TRANSPLANTATION OF CD95 LIGAND-EXPRESSING GRAFTS: Influence ... : Transplantation (original) (raw)
CD95 (also called Fas or APO-1) and its ligand (CD95L*, FasL, APO-1L) are cell-surface proteins involved in apoptosis. Their interaction has been implicated in the regulation of immune responses (1, 2). CD95 is a type I membrane protein belonging to the tumor necrosis factor receptor family and expressed on hematopoietic cells and in various tissues, including the liver, lung, intestine, and skin. CD95L is a type II integral membrane protein homologous to tumor necrosis factor and mainly expressed on activated T and NK cells.
CD95L is also constitutively expressed in rodent testis (3, 4) and eye (5). Previously, Bellgrau et al. (4) reported that testicular allografts from normal mice were not efficiently rejected when transplanted under the kidney capsule of MHC-unmatched hosts, whereas testicular allografts from gld mice, which do not express functional CD95L, were rejected. Their results suggested that CD95L expressed in the testis seems to induce apoptotic cell death of CD95-expressing recipient T cells activated in response to the allograft antigens and thus prevent the graft rejection. This led to an expectation that an organ genetically engineered to express CD95L could be used as a graft evading the rejection. Recently, Lau et al. (6) demonstrated that the engraftment of pancreatic islet allografts was facilitated by cotransplantation of syngeneic myoblasts genetically engineered to express CD95L. Griffith et al. (5) indicated that CD95L in the anterior chamber of the eye acted as an anti-inflammatory agent in a murine virus-induced keratitis model by inducing apoptosis of infiltrating inflammatory cells.
Contradictory to these observations, we recently found that CD95L-expressing cells were acutely rejected when subcutaneously (s.c.) or intraperitoneally (i.p.) transplanted in nu/nu or syngeneic mice (7, 8). The rejection was T cell-independent and seemed to be mediated mainly by granulocytes because it was abrogated by depletion of granulocytes and a severe neutrophil infiltration was observed at the rejection site (8).
In this study, we investigated whether the CD95L-expressing cells that were rejected when inoculated s.c. or i.p. could be accepted when transplanted under the kidney capsule where testicular allografts were accepted. The results indicate that the kidney capsule is a peculiar site where the CD95L-mediated inflammatory rejection does not occur efficiently, but the CD95L-expressing xeno- or allografts could not be protected from rejection even at this site, suggesting a limited immunosuppressive effect of CD95L.
MATERIALS AND METHODS
Mice. Five-week-old female DBA/2, C57BL/6 (B6), BALB/c, and BALB/c nu/nu mice were purchased from Charles River Japan Inc. (Atsugi, Japan) and used as recipients.
Cells. Baby hamster kidney (BHK) fibroblast cell line and a thymic lymphoma cell line, L5178Y of DBA/2 origin, were obtained from American Tissue Culture Collection (Rockville, MD). These cells are CD95-negative as assessed by fluorescence-activated cell sorting and resistant to CD95L-induced lysis (8 and data not shown). Human and mouse CD95L (hCD95L and mCD95L) transfectants from these cells were generated as previously described (9). Briefly, hCD95L and mCD95L cDNA were obtained by reverse transcription-polymerase chain reaction from total RNA of activated T cells by using the CD95L 5′ and 3′ primers according to the published sequence. The polymerase chain reaction products were transferred into a mammalian expression vector and then transfected into BHK and L5178Y by electroporation. The transfectants (hCD95L/BHK, mCD95L/BHK, hCD95L/L5178Y, and mCD95L/L5178Y) were selected with 1 mg/ml G418 and cloned by limiting dilution. The cells were maintained in RPMI 1640 medium (Nissui, Tokyo, Japan) containing 10% fetal bovine serum, 2 mM glutamine, 100 μg/ml streptomycin, and 100 U/ml penicilin.
3H-TdR release assay. Cytotoxic activity of the parental cells and the CD95L transfectants (5×104 cells) was tested against a mouse T lymphoma WR19L and its human CD95 transfectant (CD95/WR19L) (1×104 cells), which were prelabeled with 3H-TdR in the presence or absence of 10 μg/ml anti-mCD95L monoclonal antibody (mAb) (K10; 10). After a 6-hr incubation at 37°C, intact cells were harvested using a Micro 96 Harvester (Skatron, Lier, Norway), and radioactivity was measured on a microplate beta counter (Micro Beta Plus, Wallac, Turku, Finland). Percent cytotoxicity was calculated as follows: {(cpm without effector-cpm with effector)/cpm without effector}× 100.
Grafting. For subcutaneous transplantation, 2×106 cells in 0.2 ml of phosphate-buffered saline were injected into the backs of mice. Some mice were s.c. injected with mCD95L transfectants were i.p. administered with 200 μg of anti-mCD95L mAb (K10) three times per week. For the transplantation under the kidney capsule, cells were centrifuged, and the pellets containing about 1×105 cells were transplanted into the renal subcapsular space as described for pancreatic islet graft (11). Recovered grafts were fixed in 10% formalin and embedded in paraffin. Sections (5 μm) were stained with hematoxylin-eosin for light microscopic examination. The graft survival was estimated by visual inspection and confirmed by histological examination.
RESULTS
The CD95L transfectants we established grew well in vitro. No difference in the growth rate was observed between hCD95L/BHK or mCD95L/BHK and parental BHK cells (Fig. 1A). The hCD95L- and mCD95L-transfected BHK cells exhibited high CD95-dependent cytotoxic activity, but the parental BHK cells did not (Fig. 1B), which indicates that these CD95L transfectants express functional CD95L. An anti-mCD95L mAb (K10) completely blocked the cytotoxic activity of mCD95L/BHK cells (Fig. 1B). Similar growth rates and cytotoxic activities of L5178Y cells and its CD95L transfectants were also observed as previously demonstrated (8).
In Table 1, survival of s.c. transplanted cells is indicated as tumor growth. As we previously reported (7, 8), when the CD95L-bearing BHK or L5178Y cells were s.c. transplanted to immunodeficient BALB/c nu/nu or syngeneic DBA/2 mice, respectively, they were completely rejected, whereas the parental cells were accepted and grew progressively. Administration of an anti-mCD95L mAb reversed the rejection of mCD95L/BHK or mCD95L/L5178Y cells, as anti-hCD95L mAb (NOK-1) did against hCD95L transfectants (8), which indicates that the rejection was induced by the transfected CD95L. When these CD95L transfectants were s.c. transplanted in immunocompetent xenogeneic BALB/c or allogeneic B6 mice, they were rejected with no delay compared with the rejection of their parental cells. Histological examination at 7 days after the transplantation in nu/nu mice showed survival of the parental BHK cells with mitoses (Fig. 2A). In contrast, the CD95L transfectants were scattered, and massive neutrophil infiltration was noted (Fig. 2B) as previously described (7).
Then, we transplanted the cells under the kidney capsule where testicular allografts were accepted without any immunosuppression (4) and observed their survival for 3 weeks (Table 2). When parental BHK or L5178Y cells and their CD95L transfectants were transplanted under the kidney capsule of nu/nu or syngeneic DBA/2 mice, respectively, both grew with mitoses and formed solid tumors that were almost free from cellular infiltrates (Fig. 2, C and D). However, it was noted that the growth of CD95L transfectants was generally slower than that of parental cells, and some of the CD95L/L5178Y grafts disappeared by day 21 without notable cellular infiltration.
When the parental cells and their CD95L transfectants were transplanted under the kidney capsule of xenogeneic BALB/c or allogeneic B6 mice, all of them were acutely rejected with marked polymorphonuclear and mononuclear cell infiltration (Fig. 2, E and F), and no tumor formation was observed at day 7 (Table 2). To confirm the subcutaneous rejection, we injected hCD95L/BHK cells in the skin of nu/nu mice that were holding the hCD95L/BHK graft under the kidney capsule. The s.c. implanted cells were completely rejected, while the CD95L transfectants were still maintained under the kidney capsule (not shown).
DISCUSSION
Local elimination of activated T cells by CD95L expressed on transplanted grafts has been proposed to be a beneficial strategy for preventing rejection. This notion was suggested by recent observation in testicular allografts (4). In the present study, we used xenogeneic fibroblasts and allogeneic T lymphoma cells as CD95L-expressing grafts and found that they were not protected from allogeneic or xenogeneic rejection when transplanted s.c. or under the kidney capsule. However, it was reported that coexpression of CD95L by syngeneic cells protected the nearby transplanted allografts from rejection (6). Thus, it seems likely that direct expression of CD95L on the graft may not be solely sufficient to prevent immunological rejection. Testicular grafts might have some additional factor to CD95L, which contributes to protecting rejection.
Recently, Allison et al. (12) generated transgenic mice which express CD95L on their pancreatic β cells and transplanted the CD95L-expressing islets under the kidney capsule of allogeneic mice. Unexpectedly, survival of the CD95L-expressing islet allografts was not prolonged at all. In these grafts, massive lymphocytic infiltrates were noted as well as in control nontransgenic grafts. Their results also indicate a difficulty in protecting rejection by direct CD95L expression on the grafts.
There are substantial similarities between the observations by Allison et al. (12) and those of ours. They reported that the CD95L-expressing islets maintained under the kidney capsule of the syngeneic hosts were smaller than nontransgenic grafts, and 50% of them finally disappeared by day 30 after transplantation. This is consistent with our present observation that CD95L-expressing L5178Y cells grew slower under the kidney capsule of syngeneic host, and some of them disappeared by day 21 after transplantation. It has been reported that CD95 can be induced on mouse islet cells by interleukin-1 and possibly other cytokines (13) that can induce the death of CD95L-expressing mouse islet cells by engagement of the CD95L. Although we could not detect CD95 expression on L5178Y cells in vitro (8) and in vivo by immunohistochemical analysis (data not shown), it is possible that micro-expression of CD95 was induced was induced on some of these cells in vivo, which may have led to a slow process of their death. Alternatively, CD95L expression itself might result in the damage of the transfectants grafted under the kidney capsule, as it has been reported that high levels of expression of some transgenes are toxic for islet β cells (14). Furthermore, Allison et al. found a granulocytic infiltration in the pancreata of newborn CD95L transgenic mice (12). Similarly, our CD95L-expressing cells induced a granulocytic infiltration when transplanted s.c. (Fig. 2B), which seemed to be a direct action of CD95L upon granulocytes (8).
In this study, we demonstrated the CD95L transfectants that were s.c. rejected could be accepted when transplanted under the kidney capsule of nu/nu or syngeneic mice. This suggests that the CD95L-mediated, T cell-independent inflammatory rejection could not be efficiently operative in the renal subcapsular space. This is consistent with the observation by Lau et al. (6) that syngeneic myoblasts expressing CD95L were maintained under the kidney capsule over 44 days after transplantation. Historically, the renal subcapsular space has been considered to be a beneficial site for transplantation. For example, syngeneic islet grafts were maintained under the kidney capsule while the same grafts were s.c. rejected (15). However, it remains unclear why the CD95L-mediated granulocytic infiltration was not operative under the kidney capsule. Some local tissue factor may play a key role in the establishment and maintenance of immune privilege, which modifies the induction and enhancement of inflammation. Especially, different tissues might produce different cytokines in response to a stimulation. For example, transforming growth factor-β, which has been implicated in the immune-privileged status of the anterior chamber (16), is known to suppress neutrophil functions and can be produced by kidney cells (17). Such a suppressive factor may contribute to the prevention of CD95L-mediated inflammation in the renal subcapsular space. Alternatively, it is possible that some cytokine that is specifically produced by skin cells might enhance the CD95L-mediated inflammation.
The mice transplanted with CD95L transfectants showed no systemic abnormalities, as also demonstrated by Lau et al. (6). The strategy facilitating the graft acceptance by local expression of CD95L without inducing systemic toxicity is interesting in the light of clinical transplantation. However, it seems difficult to achieve this instantly, as CD95L-expressing cells and pancreatic islets (12) were completely rejected in allogeneic and xenogeneic recipients even when transplanted under the kidney capsule; this is possibly due to host immune responses to transplantation antigens. Furthermore, it should be considered that the rejection of CD95L-expressing cells in syngeneic hosts is substantially influenced by the site of transplantation.
In conclusion, our present results indicate that CD95L-expressing syngeneic grafts can be maintained in the renal subcapsular space, but CD95L-expressing allo- and xenografts cannot be protected from immunological rejection. Elucidation of some additional factor that confers the CD95L-mediated immune privilege on the testis and eye would be helpful to establish the CD95L-based strategy for graft protection.
Acknowledgments. The authors thank Dr. J. Allison for helpful discussions and critical reading of the manuscript and Misa Seino for secretarial assistance.
Characterization of CD95L transfectants. (A) In vitro growth of BHK, mCD95L/BHK, and hCD95L/BHK cells. The cells were seeded at 2×104/ml on day 0 and cultured for 4 days. Number of cells was counted daily. Data indicate mean ± SD of three independent cultures. (B) Cytotoxic activity of BHK, mCD95L/BHK, and hCD95L/BHK cells. Cytotoxic activity of BHK and the CD95L transfectants (5×104 cells) was tested against CD95- WR19L and its human CD95 transfectants (CD95/WR19L) (1×104 cells) by a 6-hr 3H-TdR release assay in the presence or absence of 10 μg/ml anti-mCD95L mAb (K10). Data indicate mean ± SD of triplicate wells. Representative data are from three independent experiments.
Histological examination of the grafts. BHK (A, C, E) or hCD95L/BHK (B, D, F) cells were s.c. transplanted (A, B) or placed under the kidney capsule (C-F) in BALB/c nu/nu (A-D) or BALB/c mice (E, F). Seven days (A, B, E, F) or 21 days (C, D) after the transplantation, the specimens were dissected, and paraffin sections were stained with hematoxylin and eosin. The arrows indicate glomerulus. Original magnification: A and B, ×200; C and D, ×50; E and F, ×100.
Footnotes
This work was supported by grants from the Science and Technology Agency, the Ministry of Education, Science and Culture, and the Ministry of Health, Japan.
Abbreviations: BHK, baby hamster kidney; CD95L, CD95 ligand; hCD95L, human CD95L; i.p., intraperitoneally; mAb, monoclonal antibody; mCD95L, mouse CD95L; s.c., subcutaneously.
REFERENCES
1. Nagata S, Golstein P. The CD95 death factor. Science 1995; 267: 1449.
2. Yagita H, Hanabuchi S, Asano Y, et al. Fas-mediated cytotoxicity: a new immunoregulatory and pathogenic function of Th1 CD4+ T cells. Immunol Rev 1995; 146: 223.
3. Suda T, Takahashi T, Golstein P, et al. Molecular cloning and expression of the CD95 ligand, a novel member of the tumor necrosis factor family. Cell 1993; 75: 1169.
4. Bellgrau D, Gold D, Selawry H, et al. A role for CD95 ligand in preventing graft rejection. Nature 1995; 377: 630.
5. Griffith TS, Brunner T, Fletcher SM, et al. CD95 ligand-induced apoptosis as a mechanism of immune privilege. Science 1995; 270: 1189.
6. Lau HT, Yu M, Fontana A, et al. Prevention of islet allograft rejection with engineered myoblasts expressing CD95L in mice. Science 1996; 273: 109.
7. Yagita H, Seino K, Kayagaki N, et al. CD95 ligand in graft rejection. Nature 1996; 379: 682.
8. Seino K, Kayagaki N, Okumura K, et al. Antitumor effect of locally produced CD95 ligand. Nature Med 1997; 3: 165.
9. Kayagaki N, Kawasaki A, Ebata T, et al. Metalloproteinase-mediated release of human CD95 ligand. J Exp Med 1995; 182: 1777.
10. Kayagaki N, Yamaguchi N, Nagao F, et al. Polymorphism of murine CD95 ligand that affects the biological activity. Proc Natl Acad Sci USA 1997; 94: 3914.
11. Gotoh M, Maki T, Kiyoizumi T, et al. An improved method for isolation of mouse pancreatic islets. Transplantation 1985; 40: 437.
12. Allison J, Georgiou HM, Strasser A, et al. Transgenic expression of CD95 ligand on islet β cells induces a granulocytic infiltration, but does not confer immune privilege upon islet allografts. Proc Natl Acad Sci USA 1997; 94: 3943.
13. Yamada K, Takane-Gyotoku N, Yuan X, et al. Mouse islet cell lysis mediated by interleukin-1-induced Fas. Diabetologia 1996; 39: 1306.
14. Allison J, Malcolm L, Culvenor J, et al. Overexpression of β2-microglobulin in transgenic mouse islet β cells results in defective insulin secretion. Proc Natl Acad Sci USA 1991; 88: 2070.
15. Kemp CB, Knight MJ, Scharp DW, et al. Effect of transplantation site on the results of pancreatic islet isografts in diabetic rats. Diabetologia 1973; 9: 486.
16. Wilbanks GA, Mammolenti M, Streilein JW. Studies on the induction of anterior chamber-associated immune deviation (ACAID): III. Induction of ACAID depends upon intraocular transforming growth factor-β. Eur J Immunol 1992; 22: 165.
17. Sharma K, Ziyadeh FN. The transforming growth factor-β system and the kidney. Semin Nephrol 1993; 13: 116.
© Williams & Wilkins 1997. All Rights Reserved.