Mechanism of liver allograft rejection: The indirect recognition pathway

Mechanism

of Liver

The Indirect

Recognition

Elvira Renna Molajoni, Annamaria Orlandini, Eric Ho, Zhuoru Liu, Raffaello Cortesini

Allograft

Pathway

Paola Cinti, Jacopo Molajoni, Nicole Suciu-Foca,

ABSTRACT: Transplant rejection is mediated by the direct and indirect pathways. To explore the role of the indirect recognition pathway in the rejection of liver allografts, T cells obtained from peripheral blood were expanded in medium containing IL-2 and tested in LDA for reactivity to synthetic peptides corresponding to the hypervariable regions of the mismatched HLA-DR antigen(s) of the donor. Serial investigations of 17 recipients showed that T-cell reactivity to donor HLA-DR peptides was strongly associated with acute rejection episodes. In recipients carrying a graft that was mismatched by two HLA-DR alleles, a single donor antigen was targeted dur-

ABBREVIATIONS MHC MLC DC IFN-y APC DTH LDA PBMC

Rejection:

major histocompatibility complex mixed lymphocyte culture dendritic cells interferon gamma antigen-presenting cell delayed-type hypersensitivity limiting dilutions assays peripheral blood mononuclear cells

Sorina and

Tugulea,

allopeptide reactivity ing primary rejection, although against the second HLA-DR antigen was observed during subsequent episodes of acute rejection. The finding that allopeptide reactivity occurs early following transplantation and is predictive of rejection is consistent with the notion that processing of donor alloantigens by recipient APCs activates the indirect T-cell recognition pathway that plays a major role in initiating and amplifying allograft rejection. Human Immnology 53, 57-63 (1997). 0 American Society for Histocompatibility and Immunogenetics, 1997.

CBA Aza 6-MP HPLC PCR-SSOP IL-2

cyclosporine azathioprine 6-methylprednisolone high performance liquid chromatography polymerase chain reactionsequence-specific oligonucleotide interleukin 2

INTRODUCTION The major obstacle in achieving long-term success in liver transplantation resides in rejection. In spite of considerable progress in the management of patients with liver allografts, rejection occurs in 50% to 82% of re-

From the Unizwsita’ Deglt Stud2 dr Roma “La Sapienza. ” lnstituto di II Chma Chiruvgica. (E.R.M.. P.C.. A.O.. R.C.). Sewzro Trapianti d’0pno. Roma, the Consiglio Naziotrale de//e Ric!mche (J. M.). Roma. Italy. and the College of Physiclam 6 Surgeons oj* Columbta Unmrstty (S.T.. E.H.. Z.L.. N.S.-F.). Neu, York. NY. USA. Address reprint requests to Dr. Ehira Renna Molajoni, Unherszta De& Studi di Roma “La Sapienza. “ lnstituto di I1 Clinica Chiwrgica. Laboratorio dz Inmunologta de/ Trapianti. Policlini‘-o Umberto 1. 00161 Roma. Italla. Ret-erved December 26. 1996: amped Jarmarl 10. 1997. Human Immunology 0 Amencan Soc-lety

53, 57-63 (1997) for Histocompatlbhty

and

Immunogenerlcs,

1997

cipients and is responsible for the loss of 5% to 15% of grafts [l]. Acute rejection of liver allografts develops early following transplantation and is characterized by lymphocytic infiltration of portal tracts associated with bile duct damage and inflammation of portal and hepatic venular endothelium. Up to 20% of patients with acute rejection develop chronic rejection that is characterized by loss of intrahepatic bile ducts and obliterative vasculopathy of large- and medium-sized arteries C2-41. Both CD4 and CD8 T cells have been identified at the site of rejection and were shown to display donor-specific reactivity. However, their relative contribution to rejection has not yet been defined [s-lo]. Although many cells in organ allografts express major histocompatibility 019%8859/97/$17.00 PI1 SO198-8859(97)00029-3

E. R. Molajoni et al.

58

complex (MHC) classI and classII antigens, they may not be able to deliver the costimulatory signals necessary for T-cell activation. Evidence has been accumulating that the principal allogeneic cell type responsible for stimulating T-cell proliferation in vitro and for initiating allograft rejection in viva is the bone-marrow derived dendritic cell (DC) or its associatedlineage 1111. Clustering of responding lymphocytes around allogeneic dendritic cells occurs both in mixed lymphocyte culture (MCL) and in organ allografts, suggesting that direct recognition by recipient T cells of MHC classI and class II antigens expressed on passengerDC of donor origin triggers the rejection process {12-141. Study of liver allograft rejection in untreated rats showed that a proportion of passengerDC from the donor are releasedin the recipient circulation shortly after transplantation and that they migrate into the recipient spleen where they trigger T-cell activation and proliferation 1151. Because lymphocyte transformation and proliferation was also seen in regional lymph nodes, in which no donor DC were found, and becausethe normal life span of these cells, both in the graft and in the periphery, is relatively short, it was suggested that direct recognition may not be the only pathway for allosensitization, and that indirect allorecognition may also be involved in rejection [15]. T cells may recognize, via the indirect pathway, allopeptides resulting from the processing of soluble or particulate donor MHC antigens released by the graft and processedby host antigen-presenting cells (APC) in regional lymph nodes115]. The observations made in the unmodified rat recipient of liver allografts may also be relevant to human transplantation. Thus, it is possible that liver allograft rejection in immunosuppressed recipients is initiated by CD4 T helper cells stimulated via the indirect recognition pathway. Antigen shedding because of harvesting injury, preservation, and reperfusion ischemia could promote processing by host APC and subsequent presentation of allopeptides bound to host MHC class II antigens to CD4 T helper cells. To explore the hypothesis that rejection may be initiated by T cells activated via the indirect pathway we have undertaken a systematic study of the ability of recipients’ T cells to recognize donor HLA-DR peptides following liver transplantation. In viuo activated T cells, from patients’ peripheral blood, were expanded in IL-2 and tested in limiting dilutions assays(LDA) for reactivity to synthetic peptides corresponding to each of the mismatched HLA-DR antigens of the donor. The results provided evidence that indirect alloreactivity against one of the donor’s HLADR antigens occurs early following transplantation and is predictive of rejection. The data are consistent with the notion that sensitization via the indirect pathway occurs first in the peripheral lymphoid tissue and that the al-

lopeptide reactive T cells migrate to the graft causing rejection by DTH-reactions or by producing the cytokines required for the proliferation of CDS effector cells. MATERIALS

AND METHODS

Patients and Immunosuppression Peripheral blood mononuclear cells (PBMC) obtained from 17 patients (15 males and 2 females) who underwent orthotopic liver transplantation at the University of Rome “La Sapienza” were tested for reactivity against donor-specific synthetic peptides. All patients received combined triple-drug immunosuppressive therapy: cyclosporine (CyA), azathioprine (Aza), and 6-methylprednisone (6-MP). The CyA was started at a dose of 7 mgi kg/day; the subsequent dosage was determined after measurementof whole blood CyA levels (the therapeutic level was considered in the range of 150 to 250 ng/mL). Aza and 6-MP were given at low dosages. Peptide Synthesis Synthetic peptides corresponding to residues 1-19, 2139, and 62-80 of DRPl chain from 32 HLA-DR alleles (DRBl*OlOl, DRBl*0102, DRBl”0301, DR@1*0302, DRPl”0401, DRBl*0402, DRpl*0403, DRB1*0404, DRpl*0405, DRBl”0407, DRpl*0408, DRpl*0701, DRBl”0801, DRpl*0802, DRpl”0803, DRpl*0804, DRB1*0901, DRPl”1001, DRPl”1101, DRB1*1102, DRBl”1104, DRPl”1201, DRBl”1301, DRBl*1302, DRBl”1303, DRB1*1401, DRpl*1402, DRBl”1501, DRPl”1502, DRBl”1503, DRPl”1601, and DRPl*1602) were obtained from Chiron Mimotopes (Space Imp./Exp. Mi., Italy) and Peptide Innovations (Raleigh, NC). The purity of peptides was higher than 90% as indicated by high performance liquid chromatography (HPLC) and mass spectrometry. Synthetic peptides were dissolved in RPM1 I640 medium at a concentration of 2 mgiml, as previously described 1161. Limiting T Cells

Dilution

Analysis of Allopeptide

Reactive

PBMCs were grown for 7 days in round bottom 96-well trays at concentrations of 4 x lo”, 2 x lOI, and 1 x lo* cell per well (24 wells for each concentration) in RPM1 I640 medium supplemented with 10% human serum and 50 units of r-IL-2iml (Boehringer Mannheim, Indianapolis, IN). On day 7, plates were washed three times and each culture was split into three aliquots. Two plates were used for testing T-cell reactivity to a cocktail of peptides (1 pM of each peptide) corresponding to each of the mismatched HLA-DR antigens of the donor in the presenceof irradiated autologous PBMCs (5 x 10”iwell). The third plate, to which only self PBMCs (without peptides) were added, served as a negative con-

Indirect

Recognition

of Liver Allografts

59

trol. After 48 hours of incubation the cultures were labeled with [3H]TdR and harvested 18 hours later. Wells were scored positive if cpm in cultures with allopeptides exceeded 2,000, and were at least three times higher than cpm in replicate control cultures without the peptides. The frequency of allopeptide reactive Th cells was calculated as previously described 1161. LDA was considered positive if more than 1.5 x lo-” Th cells reactive to one of the donor’s HLA-DR antigens were detected. HLA

Typing

All recipients and donors included in this study were typed for HLA-DR antigens by conventional serology and by molecular methodology using polymerase chain reaction and sequence-specific oligonucleotide probes (PCR-SSOP). Statistics Statistical analyses were performed using BMDP statistical software (BMDP Statistical Software Inc., Los Angeles, CA). The incidence of rejection two weeks following LDA was calculated by the product limit method. Differences between groups were compared using Breslow statistics. A test of linear trend was applied to the analysis of the relationship between allopeptide reactivity and progression of acute rejection. Diagnosis

and Treatment

of Acute

Rejection

Rejection episodes were suspected in cases of scant production of light bile or graft dysfunction as defined by rising serum levels of bilirubin or hepatic enzymes without evidence of mechanical causes or infection. Doppler ultrasonography was done when indicated to rule out hepatic artery or portal vein thrombosis. If suspicion of a vascular complication prevailed, diagnosis had to be confirmed by angiography. Bile leakage or biliary obstruction were excluded by cholangiography. The treatment of rejection consisted of lg of 6-methylprednisolone for 3 consecutive days.

indicative of activation of the indirect allorecognition pathway. Allopeptide reactivity was observed in 23 of 47 samples of PBMC studies in LDA. The frequency of acute rejection episodes was significantly higher when allopeptide reactive T cells were present in the circulation than when they were absent (p < 0.0001) (Fig. 1). Tests of linear trend showed a significant correlation between allopeptide reactivity in the peripheral blood and progression of the rejection process (p < 0.0001, Table 1). Three patients showed no evidence of rejection up to 170 days of observation. In 14 recipients primary rejection episodes were diagnosed within the first month following transplantation. Four of these patients displayed secondary rejection episodes at later times (Table 1). In 9 cases the results of LDA were positive within 5 to 10 days (mean = 6 days) prior to the clinical diagnosis of acute rejection, indicating that the presence of allopeptide reactive T cells in the circulation is predictive of rejection. In 5 of these 9 cases, allopeptide reactive T cells were detected in the patient’s peripheral blood at the time of transplantation. Each of these 5 patients had an acute rejection episode during the first week following transplantation suggesting that they had been presensitized to the donor’s HLA-DR antigens by previous transfusions and that early rejection was related to T-cell presensitization to donor HLA-DR peptides. Allopeptide reactivity was invariably found in LDAs performed at the time of rejection (8 of 8) and in 71% of the tests (5 of 7) done within one week of successful treatment of a rejecFIGURE 1 Association between acute rejection of liver allografts and Th-reactivity to donor allopeptides. Serial blood samples obtained from 17 patients were tested in LDA for reactivity to mismatched HLA-DR antigens of the donor. LDA was considered positive if more than 1.5 x 10m6 allopeptide reactive Th cells were detected. Rejection was monitored by functional tests and considered to have occurred when responding to steroid treatment. Incidence of rejection one month after LDA was calculated by the product-limit methods.

RESULTS Relationship Reactivity

LDA+ (n=23) Between To HLA-DR

Acute Rejection and T-Cell Peptides of Donor Origin

To explore the kinetics of allopeptide reactivity in recipients of liver transplants, PBMCs obtained at various times posttransplantation were expanded in medium with t-IL-2 and then tested for reactivity to synthetic peptides corresponding to the hypervariable region of the mismatched HLA-DR antigen(s) of the donor. A minimum of two sequential samples of blood (mean 2.8) were obtained from each patient during the early posttransplantation period. T-cell reactivity in response to one or both sets of donor HLA-DR peptides was considered

LDA- (m24)

Time Interval (days) between LDA Test andHistokgkDiiofRejectton Figum 1

E. R. Molajoni

60

TABLE

1

Relationship between clinical status

T-cell

reactivity

to donor

Time Relative

Rejection None Single

episode

No. patients

No. LDA

LDA results

3

7

+

.O

24

+

Two episodes

4

16

+

Total

7

47

+

tion episode. Only one of 23 formed during rejection free more was positive, indicating to the absence of allopeptide Specificity

of Allopeptide

Before onset

6 0 3 0 9 0

and

testing

to rejection

Concommitant

determinations periods of one that quiescence reactive T cells

Reactive

of LDA

allopeptides

et al.

2 0 6 0 8 0

(4%) permonth or was related (Table 1).

T cells

To identify the HLA-DR specificity of indirect allorecognition T cells from liver allograft recipients who were mismatched by two HLA-DR antigens, were tested in LDA for reactivity against individual mixtures of synthetic peptides corresponding to the hypervariable region of each of the donor’s HLA-DR antigens. Although seven recipients carried grafts differing by two DR antigens, self-MHC restricted Th reactivity was directed against only one of the donors’ antigens at the time of a primary rejection episode (Table 2 and Fig. 2). Intermolecular spreading of T-cell epitopes occurred, however, during secondary rejection in two patients (Fig. 2A,B) as evidenced by the development of T-cell reactivity against the previously “ignored” HLA-DR peptides. Hence, although the alloimmune response was originally directed against a sole alloantigen of the donor it eventually spread to include other antigens expressed by the same target. In patients with a graft mismatched by a single HLA-DR antigen the precursor frequency of allopeptide reactive T cells seemed to be higher during a secondary, compared to primary episode of rejection (Fig. 2E). LDA studies of indirect allorecognition of donor antigens at the time of resolving rejection, showed either a significant decrease in frequency (Fig. lC,F) or the complete disappearance from the circulation of allopeptide reactive T cells (Fig. 1D). DISCUSSION Allograft rejection is common following liver transplantation and remains a major cause of morbidity and graft failure 131. The characteristic histopathologic

Resolving

,“,~~~~~a 1 6 0 10 0 6 1 22

i 1 0 1 i 2

changes include the triad of portal tract inflammation, bile duct injury, and hepatic cell endotheliatis. The mechanism of allograft rejection remains incompletely understood, yet it is likely that a complex network of interaction between helper, cytotoxic, and inflammatory cells participates in the rejection process 15-l 11. Rejection occurs most frequently during the early postoperative period and can be reversed by immunosuppressive therapy. It is currently accepted that both the direct and the indirect pathway of allorecognition are involved in allograft rejection. Although direct recognition is presumed to play a prevailing role for as long as passenger leukocytes of donor origin are present in the graft, indirect recognition seems to promote chronic rejection damaging the graft at later times when TABLE

2

Specificity of T-cells donor allopeptide

reactivity

to

Allopeptide HLA-DRP Case A B C D E F G H f K L M N

Recipient 0101:1301 0101/0401 070111301 040 l/ 1 30 1 110111101 160111601 1401~1401 1001~1502 0101~1101 0701:1601 1001:1102 0101/l 101 0701/1502 15Olil101

1 genotype Donor 070111101 030111101 0101/1501 0101/1401 030111101 0 10 110 10 1 070111601 010110401 0301/1401 1101/l 101 1601/1601 0301/1101 14Oli1101 0401i1501

Specificity

of T-cell

reactivity

lo RejectIon

2’ Rejection

DR7 DR3 DRl DRl DR.3 DRl DRlh DR4 DR.3 DRll DRl6 DR3 DR14 DR4

DRll DRll DRl DR3

Indirect

Recognition

61

of Liver Allografts

Rec. DR : 0101,1301 Don. DR : 1101,0701

6,

60

1’R

QS

Rec. DR : 0401,1401 Don. DR : 0401,0101 A

A

119

246

2*m

as

Rec. DR : 0101,0401 Don. DR : 0301,1101

Days Clin.status

I

w

46

1.R

RR

2.BR

0 I

QS

Rec. DR: : 1101, 1101 Don. DR: : 1101,0301

6

Clin.

E

6 5 4 3 2 1 0 3

7

27

1”BR

R

PR

-Ye Clin. status

QS

Rec. DR : 0701,1301 Don. DR : 0101,1501

A,

w

RR

2’R

QS

Clin.status

Rec. DR : 1601,1601 Don. DR : OlOl,OiOl

.1

a

days

Clin. status

l*R

Baseline

RR

70 QS

/

Days Clin. status

FIGURE 2 Kinetics of allopeptide reactivity in patients with acute rejection episodes of liver allografts. PBMCs were collected from each pacienr at various cjmes after transplantation and tested in LDA for reacrlvity against peprldes correspondlog to each of the mismatched HLA-DR antigens of the donor. Th-frequency was determlned as described in reference fI 61. R, BR, RR, and respectively. QS indicate rejection, before rejection, resolving rejection, and quiescence,

62

donor APCs have migrated to the periphery Cl 1, 161. Recent studies have shown, however, that in heart allograft recipients activation of the indirect pathway of alloreactivity is an early event preceding or accompanying the development of acute rejection episodes C161. The present investigation extends these findings to liver transplantation demonstrating for the first time that allopeptide reactive T cells specific for the mismatched HLA-DR antigens of the donor are present in the recipient’s peripheral circulation during acute rejection. The highly significant correlation between rejection and allopeptide reactivity, as well as the fact that a positive LDA was almost invariably predictive of rejection, renders this assay important not only for monitoring the immune status of the host but also for differentiating rejection from other nonimmunologic events that can alter the function of the graft. Such events include infection, ischemia, biliary obstruction, and recurrent disease[4]. Although accurate differentiation of these causesis crucial to the subsequent management of the patient, there are few objective measuresof rejection. None of the immunologic parameters that have been tested so far, such as cell surface phenotype and growth capacity of mononuclear cells infiltrating the graft, reactivity of T cells from peripheral blood to donor APC, and antibody production proved useful for prediction and differential diagnosis of rejection [6, 17-191. To our knowledge measurement in LDA of T-cell alloreactivity against donor HLA peptides provides the first specific test for differential diagnosis of rejection by use of peripheral blood. The finding that significant changes in the functional makeup of circulating T cells, such as the expansion of T-cell clones reactive to donor allopeptides, take place before intragraft immune events become detectable suggeststhat the inductive phase,e.g. the afferent arm of the rejection reaction, occurs primarily in the peripheral lymphoid tissue. However, because in at least 5 patients with early (first week) rejection, allopeptide reactive T cells specific for the donor were already present in the circulation at the time of transplantation, it is possible that T-cell allosensitization has been induced by transfusions rather than by antigens shed from the graft. Allopeptide reactive T-cell clones may migrate to the graft and produce interferon (IFN-y) upregulating the expression of MHC class I and class II antigens on bile duct epithelium and central vein endothelium. These cells may then become primary targets of CD4 and CD8 effector cells with direct recognition ability Lb]. Target cell killing may further augment the amount of shed alloantigen, available for uptake, processing by host APCs and recognition by allopeptide

E. R. Molajoni

et al.

specific T cells. Allopeptide reactive cells may provide the cytokines required for DTH responses,cytotoxic cell growth, and maturation of alloantibody-producing B cells. Of particular interest is the observation that T-cell alloreactivity during primary rejection episodesof heart as well as liver allografts is directed against one rather than both HLA-DR antigens of completely mismatched donors. This biased recognition of a sole alloantigen may reflect differential processingand presentation of specific peptides, as previously suggested [lb]. Because subsequent rejections are accompanied by spreading of the responseto the second DR antigen, this finding has important implications for the design of antigen-specific immunosuppressive therapy. It is well recognized that rejection free patients acquire a state of specific hyporesponsivenesstoward their liver transplants as evidenced by the low level of immunosuppression that they require. It was suggested that specific tolerance is induced becauseof the anatomical structure and size of the liver, differences in expressionof MHC class I and II antigens, secretion of soluble HLA antigens, induction of suppressor cells, chimerism, or induction of anti-idiotypic antibodies 1201. Our finding that activation of the indirect allorecognition pathway occurs early following transplantation and is predictive of rejection suggeststhat liver transplants elicit T-cell alloreactivity against donor HLA antigens with the same or greater efficiency as other solid organs, such as the heart for example. It is possible, however, that chronic exposure of allopeptide reactive T cells to high concentrations of soluble MHC alloantigens results in the induction of high zone tolerance via a mechanism of anergy or apoptosis. If so, early treatment of transplant recipients with high concentrations of donor-specific allopeptides may represent a viable option for prevention of rejection 1211.

ACKNO\Y;LEDGMENTS

We thank Drs. G. Cimino and L. Elia for the excellent assistance in HLA molecular typing. This work was supported by grants from Consorzio Interuniversitario per i Trapianti d’organo, Rome, Italy, Consiglio Nazionale delle Ricerche, Rome, Italy and the National Institutes of Health 5-ROlAI25210-10. REFERENCES 1. Vierling M. Immunologic mechanisms of hepatic allograft rejection. Sem Liver Dis 12:16, 1992. 2. Hubscher SG. Pathology of liver Transpl Immunol 2:118, 1994.

allograft

rejection.

3. Nawaz S, Fennel1 RH. Apoptosis of bile duct epithelial cells in hepatic allograft rejection. Histopathol 25:137, 1994.

Indirect

Recognition

of Liver Allografts

63

4. Lowes JR, Hubscher SG, Neuberger JM. Chronic rejection of the liver allograft. Adv Liver Transpl 22:401, 1773. 5. Krams SM, Falco DA, Villanueva JC, Rabkin J, Tomlanovich SJ, Vincenti F, Amend WJC, Melzer J, Garovoy MR, Roberts JP, Ascher NL, Martinez OM. Cytokine and T cell receptor gene expression at the site of allograft rejection. Transplantation 53: 15 1, 1772. 6. Krams SM, Ascher NL, Martinez OM. New Immunologic insights into mechanisms of allograft rejection. Adv Liver Transpl 22:381, 1773.

15. Demetris AJ, Qian S, Sun H, Fung JJ, Yagihashi A, Murase N, Iwaki Y, Gambrel1 B, Starzl TE. Early events in liver allograft rejection. Am J Path01 138:609, 1771.

7. Knechtle SJ, Wolfe JA, Burchette J, Sanfilippo F, Bollinger RR. Infiltrating cell phenotypes and patterns associated with hepatic allogtaft rejection or acceptance. Transplantation 43: 169, 1987.

16. Liu Z, Colovai AI, Tugulea S, Reed EF, Fisher PE, Mancini D, Rose EA, Cortesini R, Michler RE, Suciu-Foca N. Indirect recognition of donor HLA-DR peptides in organ allograft rejection. J Clin Invest 98:1150, 1976.

8. Fung JJ, Zeevi A, Markus B, Zerbe TR, Duquesnoy RJ. Dynamics of allospecific T lymphocyte infiltration in vascularized human allografts. Immunol Res 5:149, 1986.

17. McCaughan W, Davies S, Waugh J, Painter D, Gallagher ND, Thompson J, Sheil AGR. Cell surface phenotype of mononuclear cells infiltrating bile ducts during acute and chronic liver allograft rejection. Transplant Proc 2 1:2201, 1987.

9. Halloran PF, Autenried P, Ramassar V, Urmson J, Cockfield S. Local T cell responses induce widespread MHC expression: evidence that IFN-y induces its own expression remote sites. J Immunol 148:3837, 1972. 10. Fung JJ, Zeevi A, Starzl TE, Demetris J, Iwatsuki S, Duquesnoy RJ. Functional characterization of infiltrating T lymphocytes in human hepatic allografts. Hum Immuno1 16:182, 1986. 11. Bradley JA. Indirect T cell recognition in allograft tion. Intern Rev Immunol 13:245, 1996.

rejec-

12. Steinman RM, Cohn ZA. Identification of a novel cell type in peripheral lymphoid organs of mice. I. Morphology, quantitation tissue distribution. J Exp Med 137: 1142, 1773. 13. Steinman

RM, Inaba K, Schuler G, Witmer

M. Stimula-

tion of the immune response: Contribution of dendritic cells. Mechanisms of host resistance to infectious agents, tumors and allografts. In: RM Steinman, RJ North (Eds) New York, Rockefeller University Press, 1786, pp 71. 14. Forbes RD, Parfrey NA, Gomresail M, Darden AG, Guttmann RD. Dendritic cell-lymphoid aggregation and major histocompatibility antigen expression during rat cardiac allograft rejection. J Exp Med 164:1237, 1786.

18. Kolbeck PC, Wood RP, Markin RS. The immunopathology and clinical relevance of lymphocyte cultures in liver transplantation. Modern Path01 6:307, 1973. 17. Farges 0, Buffello D, Shi, YM, Berth A, Bismuth H. Anti-donor antibody class switching after liver transplantation. Transplantation 6:276, 1775. 20. Duffy BF, Mathew JM, Flye MW, Mohanakuma T. Development of autoantibodies to T cell clonotypic structutes in a liver-kidney allograft recipient. Transplantation 56:212, 1973. 21. Colovai AI, Molajoni ER, Cortesini R, Suciu-Foca N. New approaches to specific immunomodulation in transplantation. Int Rev Immunol 13:161, 1776.