapsed posttransplant lymphoproliferative disor-ders (PTLD) after treatment with single-agent rituximab. Trans-plantation 2007;83:912–918.

3. Egawa H, Teramukai S, Haga H, et al. Present status of ABO-incompatible living donor liver transplantation in Japan. Hepatol-ogy 2008;47:143–152.

4. Fan K, Wang H, Wei H, et al. Blockade of LIGHT/HVEM andB7/CD28 signaling facilitates long-term islet graft survival withdevelopment of allospecific tolerance. Transplantation 2007;84:746–754.

5. Isaacs JD. T cell immunomodulation—the Holy Grail of therapeu-tic tolerance. Curr Opin Pharmacol 2007;7:418–425.

Received November 26, 2007. Accepted February 12, 2008.Address requests for reprints to: Hugo R. Rosen, MD, Waterman

rofessor of Medicine and Immunology, GI & Hepatology Division,-158, Academic Office Building 1, 12631 East 17th Avenue, Room614, PO Box 6511, Aurora, Colorado 80045. e-mail: [email protected]; fax: (303) 724-1891.
Supported by a VA Merit Review grant and NIH U19 A1066328.
mailto:[email protected]

mailto:[email protected]

transplantation immunology: what the clinician needs to know … · 2017-05-06 · transplantation...

Documents