| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Division of Immunology, Beckman Research Institute, City of Hope Duarte, CA 91010
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
and cell contact with target cells. These Tr cells can mediate their cell contact-dependent regulatory function by secreting IFN- which stimulates APCs to produce NO. NO is necessary for the Tr cells to inhibit the proliferation of pathogenic T cells and the development of diabetes. Therefore, we have identified a novel mechanism by which these Tr cells can exert their regulatory function. These results also provide an explanation as to why IFN- may play both pathogenic and immunomodulatory roles in autoimmune diseases. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
In addition to CD4+CD25+ Tr cells, our studies and those of others have demonstrated that it is possible to induce Ag-specific Tr cells (Ag-induced Tr cells) after treating the animals with Ags. Both natural and Ag-induced Tr cells are potent regulators that can effectively induce immune tolerance in animals (1, 2, 3, 4, 11, 12). Further studies have demonstrated that the in vitro regulatory function of natural Tr cells is cell contact-dependent (2, 13, 14), whereas the regulatory type 1 Tr cells (Tr1 cells) or IL-10-producing Tr cells are dependent on cytokine secretion and are independent of cell-cell contact (4, 5, 11, 12, 15). Compared with a polyclonal population of natural Tr cells, natural Tr cells stimulated by an Ag are more potent at blocking pathogenic immune responses (9, 10, 16, 17). Moreover, Ag-induced Tr cells, such as the IL-10-producing Tr cells, possess a comparable regulatory function relative to that of the natural Tr cells (18). Therefore, it is desirable to induce Ag-specific Tr cells if one wants to use these cells as potent regulators to treat autoimmune diseases.
Despite the extensive studies, the mechanisms underlying the regulatory function of various populations of Tr cells remain largely unknown. In this study, to address this question, we describe our studies on a population of potent Tr cells isolated from the diabetes-resistant nonobese resistant (NOR) mouse strain using an autoantigen-specific class II MHC I-Ag7 tetramer. The NOR mice are a MHC-matched diabetes-resistant recombinant congenic strain of NOD mice (19, 20). Both strains of mice express the disease-associated I-Ag7 molecules; however, NOR mice are resistant to diabetes which may be because these mice have a relatively robust Tr cell function (19). We have studied the NOR mouse CD4+ T cells specific for an immunodominant peptide p206–220 of a major autoantigen, glutamic acid decarboxylase (GAD), involved in type 1 diabetes. These GAD p206-specific T cells (NR206 T cells) display a Th1 phenotype and express the transcription factor T-bet, critical for Th1 cell function and development (21). However, rather than functioning as pathogenic Th1 cells, NR206 T cells function as Tr cells and can inhibit diabetes. Their regulatory function is dependent on both IFN- secretion and cell contact with APCs and the target cells. We provide evidence demonstrating that NR206 Tr cells mediate their cell contact-dependent regulatory function by secreting IFN- which stimulates APCs to produce NO. NO then suppresses the proliferation of pathogenic T cells and inhibits diabetes.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
NOR mice were purchased from The Jackson Laboratory and housed in a specific pathogen-free environment in the animal facility at the Beckman Research Institute, City of Hope. The BDC2.5 transgenic mice were a gift from Drs. D. Mathis and C. Benoist (Joslin Diabetes Center/Harvard Medical School, Boston, MA). All animals used were 7–8 wk of age.
Peptides and immunization
GAD65 p206 (TYEIAPVFVLLEYVT) and BDC2.5 T cell-stimulating peptide p79 (1040–79) (11, 22, 23) were synthesized at the Beckman Research Institute, City of Hope, and purified using reverse-phase HPLC to a purity of >90%. Animals were immunized i.p. with 100 µg of peptide emulsified in an equal volume of IFA (Sigma-Aldrich) on days 0 and 7. Spleens were removed from animals on day 14 for additional experiments.
Production of class II MHC tetramers
Production and initial use of the tetAg7/p206 and tetAg7/p79 have been previously described (11, 23, 24).
Isolation and staining of tetramer+ T cells and cytokine assays
Splenocytes isolated from immunized animals were pooled together and cultured in Click’s medium (Invitrogen Life Technologies) plus peptide for 3 days. Live cells were further incubated in RPMI 1640 medium supplemented with 5% FBS and IL-2 before being separated into tetramer+ and tetramer– cells. CD4+, tetramer+ T cells were isolated using FACS and magnetic beads (Miltenyi Biotec). Purified tetramer+ cells were maintained in complete medium (RPMI 1640 medium supplemented with 5% FBS plus IL-2) before they were used for assays. For longer cell cultures, the T cells were restimulated with peptide plus irradiated APCs (3000 rad) 2–3 wk after the last in vitro Ag stimulation.
Staining of the cells using tetramer has been previously described (11, 23, 24). Briefly, T cells were stained with PE-labeled tetramer plus the anti-TCR 
-chain Ab, H57, at 37°C for 1–2 h and analyzed by FACS using FACSCalibur (BD Biosciences). All the Abs used for FACS and for in vitro inhibition assays were purchased from BD Pharmingen.
The IL-2 bioassay has been previously described (11, 23, 24). For ELISA, cell culture supernatant was harvested after incubating cells with peptides for 24 h. Capture ELISA (assay kit for various cytokines purchased from BD Pharmingen) was used to measure the amount of cytokines produced in the cell culture according to the manufacturer’s instruction.
Western blot analysis
Cytosolic extracts were prepared by lysing cells at 4°C in lysis buffer (20 mM Tris, 1 mM EDTA, 150 mM NaCl, 5 mM iodoacetamide, 1 mM Na3VO4, 1% Triton X-100 or Brij 35, 1 mM PMSF, small peptidase inhibitors). Cell lysates were then loaded onto a 10% SDS-PAGE and separated proteins were transferred onto polyvinylidene difluoride membrane (Millipore). The membrane was blocked in PBS with 10% dry milk and 0.1% Tween 20, and incubated with primary Abs followed by a HRP-conjugated secondary Ab. The target proteins were visualized using an enhanced chemifluorescence kit (Pierce).
Adoptive transfer of T cells into NOD/scidmice
For one set of adoptive transfer experiments, 7- to 8-wk-old NOD/scid mice received a single i.v. injection of NR206 T cells, NR206 T cells plus NOD splenocytes, p206-activated NR206 T cells plus NOD splenocytes, or NOD splenocytes alone. NOD splenocytes were transferred at 1 x 107/mouse or mixed with equal numbers of NR206 T cells and then cotransferred into NOD/scid mice. The splenocytes used in the adoptive transfer experiments were isolated from nondiabetic but not from diabetic NOD mice. Recipient mice were monitored for up to 30 wk of age and were considered diabetic after 2 consecutive wk of glycosurea >2% and blood glucose level >250 µg/ml.
In a separate set of experiments, NOD/scid mice received an i.v. injection of NR206 T cells, NR206 T cells plus activated BDC2.5 splenocytes, or BDC2.5 splenocytes alone. BDC2.5 splenocytes were activated overnight with p79 (1 µg/ml) before being transferred into NOD/scid mice. In some experiments, activated BDC2.5 splenocytes were preincubated with the NO inhibitor NG-monomethyl-L-arginine (L-NMMA) or an inert control analog D-NMMA (Alexis Biochemicals) for 2 h, and were then mixed with NR206 T cells before being transferred into NOD/scid mice. Recipient mice were monitored for up to 8 wk after cell transfer and were considered diabetic after two consecutive days of glycosurea >2% and blood glucose level >250 µg/ml.
Histology was performed on pancreatic sections from diabetic and nondiabetic recipient mice. Tissue sections were stained using a standard H&E staining.
RT-PCR
For Foxp3 gene expression analysis, total RNA was prepared from cells using the RNAzol reagent (Biotecx) and first-strand cDNA synthesized using the Superscript Preamplification System (Invitrogen Life Technologies). The RT-PCR was performed using the cDNA as the template plus primers for Foxp3. The annealing PCR temperature was 60°C and the primer sequences were: 5'-ATGCCCAACCCTAGGCCAGCCAAG-3' (forward) and 5'-AGGGCAGGGATTGGAGCACTTGT-3' (reverse).
NO measurements
The NO production was measured using the Greiss assay (25). Briefly, supernatant (100 µl) from cells cultured for 24 h or 4 days were added to 100 µl of a 1:1 mixture of 1% sulfanilamide dihydrochloride in 5% H3PO4 buffer and 0.1% naphthylethylenediamine dihydrochloride (Sigma-Aldrich). The reaction was incubated at room temperature for 10 min, and the results of the reaction were measured at OD550 nm using an ELISA reader. Nitrate contents in the reaction were calculated with reference to a sodium nitrite standard curve having a detection threshold of 1 µM.
In vitro inhibition assays using cocultures
CD4+ T cells from BDC2.5 TCR transgenic mouse splenocytes were purified using magnetic beads (Miltenyi Biotec) and labeled with CFSE. Briefly, CD4+ T cells (10 x 106 cells/ml) in serum-free PBS were incubated with CFSE (0.8 µM) for 10 min at 37°C, washed, and used for further analyses.
CFSE-labeled CD4+ T cells from BDC2.5 mice (the target cells) cultured in the absence of NR206 T cells with or without p79 (1 µg/ml) were used as the controls. The CFSE-labeled cells were also cultured in the presence of NR206 T cells and p79 (1 µg/ml), with or without GAD p206. Irradiated (3000 rad) CD4+ T cell-depleted cells were used as the APCs. The cells were cultured in RPMI 1640 medium containing 5% FCS for 4 days in 96-well plates. Cells were then washed and stained with the tetAg7/p79 tetramer (the generation and use of tetAg7/p79 have been previously described; Ref.23) and an anti-CD4 Ab. The effect of NR206 T cells on BDC2.5 T cell proliferation was analyzed using FACS to determine the intensity of the CFSE in the cells and whether the cell division was altered as compared with that of the control cells.
To determine the effect of different cytokines on target cell proliferation, anti-cytokine Abs were added to the cell cultures of in vitro inhibition assays. In these assays, a saturating amount of Abs against IL-10 or IFN- (24 µg/ml) and a saturating amount of a soluble recombinant sTNFRI (20 µg/ml), which blocks the function of TNF, were added to the culture and the cells were incubated for 4 days before being harvested for analyses.
To determine the effect of NO on target cell proliferation, NO inhibitors were added to the cell cultures of the in vitro inhibition assays. The procedures were the same as described above in the in vitro inhibition assays, except that a saturating amount (0.8 mM) of L-NMMA (an active NO inhibitor) or D-NMMA (an inactive NO inhibitor analog) were added to the culture.
Transwell assays
NOD mouse spleen cells (5 x 105 cells/well) were cultured outside of a Transwell (Corning) in the lower well using a 24-well plate. NR206 T cells were cultured in the Transwell (the upper well) in the presence of irradiated CD4+ T cell-depleted APCs. All the cells were activated by PMA (100 ng/ml) and ionomycin (5000 ng/ml) in the culture. After 3 days, [3H]thymidine (1 µCi/well) was added to the lower well and the cells were harvested 24 h later. Thymidine incorporation was determined using a Wallac cell harvester.
Transwell assay was also performed using CFSE-labeled CD4+ BDC 2.5 T cells (5 x 105 cells/well) that were cultured in the lower well in the presence or absence of irradiated APCs, and NR206 T cells were cultured in the upper well. In this case, BDC2.5 T cells were activated with p79 (0.1 µg/ml). After 4 days, cells were washed and stained with tetAg7/p79 and an anti-CD4 Ab.
Statistical analyses
We have used the Student t test, Wilcoxon test, Kaplan-Meier test, and 
2 test for statistical analyses to calculate the statistical significance for differences between different experimental groups. A value of p < 0.05 was considered to be statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
We have isolated p206-specific CD4+ T cells (NR206 cells) from p206-immunized NOR mice using an I-Ag7 tetramer specific for this peptide (tetAg7/p206). Two T cell lines isolated from two independent experiments demonstrated a similar phenotype and function, and we report here the analysis of one of these two cell lines. NR206 cells were p206-specific because: 1) they were stained by tetAg7/p206 and not by a control tetramer (tetAg7/p221) specific for another GAD peptide, p221, and 2) they secreted IL-2 in response to p206 but not to an irrelevant peptide such as p79 (Fig. 1, a and b). ELISA showed that NR206 T cells secreted IFN- in response to p206, with very little IL-10, and no IL-4 in response to p206 (Fig. 1c). The amount of IFN- produced by NR206 T cells was significantly (
10-fold) lower than that produced by the BDC2.5 Th1 cells (data not shown). The NR206 cells also secreted a small amount of TNF-
(<600 pg/ml, data not shown). The intracellular cytokine staining results further showed that >60% of the cells produced IFN-, and 24% and 3% of the IFN--secreting T cells also coproduced TNF- or IL-10, respectively (data not shown). It is conceivable that the culture conditions used to expand these T cells could have modified their phenotype. Nevertheless, using the same culture condition to expand and isolate T cells, we have isolated several NOD mouse T cell lines bearing varied cytokine secretion profiles and functions (11, 12, 23). Moreover, T cells freshly isolated from p206-immunized NOR mice displayed a cytokine secretion profile similar to that of NR206 cells, and they also secreted a significant amount of IFN- but not IL-10 and IL-4 in response to p206 (Fig. 1d). Therefore, the culture conditions do not favor the selection of T cells producing IFN- or result in an alteration of the cells’ function. Altogether, these results suggest that NR206 cells have acquired their IFN--secreting phenotype in vivo before in vitro cell culture.
|
, we performed Western blot analyses to determine whether they expressed the transcription factor T-bet that which plays a critical role in the function and differentiation of Th1 cells (21). Nonactivated NR206 cells expressed T-bet at a level comparable to that detected in the control CD4+ Th1 cells (Fig. 1f). In comparison, the N206 T cells, a p206-specific T cell line isolated from NOD mice and described in our previous studies (11), did not express T-bet. Taken together, these results suggest that NR206 cells exhibit a Th1 cell phenotype. Adoptive transfer of NR206 cells could inhibit diabetes induced by pathogenic T cells
To determine the in vivo role of NR206 cells during diabetes development, we performed several adoptive transfer experiments. The results showed that transferring NR206 cells alone into NOD/scid mice did not induce diabetes in the recipients (data not shown). However, to our surprise, although all of the NOD/scid mice receiving NOD mouse splenocytes became diabetic by 25 wk of age, cotransfer of nonactivated NR206 cells with NOD splenocytes significantly delayed the onset of diabetes for 7 wk, and 60% of the recipients remained diabetes-free at the age of 30 wk (Fig. 2a). Moreover, p206-activated NR206 cells further delayed diabetes onset by an additional 5 wk when compared with that of nonactivated NR206 cells, and 67% of the animals remained diabetes-free at the age of 30 wk. Histological studies of the recipients’ pancreas showed that the diabetes-free mice receiving NR206 cells contained normal islets, suggesting that NR206 cells are able to inhibit insulitis as well (data not shown).
|
We then sought to determine whether NR206 cells could still inhibit diabetes when they were transferred into recipients at different time points. The results showed that NR206 T cells were able to inhibit insulitis and diabetes when they were transferred into recipient mice within 2 days but not 4 days after the transfer of p79-activated BDC2.5 T cells (Fig. 2c). Nevertheless, the diabetes onset was significantly delayed for 20 days when NR206 cells were transferred 4 days after the transfer of BDC2.5 T cells. Moreover, none of the mice receiving NR206 cells before the transfer of p79-activated BDC2.5 T cells developed diabetes (Fig. 2d). Altogether, these in vivo studies demonstrate that, rather than functioning as pathogenic cells and causing diabetes, NR206 cells function as Tr cells and can inhibit insulitis and diabetes.
NR206 cells are IFN- and cell contact-dependent Tr cells
We have performed in vitro experiments to determine whether NR206 cells are indeed Tr cells that can suppress the proliferation of other T cells, and to determine the mechanisms underlying their regulatory function. NR206 cells were cocultured with CFSE-labeled CD4+ BDC2.5 T cells plus p79. The initial results showed that nonactivated NR206 cells suppressed the proliferation of BDC2.5 T cells in response to p79, and activation of NR206 cells by p206 further enhances their suppressive effect (Fig. 3a). We then determined the effect of cytokine neutralizing Abs on the regulatory function of NR206 cells. The results showed that, in the presence of an anti-IFN- Ab, the percentage of CD4+ BDC2.5 T cells that underwent less than three rounds of cell division decreased from 55% to 19% (Fig. 3b). In contrast, the addition of an anti-IL10 Ab or a soluble recombinant sTNFRI protein, which can block the function of TNF-, did not restore the proliferation of BDC2.5 T cells. These results indicate that the regulatory function of NR206 cells requires IFN-.
|
in the cell culture can inhibit the proliferation of BDC2.5 T cells, we performed an "add in" experiment. In these experiments, we added various amounts of rIFN- (100–1.6 ng/ml) into the cell cultures containing CFSE-labeled BDC2.5 T cells plus APCs and p79. Our results demonstrated that the proliferation of BDC2.5 T cells was not suppressed by the addition of rIFN- (data not shown). Taken together, these results demonstrate that the regulatory function of the NR206 Tr cells is not only dependent on IFN-, but also cell contact. The mere presence of a large amount of IFN- in the assay does not result in the inhibition of the proliferation of BDC2.5 cells.
It has been shown that the transcription factor Foxp3 is necessary for the differentiation and function of the natural CD4+CD25+ Tr cells (7, 8, 28). Although the in vitro regulatory function of these Foxp3-expressing Tr cells is cell contact-dependent, their in vivo function may require the production of cytokines (1, 2, 4, 14). Because NR206 cells express CD25 and are dependent on cell contact for their regulatory function, we used RT-PCR to examine whether they also expressed Foxp3. The results showed that NR206 cells did not express Foxp3 (Fig. 3e), suggesting that NR206 cells are a distinct population from the Foxp3-positive natural CD4+CD25+ Tr cells. In addition, it has been previously shown that TGF- is an important cytokine for the regulatory function of Foxp3-expressing CD4+CD25+ Tr cells (29, 30, 31). Therefore, we have determined whether NR206 T cells produce TGF-. The results showed that NR206 T cells did not secrete TGF- nor did they express membrane-bound TGF- (data not shown). Taken together with the results that NR206 cells do not express Foxp3, these studies indicate that NR206 cells do not function through TGF-. Therefore, they represent a unique population of Tr cells that have not been previously described.
It is unclear why NR206 cells produce IFN- but function as Tr cells. One possibility is that the amount of IFN- secreted by NR206 cells (<10 ng/ml) is 10-fold less than that produced by real Th1 cells (typically >100 ng/ml). Previous reports have demonstrated the possibility that IFN- can be immunosuppressive and that lower amounts of IFN- have a potent suppressive effect on the migration of T cells (32, 33, 34, 35). To address this question and to investigate the potential in vivo mechanisms of how NR206 T cells may inhibit diabetes development in adoptive transfer experiments, we determined whether NR206 T cells could block the migration of BDC2.5 T cells into lymph nodes after they were cotransferred into recipient animals. The presence of BDC2.5 T cells in the lymph nodes of recipient mice was monitored using the tetAg7/p79 tetramer. Our results demonstrated that, compared with that of mice receiving just the BDC2.5 splenocytes, the number of BDC2.5 cells (CD4+, tetAg7/p79+ T cells) present in the lymph nodes of recipient mice was significantly reduced 6-fold at 24 h after the transfer (Fig. 3f). Furthermore, some NR206 T cells could also migrate to the lymph nodes and be detected at 24 h after the transfer (Fig. 3f). Therefore, NR206 T cells can effectively inhibit the migration of BDC2.5 T cells to lymph nodes. These in vivo studies have demonstrated that, in addition to suppressing the proliferation of other T cells as shown in the in vitro assays, inhibition of diabetogenic T cell migration could also be a mechanism of NR206 T cells to inhibit diabetes development in recipient animals. The results are also consistent with our data that there was essentially no insulitis detected in the protected mice. Altogether, although NR206 T cells produce IFN- in response to p79 stimulation, these T cells function as Tr cells but not as Th1 cells.
IFN- produced by NR206 cells correlates with the production of NO in the presence of APCs
It is not clear how the regulatory function of NR206 cells depends on not only IFN- but also cell contact. One hypothesis to explain these results is that IFN- produced by NR206 cells does not directly suppress the proliferation of pathogenic T cells and inhibit diabetes. Instead, IFN- may activate a second cell population with which NR206 cells need to interact to exert their regulatory functions. One such candidate cell population is APCs. APCs, such as dendritic cells or macrophages, may function as regulatory cells involving the production of NO (25, 36, 37, 38). It has been shown that NO plays a critical role in multiple biological or pathological conditions and can contribute to immunosuppression as a result of infection (39, 40, 41, 42).
We have performed several experiments to investigate whether the production of NO is involved in the IFN--dependent regulatory function of NR206 cells. Our initial studies demonstrated that the secretion of IFN- by NR206 cells correlated with the production of NO. The amount of IFN- and NO produced in these experiments is proportional to the amount of p206 used to stimulate NR206 cells (Fig. 4, a and b). Moreover, without being activated by p206, NR206 cells were able to produce a larger amount of IFN- with increasing amounts of IL-2 present in the cell culture (Fig. 4c). These results suggest that both Ag and IL-2 produced by other T cells can stimulate NR206 cells to secrete IFN- which may induce NO production.
|
and APCs for the production of NO. Our results showed that, when cultured alone, neither APCs nor NR206 cells produced NO (treatments 1–3, Fig. 4d). Culturing NR206 cells with APCs in the absence of p206 or IL-2 resulted in a significant increase of NO production when compared with that of culturing NR206 cells alone (treatments 3 and 4). In the presence of IL-2 without APCs, NR206 cells were able to produce a larger amount of IFN- but not NO (treatment 5). In comparison, the production of both IFN- and NO was significantly increased in the presence of APCs (treatment 6). Moreover, similar results were observed when NR206 cells were stimulated by p206 with or without IL-2 in the cell culture media (treatments 7–10). In these assays, the production of NO was significantly increased only in the presence of APCs.
We also performed additional experiments to determine whether the amount of IFN- produced in these cell culture experiments was sufficient for the production of NO in the absence of APCs. We have used an anti-TCR or an anti-CD3 Ab plus an anti-CD28 Ab to stimulate NR206 cells in the absence of APCs and p206 (treatments 11 and 13). Our results showed that, although these treatments significantly increased the amount of IFN- produced by NR206 cells, they did not result in an increased production of NO. The production of NO was significantly increased only when APCs were present in the cell culture (treatments 12 and 14). Altogether, these studies demonstrate that APCs are required for the production of NO and the amount produced is directly correlated with the production of the IFN- by NR206 cells.
Cell contact between NR206 cells and APCs is necessary for the production of NO and suppression of target cells
Because IFN- secretion correlates with NO production, we have performed additional in vitro experiments to determine whether IFN- produced by NR206 cells mediates its suppressive effect through NO. First, we determined that blocking of NO by its inhibitor L-NMMA abolished the suppressive effect of NR206 cells on the proliferation of CFSE-labeled BDC2.5 T cells. The results showed that, in response to p79, the percentage of nondividing or less divided (less than three cell divisions), BDC2.5 T cells increased from 12 to 84% in the presence of NR206 cells (Fig. 5a). The addition of either an anti-IFN- Ab or L-NMMA, but not an inactive analog D-NMMA, restored the proliferation of BDC2.5 T cells (Fig. 5a). Therefore, suppression of the proliferation of BDC2.5 T cells by NR206 cells is also dependent on NO.
|
and NO. These results showed that, although BDC2.5 T cells produced a large amount of IFN- in response to p79 plus APCs, their proliferation was not suppressed and there was no NO production induced in these cultures (Fig. 5, a and b). In comparison, the presence of NR206 cells induced NO production in cell cultures (Fig. 5b), which suppressed the proliferation of BDC2.5 T cells (Fig. 5a). Moreover, the addition of an anti-IFN- Ab reduced the production of both IFN- and NO, whereas the addition of L-NMMA only reduced the production of NO but not IFN- (Fig. 5b). As a control, the inactive analog D-NMMA had no effect on the production of both IFN- and NO. These studies demonstrate that the regulatory function of NR206 cells is mediated first by the production of IFN-, which then stimulate APCs in the presence of NR206 to produce NO, leading to the suppression of the target T cells’ proliferation.
Moreover, we wanted to determine why NR206 cells are dependent on not only the production of IFN- and NO but also cell contact with APCs or target T cells for exerting their regulatory function. We have repeated the transwell assays as described in Fig. 3, and measured the amount of IFN- and NO produced by cells following various treatments. IFN- was detected in the supernatant of BDC2.5 T cells stimulated with p79 plus APCs, irrespective of the presence of NR206 cells (treatments 2–5, Fig. 5c). However, when NR206 cells were separated from BDC2.5 T cells and APCs, the production of NO was not increased in all of the supernatants tested (treatments 4–5, Fig. 5c). Proliferation of BDC2.5 T cells was also not suppressed in these cell cultures (treatments 4–5). Therefore, it appears that cell contact between NR206 cells and APCs is necessary for NO production and suppression of the proliferation of target cells.
NO is required for NR206 T cells to inhibit diabetes
To examine whether NO is indeed an essential component of the regulatory function of NR206 cells for diabetes inhibition, we performed additional adoptive transfer experiments. In these experiments, we adoptively transferred nonactivated BDC2.5 mouse splenocytes or p79-activated BDC2.5 mouse splenocytes, either alone or together with NR206 cells, into NOD/scid mice. To evaluate the role of NO in diabetes development, some of the recipient mice received p79-activated BDC2.5 T cells preincubated with L-NMMA or D-NMMA for 2 h. The BDC2.5 T cells were part of unseparated splenocytes and APCs were present. These preincubated cells were then mixed with NR206 T cells before their transfer into NOD/scid mice. The results confirmed that cotransfer of NR206 T cells was able to inhibit diabetes induced by activated BDC2.5 T cells. However, preincubation of these BDC2.5 cells with L-NMMA but not D-NMMA was sufficient to abolish the protective effect of NR206 cells on diabetes. Although the onset of diabetes was significantly delayed in mice receiving L-NMMA-treated BDC2.5 cells plus NR206 cells, all of the recipient mice became diabetic within 25 days after the cell transfer (Fig. 6). These results demonstrate that blockade of NO production by L-NMMA was sufficient to prevent the suppression of BDC2.5 T cells by NR206 T cells and was able to restore the function of BDC2.5 T cells to induce diabetes in recipient animals. Therefore, these results support the conclusion that NO production by APCs is required for NR206 cells to inhibit diabetes development.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
and cell contact with their target cells. Our study also provided evidence that NR206 T cells have acquired their IFN--secreting phenotype in vivo before in vitro cell culture. Although their cell contact dependency and expression of CD25 are similar to the natural CD4+CD25+ Tr cells, NR206 Tr cells do not express Foxp3. Furthermore, the regulatory function of NR206 cells also requires APCs and the production of NO. Interestingly, both Ag and IL-2 produced by other activated T cells can stimulate NR206 cells to secrete IFN- and induce NO production. Moreover, separating NR206 Tr cells from APCs and target T cells abolished both NO production and suppression of target cells, regardless of whether a large amount of IFN- is produced by these cells. These results suggest that the mere production of IFN- by activated pathogenic Th1 cells alone is not sufficient to induce NO production and, instead, may lead to disease progression. Only the presence of NR206 Tr cells and their contact with APCs leads to NO production and to suppression of pathogenic T cells. More importantly, the in vivo inhibitory effect of NR206 Tr cells on diabetes also is NO-dependent. Therefore, these studies have identified a novel regulatory mechanism used by Tr cells through the secretion of IFN-, which then stimulates APCs to produce NO that suppresses the proliferation of pathogenic T cells and inhibits autoimmune disease.
The GAD p206 peptide has been demonstrated as a major immunogenic peptide in NOD mice (43). Our previous studies have shown that p206 can induce IL-10-depdendent Tr cells from NOD mice (11). In comparison, p206 induces IFN--, NO-, and cell contact-dependent Tr cells from NOR mice. Therefore, p206 can induce Tr cells in both NOD and NOR mice, suggesting that this GAD peptide preferentially selects for Tr cells. It is unclear why Tr cells with the same Ag specificity but isolated from different mouse strains are dependent on distinctive suppressive mechanisms. NOR mice are an I-Ag7-bearing recombinant congenic strain of NOD mice, but NOR mice do not develop diabetes (19, 20). One explanation as to why p206-specific Tr cells isolated from these two strains of mice are dependent on different regulatory mechanisms may be due to the difference in their non-class II MHC genetic background. It is possible that these non-class II MHC background genes may influence the selection of Tr cells with varied underlying regulatory mechanisms. In our unpublished studies, we have isolated GAD p286-specific T cells from NOR mice. The results obtained from these studies have shown that the p286-specific T cells also function as Tr cells. Moreover, in our previous studies, we have used the same protocol to isolate from NOD mice the T cells specific for an unrelated peptide (a mimetic epitope, p79, that is highly active in stimulating BDC2.5 T cells). These T cells do not function as Tr cells that can inhibit diabetes (23). These results demonstrate that the behavior of NR206 is unique compared with those previously described T cells. Altogether, these results suggest that T cells specific for GAD peptides such as p206 are biased toward Tr cells and that the induction of Tr cells is a GAD-dependent phenomenon. A poorly understood area in studies of Tr cells is the role of Ag specificity in the development and function of Tr cells. The results of our studies demonstrate that it is possible to use an autoantigenic peptide to induce potent disease-inhibiting Tr cells from both normal and disease-prone mouse strains.
Another interesting finding of our study is that there is a window of opportunity for these Tr cells to function as an immune modulator and effectively inhibit diabetes development. In vivo transfer of NR206 Tr cells 2 days, rather than 4 days, after the transfer of activated BDC2.5 T cells protected the recipient animals from insulitis and diabetes. Nevertheless, NR206 Tr cells were still able to significantly delay diabetes onset by 20 days if they were transferred at 4 days. In comparison, transferring NR206 Tr cells before the transfer of activated BDC2.5 T cells completely inhibited insulitis and diabetes. These results suggest that, in this accelerated model of diabetes development, it takes at least 2 days for a sufficient number of the transferred BDC2.5 T cells to reach and reside in the islets. The presence of NR206 Tr cells during this 2-day time period can effectively block the proliferation and/or migration of pathogenic T cells into the islets and inhibit diabetes. A delay in the appearance of NR206 Tr cells allows the infiltration of a small number of BDC2.5 T cells into the islets, and it may take up to an additional 20 days for these infiltrated pathogenic T cells to expand and destroy a majority of the islets leading to diabetes. Previous studies have suggested that accumulation of pathogenic T cells in the pancreatic draining lymph nodes occurs before their migration into the islets (44). Our studies have demonstrated that NR206 T cells can inhibit the migration of BDC2.5 T cells to lymph nodes. These results show that the interaction among Tr cells, APCs, and pathogenic T cells, and/or the NO production may occur before the migration of these cells into secondary lymph nodes, although it is also likely that such an interaction occurs later in pancreatic lymph nodes or other organs before the infiltration of pathogenic T cells into the islets. Overall, these results suggest that the use of Tr cells, such as the NR206 T cells, can be a useful therapeutic method to effectively delay or inhibit autoimmune diseases if the animals or patients are treated at the right time, even after the appearance in the body of a large number of activated pathogenic T cells.
The role of IFN- in the immune response is complicated (32). Its pathogenic role as a Th1 cytokine in the development of autoimmune diseases has been well documented (21, 32, 45, 46, 47). However, it can also down-regulate immune responses and contribute to immune regulation processes (32, 33, 34). Moreover, our results have demonstrated that IFN- can function as a regulatory cytokine by acting on APCs. The reason why IFN- plays these seemingly opposite roles in an immune response is unclear. In this regard, our studies may provide a unifying view explaining why IFN- may function both as a pathogenic and as an immunosuppressive factor. Based on our current results, we hypothesize that IFN- plays a pathogenic role in the absence of Tr cells or their interaction with APCs. In contrast, in the presence of Tr cells such as the NR206 Tr cells, IFN- stimulates APCs to produce NO in a cell contact-dependent manner between Tr cells and APCs. NO then serves as the effector molecule that suppresses pathogenic T cells and regulates autoimmune diseases. Therefore, the recruitment of autoantigen-specific Tr cells, such as NR206 Tr cells, into an inflamed site with IFN- production may facilitate the interaction between Tr cells and APCs. This interaction then leads to the production of NO by APCs and results in the suppression of (auto)immune responses.
Although NO functions as an important molecule in various biological processes (39, 41, 42), its role in immune regulation is not fully understood. For example, NO can function as a cytotoxic molecule and high levels of NO may be detrimental to animals and contribute to autoimmune diseases (48, 49), yet it can be critical for the protection of hosts to infections (40, 41). In contrast, autoimmune diseases are exacerbated in the absence of NO production in mice with deficient inducible NO synthase, suggesting an immune regulatory role of NO (50, 51). In support of these observations, previous studies have demonstrated that NO, which may be produced by macrophages or dendritic cells, can inhibit the proliferation of T cells (25, 52, 53). In addition to showing the immunosuppressive effect of NO, our results have further demonstrated that NO can be induced in cell cultures only in the presence of NR206 Tr cells. A likely explanation for this observation is that the interaction between ligands expressed on NR206 Tr cells and receptors expressed on APCs are essential for NO production by APCs. Production of IFN- by NR206 Tr cells is required but not sufficient to induce NO production by APCs. Future studies are necessary to fully understand the role of NO during the control of autoimmunity and to investigate the mechanisms underlying the regulatory role of NO.
Based on these studies, we propose a model to explain the interaction among NR206 Tr cells, APCs, and pathogenic T cells such as the BDC2.5 cells (Fig. 7). This model predicts that, in addition to being activated by their antigenic peptides presented by APCs, Tr cells, like the NR206 T cells, can be activated by IL-2 produced by pathogenic T cells in response to their cognate Ags. The activated Tr cells exert their regulatory function indirectly through the interaction with a bystander population of APCs which are activated by IFN-. The activated APCs then secrete NO that can directly suppress the proliferation of the target (pathogenic) T cell population resulting in disease inhibition. This model suggests that, in addition to the secretion of cytokines, an unknown specific cell-cell interaction mechanism is required for triggering the bystander APCs to initiate the subsequent downstream effect of suppression through the production of NO. The interaction between Tr cells with APCs is necessary for NO production that suppresses pathogenic T cells and autoimmune diseases. This model further suggests that, while the in vitro regulatory function of these Tr cells are Ag-nonspecific, activation of Tr cells by IL-2 resulting from an Ag-specific T cell response and the need of specific interaction between Tr cells and APCs provides a unique controlling mechanism. This mechanism may ensure that an immune suppression occurs only in the presence of Tr cells, following a pathogenic Th1 response that produces a large quantity of IL-2 and IFN-.
|
|
Acknowledgments |
|---|
|
Disclosures |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
Footnotes |
|---|
1 This work was supported in part by National Institutes of Health grants (to C.-P. L.). 
2 Address correspondence and reprint requests to Dr. Chih-Pin Liu, Division of Immunology, Beckman Research Institute, City of Hope, 1450 East Duarte Road, Duarte, CA 91010-3000. E-mail address: cliu{at}coh.org
3 Abbreviations used in this paper: Tr cell, regulatory T cell; NOR, nonobese resistant; GAD, glutamic acid decarboxylase.
Received for publication June 10, 2005. Accepted for publication January 13, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
. J. Exp. Med. 194: 629-644. 1 plays an important role in the mechanism of CD4+CD25+ regulatory T cell activity in both humans and mice. J. Immunol. 172: 834-842. 
-chain gene knockout (ZKO) mice. Th1 condition medium contains anti-mouse IL-4 (5 µg/ml), rIL-12 (5 ng/ml; R&D Systems), and rIL-2 (10 U/ml). We have previously demonstrated that ZKO mouse T cells are biased IFN-
