TCRαβ+ CD4−CD8−NK− double negative T cells (DN T cells) can act as regulatory T cells to inhibit allograft rejection and autoimmunity. Their role in graft-versus-host disease and mechanisms of suppression remain elusive. In this study, we demonstrate that DN T cells can inhibit CD4+ T cell-mediated GVHD in a semi-allogeneic model of bone marrow transplantation. Furthermore, we present evidence of a novel autocrine IFNγ signaling pathway in Fas-deficient C57BL/6.lpr (B6.lpr) DN T cells. B6.lpr DN T cells lacking IFNγ or its receptor were impaired in their ability to suppress syngeneic CD4+ T cells responding to alloantigen stimulation both in vitro and in vivo. Autocrine IFNγ signaling was required for sustained B6.lpr DN T cell IFNγ secretion in vivo and for upregulation of surface Fas ligand expression during TCR stimulation. Fas ligand (FasL) expression by B6.lpr DN T cells permitted lysis of activated CD4+ T cells and was required for suppression of GVHD. Collectively, our data indicate that DN T cells can inhibit GVHD and that IFNγ plays a critical autocrine role in controlling the regulatory function of B6.lpr DN T cells.
Citation: Juvet SC, Han M, Vanama R, Joe B, Kim EY, Zhao FL, et al. (2012) Autocrine IFNγ Controls the Regulatory Function of Lymphoproliferative Double Negative T Cells. PLoS ONE 7(10): e47732. https://doi.org/10.1371/journal.pone.0047732
Editor: Joao PB. Viola, National Cancer Institute (INCA), Brazil
Received: July 10, 2012; Accepted: September 14, 2012; Published: October 15, 2012
Copyright: © Juvet et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by Canadian Institutes of Health Research (http://cihr-irsc.gc.ca/) Grant #14431 and Canadian Cancer Society (http://www.cancer.ca/Research.aspx) Grant #020516 to LZ. SCJ received salary support from a Terry Fox Foundation/National Cancer Institute of Canada (http://www.cancer.ca/Research.aspx) Clinical Research Fellowship #18014 and the Clinician-Scientist Training Program, Department of Medicine, University of Toronto (http://www.deptmedicine.utoronto.ca/edustudies/CSTProgram.htm). LZ is the Maria H. Bacardi Chair in Transplantation at the University Health Network (www.uhn.ca). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
DN T cells express an αβ T cell receptor (TCR) but do not express either CD4 or CD8 coreceptors, nor do they express NK cell markers. This phenotype differentiates them from other unconventional T cells (e.g. NK T cells and γδ T cells). Depending on the context, DN T cells have been shown to have regulatory, innate, or pathogenic properties . In murine models, DN T cells can act as regulatory T cells (Tregs) that inhibit allo- and xenograft rejection , , ,  and autoimmune diabetes , . Our previous studies showed that TCR transgenic DN T cells attenuated CD8+ T cell-induced graft-versus-host disease (GVHD) in a single class I-mismatched mouse model . Whether DN T cells can suppress CD4+ T cell-mediated GVHD is not known. Human DN T cells inhibit autologous CD4+ and CD8+ T cell proliferation in vitro , , and the frequency of circulating DN T cells inversely correlated with the severity of GVHD in bone marrow transplantation (BMT) patients, suggesting that DN T cells might inhibit GVHD in humans .
Fas or FasL mutations in mice (lpr and gld respectively) and humans (autoimmune lymphoproliferative syndrome) exhibit lymphoproliferation, autoimmunity, and DN T cell expansion. Whereas some studies implicate DN T cells in the pathogenesis of autoimmunity in these settings, lpr DN T cells can also act as Tregs in some contexts , , for example following an infusion of allogeneic lymphocytes . This feature is shared with other murine ,  and human  DN T cells. Furthermore FasL-expressing T cells in NOD.lpr mice can resist diabetes induced by adoptively transferred T cells . Whether lpr DN T cells can inhibit syngeneic CD4+ T cell responses is of interest since the latter are autoimmune effectors in lpr and gld mice , .
IFNγ, whose inflammatory role is well described, is also immunoregulatory: it helps clear activated T cells , induces Foxp3+ Tregs , inhibits IL-17-secreting T cells , and upregulates immunoregulatory enzymes in antigen presenting cells (APCs) , . IFNγ is expressed by mouse , , rat  and human  DN T cells but its role in DN T cell function and the underlying mechanisms are not clear.
Here, we demonstrate that alloantigen-primed BALB/c and Fas-deficient B6.lpr DN T cells can act as Tregs to inhibit GVHD mediated by syngeneic CD4+ T cells in a semiallogeneic BMT model. Furthermore, we have identified a novel IFNγ-dependent autocrine mechanism that is critical for B6.lpr DN T cell-mediated immune suppression in vitro and in vivo.
DN T cells can inhibit CD4+ T cell-mediated GVHD
To date, whether non-transgenic, polyclonal DN T cells can inhibit GVHD is unknown. To address this question, lethally irradiated (B6xBALBc)F1 (CB6F1) mice (H-2b/d) were transplanted with T cell-depleted (TCD) BALB/c (H-2d) bone marrow cells alone (BM only) or with BALB/c CD4+ T cells, without (BM+CD4+) or with preactivated and enriched (by depleting B cells and CD8+ and CD4+ T cells) BALB/c DN T cells (BM+CD4++DN) as putative suppressor cells. GVHD clinical score  and recipient survival were monitored daily. Mice receiving BM alone had 100% long-term survival, while mice receiving BM+CD4+ succumbed to acute GVHD (median survival 8 days, Fig. 1A, p = 0.0003). In contrast, the CB6F1 mice treated with alloantigen-primed BALB/c cells enriched for DN T cells showed markedly reduced GVHD (5 of 7 mice survived >100 days, Fig. 1A, p = 0.0003).
A. Lethally irradiated CB6F1 mice were reconstituted with BALB/c TCD BM alone (BM only, n = 7) or with BALB/c CD4+ T cells, without (BM+CD4+, n = 6) or with BALB/c DN-enriched lymphocytes (BM+CD4++DN, n = 7). A clinical score  incorporating posture, fur texture, activity level, skin integrity and weight loss was assigned 3 times weekly and survival was monitored daily. Data from 3 independent experiments (each with 2–3 mice per group) are shown; *log rank p = 0.0003 vs. BM+CD4+. B. CB6F1 recipients conditioned as in A were reconstituted with B6 TCD BM alone (BM only) or with B6.Thy1.1 CD4+ T cells, without (BM+CD4+) or with B6.lpr DN T cells (BM+CD4++DN). Data are from 5 independent experiments, each with up to 6 mice per group; *log rank p = 0.0034 vs. BM+CD4+; **log rank p<0.0001 vs. BM+CD4+. C. Clinical score was determined 2 weeks after BMT in recipients of BM only (n = 17), BM+CD4+ (n = 24), BM+CD4++DN (n = 14) and BM+DN (n = 8). Data are derived from 5 independent experiments, each with up to 6 mice per group. Kruskal-Wallis test p<0.0001; *Dunns post test p<0.05.
Due to a very low frequency of DN T cells in normal mice, we were unable to obtain sufficient numbers of DN T cells with high purity from BALB/c mice to confirm that these results were exclusively attributable to DN T cells. On the other hand, DN T cells are greatly expanded in Fas-deficient lpr mice and can exhibit regulatory function , , although they are generally regarded as pathogenic autoimmune effector cells within the lpr context , . Whether they might inhibit GVHD, or in fact worsen the disease, is not known. Intriguingly, unlike allogeneic CD4+ or CD8+ T cells which cause severe GVHD , infusion of purified (Fig. S1) B6.lpr DN T cells pre-activated by alloantigen in vivo did not cause significant illness or mortality in lethally irradiated CB6F1 mice reconstituted with B6 (H-2b) BM (BM+DN, Fig. 1B–C). These mice experienced a mild, transient illness (clinical score ≤3, median survival >80d). To confirm that DN T cells are able to inhibit allogeneic CD4+ T cell-induced GVHD, we purified preactivated B6.lpr DN T cells and administered them to CB6F1 recipients of B6 BM and B6. Thy1.1 CD4+ T cells. Whereas recipients of BM only had no GVHD and all survived >80d (Fig. 1B) mice receiving BM+CD4+ developed acute GVHD (median survival 12d, p<0.0001). Importantly, infusion of preactivated B6.lpr DN T cells increased median survival from 12 to >80 days (log rank test; p = 0.0034), and decreased illness severity compared with BM+CD4+ treated mice (Fig. 1B–C). These data demonstrate that infusion of allogeneic B6.lpr DN T cells does not cause severe illness and can prevent death in mice undergoing CD4+ T cell-mediated GVHD.
Furthermore, GVHD protection by B6.lpr DN T cells was associated with decreased lung, liver, and intestinal infiltration by CD4+ T cells (Fig. 2A, left, middle, and right columns, respectively). Effector cell density in the intestine precluded accurate quantification, but we observed statistically significant reductions in alloreactive T cell numbers in the liver (p = 0.0033, Fig. 2B) and lungs (p = 0.0151, Fig. 2C) when B6.lpr DN T cells were included in the inoculum. Taken together, these data suggest that B6.lpr DN T cells prevent GVHD mortality by reducing CD4+ T cell infiltration of GVHD target tissues, thereby reducing organ injury.
Lethally irradiated CB6F1 recipients were reconstituted with B6 TCD BM alone (BM only) or with B6. Thy1.1 CD4+ T cells, without (BM+CD4+) or with B6.lpr DN T cells (BM+CD4++DN). Major GVHD target organs were harvested at two weeks after BMT and infiltrating donor T cells were identified by staining frozen tissue sections with anti-Thy1.1 mAb. A. Representative (3–9 mice per group) 4× photomicrographs of lung (left column), liver (middle panels), and intestine (right panels) sections are shown. The number of Thy1.1+ cells in 10 high power fields (40× objective) was reduced in mice receiving B6.lpr DN T cells (n = 7) compared with mice not receiving B6.lpr DN T cells (n = 9) in liver (B, t test p = 0.0033) and lung (C, t test p = 0.0151).
IFNγ is critical for B6.lpr DN T cell-mediated suppression of syngeneic T cell responses to alloantigen
IFNγ can contribute to GVHD pathogenesis; however, early IFNγ administration can inhibit GVHD , . DN T cells from humans , TCR transgenic mice , rats  and lpr mice  express IFNγ. To ascertain whether B6.lpr DN T cells might cause an early rise in IFNγ during GVHD, serum IFNγ levels were determined by ELISA 5 days after BMT. Compared with recipients of BM+CD4+, recipients of BM+CD4++DN had elevated serum IFNγ (Fig. 3A, p = 0.047), suggesting that IFNγ might contribute to B6.lpr DN T cell-mediated suppression of GVHD.
A. Lethally irradiated CB6F1 recipients were reconstituted with B6 TCD BM alone (BM only) or with B6. Thy1.1 CD4+ T cells, without (BM+CD4+) or with B6.lpr DN T cells (BM+CD4++DN). Five days after BMT, serum IFNγ was elevated in BM+CD4++B6.lpr DN-treated mice (n = 10) compared to BM+CD4+-treated mice (n = 7). Unpaired t test with Welch's correction, p = 0.047. B. CFSE-labelled B6. Thy1.1 CD4+ T cells were cultured with IL-2, irradiated CB6F1 splenocytes and B6.lpr DN T cells in indicated ratios, with anti-IFNγ mAb or isotype control. CFSE dilution in live Thy1.1+CD4+ cells was determined after 5 days. CFSE histograms from one of three independent experiments are shown. Numbers inside histograms reflect the percentage of cells in the CFSEhi gate. C. Compiled data from three independent experiments shown. Two-way ANOVA p = 0.004.
To determine the importance of IFNγ in B6.lpr DN T cell regulatory function, CFSE labelled Thy1.1+CD4+ T cells were co-cultured with B6.lpr DN T cells, with either neutralizing anti-IFNγ mAb or isotype control IgG. IFNγ neutralization slightly increased the proportion of proliferated CD4+ T cells in the absence of B6.lpr DN T cells (Fig. 3B, right histograms). Importantly, IFNγ neutralization markedly hampered the function of B6.lpr DN T cells (Fig. 3B–C).
To confirm the importance of DN T cell-derived IFNγ, we generated B6.lpr.IFNγ−/− mice and assessed the ability of their DN T cells to suppress CD4+ T cell proliferation. Although B6.lpr.IFNγ−/− DN T cells exerted a slight inhibition of CD4+ T cell proliferation at a 5∶1 DN:CD4+ ratio (Fig. 4A, bottom left histogram), they did not suppress this response efficiently compared with B6.lpr.IFNγ+/+ DN T cells (Fig. 4B, p = 0.0004), indicating that IFNγ secretion by B6.lpr DN T cells is involved in their inhibitory function in vitro.
A. In varying ratios, B6.lpr.IFNγ−/− or B6.lpr.IFNγ+/+ DN T cells were cultured with CFSE-labelled B6. Thy1.1 CD4+ T cells in the same conditions as Fig. 3B for 5 days. CFSE histograms, gated on Thy1.1+CD4+ cells, from one of three independent experiments are shown. Numbers inside histograms reflect the percentage of cells in the CFSEhi gate. B. Compiled data from three independent experiments are shown. Two-way ANOVA p = 0.0004. C. In varying ratios, B6.lpr DN T cells were cultured with CFSE-labelled B6. Thy1.1.IFNγR−/− or B6. Thy1.1.IFNγR+/+ CD4+ T cells in the same conditions as Fig. 3B for 5 days. Two-way ANOVA p = 0.93. D. In varying ratios, B6.lpr.IFNγ−/− or B6.lpr.IFNγ+/+ DN T cells were cultured with CFSE-labelled B6. Thy1.1 CD4+ T cells for 3 days with soluble anti-CD28 mAb and plate-bound anti-CD3 mAb. CFSE histograms from one of two independent experiments with similar results are shown. E. Data from two independent experiments are shown; two-way ANOVA p = 0.0028.
IFNγ promotes activation-induced cell death of T lymphocytes . To examine whether DN T cell-derived IFNγ might act on CD4+ T cells, we generated IFNγ receptor 1-deficient B6. Thy1.1.IFNγR−/− mice and compared the proliferation of their CD4+ T cells with that of B6. Thy1.1.IFNγR+/+ CD4+ T cells in the presence of B6.lpr DN T cells. Surprisingly, both types of CD4+ T cell were suppressed equally by B6.lpr DN T cells (p = 0.93, Fig. 4C), suggesting that inhibition of CD4+ T cell proliferation by B6.lpr DN T cells is not mediated by a direct effect of IFNγ on the former. Furthermore, we stimulated CFSE-labelled B6. Thy1.1 CD4+ T cells to proliferate using plate-bound anti-CD3 and soluble anti-CD28 antibodies. In this setting, B6.lpr DN T cells, but not B6.lpr.IFNγ−/− DN T cells, efficiently inhibited T cell proliferation (p = 0.0028, Fig. 4D–E). This observation indicates that B6.lpr DN T cells can inhibit the proliferative response of syngeneic CD4+ T cells in an IFNγ-dependent fashion, but that the B6.lpr DN T cell-derived IFNγ does not act on APCs to mediate this effect.
B6.lpr DN T cell IFNγ secretion and signaling are required for suppression of GVHD
We next tested whether IFNγ signalling might underlie B6.lpr DN T cell function. To this end, we generated B6.lpr.IFNγR−/− mice, and tested the ability of their DN T cells to suppress CD4+ T cells in vitro. To identify the specific contribution of B6.lpr DN T cell IFNγ signalling, CFSE-labelled B6. Thy1.1.IFNγR−/− CD4+ T cells were stimulated by CB6F1.IFNγR−/− splenocytes and their proliferation was measured by CFSE dilution. As shown in Fig. 5A, B6.lpr.IFNγR−/− DN T cells exhibited a decreased ability to inhibit CD4+ T cell proliferation compared with B6.lpr.IFNγR+/+ DN T cells (p = 0.004).
In varying ratios, B6.lpr.IFNγR−/− or B6.lpr.IFNγR+/+ DN T cells were cultured with CFSE-labelled B6. Thy1.1.IFNγR−/− CD4+ T cells and irradiated CB6F1.IFNγR−/− splenocytes and IL-2 for 5 days. Suppression of Thy1.1 CD4+ T cell proliferation by B6.lpr.IFNγR−/− or B6.lpr.IFNγR+/+ DN T cells was determined by flow cytometry. Data from 6 independent experiments are shown; two-way ANOVA p = 0.004. B. Lethally irradiated CB6F1 mice were reconstituted with BM only (n = 30), BM+CD4+ (n = 44), or BM+CD4+ plus B6.lpr DN (n = 34), B6.lpr.IFNγ−/− DN (n = 15), or B6.lpr.IFNγR−/− DN (n = 13). Recipients were monitored for clinical score and survival as in Fig. 1. Data from Fig. 1B (5 experiments) are included plus 4 additional independent experiments, each with 2–6 mice per group. In comparison with BM+CD4++B6.lpr DN: *log rank p = 0.016 (B6.lpr IFNγ−/− DN); ***log rank p<0.0001 (B6.lpr.IFNγR−/− DN); ***log rank p<0.0001 (BM+CD4+). C. Representative IFNγ histograms, gated on DN T cells, from one of 6 recipients of BM+CD4++B6.lpr.IFNγR+/+ DN (solid line) and one of 5 recipients of BM+CD4++B6.lpr.IFNγR−/− DN are compared with fluorescence-minus-one control (shaded). D. Compiled data from all 11 mice in C are shown. Horizontal lines represent mean MFI; t test p = 0.043. E. Two weeks after BMT, a lower number of Thy1.1+ cells were recovered from the spleens of mice treated with BM+CD4++B6.lpr DN T cells as compared with BM+CD4++B6.lpr.IFNγR−/− DN T cells. Data from 17 recipients of BM+CD4++B6.lpr.IFNγR+/+ DN T cells and 12 recipients of BM+CD4++B6.lpr.IFNγR−/− DN T cells pooled from three independent experiments are shown.
To determine whether B6.lpr DN T cells lacking either IFNγ or its receptor differ in their ability to inhibit GVHD, lethally irradiated CB6F1 recipients were reconstituted with BM alone, BM+CD4+, or BM+CD4++ B6.lpr, B6.lpr.IFNγR−/− or B6.lpr.IFNγ−/− DN T cells. Consistent with our in vitro data, mice receiving either IFNγ−/− or IFNγR−/− B6.lpr DN T cells experienced higher GVHD mortality than mice receiving IFNγ+/+ and IFNγR+/+ B6.lpr DN T cells, with median survival times of 34 days (B6.lpr.IFNγ−/− DN, p = 0.016) and 27 days (B6.lpr.IFNγR−/− DN, p<0.0001) compared with >80 days (B6.lpr DN, Fig. 5B). Taken together, these findings reveal that B6.lpr DN T cells must both secrete and respond to IFNγ in order to exert regulatory function in vitro and in vivo.
These observations prompted us to examine for evidence of an autocrine relationship between IFNγ secretion and signaling in B6.lpr DN T cells. Two weeks after BMT, splenocytes from CB6F1 recipients of BM+CD4++B6.lpr.IFNγR−/− or B6.lpr.IFNγR+/+ DN T cells were restimulated in vitro and IFNγ expression in the DN T cell compartment was examined. Consistent with an autocrine process, B6.lpr.IFNγR−/− DN T cells failed to synthesize as much IFNγ as B6.lpr DN T cells as determined by flow cytometry (median fluorescence intensity 1737±710 vs. 4370±830, p = 0.04, Fig. 5C–D).
We next assessed whether the absence of IFNγ signaling in B6.lpr DN T cells might be associated with an impaired ability to suppress donor T cells in the spleen. We observed a significantly lower number of Thy1.1+ CD4+ cells in mice receiving B6.lpr.IFNγR+/+ vs. B6.lpr.IFNγR−/− DN T cells (p = 0.0488, Fig. 5E). Taken together, these data indicate that in B6.lpr DN T cells unable to respond to IFNγ, IFNγ synthesis in response to TCR stimulation is profoundly impaired, and that this impairment is associated with a decreased ability to inhibit the expansion of alloreactive T cells in vivo during GVHD.
Autocrine IFNγ signaling increases surface FasL expression by B6.lpr DN T cells
Our previous work revealed that syngeneic Fas-sufficient T cells are suppressed by activated B6.lpr DN T cells at least partially via direct FasL-mediated cytotoxicity . To determine whether IFNγ might promote surface expression of FasL in B6.lpr DN T cells as shown in human CD4+ T cells , B6.lpr, B6.lpr.IFNγ−/−, and B6.lpr.IFNγR−/− lymphocytes were activated and examined for FasL expression. Activated B6.lpr DN T cells upregulated surface FasL over 48h (Fig. 6A, first panel and Fig. 6B) while B6.lpr.IFNγ−/− and B6.lpr.IFNγR−/− DN T cells did not (Fig. 6A, second and third panels and Fig. 6B). Importantly, all 3 types of DN T cells expressed similar levels of CD69 and CD25 as a result of this stimulation (data not shown). Consistent with a role for IFNγ signalling, supplementation of B6.lpr.IFNγ−/− DN T cell cultures with IFNγ partially restored surface FasL expression, although not to levels seen in IFNγ-sufficient B6.lpr DN T cells (Fig. 6A, fourth panel and Fig. 6B). Similarly, neutralization of IFNγ during activation of B6.lpr DN T cells reduced surface FasL expression (Fig. 6A, fifth panel and Fig. 6C). The observed differences in FasL expression were not due to altered proteolytic processing of FasL, as the addition of the metalloproteinase inhibitor TAPI-1 during the final 18h of culture resulted in slight increases in FasL expression in all three types of DN T cells (Fig. S2). Interestingly, all three types of DN T cells had similar intracellular FasL expression (data not shown), consistent with the notion that IFNγ does not influence FasL protein synthesis in B6.lpr DN T cells, but promotes its cell surface expression.
A. B6.lpr (first panel), B6.lpr.IFNγ−/− (second panel) and B6.lpr.IFNγR−/− (third panel) spleen and lymph node cells were cultured on plate-bound anti-CD3 with 5 µg/mL soluble anti-CD28 and IL-2 for 48h. B6.lpr.IFNγ−/− cells were also given 100ng/mL IFNγ (fourth panel), and B6.lpr DN T cells were also treated with anti-IFNγ mAb (fifth panel) or isotype control (not shown). Representative examples of surface FasL expression (gated on DN T cells) are shown in comparison with fluorescence-minus-one controls (shaded). Numbers in histograms reflect MFI. B. Compiled data from 12 B6.lpr, 12 B6.lpr.IFNγ−/−, and 6 B6.lpr.IFNγR−/− mice in 6 separate experiments are shown. One-way ANOVA p = 0.0098; *Bonferroni post test p<0.05 C. B6.lpr DN T cells were activated as in A for 48h with either IFNγ neutralizing antibody or isotype control. Data from 2 separate experiments with 5 individual mice are shown. Paired t test p = 0.0064.
To investigate whether autocrine IFNγ signalling is involved in the cytotoxic function of these cells, activated B6.lpr, B6.lpr.IFNγ−/−, and B6.lpr.IFNγR−/− DN T cells were used as effectors to kill preactivated B6 CD4+ T cells. While B6.lpr DN T cells efficiently killed the targets, B6.lpr.IFNγ−/− DN T cells exhibited reduced cytotoxicity (p = 0.009, Fig. 7A). B6.lpr.IFNγR−/− DN T cells, which do not express FasL, were not able to kill activated syngeneic CD4+ T cells (p<0.0001, Fig. 7A). These data support the hypothesis that autocrine IFNγ is required for killing of activated syngeneic CD4+ T cells by B6.lpr DN T cells. Consistent with these findings, whereas B6.lpr DN T cells suppressed Fas-sufficient B6.Thy1.1 CD4+ T cells, they were unable to suppress Fas-deficient B6.lpr CD4+ T cells effectively (Fig. 7B, p<0.0001). Furthermore, using 7AAD to identify dead cells, we observed a dose-dependent increase in cell death among proliferated B6. Thy1.1 CD4+ T cells, but not among B6.lpr CD4+ T cells (Fig. S3).
A. Preactivated 3H-thymidine labelled CD4+ T cells were cultured for 18h with preactivated B6.lpr, B6.lpr.IFNγ−/−, or B6.lpr.IFNγR−/− DN T cells at indicated ratios. Specific cytotoxicty was determined based on retention of 3H-thymidine in viable targets. Data are compiled from 4 experiments in which, respectively, B6.lpr (circles), B6.lpr. IFNγ−/− (squares), or B6.lpr.IFNγR−/− (triangles) DN T cells were used in 4, 3, and 2 experiments. Compared with B6.lpr DN, *Two-way ANOVA p = 0.0009 and **p<0.0001. B. In varying ratios, B6.lpr DN T cells were cultured with CFSE-labelled B6. Thy1.1 (Fas+/+) or B6.lpr (Faslpr/lpr) CD4+ T cells, irradiated CB6F1 splenocytes and IL-2. CFSE dilution in live (7-AAD−) cells was determined after 5 days. Data from 2 independent experiments are shown; two-way ANOVA p<0.0001. C. Lethally irradiated CB6F1 mice received BM only (n = 5), BM+CD4+ (n = 8), BM+CD4++2.5×106 B6.lpr DN (n = 10, FasL+) or 2.5×106 B6.gld DN (n = 12, FasL−). Data are from 2 independent experiments, each with 2–6 mice per group. *Log rank p = 0.006 vs. B6.gld DN.
Finally, to confirm the importance of B6.lpr DN T cell FasL expression in GVHD suppression, lethally irradiated CB6F1 mice were transplanted with BM+CD4+ and either B6.lpr (FasL+) or B6.gld (lacking functional FasL) DN T cells. In these experiments, a dose of 2.5×106, rather than 5×106, DN T cells was given to each recipient. Despite this lower dose, B6.lpr DN T cells significantly enhanced GVHD survival compared with B6.gld DN T cells (Fig. 7C; log rank test p = 0.0006), whereas B6.gld DN T cell-treated mice had a rate of survival comparable to mice treated with BM+CD4+ (Fig. 7C).
Collectively these data reveal that B6.lpr DN T cell regulatory function is dependent on autocrine IFNγ signaling and on expression of a functional FasL molecule in vitro and during GVHD in vivo.
In this study, we demonstrated that alloantigen-primed DN T cells can inhibit CD4+ T cell-induced GVHD in a semi-allogeneic mouse model. DN T cells from humans and rodents synthesize IFNγ , , , , but its role in DN T cell function has been unclear. Here we have shown that IFNγ secretion and signaling in B6.lpr DN T cells is required for these cells to act as Tregs toward alloreactive CD4+ T cells in vitro and in vivo. Since B6.lpr DN T cells were able to suppress proliferation of CD4+ T cells from both B6. Thy1.1.IFNγR−/− and B6. Thy1.1.IFNγR+/+ mice (Fig. 4C), and did not require APCs for this effect in vitro (Fig. 4D–E), our findings suggested that IFNγ likely acts in an autocrine fashion in B6.lpr DN T cells. Although our data do not exclude a role for B6.lpr DN T cell-derived IFNγ acting on APCs in vivo, the following findings further support a role for IFNγ autocrine signalling in the regulatory function of these cells in vivo: 1) B6.lpr.IFNγR−/− DN T cells failed to synthesize significant IFNγ in vivo (Fig. 5C–D); 2) expression of both IFNγ and its receptor by B6.lpr DN T cells was important for B6.lpr DN T cell-mediated suppression of GVHD (Fig. 5B and Fig. 5E); and 3) both IFNγ secretion and signaling were required for B6.lpr DN T cell surface FasL expression (Fig. 6), which was also required for GVHD inhibition (Fig. 7C). Collectively, these data support the notion that B6.lpr DN T cells are required to release IFNγ and respond to this cytokine by upregulating surface FasL upon TCR stimulation, in order to inhibit alloimmune responses by syngeneic Fas-sufficient CD4+ T cells. We must concede, however, that our data do not definitively demonstrate that IFNγ-mediated FasL surface expression in B6.lpr DN T cells is the mechanism by which CD4+ T cells were inhibited during GVHD. Confirmation of this phenomenon would require an approach such as inhibition of FasL expression in IFNγ- and IFNγR-sufficient B6.lpr DN T cells, or alternatively, rescue of the regulatory function of IFNγR-deficient B6.lpr DN T cells by overexpression of FasL in these cells.
Hill and colleagues induced cardiac allograft tolerance by treating rats with autologous immature DCs and LF15-0195 . Interestingly, IFNγ secretion by splenic DN T cells was required for tolerance, and IFNγ neutralization in vivo resulted in allograft rejection. IFNγ+ cells in the spleen were visualized in contact with apoptotic T cells . These findings demonstrate a regulatory role for DN T cell-secreted IFNγ in a different species and suggest that it is involved in the deletion of effector T cells. However, the mechanism by which IFNγ was acting in that model was not determined. Our data reveal that in B6.lpr DN T cells, IFNγ acts in an autocrine fashion and up-regulates surface expression of the key effector molecule FasL. Whether a similar process occurs in DN T cells from other mouse strains or species requires further investigation.
Our findings are also in keeping with recent studies on the role of IFNγ in CD4+ Foxp3+ Tregs. The latter must release IFNγ shortly after alloantigen encounter to prevent allograft rejection . Wei and colleagues  showed that Foxp3+ Tregs lacking either IFNγ or IFNγR could not prevent allograft rejection, accompanied by reduced STAT1 activation in these cells. Hence, a similar autocrine IFNγ pathway appears to be operating in alloantigen-reactive Foxp3+ Tregs.
We have demonstrated that autocrine IFNγ upregulates B6.lpr DN T cell surface expression of FasL; the latter is required for their ability to kill activated T cells and suppress GVHD (Fig. 6–7 and Fig. S3). Only B6.lpr DN T cells capable of secreting and responding to IFNγ were able to express high levels of surface FasL (Fig. 6A). Furthermore, inhibition of metalloproteinase activity resulted in only a slight increase in FasL expression in B6.lpr DN T cells regardless of their ability to synthesize or respond to IFNγ (Fig. S2). These data are consistent with a previous report that FasL is stored within secretory lysosomes in T cells, and subsequently externalized upon activation . Our findings therefore suggest that in B6.lpr DN T cells, TCR stimulation triggers IFNγ secretion, resulting in autocrine IFNγ signaling and thus externalization of pre-synthesized FasL.
Interestingly, cytotoxicity of B6.lpr DN T cells toward activated CD4+ T cells was only mildly impaired by IFNγ deficiency but completely abrogated by IFNγR deficiency (Fig. 7A). Since antigen re-encounter by previously primed effector T cells results in IFNγ secretion , we speculate that IFNγ may be provided partially by the activated CD4+ cells in the same culture. B6.lpr.IFNγR−/− DN T cells, unable to respond to IFNγ, would thus not be able to externalize FasL. This failure to upregulate surface FasL presumably explains their lack of cytotoxic effect.
Deletion of IFNγ in MRL.lpr mice resulted in reduced autoimmunity, lymphoproliferation and mortality . Indeed, IFNγ has been shown to be involved in glomerulonephritis and other autoimmune phenomena in this strain , . In contrast, lpr mice with other strain backgrounds, including C57BL/6, exhibit lymphoproliferation but only mild autoimmune manifestations in comparison with the MRL background , . We have shown here that autocrine IFNγ regulates FasL surface expression in B6.lpr DN T cells; as such, owing to the lack of Fas, deletion of IFNγ or its receptor would not be predicted to influence lymphoproliferation in B6.lpr mice. Further, owing the fact that autoimmunity is already mild or absent in this strain background, an effect of IFNγ deletion on this aspect of the lpr phenotype would be difficult to detect. Indeed, we observed no differences in the longevity or lymphocyte counts of age-matched B6.lpr, B6.lpr.IFNγ−/−, and B6.lpr.IFNγR−/− mice (data not shown).
A further important observation made in this study is that alloantigen-primed B6.lpr DN T cells caused only a very mild illness when infused into lethally irradiated allogeneic recipients (Fig. 1B–C). In contrast, transfer of anti-CD3 and IL-23-activated B6.lpr lymphocytes into syngeneic RAG1−/− mice produced severe glomerulonephritis , and DN T cells were shown to produce larger amounts of IL-17 than other MRL.lpr T cell subsets . Moreover, DN T cells have been shown to provide help to B cells in human lupus . Hence, there is strong evidence to suggest that DN T cells contribute to the pathogenesis of lupus-like disease in lpr mice. Our data therefore support the premise that the DN T cell compartment of lpr mice contains the potential for diverse immunological functions, depending on the context. Whether these diverse functions are attributable to discrete, as-yet unidentified subsets within this compartment, remains to be determined.
Taken together, our data demonstrate that alloantigen-primed DN T cells can inhibit allogeneic CD4+ T cell-induced GVHD in mice, and that IFNγ plays a critical autocrine role in controlling the regulatory function of B6.lpr DN T cells. Autocrine IFNγ signaling is important for sustained B6.lpr DN T cell IFNγ secretion and for upregulation of surface Fas ligand expression, which in turn leads to killing of activated anti-host Fas+ CD4+ T cells. The role of DN T cell-secreted IFNγ in Fas- and FasL-deficient autoimmune lymphoproliferative states is unclear, and requires further investigation.
Materials and Methods
Animals were housed in the Toronto Medical Discovery Tower under specific pathogen-free conditions. The animal use protocol (#741) was approved by the University Health Network Animal Care Committee. Animal care was conducted in accordance with the policies and guidelines of the Canadian Council on Animal Care and the Province of Ontario's Animals for Research Act.
C57BL/6 (B6, H-2b), BALB/c (H-2d) and (B6xBALB/c)F1 (CB6F1, H-2b/d) mice were from Jackson Laboratories (Bar Harbor, ME) or Harlan Sprague Dawley (Indianapolis, IN). B6 mice carrying the Faslpr/lpr mutation (B6.lpr), the FasLgld/gld mutation (B6.gld), Thy1.1-congenic B6 mice (B6. Thy1.1), and IFNγR-targeted B6 and BALB/c mice were from Jackson Laboratories and bred in-house. IFNγ-targeted B6 mice (B6.IFNγ−/−) were a gift of Dr. D. Kelvin and bred in-house. B6. Thy1.1.IFNγR−/−, B6.lpr.IFNγ−/−, B6.lpr.IFNγR−/−, and CB6F1.IFNγR−/− mice were generated by crossing appropriate strains.
Antibodies and flow cytometry
These mAbs were from eBioscience (San Diego, CA) or BioLegend (San Diego, CA): PE-conjugated anti-IFNγ, anti-CD4, anti-CD8, anti-NK1.1, anti-CD11b, anti-CD11c, anti-CD19, anti-Gr1, Ter119, and anti-γδ TCR; allophycocyanin-conjugated anti-Thy1.1 and anti-CD4; FITC-conjugated anti-TCRβ and anti-IFNγ; biotinylated anti-CD119 (IFNγR) and anti-FasL; purified anti-CD28. Biotinylated antibodies were detected using PE-Cy7-conjugated streptavidin (eBioscience). Anti-CD3 antibody (clone 145-2C11) was prepared from a hybridoma in house. Anti-mouse IFNγ mAb (clone H22) was from R&D Systems (Minneapolis, MN), and IgG isotype control was from eBioscience. Flow cytometry was performed on a Cytomics FC500 (Beckman Coulter, Mississauga ON), an LSR II (Becton Dickinson, Mississauga ON), or an Accuri C6 (Accuri Cytometers, Ann Arbor MI). Dead cells were excluded with propidium iodide (Sigma-Aldrich, Oakville ON), fixable viability dye eFluor450 or 7-aminoactinomycin D (7-AAD, eBioscience).
Cell purification, BMT and GVHD induction
TCD BM (>99% T cell free) was obtained by treating erythrocyte-free BM cells with anti-Thy1.2 ascites (TIB-107 hybridoma, ATCC, Manassas, VA) and Low-Tox M rabbit complement (Cedarlane Labs, Burlington ON). CD4+ T cells (>90% pure) were purified from B6. Thy1.1 spleen and lymph node cells with anti-CD4 microbeads (Miltenyi Biotec, Auburn CA). To obtain DN T cells for BMT, BALB/c, B6.lpr, B6.lpr.IFNγ−/−, or B6.lpr.IFNγR−/− mice aged 8-12 weeks were infused with 40×106 CB6F1 splenocytes to activate and expand DN T cells , . B6.lpr DN T cells were purified by removing CD4+, CD8+, NK1.1+, CD19+, CD11b+, CD11c+, γδTCR+, and Ter119+ populations with PE-conjugated mAbs and anti-PE microbeads (Miltenyi Biotec; resulting population >99% PE− and 75–90% TCRβ+CD4−CD8−NK1.1− cells; Fig. S1). BALB/c DN T cells were enriched by removing B cells by adherence to IgG-coated plates followed by depletion of CD8+ and CD4+ T cells with anti-CD4 and anti-CD8 hybridoma supernatants (RL172-4 and 3.168, respectively) and Low-Tox M rabbit complement (final preparation up to 50% DN T cells with residual B cells constituting the remainder). Cell number was adjusted based DN T cell purity to ensure that a consistent number of DN T cells were used. Male CB6F1 mice aged 6–10 weeks received 13Gy γ-irradiation (divided doses, >4h apart) in a Gammacell 40 137Cs irradiator (MDS Nordion, Ottawa ON) and were infused with 2×106 TCD BM alone or with 106 CD4+ T cells, with/without 2.5–5×106 DN T cells. Survival was monitored daily. Weights and clinical scores  were determined 2–3 times weekly. Moribund mice (score >6 or weight loss >25%) were sacrificed. Dead mice were assigned the maximum clinical score of 10. In some experiments, mice were sacrificed 2 weeks post-BMT to obtain tissues.
Histology and immunohistochemistry
Tissue specimens were snap-frozen and stored at −80°C. Frozen sections were stained with biotinylated anti-Thy1.1 mAb (eBioscience) and horseradish peroxidase-conjugated ultrastreptavidin (ID labs, London ON). Colour was developed with NovaRed (Vector Labs, Burlington ON), with haematoxylin counterstaining. Sections were examined using a Nikon Eclipse E200 microscope (Nikon Canada Inc., Mississauga ON). A blinded observer (C.J.) counted Thy1.1+ cells in 10 random high-powered (40× objective) fields. Photomicrographs were acquired using a Leica DM2000 microscope and Infinity 1 camera (Lumenera Corporation, Ottawa ON). In parallel, H&E-stained paraffin sections were examined by a pathologist (O.A.) in blinded fashion to confirm GVHD.
CD4+ T cells (107 cells/mL) were incubated (10 minutes, 37°C) in PBS containing 1 μM CFSE (Invitrogen, Carlsbad CA), followed by quenching with FBS (Gibco, Carlsbad CA). Cells were washed in α-minimum essential medium with 10% FBS, 50 mM β-mercaptoethanol, 0.1mg/mL penicillin and 0.1 mg/mL streptomycin (CM). 105 CFSE labelled cells were cultured for 5 days with 2×105 irradiated (20Gy) CB6F1 splenocytes and 50 U/mL recombinant human IL-2 (Proleukin, Chiron Corporation, Emeryville CA). Some experiments were performed using plate bound anti-CD3 and soluble anti-CD28 (5 μg/mL) antibodies instead of irradiated CB6F1 splenocytes. Purified DN T cells were added in varying ratios. Propidium iodide-negative responder cells were identified with anti-CD4 and/or anti-Thy1.1 antibodies and CFSE and subjected to flow cytometry. Percent suppression was calculated using the formula: [(%CFSEhi (DN+CD4+) – %CFSEhi (CD4+ only))/(100-%CFSEhi (CD4+ only))] ×100%. In some experiments, anti-IFNγ mAb (2.5 μg/mL) or isotype control were added. In others, 7-AAD staining and CFSE dilution were jointly examined. Cell death in the presence of B6.lpr DN T cells was compared to cell death in their absence by calculating fold change, defined as (%7-AAD+ (DN+CD4+))/(%7-AAD+ (CD4+ only)).
Two million splenocytes were cultured for 4 h in 2 mL CM containing GolgiStop (monensin, 1.5 μL/mL, BD Biosciences, Mississauga ON), with/without 50 ng/mL PMA and 500 ng/mL ionomycin (Sigma-Aldrich). Cells were stained for TCRβ, CD4, CD8, and NK1.1. IFNγ staining was performed with an intracellular staining kit (eBioscience).
The JAM assay  was used to quantitate cytotoxicity as described previously . Briefly, 2×106 B6 CD4+ T cells were added to 10×106 irradiated CB6F1 splenocytes and 50 U/mL IL-2 in 2 mL CM. After 4 days, 10 μCi/mL 3H-thymidine (Perkin-Elmer, Woodbridge ON) was added for 18 h. In parallel, 5×106 B6.lpr, B6.lpr.IFNγ−/−, or B6.lpr.IFNγR−/− DN T cells were activated in similar fashion without 3H-thymidine. Then, 104 3H-thymidine-labelled CD4+ T cells were co-cultured in duplicate or triplicate 200 μL cultures with 5×104 fresh irradiated CB6F1 splenocytes and B6.lpr DN T cells in varying ratios. The cpm of retained DNA, measured on a TopCount NXT (Perkin-Elmer), was used as an index of cell survival. At each DN:CD4+ ratio, the percent CD4+ killing was determined using the formula: [(cpm(CD4+ only) – cpm(DN+CD4+))/cpm(CD4+ alone)] ×100%.
Serum samples were assayed in duplicate for IFNγ content using a Quantikine Mouse IFNγ ELISA kit (R&D Systems), according to the manufacturer's instructions.
DN T cell FasL expression
Five million spleen and lymph node cells from B6.lpr, B6.lpr.IFNγ−/−, and B6.lpr.IFNγR−/− mice were cultured in 2 mL CM in plates coated with anti-CD3 mAb, along with 50 U/mL IL-2 and 5 μg/mL soluble anti-CD28 mAb with/without 100 ng/mL recombinant mouse IFNγ (R&D Systems). In some cases anti-IFNγ antibody (2.5 μg/mL) or isotype control were included; in others, 50 µM TAPI-1  (Peptides International, Louisville KY) or DMSO vehicle were added for the final 18h of culture. After 48 h, cells were stained for TCRβ, CD4, CD8, and NK1.1. Half of the cells were stained for surface FasL expression and fixed; the remainder were stained intracellularly for FasL.
Flow cytometry data were analyzed and presented with FlowJo 7 (Treestar, Ashland OR). Graphical presentation and statistical analysis of data were performed with Prism 5.0 (GraphPad, La Jolla CA). Survival analysis was performed using the log-rank test. Variance between two groups was assessed with either Student's t test (in some cases with Welch's correction for unequal variances) or the Mann-Whitney test; three or more groups were analyzed by ANOVA, in some cases with Bonferroni's post-test to compare groups, or the Kruskal-Wallis test in the case of non-parametric data. A p value of less than 0.05 was considered statistically significant. Error bars in all graphs represent standard deviations.
Purification of B6.lpr DN T cells. Pooled spleen and lymph node cells from B6.lpr mice were incubated with PE-conjugated antibodies to CD4, CD8, NK1.1, CD11b, CD11c, Ter119, γδTCR, and CD19, washed, and incubated with anti-PE microbeads. PE+ cells were then removed using LD columns. Aliquots of the pre-column population (left panels) and the negative fraction (right panels) were stained with FITC-conjugated TCRβ antibody and analyzed by flow cytometry. Two examples of DN T cell purification from B6.lpr mice are shown.
Matrix metalloproteinase inhibition results in a slight increase in FasL expression on B6.lpr DN T cells, regardless of their ability to secrete and respond to IFNγ. Spleen and lymph node cells from B6.lpr, B6.lpr.IFNγ−/−, and B6.lpr.IFNγR−/− mice were activated as in Fig. 6A with plate-bound anti-CD3, soluble anti-CD28, and IL-2 for 48h. During the final 18h of culture, either 50 μM TAPI-1 (dotted lines) or its DMSO vehicle (solid lines) was added to the cultures. Cells were then stained for TCRβ, CD4, CD8, NK1.1, and FasL prior to fixation and analysis by flow cytometry. Each histogram shows data from one mouse per genotype; data are from one of two experiments each with 2 mice per genotype.
Fas expression by proliferating CD4+ T cells determines their susceptibility to killing by activated DN T cells. CFSE-labelled B6.lpr (FasLPR/LPR) or B6. Thy1.1 (Fas+/+) CD4+ T cells were cultured with irradiated CB6F1 splenocytes and IL-2, without or with B6.lpr DN T cells for 5 days. These data are from the same experiment as in Fig. 7B. Responder cells, identified by CD4 (or Thy1.1) and CFSE, were stained with 7-AAD and analyzed by flow cytometry. Histograms at left show 7-AAD staining of divided (CFSE-diluted) cells for Fas+/+ Thy1.1 cells (top row) and FasLPR/LPR cells (bottom row). Numbers inside histograms reflect the percentage of 7-AAD+ cells within the gate. The fold increase in dead CD4+ T cells, defined as the percent 7-AAD+ within the divided population divided by the percent 7-AAD+ within the undivided population, is shown in the graph at right. Data are derived from duplicate wells in one of two experiments with similar results. Two-way ANOVA p<0.0001; Bonferroni post test ***p<0.001; **p<0.01; *p<0.05.
We would like to thank Dr. Aleksandra Kowalczyk for critically reviewing an earlier version of this manuscript.
Conceived and designed the experiments: SCJ MH EYK LZ. Performed the experiments: SCJ MH RV BJ FLZ CJ. Analyzed the data: SCJ MH OA LZ. Wrote the paper: SCJ LZ.
- 1. D'Acquisto F, Crompton T (2011) CD3(+)CD4(−)CD8(−) (double negative) T cells: Saviours or villains of the immune response? Biochem Pharmacol 82: 333–340.
- 2. Zhang ZX, Yang L, Young KJ, DuTemple B, Zhang L (2000) Identification of a previously unknown antigen-specific regulatory T cell and its mechanism of suppression. Nat Med 6: 782–789.
- 3. Zhang D, Yang W, Degauque N, Tian Y, Mikita A, et al. (2007) New differentiation pathway for double-negative regulatory T cells that regulates the magnitude of immune responses. Blood 109: 4071–4079.
- 4. Chen W, Ford MS, Young KJ, Cybulsky MI, Zhang L (2003) Role of double-negative regulatory T cells in long-term cardiac xenograft survival. J Immunol 170: 1846–1853.
- 5. Lee BP, Mansfield E, Hsieh SC, Hernandez-Boussard T, Chen W, et al. (2005) Expression profiling of murine double-negative regulatory T cells suggest mechanisms for prolonged cardiac allograft survival. J Immunol 174: 4535–4544.
- 6. Ford MS, Chen W, Wong S, Li C, Vanama R, et al. (2007) Peptide-activated double-negative T cells can prevent autoimmune type-1 diabetes development. Eur J Immunol 37: 2234–2241.
- 7. Hillhouse EE, Beauchamp C, Chabot-Roy G, Dugas V, Lesage S (2010) Interleukin-10 limits the expansion of immunoregulatory CD4−CD8− T cells in autoimmune-prone non-obese diabetic mice. Immunol Cell Biol 88: 771–780.
- 8. Young KJ, DuTemple B, Phillips MJ, Zhang L (2003) Inhibition of graft-versus-host disease by double-negative regulatory T cells. J Immunol 171: 134–141.
- 9. Fischer K, Voelkl S, Heymann J, Przybylski GK, Mondal K, et al. (2005) Isolation and characterization of human antigen-specific TCR alpha beta+ CD4(−)CD8− double-negative regulatory T cells. Blood 105: 2828–2835.
- 10. Voelkl S, Gary R, Mackensen A (2011) Characterization of the immunoregulatory function of human TCR-alphabeta+ CD4− CD8− double-negative T cells. Eur J Immunol 41: 739–748.
- 11. McIver Z, Serio B, Dunbar A, O'Keefe CL, Powers J, et al. (2008) Double-negative regulatory T cells induce allotolerance when expanded after allogeneic haematopoietic stem cell transplantation. Br J Haematol 141: 170–178.
- 12. Ford MS, Young KJ, Zhang Z, Ohashi PS, Zhang L (2002) The immune regulatory function of lymphoproliferative double negative T cells in vitro and in vivo. J Exp Med 196: 261–267.
- 13. Hamad AR, Mohamood AS, Trujillo CJ, Huang CT, Yuan E, et al. (2003) B220+ double-negative T cells suppress polyclonal T cell activation by a Fas-independent mechanism that involves inhibition of IL-2 production. J Immunol 171: 2421–2426.
- 14. Young KJ, Yang L, Phillips MJ, Zhang L (2002) Donor-lymphocyte infusion induces transplantation tolerance by activating systemic and graft-infiltrating double-negative regulatory T cells. Blood 100: 3408–3414.
- 15. Kim S, Kim KA, Hwang DY, Lee TH, Kayagaki N, et al. (2000) Inhibition of autoimmune diabetes by Fas ligand: the paradox is solved. J Immunol 164: 2931–2936.
- 16. Giese T, Davidson WF (1994) Chronic treatment of C3H-lpr/lpr and C3H-gld/gld mice with anti-CD8 monoclonal antibody prevents the accumulation of double negative T cells but not autoantibody production. J Immunol 152: 2000–2010.
- 17. Jevnikar AM, Grusby MJ, Glimcher LH (1994) Prevention of nephritis in major histocompatibility complex class II-deficient MRL-lpr mice. J Exp Med 179: 1137–1143.
- 18. Refaeli Y, Van Parijs L, Alexander SI, Abbas AK (2002) Interferon gamma is required for activation-induced death of T lymphocytes. J Exp Med 196: 999–1005.
- 19. Wang Z, Hong J, Sun W, Xu G, Li N, et al. (2006) Role of IFN-gamma in induction of Foxp3 and conversion of CD4+ CD25- T cells to CD4+ Tregs. J Clin Invest 116: 2434–2441.
- 20. Park H, Li Z, Yang XO, Chang SH, Nurieva R, et al. (2005) A distinct lineage of CD4 T cells regulates tissue inflammation by producing interleukin 17. Nat Immunol 6: 1133–1141.
- 21. Ren G, Su J, Zhao X, Zhang L, Zhang J, et al. (2008) Apoptotic cells induce immunosuppression through dendritic cells: critical roles of IFN-gamma and nitric oxide. J Immunol 181: 3277–3284.
- 22. Jung ID, Lee CM, Jeong YI, Lee JS, Park WS, et al. (2007) Differential regulation of indoleamine 2,3-dioxygenase by lipopolysaccharide and interferon gamma in murine bone marrow derived dendritic cells. FEBS Lett 581: 1449–1456.
- 23. Murray L, Martens C (1989) The abnormal T lymphocytes in lpr mice transcribe interferon-gamma and tumor necrosis factor-alpha genes spontaneously in vivo. Eur J Immunol 19: 563–565.
- 24. Hill M, Thebault P, Segovia M, Louvet C, Beriou G, et al. (2011) Cell therapy with autologous tolerogenic dendritic cells induces allograft tolerance through interferon-gamma and epstein-barr virus-induced gene 3. Am J Transplant 11: 2036–2045.
- 25. Fischer K, Voelkl S, Heymann J, Przybylski GK, Mondal K, et al. (2005) Isolation and characterization of human antigen-specific TCRalphabeta+ CD4−CD8− double-negative regulatory T cells. Blood 105: 2828–2835.
- 26. Cooke KR, Kobzik L, Martin TR, Brewer J, Delmonte J Jr, et al. (1996) An experimental model of idiopathic pneumonia syndrome after bone marrow transplantation: I. The roles of minor H antigens and endotoxin. Blood 88: 3230–3239.
- 27. Kyttaris VC, Zhang Z, Kuchroo VK, Oukka M, Tsokos GC (2010) Cutting edge: IL-23 receptor deficiency prevents the development of lupus nephritis in C57BL/6-lpr/lpr mice. J Immunol 184: 4605–4609.
- 28. Zhang Z, Kyttaris VC, Tsokos GC (2009) The role of IL-23/IL-17 axis in lupus nephritis. J Immunol 183: 3160–3169.
- 29. Blazar BR, Taylor PA, Vallera DA (1997) CD4+ and CD8+ T cells each can utilize a perforin-dependent pathway to mediate lethal graft-versus-host disease in major histocompatibility complex-disparate recipients. Transplantation 64: 571–576.
- 30. Sykes M, Szot GL, Nguyen PL, Pearson DA (1995) Interleukin-12 inhibits murine graft-versus-host disease. Blood 86: 2429–2438.
- 31. Yang YG, Dey BR, Sergio JJ, Pearson DA, Sykes M (1998) Donor-derived interferon gamma is required for inhibition of acute graft-versus-host disease by interleukin 12. J Clin Invest 102: 2126–2135.
- 32. Li X, McKinstry KK, Swain SL, Dalton DK (2007) IFN-gamma acts directly on activated CD4+ T cells during mycobacterial infection to promote apoptosis by inducing components of the intracellular apoptosis machinery and by inducing extracellular proapoptotic signals. J Immunol 179: 939–949.
- 33. Boselli D, Losana G, Bernabei P, Bosisio D, Drysdale P, et al. (2007) IFN-gamma regulates Fas ligand expression in human CD4+ T lymphocytes and controls their anti-mycobacterial cytotoxic functions. Eur J Immunol 37: 2196–2204.
- 34. Sawitzki B, Kingsley CI, Oliveira V, Karim M, Herber M, et al. (2005) IFN-gamma production by alloantigen-reactive regulatory T cells is important for their regulatory function in vivo. J Exp Med 201: 1925–1935.
- 35. Wei B, Baker S, Wieckiewicz J, Wood KJ (2010) IFN-gamma triggered STAT1-PKB/AKT signalling pathway influences the function of alloantigen reactive regulatory T cells. Am J Transplant 10: 69–80.
- 36. Bossi G, Griffiths GM (1999) Degranulation plays an essential part in regulating cell surface expression of Fas ligand in T cells and natural killer cells. Nat Med 5: 90–96.
- 37. Bradley LM, Duncan DD, Tonkonogy S, Swain SL (1991) Characterization of antigen-specific CD4+ effector T cells in vivo: immunization results in a transient population of MEL-14-, CD45RB- helper cells that secretes interleukin 2 (IL-2), IL-3, IL-4, and interferon gamma. J Exp Med 174: 547–559.
- 38. Balomenos D, Rumold R, Theofilopoulos AN (1998) Interferon-gamma is required for lupus-like disease and lymphoaccumulation in MRL-lpr mice. J Clin Invest 101: 364–371.
- 39. Haas C, Ryffel B, Le Hir M (1997) IFN-gamma is essential for the development of autoimmune glomerulonephritis in MRL/Ipr mice. J Immunol 158: 5484–5491.
- 40. Ma L, Chan KW, Trendell-Smith NJ, Wu A, Tian L, et al. (2005) Systemic autoimmune disease induced by dendritic cells that have captured necrotic but not apoptotic cells in susceptible mouse strains. Eur J Immunol 35: 3364–3375.
- 41. Kelley VE, Roths JB (1985) Interaction of mutant lpr gene with background strain influences renal disease. Clin Immunol Immunopathol 37: 220–229.
- 42. Vidal S, Kono DH, Theofilopoulos AN (1998) Loci predisposing to autoimmunity in MRL-Fas lpr and C57BL/6-Faslpr mice. J Clin Invest 101: 696–702.
- 43. Shivakumar S, Tsokos GC, Datta SK (1989) T cell receptor alpha/beta expressing double-negative (CD4−/CD8−) and CD4+ T helper cells in humans augment the production of pathogenic anti-DNA autoantibodies associated with lupus nephritis. J Immunol 143: 103–112.
- 44. Matzinger P (1991) The JAM test. A simple assay for DNA fragmentation and cell death. J Immunol Methods 145: 185–192.
- 45. Schulte M, Reiss K, Lettau M, Maretzky T, Ludwig A, et al. (2007) ADAM10 regulates FasL cell surface expression and modulates FasL-induced cytotoxicity and activation-induced cell death. Cell Death Differ 14: 1040–1049.