It is postulated that IFN-γ production hinders long-term acceptance of transplanted organs. To test this hypothesis, we compared survival of skin and heart allografts in wild-type (IFN-γ+/+) mice to that in IFN-γ gene knockout (IFN-γ−/−) mice. We found that perioperative blockade of the CD28 and/or CD40 ligand T cell costimulation pathways induces long-term skin and heart allograft survival in IFN-γ+/+ recipients but fails to do so in IFN-γ−/− mice or in wild-type mice treated with IFN-γ-neutralizing Ab at the time of transplantation. In vitro studies showed that endogenously produced IFN-γ down-regulates T cell proliferation and CTL generation in MLCs. These actions of IFN-γ were not mediated by TNF-α production or Fas-Fas ligand interactions. In vivo studies revealed exaggerated expansion and, subsequently, impaired deletion of superantigen-reactive T lymphocytes in IFN-γ−/− mice injected with staphylococcal enterotoxin B. Taken together, our findings indicate that IFN-γ does not hinder but instead facilitates induction of long-term allograft survival possibly by limiting expansion of activated T cells.

IFN-γ is an immunoregulatory cytokine secreted by activated T lymphocytes and NK cells (1, 2). It enhances Ag presentation by up-regulating MHC expression and promotes cellular immunity by activating macrophages, NK cells, and Th1 lymphocytes (1, 2). It is therefore postulated that IFN-γ mediates acute transplant rejection while IFN-γ neutralization favors long-term engraftment. This hypothesis is supported by studies demonstrating increased IFN-γ expression in acutely rejected cardiac, renal, and pancreatic islet cell allografts, and diminished IFN-γ expression in long-term surviving transplants (3, 4, 5). Moreover, IFN-γ administration has been shown to precipitate acute rejection in animals previously rendered tolerant to donor Ags (6).

Other studies, however, provided evidence that IFN-γ is not essential for acute rejection. For example, in vivo neutralization of IFN-γ failed to prolong survival of fully allogeneic skin grafts in rhesus monkeys or MHC class I disparate skin grafts in mice (7, 8). We recently demonstrated that IFN-γ gene-knockout (IFN-γ−/−)3 mice reject fully allogeneic, vascularized cardiac transplants at the same rate as wild-type (IFN-γ+/+) recipients (9). Furthermore, Dalton et al. observed increased proliferation and CTL activity among IFN-γ−/− splenocytes stimulated with allogeneic cells suggesting that endogenously produced IFN-γ down-regulates alloimmune responses in vitro (10).

In this study, we tested the hypothesis that induction of long-term allograft acceptance is facilitated in the absence of IFN-γ by comparing survival of skin and heart allografts between IFN-γ+/+ and IFN-γ−/− recipient mice treated with inhibitors of T cell costimulation. Blocking B7-CD28 and CD40-CD40L T cell costimulation pathways in rodents leads to long-term graft survival and in some cases donor-specific tolerance (5, 11, 12, 13). We report that, contrary to the hypothesis, induction of long-term allograft survival by T cell costimulation blockade is dependent on IFN-γ expression. We also provide evidence that endogenously produced IFN-γ limits expansion of activated T cells, a mechanism by which this cytokine could facilitate long-term acceptance of transplanted organs.

Male C3H/He (H-2k), C3H/He-FasLgld, BALB/c (H-2d), C57BL/6 (H-2b), and IFN-γ−/− C57BL/6 mice were purchased from The Jackson Laboratory, Bar Harbor, ME. IFN-γ−/− BALB/c mice were provided by Dr. David W. Pascual (Montana State University, Bozeman, MT) with permission from Drs. D. Dalton and T. Stewart (Genentech, South San Francisco, CA) (10). All IFN-γ−/− mice were bred at Emory University (Atlanta, GA)/Veterans Affairs Medical Center animal facility in microisolators supplied with sterile food and water. Inactivation of IFN-γ gene function in these animals was confirmed by performing IFN-γ-specific ELISA (Genzyme, Cambridge, MA) on splenocyte supernatants collected at 0, 24, 48, 72, and 96 h after Con A stimulation (3 μg/ml) (9).

Human rCTLA4Ig, which blocks B7-CD28 T cell costimulation pathway (11), was provided by Dr. Peter S. Linsley (Bristol-Myers Squibb Pharmaceutical Research Institute, Seattle, WA). Anti-mouse gp39 Ab (MR1), which blocks CD40-CD40L interaction (14), was provided by Drs. Christian P. Larsen and Thomas C. Pearson (Emory University School of Medicine). Neutralizing, monoclonal, hamster anti-mouse IFN-γ Ab was purchased from Genzyme, Boston, MA. Reagents were diluted in endotoxin-free PBS (Life Technologies, New Haven, CT) and sterile filtered before use.

Fully allogeneic (MHC class I and class II disparate) BALB/c tail skin grafts were transplanted to the dorsal trunk area of 6- to 8-wk-old male IFN-γ−/− and IFN-γ+/+ C57BL/6 recipients (13). Mice either were left untreated or received 250 μg of human CTLA4Ig plus 250 μg MR1 i.p. on days 0, 2, 4, and 6 post-transplantation. Skin rejection was defined as >90% graft necrosis. Skin allograft survival greater than 50 days was considered long-term survival. Fully allogeneic, vascularized, heterotopic heart transplantation was performed as described (15) using 8- to 10-wk-old male C3H/He donors and 6- to 8-wk-old male IFN-γ−/− and IFN-γ+/+ BALB/c recipients. Mice were either left untreated or received 200 μg human CTLA4Ig i.p. on the second day post-transplantation. A separate group of IFN-γ+/+ recipients treated with CTLA4Ig also received 500 μg hamster anti-mouse IFN-γ-neutralizing mAb (Genzyme) i.p. 3 days before and 4 days after transplantation. Rejection was identified by loss of palpable cardiac contractions at which time the recipient was killed and the allograft removed for analysis. Cardiac allograft survival for longer than 100 days was considered long-term survival. All procedures conformed to IACUC standards. Statistical analysis of survival data was performed using the Mann-Whitney U test.

Formalin-fixed, paraffin-embedded cardiac allograft tissue removed at the time of rejection was stained with hematoxylin and eosin or with periodic acid-Schiff and examined by a pathologist (F. K. Baddoura) who was unaware of recipients’ identity. Acute cellular rejection was graded according to the International Society for Heart Transplantation criteria (16).

Cardiac allograft and spleen tissue, resected at time of rejection, was snap-frozen in liquid nitrogen. RNA was extracted in guanidinium salt solution and purified by the CsCl gradient method (17). Five micrograms of total RNA were reverse transcribed using oligo(dT) primers and Superscript reverse transcriptase according to the manufacturer’s instructions (Life Technologies). Ten percent of cDNA was then subjected to 30 cycles of PCR amplification in a Perkin-Elmer Thermocycler 480 (Perkin-Elmer, Foster City, CA) using mouse IFN-γ-specific primer pairs (3). Fifteen percent of each PCR reaction was electrophoresed on 2% SeaKem LE agarose gels (American Bioanalytical, Natick, MA) and stained with ethidium bromide. RT-PCR controls included “no RNA” (blank) and “no reverse transcriptase” reactions.

Mouse splenocytes were enriched for T lymphocytes by applying to nylon wool columns (Polysciences, Warrington, PA) (18). One-way primary MLR was performed by incubating 4 × 105 IFN-γ+/+ or IFN-γ−/− BALB/c (H-2d) T cells with 2.5 × 104 mitomycin-treated C3H/He (H-2k) splenocytes in complete RPMI 1640 medium (10% heat-inactivated FCS, 2 mM l-glutamine, 1% nonessential amino acids, 1% sodium pyruvate, 10 mM HEPES buffer, 50 μM 2-ME, 100 U/ml penicillin, and 100 μg/ml streptomycin) (Life Technologies) at 37°C and 5% CO2 using 96-well round-bottom plates (Corning, New York, NY). After 3 to 6 days, 0.25 μCi [3H]TαR was added to each well (total well volume, 200 μl) and cells were harvested 6 h later. Net [3H]TαR uptake was determined by subtracting uptake in control wells (responder and stimulator cells cultured separately) from that in MLR wells. Proliferative responses of C3H/He and C3H/He-FasLgld T lymphocytes to BALB/c stimulators were studied in a similar fashion. Mouse rIFN-γ, neutralizing anti-mouse IFN-γ mAb, and neutralizing polyclonal rabbit anti-mouse TNF-α antiserum used in MLR experiments were obtained from Genzyme.

Splenocytes from IFN-γ+/+ or IFN-γ−/− BALB/c mice were stimulated with mitomycin-treated C3H/He splenocytes as described in the previous section. Five days later, nonadherent cells were harvested and assayed for cytotoxic activity by incubating with either 1 × 105 LK35.2 (H-2d,k) or P815 (H-2d) target cells (American Type Culture Collection, Rockville, MD) for 2.5 h (19). Target cells were preloaded with Calcein-AM (Molecular Probes, Eugene, OR) and calcein release was measured in a LS50B Luminescence Spectrometer (Perkin-Elmer) (20). Experiments in which spontaneous calcein release was greater than 30% of maximum release were rejected. Allospecific cytotoxic activity was calculated according to the following formula: % specific lysis = 100 × [(test release − spontaneous release)/(maximum release − spontaneous release)].

IFN-γ+/+ and IFN-γ−/− BALB/c male mice (6 to 8 wk old) were immunized with either PBS (Life Technologies) or 50 μg of Staphylococcus enterotoxin B (SEB) (Sigma) i.p. on day 0. To study SEB-induced T cell expansion, mice were killed on day 3 and spleen cell suspensions were analyzed by single color flow cytometry using an Ab to Vβ8.1Vβ8.2 (biotin conjugated) followed by streptavidin-cychrome. To study SEB-induced T cell deletion, mice were immunized with either PBS or 50 μg SEB i.p. on day 0 followed by either PBS or 25 μg SEB i.p. on day 3. Mice were killed on day 10 and spleen cell suspensions were analyzed by two-color flow cytometry using Abs to murine CD4 (FITC conjugated), CD8 (phycoerythrin conjugated), and either Vβ6 (biotin conjugated) or Vβ8.1Vβ8.2 (biotin conjugated) followed by streptavidin-cychrome. All flow cytometry reagents were purchased from PharMingen (San Diego, CA).

Differences in allograft survival were analyzed using the Mann-Whitney U test. All other statistical comparisons were performed using analysis of variance (ANOVA, Fisher, and Scheffé tests).

Mean survival time (MST ± SE) of tail skin allografts transplanted to untreated IFN-γ−/− mice (11 ± 1 days; n = 3) was comparable with that in untreated IFN-γ+/+ recipients (8 ± 1 days; n = 3). In contrast, administration of human rCTLA4Ig and MR1 Ab induced long-term skin allograft survival in 60% of IFN-γ+/+ recipients (MST = 56 ± 10 days; n = 7) but failed to do so in the IFN-γ−/− group (MST = 18 ± 2 days; n = 7) (Fig. 1,A) (p = 0.004, Mann-Whitney U test). Similarly, CTLA4Ig treatment resulted in long-term cardiac allograft survival in 25% of IFN-γ+/+ mice (MST >51 ± 15 days; n = 8) but in none of the IFN-γ−/− recipients (MST = 16 ± 2 days; n = 8) (Fig. 1,B) (p = 0.007, Mann-Whitney U test). CTLA4Ig also failed to induce long-term cardiac allograft survival in IFN-γ+/+ mice treated with IFN-γ-neutralizing Abs around the time of transplantation (MST = 10 ± 1 days; n = 3) (Fig. 1,B) (p = 0.01, Mann-Whitney U test, when compared with CTLA4Ig-treated IFN-γ+/+ recipients who did not receive neutralizing Abs). Histologic examination performed at the time of cessation of heart beat confirmed that allograft failure resulted from acute cellular rejection. Routine histopathologic staining did not reveal differences between hearts rejected by IFN-γ+/+ or IFN-γ−/− recipients. In both groups, grade III to IV cellular rejection characterized by mononuclear cell infiltrate, myocardial cell necrosis, and disruption of myocardial architecture was observed. RT-PCR analysis confirmed absence of IFN-γ mRNA expression in cardiac allograft and spleen tissue removed from IFN-γ−/− mice at the time of rejection (Fig. 2).

FIGURE 1.

Effect of endogenous IFN-γ on long-term allograft survival. A, Skin allograft survival (BALB/c, H-2d to C57BL/6, H-2b) in IFN-γ+/+ (•) and IFN-γ−/− (○) mice treated with CTLA4Ig and MR1 on days 0, 2, 4, and 6 after transplantation. Rejection was defined as greater than 90% graft necrosis. B, Cardiac allograft survival (C3H/He, H-2k to BALB/c, H-2d) in IFN-γ+/+ (•) and IFN-γ−/− (○) mice treated with a single dose of CTLA4Ig on day 2 after transplantation. A separate group of IFN-γ+/+ recipients (▪) was also treated with IFN-γ-neutralizing Ab (500 μg i.p.) 3 days before and 4 days after transplantation. Rejection was defined as loss of palpable contractions in the transplanted heart.

FIGURE 1.

Effect of endogenous IFN-γ on long-term allograft survival. A, Skin allograft survival (BALB/c, H-2d to C57BL/6, H-2b) in IFN-γ+/+ (•) and IFN-γ−/− (○) mice treated with CTLA4Ig and MR1 on days 0, 2, 4, and 6 after transplantation. Rejection was defined as greater than 90% graft necrosis. B, Cardiac allograft survival (C3H/He, H-2k to BALB/c, H-2d) in IFN-γ+/+ (•) and IFN-γ−/− (○) mice treated with a single dose of CTLA4Ig on day 2 after transplantation. A separate group of IFN-γ+/+ recipients (▪) was also treated with IFN-γ-neutralizing Ab (500 μg i.p.) 3 days before and 4 days after transplantation. Rejection was defined as loss of palpable contractions in the transplanted heart.

Close modal
FIGURE 2.

RT-PCR analysis of IFN-γ and hypoxanthine phosphoribosyltransferase mRNA in A) cardiac allograft or B) spleen tissue removed at the time of rejection. All recipients were treated with a single dose of CTLA4Ig on day 2 after transplantation. cDNA was subjected to 30 cycles of PCR, and 15% of each PCR reaction was electrophoresed on 2% agarose gels and stained with ethidium bromide. L = DNA ladder (*X174RF DNA/HaeIII fragments or 1 kb); +/+ = IFN-γ+/+ recipients; −/− = IFN-γ−/− recipients; and +/+* = IFN-γ+/+ recipients treated with IFN-γ-neutralizing Ab perioperatively.

FIGURE 2.

RT-PCR analysis of IFN-γ and hypoxanthine phosphoribosyltransferase mRNA in A) cardiac allograft or B) spleen tissue removed at the time of rejection. All recipients were treated with a single dose of CTLA4Ig on day 2 after transplantation. cDNA was subjected to 30 cycles of PCR, and 15% of each PCR reaction was electrophoresed on 2% agarose gels and stained with ethidium bromide. L = DNA ladder (*X174RF DNA/HaeIII fragments or 1 kb); +/+ = IFN-γ+/+ recipients; −/− = IFN-γ−/− recipients; and +/+* = IFN-γ+/+ recipients treated with IFN-γ-neutralizing Ab perioperatively.

Close modal

To investigate the response of IFN-γ−/− T lymphocytes to allogeneic stimulation, we measured their proliferation and CTL activity in primary, one-way MLR. As shown in Figure 3,A, IFN-γ−/− T cells had significantly higher [3H]TαR uptake than IFN-γ+/+ T lymphocytes (p < 0.05, ANOVA). This difference was apparent between the third and sixth days of the MLR. Addition of mouse rIFN-γ significantly reduced [3H]TαR uptake by IFN-γ−/− responders while IFN-γ-neutralizing Ab significantly increased [3H]TαR uptake by IFN-γ+/+ responders (Fig. 3,A) (p < 0.05, ANOVA). TNF-α neutralization did not block IFN-γ-mediated suppression of [3H]thymidine uptake in either IFN-γ+/+ or IFN-γ−/− MLRs (Fig. 3,B) (p > 0.05, ANOVA). Moreover, C3H/He-FasLgld responders, which lack functional FasL, displayed the same increase in [3H]TαR uptake as C3H/He responders when IFN-γ-neutralizing Ab was added to the MLR (Fig. 3,C) (p > 0.05, ANOVA). We also observed that allospecific CTL activity was significantly greater in cultures in which IFN-γ was either absent or neutralized with mAb (Fig. 4). Enhanced [3H]thymidine uptake and increased CTL activity in IFN-γ−/− MLR were not due to increased IL-2 secretion (Fig. 5). In fact, IL-2 production was higher in IFN-γ+/+ MLRs. As expected, IFN-γ was present in supernatants of IFN-γ+/+ MLRs (1.5 ng/ml at 48 h) but was absent in IFN-γ−/− MLRs. IL-4 was below detection limit of the assay (<5 pg/ml) in either culture.

FIGURE 3.

Effect of IFN-γ on [3H]TαR uptake in primary, one-way MLR. A, T cell-enriched BALB/c (H-2d) IFN-γ+/+ or IFN-γ−/− splenocytes were stimulated with mitomycin-treated C3H/He (H-2k) splenocytes, and [3H]TαR uptake was measured on days 3 through 6. •, IFN-γ+/+ responders; ○, IFN-γ−/− responders; ▪, IFN-γ+/+ responders in the presence of 2 μg/ml IFN-γ-neutralizing Ab; □, IFN-γ−/− responders in the presence of 500 U/ml mouse rIFNγ. Results shown are mean ± SD of six experiments. B, T cell-enriched splenocytes from BALB/c IFN-γ+/+ (closed bars) or IFN-γ−/− (open bars) mice were stimulated with mitomycin-treated C3H/He splenocytes in the presence (+) or absence (−) of reagents indicated on the horizontal axis (IFN-γ, 500 U/ml; anti-IFN-γ, 10 μg/ml; anti-TNF-α, 2 μg/ml). [3H]TαR uptake was measured on day 4 of the MLR. Results shown are mean ± SD of three experiments. C, T cell-enriched splenocytes from C3H/He (stippled bars) or C3H/He-FasLgld (striped bars) mice were stimulated with mitomycin-treated BALB/c splenocytes alone (medium) or in the presence of either 500 U/ml mouse rIFNγ or 10 μg/ml IFN-γ-neutralizing Ab (anti-IFN-γ). [3H]TαR uptake was measured on day 4 of the MLR. Results are mean ± SD of three experiments. Statistical analyses are described in the text.

FIGURE 3.

Effect of IFN-γ on [3H]TαR uptake in primary, one-way MLR. A, T cell-enriched BALB/c (H-2d) IFN-γ+/+ or IFN-γ−/− splenocytes were stimulated with mitomycin-treated C3H/He (H-2k) splenocytes, and [3H]TαR uptake was measured on days 3 through 6. •, IFN-γ+/+ responders; ○, IFN-γ−/− responders; ▪, IFN-γ+/+ responders in the presence of 2 μg/ml IFN-γ-neutralizing Ab; □, IFN-γ−/− responders in the presence of 500 U/ml mouse rIFNγ. Results shown are mean ± SD of six experiments. B, T cell-enriched splenocytes from BALB/c IFN-γ+/+ (closed bars) or IFN-γ−/− (open bars) mice were stimulated with mitomycin-treated C3H/He splenocytes in the presence (+) or absence (−) of reagents indicated on the horizontal axis (IFN-γ, 500 U/ml; anti-IFN-γ, 10 μg/ml; anti-TNF-α, 2 μg/ml). [3H]TαR uptake was measured on day 4 of the MLR. Results shown are mean ± SD of three experiments. C, T cell-enriched splenocytes from C3H/He (stippled bars) or C3H/He-FasLgld (striped bars) mice were stimulated with mitomycin-treated BALB/c splenocytes alone (medium) or in the presence of either 500 U/ml mouse rIFNγ or 10 μg/ml IFN-γ-neutralizing Ab (anti-IFN-γ). [3H]TαR uptake was measured on day 4 of the MLR. Results are mean ± SD of three experiments. Statistical analyses are described in the text.

Close modal
FIGURE 4.

Effect of endogenous IFN-γ on CTL generation in primary, one-way MLR. Allospecific CTL activity was determined on the fifth day of a C3H/He (H-2k)→BALB/c (H-2d) MLR by measuring calcein released from LK35.2 (H-2d,k) target cells. •, IFN-γ+/+ responders; ○, IFN-γ−/− responders; ▪, IFN-γ+/+ responders stimulated in the presence of 10 μg/ml IFN-γ-neutralizing Ab. Results shown are mean ± SD of three experiments. Nonspecific lysis of P815 (H-2d) target cells was <2%. *p < 0.05 compared with IFN-γ+/+ (•) CTL activity.

FIGURE 4.

Effect of endogenous IFN-γ on CTL generation in primary, one-way MLR. Allospecific CTL activity was determined on the fifth day of a C3H/He (H-2k)→BALB/c (H-2d) MLR by measuring calcein released from LK35.2 (H-2d,k) target cells. •, IFN-γ+/+ responders; ○, IFN-γ−/− responders; ▪, IFN-γ+/+ responders stimulated in the presence of 10 μg/ml IFN-γ-neutralizing Ab. Results shown are mean ± SD of three experiments. Nonspecific lysis of P815 (H-2d) target cells was <2%. *p < 0.05 compared with IFN-γ+/+ (•) CTL activity.

Close modal
FIGURE 5.

IL-2 production by BALB/c (H-2d) IFN-γ+/+ (•) or IFN-γ−/− (○) T cell-enriched splenocytes stimulated with mitomycin-treated C3H/He (H-2k) splenocytes in a primary, one-way MLR. Results shown are mean ± SD of three experiments. *p < 0.05.

FIGURE 5.

IL-2 production by BALB/c (H-2d) IFN-γ+/+ (•) or IFN-γ−/− (○) T cell-enriched splenocytes stimulated with mitomycin-treated C3H/He (H-2k) splenocytes in a primary, one-way MLR. Results shown are mean ± SD of three experiments. *p < 0.05.

Close modal

Injecting SEB superantigen into mice causes rapid proliferation followed by death of T cells bearing Vβ8 TCR (21). We therefore compared SEB-induced clonal expansion and deletion of Vβ8+ T cells in IFN-γ+/+ mice to that in IFN-γ−/− mice. Percent change in Vβ8+ cells was calculated in reference to control mice that received PBS. As shown in Figure 6,A, expansion of Vβ8+ splenocytes 3 days after a single SEB injection was significantly greater in the IFN-γ−/− group. Moreover, percent deletion of Vβ8+ T cells (CD4+ or CD8+) 7 days after rechallenge with SEB was lower in IFN-γ−/− mice suggesting that a greater number of Vβ8+ T cells persisted in the absence of IFN-γ (Fig. 6,B). Small compensatory changes in the non-SEB-reactive Vβ6 subset were observed in both experiments (Fig. 6, A and B, bar graphs).

FIGURE 6.

Effect of IFN-γ on superantigen-induced clonal T cell expansion and deletion in vivo. A, Expansion of Vβ8+ splenocytes was determined by single-color flow cytometry (histograms) 3 days after injecting mice i.p. with either SEB or PBS. The percentage of change in Vβ8+ or Vβ6+ splenocytes in SEB-injected IFN-γ+/+ (closed bars) or IFN-γ−/− (open bars) mice relative to animals that received PBS is shown in the bar graph. B, Mice were immunized with SEB or PBS on days 0 and 3. Seven days later, deletion of Vβ8+CD4+ or Vβ8+CD8+ splenocytes was determined by two-color flow cytometry (density plots). The percentage of change in Vβ8+ or Vβ6+ T lymphocytes in SEB-injected IFN-γ+/+ (closed bars) or IFN-γ−/− (open bars) mice relative to animals that received PBS is shown in the bar graph. Results are mean ± SD of three Vβ8 experiments and one representative Vβ6 experiment. *p < 0.05; p = 0.06 when the IFN-γ+/+ group (closed bars) is compared with the IFN-γ−/− group (open bars).

FIGURE 6.

Effect of IFN-γ on superantigen-induced clonal T cell expansion and deletion in vivo. A, Expansion of Vβ8+ splenocytes was determined by single-color flow cytometry (histograms) 3 days after injecting mice i.p. with either SEB or PBS. The percentage of change in Vβ8+ or Vβ6+ splenocytes in SEB-injected IFN-γ+/+ (closed bars) or IFN-γ−/− (open bars) mice relative to animals that received PBS is shown in the bar graph. B, Mice were immunized with SEB or PBS on days 0 and 3. Seven days later, deletion of Vβ8+CD4+ or Vβ8+CD8+ splenocytes was determined by two-color flow cytometry (density plots). The percentage of change in Vβ8+ or Vβ6+ T lymphocytes in SEB-injected IFN-γ+/+ (closed bars) or IFN-γ−/− (open bars) mice relative to animals that received PBS is shown in the bar graph. Results are mean ± SD of three Vβ8 experiments and one representative Vβ6 experiment. *p < 0.05; p = 0.06 when the IFN-γ+/+ group (closed bars) is compared with the IFN-γ−/− group (open bars).

Close modal

We examined in this study whether induction of long-term allograft survival is facilitated in the absence of IFN-γ. Surprisingly, we found that T cell costimulation blockade induces long-term acceptance of skin and heart allografts in wild-type recipients but fails to do so in IFN-γ−/− mice or in wild-type mice treated with IFN-γ-neutralizing Ab. These findings indicate that IFN-γ is required for successful induction of long-term allograft survival. Our results are consistent with other experimental systems in which endogenous IFN-γ was found to have immunosuppressive actions. For example, (IFN-γR−/−) mice were more susceptible to collagen-induced or Staphylococcus aureus-induced arthritis than their wild-type counterparts (22, 23, 24). Similarly, IFN-γR−/− mice were hypersensitive to anti-CD3-induced cytokine release syndrome (25), and IFN-γ−/− or IFN-γR−/− mice had exaggerated mortality following induction of experimental allergic encephalomyelitis (26, 27). IFN-γ production also accounted for the generalized immunosuppression that accompanies murine graft-vs-host disease (28).

We also addressed in this report the mechanisms by which IFN-γ could contribute to long-term allograft acceptance. We found that IFN-γ is an endogenous inhibitor of T cell proliferation and CTL generation in primary MLRs. It also inhibited superantigen-induced clonal expansion of Vβ8+ T lymphocytes in vivo. These observations suggest that IFN-γ regulates immune responses by limiting expansion of activated T lymphocytes. In fact, previous studies have shown that exogenous IFN-γ suppresses proliferation of several cell types including T cell lines (2, 10, 29). It has also been proposed that IFN-γ promotes death of activated T lymphocytes, particularly cloned Th1 cells (30). We found, however, that IFN-γ-induced inhibition of [3H]TαR uptake in MLRs is not mediated through TNF-α-TNF-αR or Fas-FasL cell death pathways and was not accompanied by preferential decrease in Th1-derived cytokines. Others also failed to demonstrate a pro-apoptotic effect of IFN-γ in primary T cell cultures (31, 32, 33). Additional mechanisms by which IFN-γ could exert immunosuppressive effects include stimulation of macrophage hydrogen peroxide and nitric oxide production, which in turn inhibit T cell proliferation (34, 35). It has also been suggested that IFN-γ contributes to T cell anergy by inducing MHC class II molecules on nonprofessional APCs (36) or by enhancing “natural suppressor cell” activity (37).

Although endogenous IFN-γ appears to limit T cell proliferation in unmodified MLRs, accelerated acute allograft rejection was not observed in untreated IFN-γ−/− mice (9). It is possible that the immunoregulatory functions of IFN-γ are masked by the extreme magnitude and plurality of the cytokine response observed in unmodified allograft recipients. Discrepancies between in vitro and in vivo findings relating to acute allograft rejection have been demonstrated in other cytokine gene knockout mice (38, 39). For example, unmodified IL-2 gene knockout mice acutely reject pancreatic islet cell and cardiac allografts (our unpublished observations) despite near absence of T cell proliferation in a primary MLR (39).

We also demonstrated in this study that administration of IFN-γ-neutralizing Ab to wild-type recipients prevents induction of long-term allograft survival. This indicates that failure to achieve allograft acceptance in IFN-γ−/− mice did not result from overcompensation by a redundant cytokine system such as TNF-α, which could have developed during embryogenesis of these mice. In fact, we did not detect immunoreactive TNF-α in sera from IFN-γ−/− recipients, and TNF-α production by allostimulated IFN-γ−/− splenocytes was similar to that by IFN-γ+/+ splenocytes (our unpublished observations). Studies which examined the role of exogenous IFN-γ in transplantation tolerance have yielded conflicting results. Paineau et al. found that administration of rIL-2 shortened cardiac allograft survival in cyclosporin-treated rats when given alone but failed to do so when given with rIFN-γ, leading the authors to suggest that IFN-γ has immunosuppressive effects (40). The same group, however, reported later that rat rIFN-γ, given 1 wk after a tolerizing regimen of donor blood transfusions, abolished tolerance to allografts transplanted between congenic rat strains (6). Chen et al. also demonstrated that rIFN-γ counteracts neonatal tolerance induction in mice (41). Exogenous IFN-γ, therefore, could promote or hinder long-term allograft survival depending on time of administration and tolerance model studied.

The mechanisms by which perioperative T cell costimulation blockade leads to long-term allograft survival or transplantation tolerance are not known. It has been proposed that CTLA4Ig promotes Th1 to Th2 immune deviation and thus protects against allograft rejection (5). We observed in this study that CTLA4Ig-induced long-term allograft survival is not facilitated but, instead, hindered by the absence of IFN-γ. Moreover, IL-4−/− mice, which are deficient in Th2 cytokine production, accepted heart allografts indefinitely when treated with CTLA4Ig (38). Taken together, these data strongly suggest that immune deviation is not a principal mechanism by which CTLA4Ig induces long-term allograft survival. Using a TCR transgenic model, Judge et al. demonstrated that CTLA4Ig suppresses immune responses partly by inhibiting expansion of Ag-reactive cells (42). Our results extend this observation by providing evidence that endogenous IFN-γ facilitates CTLA4Ig-induced long-term allograft survival possibly by down-regulating proliferation of activated T cells. In addition to limiting lymphocyte expansion, costimulation blockade also induces a state of ignorance or anergy in residual Ag-reactive T cells (13, 42, 43). Larsen et al. proposed that silencing of donor-specific T lymphocytes in mice treated with inhibitors of B7-CD28 and CD40-CD40L interactions is an active process requiring signaling through the TCR because concomitant cyclosporin A administration resulted in premature rejection of skin grafts (13). Cyclosporin A suppresses synthesis of cytokines such as IFN-γ, which are secreted by activated T cells (44). We demonstrated in this study that IFN-γ is critical for achieving long-term allograft survival, further suggesting that tolerance induction depends on stimulation of T lymphocytes by alloantigen. These observations are pertinent to testing CTLA4Ig and MR1 in clinical transplantation in which the majority of patients are treated with cyclosporin A.

1

This work was supported by grants from the Carlos and Marguerite Mason Trust (F.G.L., C.P.L., T.C.P.), National Institutes of Health (1R29 AI41643-01, F.G.L.), and Veterans Affairs Merit Review (F.G.L.). C.P.L. and T.C.P. are also supported in part by National Institutes of Health Grants 1R29 AI33588-01, 1R01 DK50762-01, and 1R01 AI40519-01.

3

Abbreviations used in this paper: IFN-γ−/−, IFN-γ gene knockout; IFN-γ+/+, wild-type strain; SEB, staphylococcal enterotoxin B; MST, mean survival time; L, ligand (as in CD40L).

1
Boehm, U., T. Klamp, M. Groot, J. Howard.
1997
. Cellular responses to interferon.
Annu. Rev. Immunol.
15
:
749
2
Billiau, A..
1996
. Interferon.
Adv. Immunol.
62
:
61
3
Takeuchi, T., R. P. Lowry, B. Konieczny.
1992
. Heart allografts in murine systems.
Transplantation
53
:
1281
4
O’Connell, P. J., A. Pacheco-Silva, P. W. Nickerson, R. A. Muggia, M. Bastos, V. E. Rubin-Kelley, T. B. Strom.
1993
. Unmodified pancreatic islet allograft rejection results in the preferential expression of certain T cell activation transcripts.
J. Immunol.
150
:
1093
5
Sayegh, M. H., E. Akalin, W. W. Hancock, M. E. Russell, C. B. Carpenter, P. S. Linsley, L. A. Turka.
1995
. CD28-B7 blockade after alloantigenic challenge in vivo inhibits Th1 cytokines but spares Th2.
J. Exp. Med.
181
:
1869
6
Bugeon, L., M. C. Cuturi, M. M. Hallet, J. Paineau, D. Chabannes, J. P. Soulillou.
1992
. Peripheral tolerance of an allograft in adult rats: characterization by low interleukin-2 and interferon-gamma mRNA levels and by strong accumulation of major histocompatibility complex transcripts in the graft.
Transplantation
54
:
219
7
Stevens, H. P. J. D., T. H. Van der Kwast, P. Van der Meide, V. D. Vuzevski, W. A. Buurman, M. Jonker.
1990
. Synergistic immunosuppression effects of monoclonal antibodies specific for interferon.
Transplantation
50
:
856
8
Rosenberg, A. S., D. S. Finbloom, T. G. Maniero, P. H. Van der Meide, A. Singer.
1990
. Specific prolongation of MHC class 2 disparate skin allografts by in vivo administration of anti-IFN.
J. Immunol.
144
:
4648
9
Saleem, S., B. T. Konieczny, R. P. Lowry, F. K. Baddoura, F. G. Lakkis.
1996
. Acute rejection of vascularized heart allografts in the absence of IFN.
Transplantation
62
:
1908
10
Dalton, D. K., S. Pitts-Meek, S. Keshav, I. S. Figari, A. Bradley, T. Stewart.
1993
. Multiple defects of immune cell function in mice with disrupted interferon.
Science
259
:
1793
11
Linsley, P. S., P. M. Wallace, J. Johnson, M. G. Gibson, J. L. Greene, J. A. Ledbetter, C. Singh, M. A. Tepper.
1992
. Immunosuppression in vivo by a soluble form of the CTLA-4 T cell activation molecule.
Science
257
:
792
12
Pearson, T. C., D. Z. Alexander, K. J. Winn, P. S. Linsley, R. P. Lowry, C. P. Larsen.
1994
. Transplantation tolerance induced by CTLA4-Ig.
Transplantation
57
:
1701
13
Larsen, C., E. Elwood, D. Alexander, S. Ritchie, R. Hendrix, C. Tucker-Burden, H.-R. Cho, A. Aruffo, D. Hollenbaugh, P. Linsley, K. Winn, T. Pearson.
1996
. Long-term acceptance of skin and cardiac allografts after blocking CD40 and CD28 pathways.
Nature
381
:
434
14
Noelle, R. J., M. Roy, D. M. Shepherd, I. Stamenkovic, J. A. Ledbetter, A. Aruffo.
1992
. A 39-kDa protein on activated helper T cells binds CD40 and transduces the signal for cognate activation of B cells.
Proc. Natl. Acad. Sci. USA
89
:
6550
15
Corry, R. J., H. J. Winn, P. S. Russel.
1973
. Primarily vascularized allografts of hearts in mice: the role of H-2D, H-2K, and non H-2 antigens.
Transplantation
16
:
343
16
Billingham, M. E., N. R. B. Cary, M. E. Hammond, J. Kemnitz, C. Marboe, H. A. McCallister, D. C. Snovar, G. L. Winters, A. Zerbe.
1990
. A working formulation for the standardization of nomenclature in the diagnosis of heart and lung rejection: heart rejection study group.
J. Heart Transplant.
9
:
587
17
Kingston, R. E..
1994
. Guanidinium method for total RNA preparation. J. E. Coligan, and A. M. Kruisbeek, and D. H. Margulies, and E. M. Shevach, and W. Strober, eds.
Current Protocols in Immunology
10.11.5
Wiley, New York.
18
Hathcock, K. S..
1994
. Isolation of mouse mononuclear cells. J. E. Coligan, and A. M. Kruisbeek, and D. H. Margulies, and E. M. Shevach, and W. Strober, eds.
Current Protocols in Immunology
3.1.1
Wiley, New York.
19
Wunderlich, J., G. Shearer.
1991
. Induction and measurement of cytotoxic T lymphocyte activity. J. E. Coligan, and A. M. Kruisbeek, and D. H. Margulies, and E. M. Shevach, and W. Strober, eds.
Current Protocols in Immunology
3.11.1
Wiley, New York.
20
Biddison, W. E., R. Lichtenfels, M. Adibzadeh, R. Martin.
1994
. Measurement of CTL activity by the calcein release methos. J. E. Coligan, and A. M. Kruisbeek, and D. H. Margulies, and E. M. Shevach, and W. Strober, eds.
Current Protocols in Immunology, Suppl. 17
7.17.4
Wiley, New York.
21
Scherer, M. T., L. Ignatowicz, G. M. Winslow.
1993
. Superantigens: bacterial and viral proteins that manipulate the immune system.
Annu. Rev. Cell Biol.
9
:
101
22
Manoury-Schwartz, B., G. Chiocchia, N. Bessis, O. Abehsira-Amar, F. Batteux, S. Muller, S. Huang, M.-C. Boissier, C. Fournier.
1997
. High susceptibility to collagen-induced arthritis in mice lacking IFN.
J. Immunol.
158
:
5501
23
Vermeire, K., H. Heremans, M. Vandeputte, S. Huang, A. Billiau, P. Matthys.
1997
. Accelerated collagen-induced arthritis in IFN.
J. Immunol.
158
:
5507
24
Zhao, Y.-X., A. Tarkowski.
1995
. Impact of interferon.
J. Immunol.
155
:
5736
25
Matthys, P., G. Froyen, L. Verdot, S. Huang, H. Sobis, J. Van Damme, B. Vray, M. Aguet, A. Billiau.
1995
. IFN.
J. Immunol.
155
:
3823
26
Ferber, I. A., S. Brocke, C. Taylor-Edwards, W. Ridgway, C. Dinisco, L. Steinman, D. Dalton, C. G. Fathman.
1996
. Mice with a disrupted IFN.
J. Immunol.
156
:
5
27
Willenborg, D. O., S. Fordham, C. C. A. Bernard, W. B. Cowden, I. A. Ramshaw.
1996
. IFN-γ plays a critical down-regulatory role in the induction and effector phase of myelin oligodendrocyte glycoprotein-induced autoimmune encephalomyelitis.
J. Immunol.
157
:
3223
28
Wall, D. A., S. D. Hamberg, S. J. Burakoff, D. S. Reynolds, A. K. Abbas, L. M. Ferrara.
1988
. Immunodeficiency in graft-versus-host disease. I. Mechanism of immune suppression.
J. Immunol.
140
:
2970
29
Gajewski, T. F., F. W. Fitch.
1988
. Anti-proliferative effect of IFN.
J. Immunol.
140
:
4245
30
Liu, Y., C. A. Janeway.
1990
. Interferon.
J. Exp. Med.
172
:
1735
31
Russell, J. H., B. J. Rush, S. I. Abrams, R. Wang.
1992
. Sensitivity of T-cells to anti-CD3-stimulated suicide is independent of functional phenotype.
Eur. J. Immunol.
22
:
1655
32
Wesselborg, S., O. Janssen, D. Kabelitz.
1993
. Induction of activation-driven death (apoptosis) in activated but not resting peripheral blood T cells.
J. Immunol.
150
:
4338
33
Radvanyi, L. G., G. B. Mills, R. G. Miller.
1993
. Religation of the T cell receptor after primary activation of mature T cells inhibits proliferation and induces apoptotic cell death.
J. Immunol.
150
:
5704
34
Metzger, Z., J. T. Hoffeld, J. J. Oppenheim.
1980
. Macrophage-mediated suppression. I. Evidence for participation of both hydrogen peroxide and prostaglandin in suppression of murine lymphocyte proliferation.
J. Immunol.
124
:
983
35
Albina, J. E., J. A. Abate, W. L. J. Henry.
1991
. Nitric oxide production is required for murine resident peritoneal macrophages to suppress mitogen-stimulated T cell proliferation. Role of IFN.
J. Immunol.
147
:
144
36
Grabbe, S., S. Bruvers, S. Beissert, R. D. Granstein.
1994
. Interferon-gamma inhibits tumor antigen presentation by epidermal antigen-presenting cells.
J. Leukocyte Biol.
55
:
695
37
Huchet, R., M. Bruley-Rosset, C. Mathiot, D. Grandjon, O. Halle-Pannenko.
1993
. Involvement of IFN-γ and transforming growth factor.
J. Immunol.
150
:
2517
38
Lakkis, F. G., B. T. Konieczny, S. Saleem, F. K. Baddoura, P. S. Linsley, D. Z. Alexander, R. P. Lowry, T. C. Pearson, C. P. Larsen.
1997
. Blocking the CD28-B7 T cell costimulation pathway induces long-term cardiac allograft acceptance in the absence of interleukin-4.
J. Immunol.
158
:
2443
39
Steiger, J., P. W. Nickerson, W. Steurer, M. Moscovitch-Lopatin, T. B. Strom.
1995
. IL-2 knockout recipient mice reject islet cell allografts.
J. Immunol.
155
:
489
40
Paineau, J., C. Priestley, J. Fabre, S. Chevalier, P. van der Meide, H. Schellekens, Y. Jacques, J. P. Soulillou.
1991
. Effect of recombinant interferon gamma and interleukin-2 and of monoclonal antibody against interferon gamma on the rat immune response against heart allografts.
J. Heart Lung Transplant.
10
:
424
41
Chen, N., Q. Gao, E. H. Field.
1996
. Prevention of TH1 responses is critical for tolerance.
Transplantation
61
:
1076
42
Judge, T. A., A. Tang, L. M. Spain, J. Deans-Gratiot, M. H. Sayegh, L. A. Turka.
1996
. The in vivo mechanism of action of CTLA4Ig.
J. Immunol.
156
:
2294
43
Griggs, N. D., S. S. Agersborg, R. J. Noelle, J. A. Ledbetter, P. S. Linsley, K. S. K. Tung.
1996
. The relative contribution of CD28 and gp39 costimulatory pathways in the clonal expansion and pathogenic acquisition of self-reactive T cells.
J. Exp. Med.
183
:
801
44
Andersson, J., S. Nagy, C. G. Groth, U. Andersson.
1992
. Effects of FK506 and cyclosporine A on cytokine production studied in vitro at a single-cell level.
Immunology
75
:
136