|
|
||||||||
Department of Surgery, Washington University School of Medicine, St. Louis, MO 63110
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
There is now a large body of evidence that the responsible regulatory lymphocyte population is a CD4+ T cell (2, 4, 5, 6, 11, 12, 13). However, there is evidence that CD8+ T lymphocytes can also exert regulatory activity (14, 15, 16, 17, 18, 19, 20). Thus, the regulatory cells contributing to the tolerant state appear to be of varying phenotypes, and their mechanism of action remains obscure. Therefore, a careful identification of the regulatory cells participating in tolerance induction will lead to a better understanding of the underlying regulatory mechanisms and then be used to actively enhance graft acceptance.
Donor-specific transfusions (DST)3 induce both experimental and clinical donor-specific allograft tolerance (15, 18, 19, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34). However, the mechanisms of graft acceptance after DST are still unclear. The presence of donor-specific CTL in cardiac or renal allografts accepted after DST (21, 22, 28) indicates that clonal deletion of alloreactive T cells probably does not occur. Thus, other cells appear to be necessary to regulate these potentially reactive cells.
Recently, we have found that tolerance induced by DST, without mAb treatment, in LEW to DA rat heart transplantation can subsequently be transferred to a naive animal by the adoptive transfer of the tolerant recipients splenocytes (35). In this study, we characterize the regulatory cells in this DST-induced infectious tolerance.
| Materials and Methods |
|---|
|
|
|---|
Inbred male rats, weighing 200250 g, were purchased from Harlan Sprague-Dawley (Indianapolis, IN), maintained under standard conditions, and allowed to drink water and eat rodent chow ad libitum. LEW (RT1l) and DA (RT1a) rats served as donor and recipient, respectively, for cardiac transplantation and Brown Norway rats (BN; RT1n) were used as third-party donor controls. The care and use of laboratory animals conformed to the National Institutes of Health and Washington University guidelines.
DST
Single-cell suspensions of LEW splenocytes were prepared in ice-cold RPMI 1640 (Life Technologies, Gaithersburg, MD) supplemented with 10 mmol/L HEPES and 1% penicillin-streptomycin. RBC were lysed with 0.83% Tris-ammonium chloride buffer at 37°C for 5 min, and the remaining cells were washed twice with RPMI 1640. Splenocytes (100 x 106 in 1 ml of PBS) were injected i.v. into each rat via the penile vein as a source of DST. Seven days after DST, a LEW to DA heart transplantation was performed to produce long tem recipients to provide tolerant cells for adoptive transfer.
Cardiac transplantation and induction of primary tolerance
Heterotopic cardiac transplantation was performed using the modified technique of Ono and Lindsey (36). Briefly, the donor thoracic aorta and pulmonary artery were anastomosed to the infrarenal recipient aorta and inferior vena cava, respectively. Graft survival was assessed by daily palpation and rejection was confirmed histologically.
Adoptive transfer study
To detect the presence of regulatory cells, a naive DA recipient was gamma-irradiated with 300 or 450 rad, and 100 x 106 splenocytes harvested from a tolerant DA rat bearing a LEW heart for >60 days were adoptively transferred on the following day. A LEW heart was transplanted 24 h later. Irradiation was performed with a 137Cs irradiator (MarK I, model 30; J. L. Shepherd & Associates, San Fernando, CA). In some experiments, positive selection or depletion of CD4+ and/or CD8+ cells were performed with MACS Rat CD4 and/or CD8 MicroBeads (Miltenyi Biotec, Auburn, CA) before adoptive transfer. The resulting cell population after MACS manipulation was characterized by flow cytometry.
MLC
Stimulator cells (2 x 107/ml) were incubated with 50 µg/ml mitomycin C (MMC) for 40 min at 37°C and washed four times with RPMI 1640. Responder cells (2.5 x 105) and MMC-treated allogeneic or syngeneic splenocytes (2.5 x 105) were cocultured in 200 µl of RPMI 1640 supplemented with 10% FBS, 1% L-glutamine, 100,000 U/L penicillin-streptomycin, 10 mM HEPES, 1% sodium pyruvate, 1% nonessential amino acids, and 1 x 10-5 M 2-mercaptoethanol (complete RPMI) in a 96-well U-bottom microtiter plate for 4 days at 37°C in a 95% air, 5% CO2 humidified atmosphere. [3H]Thymidine (1.0 µCi/well; ICN Pharmaceuticals, Costa Mesa, CA) was added 18 h before harvesting, and [3H]thymidine incorporation was determined in a liquid scintillation counter (model 1450; Microbeta, Gaithersburg, MD). Each sample was performed in triplicate. Stimulation index (S.I.) was calculated by S.I. = (cpm from allogeneic stimulation/cpm from syngeneic stimulation).
Direct and indirect CTL assay
Freshly isolated splenocytes were used as effector cells in the direct CTL assay without in vitro stimulation. For indirect CTL, responder (5 x 106/ml) and MMC-treated stimulator splenocytes (5 x 106/ml) were cocultured in complete RPMI at 37°C in a 95% air, 5% CO2 humidified atmosphere in a 25-ml flask. After 7 days, the effector cells were washed and resuspended in fresh complete RPMI. The adenovirus-transformed LEW rat strain (RT1l) cell line A2/ASREB/IP/F4 (As-F4) (37) provided by Dr. D. Bellgrau (Denver, CO) was used as LEW target cells. As specificity controls, LEW, DA, and BN splenocyte blasts cultured with 5 µg/ml Con A for 23 days were also used as target cells. Target cells (5 x 106) were labeled with 200 µCi of 51Cr (Na51CrO4; DuPont NEN, Wilmington, DE) in 100 µl of complete RPMI for 1 h at 37°C in 5% CO2. After four washings with complete-RPMI, 1 x 104 labeled target cells in 100 µl of complete RPMI were added to a 96-well microtiter plate in E:T ratios of 100:1, 30:1, 10:1, and 3:1. After incubation for 4 h at 37°C in 5% CO2, the plates were centrifuged at 500 x g, and 100 µl of the supernatant were harvested and counted in a gamma counter (1272 Clinigamma; LKB Instruments, Turku, Finland). The mean of triplicate samples was calculated, and percent lysis was determined according to the equation: % lysis = [(a - b)/(c - b)] x 100, where a = 51Cr release from target cells mixed with effector cells, b = 51Cr release from target cells alone, and c = 51Cr release after 0.1% Triton X-100 lysis.
In vivo tracking of adoptively transferred lymphocytes
Labeling of splenocytes with CFSE was performed (38) by resuspending splenocytes in PBS at 1 x 107 cell/ml and incubating with CFSE (Molecular Probes, Eugene, OR) at a final concentration of 5 µM for 10 min at 37°C. Labeling was terminated by the addition of FBS (10% of total volume). Splenocytes were washed twice with complete RPMI and resuspended in PBS at 1 x 108 cell/ml for i.v. injection. CFSE-labeled splenocytes (100 x 106) from either a naive or a DST-treated LTS (graft acceptance for >60 days) DA were adoptively transferred into a 450-rad irradiated naive DA rat that was transplanted with a LEW heart 24 h later. Seven days after heart transplantation, splenocytes and heart graft-infiltrating lymphocytes (GIL) were isolated for flow cytometric analysis. As a control, irradiated (450 rad) and nonirradiated DA rats were adoptively transferred with CFSE-labeled naive DA splenocytes without heart transplantation, and 8 days later splenocytes were also isolated for flow cytometric analysis.
Isolation of GIL
Graft tissue was minced through a 50-µm pore size stainless steel sieve into RPMI 1640 supplemented with 10% FBS. The tissue suspension was then digested with 1 mg/ml collagenase (Sigma-Aldrich, St. Louis, MO) for 60 min at 37°C. After two washes, viable lymphocytes were separated on a Ficoll gradient (Histopaque-1083; Sigma-Aldrich). Cells were washed three times with RPMI 1640 supplemented with 10% FBS and used for additional experiments.
Flow cytometric analysis
Cells were incubated with Cy-Chrome-conjugated anti-rat CD4 mAb (mouse IgG2a, clone OX35), FITC-conjugated anti-rat CD8
mAb (mouse IgG1, clone OX8), and PE-conjugated anti-rat CD3 mAb (mouse IgG3, clone G4.18) (all from BD PharMingen, San Diego, CA) for 30 min at 4°C. The cells were washed, resuspended in PBS, and analyzed by FACScan (BD Biosciences, Mountain View, CA) using CellQuest (BD Biosciences) software.
| Results |
|---|
|
|
|---|
Adoptive transfer of 100 x 106 splenocytes from DST-treated LTS (graft accepted for >60 days) into a 300-rad (n = 5)- or 450-rad (n = 10)-irradiated naive DA recipient resulted in the acceptance of all LEW cardiac allografts (mean survival time (MST) > 100 days in both groups), indicating the presence of donor regulatory cells (Fig. 1). To demonstrate donor specificity, third-party BN hearts were also transplanted into DA recipients that had been irradiated with 300 or 450 rad and given splenocytes from DA long term survivors (LTS) with a LEW heart. Rejection of all BN grafts in MST of 11.3 ± 0.6 days (300 rad, n = 3) and 23.8 ± 8.3 days (450 rad, n = 4) showed that this transferable tolerance is donor specific. Irradiation of the DA recipient, without adoptive cell transfer, moderately prolonged LEW heart graft survival to a MST of 17.3 ± 4.9 or 18.8 ± 7.7 days, respectively, in 300 rad (n = 3) or 450 rad (n = 10; 1 of 10 recipient did not reject) irradiated DA recipients. Irradiation of recipients was necessary to detect infectious tolerance because a nonirradiated DA recipient that received adoptive transfer of 100 x 106 splenocytes from a DST-treated LTS rejected a LEW heart in MST of 8.3 ± 0.6 days (n = 3).
|
|
|
The adoptive transfer of cells from a DA rat receiving only LEW DST, but no heart transplant, did not lead to LEW heart graft acceptance in any irradiated naive recipient. Splenocytes (100 x 106) adoptively transferred from a DA rat 7 days after DST or a DA rat >60 days after DST pretreatment without LEW heart transplantation did not prevent rejection of LEW heart grafts in MST of 10.5 ± 5.4 and 5.6 ± 0.5 days, respectively (Fig. 3). Therefore, DST alone was ineffective without the presence of a heart allograft in generating the regulatory cells of this infectious tolerance.
|
Zhai et al. (13) has reported that lymphocytes capable of transferring tolerance were compartmentalized only in the spleen, and were not present in the lymph nodes (LNs) of LTS after anti-CD4 mAb treatment. To determine whether this occurred in our DST tolerant model, 100 x 106 LN cells from DST-treated DA LTS were adoptively transferred into a 450-rad-irradiated DA recipient 24 h before a LEW heart transplant. All rats given 100 x 106 LN cells (n = 5) accepted cardiac allografts for >100 days, similar to that seen when 100 x 106 LTS splenocytes were transferred. We also confirmed that 50 x 106 LN cells or splenocytes could transfer tolerance into a 450-rad-irradiated DA rat. Therefore, both splenocytes and LN cells were equally effective in transferring tolerance to a DA heart graft.
GIL were more effective for transfer of tolerance
Considering that graft-infiltrating alloreactive recipient cells encounter the donor Ag at the graft site, it is possible that more regulatory cells reside in the graft. To examine this possibility, we transferred isolated GIL from a recipient accepting a LEW heart graft for >60 days into a 450-rad-irradiated DA rat (Fig. 4). Because only a small number of GIL can be harvested from long term accepted heart grafts, all of the GIL harvested from a single heart graft were adoptively transferred (0.33 x 106). However, two of the four subsequent LEW cardiac allografts were accepted. Because 30 x 106 LTS splenocytes were necessary to transfer tolerance, GIL were 10 to 100 times more effective in transferring tolerance. To avoid losing cells in GIL isolation, we harvested the tolerant graft 60 days after heart transplantation and retransplanted into a 450-rad-irradiated DA recipient. Three of four LEW retransplanted grafts were accepted. To examine whether lymphocyte populations changed in the GIL, we performed FACS analysis of GILs and splenocytes from LTS. The proportions of these subsets are similar in the two populations (31.84% vs 41.37% for CD4+ T cells, 3.27% vs 4.12% for CD25+CD4+ T cells, and 10.72% vs 9.18% for CD8+ T cells).
|
When 2.5 x 105 responder LN cells from 1) naive DA, 2) DST-treated DA LTS, or 3) DA 7 days after DST without heart transplant were cultured with 2.5 x 105 MMC-treated LEW (allogeneic), DA (syngeneic), or BN (third-party) stimulator splenocytes for 4 days, there was no significant difference in the proliferative response among these three groups (Fig. 5). To determine whether lymphocytes from 2) DST-treated DA LTS or 3) DA 7 days after DST (suppressed 1) naive DA lymphocytes in vitro, 1.25 x 105 naive DA lymph node cells mixed with the same number of lymph node cells from 2) DST-treated DA LTS or 3) DA 7 days after DST (total 2.5 x 105 responder cells) were cocultured with 2.5 x 105 MMC-treated stimulator splenocytes for 4 days. There was no suppressive effect by lymphocytes from DST-treated DA LTS or DA 7 days after DST on the normal DA response to LEW rat Ags (Fig. 5).
|
The development of indirect CTL activity by splenocytes was examined in the following groups: group 1, naive DA; group 2, DA rejecting a LEW heart 12 days after transplantation without DST-pretreatment; group 3, DA 7 days after LEW DST without LEW heart transplant; group 4, LEW DST-treated DA LTS (>60 days); and group 5, LTS after the first adoptive transfer (Fig. 6A). CTL activity observed in the splenocytes from DST-treated LTS (group 4, 70% at E:T of 100:1) was equal to that seen in an untreated animal rejecting a graft (group 2) or treated only with DST (group 3) and higher than that seen with nontransplanted naive DAsplenocytes (group 1). The indirect cytotoxicity of splenocytes from a LTS after the first adoptive transfer (group 5) was also equivalent to that of naive DA splenocytes (group 1). The specificity of CTL generated from DST-treated LTS splenocytes was donor specific in that cytotoxicity was observed against LEW Con A blasts but not against DA or BN blasts (Fig. 6B). No direct cytotoxicity (without in vitro stimulation) was observed with freshly isolated splenocytes from any of groups 15 (Fig. 6C).
|
Adoptively transferred CFSE-labeled naive DA splenocytes were barely detectable (0.20%) in the spleen of nonirradiated naive DA recipients 8 days after adoptive transfer without heart transplantation (Fig. 7A). The percent of adoptively transferred DA splenocytes detected in DA recipients irradiated with 450 rad before adoptive transfer increased to 3.07% (Fig. 7B). Adoptively transferred splenocytes from naive DA or DST-treated DA LTS settling in the recipient spleen were further increased (12.99 and 13.48%, respectively) when a DA recipient was transplanted with a LEW heart (Fig. 7, C and D). The similar proliferation of adoptively transferred splenocytes from naive DA or DST-treated DA LTS, determined by the sequential loss of CFSE fluorescence intensity with cell division, was observed. The percent of adoptively transferred T cells, identified by CFSE and CD3 staining, in the spleen 7 days after heart transplantation after adoptive transfer of either CFSE-labeled naive or LTS splenocytes was equivalent (6.79 and 6.77%, respectively). Higher numbers of adoptively transferred DA LTS T cells was observed in the LEW cardiac GIL than adoptively transferred naive DA T cells (2.95 and 1.57%, respectively) (Fig. 7, E and F).
|
We examined the relative contribution of CD4+ or CD8+ cells to transfer infectious tolerance. Splenocytes harvested from DST-treated LTS were depleted of CD4+ and/or CD8+ cells with magnetic beads before transfer to a 450-rad-irradiated DA recipient (Fig. 8A). Thirty million depleted splenocytes were used because 30 x 106 unfractionated DST-treated LTS splenocytes uniformly transferred tolerance (Fig. 2). When CD8+-depleted splenocytes were transferred, all (n = 5) of the transplanted LEW cardiac allografts were accepted. However, when CD4+-depleted splenocytes (n = 10) were transferred, only 40% of LEW allografts were accepted. The depletion of both CD4 and CD8 populations removed all T lymphocytes and resulted in the rejection of all (n = 5) allografts in a MST of 14.0 ± 2.3 days. We also adoptively transferred positively selected CD4+ or CD8+ T cells (Fig. 8B). Because CD4+ or CD8+ cells comprise 40% or 10% of splenocytes, respectively, we transferred 40 x 106 CD4+ cells or 10 x 106 CD8+ cells to reflect the actual number of these populations in 100 x 106 unfractionated splenocytes. Adoptive transfer of 40 x 106 (n = 6) or 10 x 106 (n = 3) CD4+ cells resulted in the acceptance of all transplanted LEW hearts, respectively, whereas only 62.5% of the LEW allografts were accepted when 10 x 106 CD8+ cells (n = 8) were transferred. These results confirmed that although CD4+ T cells are the primary regulatory population that transfers infectious tolerance, CD8+ cells also provide a less potent regulatory population.
|
| Discussion |
|---|
|
|
|---|
Our study found equal transfer of tolerance by lymph node cells and splenocytes. This contrasts with the reported compartmentalization of the tolerogeneic rat lymphocytes in the spleen, but not in lymph nodes, in recipients given peritransplant nondepleting anti-CD4 mAb (RIB 5/2) (13). This difference may reflect the different methods of primary tolerance induction. Zhai et al. (13) exposed the recipient rat to a skin graft 1 week before heart transplantation under the cover of 10 injections of anti-CD4 mAbs from the day of skin grafting to 3 wk after heart transplantation. In contrast, we gave only a single preoperative DST 7 days before heart transplantation.
Recently, the ability of GIL to transfer tolerance has been reported (20, 39, 40). Sawitzki et al. (39) reported that GIL in the tolerant rat kidney graft induced by anti-CD4 mAbs are enriched for regulatory cells when compared with splenocytes. GIL (2 x 106) transferred tolerance equivalent to >50 x 106 splenocytes. Zhou et al. (20) also reported that, after oral administration of donor splenocytes, renal allograft GIL could adoptively transfer tolerance to a naive animal. Graca et al. (40) showed that CBA/Ca T cells infiltrating a B10.BR skin graft tolerized by anti-CD4 and anti-CD8 mAb treatment could expand when the B10.BR skin graft was retransplanted into T cell-deficient RAG1-/--CBA mouse and could prevent the rejection of a subsequent fresh B10.BR skin graft by transfused naive CBA/Ca splenocytes. Our transfer of tolerance with GIL isolated from tolerant heart grafts shows the presence of regulatory cells in the GIL. We also found that GIL are much more effective than splenocytes or LN cells in transferring tolerance.
Quigley et al. (23) reported that DA lymph node cells harvested 7 days after LEW DST suppressed proliferation to LEW stimulator cells in 4-day MLC. In contrast, our MLC assays showed no suppression of proliferation by DA lymph node cells from either DST-treated LTS or DA 7 days after DST. However, in vivo tracking of adoptively transferred CFSE-labeled splenocytes showed equivalent proliferation of adoptively transferred DST-treated LTS and naive DA splenocytes in the spleen of the recipient. A higher percent of DST-treated LTS splenocytes was observed in heart GIL 7 days after heart transplantation than naive splenocytes. Thus, the proliferation of splenocytes from DST-treated LTS in response to donor Ag occurs in recipients and expands the regulatory population.
Recently, Lin et al. (41) reported that when adoptively transferred into tolerant mice, CD8+ T cells monospecific for the tolerated transplant Ag could proliferate and accumulate to the same extent as in a naive host but, in contrast, could not cause graft rejection, express IFN-
, or generate CTLs. They suggested that the tolerance was mediated by regulatory T cells that censored immune effector functions rather than suppressing the induction of T cell responses. Unlike their Ab-induced tolerance, our DST-induced tolerance generated CTL after in vitro stimulation. Equal numbers of graft-infiltrating cells (32) and an equivalent level of anti-donor cytotoxic activity (21, 22, 28) have been reported for DST-treated recipients compared with untreated recipients. Our results showed that DST-induced LTS splenocytes generated cytotoxicity after in vitro stimulation, equal to that seen in splenocytes from a rat rejecting a graft without DST pretreatment or a rat treated with DST alone but without heart transplantation. Together, these findings suggest that, after initial preoperative DST stimulation and subsequent heart transplantation, persisting CTL precursors (CTLp) are regulated to achieve tolerance. Generation of our indirect cytotoxicity in splenocytes from DST-treated LTS reflects the in vitro differentiation of LTS CTLp into effector CTL by release from this in vivo regulation.
The method used for inducing tolerance may also influence the phenotype of the regulatory T cells imparting infectious tolerance. Waldmanns group showed that tolerant cells selectively depleted of CD4+ cells failed to transfer the mAb-induced tolerance (2, 4, 5). Woods group found that pretreatment with DST alone was not sufficient to establish the tolerance and the addition of depleting anti-CD4 mAb was necessary to induce a regulatory population of murine lymphocytes (CD45RBlowCD25+CD4+ cells) that could adoptively transfer the donor specific tolerance to a subsequent naive mouse heart recipient (10, 11, 12). In skin-sensitized rats tolerized with a nondepleting anti-CD4 mAb (RIB 5/2), Onodera et al. (6) showed that the depletion of adoptively transferred CD4+, but not CD8+, cells prevented the development of infectious tolerance. Regardless of the animal model used, CD4 cells have uniformly been reported to exert regulatory effect.
Our failure to induce infectious tolerance by depletion of both CD4+ and CD8+ splenocytes indicates that the regulatory cells reside within the T cell population. CD8-depleted splenocytes or positively selected CD4+ cells uniformly transferred tolerance also clearly identified CD4+ cells as regulatory cells. However, we also found that CD4-depleted cells or positively selected CD8+ cells could transfer tolerance to 40 or 62.5% of the naive recipients, respectively. Recently, Nicolls et al. (42) reported that the ability to transfer established tolerance to islet allograft induced by anti-CD154/anti-LFA-1 therapy was only partially inhibited by depleting CD4 T cells from tolerant cells. The suppressive effect of CD8+ T cells has been previously reported in several transplantation models (14, 15, 16, 17, 18, 19, 20). Padberg et al. (14) demonstrated that both CD4+ and CD8+ T cells from LTS, induced by DST plus hyperimmune serum treatment, suppressed rat heart transplant rejection when adoptively transferred to a naive recipient. Oluwole et al. (15) showed that adoptive transfer of 5 x 106 LEW DST induced LTS ACI CD8+ T cells prolonged subsequent LEW heart graft survival in an unmodified ACI recipient. Douillard et al. (18) found that the expansion of a CD8+ clone bearing the V
18-D
1-J
2.7 TCR gene rearrangement specific for donor MHC was detected in each LEW.1A rat that was tolerized to LEW.1W hearts after DST. They also reported that anti-TCR V
18-D
1-J
2.7 DNA vaccination abolished this DST induced allograft tolerance (19). Recently, Zhou et al. (20) reported the generation of CD8+ regulatory GIL in a renal allograft after oral administration of donor splenocytes. They showed that although those CD8+ graft-infiltrating cells expressed increased direct CTL activity, they could adoptively transfer allograft tolerance into a naive recipient.
In conclusion, despite the enhanced in vitro immunological reactivity against donor Ag, rats treated with DST alone accept a donor-specific heart transplant and develop an infectious tolerance similar to that achieved in previous studies using in vivo mAb administration. We found that both CD4+ and CD8+ populations were responsible for the regulatory mechanism in DST-induced tolerance, with the CD4+ population playing the major role. This infectious tolerance can be expanded and serially transferred to subsequent naive cardiac recipients.
| Acknowledgments |
|---|
| Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. M. Wayne Flye, Department of Surgery, Washington University School of Medicine, One Barnes Hospital Plaza, Suite 5103, St. Louis, MO 63110. E-mail address: flyew{at}msnotes.wustl.edu ![]()
3 Abbreviations used in this paper: DST, donor-specific transfusion; CTLp, CTL precursor; GIL, graft-infiltrating lymphocytes; LN, lymph node; LTS, long term survivor; MMC, mitomycin C; MST, mean survival time; h, human. ![]()
Received for publication February 27, 2003. Accepted for publication April 25, 2003.
| References |
|---|
|
|
|---|
rearrangement in heart allografts. J. Immunol. 157:1250.[Abstract]
mRNA levels and by strong accumulation of major histocompatibility complex transcripts in the graft. Transplantation 54:219.[Medline]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |