Impaired Recall of CD8 Memory T Cells in Immunologically Privileged Tissue
http://www.100md.com
免疫学杂志 2005年第1期
Abstract
Foreign Ags that enter immunologically privileged sites such as the eye, brain, and testis persist for an extended period of time, whereas the same Ags are rapidly eliminated at conventional sites. Immune privilege, therefore, provides unwanted refuge for pathogens and tumor cells but is beneficial for the survival of allogeneic grafts. In this study, we asked whether memory T cells can eliminate foreign Ags deposited at an immunologically privileged site by studying CD8 memory T cell-mediated rejection of pancreatic islet allografts placed either in the testis (a privileged organ) or under the kidney capsule (a nonprivileged site) of diabetic mice. We found that CD8 memory T cells reject intratesticular grafts at a significantly slower rate than the rejection of intrarenal grafts. Delayed graft rejection in the testis was not due to reduced homing or proliferation of memory T cells but due to their increased apoptosis at that site. Apoptosis was mediated by the combined actions of two TNFR family members that are up-regulated on activated memory T cells, Fas, and CD30. Therefore, memory T cells survey immunologically privileged tissues but are subject to the immunosuppressive mechanisms present at these sites.
Introduction
Foreign Ags deposited in areas of the body known as immunologically privileged sites persist indefinitely or are eliminated slowly by the immune system (1, 2). Typical examples of immunologically privileged organs are the testes, ovaries, brain, and eyes, where immune privilege is maintained by mechanisms that attenuate both innate and adaptive immune responses. These mechanisms include reduced migration of dendritic cells to the lymph nodes and spleen due to paucity of draining lymphatics or the presence of blood-tissue barriers (3); the production of immunosuppressive and anti-inflammatory cytokines such as TGF- (4); and the local expression of death-inducing molecules, such as Fas ligand (FasL),3 which eliminate effector lymphocytes (5). Although the mechanisms of immune privilege were originally identified in the anterior chamber of the eye (2), they are also relevant to other privileged organs. For example, Sertoli cells of the testis express both TGF- and FasL (although the latter is controversial), thus creating an environment in which foreign Ags can persist either indefinitely or for an extended duration (6, 7). It is postulated that immune privilege confers a survival advantage on the organism by protecting vital organs from excessive inflammation if infection were to occur. In contrast, immune privilege poses a challenge to the host’s immune system by providing a refuge for unwanted tumor cells and pathogens. This has been demonstrated in HSV infection, whereby entry of the virus into the anterior chamber of the eye leads to tolerance rather than immunity to the viral Ags (8).
In this paper, we asked whether memory T cells, which are central to host immune protection, are capable of seeking out and eliminating foreign Ags at immunologically privileged sites. Unlike naive T cells, memory T cell activation is not dependent on secondary lymphoid organs as they can migrate and respond to Ag in nonlymphoid, peripheral tissues (9, 10). For example, memory T cells cause the rejection of allogeneic grafts in a mouse that lacks all secondary lymphoid organs, whereas naive T cells fail to do so (11, 12), indicating that memory T cells can eliminate foreign Ags that reside exclusively in nonlymphoid compartments. However, it is not known whether memory T cells enter immunologically privileged sites in sufficient numbers to mount a protective immune response, or whether, once they have reached an immunologically privileged site, their response is attenuated by local immunosuppressive mechanisms. Using a murine pancreatic islet cell transplantation model in which allograft rejection is mediated by CD8 memory T cells, we provide direct evidence that CD8 memory T cells migrate efficiently to an immunologically privileged site (the testis) but mount an attenuated response at that site.
Materials and Methods
Mice
B6 mice homozygous for the mutation that leads to alymphoplasia (aly/aly) were purchased from CLEA Japan (13). 2C TCR-transgenic B6 mice on a RAG-2 knockout (Rag2–/–) background (2C.Rag–/–) were generated by backcrossing 2C transgenic mice onto Rag2–/– mice (The Jackson Laboratory). CD30–/– B6 mice were a gift from Dr. T. Mak (University of Toronto, Toronto, Canada). lpr, wild-type (wt) BALB/c, and wt B6 mice were purchased from The Jackson Laboratory. All mice were housed in a specific pathogen-free environment.
Pancreatic islet isolation
Islet donors were 6- to 8-wk-old BALB/c (H-2d) mice. Donor mice were anesthetized, and a median laparotomy was performed. The common bile duct was exposed and clamped at the site where it feeds into the duodenum. A volume of 3 ml of collagenase V (Sigma-Aldrich) (1 mg/ml in PBS) was injected into the common bile duct to distend the pancreas and initiate enzymatic digestion. The pancreas was then removed, minced, and subjected to stationary digestion by incubation in a 37°C water bath for 13 min (14). A volume of 10 ml of cold HBSS with 10% calf serum was added to stop digestion, and the pancreatic tissue was shaken vigorously for 30 s to suspend the islets. Islets were then washed three times in ice-cold HBSS by centrifugation at 1000 rpm for 5 min, and the crude preparation was filtered through a hand-held 100-μm nylon cell strainer (BD Biosciences). After flushing with HBSS, the strainer was turned upside-down over a petri dish and rinsed with HBSS to wash the islets into the dish. Islets were counted and picked up using a 23G needle.
Mouse splenectomy and islet transplantation
Allograft recipients were 6- to 8-wk-old splenectomized aly/aly B6 mice or wt B6 mice (H-2b). Splenectomy was performed by opening the abdomen between the left inferior rib and hip under anesthesia, and the entire spleen was removed intact after ligating the splenic vein and artery at the hilum. The peritoneal membrane and skin were then closed separately. Two weeks later, splenectomized mice underwent islet transplantation either under the kidney capsule or in the testis. Briefly, mice were anesthetized, and a right flank incision was made. Islets (350–400 per mouse) were injected into the subcapsular space of the right kidney using a 23G needle. The needle was pulled out slowly, and the incision was closed with skin staples. For intratesticular islet transplantation, the same number of islets were injected into one testis using a 23G needle. A lower abdominal incision was then made, and both testicles were pulled into the abdominal cavity through small openings in the inguinal areas. These openings were then sutured to prevent the testicles from migrating back into the scrotum. Recipient mice were rendered diabetic by a single injection of streptozotocin (Sigma-Aldrich) (180 mg/kg weight) 10–14 days before transplantation. Primary graft function was defined as blood glucose under 200 mg/dl at 24 h after transplantation. Graft rejection was defined as a rise in blood glucose to >300 mg/dl for 3 consecutive days in mice that had primary graft function. Differences in allograft survival were analyzed using the Mann-Whitney U test.
Memory T cell preparation and adoptive transfer
Memory CD8+ T cells were purified from B6 wt, transgenic 2C-Rag–/–, or CD30–/– spleens 10 wk after immunization with 1 x 107 BALB/c RBC-lysed splenocytes per mouse (15). Briefly, CD8+ T cells were positively selected by autoMACS (Miltenyi Biotec) and incubated with anti-CD8-PE and anti-CD44-FITC Abs (BD Pharmingen), and the memory population was sorted after gating on CD8+CD44high cells (FACSVantage; BD Biosciences). The purity of CD8+CD44high cells was typically >95%. The memory phenotype of this population was further confirmed by staining with anti-CD62L-allophycocyanin or anti-CD25-allophycocyanin (BD Pharmingen) (11). 2C memory cell fractions were assessed by first incubating the cells with 1B2 (mouse IgG1, a clonotypic mAb recognizing the transgenic TCR) followed by anti-mouse IgG1-biotin and streptavidin-PerCP (Zymed). A total of 1 x 106 wt, CD30–/–, or 2C memory CD8+ T cells were adoptively transferred to recipient mice via tail vein injection 24 h after islet transplantation.
In vivo Ab treatment
To block FasL-Fas and CD30L-CD30 interactions, splenectomized aly/aly mice, transplanted with BALB/c islets 1 day earlier, were injected i.p. with 0.125 mg of purified hamster anti-mouse FasL mAb (clone MF3; BD Pharmingen) and/or 0.125 mg of anti-CD30L mAb (BD Pharmingen) on days 0, 2, 4, and 6 relative to the adoptive transfer of memory CD8+ T cells. To block T cell costimulation and prevent acute allograft rejection mediated by primary alloimmune responses, wt mice were treated with 0.25 mg of MR1 (anti-CD40L mAb) and CTLA4-Ig (generated in the laboratory of C. P. Larsen) on days 0, 2, 4, and 6 relative to islet transplantation.
Isolation of graft-infiltrating T cells
Graft-infiltrating cells were isolated as described previously (16). Briefly, the kidney and testis transplanted with islet allografts were first perfused in situ with heparinized 0.9% saline. They were then minced and digested at 37°C for 30 min in 20 ml of RMPI 1640 medium containing 10% FCS and 250 U/ml collagenase (Sigma-Aldrich). Cells were washed twice with HBSS after digestion. To clear the debris, cell suspensions were rapidly passed down a loosely packed glass wool column (300 mg of sterile glass wool in a 10-ml syringe), mixed with Percoll solution (Sigma-Aldrich) to a concentration of 30%, and centrifuged at 2000 rpm for 15 min at room temperature. The pellet was washed and resuspended before flow analysis. For detection of T cell apoptosis, cells were analyzed directly without Percoll purification.
BrdU labeling of T cells
Splenectomized aly/aly mice were pulsed i.p. with 0.8 mg of BrdU (Sigma-Aldrich) 7 days after transplantation (6 days after T cell adoptive transfer). One day later, mice were sacrificed and graft-infiltrating cells were isolated as described above. Cell surface staining was performed using anti-CD8-PE, 1B2, anti-mouse IgG1-biotin, and streptavidin-PerCP. Cells were then fixed in 70% ethanol followed by 1% paraformaldehyde and incubated with 50 U/ml DNase I (Sigma-Aldrich) for 10 min at room temperature. Cells were finally stained with anti-BrdU-FITC Ab (BD Biosciences) and analyzed by flow cytometry.
Analysis of T cell apoptosis
Splenectomized aly/aly mice were sacrificed 8 days after transplantation (7 days after T cell adoptive transfer), and graft-infiltrating cells were isolated as above. Cell surface staining was performed using anti-CD8-PE, 1B2, anti-mouse IgG1-biotin, and streptavidin-PerCP. Cells were then fixed in 2% paraformaldehyde, permeabilized with 0.1% Triton X-100 in 0.1% sodium citrate, and labeled with fluorescein-tagged dUTP by the TUNEL method according to the manufacturer’s instructions (In Situ Cell Death Detection kit; Boehringer Mannheim). Cells were finally analyzed by flow cytometry.
Results
CD8 memory T cells cause delayed elimination of foreign Ag at an immunologically privileged site
To investigate whether memory T cells eliminate foreign Ags at an immunologically privileged site, we used an established murine transplantation model in which allograft rejection is mediated by memory T cells. In this model, allogeneic tissue is transplanted to splenectomized alymphoplastic (aly/aly) mice that lack all secondary lymphoid organs (12, 13). These mice do not mount a primary alloimmune response and do not reject an allograft unless they receive allospecific effector or memory T cells (11, 12). Therefore, we transplanted BALB/c (H-2d) pancreatic islet cells either in the testis (an immune-privileged organ) or under the kidney capsule (a nonprivileged site) of splenectomized aly/aly mice (H-2b) that had been rendered diabetic by streptozotocin injection. Euglycemia (blood glucose, <200 mg/dl) within 24 h of transplantation was indicative of technically successful transplantation, at which time the mice received CD8+CD44high memory T cells isolated from B6 mice immunized with BALB/c splenocytes 10 wk earlier. Control recipients received an equal number of CD8+CD44low naive B6 T cells. Allograft rejection, defined as blood glucose >300 mg/dl for 3 consecutive days, was confirmed by histology. As shown in Fig. 1a, allograft rejection was not observed in control mice that received naive T cells (median survival time (MST) >120 days). This result is consistent with our previous studies in which adoptively transferred naive T cells failed to reject either skin or heart allografts in splenectomized aly/aly recipients (11, 12). In contrast, rejection was observed in all mice that received CD8 memory T cells, irrespective of whether pancreatic islet cell grafts were placed in the testis or under the kidney capsule. However, the rejection of intratesticular grafts was significantly delayed compared with that of intrarenal grafts (MST = 39 vs 16 days, respectively; p < 0.05).
FIGURE 1. Pancreatic islet allograft survival following the transfer of memory or naive CD8 T cells. BALB/c islets were transplanted in the testis or under the kidney capsule of splenectomized aly/aly (a) or wt (b) B6 mice rendered diabetic by streptozotocin injection. One day later, CD8 memory (CD44high) or CD8 naive (CD44low) T cells were transferred to islet recipients. Islet allograft survival was monitored by measuring blood glucose levels. Blood glucose >300 mg/dl for 3 consecutive days was indicative of allograft rejection. wt mice also received CTLA4-Ig and MR1 (anti-CD40L mAb) at the time of transplantation to block naive T cell responses.
The results shown in Fig. 1a were generated in aly/aly mice that are homozygous for a mutation in the NF-B-inducing kinase (17). In addition to the absence of secondary lymphoid tissues, this mutation could potentially influence allograft rejection mediated by exogenous effector or memory T cells (18). To rule out this possibility, we transferred CD8 memory T cells to wt B6 recipients 2 days after transplanting BALB/c pancreatic islet cells either in the testis or under the kidney capsule. The recipients were treated at the time of transplantation with CTLA4-Ig and MR1 (anti-CD40L mAb) to block the CD28/B7 and CD40/CD40L costimulatory pathways, respectively. Because these pathways are required for the activation of naive but not Ag-experienced T cells (19, 20), the transfer of memory but not naive T cells is expected to mediate rejection in this model. As shown in Fig. 1b, naive T cells did not cause rejection (MST > 120 days), whereas CD8 memory T cells caused rapid rejection of intrarenal grafts (MST = 14 days) and delayed rejection of intratesticular grafts (MST = 44 days; p < 0.05). These findings confirm our results in the splenectomized aly/aly model (Fig. 1a). Therefore, CD8 memory T cells are capable of eliminating foreign Ags at an immunologically privileged organ, albeit at a significantly slower rate than that observed at a nonprivileged site.
Ag-dependent homing of CD8 memory T cells to an immunologically privileged site
Because immunologically privileged sites possess tissue-blood barriers that interfere with lymphocyte migration (2), we asked whether delayed rejection of intratesticular allografts was due to reduced homing of CD8 memory T cells to the testis. To answer this question, we adoptively transferred CD8+CD44high memory T cells (harvested from Thy1.1+ B6 mice immunized with BALB/c splenocytes) to Thy1.2+ B6 wt mice 1 day after transplanting pancreatic islet cells either in the testis or under the kidney capsule. An equal number of memory T cells was transferred to control mice injected with PBS at the same sites. Thy1.1+CD8+ cells that infiltrated the transplantation/injection site were quantitated 24 and 48 h after cell transfer. As shown in Fig. 2, comparable numbers of these cells were found in the testis and the kidney at both time points, indicating that CD8 memory T cell homing to immunologically privileged sites is not impaired. The migration of CD8 memory T cells to either the testis or the kidney, however, was dependent on the presence of alloantigen (islets) at these sites (Fig. 2).
FIGURE 2. Alloantigen-dependent migration of CD8 memory T cells to the testis and the kidney. Thy1.1+CD8+ memory T cells (CD44high) were transferred to Thy1.2+ wt B6 mice 1 day after transplanting BALB/c islets in the testis or under the kidney capsule. Control mice were injected with PBS at the same sites. Thy1.1+CD8+ T cells that infiltrated the transplantation/injection site were quantitated by flow analysis 24 and 48 h after cell transfer. Results are shown as percentage of gated lymphocytes (a) or as absolute number (b) of infiltrating CD8+Thy1.1+ T cells per site. In the experiment depicted here, infiltrating T cells were pooled from the testes or kidneys of three mice per group. The results were confirmed in two additional experiments (n = 3).
Enhanced apoptosis of CD8 memory T cells at an immunologically privileged site
To further explore why memory T cell recall is attenuated at an immunologically privileged site, we studied the proliferation and apoptosis of allospecific CD8 memory T cells that migrated to the testis and the kidney after islet cell transplantation. In these experiments, 2C CD8 TCR-tg memory T lymphocytes, which react to Ld MHC class I Ag on BALB/c cells (21), were transferred to splenectomized aly/aly mice 1 day after transplantation, and their in vivo proliferation and apoptosis were studied 7 days later. 2C CD8 TCR-tg T cells are identified by the clonotypic Ab 1B2 and, therefore, are referred to as CD8+1B2+ (21). As shown in Fig. 3, 50% of CD8+1B2+ cells had undergone proliferation 1 wk after encountering alloantigen in either the testis or the kidney (52 ± 2 and 51 ± 3%, respectively; n = 4/group) (Fig. 3). In contrast, their apoptosis was 3-fold higher in the testis than in the kidney (20 ± 3 vs 7 ± 2%, respectively; n = 4/group) (Fig. 3). Increased T cell apoptosis rate in the testis correlated with lower recovery of graft-infiltrating, allospecific 2C T cells from that site at 1 wk posttransfer (0.7 ± 0.1 x 104 CD8+1B2+ cells recovered from the testis vs 2 ± 0.2 x 104 from the kidney; n = 4/group). These results indicate that delayed elimination of foreign Ag at an immunologically privileged site is due to enhanced apoptosis, and not reduced proliferation, of activated, Ag-specific CD8 memory T cells.
FIGURE 3. CD8 memory T cell proliferation and apoptosis in the testis and the kidney. CD8 TCR-tg (2C) memory T cells (CD44high) were transferred to splenectomized aly/aly mice 1 day after transplanting BALB/c pancreatic islets in the testis or under the kidney capsule. Infiltrating T cells were harvested from the testes or kidneys 1 wk after cell transfer, and their proliferation and apoptosis were quantitated by BrdU uptake and TUNEL labeling, respectively, after gating on the CD8 TCR-tg cells (CD8+1B2+ cells shown in top panel). In the experiment depicted here, infiltrating T cells were pooled from the testes and kidneys of two mice per group. Results were confirmed in three additional experiments.
Enhanced apoptosis of CD8 memory T cells at an immunologically privileged site is mediated by Fas and CD30
FasL expression in the testis and anterior chamber of the eye contributes to immune privilege by inducing the apoptosis of effector T cells (5). We therefore examined whether FasL-Fas interaction is responsible for enhanced apoptosis of memory T cells activated in the testis by treating transplanted mice with anti-FasL blocking Ab at the time of transferring 2C CD8 TCR-tg memory T cells. In these experiments, FasL blockade neither reduced the rate of Ag-specific CD8 memory T cell apoptosis in the testis (Fig. 4a), nor significantly accelerated the rejection of intratesticular islet allografts when compared with the mouse group that received isotype control Ab (b). Importantly, CD8+ memory T cells isolated from alloimmunized lpr (Fas mutant) mice also failed to significantly accelerate the rejection of intratesticular islet allografts despite a trend toward shorter graft survival in this group as well as in mice that received anti-FasL Ab. We then asked whether another member of the TNFR family, CD30, contributes to the death of activated CD8 memory T cells. CD30 is expressed on activated T cells, delivers apoptotic signals, and has been recently shown to mediate the suppression of CD8 memory T cells by CD4+CD25+ regulatory T cells (15, 22, 23). After transplanting islets into splenectomized aly/aly mice, CD30–/– CD8 memory T cells underwent apoptosis in the testis and mediated the rejection of islet allografts with the same reduced tempo as wt CD8 memory T cells (Fig. 4, a and b). The inability of either FasL-Fas or CD30L-CD30 blockade alone to significantly alter apoptosis or rejection could not be attributed to the lack of Fas and CD30 on activated memory T cells because both molecules were present on adoptively transferred, allospecific (CD8+1B2+) T cells that infiltrated the testis following transplantation (Fig. 4c). Importantly, the apoptosis of memory T cells was significantly reduced, and allograft survival in the testis was significantly shortened, when anti-FasL was administered to mice that received CD30–/– CD8 memory T cells or when FasL and CD30L were simultaneously blocked at the time of memory T cell transfer (Fig. 4, a and b). Commensurate with reduced T cell apoptosis, more 2C T cells were recovered from the testis when both Fas and CD30 were blocked (2 x 104 CD8+1B2+ cells vs 0.8 x 104 in isotype control Ab-treated mice; n = 2/group). Isotype control Ab did not accelerate allograft rejection mediated by CD8 memory T cells (MST = 37 days; range, 33–48 days; n = 4). Therefore, at least two death receptors, Fas and CD30, cooperate to induce the apoptosis of activated CD8 memory T cells at an immunologically privileged site.
FIGURE 4. Role of Fas and CD30 in CD8 memory T cell apoptosis in the testis. a, CD8 TCR-tg (2C) memory T cells were transferred and analyzed for proliferation and apoptosis as described in Fig. 3 except that anti-FasL, anti-CD30L, or both Abs were administered at the time of cell transfer. In the experiment depicted here, infiltrating T cells were pooled from the testes or kidneys of two mice per group. Results were confirmed in one additional experiment. b, To study the effect of FasL and/or CD30 neutralization on intratesticular allograft survival, BALB/c pancreatic islets were transplanted in the testes of splenectomized aly/aly mice and neutralizing Abs were administered at the time CD8 memory T cells were transferred. Allograft survival was monitored by measuring blood glucose levels. c, Fas and CD30 expression on CD8 TCR-tg (2C) T cells that infiltrated the testes of transplanted mice. Fas and CD30 expression (filled histograms) were determined after gating on 2C T cells (CD8+1B2+). Dotted line represents isotype control Ab staining and Fas/CD30 expression on naive 2C T cells.
Discussion
Using an allogeneic cell transplantation model, we have shown that CD8+ memory T cells migrate to an immunologically privileged site and mount a productive immune response that leads to elimination of the foreign Ag. However, Ag elimination at the privileged site occurred in a delayed fashion because of increased apoptosis of memory T cells. Apoptosis was mediated by the combined actions of two TNFR family members that are up-regulated on activated memory T cells, Fas and CD30. Our findings, therefore, indicate that CD8+ memory T cells survey immunologically privileged sites effectively, but that their response is hampered by immunosuppressive mechanisms present at these sites.
Memory T cells respond to and eliminate Ags more efficiently than their naive counterparts as they enter both lymphoid and nonlymphoid compartments, whereas the migration of naive T cells is largely restricted to secondary lymphoid organs (9, 10, 11). We demonstrated in this study that CD8 memory T cells also migrate to an immunologically privileged site, the testis, suggesting that foreign Ags that reside in privileged tissues are subject to immune surveillance by memory T cells. This conclusion is consistent with a recent study showing that HSV-specific CD8+ memory T cells are selectively activated and retained in latently infected sensory neuroganglia, an anatomic location previously considered to be ignored by the host immune system (24). However, our experiments demonstrate that memory T cells not only migrate to an immunologically privileged site but also proliferate and eliminate the foreign Ag that resides there. This implies that vaccines designed to generate T cell memory against microbial pathogens or tumor cells could confer protection to the host even when the foreign Ag seeks refuge in a privileged organ.
Although in our experiments CD8+ memory T cells consistently rejected allografts transplanted at an immunologically privileged site (the testis), the rejection tempo was significantly slower than that observed for grafts placed at a conventional site (under the renal capsule). Slower rejection was associated with increased apoptosis of CD8+ memory T cells that migrated to the testis, suggesting that, like effector T cells, memory T cells are subject to immunoregulatory mechanisms present at privileged sites (5, 6, 7). This finding appears unexpected because memory CD8+ T cells are thought to be resistant to apoptosis, because they express antiapoptotic molecules (25). Grayson et al. (26) have shown that secondary effector T cells generated from CD8+ memory T cells express more Bcl-2 and undergo much less contraction than primary effectors generated from naive cells. In our study, we found that CD8+ memory T cells activated in the testis up-regulate two TNFR family receptors that mediate lymphocyte apoptosis: Fas and CD30. Blocking both receptors prevented the apoptosis of CD8+ memory T cells and reduced the survival of intratesticular allografts, whereas blocking only one receptor at a time did not significantly alter either apoptosis or allograft survival. Therefore, Fas engagement alone is not sufficient for inducing the apoptosis of Ag-stimulated memory T cells, providing further evidence that the memory population is relatively resistant to apoptotic signals.
CD30, a member of the TNFR superfamily, is expressed on activated T and B lymphocytes and certain regulatory T cell populations (15, 22, 23). The ligand for CD30, CD153 or CD30L, is expressed on activated T cells, neutrophils, eosinophils, and resting B cells (22). Earlier in vitro studies have shown that the engagement of CD30 by its ligand provides costimulatory signals to activated T cells (27, 28), whereas other studies provided evidence that CD30 delivers proapoptotic signals (29, 30). In this study, we found that CD30 is up-regulated on CD8+ memory T cells following antigenic stimulation, and that it cooperates with Fas to mediate CD8+ memory T cell apoptosis at an immunologically privileged site. The latter observation suggests that CD30L, like FasL, is preferentially expressed on privileged tissues where it provides protection against immune attack. This possibility remains to be tested. More importantly, our findings add to the mounting evidence that CD30L-CD30 interactions regulate immune responses either by activating suppressor T cells (15) or by inducing the apoptosis of effector T cells (30).
Impaired recall of memory T cells in immunologically privileged tissue, as demonstrated by our study, is commensurate with the original studies by Ksander and Streilein (31, 32), who demonstrated impaired differentiation and function of effector T cells in the anterior chamber of the eye. In one study (32), these investigators showed that the progressive growth of allogeneic tumor cells injected into the anterior chamber of the eye is due in part to failure of infiltrating precursor CTL to differentiate in situ into fully functional effector T cells. Therefore, it is possible that impaired recall of CD8+ memory T cells in the testis could also be attributed in part to defective differentiation of these cells into cytotoxic effectors.
Transplanting allogeneic tissues at immunologically privileged sites has been repeatedly attempted to induce donor-specific immunologic tolerance (1). However, many of these attempts have failed to achieve the desired goal and instead led to prolongation of allograft survival but not indefinite allograft acceptance. The results presented in our study underscore the significant hurdle that memory T cells pose in transplantation, mainly because of their widespread migration routes and their relative resistance to apoptosis.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health Grants AI49466 (to F.G.L.), AI44644 (to F.G.L. and C.P.L.), AI45485 (to D.M.R.), and a Juvenile Diabetes Research Foundation International career award (to Z.D.). I.W.N. is supported by a fellowship from the National Kidney Foundation.
2 Address correspondence and reprint requests to Dr. Fadi G. Lakkis, Yale University School of Medicine, 330 Cedar Street, P.O. Box 208029, New Haven, CT 06520. E-mail address: fadi.lakkis@yale.edu
3 Abbreviations used in this paper: FasL, Fas ligand; wt, wild type; MST, median survival time.
Received for publication August 5, 2004. Accepted for publication October 20, 2004.
References
Streilein, J.. 2003. Ocular immune privilege: therapeutic opportunities from an experiment of nature. Nat. Rev. Immunol. 3:879.
Green, D. R., T. A. Ferguson. 2001. The role of Fas ligand in immune privilege. Nat. Rev. Mol. Cell Biol. 2:917.
Medawar, P.. 1948. Immunity to homologous grafted skin. III. The fate of skin homografts transplanted to the brain, to subcutaneous tissue, and to the anterior chamber of the eye. Br. J. Exp. Pathol. 129:58.
Chen, J. J., Y. Sun, G. J. Nabel. 1998. Regulation of the proinflammatory effects of Fas ligand (CD95L). Science 282:1714.
Griffith, T. S., T. Brunner, S. M. Fletcher, D. R. Green, T. A. Ferguson. 1995. Fas ligand-induced apoptosis as a mechanism of immune privilege. Science 270:1189.
Suarez-Pinzon, W., G. S. Korbutt, R. Power, J. Hooton, R. V. Rajotte, A. Rabinovitch. 2000. Testicular Sertoli cells protect -cells from autoimmune destruction in NOD mice by a transforming growth factor--dependent mechanism. Diabetes 49:1810.
Bellgrau, D., D. Gold, H. Selawry, J. Moore, A. Franzusoff, R. C. Duke. 1995. A role for CD95 ligand in preventing rejection. Nature 377:630.
Griffith, T., X. Yu, J. Herndon, D. Green, T. Ferguson. 1996. CD95-induced apoptosis of lymphocytes in an immune privileged site induces immunological tolerance. Immunity 5:7
Masopust, D., V. Vezys, A. L. Marzo, L. Lefrancois. 2001. Preferential localization of effector memory cells in nonlymphoid tissue. Science 291:2413.
Reinhardt, R. L., A. Khoruts, R. Merica, T. Zell, M. K. Jenkins. 2001. Visualizing the generation of memory CD4 T cells in the whole body. Nature 410:101.
Chalasani, G., Z. Dai, B. T. Konieczny, F. K. Baddoura, F. G. Lakkis. 2002. Recall and propagation of allospecific memory T cells independent of secondary lymphoid organs. Proc. Natl. Acad. Sci. USA 99:6175.
Lakkis, F. G., A. Arakelov, B. T. Konieczny, Y. Inoue. 2000. Immunologic "ignorance" of vascularized organ transplants in the absence of secondary lymphoid tissue. Nat. Med. 6:686.
Miyawaki, S., Y. Nakamura, H. Suzuka, M. Koba, R. Yasumizu, S. Ikehara, Y. Shibata. 1994. A new mutation, aly, that induces a generalized lack of lymph nodes accompanied by immunodeficiency in mice. Eur. J. Immunol. 24:429.
Salvaggio, P., S. Deng, C. Ariyan, I. Millet, W. Zawalich, G. Basadonna, D. Rothstein. 2002. Islet filtration: a simple and rapid new purification procedure that avoids Ficoll and improves islet mass and function. Transplantation 74:877.
Dai, Z., Q. Li, Y. Wang, G. Gao, L. Diggs, G. Tellides, F. Lakkis. 2004. CD4+CD25+ regulatory T cells suppress allograft rejection mediated by memory CD8+ T cells via a CD30-dependent mechanism. J. Clin. Invest. 113:310.
Zhou, J., R. I. Carr, R. S. Liwski, A. W. Stadnyk, T. D. G. Lee. 2001. Oral exposure to alloantigen generates intragraft CD8+ regulatory cells. J. Immunol. 167:107.
Shinkura, R., K. Kitada, F. Matsuda, K. Tashiro, K. Ikuta, M. Suzuki, K. Kogishi, T. Serikawa, T. Honjo. 1999. Alymphoplasia is caused by a point mutation in the mouse gene encoding Nf-b-inducing kinase. Nat. Genet. 22:74.
Fagarasan, S., R. Shinkura, T. Kamata, F. Nogaki, K. Ikuta, K. Tashiro, T. Honjo. 2000. Alymphoplasia (aly)-type nuclear factor B-inducing kinase causes defects in secondary lymphoid tissue chemokine receptor signaling and homing of peritoneal cells to the gut-associated lymphatic tissue system. J. Exp. Med. 191:1477.
Bowen, M., R. Lee, G. Miragliotta, S. Nam, E. R. Podack. 1996. Structure and expression of murine CD30 and its role in cytokine production. J. Immunol. 156:442.
Gilfilan, M., P. Noel, E. Podack, S. Reiner, C. Thompson. 1998. Expression of costimulatory receptor CD30 is regulated by both CD28 and cytokines. J. Immunol. 160:2180.
Lee, S. Y., C. G. Park, Y. Choi. 1996. T cell receptor-dependent cell death of T cell hybridomas mediated by CD30 cytoplasmic domain in association with tumor necrosis factor receptor-associated factors. J. Exp. Med. 183:669.
Telford, W. G., S.-Y. Nam, E. R. Podack, R. A. Miller. 1997. CD30-regulated apoptosis in murine CD8 T cells after cessation of TCR signals. Cell. Immunol. 182:125.
Ksander, B., J. Streilein. 1989. Analysis of cytotoxic T cell responses to intracameral allogeneic tumors. Invest. Ophthalmol. Vis. Sci. 30:323.
Ksander, B., J. Streilein. 1990. Failure of infiltrating precursor cytotoxic T cells to acquire direct cytotoxic function in immunologically privileged sites. J. Immunol. 145:2057.(Zhenhua Dai, Isam W. Nasr)
濠电儑绲藉ú鐘诲礈濠靛洤顕遍柛娑卞枤椤╃兘鏌涘☉鍗炲閺夆晜妫冮弻娑樷枎韫囨挴鍋撴禒瀣劦妞ゆ巻鍋撻柛鐘崇〒濡叉劕鈹戦崶鈹炬灃閻庡箍鍎卞Λ娑㈠焵椤掑鐏︽鐐差儔楠炲洭顢旈崨顓炵哎濠电偠鎻徊鎯洪幋鐘典笉闁挎繂鎷嬮崵鍫澪旈敂绛嬪劌闁哥偞鎸抽弻鏇㈠幢閺囩姴濡介柣銏╁灠缂嶅﹪骞婇敓鐘茬疀妞ゆ挾鍋熸禒鎰版⒑閸︻厐鐟懊洪妶鍥潟闁冲搫鎳庤繚闂佺ǹ鏈粙鎺楁倵椤斿墽纾奸柡鍐ㄥ€稿暩婵犫拃鍕垫疁鐎殿喖鐖煎畷姗€濡歌閸撴垶绻涚€涙ḿ鐭婂Δ鐘叉憸閺侇噣顢曢敂钘夘€涘┑锛勫仜婢х晫绮欐繝鍥ㄧ厸濠㈣泛锕ら弳鏇熸叏閻熼偊妯€闁轰礁绉撮悾婵嬪礃椤垳鎴烽梻浣筋嚃閸犳捇宕濊箛娑辨晣缂備焦岣块埢鏃堟煟閹寸儑渚涢柛鏂垮暣閺岋繝宕掑顓犵厬缂備焦顨呴ˇ閬嶅焵椤掑喚娼愮紒顔肩箻閿濈偤鏁冮崒姘卞摋闁荤娀缂氬▍锝囩矓閸喓鈧帒顫濋鐘闂侀潧娲ゅú銊╁焵椤掑偆鏀版繛澶嬬洴瀹曘垽濡堕崶銊ヮ伕閻熸粎澧楃敮妤咃綖婢舵劖鍋ｉ柛銉娑撹尙绱掓潏銊х畼闁归濞€閹粓鎸婃径澶岀梾濠电偛顕慨楣冨春閺嶎厼鍨傞柕濞炬櫆閸嬨劌霉閿濆懎鏆熸俊顖氱墦濮婃椽顢曢敐鍡欐闂佺粯鎼换婵嬬嵁鐎ｎ喖绠ｆ繝濠傚閹枫劑姊洪幐搴ｂ槈闁哄牜鍓熷畷鐟扳堪閸曨収娴勫銈嗗笂閻掞箓寮抽鍫熺厱闁瑰搫绉村畵鍡涙煃瑜滈崜姘潩閵娾晜鍋傞柨鐔哄Т鐟欙箓骞栭幖顓炵仯缂佲偓婢跺⊕褰掑礂閸忚偐娈ら梺缁樼箖閻╊垰鐣烽敓鐘茬闁肩⒈鍓氶鎴︽⒑鐠団€虫灁闁告柨楠搁埢鎾诲箣閻愭潙顎撳┑鐘诧工閸燁垶骞嗛崒姣綊鎮╅幓鎺濆妷濠电姭鍋撻柟娈垮枤绾鹃箖鏌熺€电ǹ啸鐟滅増鐓￠弻娑㈠箳閺傚簱鏋呭┑鐐叉噹闁帮絾淇婇幘顔芥櫢闁跨噦鎷�Foreign Ags that enter immunologically privileged sites such as the eye, brain, and testis persist for an extended period of time, whereas the same Ags are rapidly eliminated at conventional sites. Immune privilege, therefore, provides unwanted refuge for pathogens and tumor cells but is beneficial for the survival of allogeneic grafts. In this study, we asked whether memory T cells can eliminate foreign Ags deposited at an immunologically privileged site by studying CD8 memory T cell-mediated rejection of pancreatic islet allografts placed either in the testis (a privileged organ) or under the kidney capsule (a nonprivileged site) of diabetic mice. We found that CD8 memory T cells reject intratesticular grafts at a significantly slower rate than the rejection of intrarenal grafts. Delayed graft rejection in the testis was not due to reduced homing or proliferation of memory T cells but due to their increased apoptosis at that site. Apoptosis was mediated by the combined actions of two TNFR family members that are up-regulated on activated memory T cells, Fas, and CD30. Therefore, memory T cells survey immunologically privileged tissues but are subject to the immunosuppressive mechanisms present at these sites.
Introduction
Foreign Ags deposited in areas of the body known as immunologically privileged sites persist indefinitely or are eliminated slowly by the immune system (1, 2). Typical examples of immunologically privileged organs are the testes, ovaries, brain, and eyes, where immune privilege is maintained by mechanisms that attenuate both innate and adaptive immune responses. These mechanisms include reduced migration of dendritic cells to the lymph nodes and spleen due to paucity of draining lymphatics or the presence of blood-tissue barriers (3); the production of immunosuppressive and anti-inflammatory cytokines such as TGF- (4); and the local expression of death-inducing molecules, such as Fas ligand (FasL),3 which eliminate effector lymphocytes (5). Although the mechanisms of immune privilege were originally identified in the anterior chamber of the eye (2), they are also relevant to other privileged organs. For example, Sertoli cells of the testis express both TGF- and FasL (although the latter is controversial), thus creating an environment in which foreign Ags can persist either indefinitely or for an extended duration (6, 7). It is postulated that immune privilege confers a survival advantage on the organism by protecting vital organs from excessive inflammation if infection were to occur. In contrast, immune privilege poses a challenge to the host’s immune system by providing a refuge for unwanted tumor cells and pathogens. This has been demonstrated in HSV infection, whereby entry of the virus into the anterior chamber of the eye leads to tolerance rather than immunity to the viral Ags (8).
In this paper, we asked whether memory T cells, which are central to host immune protection, are capable of seeking out and eliminating foreign Ags at immunologically privileged sites. Unlike naive T cells, memory T cell activation is not dependent on secondary lymphoid organs as they can migrate and respond to Ag in nonlymphoid, peripheral tissues (9, 10). For example, memory T cells cause the rejection of allogeneic grafts in a mouse that lacks all secondary lymphoid organs, whereas naive T cells fail to do so (11, 12), indicating that memory T cells can eliminate foreign Ags that reside exclusively in nonlymphoid compartments. However, it is not known whether memory T cells enter immunologically privileged sites in sufficient numbers to mount a protective immune response, or whether, once they have reached an immunologically privileged site, their response is attenuated by local immunosuppressive mechanisms. Using a murine pancreatic islet cell transplantation model in which allograft rejection is mediated by CD8 memory T cells, we provide direct evidence that CD8 memory T cells migrate efficiently to an immunologically privileged site (the testis) but mount an attenuated response at that site.
Materials and Methods
Mice
B6 mice homozygous for the mutation that leads to alymphoplasia (aly/aly) were purchased from CLEA Japan (13). 2C TCR-transgenic B6 mice on a RAG-2 knockout (Rag2–/–) background (2C.Rag–/–) were generated by backcrossing 2C transgenic mice onto Rag2–/– mice (The Jackson Laboratory). CD30–/– B6 mice were a gift from Dr. T. Mak (University of Toronto, Toronto, Canada). lpr, wild-type (wt) BALB/c, and wt B6 mice were purchased from The Jackson Laboratory. All mice were housed in a specific pathogen-free environment.
Pancreatic islet isolation
Islet donors were 6- to 8-wk-old BALB/c (H-2d) mice. Donor mice were anesthetized, and a median laparotomy was performed. The common bile duct was exposed and clamped at the site where it feeds into the duodenum. A volume of 3 ml of collagenase V (Sigma-Aldrich) (1 mg/ml in PBS) was injected into the common bile duct to distend the pancreas and initiate enzymatic digestion. The pancreas was then removed, minced, and subjected to stationary digestion by incubation in a 37°C water bath for 13 min (14). A volume of 10 ml of cold HBSS with 10% calf serum was added to stop digestion, and the pancreatic tissue was shaken vigorously for 30 s to suspend the islets. Islets were then washed three times in ice-cold HBSS by centrifugation at 1000 rpm for 5 min, and the crude preparation was filtered through a hand-held 100-μm nylon cell strainer (BD Biosciences). After flushing with HBSS, the strainer was turned upside-down over a petri dish and rinsed with HBSS to wash the islets into the dish. Islets were counted and picked up using a 23G needle.
Mouse splenectomy and islet transplantation
Allograft recipients were 6- to 8-wk-old splenectomized aly/aly B6 mice or wt B6 mice (H-2b). Splenectomy was performed by opening the abdomen between the left inferior rib and hip under anesthesia, and the entire spleen was removed intact after ligating the splenic vein and artery at the hilum. The peritoneal membrane and skin were then closed separately. Two weeks later, splenectomized mice underwent islet transplantation either under the kidney capsule or in the testis. Briefly, mice were anesthetized, and a right flank incision was made. Islets (350–400 per mouse) were injected into the subcapsular space of the right kidney using a 23G needle. The needle was pulled out slowly, and the incision was closed with skin staples. For intratesticular islet transplantation, the same number of islets were injected into one testis using a 23G needle. A lower abdominal incision was then made, and both testicles were pulled into the abdominal cavity through small openings in the inguinal areas. These openings were then sutured to prevent the testicles from migrating back into the scrotum. Recipient mice were rendered diabetic by a single injection of streptozotocin (Sigma-Aldrich) (180 mg/kg weight) 10–14 days before transplantation. Primary graft function was defined as blood glucose under 200 mg/dl at 24 h after transplantation. Graft rejection was defined as a rise in blood glucose to >300 mg/dl for 3 consecutive days in mice that had primary graft function. Differences in allograft survival were analyzed using the Mann-Whitney U test.
Memory T cell preparation and adoptive transfer
Memory CD8+ T cells were purified from B6 wt, transgenic 2C-Rag–/–, or CD30–/– spleens 10 wk after immunization with 1 x 107 BALB/c RBC-lysed splenocytes per mouse (15). Briefly, CD8+ T cells were positively selected by autoMACS (Miltenyi Biotec) and incubated with anti-CD8-PE and anti-CD44-FITC Abs (BD Pharmingen), and the memory population was sorted after gating on CD8+CD44high cells (FACSVantage; BD Biosciences). The purity of CD8+CD44high cells was typically >95%. The memory phenotype of this population was further confirmed by staining with anti-CD62L-allophycocyanin or anti-CD25-allophycocyanin (BD Pharmingen) (11). 2C memory cell fractions were assessed by first incubating the cells with 1B2 (mouse IgG1, a clonotypic mAb recognizing the transgenic TCR) followed by anti-mouse IgG1-biotin and streptavidin-PerCP (Zymed). A total of 1 x 106 wt, CD30–/–, or 2C memory CD8+ T cells were adoptively transferred to recipient mice via tail vein injection 24 h after islet transplantation.
In vivo Ab treatment
To block FasL-Fas and CD30L-CD30 interactions, splenectomized aly/aly mice, transplanted with BALB/c islets 1 day earlier, were injected i.p. with 0.125 mg of purified hamster anti-mouse FasL mAb (clone MF3; BD Pharmingen) and/or 0.125 mg of anti-CD30L mAb (BD Pharmingen) on days 0, 2, 4, and 6 relative to the adoptive transfer of memory CD8+ T cells. To block T cell costimulation and prevent acute allograft rejection mediated by primary alloimmune responses, wt mice were treated with 0.25 mg of MR1 (anti-CD40L mAb) and CTLA4-Ig (generated in the laboratory of C. P. Larsen) on days 0, 2, 4, and 6 relative to islet transplantation.
Isolation of graft-infiltrating T cells
Graft-infiltrating cells were isolated as described previously (16). Briefly, the kidney and testis transplanted with islet allografts were first perfused in situ with heparinized 0.9% saline. They were then minced and digested at 37°C for 30 min in 20 ml of RMPI 1640 medium containing 10% FCS and 250 U/ml collagenase (Sigma-Aldrich). Cells were washed twice with HBSS after digestion. To clear the debris, cell suspensions were rapidly passed down a loosely packed glass wool column (300 mg of sterile glass wool in a 10-ml syringe), mixed with Percoll solution (Sigma-Aldrich) to a concentration of 30%, and centrifuged at 2000 rpm for 15 min at room temperature. The pellet was washed and resuspended before flow analysis. For detection of T cell apoptosis, cells were analyzed directly without Percoll purification.
BrdU labeling of T cells
Splenectomized aly/aly mice were pulsed i.p. with 0.8 mg of BrdU (Sigma-Aldrich) 7 days after transplantation (6 days after T cell adoptive transfer). One day later, mice were sacrificed and graft-infiltrating cells were isolated as described above. Cell surface staining was performed using anti-CD8-PE, 1B2, anti-mouse IgG1-biotin, and streptavidin-PerCP. Cells were then fixed in 70% ethanol followed by 1% paraformaldehyde and incubated with 50 U/ml DNase I (Sigma-Aldrich) for 10 min at room temperature. Cells were finally stained with anti-BrdU-FITC Ab (BD Biosciences) and analyzed by flow cytometry.
Analysis of T cell apoptosis
Splenectomized aly/aly mice were sacrificed 8 days after transplantation (7 days after T cell adoptive transfer), and graft-infiltrating cells were isolated as above. Cell surface staining was performed using anti-CD8-PE, 1B2, anti-mouse IgG1-biotin, and streptavidin-PerCP. Cells were then fixed in 2% paraformaldehyde, permeabilized with 0.1% Triton X-100 in 0.1% sodium citrate, and labeled with fluorescein-tagged dUTP by the TUNEL method according to the manufacturer’s instructions (In Situ Cell Death Detection kit; Boehringer Mannheim). Cells were finally analyzed by flow cytometry.
Results
CD8 memory T cells cause delayed elimination of foreign Ag at an immunologically privileged site
To investigate whether memory T cells eliminate foreign Ags at an immunologically privileged site, we used an established murine transplantation model in which allograft rejection is mediated by memory T cells. In this model, allogeneic tissue is transplanted to splenectomized alymphoplastic (aly/aly) mice that lack all secondary lymphoid organs (12, 13). These mice do not mount a primary alloimmune response and do not reject an allograft unless they receive allospecific effector or memory T cells (11, 12). Therefore, we transplanted BALB/c (H-2d) pancreatic islet cells either in the testis (an immune-privileged organ) or under the kidney capsule (a nonprivileged site) of splenectomized aly/aly mice (H-2b) that had been rendered diabetic by streptozotocin injection. Euglycemia (blood glucose, <200 mg/dl) within 24 h of transplantation was indicative of technically successful transplantation, at which time the mice received CD8+CD44high memory T cells isolated from B6 mice immunized with BALB/c splenocytes 10 wk earlier. Control recipients received an equal number of CD8+CD44low naive B6 T cells. Allograft rejection, defined as blood glucose >300 mg/dl for 3 consecutive days, was confirmed by histology. As shown in Fig. 1a, allograft rejection was not observed in control mice that received naive T cells (median survival time (MST) >120 days). This result is consistent with our previous studies in which adoptively transferred naive T cells failed to reject either skin or heart allografts in splenectomized aly/aly recipients (11, 12). In contrast, rejection was observed in all mice that received CD8 memory T cells, irrespective of whether pancreatic islet cell grafts were placed in the testis or under the kidney capsule. However, the rejection of intratesticular grafts was significantly delayed compared with that of intrarenal grafts (MST = 39 vs 16 days, respectively; p < 0.05).
FIGURE 1. Pancreatic islet allograft survival following the transfer of memory or naive CD8 T cells. BALB/c islets were transplanted in the testis or under the kidney capsule of splenectomized aly/aly (a) or wt (b) B6 mice rendered diabetic by streptozotocin injection. One day later, CD8 memory (CD44high) or CD8 naive (CD44low) T cells were transferred to islet recipients. Islet allograft survival was monitored by measuring blood glucose levels. Blood glucose >300 mg/dl for 3 consecutive days was indicative of allograft rejection. wt mice also received CTLA4-Ig and MR1 (anti-CD40L mAb) at the time of transplantation to block naive T cell responses.
The results shown in Fig. 1a were generated in aly/aly mice that are homozygous for a mutation in the NF-B-inducing kinase (17). In addition to the absence of secondary lymphoid tissues, this mutation could potentially influence allograft rejection mediated by exogenous effector or memory T cells (18). To rule out this possibility, we transferred CD8 memory T cells to wt B6 recipients 2 days after transplanting BALB/c pancreatic islet cells either in the testis or under the kidney capsule. The recipients were treated at the time of transplantation with CTLA4-Ig and MR1 (anti-CD40L mAb) to block the CD28/B7 and CD40/CD40L costimulatory pathways, respectively. Because these pathways are required for the activation of naive but not Ag-experienced T cells (19, 20), the transfer of memory but not naive T cells is expected to mediate rejection in this model. As shown in Fig. 1b, naive T cells did not cause rejection (MST > 120 days), whereas CD8 memory T cells caused rapid rejection of intrarenal grafts (MST = 14 days) and delayed rejection of intratesticular grafts (MST = 44 days; p < 0.05). These findings confirm our results in the splenectomized aly/aly model (Fig. 1a). Therefore, CD8 memory T cells are capable of eliminating foreign Ags at an immunologically privileged organ, albeit at a significantly slower rate than that observed at a nonprivileged site.
Ag-dependent homing of CD8 memory T cells to an immunologically privileged site
Because immunologically privileged sites possess tissue-blood barriers that interfere with lymphocyte migration (2), we asked whether delayed rejection of intratesticular allografts was due to reduced homing of CD8 memory T cells to the testis. To answer this question, we adoptively transferred CD8+CD44high memory T cells (harvested from Thy1.1+ B6 mice immunized with BALB/c splenocytes) to Thy1.2+ B6 wt mice 1 day after transplanting pancreatic islet cells either in the testis or under the kidney capsule. An equal number of memory T cells was transferred to control mice injected with PBS at the same sites. Thy1.1+CD8+ cells that infiltrated the transplantation/injection site were quantitated 24 and 48 h after cell transfer. As shown in Fig. 2, comparable numbers of these cells were found in the testis and the kidney at both time points, indicating that CD8 memory T cell homing to immunologically privileged sites is not impaired. The migration of CD8 memory T cells to either the testis or the kidney, however, was dependent on the presence of alloantigen (islets) at these sites (Fig. 2).
FIGURE 2. Alloantigen-dependent migration of CD8 memory T cells to the testis and the kidney. Thy1.1+CD8+ memory T cells (CD44high) were transferred to Thy1.2+ wt B6 mice 1 day after transplanting BALB/c islets in the testis or under the kidney capsule. Control mice were injected with PBS at the same sites. Thy1.1+CD8+ T cells that infiltrated the transplantation/injection site were quantitated by flow analysis 24 and 48 h after cell transfer. Results are shown as percentage of gated lymphocytes (a) or as absolute number (b) of infiltrating CD8+Thy1.1+ T cells per site. In the experiment depicted here, infiltrating T cells were pooled from the testes or kidneys of three mice per group. The results were confirmed in two additional experiments (n = 3).
Enhanced apoptosis of CD8 memory T cells at an immunologically privileged site
To further explore why memory T cell recall is attenuated at an immunologically privileged site, we studied the proliferation and apoptosis of allospecific CD8 memory T cells that migrated to the testis and the kidney after islet cell transplantation. In these experiments, 2C CD8 TCR-tg memory T lymphocytes, which react to Ld MHC class I Ag on BALB/c cells (21), were transferred to splenectomized aly/aly mice 1 day after transplantation, and their in vivo proliferation and apoptosis were studied 7 days later. 2C CD8 TCR-tg T cells are identified by the clonotypic Ab 1B2 and, therefore, are referred to as CD8+1B2+ (21). As shown in Fig. 3, 50% of CD8+1B2+ cells had undergone proliferation 1 wk after encountering alloantigen in either the testis or the kidney (52 ± 2 and 51 ± 3%, respectively; n = 4/group) (Fig. 3). In contrast, their apoptosis was 3-fold higher in the testis than in the kidney (20 ± 3 vs 7 ± 2%, respectively; n = 4/group) (Fig. 3). Increased T cell apoptosis rate in the testis correlated with lower recovery of graft-infiltrating, allospecific 2C T cells from that site at 1 wk posttransfer (0.7 ± 0.1 x 104 CD8+1B2+ cells recovered from the testis vs 2 ± 0.2 x 104 from the kidney; n = 4/group). These results indicate that delayed elimination of foreign Ag at an immunologically privileged site is due to enhanced apoptosis, and not reduced proliferation, of activated, Ag-specific CD8 memory T cells.
FIGURE 3. CD8 memory T cell proliferation and apoptosis in the testis and the kidney. CD8 TCR-tg (2C) memory T cells (CD44high) were transferred to splenectomized aly/aly mice 1 day after transplanting BALB/c pancreatic islets in the testis or under the kidney capsule. Infiltrating T cells were harvested from the testes or kidneys 1 wk after cell transfer, and their proliferation and apoptosis were quantitated by BrdU uptake and TUNEL labeling, respectively, after gating on the CD8 TCR-tg cells (CD8+1B2+ cells shown in top panel). In the experiment depicted here, infiltrating T cells were pooled from the testes and kidneys of two mice per group. Results were confirmed in three additional experiments.
Enhanced apoptosis of CD8 memory T cells at an immunologically privileged site is mediated by Fas and CD30
FasL expression in the testis and anterior chamber of the eye contributes to immune privilege by inducing the apoptosis of effector T cells (5). We therefore examined whether FasL-Fas interaction is responsible for enhanced apoptosis of memory T cells activated in the testis by treating transplanted mice with anti-FasL blocking Ab at the time of transferring 2C CD8 TCR-tg memory T cells. In these experiments, FasL blockade neither reduced the rate of Ag-specific CD8 memory T cell apoptosis in the testis (Fig. 4a), nor significantly accelerated the rejection of intratesticular islet allografts when compared with the mouse group that received isotype control Ab (b). Importantly, CD8+ memory T cells isolated from alloimmunized lpr (Fas mutant) mice also failed to significantly accelerate the rejection of intratesticular islet allografts despite a trend toward shorter graft survival in this group as well as in mice that received anti-FasL Ab. We then asked whether another member of the TNFR family, CD30, contributes to the death of activated CD8 memory T cells. CD30 is expressed on activated T cells, delivers apoptotic signals, and has been recently shown to mediate the suppression of CD8 memory T cells by CD4+CD25+ regulatory T cells (15, 22, 23). After transplanting islets into splenectomized aly/aly mice, CD30–/– CD8 memory T cells underwent apoptosis in the testis and mediated the rejection of islet allografts with the same reduced tempo as wt CD8 memory T cells (Fig. 4, a and b). The inability of either FasL-Fas or CD30L-CD30 blockade alone to significantly alter apoptosis or rejection could not be attributed to the lack of Fas and CD30 on activated memory T cells because both molecules were present on adoptively transferred, allospecific (CD8+1B2+) T cells that infiltrated the testis following transplantation (Fig. 4c). Importantly, the apoptosis of memory T cells was significantly reduced, and allograft survival in the testis was significantly shortened, when anti-FasL was administered to mice that received CD30–/– CD8 memory T cells or when FasL and CD30L were simultaneously blocked at the time of memory T cell transfer (Fig. 4, a and b). Commensurate with reduced T cell apoptosis, more 2C T cells were recovered from the testis when both Fas and CD30 were blocked (2 x 104 CD8+1B2+ cells vs 0.8 x 104 in isotype control Ab-treated mice; n = 2/group). Isotype control Ab did not accelerate allograft rejection mediated by CD8 memory T cells (MST = 37 days; range, 33–48 days; n = 4). Therefore, at least two death receptors, Fas and CD30, cooperate to induce the apoptosis of activated CD8 memory T cells at an immunologically privileged site.
FIGURE 4. Role of Fas and CD30 in CD8 memory T cell apoptosis in the testis. a, CD8 TCR-tg (2C) memory T cells were transferred and analyzed for proliferation and apoptosis as described in Fig. 3 except that anti-FasL, anti-CD30L, or both Abs were administered at the time of cell transfer. In the experiment depicted here, infiltrating T cells were pooled from the testes or kidneys of two mice per group. Results were confirmed in one additional experiment. b, To study the effect of FasL and/or CD30 neutralization on intratesticular allograft survival, BALB/c pancreatic islets were transplanted in the testes of splenectomized aly/aly mice and neutralizing Abs were administered at the time CD8 memory T cells were transferred. Allograft survival was monitored by measuring blood glucose levels. c, Fas and CD30 expression on CD8 TCR-tg (2C) T cells that infiltrated the testes of transplanted mice. Fas and CD30 expression (filled histograms) were determined after gating on 2C T cells (CD8+1B2+). Dotted line represents isotype control Ab staining and Fas/CD30 expression on naive 2C T cells.
Discussion
Using an allogeneic cell transplantation model, we have shown that CD8+ memory T cells migrate to an immunologically privileged site and mount a productive immune response that leads to elimination of the foreign Ag. However, Ag elimination at the privileged site occurred in a delayed fashion because of increased apoptosis of memory T cells. Apoptosis was mediated by the combined actions of two TNFR family members that are up-regulated on activated memory T cells, Fas and CD30. Our findings, therefore, indicate that CD8+ memory T cells survey immunologically privileged sites effectively, but that their response is hampered by immunosuppressive mechanisms present at these sites.
Memory T cells respond to and eliminate Ags more efficiently than their naive counterparts as they enter both lymphoid and nonlymphoid compartments, whereas the migration of naive T cells is largely restricted to secondary lymphoid organs (9, 10, 11). We demonstrated in this study that CD8 memory T cells also migrate to an immunologically privileged site, the testis, suggesting that foreign Ags that reside in privileged tissues are subject to immune surveillance by memory T cells. This conclusion is consistent with a recent study showing that HSV-specific CD8+ memory T cells are selectively activated and retained in latently infected sensory neuroganglia, an anatomic location previously considered to be ignored by the host immune system (24). However, our experiments demonstrate that memory T cells not only migrate to an immunologically privileged site but also proliferate and eliminate the foreign Ag that resides there. This implies that vaccines designed to generate T cell memory against microbial pathogens or tumor cells could confer protection to the host even when the foreign Ag seeks refuge in a privileged organ.
Although in our experiments CD8+ memory T cells consistently rejected allografts transplanted at an immunologically privileged site (the testis), the rejection tempo was significantly slower than that observed for grafts placed at a conventional site (under the renal capsule). Slower rejection was associated with increased apoptosis of CD8+ memory T cells that migrated to the testis, suggesting that, like effector T cells, memory T cells are subject to immunoregulatory mechanisms present at privileged sites (5, 6, 7). This finding appears unexpected because memory CD8+ T cells are thought to be resistant to apoptosis, because they express antiapoptotic molecules (25). Grayson et al. (26) have shown that secondary effector T cells generated from CD8+ memory T cells express more Bcl-2 and undergo much less contraction than primary effectors generated from naive cells. In our study, we found that CD8+ memory T cells activated in the testis up-regulate two TNFR family receptors that mediate lymphocyte apoptosis: Fas and CD30. Blocking both receptors prevented the apoptosis of CD8+ memory T cells and reduced the survival of intratesticular allografts, whereas blocking only one receptor at a time did not significantly alter either apoptosis or allograft survival. Therefore, Fas engagement alone is not sufficient for inducing the apoptosis of Ag-stimulated memory T cells, providing further evidence that the memory population is relatively resistant to apoptotic signals.
CD30, a member of the TNFR superfamily, is expressed on activated T and B lymphocytes and certain regulatory T cell populations (15, 22, 23). The ligand for CD30, CD153 or CD30L, is expressed on activated T cells, neutrophils, eosinophils, and resting B cells (22). Earlier in vitro studies have shown that the engagement of CD30 by its ligand provides costimulatory signals to activated T cells (27, 28), whereas other studies provided evidence that CD30 delivers proapoptotic signals (29, 30). In this study, we found that CD30 is up-regulated on CD8+ memory T cells following antigenic stimulation, and that it cooperates with Fas to mediate CD8+ memory T cell apoptosis at an immunologically privileged site. The latter observation suggests that CD30L, like FasL, is preferentially expressed on privileged tissues where it provides protection against immune attack. This possibility remains to be tested. More importantly, our findings add to the mounting evidence that CD30L-CD30 interactions regulate immune responses either by activating suppressor T cells (15) or by inducing the apoptosis of effector T cells (30).
Impaired recall of memory T cells in immunologically privileged tissue, as demonstrated by our study, is commensurate with the original studies by Ksander and Streilein (31, 32), who demonstrated impaired differentiation and function of effector T cells in the anterior chamber of the eye. In one study (32), these investigators showed that the progressive growth of allogeneic tumor cells injected into the anterior chamber of the eye is due in part to failure of infiltrating precursor CTL to differentiate in situ into fully functional effector T cells. Therefore, it is possible that impaired recall of CD8+ memory T cells in the testis could also be attributed in part to defective differentiation of these cells into cytotoxic effectors.
Transplanting allogeneic tissues at immunologically privileged sites has been repeatedly attempted to induce donor-specific immunologic tolerance (1). However, many of these attempts have failed to achieve the desired goal and instead led to prolongation of allograft survival but not indefinite allograft acceptance. The results presented in our study underscore the significant hurdle that memory T cells pose in transplantation, mainly because of their widespread migration routes and their relative resistance to apoptosis.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health Grants AI49466 (to F.G.L.), AI44644 (to F.G.L. and C.P.L.), AI45485 (to D.M.R.), and a Juvenile Diabetes Research Foundation International career award (to Z.D.). I.W.N. is supported by a fellowship from the National Kidney Foundation.
2 Address correspondence and reprint requests to Dr. Fadi G. Lakkis, Yale University School of Medicine, 330 Cedar Street, P.O. Box 208029, New Haven, CT 06520. E-mail address: fadi.lakkis@yale.edu
3 Abbreviations used in this paper: FasL, Fas ligand; wt, wild type; MST, median survival time.
Received for publication August 5, 2004. Accepted for publication October 20, 2004.
References
Streilein, J.. 2003. Ocular immune privilege: therapeutic opportunities from an experiment of nature. Nat. Rev. Immunol. 3:879.
Green, D. R., T. A. Ferguson. 2001. The role of Fas ligand in immune privilege. Nat. Rev. Mol. Cell Biol. 2:917.
Medawar, P.. 1948. Immunity to homologous grafted skin. III. The fate of skin homografts transplanted to the brain, to subcutaneous tissue, and to the anterior chamber of the eye. Br. J. Exp. Pathol. 129:58.
Chen, J. J., Y. Sun, G. J. Nabel. 1998. Regulation of the proinflammatory effects of Fas ligand (CD95L). Science 282:1714.
Griffith, T. S., T. Brunner, S. M. Fletcher, D. R. Green, T. A. Ferguson. 1995. Fas ligand-induced apoptosis as a mechanism of immune privilege. Science 270:1189.
Suarez-Pinzon, W., G. S. Korbutt, R. Power, J. Hooton, R. V. Rajotte, A. Rabinovitch. 2000. Testicular Sertoli cells protect -cells from autoimmune destruction in NOD mice by a transforming growth factor--dependent mechanism. Diabetes 49:1810.
Bellgrau, D., D. Gold, H. Selawry, J. Moore, A. Franzusoff, R. C. Duke. 1995. A role for CD95 ligand in preventing rejection. Nature 377:630.
Griffith, T., X. Yu, J. Herndon, D. Green, T. Ferguson. 1996. CD95-induced apoptosis of lymphocytes in an immune privileged site induces immunological tolerance. Immunity 5:7
Masopust, D., V. Vezys, A. L. Marzo, L. Lefrancois. 2001. Preferential localization of effector memory cells in nonlymphoid tissue. Science 291:2413.
Reinhardt, R. L., A. Khoruts, R. Merica, T. Zell, M. K. Jenkins. 2001. Visualizing the generation of memory CD4 T cells in the whole body. Nature 410:101.
Chalasani, G., Z. Dai, B. T. Konieczny, F. K. Baddoura, F. G. Lakkis. 2002. Recall and propagation of allospecific memory T cells independent of secondary lymphoid organs. Proc. Natl. Acad. Sci. USA 99:6175.
Lakkis, F. G., A. Arakelov, B. T. Konieczny, Y. Inoue. 2000. Immunologic "ignorance" of vascularized organ transplants in the absence of secondary lymphoid tissue. Nat. Med. 6:686.
Miyawaki, S., Y. Nakamura, H. Suzuka, M. Koba, R. Yasumizu, S. Ikehara, Y. Shibata. 1994. A new mutation, aly, that induces a generalized lack of lymph nodes accompanied by immunodeficiency in mice. Eur. J. Immunol. 24:429.
Salvaggio, P., S. Deng, C. Ariyan, I. Millet, W. Zawalich, G. Basadonna, D. Rothstein. 2002. Islet filtration: a simple and rapid new purification procedure that avoids Ficoll and improves islet mass and function. Transplantation 74:877.
Dai, Z., Q. Li, Y. Wang, G. Gao, L. Diggs, G. Tellides, F. Lakkis. 2004. CD4+CD25+ regulatory T cells suppress allograft rejection mediated by memory CD8+ T cells via a CD30-dependent mechanism. J. Clin. Invest. 113:310.
Zhou, J., R. I. Carr, R. S. Liwski, A. W. Stadnyk, T. D. G. Lee. 2001. Oral exposure to alloantigen generates intragraft CD8+ regulatory cells. J. Immunol. 167:107.
Shinkura, R., K. Kitada, F. Matsuda, K. Tashiro, K. Ikuta, M. Suzuki, K. Kogishi, T. Serikawa, T. Honjo. 1999. Alymphoplasia is caused by a point mutation in the mouse gene encoding Nf-b-inducing kinase. Nat. Genet. 22:74.
Fagarasan, S., R. Shinkura, T. Kamata, F. Nogaki, K. Ikuta, K. Tashiro, T. Honjo. 2000. Alymphoplasia (aly)-type nuclear factor B-inducing kinase causes defects in secondary lymphoid tissue chemokine receptor signaling and homing of peritoneal cells to the gut-associated lymphatic tissue system. J. Exp. Med. 191:1477.
Bowen, M., R. Lee, G. Miragliotta, S. Nam, E. R. Podack. 1996. Structure and expression of murine CD30 and its role in cytokine production. J. Immunol. 156:442.
Gilfilan, M., P. Noel, E. Podack, S. Reiner, C. Thompson. 1998. Expression of costimulatory receptor CD30 is regulated by both CD28 and cytokines. J. Immunol. 160:2180.
Lee, S. Y., C. G. Park, Y. Choi. 1996. T cell receptor-dependent cell death of T cell hybridomas mediated by CD30 cytoplasmic domain in association with tumor necrosis factor receptor-associated factors. J. Exp. Med. 183:669.
Telford, W. G., S.-Y. Nam, E. R. Podack, R. A. Miller. 1997. CD30-regulated apoptosis in murine CD8 T cells after cessation of TCR signals. Cell. Immunol. 182:125.
Ksander, B., J. Streilein. 1989. Analysis of cytotoxic T cell responses to intracameral allogeneic tumors. Invest. Ophthalmol. Vis. Sci. 30:323.
Ksander, B., J. Streilein. 1990. Failure of infiltrating precursor cytotoxic T cells to acquire direct cytotoxic function in immunologically privileged sites. J. Immunol. 145:2057.(Zhenhua Dai, Isam W. Nasr)