0041-1337/98/6607-887$03.00/0 TRANSPLANTATION Vol. 66, 887– 893, No. 7, October 15, 1998 Copyright © 1998 by Williams & Wilkins Printed in U.S.A. Fas LIGAND, TUMOR NECROSIS FACTOR- EXPRESSION, AND APOPTOSIS DURING ALLOGRAFT REJECTION AND TOLERANCE R ´ EGIS JOSIEN, 1,2 MARKUS M ¨ USCHEN, 3 EMMANUELLE GILBERT,PATRICE DOUILLARD, 4 JEAN-MARIE HESLAN,JEAN-PAUL SOULILLOU, AND MARIA-CRISTINA CUTURI 5 INSERM U437 “Immunointervention dans les Allo et X´ enotransplantations” and ITERT “Institut de Transplantation et de Recherche en Transplantation,” 44035 Nantes Cedex 01, France Background. Cytotoxic T cells can induce target cell lysis and apoptosis by different pathways. The inter- actions of CD95 antigen (Fas) with its ligand (CD95L) and of tumor necrosis factor (TNF)- with its receptor (TNF-R1) lead to apoptotic cell death. Recently, con- flicting studies have been published concerning the expression and the role of CD95L in allograft rejection and tolerance. Methods. In this study, the intragraft expression of CD95/CD95L and TNF- and the frequency and distri- bution of apoptotic cells were compared in a model of heterotopic cardiac allograft in the rat in which recip- ients were either not treated (acute rejection) or pre- treated with donor-specific blood transfusion (toler- ant). Results. In the acutely rejected allografts, a peak in the expression of CD95L and TNF- and in the number of apoptotic cells was observed during the first week after transplantation; apoptotic cells were confined to graft-infiltrating cells. In the tolerated allografts, how- ever, levels of graft-infiltrating cell apoptosis and CD95L and TNF- expression during the same period of time were dramatically lower. The expression of Fas was constitutive and was not modulated during acute rejection or tolerance. Conclusion. This down-regulation of CD95L and TNF- in allografts rendered tolerant by donor-spe- cific transfusion suggests a role for apoptosis-induc- ing pathways in acute allograft rejection. The role of anti-donor cytotoxic T lymphocytes (CTLs*) in allograft rejection and tolerance remains a subject of contro- versy (1). While some authors have demonstrated a correla- tion between the occurrence of high CTL frequency in circu- lating blood and the incidence of allograft rejection (2), specific anti-donor CTLs have also been found both in hu- mans and rodents in the blood and allografts of “tolerant” recipients (2–4). CTLs can lyse their target cells via the Ca ++ -dependent granule exocytosis pathway (perforin/gran- zyme system) and/or via the Ca ++ -independent CD95 (Fas)/ CD95 ligand (FasL) pathway (5–7). In addition, it has re- cently been shown that tumor necrosis factor (TNF)- is released by CD8 + CTLs (8) and may also contribute to CTL- mediated cytotoxicity under certain conditions, as suggested in perforin-FasL– deficient mice (9). The final mechanism used by CTLs to induce target cell lysis is apoptosis (5), which can be mediated by different molecules including granzyme/ perforin, CD95/CD95L, and TNF- (1, 6, 7). The CD95 anti- gen and the p55 TNF receptor (TNF-R1) are related mole- cules that contain a death domain in their intracytoplasmic tail and can signal apoptosis (6, 7, 10, 11). It has been sug- gested that apoptosis has a role to play in liver allograft rejection (12), but its role in other models of allograft rejec- tion has remained controversial (13, 14). Several studies have found increased expression of CD95L and TNF- during acute rejection of allografts in rodents (15, 16) and in humans (17, 18). Larsen and colleagues (15) have shown that despite the up-regulation of CD95L during acute allograft rejection, the absence of an intact CD95/CD95L pathway did not alter the tempo of rejection, suggesting that CD95/CD95L interac- tions were not essential mediators of T-induced allograft damage. They suggested, instead, a role for FasL in the regulation of the immune response (15). By contrast, using the same strain combination (Fas-deficient lpr or FasL-defi- cient gld mice as donor or recipient), Seino et al. (16) reported results that pointed to CD95L as making a substantial con- tribution to cardiac allograft rejection. Interestingly, this effect was confined to the expression of CD95/CD95L by recipient cells and was independent of CD95 expression by grafts. In a previous study, using a model of heart allograft toler- ance induced by transfusion of donor strain blood, we showed that heart allografts were tolerated despite the strong anti- donor cytotoxicity that graft-infiltrating cells displayed in vitro and the high levels of perforin and granzyme A mRNA expressed in vivo (3). In this study, using the same model, we investigated intragraft expression of CD95/CD95L and TNF-, two molecules also involved in CTL target lysis, and measured the proportion of apoptotic cells. During the first week after transplantation, a strong and transient up-regu- lation of CD95L and TNF- was observed in the rejected allografts from untreated recipients. Expression of both cor- related with the number of apoptotic cells seen in these animals. In contrast, in heart allografts from donor-specific transfusion (DST)-treated recipients, CD95L and TNF- 1 The first two authors contributed equally to this work. 2 Current affiliation: Laboratory of Cellular Physiology and Im- munology, The Rockefeller University, New York, NY 10021. 3 Current affiliation: Institut fu ¨ r Physiologische Chemie 1, Hein- rich-Heine Universita ¨t, 40225 Du ¨ sseldorf, Germany. 4 Current affiliation: Brigham and Women’s Hospital, Boston, MA 02115. 5 Address correspondence to: Maria Cristina Cuturi, INSERM U437, Immeuble Jean Monnet. 30, Boulevard Jean Monnet, 44035 Nantes Ce ´dex 01, France. E-mail: ccuturi@sante.univ-nantes.fr. * Abbreviations: CTL, cytotoxic T lymphocyte; DST, donor-specific blood transfusion; HPRT, hypoxanthinguanine phosphoribosyltrans- ferase; GIC, graft-infiltrating cells; IL, interleukin; PCR, polymerase chain reaction; RT, reverse transcriptase; TNF, tumor necrosis fac- tor; TUNEL, terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling. 887