A Self MHC Class II -Chain Peptide Prevents Diabetes in Nonobese Diabetic Mice1 Pratibha Chaturvedi, Babita Agrawal, Marc Zechel, Edwin Lee-Chan, and Bhagirath Singh2 We explored T cell responses to the self class II MHC (I-Ag7) -chain-derived peptides in diabetic and prediabetic nonobese diabetic (NOD) mice. We found that one of these immunodominant epitopes of the -chain of I-Ag7 molecule, peptide 54 –76, could regulate autoimmunity leading to diabetes in NOD mice. T cells from prediabetic young NOD mice do not respond to the peptide 54 –76, but T cells from diabetic NOD mice proliferated in response to this peptide. T cells from older nondiabetic mice or mice protected from diabetes do not respond to this peptide, suggesting a role for peptide 54 –76-specific T cells in pathogenesis of diabetes. We show that this peptide is naturally processed and presented by the NOD APCs to self T cells. However, the peptidespecific T cells generated after immunization of young mice regulate autoimmunity in NOD mice by blocking the diabetogenic cells in adoptive transfer experiments. The NOD mice immunized with this peptide are protected from both spontaneous and cyclophosphamide-induced insulin-dependent diabetes mellitus. Immunization of young NOD mice with this peptide elicited T cell proliferation and production of Th2-type cytokines. In addition, immunization with this peptide induced peptide-specific Abs of IgG1 isotype that recognized native I-Ag7 molecule on the cell surface and inhibited the T cell proliferative responses. These results suggest that I-Ag7(54 –76) peptide-reactive T cells are involved in the pathogenesis of diabetes. However, immunization with this peptide at young age induces regulatory cells and the peptide-specific Abs that can modulate autoimmunity in NOD mice and prevent spontaneous and induced diabetes. The Journal of Immunology, 2000, 164: 6610 – 6620.
I
nsulin-dependent diabetes mellitus (IDDM)3 is a T cell-mediated autoimmune disease characterized by the destruction of insulin-producing  cells in the islets of Langerhans (1, 2). Islet  cell-specific T cells play an important role in the pathogenesis of IDDM (3). MHC class II molecules are critical self molecules playing central role in the induction and regulation of an immune response (4). They are also present in the thymus at the time of negative and positive selection, required for the acquisition of T cell repertoire during development. The class II MHC molecules have been implicated in the pathogenesis of diabetes mellitus and other autoimmune diseases (5). Susceptibility to diabetes is strongly associated with the expression of class II -chain that lacks the usual acidic aspartate residue at position 57 (6). The expression of transgenic class II -chain with aspartate at position 57, transgenic I-Ak, and I-E has been shown to protect mice from developing diabetes (7–9). Self tolerance in the immune system is essential for the self/nonself discrimination and maintenance of integrity of self. The general mechanisms proposed for achieving self tolerance are clonal deletion (10 –12) or clonal anergy (13–16) of self-reactive T cell clones. These mechanisms may operate at various stages of T cell development. The self Ags involved in Department of Microbiology and Immunology and John P. Robarts Research Institute, University of Western Ontario, London, Ontario, Canada Received for publication October 12, 1999. Accepted for publication March 31, 2000. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by grants from the Juvenile Diabetes Foundation International, the Medical Research Council of Canada, and the Canadian Diabetes Association. P.C. is recipient of a Juvenile Diabetes Foundation International Postdoctoral Fellowship. 2 Address correspondence and reprint requests to Dr. Bhagirath Singh, Department of Microbiology and Immunology, University of Western Ontario, London, Ontario, Canada N6A5C1. E-mail address:
[email protected] 3 Abbreviations used in this paper: IDDM, insulin-dependent diabetes mellitus; Cy, cyclophosphamide; NOD, nonobese diabetic mice; NOR, nonobese diabetes-resistant mice; PPD, purified protein derivative.
Copyright © 2000 by The American Association of Immunologists
tolerance induction are combinations of self peptides/MHC molecules and self proteins that require the processing and presentation in the context of self MHC molecules. Direct evidence for the existence of self peptides/self MHC molecule complexes comes from functional studies (17, 18) as well as from studies in which naturally processed peptides were acid eluted from affinity-purified class II molecules (19). Although self tolerance is necessary to prevent autoimmunity, overwhelming T cell depletion or clonal anergy has to be avoided to provide a functionally diverse T cell repertoire. Using a number of peptides from the ␣-helical and -pleated region of the Ag-binding groove of the MHC class II (I-A) molecules, we and others have examined the T cell responses to self I-A molecules (20, 21). It has also been found that exogenously added peptides of self Ags are processed into forms that are recognized by self T cells (22). MHC class II (I-Ag7)-derived synthetic peptides from NOD mice bind to syngeneic and allogeneic MHC class II molecules (23). Based on these studies, we used a number of peptides from the third hypervariable region of I-Ag7 -chain and tested the response of NOD mice toward these self MHC peptides. We found that one of the peptides corresponding to the region 54 –76 did not stimulate proliferation of T cells from young NOD mice. However, this peptide induced proliferative response in unprimed old diabetic NOD mice, suggesting a breakdown of tolerance to this self MHC peptide with age and disease status and role of peptide-specific cells in pathogenesis of diabetes. The peptide I-Ag7(54 –76) represents an immunodominant region on I-Ag7 molecule. This region has also been implicated in contacting the TCR in the recognition of MHC-peptide complex (24). We found that immunization of NOD mice at a young age with this peptide was associated with production of IgG1 Ab and Th2 responses and protected mice from diabetes. However, spontaneous breakdown of tolerance to this peptide in older mice leads to a Th1 dominant response that may contribute to the development of diabetes in NOD mice. Therefore, immunization with this self MHC peptide at young age alters the natural history of the disease, 0022-1767/00/$02.00
The Journal of Immunology
6611
Table I. Synthetic peptides used in this study Synthetic Peptide
I-A (1–14) I-Ag7(48–60) I-Ag7(54–76) I-Ag7(82–95) Proinsulin(24–36) OVA(323–339) g7
Sequence of Peptide
GDSERHFVHQFKGE RAVTELGRHSAEY GRHSAEYYNKQYLERTRAELDTA EETEVPTSLRRLEQ FFYTPKSRREVED ISQAVHAAHAEINEAGR
modulates the autoimmune responses, and prevents the development of IDDM.
Materials and Methods Mice Female NOD/Lt and NOR mice were bred in the animal facility at the John P. Robarts Research Institute and the University of Western Ontario (London, Ontario, Canada). Female BALB/c mice were purchased from The Jackson Laboratory (Bar Harbor, ME).
Antigens Peptide Ags used in this study were synthesized in this laboratory, as previously described (20), using the Merrifield solid-phase peptide synthesis technique on a ABI 431A peptide synthesizer (Applied Biosystems, Mississauga, Ontario, Canada). The crude peptides were purified by reversephase HPLC on a semipreparative synchropak RP-P C18 (250 ⫻ 10 mm ID) column using a linear gradient from 0.1% trifluoroacetic acid in water to 0.1% trifluoroacetic acid in acetonitrile (1% of the second solvent/min). Peptide purity and composition were confirmed by amino acid analysis. For functional assays, peptides were dissolved in saline by adjusting the pH with 0.1 M NaOH and were sterilized by filtration through a 0.22-m filter. The sequences of the peptides used in this study are presented in Table I. Purified protein derivative of tuberculin (PPD) was obtained from Statens Serum Institute (Copenhagen, Denmark) and Con A from Sigma (St. Louis, MO).
Immunization and T cell proliferation assay For T cell proliferation, 3- to 4-wk-old female NOD mice were immunized s.c. with 50 g peptide in 25 l saline, emulsified in 25 l CFA (Sigma) in each hind footpad. After 10 days, draining popliteal lymph nodes were collected and a single cell suspension was made. Cells (2 ⫻ 105) were then cultured in 96-well flat-bottom plates (Becton Dickinson, Bedford, MA) with various I-Ag7 -chain peptides (50 g/ml) or PPD (40 g/ml) in 200 l of RPMI (Life Technologies, Grand Island, NY) supplemented with 5 ⫻ 10⫺5 M 2-ME, 10 mM HEPES, 2 mM glutamine, 5 U/ml penicillin-streptomycin, and 10% heat-inactivated FCS (HyClone Laboratories, Logan, UT). After 3 days, cultures were pulsed with 1 Ci/well of [3H]thymidine (NEN-DuPont, Boston, MA) for 16 –20 h. Incorporation of [3H]thymidine was measured using a liquid scintillation counter (LKB Instruments, Gaithersburg, MD).
Proliferation assay using T cells from unprimed NOD mice For experiments with unprimed mice, spleens were taken out and single cell suspensions were prepared. T cells were purified using nylon wool columns. Briefly, 1 ⫻ 108 spleen cells were suspended in 1 ml RPMI (Life Technologies) supplemented with 5 ⫻ 10⫺5 M 2-ME, 10 mM HEPES, 2 mM glutamine, 5 IU/ml penicillin-streptomycin, and 10% heat-inactivated FCS (HyClone Laboratories). The cells were then loaded onto a 10-ml column containing 0.6 g of nylon wool (Robbins Scientific, Sunnyvale, CA). After 45-min incubation at 37°C in the presence of 5% CO2, the columns were washed with warm RPMI and the effluent containing T cells was collected. T cells (5 ⫻ 105) were then cultured in 96-well flat-bottom plates (Becton Dickinson, Bedford, MA) with various I-Ag7 -chain peptides (50 g/ml) or Con A (5 g/ml) in 200 l of medium. Irradiated (3000 rad) spleen cells (106 cells/well) from normal syngeneic mice (8 –9 wk old) were used as a source of APCs. After 4 days, cultures were pulsed with 1 Ci/well of [3H]thymidine (NEN-DuPont) for 16 –20 h. Incorporation of [3H]thymidine was measured using a liquid scintillation counter (LKB Instruments).
I-Ag7(54 –76) peptide-specific T cell line Female NOD mice (3– 4 wk old) were immunized with I-Ag7(54 –76) peptide (50 g/footpad) emulsified in CFA. After 10 days, popliteal lymph
nodes were harvested and T cells were purified using nylon wool. T cells (4 ⫻ 106) were cultured for 4 days with I-Ag7(54 –76) peptide (50 g/ml) in presence of irradiated spleen cells as APCs (1 ⫻ 106). Cells were washed and incubated in medium alone at 37°C for 7 days. This process was repeated twice. Cells were collected and dead cells were removed using lympholyte M (Cedarlane, Hornby, Ontario, Canada). T cells (2 ⫻ 106) were cultured with the peptide (50 g/ml) in presence of irradiated APCs. Two days later, cells were diluted in 100 ml medium and expanded in presence of 15 U/ml IL-2 (Becton Dickinson). After 10 days, cells were used for the experiments or restimulated to maintain the cell line. For control, a GAD 67-specific T cell line was used.
Cytokine assay Female NOD mice (3– 4 wk old) were immunized with I-Ag7(54 –76) (50 g/footpad) emulsified in CFA. After 10 days, popliteal lymph nodes were harvested and cells (2 ⫻ 106) were cultured in presence of I-Ag7(54 –76) peptide (50 g/ml) or PPD (40 g/ml) in 24-well plates (Becton Dickinson). Culture supernatants were collected after 24 h and assayed for the presence of IL-2, IL-4, and IFN-␥ using cytokine-specific ELISA. Briefly, ELISA plates (Becton Dickinson) were coated with 1 g/ml anti-cytokine Ab (PharMingen Canada, Mississauga, Ontario) overnight at 4°C. Plates were washed and blocked with 5% BSA for 2 h at room temperature. Supernatants (100 l) from different groups were added to the plates and incubated overnight at 4°C. Plates were washed and incubated further with 1 g/ml biotinylated anti-cytokine Ab (PharMingen Canada) for 2 h at room temperature. After washing, streptavidin-alkaline phosphatase (1:1000) was added to the wells and incubated for an additional 30-min incubation. Plates were washed and developed using p-nitrophenyl phosphate substrate (Sigma). Plates were read at 405 nm using a Bio-Rad (Richmond, CA) ELISA plate reader. Standard curves were obtained using recombinant cytokines.
Immunization of mice for Ab production Female NOD mice (3– 4 wk old) were immunized s.c. with 25 l (50 g) of I-Ag7(54 –76) peptide emulsified with equal volume of IFA in one hind footpad. Our previous studies have shown that CFA immunization protects NOD mice from diabetes (25). Therefore, IFA was used for immunization. Two weeks later, mice were reimmunized with 50 g of the same peptide in IFA i.p. Serum was collected 2 wk after the second injection and tested for the presence of peptide-specific Abs. Sera from mice immunized with saline or OVA(323–339) peptide emulsified with IFA and sera from diabetic NOD mice were used as controls.
Detection of peptide-specific Abs Peptide-specific Abs were detected using ELISA assays. Briefly, the I-Ag7(54 –76) peptide (1 g/ml) was immobilized in 96-well flat-bottom ProBind plates (Becton Dickinson) by overnight incubation at 4°C, followed by blocking at 37°C with 3% BSA (Boehringer Mannheim, Laval, Quebec, Canada) for 2 h. Serum diluted in 100 l of PBS was added to each well and incubated for 1 h at 37°C. Plates were washed, and developed using 100 l of alkaline phosphatase-conjugated goat anti-mouse IgG (Caltag Laboratories, San Francisco, CA) in the presence of p-nitrophenyl phosphate substrate (Sigma). Anti-isotype-specific Abs (IgG1) were used in an ELISA to determine the isotype of the Abs generated. Plates were read at 405 nm.
Flow-cytometric analysis Spleen cells (1 ⫻ 106) from NOD female mice (3– 4 wk old) were incubated with serum from I-Ag7(54 –76) peptide-immunized mice, washed, and stained with FITC-conjugated goat anti-mouse IgGFc (Jackson ImmunoResearch, West Grove, PA). mAb 10.2.16, which cross-reacts with I-Ag7, was used as positive control. Sera from CFA saline- or IFA salineimmunized, or diabetic NOD mice were used as controls for these studies.
Induction of diabetes in NOD mice by cyclophosphamide treatment Female NOD mice, 4 – 6 wk of age, were injected with a single dose of cyclophosphamide (Cy) (Sigma) i.p. at 200 mg/kg of body weight. Mice were then randomly divided into four groups. One group was injected with the I-Ag7(54 –76) peptide (100 g/mouse) emulsified in IFA after 3 days of Cy injection. For control, one group was injected with either saline or OVA(323–339) peptide emulsified in IFA. Mice were monitored three times per week for glycosuria and regarded as overtly diabetic based on two consecutive positive (⬎11.5 mmol) glycosuria tests.
6612
AUTOIMMUNE RESPONSE TO SELF MHC CLASS II EPITOPES
Adoptive transfer of spleen cells for disease modulation Female NOD mice (3– 4 wk old) were immunized i.p. with 100 l (100 g) of I-Ag7(54 –76) peptide emulsified with equal volume of IFA. For control, mice were immunized with saline emulsified with IFA. After 10 days, spleens were harvested and a single cell suspension was prepared. Spleen cell suspension was also prepared from spontaneously diabetic NOD mice. Splenocytes (10 ⫻ 106) from I-Ag7(54 –76) peptide or saline-immunized mice were mixed with splenocytes (10 ⫻ 106) from diabetic mice (1:1) and injected at 20 ⫻ 106 cells per mouse into 4- to 6-wk-old NOD-SCID mice. Control mice were injected with splenocytes (10 ⫻ 106) from diabetic NOD mice alone. Recipient mice were then monitored three times per week for glycosuria and regarded as overtly diabetic based on two consecutive positive (⬎11.5 mmol) glycosuria tests.
Histology Mice were sacrificed when they developed diabetes or at the end of the study. Pancreata were removed, fixed in 10% Formalin, and embedded in paraffin. Sections were cut and stained with hematoxylin-eosin. The severity of lymphocytic infiltration in islets was determined by light microscopy. Seven to eleven islets were examined for each section and scored as follows: 0 ⫽ no infiltration; 1 ⬍ 25% infiltration; 2 ⫽ 25–50% infiltration; 3 ⫽ 50 –75% infiltration; 4 ⬎ 75% infiltration.
Statistical analysis Results were analyzed using Mann-Whitney Rank Sum test or ANOVA on Ranks test, in which p value of 0.05 or less was considered significant.
Results
Response of NOD T cells to self I-Ag7-derived peptides We have reported earlier that a number of peptides from self MHC class II molecules induce proliferation of T cells (20). It has also been shown that IAg7 peptides 1–16 and 52–77 bind to self I-Ag7 molecules (23). Therefore, experiments were conducted to examine the proliferative response of T cells from NOD mice to various self I-Ag7 peptides. Splenic T cells from unprimed young NOD female (3– 4 wk) mice were cultured with various I-Ag7 peptides and proliferation was assayed. The results show that the I-Ag7 peptides corresponding to regions 1–14, 48 – 60, 54 –76, and 82–95 did not induce proliferation of T cells from young unprimed NOD mice (Fig. 1A). T cells from unprimed diabetic NOD (female, ⬎6mo-old) mice did not proliferate in response to syngeneic I-Ag7 peptides 1–14, 48 – 60, and 82–95, but there was a considerable response to peptide 54 –76 (Fig. 1B). These results suggest that the young mice are tolerant to self I-Ag7 peptides, but the tolerance to the peptide corresponding to region 54 –76 is broken at an older age. Furthermore, to find out whether the immunization with these peptides will induce a response in young NOD mice, 3- to 4-wkold female NOD mice were immunized with various peptides, and proliferation was assayed. Immunization with the peptide I-Ag7 1–14 induced a weak proliferative response, whereas I-Ag7 48 – 60 and 82–95 were nonimmunogenic. However, immunization of young NOD mice with peptide 54 –76 induced a strong proliferative response, suggesting that tolerance to this peptide can be broken at an early age by immunization (Fig. 1C). Response of T cells from nondiabetic NOD mice to I-Ag7 (54 –76) peptide
FIGURE 1. T cell proliferative response of NOD mice toward syngeneic I-Ag7 peptides. A, Response of young (3- to 4-wk-old) unprimed female NOD mice to various I-Ag7 peptides. T cells were purified from the spleens of unprimed young NOD mice and cultured in the presence of various I-Ag7 peptides. Proliferation was assayed, as described in Materials and Methods. B, Response of unprimed old (⬎6-mo-old) NOD female mice to various I-Ag7 peptides. Purified T cells from spleen were cultured with different peptides, and proliferation was assayed, as described in Materials and Methods. C, Response of young (3- to 4-wk-old) NOD female mice after immunization with I-Ag7 peptides. Young NOD female mice were immunized with 50 g of various I-Ag7 peptides emulsified in (50 l) CFA in both the hind footpads. Lymph nodes were harvested after 10 days, and single cell suspension was prepared. These cells were then cultured in the presence of the peptides. Proliferation was assayed by measuring [3H]thymidine uptake, as described in Materials and Methods. As negative control, T cells were cultured with APCs in the absence of the peptides. As a positive control, Con A (5 g/ml) was used for unprimed mice and PPD (40 g/ml) for peptide/CFA-primed mice, as described in Materials and Methods. Results represent mean of triplicate cultures ⫾ SD.
Additional experiments were done to determine the role of peptide 54 –76-induced T cells in the pathogenesis of IDDM. Proliferative response of T cells from NOD mice protected from diabetes by CFA and insulin B (9 –23) peptide treatment to I-Ag7(54 –76) peptide was measured. The results presented in Fig. 2A show that T cells from NOD mice protected from diabetes did not proliferate in response to peptide 54 –76. Response of T cells from older nondiabetic male and female NOD and female NOR mice to peptide 54 –76 was also measured. The data presented in Fig. 2B show that
T cells from these nondiabetic mice did not proliferate in response to peptide 54 –76 either. The absence of T cell response to peptide 54 –76 in nondiabetic older mice and mice protected from diabetes suggests that T cells responding to self I-Ag7(54 –76) peptide contribute to the pathogenesis of diabetes. Lack of response in NOR mice further confirmed that the T cells from nondiabetesprone mice with same MHC as NOD do not respond to this peptide.
The Journal of Immunology
FIGURE 2. A, Proliferative response of T cells from NOD mice protected from diabetes to I-Ag7(54 –76) peptide. NOD female mice (3– 4 wk old) were immunized with either CFA or insulin B (9 –23) peptide emulsified in IFA. Mice were then observed for the development of diabetes. Mice that did not develop diabetes after 30 wk of injection were sacrificed, and spleens were harvested. Primed T cells proliferation was assayed by measuring [3H]thymidine uptake, as described in Materials and Methods. Results represent mean of triplicate cultures ⫾ SD. B, Response of T cells from nondiabetic mice to I-Ag7 54 –76 peptide. Splenocytes from unprimed female NOR (12–16 wk old), male NOD (12–16 wk old), or nondiabetic female NOD (16 –20 wk old) mice were cultured with peptide 54 –76. Proliferation was assayed by measuring [3H]thymidine uptake, as described in Materials and Methods. Results represent mean of triplicate cultures ⫾ SD.
Immunization with I-Ag7(54 –76) peptide induces Th2-like response To determine the subset of Th cells generated in response to the I-Ag7(54 –76) peptide, young NOD mice were immunized with the peptide. After 10 days, draining lymph nodes were collected and single cell suspension was prepared. Cells were then cultured in the presence of peptide, and culture supernatants were tested for the presence of IL-2, IL-4, and IFN-␥. The cytokine profile of spleen cells from unimmunized diabetic mice after stimulation with this peptide was also tested. Fig. 3 shows that cells from young mice immunized with the I-Ag7(54 –76) peptide secreted IL-4, but little or no IL-2 and IFN-␥. On the other hand, cells from diabetic NOD mice cultured with this peptide secreted significant amounts of IL-2 and IFN-␥, but no IL-4. The cytokine profile of cells cultured with PPD after immunization with CFA was Th1 type with large amounts of IFN-␥ and IL-2, but no IL-4. This suggested that immunization with peptide 54 –76 at young age induces a Th2-type response in contrast to a Th1-type response generated in the case of spontaneous breakdown of tolerance with age and disease status.
6613
FIGURE 3. Cytokine profile of the T cells generated in response to I-Ag7(54 –76) peptide immunization. Young (3- to 4-wk-old) NOD mice were immunized with 50 g of the I-Ag7(54 –76) peptide emulsified in CFA in both hind footpads. After 10 days, popliteal lymph nodes were harvested and cultured in the presence of 50 g/ml of the peptide 54 –76. Culture supernatants were collected after 24 h and tested for the presence of IL-2, IL-4, and IFN-␥, as described in Materials and Methods. T cells from unprimed diabetic NOD female mice were cultured with I-Ag7(54 – 76) peptide and supernatants were collected. Supernatants were then assayed for the presence of IL-2, IL-4, and IFN-␥, as described in Materials and Methods. Results are presented as mean units of cytokine in triplicate cultures ⫾ SD.
Ab response to self I-Ag7(54 –76) peptide To investigate whether immunization with I-Ag7(54 –76) peptide induces an Ab response, mice were immunized with peptide emulsified in IFA for Ab production, and peptide-specific Abs were detected using ELISA. Results presented in Fig. 4 show that immunization with the peptide 54 –76 induced a strong peptide-specific Ab response. Furthermore, isotype of the Abs generated in response to the peptide was determined. It was also observed that immunization with the I-Ag7(54 –76) peptide induced IgG1 isotype of Abs (Fig. 4). The serum obtained from mice immunized with IFA-saline, IFA-OVA(323–339), and diabetic mice showed no reactivity to I-Ag7(54 –76) peptide in both the assays. This suggested that Abs generated were specific for the I-Ag7(54 –76) peptide.
6614
AUTOIMMUNE RESPONSE TO SELF MHC CLASS II EPITOPES periments. The results presented in Fig. 6 show that antiI-Ag7(54 –76) Ab inhibited the response of T cells to PPD Ag, while serum from IFA-saline-immunized mice did not inhibit the response. This suggests that anti I-Ag7(54 –76) Ab binds to native I-A molecules and inhibits the presentation of Ag to T cells in dose-dependent fashion. Prevention of Cy-induced diabetes by I-Ag7(54 –76) peptide Because immunization with this peptide shifted the T cell response toward Th2 type, additional experiments were done to determine its effect on the development of diabetes in NOD mice. Cy at a dose of 200 mg/kg of body weight precipitates overt diabetes in NOD mice, which is similar to spontaneous diabetes. As shown in Fig. 7, a single dose of 200 mg/kg Cy caused a rapid onset of diabetes in young NOD females, and 80% mice became diabetic after 25 days of Cy treatment. The incidence of diabetes in mice treated with Cy, followed by an injection of OVA peptide (323– 339) or IFA-saline emulsion, was similar, indicating that IFA or the control OVA peptide did not influence the onset of disease. On the other hand, none of the mice treated with Cy followed by an injection of the I-Ag7(54 –76) peptide developed diabetes (Fig. 7). Histology of the pancreas showed a significant decrease in the islet infiltration after treatment with the peptide compared with saline treatment (Table II). The infiltration seen in some islets in peptide-treated group is similar to that observed in NOD mice protected from diabetes by other treatments such as CFA (25) and probably represents immunoregulatory Th2-like cells.
FIGURE 4. Ab response of NOD mice to I-Ag7(54 –76) peptide. NOD mice were immunized with the I-Ag7(54 –76) peptide (50 g) emulsified in 50 l IFA in one hind footpad. After 2 wk, a second injection of the peptide (50 g) emulsified in IFA (50 l) was given i.p. Serum was collected after 2 wk and tested for A, peptide-specific Abs using ELISA. B, Isotype of the peptide-specific Abs, IgG1 was determined using isotypespecific ELISA, as described in Materials and Methods. As control, sera from IFA-saline-immunized, IFA-OVA(323–339) peptide-immunized, and diabetic NOD mice were used.
Recognition of I-A molecules by anti-I-Ag7(54 –76) peptide Ab on NOD spleen cells Additional experiments were done to investigate whether anti-IAg7(54 –76) peptide-specific Abs would recognize the native I-A molecules on the cell surface. It was found that the peptide-specific Abs bound to 37.9% of the NOD spleen cells, while Ab 10.2.16 stained 44.5% of the spleen cells. In contrast, serum from IFAsaline-immunized mice stained only 9.7% of the cells, and serum from diabetic NOD mice stained 12.1% cells (Fig. 5). The peptidespecific Ab did not stain the spleen cells obtained from BALB/c mice (data not shown). These results show that anti-I-Ag7(54 – 76) peptide-specific Ab can recognize the native I-A molecules on the cell surface. In vitro inhibition of Ag presentation to T cells by anti-I-Ag7(54 –76) peptide-specific Abs Additional experiments were done to find out whether antiI-Ag7(54 –76) Ab can inhibit the proliferative response of T cells to recall Ags. NOD mice were immunized with 50 l of Mycobacterium tuberculosis preparation of CFA emulsified with saline in both the hind footpads. After 10 days, popliteal lymph nodes were collected and single cell suspension was prepared. Cells were cultured with recall Ag PPD with or without anti-sera collected from I-Ag7(54 –76) peptide-immunized or IFA-saline-immunized NOD mice. Ab 10.2.16 was used as positive control in these ex-
Prevention of spontaneous diabetes by immunization with I-Ag7(54 –76) peptide Young NOD female (3– 4 wk) mice were given two injections (2 wk apart) of the peptide emulsified in IFA and observed for the development of overt diabetes. For control, mice were injected with saline or a control proinsulin B (24 –36) peptide emulsified in IFA. The proinsulin B (24 –36) peptide induces T cell responses and has been implicated in the pathogenesis of IDDM (26). This peptide bears marked similarity to GAD 65 (506 –518) peptide (26). The data presented in Fig. 8 show that 80% of mice in the IFA-saline group developed diabetes by 12–14 wk after injection, while the incidence reached 100% by 16 –18 wk after injection in mice treated with control proinsulin (24 –36) peptide. The disease incidence in the I-Ag7(54 –76) peptide-treated group was 30% by 20 wk and 40% by 40 wk after injection. The histology of pancreas from saline-treated mice showed massive infiltration of islets, but the pancreas from I-Ag7(54 –76) peptide-treated mice showed only periinsulitis (Fig. 9). Histology of the pancreas showed a significant decrease in the islet infiltration after treatment with the peptide compared with saline or proinsulin (24 –36) treatment (Table III). This suggests that islet-infiltrating cells are regulatory cells generated in response to I-Ag7(54 –76) peptide, as observed in NOD mice protected with other treatments such as CFA immunization (25). Furthermore, our data presented in Fig. 5 showed that I-Ag7(54 –76) peptide-specific Abs were detectable in these mice as late as 9 mo after peptide immunization. These Abs bind to the MHC class II molecules (Fig. 5) and inhibit the presentation of autoantigens (Fig. 6), resulting in the down-regulation of immune responses leading to protection from disease. Adoptive transfer of diabetes An adoptive transfer protocol was used to determine whether cells generated in response to I-Ag7(54 –76) peptide immunization are able to inhibit the transfer of diabetes by diabetogenic splenocytes. It was observed that cotransfer of splenocytes from I-Ag7(54 –76)
The Journal of Immunology
6615
FIGURE 5. Recognition of native I-Ag7 molecules by I-Ag7(54 –76) peptide-specific Abs. Spleen cells from normal NOD mice were stained for flow-cytometric analysis with sera from I-Ag7(54 –76) peptideimmunized mice collected at various time points. As control, sera from IFA-saline-immunized, CFA-saline-immunized, and diabetic NOD mice were used. As positive control, 10.2.16 mAb was used. Cells were analyzed by flow cytometry, as described in Materials and Methods.
peptide-immunized mice with diabetogenic splenocytes significantly delayed the onset of diabetes as compared with splenocytes from saline-injected mice. After 12 wk, all the mice in control group (diabetogenic spleen cells plus splenocytes from saline-in-
jected mice) became diabetic, whereas only 40% of mice cotransferred with peptide-specific cells plus diabetogenic splenocytes tested positive for diabetes after 12 wk of cell transfer (Fig. 10). These results suggest that cells generated in response to
FIGURE 6. Inhibition of in vitro proliferative response by I-Ag7(54 –76) peptide-specific Ab. Mice were immunized with CFA-saline emulsion and lymph nodes were harvested after 10 days. Cells were cultured with PPD (40 g/ml), as recall Ag present in CFA, in the presence or absence of serum from IFA-I-Ag7(54 –76) peptide-immunized mice, IFA-saline-immunized mice, or 10.2.16 Ab (supernatant from 10.2.16 B cell hybridoma). Proliferation was assayed by measuring [3H]thymidine uptake, as described in Materials and Methods. Results represent mean cpm of triplicate cultures ⫾ SD.
6616
AUTOIMMUNE RESPONSE TO SELF MHC CLASS II EPITOPES
FIGURE 7. Prevention of Cy-induced diabetes by I-Ag7(54 –76) peptide immunization. NOD mice were given single dose of Cy (200 mg/kg body weight) for accelerated onset of diabetes. After 3 days, a single injection of I-Ag7(54 –76) peptide (100 g) emulsified in IFA was given i.p. As control, mice injected with saline or OVA(323–339) peptide emulsified in IFA were used. Mice were then monitored three times per week for 45 days for glycosuria and regarded as overtly diabetic based on two consecutive positive (⬎11.5 mmol) glycosuria tests. Statistical analysis was performed using ANOVA on Ranks test, in which p ⬍ 0.0001 for I-Ag7(54 – 76) peptide-treated group compared with other groups. Difference between all the other groups was not significant.
I-Ag7(54 –76) peptide down-regulate the pathogenic cells and thus inhibit the induction of diabetes by the diabetogenic cells. Peptide I-Ag7(54 –76) is naturally processed and presented in vivo To determine whether this peptide fragment is naturally processed and presented in vivo, an I-Ag7(54 –76) peptide-specific T cell line was generated and used for proliferation assays. T cells from young female NOD (4 wk old) and diabetic NOD mice were also used. T cells (5 ⫻ 105) were cultured with irradiated splenocytes (1 ⫻ 106) from 12-wk-old NOD female mice, and proliferation was assayed. The data presented in Fig. 11 show that T cells from diabetic NOD mice and I-Ag7(54 –76) peptide-specific T cell line proliferated when cultured with splenocytes in absence of exogenously added peptide. The peptide-specific T cell line also proliferated in response to splenocytes from NOR mice. T cells from diabetic NOD mice also proliferated when cultured with splenocytes from NOR, NOD-SCID, and male NOD mice (data not shown). However, peptide 54 –76-specific T cells did not prolifTable II. Immunization with I-Ag7(54 –76) peptide prevents insulitis induced by Cy treatment Severity of Insulitisb (% of islets infiltrated) Groupsa
0
1
2
3
4
Cy treatment Cy followed by I-Ag7(54–76) peptide treatment
18.5 80.7
19.5 11.9
42.7 3.7
15.8 3.7
3.5 0
a NOD mice were treated with Cy (200 mg/kg of body weight) for the accelerated induction of diabetes. For prevention of disease, one group was injected with self g7 I-A (54 –76) peptide (100 g) emulsified in IFA i.p. 3 days after the Cy injection as described in Materials and Methods. b The severity of insulitis was defined as: 0 ⫽ no infiltration, 1 ⫽ ⬍25% infiltration, 2 ⫽ 25–50% infiltration, 3 ⫽ 50 –75% infiltration, and 4 ⫽ ⬎75% infiltration.
FIGURE 8. Prevention of spontaneous diabetes by immunization with I-Ag7(54 –76) peptide. Young NOD mice (3– 4 wk old) were immunized with I-Ag7(54 –76) peptide emulsified in IFA for Ab production, as described in Materials and Methods. As control, mice were immunized with saline or a proinsulin (24 –36) peptide emulsified in IFA. Mice were then monitored three times per week for glycosuria and regarded as overtly diabetic based on two consecutive positive (⬎11.5 mmol) glycosuria tests. Statistical analysis was performed using ANOVA on Ranks test, in which p ⬍ 0.0001 for I-Ag7(54 –76) peptide-treated group compared with other groups. The difference between the control groups was not significant.
erate when cultured with BALB/c splenocytes, suggesting that the response is specific. T cells from young (4-wk-old) NOD mice did not proliferate when cultured with splenocytes in absence of peptide. A GAD 67-specific T cell line did not proliferate when cultured with splenocytes from NOD mice in absence of GAD Ag, suggesting that peptide 54 –76-specific T cell line responds to peptide 54 –76 presented by APCs specifically. These results suggest that the I-Ag7(54 –76) peptide is naturally processed and presented by APCs in NOD mice. These APCs may activate peptide 54 –76-specific T cells that contribute to the pathogenesis of diabetes.
Discussion The process of thymic selection during T cell ontogeny ensures the deletion/inactivation of potentially autoreactive T lymphocytes (10, 11, 27). Therefore, it is generally expected that all potential self determinants or self peptides that bind to MHC molecules with high affinity would be nonimmunogenic due to clonal deletion or inactivation of the corresponding T cells in the periphery. However, using a foreign Ag system, it has been shown that T cells specific for minor or cryptic determinants can escape tolerance induction (28). This escape could result from ineffective in vivo processing, and failure to assemble enough complexes of the minor determinants and MHC molecules. Peptide/MHC complexes of self Ags in the thymus allow for positive and negative selection of thymocytes. However, exogenously added and internalized self proteins give rise to immune response to such peptides (22). We and others have reported (20, 21) the presence of self MHC peptide-reactive T cells in syngeneic mice. Our results further suggested that syngeneic T cells could be primed by self MHC peptides presented by APC in vivo in naive mice. In this study, we have found that a number of peptides from the self I-Ag7 did not induce T cell proliferation in young NOD mice. However, a considerable response to a self I-Ag7 peptide corresponding to region
The Journal of Immunology
6617
FIGURE 9. Histology of the pancreas from an NOD mouse treated with I-Ag7(54 –76) peptide or saline emulsified in IFA. Pancreatic tissue sections were stained with hematoxylin and eosin. Pancreatic sections from saline-treated mice show heavy infiltration and destruction of islets. Sections of pancreas from peptide-treated mice show confined periinsulitis with no destruction of islet cells.
presented by NOD APCs. T cells from 4-wk-old NOD mice did not proliferate when cultured with splenocytes from 12-wk-old NOD mice, but proliferation was observed with T cells from diabetic mice and I-Ag7(54 –76) peptide-specific T cell line. The I-Ag7(54 –76) peptide-specific T cell line also proliferated when cultured with splenocytes from NOR mice, but not in response to splenocytes from BALB/c mice. GAD 67-specific T cell line did not proliferate when cultured with splenocytes alone, suggesting
54 –76 was found in T lymphocytes isolated from old mice (⬎6 mo old). The response to this peptide was absent in T cells from old nondiabetic (male NOD, female NOD, and NOR mice) and NOD mice protected from disease using CFA and insulin B (9 –23) peptide treatment. These results suggest a possible role for peptide 54 –76-specific T cells in the pathogenesis of type I diabetes. To determine the relevance of I-Ag7(54 –76) peptide in vivo, we investigated whether this peptide is naturally processed and
Table III. Immunization with I-Ag7(54 –76) peptide prevents insulitis Severity of Insulitisb (% of islets infiltrated) Groupsa
0
1
2
3
4
IFA-saline treatment IFA-proinsulin(24–36) peptide treatment IFA-I-Ag7(54–76) peptide treatment
17.7 15.9 71.5
14.0 14.4 9.4
20.6 32.6 14.8
38.3 15.8 2.7
9.0 20.9 1.3
a NOD mice were treated with self I-Ag7(54 –76) peptide (100 g) emulsified in IFA i.p. as described in Materials and Methods. For control, one group of mice was injected with saline emulsified in IFA and another group was injected with proinsulin peptide emulsified in IFA. Mice were then observed for the development of diabetes. Mice were sacrificed when they became diabetic (control groups) or 10 mo after the injection (peptide-treated group). Pancreata were removed, fixed in 10% formalin, and embedded in paraffin. Sections were cut and stained with hematoxylin-eosin. The severity of lymphocytic infiltration in islets was determined by light microscopy. b The severity of insulitis was defined as: 0 ⫽ no infiltration, 1 ⫽ ⬍25% infiltration, 2 ⫽ 25–50% infiltration, 3 ⫽ 50 –75% infiltration, and 4 ⫽ ⬎75% infiltration.
6618
FIGURE 10. Splenocytes from I-Ag7(54 –76) peptide-immunized mice reduce adoptive transfer of diabetes by diabetogenic NOD splenocytes. NOD mice were injected with I-Ag7(54 –76) peptide or saline emulsified in IFA, i.p. After 10 days, spleens were harvested and single cell suspensions were prepared. Splenocytes (10 ⫻ 106) from spontaneously diabetic NOD mice were mixed (1:1) with (10 ⫻ 106) splenocytes from either I-Ag7(54 –76) peptide-immunized mice or saline-immunized mice. These cells (20 ⫻ 106) were then injected i.v. in NOD-SCID mice. Recipient mice were monitored three times per week for glycosuria for 12 wk and regarded as overtly diabetic based on two consecutive positive (⬎11.5 mmol) glycosuria tests. Statistical analysis was performed using MannWhitney Rank Sum test, in which p ⬍ 0.01.
that response is specific for peptide 54 –76-specific cell line. T cells from diabetic mice and peptide-specific T cell line also proliferated in response to splenocytes from older nondiabetic female NOD, NOD-SCID, and male NOD mice (data not shown). Possibly, presentation of this peptide by APCs primes the small population of T cells in young NOD mice. During the course of disease, release of cross-reactive autoantigen may cause the expansion of these peptide-specific T cells, resulting in the observed proliferation. The relevance of a T cell response to self I-Ag7(54 –76) peptide in autoimmune condition may be indirect and could be related to other genetic factors because autoimmune diabetes is a multigenic disease (33). A direct link between class II MHC molecules and autoantigens is also possible. It has been reported that NOD mice have Abs directed against a 58-kDa islet Ag identified as peripherin, which cross-reacts with MHC class II gene products (34). Possibly, during the course of the disease, peripherin is released from damaged islets and already existing small population of self I-Ag7(54 –76) peptide-reactive T cells becomes activated due to cross-reactivity and contributes to the pathogenesis of diabetes. We also found that T cells from NOD mice protected from diabetes and the nondiabetic (male NOD, female NOD, and NOR) mice did not proliferate in response to I-Ag7(54 –76) peptide. APCs from male NOD and NOR mice induced proliferation of peptide-specific T cell line and T cells from diabetic mice, suggesting that this peptide is naturally processed and presented in these mice as well. Because  cells in these mice are not destroyed, cross-reactive autoantigens are not released and peptide-specific T cell population does not expand. These results again confirm the involvement of peptide-specific cells in the pathogenesis. There have been reports indicating that self-reactive T cells are causally involved in the pathogenesis of type I diabetes (35, 36). However, all the T cells infiltrating the islets are not destructive
AUTOIMMUNE RESPONSE TO SELF MHC CLASS II EPITOPES
FIGURE 11. Peptide I-Ag7(54 –76) is naturally processed and presented in NOD mice. An I-Ag7(54 –76) peptide-specific T cell line was generated, as described in Materials and Methods. Spleens were harvested from unprimed 4-wk-old and diabetic NOD female mice, and T cells were purified, as described in Materials and Methods. T cells (5 ⫻ 105) were then cultured with irradiated (3000 rad) splenocytes from 12-wk-old NOD female mice for 72 h at 37°C in the presence or absence of any exogenous peptide. Cultures were pulsed with [3H]thymidine (1 Ci/well) for additional 16 –20 h. Cultures were harvested and proliferation was assayed by measuring the [3H]thymidine uptake, as described in Materials and Methods. Results represent triplicate cultures ⫾ SD.
(37). A number of transgenic mice in which  cells express certain cytokines constitutively show extensive lymphocytic infiltration without overt diabetes (38 – 40). These studies suggest the possibility that a proportion of T cells present within an islet downregulates the anti-self immune response, as opposed to causing its destruction. This balance in the immune response may be maintained by the ratio of Th1 to Th2 cells. Th1 cells producing IFN-␥ and IL-2 induce the cytotoxic response, and Th2 cells producing IL-4, IL-6, and IL-10 appear to promote humoral responses (41, 42). There is evidence for the importance of this balance in IDDM (1, 2, 37). Immunization of young NOD mice with the I-Ag7(54 – 76) peptide induced a Th2-like response with secretion of little or no IL-2 and IFN-␥ and large amounts of IL-4. Furthermore, peptide-specific Abs were found to be of IgG1 isotype, suggesting the generation of Th2-like response. The peptide-specific Abs recognized native I-A molecules on the cell surface and inhibited the activation of T cells. However, the spontaneous T cell response to this peptide in diabetic mice is of Th1 type. This difference in the outcome of response to the same Ag depending on the age of mice and disease status could be due to the Ag concentration and increased number of peptide 54 –76-specific cells. During the course of disease, a number of autoantigens are released after  cell destruction, and peptide-specific cells may expand due to the crossreactivity. These T cells may recognize more than one cross-reactive Ag presented by APCs shifting the response toward Th1 type due to the increased ligand density on the surface of APCs, and these cells may contribute to the pathogenesis of diabetes. However, when young mice are immunized, only peptide-specific T cell response may be induced due to the absence of cross-reactive autoantigens as most of the autoantigens are released after  cell destruction. This may also result in the decreased ligand density on the surface of APCs shifting the response toward protective Th2. Indeed, it has been reported that there is a difference in optimal ligand density on APC for Th1 and Th2 activation. High ligand
The Journal of Immunology density on APCs activates Th1 cells, while Th2 cells are activated by low ligand density on the APCs (43, 44). Thus, immunization of young NOD mice with the I-Ag7(54 – 76) peptide may prevent onset of diabetes by inducing protective Th2-like response, and peptide-specific Abs that bind to I-Ag7 and inhibit presentation of autoantigens to T cells. We used this peptide to prevent onset of diabetes in NOD mice using spontaneous as well as Cy-accelerated diabetes model. In the spontaneous model, only 40% of mice became diabetic after immunization with the I-Ag7(54 –76) peptide. The peptide-specific Ab response declined with age, although was still detectable as late as 9 mo after injection. When mice became diabetic, the Ab titer was much reduced. The peptide-specific Abs bind to I-Ag7 molecules and inhibit the presentation of autoantigens to T cells, resulting in down-regulation of pathogenic response. Possibly, to completely protect mice from diabetes, a high Ab titer is needed. Indeed, injection of anticlass II (anti-I-Ag7) Ab has been shown to block IDDM in NOD mice (45). The presence of I-Ag7(54 –76) peptide-specific Ab does not seem to compromise the health of mice, as evident from our spontaneous disease protection studies. Mice were healthy up to the age of 10 mo, when they became diabetic. A single injection of I-Ag7(54 –76) peptide 3 days after Cy treatment also protected mice from Cy-accelerated diabetes. Cy treatment increases Th1 cells, IFN-␥ production, and NO production in NOD mice (46). We postulate that regulatory cells (Th2) are induced when these mice are immunized with the I-Ag7 peptide and block the generation of Th1 cells. The periinsulitis observed in the pancreas of mice protected with I-Ag7 peptide in both Cy-accelerated and spontaneous models of diabetes may be due to the accumulation of these regulatory cells. Similar periinsulitis is observed in NOD mice that are protected from diabetes by CFA/ bacillus Calmette-Guerin treatment (25). The other possible mechanism, as discussed earlier, could be that immunization with self I-Ag7(54 –76) peptide induces peptide-specific Abs that bind to the I-A molecules on the APCs down-regulating the presentation of autoantigens. The down-regulation of Ag presentation may also result in the inhibition of the generation of new population of autoreactive T cells (Th1) shifting the response toward protective Th2-like response. Our hypothesis that the down-regulation of pathogenic Th1 cells by Th2 cells generated in response to peptide represents the mechanism of protection was confirmed by adoptive transfer experiments. We found that cotransfer of diabetogenic cells with peptide-specific cells (1:1) delayed the onset of diabetes and only 40% mice became diabetic. Possibly, increasing the ratio of peptide-specific cells to diabetogenic cells may result in complete blockade of disease transfer. The generation of an immune response to self MHC peptides has been observed in a number of systems using synthetic peptides from MHC class I and II molecules (20, 21, 47). The incomplete tolerance to self components may not be normally pathogenic. However, genetic and environmental factors or age-dependent changes may lead to an altered autoreactive T cell repertoire. Our results show that age of mice and disease status can lead to generation of responses to a self MHC class II peptide that is pathogenic. We suggest that down-regulation of anti-self responses using self I-A peptides at a younger age may be one way of modulating the immune system and preventing autoimmune diseases.
References 1. Bach, J. F. 1994. Insulin-dependent diabetes mellitus as an autoimmune disease. Endocr. Rev. 15:516. 2. Delovitch, T. L., and B. Singh. 1997. The nonobese diabetic mouse as a model of autoimmune diabetes: immune dysregulation gets the NOD. Immunity 7:727.
6619 3. Haskins, K. A., and M. McDuffie. 1990. Acceleration of diabetes in young NOD mice by CD4⫹ islet specific T cell clone. Science 249:1433. 4. Schwartz, R. H. 1985. T lymphocytes recognition of antigen in association with gene products of major histocompatibility complex. Annu. Rev. Immunol. 3:237. 5. Nepom, G. T., and H. Erlich. 1991. MHC class-II molecules and autoimmunity. Annu. Rev. Immunol. 9:493. 6. Todd, J. A., J. I. Bell, and H. O. McDevitt. 1987. HLA-DQ  gene contributes to susceptibility and resistance to insulin-dependent diabetes mellitus. Nature 329: 599. 7. Nishimoto, H., H. Kikutani, K. Yamamura, and T. Kishimoto. 1987. Prevention of autoimmune insulitis by expression of I-E molecules in NOD mice. Nature 328:432. 8. Miyazaki, T., M. Uno, M. Uehira, H. Kikutani, T. Kishimoto, M. Kimoto, H. Nishimoto, J. Miyazai, and K.-I. Yamamura. 1990. Direct evidence for the contribution of the unique I-ANOD to the development of insulitis in non-obese diabetic mice. Nature 345:722. 9. Lund, T., L. O’Reily, P. Hutchings, O. Kanagawa, E. Simpson, R. Gravely, P. Chandler, J. Dyson, J. K. Picard, and A. Edwards. 1990. Prevention of insulindependent diabetes mellitus in nonobese diabetic mice by transgenes encoding modified I-A -chain or normal I-E ␣-chain. Nature 345:727. 10. Kappler, J. W., N. Roehm, and P. Marrack. 1987. T cell tolerance by clonal elimination in the thymus. Cell 49:273. 11. Kappler, J. W., U. Staerz, J. White, and P. Marrack. 1988. Self tolerance eliminates T cells specific for Mls-modified products of the major histocompatibility complex. Nature 332:35. 12. MacDonald, H. R., R. Schneider, R. K. Lees, R. C. Howe, H. Acha-Orbea, H. Festenstein, R. M. Zinkernagel, and H. Hengartner. 1988. T-cell receptor V use predicts reactivity and tolerance to Mlsa-encoded antigens. Nature 332:40. 13. Schwartz, R. H. 1990. A cell culture model for T lymphocyte clonal anergy. Science 248:1349. 14. Ramsdell, F., T. Lantz, and B. J. Fowlkes. 1989. A nondeletional mechanism of thymic self tolerance. Science 246:1038. 15. Roberts, J. L., S. O. Sharrow, and A. Singer. 1990. Clonal deletion and clonal anergy in the thymus induced by cellular elements with different radiation sensitivities. J. Exp. Med. 171:935. 16. Mueller, D. L., M. K. Jenkins, and R. H. Schwartz. 1989. Clonal expansion versus functional clonal inactivation: a costimulatory signaling pathway determines the outcome of T cell antigen receptor occupancy. Annu. Rev. Immunol. 7:445. 17. Lorenz, R. G., and P. M. Allen. 1988. Direct evidence for functional self-protein/ Ia-molecule complexes in vivo. Proc. Natl. Acad. Sci. USA 85:5220. 18. Lorenz, R., and P. M. Allen. 1988. Processing and presentation of self proteins. Immunol. Rev. 106:115. 19. Rudensky, A. Y., P. P. Hurlburt, S. C. Hong, and C. A. Janeway Jr. 1991. Sequence analysis of peptides bound to MHC class II molecules. Nature 353:622. 20. Agrawal, B., M. Manickasundari, E. Fraga, and B. Singh. 1991. T cells that recognize peptide sequences of self MHC class II molecules exist in syngeneic mice. J. Immunol. 147:383. 21. Benichou, G., P. A. Takizawa, P. T. Ho, C. C. Killion, C. A. Olson, M. McMillan, and E. E. Sercarz. 1990. Immunogenicity and tolerogenicity of self major histocompatibility complex peptides. J. Exp. Med. 172:1341. 22. Barlow, A. K., X. He, and C. Janeway Jr. 1998. Exogenously provided peptides of a self-antigen can be processed into forms that are recognized by self-T cells. J. Exp. Med. 187:1403. 23. Feili-Hariri, M., H. Kao, T. A. Mietzner, and P. A. Morel. 1996. Functional consequences of the binding of MHC class II-derived peptides to MHC class II. Int. Immunol. 8:1857. 24. Brown, J. H., T. Jardetzky, M. A. Saper, B. Samraoui, P. J. Bjorkman, and D. C. Wiley. 1988. A hypothetical model of the foreign antigen binding site of class II histocompatibility molecules. Nature 332:845. 25. Qin, H.-Y., W. J. Sadelain, C. Hitchon, J. Lauzon, and B. Singh. 1993. Complete Freund’s adjuvant-induced T cells prevent the development and adoptive transfer of diabetes in nonobese diabetic mice. J. Immunol. 150:2072. 26. Rudy, G., N. Stone, L. C. Harrison, P. G. Colman, P. McNair, V. Brusic, M. B. French, M. C. Honeyman, B. Tait, and A. M. Lew. 1995. Similar peptides from two  cell autoantigens, proinsulin and glutamic acid decarboxylase, stimulate T cells of individuals at risk for insulin-dependent diabetes. Mol. Med. 1:625. 27. Kisielow, P., H. Bluthmann, U. D. Staerz, M. Steinmetz, and H. von Boehmer. 1988. Tolerance in T-cell-receptor transgenic mice involves deletion of nonmature CD4⫹ 8⫹ thymocytes. Nature 333:742. 28. Gammon, G., and E. E. Sercarz. 1989. How some T cells escape tolerance induction. Nature 342:183. 29. Roudier, J., A. Sette, A. Lamont, S. Albani, J. G. Karras, and D. A. Carson. 1991. Tolerance to a self peptide from the third hypervariable region of the Es  chain: implications for molecular mimicry models of autoimmune disease. Eur. J. Immunol. 21:2063. 30. Rocken, M., J. F. Urban, and E. M. Shevach. 1992. Infection breaks T-cell tolerance. Nature 359:79. 31. Ramsdell, F., and B. J. Fowlkes. 1992. Maintenance of in vivo tolerance by persistence of antigen. Science 257:1130. 32. Ohteki, T., R. Okuyama, S. Seiki, T. Abo, K. Sugiura, A. Kusumi, T. Ohmori, H. Watanabe, and K. Kumagai. 1992. Age dependent increase of extrathymic T cells in the liver and their appearance in the periphery of older mice. J. Immunol. 149:1562. 33. Wicker, L. S., J. A. Todd, and L. B. Peterson. 1995. Genetic control of autoimmune diabetes in the NOD mouse. Annu. Rev. Immunol. 13:179.
6620 34. Boitard, C., M. C. Villa, H. Becourt, P. Gia, C. Huc, P. Sempe, M. M. Portier, and J. F. Bach. 1992. Peripherin: an islet antigen that is cross reactive with nonobese diabetic mouse class II gene products. Proc. Natl. Acad. Sci. USA 89:172. 35. Bendelac, A., C. Carnaud, C. Boitard, and J. F. Bach. 1987. Syngeneic transfer of autoimmune diabetes from diabetic NOD mice to healthy neonates: requirement for both L3T4⫹ and Lyt-2⫹ T cells. J. Exp. Med. 166:823. 36. Wicker, L. S., B. J. Miller, and Y. Mullen. 1986. Transfer of autoimmune diabetes mellitus with splenocytes from nonobese diabetic (NOD) mice. Diabetes 35:855. 37. Shehadeh, N. N., F. LaRosa, and K. J. Lafferty. 1993. Altered cytokine activity in adjuvant inhibition of autoimmune diabetes. J. Autoimmun. 6:291. 38. Higuchi, Y., P. Herrera, P. Muniesa, J. Huarte, D. Belin, P. Ohashi, P. Aichele, L. Orci, J. D. Vassalli, and P. Vassalli. 1992. Expression of a tumor necrosis factor ␣ transgene in murine pancreatic -cells results in severe and permanent insulitis without evolution towards diabetes. J. Exp. Med. 176:1719. 39. Heath, W. R., J. Allison, M. W. Hoffmann, G. Schonrich, G. Hammerling, B. Arnold, and J. F. Miller. 1992. Autoimmune diabetes as a consequence of locally produced interleukin-2. Nature 359:547. 40. Picarella, D. E., A. Kratz, C. B. Li, N. H. Ruddle, and R. A. Flavell. 1993. Transgenic tumor necrosis factor (TNF)-␣ production on pancreatic islets leads to insulitis, not diabetes: distinct patterns of inflammation in TNF-␣ and TNF- transgenic mice. J. Immunol. 150:4136.
AUTOIMMUNE RESPONSE TO SELF MHC CLASS II EPITOPES 41. Ridgway, W. M., M. Fasso, A. Lanctot, C. Garvey, and C. G. Fathman. 1996. Breaking self-tolerance in nonobese diabetic mice. J. Exp. Med. 183:1657. 42. Liblau, R. S., S. M. Singer, and H. O. McDevitt. 1995. Th1 and Th2 CD4⫹ T cells in the pathogenesis of organ-specific autoimmune diseases. Immunol. Today 16:34. 43. Murray, J. S., C. Pfeiffer, J. Madri, and K. Bottomly. 1992. Major histocompatibility complex (MHC) control of CD4 T cell subset activation. II. A single peptide induces either humoral or cell-mediated response in mice of distinct MHC genotype. Eur. J. Immunol. 22:559. 44. Chaturvedi, P., Q. Yu, S. Southwood, A. Sette, and B. Singh. 1996. Peptide analogs with different affinities for MHC alter the cytokine profile of T helper cells. Int. Immunol. 8:745. 45. Singh, B., T. Dillon, E. Fraga, and J. Lauzon. 1990. Role of the first external domain of I-A  chain in immune responses and diabetes in non-obese diabetic (NOD) mice. J. Autoimmun. 3:507. 46. Rothe, H., A. Faust, U. Schade, R. Kleemann, G. Bosse, T. Hibino, S. Martin, and H. Kolb. 1994. Cyclophosphamide treatment of female non-obese diabetic mice causes enhanced expression of inducible nitric oxide synthase and interferon-␥, but not of interleukin-4. Diabetologia 37:1154. 47. Gammon, G., E. E. Sercarz, and G. Benichou. 1991. The dominant self and the cryptic self: shaping the autoreactive T-cell repertoire. Immunol. Today 12:193.
