The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Quinn, A.
Right arrow Articles by Sercarz, E. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Quinn, A.
Right arrow Articles by Sercarz, E. E.
The Journal of Immunology, 2001, 167: 1748-1757.
Copyright © 2001 by The American Association of Immunologists

MHC Class I-Restricted Determinants on the Glutamic Acid Decarboxylase 65 Molecule Induce Spontaneous CTL Activity1

Anthony Quinn*, Marcia F. McInerney{dagger} and Eli E. Sercarz2,*

* Division of Immune Regulation, La Jolla Institute for Allergy and Immunology, San Diego, CA 92121; and {dagger} University of Toledo, College of Pharmacy, Toledo, OH 43606


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
CD4+ T cell responses to glutamic acid decarboxylase (GAD65) spontaneously arise in nonobese diabetic (NOD) mice before the onset of insulin-dependent diabetes mellitus (IDDM) and may be critical to the pathogenic process. However, since both CD4+ and CD8+ T cells are involved in autoimmune diabetes, we sought to determine whether GAD65-specific CD8+ T cells were also present in prediabetic NOD mice and contribute to IDDM. To refine the analysis, putative Kd-binding determinants that were proximal to previously described dominant Th determinants (206–220 and 524–543) were examined for their ability to elicit cytolytic activity in young NOD mice. Naive NOD spleen cells stimulated with GAD65 peptides 206–214 (p206) and 546–554 (p546) produced IFN-{gamma} and showed Ag-specific CTL responses against targets pulsed with homologous peptide. Conversely, several GAD peptides distal to the Th determinants, and control Kd-binding peptides did not induce similar responses. Spontaneous CTL responses to p206 and p546 were mediated by CD8+ T cells that are capable of lysing GAD65-expressing target cells, and p546-specific T cells transferred insulitis to NOD.scid mice. Young NOD mice pretreated with p206 and p546 showed reduced CTL responses to homologous peptides and a delay in the onset of IDDM. Thus, MHC class I-restricted responses to GAD65 may provide an inflammatory focus for the generation of islet-specific pathogenesis and {beta} cell destruction. This report reveals a potential therapeutic role for MHC class I-restricted peptides in treating autoimmune disease and revisits the notion that the CD4- and CD8-inducing determinants on some molecules may benefit from a proximal relationship.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Insulin-dependent diabetes mellitus (IDDM)3 in the nonobese diabetic (NOD) mouse is a T cell-mediated autoimmune disease that serves as a model for human type I diabetes. Female NOD mice spontaneously develop insulitis at 4–5 wk of age and 70–90% go on to develop diabetes later in life. Before the onset of disease, young NOD mice develop immune responses to several islet cell Ags, including glutamic acid decarboxylase (GAD65) (1, 2). The early CD4+ T cell response to GAD65 is directed against two specific regions of the molecule (aa 509–528 and 524–543), while later responses encompass other determinants on the molecule (1). In addition, a dominant Th response to the 206–220 region of GAD65 has been described in immunized NOD mice (3, 4). CD4+ T cell responses to GAD65 have been the focus of a number of studies (5, 6) and are clearly important in the pathogenesis of IDDM (7); however, both CD4+ and CD8+ T cells are required for disease progression in NOD mice (8). NOD mice deficient in MHC class I or {beta}2-microglobulin expression are protected from both IDDM and insulitis (9, 10), and adoptive transfer experiments using T cell clones or spleen cells from diabetic mice have clearly demonstrated the necessity for both MHC class I- and class II-restricted responses in the NOD mouse model of spontaneous autoimmune diabetes (8, 11). Moreover, recent findings suggest that CD8+ T cells are not only necessary for the progression of insulitis and IDDM, but that MHC class I-restricted T cell responses may be the early initiators of islet {beta} cell damage (12). Given the prominence of GAD65 as an autoantigen in IDDM (13), and its potential role in immunotherapy, there is a need to investigate the activities of GAD-specific CD8+ T cells in IDDM.

To identify relevant CD8+ T cell-inducing determinants on an Ag, one of two approaches has usually been taken. The pepscan, which utilizes overlapping synthetic peptides representing the entire sequence of the Ag, can be used, but is often cost prohibitive with large proteins such as GAD65. Alternatively, Ag-specific T cell clones can be recovered from responding animals, cultured in vitro, and characterized to delineate the cognate determinant (14). However, because of the recent accumulation of data regarding the binding preferences of MHC class I molecules, putative determinants can now be identified based on the general peptide-binding motifs for a given MHC haplotype (15). Allele-specific preferences for certain sequence motifs can provide hints to reveal strong MHC-binding peptides and significantly reduce the number of synthetic peptides that may be required to study CD8+ T cell responses. In this study, we have used information regarding the Kd allele-specific binding motif (16, 17) to identify potential CD8+ T cell-inducing determinants on GAD65. To further refine the search, we took into consideration the findings of previous reports from this lab, wherein the dominant CD8-inducing and CD4-inducing specificities on {beta}-galactosidase were defined (18, 19) as well as the nature of the interactions of cells recognizing each determinant. Those reports suggested a proximity relationship between CD4-inducing determinants and CD8-inducing determinants such that the most important feature of the relationship between the two T cell types was the proximal orientation of the distinct determinants that they recognized. More recently, in a study of the regulatory response to aggressive autoimmunity in the experimental autoimmune encephalomyelitis model of multiple sclerosis, the importance of cooperativity between TCR peptide-specific CD4+ and CD8+ T cells was revealed (20). Once more, the determinants recognized by these two T cell subsets were relatively close to each other on the antigenic V{beta}8.2 TCR chain. In this study, we show that significant CTL responses to two Kd-restricted GAD determinants, 206–214 (p206) and 546–554 (p546), arise spontaneously in NOD mice. These determinants are proximal to previously described Th-inducing determinants (206–220 and 524–543), and their activities are related to the development of IDDM. Furthermore, treatments that lead to a reduction in these CTL responses also provide protection from cyclophosphamide-induced IDDM.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Animals

NOD mice were purchased from Taconic Farms (Germantown, NY) and bred in our facilities (specific pathogen free). To demonstrate that the observed CTL responses were not a colony-specific phenomenon, NOD mice were also purchased from The Jackson Laboratory (Bar Harbor, ME), as were NOR.Lt, BALB/c, and NOD.scid mice. NOD.scid mice were bred at the University of Toledo (Toledo, OH).

Peptides

GAD peptides p10, p48, p125, p206, p436, p505, and p546 (all 9-mers with the N-terminal amino acid designated), and control peptides hen egg white lysozyme (HEL) 91 and HEL 116 (HEL 91–99 and 116–124) were selected for analysis based on their expression of the Kd-binding motif (16). p206, p505, and p546 are proximal to the previously described I-Ag7-restricted GAD65 determinants, 509–527, 524–543, and 206–220. HEL 91 and HEL 116 are adjacent to the I-Ag7-restricted determinant HEL 91–105 and the I-Ed-restricted determinant 106–116. The Kd-restricted peptide nucleoprotein (NP) 147 (influenza nucleoprotein 147–155) (21) and the Ld-restricted peptide NP118 (lymphocytic choriomeningitis virus nucleoprotein 118–126) (22) have been previously described. The peptides were synthesized at the University of California Peptide Synthesis Laboratory (Los Angeles, CA) on an Advanced Chemtech 395 synthesizer using f-moc chemistry and purified using HPLC. Peptide purity was determined by capillary electrophoresis, and amino acid composition was verified by mass spectrometry.

CD8+ CTL responses

Single-cell suspensions were prepared from mouse spleens and were depleted of RBCs using RBC-lysing buffer (Sigma, St. Louis, MO). After washing twice with DMEM, the cells were resuspended to 1.6 x 107 cells/ml in DMEM plus 1% FBS. Peptide(s) was then added to a final concentration of 5 µg/ml, and the suspension was plated at 0.5 ml/well in 24-well culture plates and incubated overnight at 37°C in 7% CO2. The next day, 1 ml of rIL-2 medium (10 U/ml human rIL-2, Cetus, Emeryville, CA, in DMEM with 10% FBS) was added to each well. On day 5, the effectors were washed, counted, and used in a typical chromium release assay. Briefly, target cells were pulsed with 200 µCi of 51Cr (ICN Pharmaceuticals, Irvine, CA), with or without 5 µg/ml cognate peptide, for 90 min at 37°C. The target cells were then washed three times, and plated at 5 x 103 cells/well in 96-well plates. The CTL effectors were then added to the plates at 5 x 105–5 x 103 cells/well in triplicate wells, centrifuged, and incubated for 4 h at 37°C. Fifty microliters of culture supernatant was collected from each well, and the radioactivity was measured in a Trilux liquid scintillation counter (Wallac, Gaithersburg, MD). Specific cytotoxicity was calculated according to the formula: (experimental cpm - spontaneous release cpm/maximum cpm - spontaneous release cpm) x 100. P815 (Kd, Dd), EL-4 (Kb, Db) (American Type Culture Collection, Manassas, VA), and MHC class II-negative M12.C3 (Kd, Dd) (obtained from L. Glimcher, Harvard University, Boston, MA) (23) cell lines served as targets for the CTL assay. A line that expressed GAD65 (M12.C3.GAD) was produced by transfecting M12.C3 cells using electroporation (Bio-Rad, Hercules, CA) and the vector pRC/RSV (Invitrogen, Carlsbad, CA) containing the full-length rat GAD65 gene (obtained as a gift from M. Solimena, Yale University, New Haven, CT) (24). A line transfected with the pRC/RSV vector alone (M12.C3.RSV) was used as a control. GAD65 expression in the transfected line was confirmed by immunofluorescent staining using the GAD65-reactive mAb GAD6 (25).

Immunization

To establish the immunogenicity of the peptides used in this study, mice were immunized with 10 µg of peptides plus 30 µg of Th-inducing peptide (lysozyme 11–25) (26) emulsified in IFA. Fourteen days later, the spleens were processed and tested for CTL activity and proliferative responses. Spontaneous and induced splenic proliferative responses were determined as previously described (1, 27).

ELISPOT assay

ELISPOT plates (Millititer HA plates; Millipore, Bedford, MA) were coated overnight at 4°C with 2 µg/ml anti-IFN-{gamma} (BD PharMingen, La Jolla, CA). Nonspecific binding sites were blocked by incubation with 10% FBS-DMEM. Target cells, pulsed with or without peptides (1 µg/ml), were washed and seeded in ELISPOT plates at 1 x 105 cells/well. The effectors were then added at various concentrations (0.6–25 x 105 cells/well) and then incubated for 24 h at 37°C. After extensive washing with 0.05% Tween 20-PBS, biotin-conjugated anti-IFN-{gamma} (2 µg/ml; BD PharMingen) was added and incubated overnight, followed by the addition of avidin D-peroxidase (Vector Laboratories, Burlingame, CA) for 1 h at room temperature. The spots were visualized by the addition of 3-amino-9-ethylcarbazole (Sigma) substrate and enumerated with a digital image processing and analysis program (NIH Image).

Nasal treatment with peptides

Seven-day-old NOD mice were nasally treated with 30 µg of GAD peptide, as a pool (15 µg each of p206 and p546) or individually (p546), on days 7, 9, and 11. In experiments with cyclophosphamide (CY)-induced IDDM, the mice received an additional nasal dose on day 15. To check for tolerance induction, some nasally treated mice were challenged by in vivo immunization with the cognate peptide emulsified in IFA.

Cyclophosphamide-induced IDDM

To accelerate the onset of diabetes, NOD mice were treated with cyclophosphamide as previously described (28). Briefly, 10–13-wk-old mice were given a single i.p. injection of cyclophosphamide (Sigma), 200 mg/kg. The incidence of IDDM in mice was determined by daily urine analysis (Chemstrip µG; Boehringer Mannheim, Indianapolis, IN) for 3 wk, with those mice testing positive for glucosuria being confirmed by measurements of blood glucose using an Encore glucometer (Bayer, Elkhart, IN). Those mice with blood glucose >250 mg/dl were considered diabetic.

Adoptive transfer of GAD-reactive T cells

GAD 546-reactive T cells from two T cell lines (546I and 546L) were labeled with octadecyl indocarbocyanine (DiI) according to the manufacturer instructions (Molecular Probes, Eugene, OR). Briefly, a stock solution of DiI was made by dissolving the dye into DMSO (3 mg/ml). The T cells were then incubated in a dilution of the stock solution (1/1000 in PBS), 1 x 106 cells/ml, for 5 min at 37°C, followed by 15 min on ice. Each cell line was washed twice before being injected i.v. into two female NOD.scid mice (3 x 106 cells/mouse). Before transfer, the lines were checked for labeling using a fluorescence microscope. Forty-eight to 72 h after transfer, the pancreas, spleen, liver, and a kidney were collected from each animal and snap frozen in OCT medium with liquid nitrogen. Frozen tissue samples were cut into 8-µm cryosections, examined, and photographed. Fluorescent images were captured by a Laser Scanning Confocal Imaging System (Bio-Rad).

Flow cytometry analysis

For analysis of CD8, CD4, and TCR expression, T cell lines were stained with FITC-conjugated Abs (BD PharMingen) specific for CD3, CD4, CD8, or {alpha}{beta}TcR. Stained cells were analyzed on a FACScan flow cytometer (BD Biosciences, Mountain View, CA) using CellQuest software (BD Biosciences).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Spleen cells from naive NOD mice show MHC class I-restricted responses to GAD peptides p546 and p206

The evidence to date suggests that Kd-restricted, rather than Db-restricted, CTL responses are involved in {beta} cell destruction (29, 30). To identify potential CD8+ CTL determinants in the mouse GAD65 molecule, we examined its reported amino acid sequence (1) and identified sequences that contained the Kd-binding motif (17) (Table IGo). Because earlier findings suggested that CD8-inducing determinants may be located near CD4-inducing determinants (18), GAD65 peptides p206, p505, and p546 were synthesized. These peptide sequences are either juxtaposed to a GAD65 sequence previously shown to be an early inducer of Th responses in naive NOD mice (509–527 and 524–543) (1, 5) or a somewhat proximal sequence (252–269). Interestingly, p206 is contained within the 206–220 Th-inducing determinant (3, 4). Four other GAD65 peptides were also synthesized, p10, p48, p125, and p436, that express the appropriate motif, but their sequences are distal (>50 aa) to the aforementioned Th determinants. Peptides HEL 91 and HEL 116 express a Kd-binding motif and either overlap or are proximal to the I-Ag7- and I-Ed-restricted determinants 91–105 and 106–116, respectively (26).


View this table:
[in this window]
[in a new window]
  		Table I. Mouse GAD65 and lysozyme sequences that display the Kd-binding motif

 
To determine whether MHC class I-restricted responses to GAD65 were spontaneously primed in NOD mice, spleen cells from young naive mice were cultured with GAD65 or control peptides and then tested for cytokine production using the ELISPOT assay. GAD65 peptides p206 and p546 were able to incite Ag-specific IFN-{gamma} production, as demonstrated by the numerous IFN-{gamma} spot-forming cells seen upon restimulation of NOD effectors (Fig. 1Go, A and B). These IFN-{gamma}-producing cells were easily detectable at a number of plating dilutions (Fig. 1GoB) and were reproducibly found to number >100 per 105 cultured cells plated. Conversely, GAD65 peptides p10, p48, p125, p436, and p505 and the control peptides HEL 91, HEL 116, and NP147 did not induce responses above background levels (Fig. 1Go, B and C), even when high cell numbers were used in the ELISPOT (Fig. 1GoC). None of the peptides was able to induce IL-4 or IL-5 spot-forming cells (data not shown), although spleen cells cultured with anti-CD3 were positive in all cytokine assays (data not shown). M12.C3 cells, which express no MHC class II molecules, served as the APC in the IFN-{gamma} ELISPOT assay. We found no significant difference between assays in which the APC were prepulsed with peptide and washed, before being added to the effectors, and those assays in which the peptide remained throughout the culture period (data not shown). These results demonstrate that the M12.C3 cells function as the APC in the ELISPOT assay and, given that the only MHC molecule shared between M12.C3 and NOD mice is H-2 Kd, it is clear that the response is mediated by MHC class I-restricted T cells.



View larger version (11K):
[in this window]
[in a new window]
  		FIGURE 1. GAD peptides p206 and p546 induce IFN-{gamma} production in responding T cells. To determine whether GAD65 peptides or control peptides were capable of inducing IFN-{gamma} production in responding CTL, an ELISPOT assay was performed as described in Materials and Methods. A, Spleen cells from naive 4-wk-old female NOD mice were cultured in vitro for 5 days with peptide and then were restimulated by incubation with M12.C3 cells pulsed with 2 µg/ml homologous peptide. The CTL effectors were first serially diluted and then added to plates (1.7–45 x 104 cells/well) containing 1 x 105 target cells/well. Background responses (effectors cultured in the presence of M12.C3 cells without peptide pulsing) were subtracted from the values displayed above. The number of background spots ranged from 5 to 50 spots per well at the 45 x 104 effector cell concentration. This experiment is representative of four independent assays. B and C, After in vitro culture with peptide p206, p505, p546, or NP147, spleen cells from 6- to 8-wk-old female NOD mice were plated at 1.66, 5, and 15 x 104 cells/well in the ELISPOT assay. These results are representative of five experiments.

 
To examine the cytolytic activity of these GAD-reactive T cells, spleen cells from NOD mice were cultured with peptides p206, p505, or p546, and then tested for cytotoxicity against peptide-pulsed M12.C3 (H-2d) target cells. p206 and p546 were capable of activating effectors that displayed significant cytolytic activity (Table IIGo), whereas little response was observed with p505 (Table IIGo). Cytotoxicity against control targets (no peptide pulsing) was negligible (<5%, data not shown). In addition, NOD spleen cells cultured with control peptides NP147 or NP118 showed no specific cytotoxicity against target cells pulsed with homologous peptide or p206 or p546 (data not shown). The Kd molecule was shown to be the MHC restriction element for the GAD-specific CTL responses by genetic mapping using M12.C3 (H-2d) and P815 (H-2d) cells, which could serve as effective targets for p206- and p546-specific responses. Like M12.C3 cells, P815 cells do not express the NOD MHC class II molecule (data not shown). No lysis was seen with peptide-pulsed EL-4 cells (H-2b) (data not shown), indicating that the responses did not involve the Db molecule. p206- and p546-specific CTL activity were undetectable in naive BALB/c (H-2d) (Table IIGo) and NOR (H-2g7) (data not shown) mice, thus demonstrating that the spontaneous CTL responses in NOD mice were not due to an intrinsic mitogenic activity on the part of the GAD65 peptides. However, upon immunization, peptides p206, p505, p546, and NP147 were each able to prime specific CTL responses in BALB/c and NOD mice (data not shown). Moreover, after immunization, NP147 induced higher titers of specific CTL responses in both strains (data not shown) and thus appeared to be more immunogenic than p206 and p546 in NOD mice. Therefore, the spontaneous CTL responses to p206 and p546 in NOD mice were apparently not owing to nonspecific CTL activity or promoted by in vitro priming by very immunogenic peptides. We were unable to detect proliferative responses to any of the MHC class I-restricted GAD peptides. Proliferative responses of CD8+ T cells in general may be difficult to demonstrate; thus, despite being able to induce CTL responses to NP147 after immunization, we were still unable to detect proliferation to the homologous peptide in recall assays (data not shown).


View this table:
[in this window]
[in a new window]
  		Table II. NOD mice spontaneously develop CTL responses to GAD65 peptides

 
Finally, although we can quite easily detect proliferative Th recall responses to the GAD peptides 206–220, 530–543, and 538–558 from NOD mice immunized with homologous peptide, we do not see such responses in NOD mice immunized with GAD peptides p546–554 or p206–214 (data not shown). These data support the contention that the p546–554- and p206–214-specific responses are distinct from the previously described Th responses (1, 3, 4).

The expression of spontaneous GAD-specific CTL activity in neonatal, young, and diabetic NOD mice

To define the temporal expression of the spontaneous GAD-specific CTL activity, spleen cells from 4-, 10-, and 23-wk-old normal glycemic or diabetic female NOD mice were stimulated in vitro with a pool of GAD peptides p206, p505, and p546, and then tested for peptide-specific CTL activity. Specific effectors raised from normal glycemic mice displayed significant CTL activity at 100:1 (25–42%) and 20:1 (25–42%) E:T ratios (Fig. 2GoA). However, GAD-specific CTL raised from diabetic mice were less effective at killing peptide-pulsed targets (0–18%) (Fig. 2GoA). The peak CTL responses to p206 and p546 were typically seen in 6- to 10-wk-old mice (data not shown).



View larger version (23K):
[in this window]
[in a new window]
  		FIGURE 2. Spontaneous CTL responses to GAD peptides p206 and p546 are not seen in very young NOD mice. A, Spleen cells from 4-, 10-, and 23-wk-old normal glycemic female NOD mice and diabetic female NOD mice (25–30 wk of age) were stimulated in vitro with a pool of peptides p206, p546, and p505 (5 µg/ml each), then cultured and tested for CTL activity against P815 (H-2d) cells prepulsed with the homologous peptides, at E:T ratios of 100:1 and 20:1. These results are representative of two independent experiments, and mice from all groups were tested in the same assay. B, Spleen cells from 4 ({square})- and 3 (•)-wk-old female NOD mice (two mice per group) or pooled cells from 1-wk ({triangleup})-old NOD mice were stimulated in vitro with p206 or p546 and then tested for CTL activity against P815 cells pulsed with the cognate peptide. The results are representative of two experiments.

 
Because IDDM occurs spontaneously in NOD mice, the time point at which islet-specific autoimmunity is initiated is not clear, although it would be expected to precede insulitis. Thus, we examined the CTL responses to p206 and p546 in the spleen cells of NOD mice, at 1, 3, or 4 wk of age. p206- and p546-specific cytolytic activities were seen in 3-wk-old mice (2–14%), but were undetectable at 1 wk (Fig. 2GoB) or 2 wk of age (data not shown). These results show that p206- and p546-specific CTL are not always present in NOD spleen cells, but rather they are detectable just before histological insulitis and decrease after the onset of diabetes.

CTL lines raised to GAD p206 or p546 are CD8+ and cytotoxic for GAD-transfected target cells

To show that the effectors activated by p206 and p546 could recognize determinants processed from endogenously synthesized GAD65, the GAD-expressing M12.C3.GAD cell line was used as a target. Cells transfected with GAD65 were chosen as targets rather than purified islets, since islets are a mixture of at least three different cell populations ({alpha}, {beta}, and {delta}); {beta} and non-{beta} cells both will become labeled with chromium, thus masking the specific activity of the effectors. Furthermore, in our hands, chromium-loaded islet cells can be very labile and display high backgrounds (our unpublished observations).

Short-term CTL lines were produced from the spleen cells of NOD mice immunized with p206, p546, or NP147. The Ag-specific CTL lines 206A, 546A, and flu-1 were cytotoxic for target cells pulsed with cognate peptides p206, p546, or NP147, respectively (data not shown); however, only lines 206A and 546A were able to specifically lyse the GAD-expressing targets (Fig. 3GoA). In addition, a p546-specific CTL line, 546F, produced from the spleen cells of naive NOD mice by repeated in vitro stimulation with peptide-pulsed LPS blasts, was cytotoxic for M12.C3.GAD cells and M12.C3.RSV cells pulsed with peptide p546, but not M12.C3.RSV cells alone (Fig. 3GoB). These data demonstrate that p206- and p546-specific CTL recognize determinants processed from the GAD65 molecule. Additional p546-specific CTL lines (546H, I, and L) were produced and found to be primarily composed of CD8+ T cells (>90%). Similar to the splenic response, these lines produced IFN-{gamma}, but no IL-4 or IL-5, in response to antigenic stimulation with p546 (data not shown). To definitively demonstrate that CD8+ T cells mediated the spontaneous GAD-specific CTL activity, spleen cells activated by p206 or p546 were preincubated with mAbs specific for CD8, CD4, or control Ig, before being tested for cytotoxicity against peptide-pulsed targets. Greater than 90% of the CTL activity induced by each peptide was blocked by anti-CD8, while anti-CD4 Ab was similar to the Ig control in showing no inhibition (Fig. 3GoC).



View larger version (15K):
[in this window]
[in a new window]
  		FIGURE 3. Cells specific for GAD p546 or p206 are cytotoxic for M12.C3 cells that express GAD65. A, Six- to 8-wk-old female NOD mice were immunized s.c. with 10 µg of p546 ({square}), p206 (•), or NP147 ({triangleup}) in IFA (two mice per group). Ten to 14 days later, the spleen cells were stimulated in vitro with homologous peptide in IL-2-containing medium, and then tested 5 days later for cytotoxicity against target cells that express rat GAD65, M12.C3.GAD. Nonspecific activity, chromium released from target cells transfected with the vector control, M12.C3.RSV, has been subtracted from the results above. The effectors were tested at E:T ratios of 100, 25, 6.25, and 1.56:1. B, Line 546F was produced by the weekly restimulation of spleen cells from untreated NOD mice using LPS blasts pulsed with peptide p546, in IL-2-containing medium. Five days after the third in vitro restimulation, line 546F was plated at E:T ratios of 100, 33.3, 11.1, 3.7, and 1.23 with GAD65-expressing target cells M12.C3.GAD ({square}), M12.C3.RSV, the vector control cells (•), or M12.C3.RSV pulsed with GAD 546 ({diamondsuit}). C, Spleen cells from 6- to 8-wk-old female NOD mice were stimulated in vitro with p206 or p546 (5 µg/ml). Before being added to P815 (H-2d) cells, prepulsed with homologous peptide, the effectors were incubated with mAb (10 µg/ml in complete medium) specific for mouse CD4, mouse CD8, or lysozyme (Ig control) for 1 h at room temperature. The effectors were then washed once with PBS and added to the target cells to test for CTL activity at an E:T of 50:1. The percentage of specific cytotoxicity from effectors preincubated in medium alone was 24 and 18% for peptide p546- or p206-pulsed cells, respectively. No inhibition was seen with anti-CD4 or anti-lysozyme (IgG). The results are representative of three experiments.

 
Finally, to determine whether GAD 546-specific T cells could contribute to the pathogenesis of IDDM, two T cell lines (546I and 546L) were each labeled with the fluorescent tracking dye, DiI, and injected i.v. into two female NOD.scid mice. The labeling efficiency was checked by fluorescence microscopy before the transfer. Forty-eight to seventy-two hours later, the pancreas, spleen, liver, and a kidney were collected from each animal and snap frozen. Labeled 546-specific T cells were found only in the pancreatic islets (Fig. 4GoA) and spleens (data not shown) of recipient mice, and some of the islets displayed cellular damage (Fig. 4GoA). The lymphocytic infiltration was organ and islet specific since no infiltration was seen in the liver or kidney, and the labeled cells were only associated with pancreatic islets. These data provide further evidence that such cells can contribute to the early stages of IDDM. We have not been able to show that these T cells can transfer IDDM themselves, but since we are studying the early events in the disease process, the cells we are characterizing may not necessarily be the effectors responsible for the pivotal progression from insulitis to IDDM. FACS analysis shows that lines 546I (Fig. 4GoC) and 546L (data not shown) are CD8+ T cells.



View larger version (67K):
[in this window]
[in a new window]
  		FIGURE 4. CTL line GAD 546I induces insulitis in NOD.scid mice. GAD 546–554-specific CTL were labeled with the fluorescent tracking dye DiI and adoptively transferred by i.v. injection into two NOD.scid mice (5 x 106/mouse). The mice were sacrificed 72 h later, and the pancreas, spleen, liver, and kidney were snap frozen. Shown are representative 8-µm cryosections of a hematoxylin-stained pancreatic islet (A, original magnification, x20) and a fluorescent image of an adjacent section (B, original magnification, x20). C, FACS analysis of line 546I shows that it is a CD8+ T cell line. The unshaded curves represent the Ig-FITC control.

 
Nasal instillation (NI) with GAD p206 and p546 delays the onset of spontaneous IDDM in NOD mice

Treatment with CD4-inducing GAD peptides ameliorates diabetes in NOD mice (1, 5). To determine whether nasal treatment with MHC class I-restricted determinants could also afford protection from disease, neonatal NOD mice were nasally treated with p206 and p546 (n = 9) on days 7, 9, and 11 of life. Control mice were treated with matching doses of NP147 (n = 12) or were left untreated (n = 6). After weaning, all of the female animals were tested biweekly for glucosuria. Nasal treatment with GAD65 peptides was able to delay the onset of spontaneous IDDM and reduce the overall incidence in female mice; however, its protective effect appeared to wane during the extended period of the study. At 18 wk of age, 22.2% of the mice in the GAD group were diabetic, whereas 40 and 60% of the saline- and NP147-treated mice, respectively, were diabetic (Fig. 5Go). By 23 wk of age, all of the control mice (100%) were diabetic, whereas the incidence of IDDM in GAD peptide-treated mice was significantly reduced (33%; Fig. 5Go). This delay in onset of spontaneous IDDM observed in mice pretreated with GAD65 peptides 546–554 and 206–214 suggested that T cells responsive to these determinants could influence the pathogenic process. Fully one-third of the GAD peptide-treated mice showed no diabetes through the 44th week of study (Fig. 5Go).



View larger version (18K):
[in this window]
[in a new window]
  		FIGURE 5. Nasal treatment with GAD peptides p206 and p546 delays the onset of spontaneous IDDM in NOD mice. Neonatal female NOD mice were nasally treated on days 7, 9, and 11 with 30 µg of p206 and p546, 15 µg each ({triangleup}, n = 9) or 30 µg of peptide NP147 (•, n = 12), or were left untreated ({square}, n = 6). At 4 wk of age, the mice were weaned and the females were monitored for glucosuria until the age of 45 wk. At 23 wk of age, using Fisher’s exact test, the incidence of IDDM in treated mice was compared with that of untreated mice (GAD, p = 0.031; flu, p = 0.515); p < 0.05 is considered statistically significant.

 
Treatment with p546 inhibits cyclophosphamide-induced diabetes

Because the incubation period for spontaneous IDDM can be rather lengthy (35–50 wk), we suspect that additional nasal treatments might have led to a more profound influence on the course of spontaneous disease. However, in this study, we wanted to avoid the nonspecific effects that may occur when giving large doses of Ag over an extended period of time. Although a number of immunogenic components have been reported to arrest the spontaneous development of IDDM, a more stringent test of a protective therapy is its ability to block cyclophosphamide-induced IDDM (28, 31). An injection of cyclophosphamide (200 mg/kg) at 10–13 wk of age increases the incidence of IDDM in male NOD mice from 0 to 95–100%, within 2 wk of the injection. Complete penetrance is similarly seen in female NOD mice. Therefore, to synchronize the onset of IDDM and to further test the efficacy of protection provided by nasal treatment with p546, 10-wk-old neonatally treated NOD mice were given a single injection of cyclophosphamide. In mice pretreated with p546, the incidence of diabetes was reduced by 50% (Table IIIGo), while no protection was seen in animals pretreated with the control peptide NP147 (Table IIIGo). The protected mice were followed for an additional 10 days, beyond the typical 2-wk observation period, to ensure that the disease was not simply delayed; it was not.


View this table:
[in this window]
[in a new window]
  		Table III. Nasal pretreatment with GAD peptides protects NOD mice from cyclophosphamide-induced IDDM

 
Thus, in two models of diabetes in the NOD mouse, nasal treatment with MHC class I-restricted GAD determinants produced amelioration from IDDM. To determine whether such treatments had an effect on the CTL response to homologous peptides, instead of receiving cyclophosphamide at 10 wk of age, some of the mice from the groups described above (NI with p546 or NP147) were challenged by s.c. injections with homologous peptide in IFA. The Ag-specific CTL response in mice pretreated with p546 was reduced by >60% when compared with animals NI with saline (Table IVGo). Similarly, NP147-pretreated mice showed a reduction in specific CTL response following immunization with NP147 in IFA (Table IVGo). The CTL responses to NP147 was less sensitive to tolerance induction, most likely due to its strong immunogenic nature in NOD mice. In additional experiments, adult NOD mice pretreated by NI with p546 or NP147, before immunization with homologous peptide in IFA, also had reduced peptide-specific CTL responses compared with those pretreated with saline (data not shown). Therefore, the protection provided by NI with MHC class I-restricted GAD peptides was most likely mediated through the induction of Ag-specific tolerance.


View this table:
[in this window]
[in a new window]
  		Table IV. NI with MHC class-restricted peptides reduces the CTL response in mice immunized with homologous peptide

 
Treatment with CFA reduces the spontaneous CTL response to GAD peptides

Exposure to CFA has been shown to interfere with the progression of IDDM in NOD mice (31, 32). We wanted to determine whether the protective effect produced by CFA immunization could influence the spontaneous CTL response to p206 or p546. Female NOD were treated with CFA and then tested 2 wk later for CTL responses to the GAD determinants. The CFA pretreatment consistently reduced the splenic CTL response to p206 and p546 (data not shown). Although the mechanisms involved in CFA-induced protection in NOD mice are still unresolved, these findings suggest that the presumed regulatory network induced by CFA injections apparently influences these GAD-specific CTL responses. Injections with IFA or PBS had no effect on such CTL responses (data not shown).


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Autoimmune diabetes is a manifestation of permanent {beta} cell loss from the islets of Langerhans, leading to an inability to maintain glucose homeostasis. Whether the critical threshold of {beta} cell damage is due to chronic immunological assault or rather to an acute attack precipitated by some yet undescribed event, is a matter of debate (33, 34). However, one of the central issues in the discussion is the mechanism by which {beta} cells are destroyed. In this study, we show that CD8+ T cells specific for GAD65 sequences 206–214 and 546–554 secrete IFN-{gamma} upon antigenic challenge, are cytotoxic for GAD-expressing cells, and can transfer insulitis. Such MHC class I-restricted responses may provide the necessary focus for the generation of islet-specific inflammation in NOD mice. Although CD4+ T cells are requisite for spontaneous IDDM and some Th clones can transfer disease without the need for donor CD8+ T cells (35, 36), the means by which they perpetuate hyperglycemia is not clear since murine {beta} cells do not express MHC class II molecules on their surface (37, 38). It is possible that certain Th cells incite islet cell damage via the production of deleterious cytokines (39) or through Fas/Fas ligand-mediated events precipitated by interactions between activated T cells and {beta} cells (9). Alternatively, CD8+ T cells may initiate early {beta} cell damage, which then leads to the release of islet Ags and the activation of CD4+ T cells (11, 40). The islet-reactive Th cells could then home to the pancreas, secrete cytokines/chemokines that attract other inflammatory cells, and escalate the inflammatory process. Accordingly, previously described diabetogenic CD4+ T cell clones, which transfer IDDM without the need for CD8+ T cells, were themselves recovered from mice that possessed a CD8+ T cell repertoire (35, 36). Thus, while the role of CD8+ T cells as initiators or terminal effectors of {beta} cell damage in type I diabetes remains unresolved, we have demonstrated that CTL specific for the Kd-restricted GAD65 determinants 546–554 and 206–214 are detectable in the spleens of NOD mice before the onset of overt insulitis. Whether these CD8+ T cell responses require CD4 T cell help remains to be determined; however, certain determinants with sufficient affinity for MHC class I molecules are able to induce CTL responses in a Th-independent manner (41). Although we did not determine the affinity of peptides p546 and p206 for the Kd molecule, we were able to show that the immune responses to these peptides, in both naive and immunized NOD mice, were blocked by anti-CD8 Ab, but not by anti-CD4, suggesting that CD4+ T cells are not required during activation.

CTL responses to p206 and p546 were consistently detectable, but qualitatively the cytotoxicity varied from experiment to experiment (25–70%) and was always most evident at E:T ratios of 100:1–50:1. The magnitude of these responses was not so surprising given that they are directed to a self Ag and occurred in untreated mice (14). Moreover, even CD4+ T cell responses to well-described islet Ags are sometimes difficult to detect in naive mice (42). Thus, anti-islet CTL precursor frequencies in the natural course of IDDM cannot be compared with the induced responses observed when mice are given sublethal doses of an infectious virus, a large bolus of oncogenic cells, or immunized with peptides. Nonetheless, we were able to detect p206- and p546-responsive cells at E:T ratios of less than 0.5:1 when using the IFN-{gamma} ELISPOT assay, which is more sensitive than the cytotoxicity assay.

The concept that functional CD8+ T cell-inducing determinants on a protein molecule lie near CD4+ T cell-inducing determinants of the same molecule is intriguing. In the present study, a larger set of putative CD8-inducing determinants, not lying near strong CD4-inducing determinants, did not induce a spontaneous CTL response. Previously, it was shown that not all Th-induced responses to {beta}-galactosidase were suppressor T cell sensitive, but only those Th determinants residing near CD8+ suppressor T cell-inducing determinants (18), a finding that implies a regional interaction between MHC class I and class II responses. Overlapping MHC class I and class II determinants have also been described in humans (43, 44, 45). More recently, it was discovered that a diabetogenic CD8+ T cell clone recognized a site on the insulin {beta}-chain (14), which overlaps a well-described Th-inducing determinant ({beta}9–23) (36, 46). The mechanism(s) that controls and mediates this regional response is not understood, although it is tempting to suggest that the processing of neighboring MHC class I and class II determinants occurs in the same vicinity. However, not all CD8+ T cell-inducing determinants are located proximal to Th-inducing ones, just as not all CTL responses are predicated on CD40-CD40L interactions (47, 48). The observation is nonetheless interesting and may provide a useful approach for identifying relevant CD8+ T cell determinants in infectious disease, cancer, and autoimmunity.

Previously, it was shown that NOD mice treated with MHC class II-restricted peptides from GAD (1), insulin (46), or heat-shock protein 60 (49) were protected from IDDM. In this study, we found that not only did NOD mice pretreated with p546 and p206 show a delayed onset of spontaneous IDDM, but most impressively, p546 by itself was able to confer protection from CY-induced IDDM, a very aggressive model of IDDM. The extended incubation period of spontaneous IDDM may have taxed the regulation induced by our peptides such that the protection was more effective in the early stages of the experiment, but appeared to lose the ability to ameliorate the disease as the animals aged. Peripheral tolerance induced in euthymic mice by nasal treatment with peptides can be reversed after a few months by new thymic emigrants (50). The results from experiments with CY-induced IDDM, which has a shorter incubation period, support this conclusion. It should be noted that even in the p546-treated mice that succumbed to inducible diabetes, the onset of the disease occurred much later (10–12 days after CY treatment compared with 5–7 days in the saline-treated mice). Despite the protection from IDDM, we could not demonstrate a statistically significant reduction of islet infiltration in p206- and p546-pretreated animals sacrificed at 8 wk of age. It is possible that a difference might be evident at another time point, or perhaps protection may involve a qualitative change in vivo. We did not detect an increase in IL-4-, IL-5-, or IL-10-producing cells after NI with p206 or p546, which would have been indicative of immune deviation (data not shown).

Although autoimmunity to other Ags may contribute at some point to the pathogenesis of IDDM, the absence of insulitis and diabetes in mice lacking GAD65 (13) provides strong evidence that GAD is a major autoantigen in the initiation of the disease. In an effort to determine the impact of nasal treatment with the MHC class I-binding peptides on the spontaneous MHC class II-restricted responses to islet Ags, proliferative responses to GAD 524–543, the insulin {beta}-chain, or the heat shock protein 60 peptide (P277) (49) were analyzed in p206- and p546-pretreated or saline-treated controls. Despite finding that proliferative responses to GAD 524–543 were generally decreased in such mice, the individual variability in the response to each of the three Ags made the results difficult to interpret or to show statistical significance (data not shown). In any case, the use of MHC class I-restricted peptides to ameliorate IDDM in the NOD mouse model represents an important step in our efforts to treat autoimmune disease. Although a great deal of success has been achieved using MHC class II-restricted peptides therapeutically in a number of animal models of autoimmune disease, a comparable level of achievement has been difficult to approach in humans. One of the obstacles to human therapy lies in predicting and designing the appropriate MHC class II-binding peptides for an outbred population. Fortunately, MHC class I alleles typically display a more limited predilection for peptides, forming binding motifs that are easier to characterize. The use of CD8+ T cell-inducing peptides to treat autoimmunity, perhaps in addition to relevant CD4-inducing peptides, may represent a more pragmatic approach to therapy. To our knowledge, this is the first reported demonstration of MHC class I-restricted self peptides being effectively used to treat an autoimmune disease.


   Acknowledgments
 
We thank Drs. Alessandra Franco, Susanne Schneider, and Steven Schoenberger for discussions and criticisms on this manuscript. This is Publication 340 from the La Jolla Institute for Allergy and Immunology.


   Footnotes
 
1 This work was supported by National Institutes of Health Grant AI28419 and grants from the Juvenile Diabetes Foundation International. M.F.M. was supported by a Career Development Award from the American Diabetes Association. Back

2 Address correspondence and reprint requests to Dr. Eli E. Sercarz, Division of Immune Regulation, La Jolla Institute for Allergy and Immunology, 10355 Science Center Drive, San Diego, CA 92121. E-mail address: eli{at}liai.org Back

3 Abbreviations used in this paper: IDDM, insulin-dependent diabetes mellitus; CY, cyclophosphamide; DiI, octadecyl indocarbocyanine; GAD, glutamic acid decarboxylase; HEL, hen egg white lysozyme; NI, nasal instillation; NOD, nonobese diabetic; NP, nucleoprotein. Back

Received for publication October 11, 2000. Accepted for publication May 17, 2001.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Kaufman, D. L., M. Clare-Salzler, J. Tian, T. Forsthuber, G. S. Ting, P. Robinson, M. A. Atkinson, E. E. Sercarz, A. J. Tobin, P. V. Lehmann. 1993. Spontaneous loss of T-cell tolerance to glutamic acid decarboxylase in murine insulin-dependent diabetes. Nature 366:69.[Medline]
  2. Tisch, R., X. D. Yang, S. M. Singer, R. S. Liblau, L. Fugger, H. O. McDevitt. 1993. Immune response to glutamic acid decarboxylase correlates with insulitis in non-obese diabetic mice. Nature 366:72.[Medline]
  3. Chao, C. C., H. O. McDevitt. 1997. Identification of immunogenic epitopes of GAD 65 presented by Ag7 in non-obese diabetic mice. Immunogenetics 46:29.[Medline]
  4. Zechel, M. A., J. F. Elliott, M. A. Atkinson, B. Singh. 1998. Characterization of novel T-cell epitopes on 65 kDa and 67 kDa glutamic acid decarboxylase relevant in autoimmune responses in NOD mice. J. Autoimmun. 11:83.[Medline]
  5. Tian, J., M. Clare-Salzler, A. Herschenfeld, B. Middleton, D. Newman, R. Mueller, S. Arita, C. Evans, M. A. Atkinson, Y. Mullen, et al 1996. Modulating autoimmune responses to GAD inhibits disease progression and prolongs islet graft survival in diabetes-prone mice. Nat. Med. 2:1348.[Medline]
  6. Tisch, R., B. Wang, D. V. Serreze. 1999. Induction of glutamic acid decarboxylase 65-specific Th2 cells and suppression of autoimmune diabetes at late stages of disease is epitope dependent. J. Immunol. 163:1178.[Abstract/Free Full Text]
  7. Zekzer, D., F. S. Wong, O. Ayalon, I. Millet, M. Altieri, S. Shintani, M. Solimena, R. S. Sherwin. 1998. GAD-reactive CD4+ Th1 cells induce diabetes in NOD/SCID mice. J. Clin. Invest. 101:68.[Medline]
  8. Miller, B. J., M. C. Appel, J. J. O’Neil, L. S. Wicker. 1988. Both the Lyt-2+ and L3T4+ T cell subsets are required for the transfer of diabetes in nonobese diabetic mice. J. Immunol. 140:52.[Abstract]
  9. Katz, J., C. Benoist, D. Mathis. 1993. Major histocompatibility complex class I molecules are required for the development of insulitis in non-obese diabetic mice. Eur. J. Immunol. 23:3358.[Medline]
  10. Sumida, T., M. Furukawa, A. Sakamoto, T. Namekawa, T. Maeda, M. Zijlstra, I. Iwamoto, T. Koike, S. Yoshida, H. Tomioka, et al 1994. Prevention of insulitis and diabetes in {beta}2-microglobulin-deficient non-obese diabetic mice. Int. Immunol. 6:1445.[Abstract/Free Full Text]
  11. Christianson, S. W., L. D. Shultz, E. H. Leiter. 1993. Adoptive transfer of diabetes into immunodeficient NOD-scid/scid mice: relative contributions of CD4+ and CD8+ T-cells from diabetic versus prediabetic NOD.NON-Thy-1a donors. Diabetes 42:44.[Abstract]
  12. Serreze, D. V., H. D. Chapman, D. S. Varnum, I. Gerling, E. H. Leiter, L. D. Shultz. 1997. Initiation of autoimmune diabetes in NOD/Lt mice is MHC class I-dependent. J. Immunol. 158:3978.[Abstract]
  13. Yoon, J. W., C. S. Yoon, H. W. Lim, Q. Q. Huang, Y. Kang, K. H. Pyun, K. Hirasawa, R. S. Sherwin, H. S. Jun. 1999. Control of autoimmune diabetes in NOD mice by GAD expression or suppression in {beta} cells. Science 284:1183.[Abstract/Free Full Text]
  14. Wong, F. S., J. Karttunen, C. Dumont, L. Wen, I. Visintin, I. M. Pilip, N. Shastri, E. G. Pamer, Jr C. A. Janeway. 1999. Identification of an MHC class I-restricted autoantigen in type 1 diabetes by screening an organ-specific cDNA library. Nat. Med. 5:1026.[Medline]
  15. Falk, K., O. Rotzschke, S. Stevanovic, G. Jung, H. G. Rammensee. 1991. Allele-specific motifs revealed by sequencing of self-peptides eluted from MHC molecules. Nature 351:290.[Medline]
  16. Romero, P., G. Corradin, I. F. Luescher, J. L. Maryanski. 1991. H-2Kd-restricted antigenic peptides share a simple binding motif. J. Exp. Med. 174:603.[Abstract/Free Full Text]
  17. Rammensee, H. G., K. Falk, O. Rotzschke. 1993. Peptides naturally presented by MHC class I molecules. Annu. Rev. Immunol. 11:213.[Medline]
  18. Krzych, U., A. V. Fowler, E. E. Sercarz. 1985. Repertoires of T cells directed against a large protein antigen, {beta}-galactosidase. II. Only certain T helper or T suppressor cells are relevant in particular regulatory interactions. J. Exp. Med. 162:311.[Abstract/Free Full Text]
  19. Shivakumar, S., E. E. Sercarz, U. Krzych. 1989. The molecular context of determinants within the priming antigen establishes a hierarchy of T cell induction: T cell specificities induced by peptides of {beta}-galactosidase vs. the whole antigen. Eur. J. Immunol. 19:681.[Medline]
  20. Kumar, V., E. Sercarz. 1996. Dysregulation of potentially pathogenic self reactivity is crucial for the manifestation of clinical autoimmunity. J. Neurosci. Res. 45:334.[Medline]
  21. Townsend, A., C. Ohlen, J. Bastin, H. G. Ljunggren, L. Foster, K. Karre. 1989. Association of class I major histocompatibility heavy and light chains induced by viral peptides. Nature 340:443.[Medline]
  22. Whitton, J. L., A. Tishon, H. Lewicki, J. Gebhard, T. Cook, M. Salvato, E. Joly, M. B. Oldstone. 1989. Molecular analyses of a five-amino-acid cytotoxic T-lymphocyte (CTL) epitope: an immunodominant region which induces nonreciprocal CTL cross-reactivity. J. Virol. 63:4303.[Abstract/Free Full Text]
  23. Glimcher, L. H., D. J. McKean, E. Choi, J. G. Seidman. 1985. Complex regulation of class II gene expression: analysis with class II mutant cell lines. J. Immunol. 135:3542.[Abstract]
  24. Solimena, M., D. Aggujaro, C. Muntzel, R. Dirkx, M. Butler, P. De Camilli, A. Hayday. 1993. Association of GAD-65, but not of GAD-67, with the Golgi complex of transfected Chinese hamster ovary cells mediated by the N-terminal region. Proc. Natl. Acad. Sci. USA 90:3073.[Abstract/Free Full Text]
  25. Chang, Y. C., D. I. Gottlieb. 1988. Characterization of the proteins purified with monoclonal antibodies to glutamic acid decarboxylase. J. Neurosci. 8:2123.[Abstract]
  26. Deng, H., R. Apple, M. Clare-Salzler, S. Trembleau, D. Mathis, L. Adorini, E. Sercarz. 1993. Determinant capture as a possible mechanism of protection afforded by major histocompatibility complex class II molecules in autoimmune disease. J. Exp. Med. 178:1675.[Abstract/Free Full Text]
  27. Quinn, A., E. E. Sercarz. 1996. T cells with multiple fine specificities are used by non-obese diabetic (NOD) mice in the response to GAD524–543. J. Autoimmun. 9:365.[Medline]
  28. Yasunami, R., J. F. Bach. 1988. Anti-suppressor effect of cyclophosphamide on the development of spontaneous diabetes in NOD mice. Eur. J. Immunol. 18:481.[Medline]
  29. Nagata, M., K. Yokono, M. Hayakawa, Y. Kawase, N. Hatamori, W. Ogawa, K. Yonezawa, K. Shii, S. Baba. 1989. Destruction of pancreatic islet cells by cytotoxic T lymphocytes in nonobese diabetic mice. J. Immunol. 143:1155.[Abstract]
  30. Taki, T., M. Nagata, W. Ogawa, N. Hatamori, M. Hayakawa, J. Hari, K. Shii, S. Baba, K. Yokono. 1991. Prevention of cyclophosphamide-induced and spontaneous diabetes in NOD/Shi/Kbe mice by anti-MHC class I Kd monoclonal antibody. Diabetes 40:1203.[Abstract]
  31. Sadelain, M. W., H. Y. Qin, J. Lauzon, B. Singh. 1990. Prevention of type I diabetes in NOD mice by adjuvant immunotherapy. Diabetes 39:583.[Abstract]
  32. McInerney, M. F., S. B. Pek, D. W. Thomas. 1991. Prevention of insulitis and diabetes onset by treatment with complete Freund’s adjuvant in NOD mice. Diabetes 40:715.[Abstract]
  33. Eisenbarth, G. S.. 1986. Type I diabetes mellitus: a chronic autoimmune disease. N. Engl. J. Med. 314:1360.[Medline]
  34. Dilts, S. M., K. J. Lafferty. 1999. Autoimmune diabetes: the involvement of benign and malignant autoimmunity. J. Autoimmun. 12:229.[Medline]
  35. Peterson, J. D., K. Haskins. 1996. Transfer of diabetes in the NOD-scid mouse by CD4 T-cell clones: differential requirement for CD8 T-cells. Diabetes 45:328.[Abstract]
  36. Wegmann, D. R., R. G. Gill, M. Norbury-Glaser, N. Schloot, D. Daniel. 1994. Analysis of the spontaneous T cell response to insulin in NOD mice. J. Autoimmun. 7:833.[Medline]
  37. McInerney, M. F., S. Rath, Jr C. A. Janeway. 1991. Exclusive expression of MHC class II proteins on CD45+ cells in pancreatic islets of NOD mice. Diabetes 40:648.[Abstract]
  38. Kay, T. W., I. L. Campbell, L. Oxbrow, L. C. Harrison. 1991. Overexpression of class I major histocompatibility complex accompanies insulitis in the non-obese diabetic mouse and is prevented by anti-interferon-{gamma} antibody. Diabetologia 34:779.[Medline]
  39. Rabinovitch, A., W. L. Suarez-Pinzon. 1998. Cytokines and their roles in pancreatic islet {beta}-cell destruction and insulin-dependent diabetes mellitus. Biochem. Pharmacol. 55:1139.[Medline]
  40. Young, L. H., L. B. Peterson, L. S. Wicker, P. M. Persechini, J. D. Young. 1989. In vivo expression of perforin by CD8+ lymphocytes in autoimmune disease: studies on spontaneous and adoptively transferred diabetes in nonobese diabetic mice. J. Immunol. 143:3994.[Abstract]
  41. Alessandro Franco, D. T., I. Gramaglia, M. Croft, L. Cipolla, M. Meldal, H. Grey. 2000. Epitope affinity for MHC class I determines helper requirement for CTL priming. Nat. Immun. 1:145.
  42. Hanson, M. S., M. Cetkovic-Cvrlje, V. K. Ramiya, M. A. Atkinson, N. K. Maclaren, B. Singh, J. F. Elliott, D. V. Serreze, E. H. Leiter. 1996. Quantitative thresholds of MHC class II I-E expressed on hemopoietically derived antigen-presenting cells in transgenic NOD/Lt mice determine level of diabetes resistance and indicate mechanism of protection. J. Immunol. 157:1279.[Abstract]
  43. Carreno, B. M., R. V. Turner, W. E. Biddison, J. E. Coligan. 1992. Overlapping epitopes that are recognized by CD8+ HLA class I-restricted and CD4+ class II-restricted cytotoxic T lymphocytes are contained within an influenza nucleoprotein peptide. J. Immunol. 148:894.[Abstract]
  44. Ou, D., L. A. Mitchell, D. Decarie, S. Gillam, A. J. Tingle. 1997. Characterization of an overlapping CD8+ and CD4+ T-cell epitope on rubella capsid protein. Virology 235:286.[Medline]
  45. Takahashi, H., R. N. Germain, B. Moss, J. A. Berzofsky. 1990. An immunodominant class I-restricted cytotoxic T lymphocyte determinant of human immunodeficiency virus type 1 induces CD4 class II-restricted help for itself. J. Exp. Med. 171:571.[Abstract/Free Full Text]
  46. Daniel, D., D. R. Wegmann. 1996. Protection of nonobese diabetic mice from diabetes by intranasal or subcutaneous administration of insulin peptide B-9–23. Proc. Natl. Acad. Sci. USA 93:956.[Abstract/Free Full Text]
  47. Schoenberger, S. P., R. E. Toes, E. I. van der Voort, R. Offringa, C. J. Melief. 1998. T-cell help for cytotoxic T lymphocytes is mediated by CD40-CD40L interactions. Nature 393:480.[Medline]
  48. Bennett, S. R., F. R. Carbone, F. Karamalis, R. A. Flavell, J. F. Miller, W. R. Heath. 1998. Help for cytotoxic-T-cell responses is mediated by CD40 signalling. Nature 393:478.[Medline]
  49. Elias, D., I. R. Cohen. 1994. Peptide therapy for diabetes in NOD mice. Lancet 343:704.[Medline]
  50. Metzler, B., D. C. Wraith. 1999. Inhibition of T-cell responsiveness by nasal peptide administration: influence of the thymus and differential recovery of T-cell-dependent functions. Immunology 97:257.[Medline]



This article has been cited by other articles:


Home page
 Int ImmunolHome page
A. Quinn, M. McInerney, D. Huffman, B. McInerney, S. Mayo, K. Haskins, and E. Sercarz
T cells to a dominant epitope of GAD65 express a public CDR3 motif
Int. Immunol., June 1, 2006; 18(6): 967 - 979.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
C. P. Wong, L. Li, J. A. Frelinger, and R. Tisch
Early Autoimmune Destruction of Islet Grafts Is Associated with a Restricted Repertoire of IGRP-Specific CD8+ T Cells in Diabetic Nonobese Diabetic Mice
J. Immunol., February 1, 2006; 176(3): 1637 - 1644.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
G. Han, Y. Li, J. Wang, R. Wang, G. Chen, L. Song, R. Xu, M. Yu, X. Wu, J. Qian, et al.
Active Tolerance Induction and Prevention of Autoimmune Diabetes by Immunogene Therapy Using Recombinant Adenoassociated Virus Expressing Glutamic Acid Decarboxylase 65 Peptide GAD500-585
J. Immunol., April 15, 2005; 174(8): 4516 - 4524.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
R. Kennedy, A. H. Undale, W. C. Kieper, M. S. Block, L. R. Pease, and E. Celis
Direct Cross-Priming by Th Lymphocytes Generates Memory Cytotoxic T Cell Responses
J. Immunol., April 1, 2005; 174(7): 3967 - 3977.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
S. M. Lieberman, T. Takaki, B. Han, P. Santamaria, D. V. Serreze, and T. P. DiLorenzo
Individual Nonobese Diabetic Mice Exhibit Unique Patterns of CD8+ T Cell Reactivity to Three Islet Antigens, Including the Newly Identified Widely Expressed Dystrophia Myotonica Kinase
J. Immunol., December 1, 2004; 173(11): 6727 - 6734.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
T. Takaki, S. M. Lieberman, T. M. Holl, B. Han, P. Santamaria, D. V. Serreze, and T. P. DiLorenzo
Requirement for Both H-2Db and H-2Kd for the Induction of Diabetes by the Promiscuous CD8+ T Cell Clonotype AI4
J. Immunol., August 15, 2004; 173(4): 2530 - 2541.
[Abstract] [Full Text] [PDF]

Home page
 Int ImmunolHome page
A. Margalit, S. Fishman, D. Berko, J. Engberg, and G. Gross
Chimeric {beta}2 microglobulin/CD3{zeta} polypeptides expressed in T cells convert MHC class I peptide ligands into T cell activation receptors: a potential tool for specific targeting of pathogenic CD8+ T cells
Int. Immunol., November 1, 2003; 15(11): 1379 - 1387.