HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Trembleau, S. Articles by Adorini, L. Search for Related Content PubMed PubMed Citation Articles by Trembleau, S. Articles by Adorini, L.

IL-12 Administration Reveals Diabetogenic T Cells in Genetically Resistant I-E-Transgenic Nonobese Diabetic Mice: Resistance to Autoimmune Diabetes Is Associated with Binding of E-Derived Peptides to the I-A^g7 Molecule¹

Sylvie Trembleau^*, Silvia Gregori^*, Giuseppe Penna^*, Irmina Gorny and Luciano Adorini^2,*

^* Roche Milan Ricerche, Milan, Italy; and Research Institute of Molecular Pathology, Vienna, Austria

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Nonobese diabetic (NOD) and NOD-DR transgenic (tg) mice, expressing A^d:A^g7 and A^d:A^g7 plus DR:E^g7 class II molecules, respectively, both develop insulin-dependent diabetes mellitus (IDDM), whereas NOD-E tg mice expressing A^d:A^g7 plus E:E^g7 are protected. We show that IL-12 administration induces rapid IDDM onset in NOD-DR but fails to provoke insulitis and diabetes in NOD-E tg mice. Nevertheless, T cells from IL-12-treated NOD-E tg mice secrete IFN- and transfer IDDM to NOD-SCID and NOD-E-SCID recipients, demonstrating the presence of peripheral diabetogenic Th1 cells in the protected mice. Surprisingly, regulatory cells were undetectable. Moreover, E:E^g7 could substitute for DR:E^g7 in Ag presentation, arguing against mechanisms of protection involving capture of diabetogenic I-A^g7-restricted epitopes by E:E^g7molecules. Interestingly, the expression of naturally processed epitopes derived from DR- and E-chains bound to I-A^g7 is different in the two strains of tg mice, and the difference is enhanced by IL-12 administration. I-A^g7 molecules from both NOD-DR and NOD-E tg mice present the conserved DR/E 52-68 sequence, at high and low levels, respectively. In addition, only IDDM-resistant NOD-E tg mice possess APCs bearing E65-77/I-A^g7 complexes, which tolerize the specific T cells. This is associated with the selective inhibition of the response to insulinoma-associated protein 2 (IA-2), an autoantigen in IDDM. Our results support protective mechanisms based on I-A^g7 blockade by peptides unique to the E-chain, such as E65-77 and/or tolerance of diabetogenic T cells cross-reactive with E-peptide/I-A^g7 complexes.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
The nonobese diabetic (NOD)³ mouse spontaneously develops insulin-dependent diabetes mellitus (IDDM) that closely resembles the human disease (1). Susceptibility to IDDM, although under complex polygenic control, is primarily associated with genes in the MHC region (2). The class I genes of the NOD MHC consist of K^d and D^b alleles, whereas the class II region encodes an unusual I-A molecule, A^d:A^g7 (I-A^g7) (3, 4). The NOD MHC includes also a characteristic E^g7 gene, but I-E molecules are not expressed on the cell surface due to a deletion in the promoter region of the E gene (3, 5). Both the absence of I-E and homozygous expression of I-A^g7 molecules are necessary for IDDM development (3). Indeed, the introduction of other MHC molecules onto the NOD genetic background causes complete or partial protection from IDDM (6).

NOD-E transgenic (NOD-E tg) mice, which express E:E^g7 together with I-A^g7, fail to develop diabetes either spontaneously or after treatment with cyclophosphamide, an agent accelerating IDDM in wild-type NOD mice (7, 8, 9, 10, 11). The mechanisms underlying this protective effect remain unclear. I-E molecules, either directly or in association with endogenous superantigens, may lead to clonal deletion of specific V-bearing T cells (12, 13), and thereby of pathogenic T cells. However, little I-E-mediated negative selection of V families occurs in the NOD background (10, 12). Unlike originally proposed (14), protection mediated by I-E molecules could not be ascribed to deletion or anergy of T cells expressing specific V chains (10, 15). Nevertheless, deletion of I-A^g7-restricted pathogenic autoreactive T cells by tg class II molecules has been proposed to occur in the thymic medulla, independently of endogenous superantigens, via presentation of a nonautoantigenic peptide to TCR-tg diabetogenic thymocytes (16, 17). Therefore, clonal deletion as a mechanism of protection has not been ruled out.

Peripheral mechanisms hypothesizing interference of I-E molecules with the presentation of diabetogenic peptides by I-A^g7 have also been proposed. The epitope stealing model (18) predicts that I-E molecules bind pathogenic peptides with higher affinity than I-A^g7, thereby inhibiting Ag presentation by the IDDM-associated I-A^g7 molecules. The determinant capture model (19) proposes that I-E molecules bind an epitope neighboring autoantigenic determinants, preventing their generation. In the determinant displacement model (20), peptide(s) from degraded I-E molecules, present at high concentrations in the peptide-loading compartment, displace the diabetogenic peptide without necessarily having much higher affinity for I-A^g7 molecules than the displaced peptide itself. However, none of these mechanisms has so far been validated in the NOD model. Alternatively, mechanisms accounting for I-E-mediated protection from IDDM could involve both thymic and peripheral events, such as the positive selection of I-E-restricted regulatory T cells that could deviate or down-regulate Th1-mediated cell destruction (21).

DR is the human homolog of the mouse E chain (22). In E-negative mice, tg expression of DR or E similarly reconstitutes I-E-dependent T cell repertoire selection and immune responses (23). Nevertheless, quite surprisingly, NOD-DR tg mice develop IDDM with the same incidence and tempo as NOD mice (24).

In the present study, we analyzed the protective events afforded by E:E^g7 molecules. Our initial strategy aimed first at breaking the mechanism of protection with IL-12, a cytokine known to accelerate IDDM in NOD mice (25). IL-12 administration induced IDDM in NOD-DR but not in NOD-E tg mice. However, T cells from IL-12-treated NOD-E tg mice could transfer IDDM to NOD-SCID recipients, demonstrating the presence of pathogenic T cells. A major difference between NOD-DR and NOD-E tg mice was the binding and presentation of naturally processed epitopes from the tg chain by I-A^g7 molecules, highlighting possible mechanisms preventing IDDM development in NOD-E tg mice.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Mice

NOD/Lt and NOD-SCID mice from The Jackson Laboratory (Bar Harbor, ME) were isolator-reared at Charles River Breeding Laboratories (Calco, Italy) and kept under specific pathogen-free conditions in our animal facility. E^k-tg NOD mice (10) were provided by C. Benoist and D. Mathis (Josline Diabetes Center, Boston, MA). Since their original description (10), the tg mice (here, referred to as NOD-E tg) have been backcrossed onto the NOD background for 20 generations. NOD-E-SCID mice were obtained by backcrossing (NOD-E x NOD-SCID)F₁ mice onto the NOD-SCID parental strain. Backcrossed mice were screened for the absence of T and B cells, and for the presence of E molecules on CD11b⁺ cells in the peripheral blood. Homozygous HLA-DRA tg mice on the DBA/2 (H-2^d) background (26) were backcrossed onto the NOD background for over 10 generations. Backcrossed mice were selected for the presence of I-A^g7 (mAb 10.3.62), the presence of DR (mAb L243), and the absence E molecules (mAb 14.4.4S) by cytofluorometric analysis of PBLs. DR tissue distribution in tg mice is superimposable to that of E (26, 27). Transgene hemizygous NOD-DR, NOD-E, NOD-E-SCID mice, and transgene-negative littermates were used. Mice were diagnosed diabetic after two sequential measurements of blood glucose levels higher than 200 mg/dl.

Recombinant mouse IL-12

Recombinant mouse IL-12 was produced in serum-free medium by transfected Chinese hamster ovary cells and purified by sequential chromatography, as described (28). The IL-12 used in this study was >95% pure, as assessed by SDS-PAGE analysis, and the endotoxin content was <5 endotoxin U/mg IL-12, as determined by the Limulus amebocyte assay. IL-12 was diluted in PBS (Life Technologies, Grand Island, NY) containing 100 µg/ml mouse serum albumin (Sigma, St. Louis, MO), and injected in mice i.p. at 7.5 µg/kg daily for 10 or 32 days, as indicated.

Immunohistology

Pancreata were snap-frozen in Tissue Tek (Miles Laboratories, Elkhart, IN), and 5-µm-thick sections were stained with biotinylated GK1.5 anti-CD4, 53-6.7 anti-CD8, RA3-6B2 anti-B220, 14.4.4S anti-E, 17.3.3 anti-E^g7, L243 anti-DR mAbs (all purchased from BD PharMingen, San Diego, CA), or N418 anti-CD11c mAb (American Type Culture Collection, Manassas, VA), followed by streptavidin-peroxidase conjugate. 3-amino-9-ethylcarbazole (DAKO, Carpenteria, CA) was used as chromogen and hematoxylin as a counterstain.

Antigens

Mycobacterium tuberculosis heat shock protein 65-kDa (MT-hsp 65-kDa) was a gift from Dr. R. van der Zee (University of Utrecht, Utrecht, The Netherlands). The MT-hsp 65-kDa peptide 1–16, the hen egg white lysozyme peptide 10–23, and the E and DR peptides 52-68, 62-77, and 65-77 were synthesized by standard Fmoc/tBu chemistry, using side chain protection, coupling procedures, and an automated synthesizer (models 431A and 433A; Applied Biosystems, Foster City, CA). The crude peptides were purified by reversed-phase HPLC, and the sequence was confirmed by amino acid analysis and fast atom bombardment mass spectrometry or matrix assisted laser desorption/ionization time of flight (MALDI TOF, Reflex III; Bruker, Bremen, Germany). Peptide sequences are indicated in Table I. Purified recombinant mouse insulinoma-associated protein 2 (IA-2), corresponding to the intracytoplasmic region (amino acids 601–979) of the protein tyrosine phosphatase-like IA-2 (29), was a gift from Dr. A. Isacchi (Pharmacia-Upjohn, Milan, Italy). IA-2 was purified as previously described (30).

Competitive peptide binding assays to purified I-A^g7 and I-E^g7 molecules were performed as previously described (31, 32).

Isolation of CD4⁺ and CD8⁺ cells

After removal of all visible pancreatic lymph nodes, individual pancreata were digested in 3 ml of HBSS containing 1 mg/ml collagenase IV (Sigma), by shaking (200 rpm) at 37°C for 15 min. Cell suspensions were collected after diluting the enzyme with ice-cold HBSS containing 5% FCS and removing the aggregates by settling for 2 min on ice. Aggregates were further digested with collagenase IV at 0.5 mg/ml for 10 min, and at 0.25 mg/ml for 6 min. The cells obtained in these three rounds of collagenase digestion were pooled, washed three times, and passed over a MACS Pre-Separation Filter (Miltenyi Biotec, Auburn, CA) to remove cell aggregates and clumps. Single cell suspensions, obtained after collagenase digestion of pancreata, were incubated with anti-CD4 and anti-CD8 mAb-coated MicroBeads (Miltenyi Biotec) and applied onto MiniMACS separation columns (Miltenyi Biotec). Alternatively, single cell suspensions from pooled mesenteric and pancreatic lymph nodes or pooled spleens, obtained by mashing them on a nylon filter, were incubated with MicroBeads and applied onto MiniMACS separation columns. This procedure yielded a positive fraction (containing CD4⁺ plus CD8⁺ cells) and a negative fraction (devoid of CD4⁺ plus CD8⁺ cells). Immediately after isolation, CD4⁺ plus CD8⁺ cells from pancreas and/or lymph nodes, or spleen, as indicated, were adoptively transferred by i.v. injection into NOD-SCID or NOD-E-SCID mice. In some cases, when indicated, lymph node APCs (the fraction devoid of CD4⁺ and CD8⁺ cells) were cotransferred with T cells.

Cell surface staining of class II molecules

Single cell suspensions, obtained after collagenase digestion of pancreata, were incubated with anti-CD45 mAb-coated MicroBeads and applied onto MiniMACS separation columns. Then, purified CD45⁺ cells (2 x 10⁵/well) were incubated at 4°C for 20 min with appropriately diluted PE- and FITC-conjugated mAb. Alternatively, PBMC from individual mice were stimulated with IFN- (500 U/ml) and Staphylococcus aureus Cowan I (dilution 1/5000, Pansorbin cells; Calbiochem, San Diego, CA) for 48 h in U-bottom 96-well plates in RPMI 1640 supplemented with 2 mM L-glutamine (Life Technologies), 50 µg/ml gentamicin (Sigma), 50 µM 2-ME (Fluka Biochemica, Buchs, Switzerland), and 10% FCS (complete RPMI medium). Then, the PBMC were washed, incubated for 20 min with FACS medium containing 100 µg/ml mouse IgG (Sigma), and for another 20 min with PE- or FITC-conjugated anti-CD11b (Mac-1) and anti-E (14.4.4S) mAbs. FACS medium used throughout was PBS containing 5% FCS and 0.1% NaN₃. Analysis was performed with a FACScan flow cytometer (BD Biosciences, Mountain View, CA) equipped with CellQuest software.

Intracellular staining for cytokine production

Pancreatic CD8⁺ plus CD4⁺ cells were stimulated immediately after isolation with 50 ng/ml PMA and 750 ng/ml ionomycin (all obtained from Sigma) for 4 h at 37°C in complete RPMI medium, and 10 µg/ml brefeldin A (Novartis, Basel, Switzerland) was added for the last 2 h. Cells were resuspended in PBS containing 10 µg/ml brefeldin A before adding an equal volume of 4% paraformaldehyde. After fixing for 20 min, cells were kept at 4°C and stained the next day for intracellular production of cytokines. Cells were stained for IFN-, IL-4, and IL-10 using the method described by Openshaw (33) and mAbs purchased from BD PharMingen. Reagents for intracytoplasmic staining contained 1% FCS, 0.5% saponin (Sigma), and 0.1% sodium azide, and all incubations were performed at room temperature. Cells were washed, preincubated for 10 min with PBS/FCS/saponin, and then incubated with FITC-labeled rat anti-mouse IFN- (XMG1.2) and PE-labeled rat anti-mouse IL-4 (11B11), or PE-labeled rat anti-mouse IL-10 (JES5-16E3). Isotype controls were FITC- and PE-labeled rat IgG1 (R3-34). After 30 min, cells were washed twice with PBS/FCS/saponin and then with PBS containing 5% FCS without saponin to allow membrane closure. The cell surface was then stained with CyChrome-labeled anti-CD4 (L3T4) for 15 min at room temperature. Analysis was performed with a FACScan flow cytometer equipped with CellQuest software, and 10,000 events were acquired.

T cell proliferation assay

Mice were immunized s.c. in the hind footpads with 1 nmol of peptide hsp 1–16 emulsified in IFA (Difco, Detroit, MI). Alternatively, mice were immunized in the hind footpads with 30 nmol of E52-68, E62-77, or DR62-77 peptide emulsified in CFA containing H37Ra mycobacteria (Difco). Nine days later, popliteal lymph node cells (LNCs, 6 x 10⁵ cells/well) were cultured in 96-well culture plates (Costar, Cambridge, MA) with different concentrations of Ag, as indicated, in synthetic HL-1 medium (Ventrex Laboratories, Portland, ME) supplemented with 2 mM L-glutamine (Life Technologies) and 50 µg/ml gentamicin (Sigma). Cultures were incubated for 3 days in a humidified atmosphere of 5% CO₂ in air and pulsed 8 h before harvesting with 1 µCi [³H]thymidine (Amersham, Arlington Heights, IL). The incorporation of [³H]thymidine into dividing cells was measured by liquid scintillation on a beta counter.

T cell hybridomas

The T cell hybridoma 4H1 was generated by polyethylene glycol-induced fusion of hsp 1–16-immune LNCs from NOD-E tg mice with the TCR -negative variant of the BW5147 thymoma, as previously described (34). Cloned hybridoma cells were selected based on their ability to secrete IL-2 in response to hsp 1–16 peptide presented by irradiated NOD-E splenocytes. The T cell hybridoma TGP1 was generated as described above from E52-68-immune LNCs from NOD mice. T cell hybridomas (5 x 10⁴ cells/well) were incubated with NOD-E or NOD-DR spleen cells (2.5 x 10⁵ cells/well, unless otherwise stated), with or without peptides, in 96-well plates. Culture medium was RPMI 1640 supplemented with 2 mM L-glutamine, 50 µg/ml gentamicin, 50 µM 2-ME, and 10% FCS. After 24 h of culture, 50 µl of supernatant was collected from each well and transferred to culture wells containing 10⁴ IL-2-responsive CTL line (CTLL) cells. During the final 5 h of a 24-h culture, CTLL cells were pulsed with 1 µCi [³H]thymidine. The incorporation of [³H]thymidine into dividing cells was measured by liquid scintillation spectrometry.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
IL-12 administration provokes IDDM in NOD-DR but not in NOD-E tg mice

As reported by Yamane and coworkers (24), we found, although our tg mice were generated differently, that NOD-DR tg mice, in contrast to NOD-E tg mice, spontaneously develop insulitis and diabetes (data not shown). In an attempt to break protection from IDDM in NOD-E tg mice, we administered IL-12, a cytokine that enhances Th1 responses and leads to IDDM acceleration in wild-type NOD mice (25). IL-12 was administered daily for 32 days to NOD-DR and NOD-E tg mice as well as to their negative littermates, starting at 9 wk of age. This treatment induced rapid IDDM development in both NOD-DR and their negative littermates (Fig. 1). In contrast, none of IL-12-treated NOD-E tg mice developed IDDM, although this was rapidly induced in 90% of E-negative littermates (Fig. 1). Histological examination of the pancreas of IL-12-treated NOD-DR tg mice revealed a massive islet infiltration, characterized by abundant CD4⁺, CD8⁺, and CD11c⁺, but few B220⁺ cells (Fig. 2A), similar to the infiltration observed in IL-12-treated negative littermates (data not shown). In both groups, I-A^g7 class II molecules were prominently expressed on islet-infiltrating cells, as well as on cells in the exocrine pancreas (Fig. 2A and data not shown). In addition, both DR and E^g7 chains were highly expressed in NOD-DR tg mice (Fig. 2A). Conversely, all the islets from IL-12-treated NOD-E tg mice were free of insulitis (Fig. 2B). Rare CD4⁺ and a few CD8⁺ cells surrounded some islets, whereas B220⁺ cells were absent (Fig. 2B). I-A^g7 and I-E^g7 molecules were visible in the pancreas of NOD-E but, due to the absence of insulitis, to a far lower extent than in NOD-DR tg mice (Fig. 2B). However, CD4⁺, CD8⁺, and abundant B220⁺ cells were present in the omentum of NOD-E tg mice (Fig. 2B). We also compared the level of class II expression on pancreatic B cells (B220⁺) and dendritic cells (DCs) (CD11c⁺) by flow cytometry (Fig. 2C). B cells from IL-12-treated NOD-E tg mice expressed either low or high levels of I-E^g7 molecules. In contrast, B cells from IL-12-treated DR tg mice homogeneously expressed intermediate levels of I-E^g7 molecules (Fig. 2C). I-A^g7 molecules were similarly expressed in NOD-DR and NOD-E tg mice. Class II MHC molecule expression was similar on lymph node, splenic, and pancreatic B220⁺ cells (data not shown).

View larger version (25K): [in this window] [in a new window]   FIGURE 1. IL-12 administration induces rapid IDDM development in NOD-DR but not in NOD-E tg mice. Female NOD-DR (n = 8, ), female NOD-E tg mice (n = 8, •), and their I-E-negative female littermates (n = 11, , and n = 8, , respectively) were injected i.p. daily for 30 days with 7.5 µg/kg recombinant mouse IL-12 from 9 to 13.5 wk of age. A diagnosis of diabetes was made after two sequential glucose measurements higher than 200 mg/dl. The curves represent the cumulative incidence of IDDM. There was no statistical difference in the time to IDDM development (p = 0.14, by a two-tailed Gehan’s test) nor in the proportion of mice that eventually became diabetic (p = 0.51, by a two-tailed Fisher’s exact test) between NOD-DR tg mice and their negative littermates.

View larger version (95K): [in this window] [in a new window]   FIGURE 2. IL-12 administration induces insulitis in NOD-DR but not in NOD-E tg mice. Pancreatic sections from NOD-DR (A) and NOD-E (B) tg mice, injected with IL-12 for 10 consecutive days, were stained with GK1.5 anti-CD4, 53-6.7 anti-CD8, RA3-6B2 anti-B220, 14.4.4S anti-E, 17.3.3 anti-Eg7, L243 anti-DR, 10.3.62 anti-Ag7, or N418 anti-CD11c mAb as indicated. For NOD-DR tg mice, consecutive sections of two representative islets are shown. For NOD-E tg mice, an islet from the very few surrounded by CD4 and CD8 cells is shown. The other stainings are representative of the whole pancreas and omentum. In addition, the staining of CD4+, CD8+, and B220+ cells in the omentum of IL-12-treated NOD-E tg mice is shown. C, Cytofluorometric analysis of I-Ag7 and I-Eg7 expression on pancreatic APCs. CD45+ cells purified from the pancreas of IL-12-treated NOD-DR, NOD-E tg mice, and their I-E-negative littermates were double-stained with either PE-conjugated anti-B220 or anti-CD11c mAb, and FITC-conjugated anti-E (14.4.4S), anti-DR (L243), anti-Eg7 (17.3.3), or anti-Ag7 (10.3.62) mAbs. Cell surface expression of class II molecules was analyzed on B220+ and CD11c+ gated populations. The black and red lines represent the staining of I-E-negative and I-E-positive littermates, respectively.

Pancreatic CD4⁺ cells from IL-12-treated NOD-DR and NOD-E tg mice display a similar Th1-type phenotype

We next analyzed the cytokine production by pancreatic CD4⁺ cells from DR and E-tg NOD mice. Pancreatic CD4⁺ cells from IL-12-treated mice were isolated, stimulated with PMA plus ionomycin, and stained for intracytoplasmic cytokine production. It should be noted that the purification procedure isolates CD4⁺ cells derived from both the omentum and the pancreas. In IL-12-treated NOD-E, NOD-DR tg, and negative littermates, 30–40% of pancreatic CD4⁺ cells produced IFN-, and among them 10–20% also secreted IL-10 (Fig. 3). In all mice, 7–10% of pancreatic CD4⁺ cells secreted IL-10 but not IFN-, whereas IL-4 was never detectable (Fig. 3). Thus, pancreatic CD4⁺ T cells in IL-12-treated NOD, NOD-E, and NOD-DR tg mice display a similar Th1-dominated phenotype.

View larger version (37K): [in this window] [in a new window]   FIGURE 3. Cytokine production by pancreatic CD4+ cells from IL-12-treated mice. Pancreatic T cells were obtained from a pool of 10–15 NOD-DR, NOD-E tg mice, or I-E-negative littermates (all 8 wk old) injected with IL-12 for 10 consecutive days. T cells (2 x 105 cells/well) were stimulated with PMA and ionomycin for 4 h and analyzed by flow cytometry for IFN- (abscissa) and IL-4 or IL-10 (ordinate) production. Acquisition was performed on CD4+ cells. Percentage of positive cells, set according to the isotype-matched controls (data not shown), are indicated in the top right corner of each quadrant.

Pancreatic T cells from IL-12-treated NOD-E tg mice transfer IDDM to NOD-SCID and NOD-E-SCID recipients

The presence of mononuclear cells within the omentum and the limited peri-insulitis in the pancreas of IL-12-treated-NOD-E tg mice could be consistent with clonal deletion of pathogenic autoreactive T cells. To compare their diabetogenic potential, CD4⁺ and CD8⁺ T cells purified from the pancreas of IL-12-treated NOD-E tg or negative littermates were transferred into NOD-SCID recipients. IDDM developed equally well, within 4–8 wk, upon transfer of T cells from either NOD-E tg or negative littermates, demonstrating the presence of diabetogenic T cells in both groups of mice (Fig. 4). We analyzed the cytokine profile of pancreatic CD4⁺ cells from NOD-E tg mice also after transfer to NOD-SCID recipients. Interestingly, in diabetic NOD-SCID mice transferred 4 wk earlier with T cells from IL-12-treated NOD-E tg mice, pancreatic CD4⁺ cells still secreted IFN- and IL-10 (Fig. 4), and their phenotype was similar to the pancreatic Th1-type CD4⁺ cells before the transfer (Fig. 3). Thus, both protected IL-12-treated NOD-E tg and diabetic NOD-SCID mice after transfer possess pancreatic Th1-like CD4⁺ cells. These results indicate that the mechanism(s) of protection in IL-12-treated NOD-E tg mice is unlikely to reflect a deviation of pancreatic CD4⁺ cells from the Th1 to the Th2 phenotype.

View larger version (23K): [in this window] [in a new window]   FIGURE 4. Pancreatic T cells from IL-12-treated NOD-E tg mice transfer IDDM to NOD-SCID recipients. CD4+ plus CD8+ cells were purified from the pancreas of 8-wk-old NOD-E tg mice or their negative littermates (pool of 10–15 mice/group). These cells were injected i.v. into NOD-SCID recipients (0.5 x 106 cells/mouse). The cumulative IDDM incidence in 11 and 8 NOD-SCID recipients, adoptively transferred with T cells from NOD-E or from their negative littermates (indicated as NOD), is shown in the left panel. Thirty days after adoptive transfer of T cells from NOD-E tg mice, the pancreatic CD4+ cells from three recently diabetic NOD-SCID mice were purified and pooled. These CD4+ cells were stimulated with PMA and ionomycin for 4 h and analyzed by flow cytometry for IFN- (abscissa) and IL-10 (ordinate) production (right panel). Percentages of positive cells, set according to the isotype-matched control (data not shown), are indicated in the top right corner of the quadrant.

View larger version (26K): [in this window] [in a new window]   FIGURE 5. Pancreatic T cells from IL-12-treated NOD-E tg mice transfer IDDM to NOD-E-SCID recipients. Upper panels, I-E expression on activated macrophages. PBMC from individual mice were stimulated with IFN- (500 U/ml) and Staphylococcus aureus Cowan I (1/5000) for 48 h in U-bottom 96-well plates in complete RPMI medium. Then, the PBMC were stained with PE- or FITC-conjugated anti-CD11b (Mac-1) and anti-E (14.4.4S) mAbs. Cell surface expression of the E-chain was analyzed on the CD11b+ gated population. Lower panel, CD4+ plus CD8+ cells purified from the pancreas of 8-wk-old NOD-E tg mice (pool of 10–15 mice/group) were injected i.v. into either NOD-E-SCID mice or their I-E-negative NOD-SCID littermates (0.5 x 106 cells/mouse). There was no statistical difference in the time to develop IDDM (p = 0.64, by a two-tailed Gehan’s test) between NOD-SCID and NOD-E-SCID mice following transfer of T cells from NOD-E tg mice.

The diabetogenic potential of T cells from NOD-E tg mice is expressed following IL-12 administration

We next compared the pathogenic potential of T cells isolated from various organs of untreated or IL-12-treated mice. Splenic and lymph node T cells from untreated prediabetic NOD mice could transfer IDDM in NOD-SCID recipients, whereas T cells from untreated NOD-E tg mice could not, showing a total absence of pathogenicity (Table II, rows 1 and 3 vs 2 and 4). However, NOD lymph node T cells transferred diabetes with a significant delay (p = 0.003) and a much lower frequency (17 vs 100%, p = 0.004) compared with their splenic counterpart (Table II, rows 1 vs 3). This defect could be corrected by IL-12 administration, because lymph node T cells from IL-12-treated NOD mice induced IDDM rapidly (p = 0.0007) in 100% (instead of 17%, p = 0.004) of NOD-SCID recipients (Table II, rows 3 vs 5). Nevertheless, these lymph node T cells transferred disease with a significant delay (p = 0.036) of 2 wk, compared with pancreatic T cells from the same mice (Table II, rows 5 vs 7). In contrast, as mentioned above, both splenic and lymph node T cells from untreated NOD-E tg mice remained nonpathogenic after adoptive transfer into NOD-SCID recipients. Only the splenocytes and LNCs of these mice could be tested, because without IL-12 treatment there is no infiltration of T cells in the pancreas and the omentum. Intriguingly, T cells from IL-12-treated NOD-E tg mice transferred IDDM. Pancreatic T cells were as efficient as cells from IL-12-treated NOD littermates, whereas lymph node T cells transferred disease with a significant delay and a lower frequency of diabetes (Table II, row 8 vs 7 and row 6 vs 5). In summary, IL-12 administration does not break protection from IDDM in NOD-E tg mice (Fig. 1), but renders NOD-E T cells pathogenic when transferred into NOD-SCID recipients (Table II). In addition, lymph node T cells are generally less diabetogenic than the pancreatic T cells of the same mice, and this is more pronounced for IL-12-treated NOD-E tg mice. This observation of low frequency and/or aggressiveness of diabetogenic cells in lymph nodes suggests either that the pathogenic cells preferentially migrate and localize into other organs such as pancreas or spleen, or that specific mechanism(s) of tolerance occur in lymph nodes.

View this table: [in this window] [in a new window]   Table II. Comparison of the diabetogenic potential of T cells from untreated or IL-12-treated NOD-E tg mice and negative littermates

Next, to study the role of APCs in the protective mechanism(s), we cotransferred T cell-depleted LNCs together with T cells into NOD-SCID mice. The transfer of pancreatic T cells alone or together with lymph node APCs from IL-12-treated NOD-E tg donors induced a similar IDDM incidence in NOD-SCID recipients (Table II, rows 8 and 13). Cotransfer of lymph node APCs and T cells from NOD-E tg mice was less efficient than lymph node T cells alone in inducing IDDM in NOD-SCID recipients, but this difference did not reach statistical significance (Table II, rows 6 and 11). Thus, lymph node APCs are unable to interfere with the pathogenicity of pancreatic T cells, although they appear to further decrease the already low pathogenic potential of lymph node T cells.

Similar Ag processing and presentation by NOD-DR and NOD-E APCs

To further examine a possible role of APCs in the protection from IDDM, we compared Ag presentation by NOD-DR and NOD-E APCs, using as readout the activation of peptide-specific T cells. The I-E^g7-restricted 4H1 hybridoma recognizes the hsp peptide 1–16 and the core sequence hsp 4–13 (32). These peptides bind well to I-E^g7 but poorly to I-A^g7 and represent the immunodominant epitope of 65-kDa hsp presented by I-E^g7 molecules (32). Spleen cells from either NOD-DR or NOD-E tg mice presented equally well the peptide hsp 1–16 to the 4H1 T cell hybridoma (Fig. 6A). The shorter peptide hsp 4–13 was also presented similarly by DR:E^g7 or E:E^g7 molecules (Fig. 6A). Next, we compared the capacity of APCs to process the 65-kDa hsp. Spleen cells from either NOD-DR or NOD-E tg mice incubated with this protein could generate epitope(s) able to similarly activate the hsp 1–16-specific 4H1 hybridoma (Fig. 6B). In addition, specific T cells could be primed in vivo equally well by either hsp 1–16/DR:E^g7 or hsp 1–16/E:E^g7 complexes because LNCs from hsp 1–16-primed NOD-DR and NOD-E tg mice responded similarly to in vitro restimulation with this peptide (Fig. 6C). In conclusion, APCs from NOD-DR and NOD-E tg mice can activate I-E^g7-restricted T cells similarly, although they show different levels of I-E^g7 molecule expression (Fig. 2C). Thus, as in other haplotypes (23), DR:E^g7 molecules can substitute for E:E^g7 dimers in Ag presentation to T cells. This result, together with the ones described above (Fig. 5 and Table II, row 13), suggests that APCs are not responsible for IDDM protection via a direct involvement of E:E^g7 molecules.

View larger version (31K): [in this window] [in a new window]   FIGURE 6. Processing and presentation of MT-hsp 65 kDa by APCs from NOD-DR and NOD-E tg mice. A and B, T cell hybridoma 4H1 (5 x 104 cells/well) was incubated with the indicated concentrations of hsp peptide (A) or hsp protein (B) and 2.5 x 105 splenocytes from NOD-DR and NOD-E tg mice. After 24 h of culture, 50 µl of supernatant was removed and tested for IL-2 using CTLL cells. Data are presented as mean cpm of [3H]thymidine incorporation into CTLL cells from duplicate initial cultures. C, NOD-E () and NOD-DR tg mice (•) were immunized in the hind footpads with 1 nmol/mouse of hsp peptide 1–16 in IFA. Nine days later, popliteal LNCs were cultured with the indicated concentrations of peptide. Data are expressed as mean cpm of [3H]thymidine incorporation from duplicate cultures with background values (3000 cpm) subtracted. Each curve represents the mean response of two individual mice. LNCs from immunized E- and DR-negative littermates did not proliferate when cultured with the hsp 1–16 peptide (data not shown).

The peptide 52-68 from tg DR- and E-chains is endogenously processed and presented by I-A^g7 molecules in both NOD-DR and NOD-E tg mice

Naturally processed class II-derived peptides bound to self class II molecules might play a pathogenic (35, 36) or a protective (36, 37) role in autoimmunity. Intriguingly, E52-68 bound to I-A^b molecules has been associated with protection against autoimmune lupus (37, 38). Thus, to address the possible role of transgene-derived peptides in the protection against IDDM, we first examined the expression of the E52-68 epitope by APCs from NOD-E and NOD-DR tg mice. Importantly, the E52-68 sequence is identical in E- and DR-chains (22). To assess the constitutive expression of E52-68 in vivo, NOD-DR, NOD-E tg, and wild-type littermates were immunized in the hind footpads with the synthetic E/DR52-68 peptide emulsified in CFA. Nine days later, immune LNCs were restimulated in vitro with different concentrations of the E/DR52-68 peptide. LNCs from wild-type NOD mice proliferated well in response to this peptide, demonstrating that E/DR52-68 presented by I-A^g7 is immunogenic (Fig. 7). In contrast, LNCs from either NOD-E or NOD-DR tg mice failed to proliferate, indicating T cell tolerance to this epitope (Fig. 7). These results suggest that E/DR52-68 is generated from the tg chains by APCs of both NOD-DR and NOD-E tg mice.

View larger version (22K): [in this window] [in a new window]   FIGURE 7. Immunogenicity of E/DR52-68 in NOD-E, NOD-DR tg mice, and their I-E-negative littermates. NOD-E (•), NOD-DR tg mice (), and their I-E-negative littermates ( and , respectively) were immunized into the hind footpads with 30 nmol/mouse of peptide E/DR52-68 in CFA. Nine days later, popliteal LNCs were cultured with the indicated concentrations of peptide. Data are expressed as mean cpm of [3H]thymidine incorporation from duplicate cultures. Each curve represents the mean (±SE) response from five individual mice.

Comparison of E/DR52-68:I-A^g7 complex levels on splenocytes from NOD-DR and NOD-E tg mice

To quantify E/DR52-68:I-A^g7 complexes on APCs from NOD-E and NOD-DR tg mice, we generated from E/DR52-68-primed NOD mice a T cell hybridoma (TGP1) recognizing the naturally processed E/DR52-68 epitope bound to I-A^g7 molecules. This readout system allows quantification of naturally processed peptides bound to class II MHC molecules (39). TGP1 was activated by the synthetic E/DR52-68 peptide presented by spleen cells from NOD, NOD-E, and NOD-DR tg mice (Fig. 8A). In the absence of synthetic peptide, TGP1 was activated by spleen cells from NOD-E and NOD-DR tg mice but not by NOD spleen cells, indicating recognition of the naturally processed E/DR52-68 epitope bound to I-A^g7 molecules (Fig. 8B). Ab blocking experiments confirmed the restriction by I-A^g7 molecules (data not shown). These results demonstrate that the E/DR52-68 epitope represents a naturally processed determinant of the E/DR-chain, expressed in vivo on APCs from I-E-tg NOD mice and presented by I-A^g7 molecules. Surprisingly, splenocytes from NOD-E mice were much less efficient than splenocytes from NOD-DR tg mice in activating TGP1 (Fig. 8B). This was not due to an intrinsic defect of the APCs, because splenocytes from NOD-DR and NOD-E tg mice, as well as from I-E-negative littermates, could all activate TGP1 similarly when cultured with the synthetic E/DR52-68 peptide (Fig. 8A). Therefore, these results indicate that fewer E/DR52-68:I-A^g7 complexes are expressed on splenocytes from NOD-E than NOD-DR tg mice. In addition, splenocytes from IL-12-treated compared with untreated NOD-DR tg mice stimulated TGP1 slightly better, whereas splenocytes from IL-12-treated NOD-E did not activate at all the T cell hybridoma (Fig. 8C). Thus, IL-12 administration abrogates the already low capacity of NOD-E splenocytes to present the naturally processed endogenous self-epitope E52-68 bound to I-A^g7 molecules.

View larger version (40K): [in this window] [in a new window]   FIGURE 8. Expression of self E/DR52-68: I-Ag7 complexes on splenocytes from NOD-E and NOD-DR tg mice. A, Splenocytes from NOD-DR, NOD-E tg mice, and I-E-negative littermates activate the E/DR52-68:I-Ag7-specific T cell hybridoma TGP1 similarly when cultured with the synthetic peptide E/DR52-68. TGP1 (5 x 104 cells/well) was cultured with spleen cells (3 x 104 cells/well) from NOD-DR (), NOD-E (•) tg mice, or their I-E-negative littermates ( and , respectively) and different concentrations of the synthetic peptide E/DR52-68. B, Splenocytes from NOD-DR but not from NOD-E tg mice efficiently activate TGP1 cells. Graded numbers of the same splenocytes described in A were cultured with TGP1 (5 x 105 cells/well) without peptide. C, Splenocytes from IL-12-treated NOD-E tg mice fail to activate the TGP1 T cell hybridoma. TGP1 (5 x 104 cells/well) was cultured with spleen cells (5 x 105 cells/well) from NOD-DR or NOD-E tg mice, either untreated or previously injected with IL-12 for 10 consecutive days. In all cases, after 24 h of culture, 50 µl of supernatant was removed and tested for IL-2 using CTLL cells. Data are presented as mean cpm of [3H]thymidine incorporation into CTLL cells from duplicate initial cultures. D, DCs and B cells from NOD-DR but not from NOD-E tg mice efficiently activate the E/DR52-68:I-Ag7-specific T cell hybridoma. TGP1 (5 x 104 cells/well) was cultured with DCs or B cells, purified from the spleen of NOD-DR or NOD-E tg mice, as indicated. The highest number of DCs tested was 3 x 104 cells/well. After 24 h of culture, 50 µl of culture supernatant was removed and tested for IL-2 using CTLL cells. Splenic DCs and B cells obtained from DR- or E-negative littermates did not activate TGP1 cells (data not shown). Data are presented as mean cpm of [3H]thymidine incorporation into CTLL cells from triplicate initial cultures.

Generalized defect of E52-68:I-A^g7 complex expression on different APCs from NOD-E tg mice

APC-enriched cells from either lymph nodes or spleen from E compared with DR tg mice were found defective in their capacity to stimulate TGP1 cells, reflecting an overall lower expression of E52-68:I-A^g7 complexes in NOD-E tg mice (data not shown). Because DCs are the most efficient APCs in presenting endogenous naturally processed self-epitopes to class II-restricted T cells (40), we also tested the capacity of DCs and B cells from NOD-DR and NOD-E tg mice to stimulate TGP1. As expected, DCs activated TGP1 much more efficiently than B cells from the same mice (Fig. 8D). Strikingly, both DCs and B cells from NOD-E tg mice were nearly unable to activate the TGP1 cell hybridoma (Fig. 8D). Thus, NOD-E compared with NOD-DR tg mice express much lower levels of E/DR52-68:I-A^g7 complexes on different APC types and in different lymphoid tissues, indicating a generalized defect in the expression of this endogenous naturally processed T cell epitope complexed with I-A^g7 molecules. In conclusion, the diabetes-susceptible NOD-DR tg mice display higher levels of E/DR52-68:I-A^g7 complexes on both B cells and DCs than the diabetes-resistant NOD-E tg mice, indicating that the presence of E/DR52-68 peptide is not associated with IDDM protection.

IDDM-resistant NOD-E but not IDDM-sensitive NOD-DR tg mice express E65-77/I-A^g7 complexes

The low number of E52-68/I-A^g7 complexes on APC from NOD-E tg mice could be due to the preferential generation of a neighboring epitope in a mechanism called determinant capture (19). To test this hypothesis, we selected, based on their binding motifs (31, 32, 41), peptides from E and DR molecules adjacent to the 52-68 sequence, and analyzed their binding to purified I-A^g7 and I-E^g7 molecules. We found some good binders to I-E^g7 but, upon immunization of either NOD-E or NOD-DR tg mice, these peptides induced vigorous LNC proliferation, suggesting that they represent cryptic epitopes (data not shown). Interestingly, we found that E65-77 binds to I-A^g7 and not to I-E^g7, whereas the corresponding sequence in the DR-chain, DR65-77, does not bind to either of these class II molecules (Table I). LNCs from primed wild-type NOD mice proliferated strongly in response to E65-77, demonstrating that this peptide presented by I-A^g7 is highly immunogenic (Fig. 9). In contrast, LNCs from NOD-E tg mice failed to proliferate, indicating T cell tolerance to this epitope (Fig. 9). This suggests that E65-77 is generated from the tg chain by APCs from NOD-E tg mice. The DR65-77 peptide was weakly immunogenic in NOD mice (Fig. 9), likely due to its poor binding to I-A^g7 molecules. Indeed, this peptide was not able to compete for binding to purified I-A^g7 molecules (Table I). LNCs from primed NOD and NOD-DR mice proliferated similarly in response to DR65-77 peptide. Thus, DR65-77 is unable to induce specific T cell tolerance in NOD-DR tg mice, possibly because it is not naturally generated (Fig. 9). In summary, APCs from NOD-E tg mice express E65-77/I-A^g7 as well as low levels of E/DR52-68:I-A^g7 complexes. In contrast, NOD-DR tg mice express E/DR52-68:I-A^g7 only. The E65-77/I-A^g7 complexes, present exclusively in IDDM-resistant-NOD-E tg mice, tolerize the specific T cells. Therefore, not only the endogenous peptides bound to I-A^g7 but also the T cell repertoires are different in NOD-E compared with NOD-DR tg mice.

View larger version (22K): [in this window] [in a new window]   FIGURE 9. Immunogenicity of E65-77 and DR65-77 peptides in NOD-E and NOD-DR tg mice. Popliteal LNCs from E62-77-primed NOD-E mice, DR62-77-primed NOD-DR tg mice, and E62-77 or DR62-77-primed I-E-negative littermates were cultured with various concentrations of E65-77 or DR65-77 peptides, as indicated. Data are expressed as mean cpm of [3H]thymidine incorporation from duplicate cultures. Each curve represents the mean response from two to three individual mice.

Reduced response to the IDDM-associated autoantigen IA-2 in NOD-E tg mice

Similar I-A^g7-restricted T cell responses are induced in NOD and NOD-E tg mice following priming with foreign Ags emulsified in CFA, indicating that under these conditions the I-E molecule does not interfere with the presentation of I-A^g7-restricted T cell epitopes (Ref. 19 and data not shown). To determine whether the expression of the I-E molecule could affect the response to an IDDM-associated autoantigen in unprimed mice, we analyzed T cell responses to IA-2, a phosphatase-like autoantigen inducing Th1 cell responses associated with IDDM pathogenesis in the NOD mouse (30). Splenocytes from unprimed, naive NOD-DR tg and NOD mice produced similar amounts of IFN- in response to IA-2. Conversely, spleen cells from NOD-E tg mice, compared with cells from IDDM-susceptible mice, secreted significantly less IFN- when cultured with this autoantigen (Fig. 10). Thus, tg expression of E but not DR molecules leads to a decreased Th1 response to a self-Ag that induces, in the NOD mouse, an early T cell response implicated in the pathogenesis of IDDM. In conclusion, only IDDM-resistant NOD-E tg mice possess APCs bearing E65-77/I-A^g7 complexes that tolerize the specific T cells. This is associated with the selective inhibition of the response to the autoantigen IA-2, highlighting possible mechanisms preventing IDDM in NOD-E tg mice.

View larger version (25K): [in this window] [in a new window]   FIGURE 10. IA-2-induced IFN- secretion by spleen cells from NOD-DR mice, NOD-E mice, and I-E-negative littermates. Spleen cells (106/well) from 8-wk-old unprimed female mice were cultured with various concentrations of recombinant mouse IA-2, as indicated. After 48 h of culture, IFN- secretion was determined in culture supernatants by ELISA. The curves represent the response from a pool of three mice per group (left panel). The bars represent the mean (±SE) IFN- production of spleen cells from three individual mice.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The protection from IDDM is unrelated to the capacity of NOD-DR and NOD-E APCs to process and present Ag to I-E^g7-restricted T cells. Indeed, APCs from these two strains were similar in the presentation of I-E^g7-restricted hsp epitopes and in the presentation of I-E^g7-restricted E or DR-derived peptides. This is consistent with the predicted location, outside the peptide-binding groove, of the 17 residues in the first external domain differing between E and DR chains, except the conservative substitution E Val for DR Ile at position 72 (26, 42). Thus, the mechanism of protection in NOD-E tg mice can neither be "stealing" of the diabetogenic epitope(s) by E:E^g7 nor "capture" by E:E^g7 of neighboring determinants preventing the generation of diabetogenic epitope/I-A^g7 complexes. Otherwise, the DR:E^g7 molecule would have been expected to act similarly and protect against IDDM.

Specific T cells might not always recognize similarly peptides derived from the tg chains bound to DR:E^g7 and E:E^g7, because of substitutions interfering with the TCR contact residues. For example, the Glu in E for Lys in DR at position 75 replaces an acidic with a basic residue at a putative TCR contact site (26, 42). We cannot rule out subtle differences between NOD-DR and NOD-E tg mice in I-E^g7-mediated positive and negative selection of T cells. However, the only important difference we detected in the T cell repertoire of these two strains was induced by tg class II-derived peptides and was I-A^g7-restricted. Thus, we found tolerance to E65-77/I-A^g7 complexes in NOD-E, which could not occur in NOD-DR tg mice because the DR65-77 sequence presents substitutions precluding efficient binding to I-A^g7 molecules. Thymic deletion of autoreactive T cells as a mechanism accounting for protection from IDDM in NOD-E tg mice has been extensively debated. This possibility was initially proposed (14), then dismissed (10, 15, 43), and more recently reproposed (16, 17). Using NOD mice tg for a diabetogenic TCR and class II molecules, clonal deletion of diabetogenic T cells has been proposed to result from the presentation in the thymic medulla of a nonautoantigenic peptide (16, 17). The nonautoantigenic peptide(s) have not been identified, but we suggest that class II-derived peptides could be good candidates. Moreover, a similar mechanism could apply to the protection seen in NOD-E tg mice if the deleted or anergic E65-77/I-A^g7-specific T cells include a subset of highly diabetogenic T cells. In this case, the frequency of diabetogenic T cells would be sufficiently reduced in NOD-E tg to prevent IDDM development, as compared with NOD-DR tg and wild-type NOD mice. However, the NOD mouse has a diverse T cell repertoire with multiple potentially diabetogenic cells as compared with a TCR-tg mouse. Thus, even if the putative highly diabetogenic E65-77/I-A^g7-specific T cells were deleted in NOD-E tg mice, diabetogenic T cells with other specificities could still expand sufficiently, upon IL-12 treatment and transfer into NOD-SCID recipients, to cause IDDM.

Alternatively, the putative highly diabetogenic E65-77/I-A^g7-specific T cells are not deleted but actively tolerized in NOD-E tg mice. In this case, exogenous IL-12 could partially break the tolerogenic mechanism, as demonstrated in models of Ag-induced Th1 cell hyporesponsiveness (44). We have previously shown that IL-12 administration in NOD mice induces a high number of Th1 cells in the pancreas and accelerates IDDM (25), suggesting that this cytokine may increase the frequency and/or pathogenicity of diabetogenic Th1 cells. As described in the present study, pancreatic T cells from IL-12-treated NOD-E tg and NOD mice transfer IDDM to NOD-SCID recipients with a similar efficiency. Altogether, the adoptive transfer in immunocompromised hosts of pancreas/omentum-derived cells from IL-12-treated NOD-E tg mice may create conditions that synergize to increase sufficiently the frequency/aggressiveness of the diabetogenic T cells present in NOD-E tg mice, bypassing the mechanism of protection. This would also explain IDDM development upon transfer into NOD-E-SCID recipients, which express E:E^g7 molecules. This hypothesis is further supported by our observation that lymph node T cells and especially those from IL-12-treated NOD-E tg mice are very inefficient in transferring IDDM into NOD-SCID recipients. Lymph nodes possess a highly diverse T cell repertoire, and it is conceivable that, upon IL-12 administration and adoptive transfer, cells with multiple specificities could dilute potentially diabetogenic T cells.

We found two epitopes naturally processed from the E-chain in NOD-E tg mice, E52-68 and E65-77. Although most proteins possess usually only a few dominant epitopes presented by a given class II molecule (45), we cannot exclude that additional E epitopes, other than the two identified, are presented by I-A^g7 in NOD-E but not in NOD-DR tg mice. However, this would only strengthen the mechanisms described here for E65-77/I-A^g7 by modifying even more deeply the T cell repertoire in NOD-E tg mice. Thus, our study allows us to consider several mechanisms of protection mediated by I-A^g7 rather than I-E^g7 molecules. Some level of clonal deletion mediated by E65-77/I-A^g7 in NOD-E tg mice may occur in the thymus because E-derived peptides are usually expressed in this organ (46, 47). In addition, partial T cell deletion or anergy by E65-77/I-A^g7 complexes may occur peripherally (48). Moreover, alternative peripheral mechanisms can play a role. For example, E65-77 may compete for peptide binding to I-A^g7 and decrease the presentation of diabetogenic epitope(s) (49, 50). This implies that E65-77 competes more effectively than the E/DR52-68 peptide in NOD-DR tg mice. However, the E65-77 peptide could also inhibit the activation of diabetogenic T cells without interfering with the MHC-peptide-TCR pathway. E65-77 is highly homologous to rat and human class II-derived peptides, which have been shown to induce apoptosis or block cell division of activated T cells, behaving like potent immunosuppressive agents such as rapamycin (51, 52). Consistent with this possibility, the pathogenicity of NOD-E lymph node, albeit not pancreatic, T cells could be decreased by the cotransfer of NOD-E APCs, which are a potential source of the E65-77 peptide.

The susceptibility and resistance to IDDM development in NOD-DR and NOD-E tg mice is paralleled by the high and low response to the intracytoplasmic domain of the IA-2 autoantigen. A mechanistic explanation could be I-A^g7 blockade by the E65-77 peptide, lowering the response to IA-2 epitopes. However, the T cell tolerance to E65-77 could also affect the development of diabetogenic T cells. It would be interesting to determine whether the I-A^g7-restricted T cell response to E65-77 can cross-react with some IA-2 determinants. The E65-77 peptide does not present similarities to any IA-2 sequence, but cross-reactivity at the TCR level of apparently nonhomologous peptides has already been described (53). Alternatively, E65-77 may present some molecular mimicry with an autoantigen inducing a T cell response earlier than IA-2. In this case, the tolerance to this putative autoantigen would decrease the spreading of the response to other diabetogenic Ags such as IA-2. It should be noted that the response to IA-2 is profoundly decreased but not totally absent in NOD-E tg mice. Therefore, IL-12 administration may expand and activate IA-2-specific T cells in these mice, as we previously reported in NOD mice (30).

Disease inhibition mediated by an I-E transgene is usually complete (9). In agreement with this observation, we show here that even purified T cells from untreated NOD-E tg mice are unable to transfer IDDM to NOD-SCID recipients. This contrasts with results obtained with I-A transgenes, which confer a more limited protection (9, 54, 55, 56, 57, 58). Nevertheless, the introduction of I-E molecules on the NOD background by breeding with MHC congenic mice, or by reducing the copy number of the transgene, may lead to some degree of insulitis and to diabetes in a small fraction of aged mice (21, 59). It was suggested that the quantitative level of I-E expression is crucial to induce resistance to IDDM. However, the interpretation of these results is difficult because the level of I-E molecule expression correlates with the amount of E peptides generated (S. Trembleau, S. Gregori, G. Penna, and L. Adorini, unpublished observations). Moreover, the breeding of NOD with MHC congenic mice introduces not only an E-chain but also additional E and I-A alleles, which may interfere with the disease. Interestingly, the naturally processed E52-68 epitope bound to I-A^b molecules has been associated with protection against autoimmune lupus in the BXSB mouse background (37, 38). However, the exact mechanism has not been clarified. In our case, the diabetes-susceptible NOD-DR tg mice display higher levels of E/DR52-68:I-A^g7 complexes on both DCs and B cells than diabetes-resistant NOD-E tg mice, indicating that E/DR52-68 expression is not associated with protection from diabetes. The absence of insulitis and IDDM in NOD-E tg mice may instead be explained by the unique expression of E65-77/I-A^g7 complexes. Protection does not appear to involve a deviation of pancreatic T cells from the Th1 to the Th2 phenotype, as previously documented by the association of pancreatic Th2 cells and protection from IDDM in the NOD mouse treated with IL-12 antagonists (60, 61). Although some studies have reported a shift to the Th2 response associated with the presence of protective class II molecules (21, 62), this may not necessarily be the cause of their disease resistance (63). Indeed, we found a similar Th1 phenotype expressed by pancreatic CD4⁺ cells from IL-12-treated NOD-E tg mice, all completely protected from insulitis and diabetes, and NOD-SCID recipients, which all developed IDDM upon cell transfer. In addition, NOD-E tg mice, which are protected from IDDM, exhibit enhanced IFN- production in response to the proteolipoprotein epitope 56–70 and exacerbated experimental autoimmune encephalomyelitis upon immunization with proteolipoprotein, as compared with the NOD mouse (64). This shows that NOD-E tg mice are also prone to develop Th1-like responses. Importantly, it also proves that E:E^g7 expression is not associated with a general protection against autoimmune diseases in the NOD background.

In conclusion, our results show that diabetogenic T cells are present in E tg NOD mice and can be revealed by IL-12 administration and transfer into NOD-SCID recipients. In addition, our data point to a mechanism of protection based on binding of E-derived peptides to the I-A^g7 molecule, rather than on the Ag-presenting capacity of the E:E^g7 molecule. Several possible mechanisms, not mutually exclusive, associated with the expression of E65-77/I-A^g7 complexes could explain diabetes protection in NOD-E tg mice. First, blockade of I-A^g7 molecules by E65-77 may lower the presentation of diabetogenic epitope(s). Second, an active inhibitory mechanism mediated by the E65-77 peptide, independently of the MHC and TCR, may induce the inactivation or apoptosis of recently activated diabetogenic T cells. Third, tolerance either via partial deletion and/or anergy of a subset of T cells recognizing E65-77/I-A^g7 complexes may lower the frequency of diabetogenic T cells. In any case, diabetogenic T cells are not completely deleted in NOD-E tg mice and can be expanded and/or rendered pathogenic by IL-12 administration and adoptive transfer. This study illustrates how only few amino acid differences in a class II molecule not associated with disease induction can affect the development of class II-associated autoimmune diseases, highlighting the role of class II polymorphism in the pathogenesis of human autoimmune diseases. Thus, in the presence of disease-associated class II molecules, polymorphisms in other class II (and possibly class I) molecules expressed by a given individual may lead to differential generation of MHC-derived peptides, modification of the T cell repertoire, and subsequently to a different disease outcome.

	Footnotes

 
¹ This work was supported in part by European Community Contract BIO4-CT96-0077.

² Address correspondence and reprint requests to Dr. Luciano Adorini, Roche Milano Ricerche, Via Olgettina 58, I-20132 Milan, Italy. E-mail address: luciano.adorini{at}roche.com

³ Abbreviations used in this paper: NOD, nonobese diabetic; hsp, heat shock protein; MT-hsp 65-kDa, Mycobacterium tuberculosis hsp 65-kDa; IDDM, insulin-dependent diabetes mellitus; tg, transgenic; CTLL, CTL line; DC, dendritic cell; IA-2, insulinoma-associated protein 2; LNC, lymph node cell.

Received for publication February 14, 2001. Accepted for publication August 6, 2001.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Atkinson, M. A., E. H. Leiter. 1999. The NOD mouse model of type 1 diabetes: as good as it gets?. Nat. Med. 5:601.[Medline]
2. Wicker, L. S., J. A. Todd, L. B. Peterson. 1995. Genetic control of autoimmune diabetes in the NOD mouse. Annu. Rev. Immunol. 13:179.[Medline]
3. Hattori, M., J. B. Buse, R. A. Jackson, L. Glimcher, M. E. Dorf, M. Minami, S. Makino, K. Moriwaki, H. Kuzuya, H. Imura, et al 1986. The NOD mouse: recessive diabetogenic gene in the major histocompatibility complex. Science 231:733.[Abstract/Free Full Text]
4. Acha-Orbea, H., H. O. McDevitt. 1987. The first external domain of the non obese diabetic mouse class II I-A chain is unique. Proc. Natl. Acad. Sci. USA 84:2435.[Abstract/Free Full Text]
5. Acha-Orbea, H., L. Scarpellino. 1991. Non-obese diabetic and non-obese non-diabetic mice have unique MHC class II haplotypes. Immunogenetics 34:57.[Medline]
6. Kikutani, H., S. Makino. 1992. The murine autoimmune diabetes model: NOD and related strains. Adv. Immunol. 51:285.[Medline]
7. Nishimoto, H., H. Kikutani, K. I. Yamamura, T. Kishimoto. 1987. Prevention of autoimmune insulitis by expression of I-E molecules in NOD mice. Nature 329:432.
8. Uehira, M., M. Onu, J. Miyazaki, H. Nishimoto, T. Kishimoto, K. Yamamura. 1989. Development of autoimmune insulitis is prevented in E^d but not in A^k NOD transgenic mice. Int. Immunol. 1:209.[Abstract/Free Full Text]
9. Lund, T., L. A. O’Reilly, P. R. Hutchings, O. Kanagawa, E. Simpson, R. Gravely, P. J. Chandler, P. J. Dyson, J. K. Picard, A. Edwards, et al 1990. Prevention of insulin-dependent diabetes mellitus in non-obese diabetic mice by transgenes encoding modified I-A -chain or normal I-E -chain. Nature 345:727.[Medline]
10. Boehme, J., B. Schuhbaur, O. Kanagawa, C. Benoist, D. Mathis. 1990. MHC-linked protection from diabetes dissociated from clonal deletion of T cells. Science 249:293.[Abstract/Free Full Text]
11. Uno, M., T. Miyazaki, M. Uehira, H. Nishimoto, M. Kimoto, J. Miyazaki, K. Yamamura. 1992. Complete prevention of diabetes in transgenic NOD mice expressing I-E molecules. Immunol. Lett. 31:47.[Medline]
12. Tomonari, K.. 1990. Linkage between Tcrb-V and a gene responsible for deletion of Tcrb-V11⁺ T cells. Immunogenetics 32:60.[Medline]
13. Bill, J., O. Kanagawa, J. Linten, Y. Utsunomiya, E. Palmer. 1990. Class I and class II MHC gene products differentially affect the fate of V5 bearing thymocytes. J. Mol. Cell. Immunol. 4:269.[Medline]
14. Reich, E. P., R. S. Sherwin, O. Kanagawa, Jr C. A. Janeway. 1989. An explanation for the protective effect of the MHC class II I-E molecule in murine diabetes. Nature 341:326.[Medline]
15. Parish, N. M., P. Chandler, R. Quartey-Papafio, E. Simpson, A. Cooke. 1993. The effect of bone marrow and thymus chimerism between non-obese diabetic (NOD) and NOD-E transgenic mice, on the expression and prevention of diabetes. Eur. J. Immunol. 23:2667.[Medline]
16. Schmidt, D., J. Verdaguer, N. Averill, P. Santamaria. 1997. A mechanism for the major histocompatibility complex-linked resistance to autoimmunity. J. Exp. Med. 186:1059.[Abstract/Free Full Text]
17. Schmidt, D., A. Amrani, J. Verdaguer, S. Bou, P. Santamaria. 1999. Autoantigen-independent deletion of diabetogenic CD4⁺ thymocytes by protective MHC class II molecules. J. Immunol. 162:4627.[Abstract/Free Full Text]
18. Nepom, G. T.. 1990. A unified hypothesis for the complex genetics of HLA associations with IDDM. Diabetes 39:1153.[Abstract]
19. 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]
20. Boehme, J., N. Brenden, C. Riez, B. Pilstroem. 1994. Dominant disease protection in MHC-transgenic NOD mice: more than one mechanism?. The Immunologist 1994:45.
21. 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]
22. Mathis, D. J., C. O. Benoist, II V. E. Williams, M. R. Kanter, H. O. McDevitt. 1983. The murine E immune response gene. Cell 32:745.[Medline]
23. Lawrance, S. K., L. Karlsson, J. Price, V. Quaranta, Y. Ron, J. Sprent, P. A. Peterson. 1989. Transgenic HLA-DR faithfully reconstitutes I-E-controlled immune functions and induces cross-tolerance to E in E° mutant mice. Cell 58:583.[Medline]
24. Yamane, K., K. Yamamoto, Y. Yoshikawa, T. Sasazuki. 1996. Effect of the expression of DRE^NOD molecule on the development of insulitis and diabetes in the non-obese diabetic (NOD) mouse. Clin. Exp. Immunol. 103:141.[Medline]
25. Trembleau, S., G. Penna, E. Bosi, A. Mortara, M. K. Gately, L. Adorini. 1995. IL-12 administration induces Th1 cells and accelerates autoimmune diabetes in NOD mice. J. Exp. Med. 181:817.[Abstract/Free Full Text]
26. Trembleau, S., P. Giacomini, J.-C. Guéry, A. Setini, J. Hammer, A. Sette, E. Appella, L. Adorini. 1995. DR:E heterodimers in DRA transgenic mice hinder expression of E:E molecules and are more efficient in antigen presentation. Int. Immunol. 7:1927.[Abstract/Free Full Text]
27. Giacomini, P., A. Ciucci, M. R. Nicotra, C. Nastruzzi, G. Feriotto, E. Appella, R. Gambari, L. Pozzi, P. G. Natali. 1991. Tissue specific expression of the HLA-DRA gene in transgenic mice. Immunogenetics 34:385.[Medline]
28. Gately, M. K., R. R. Warrier, S. Honasoge, D. M. Carvajal, D. A. Faherty, S. E. Connaughton, T. D. Anderson, U. Sarmiento, B. R. Hubbard, M. Murphy. 1994. Administration of recombinant IL-12 to normal mice enhances cytolytic lymphocyte activity and induces production of IFN- in vivo. Int. Immunol. 6:157.[Abstract/Free Full Text]
29. Magistrelli, G., N. Covini, M. Mosca, G. Lippoli, A. Isacchi. 1995. Expression of PTP35, the murine homologue of the protein tyrosine phosphatase-related sequence IA-2, is regulated during cell growth and stimulated by mitogens in 3T3 fibroblasts. Biochem. Biophys. Res. Commun. 217:154.[Medline]
30. Trembleau, S., G. Penna, S. Gregori, G. Magistrelli, A. Isacchi, L. Adorini. 2000. Early Th1 response in unprimed nonobese diabetic mice to the tyrosine phosphatase-like insulinoma-associated protein 2, an autoantigen in type 1 diabetes. J. Immunol. 165:6748.[Abstract/Free Full Text]
31. Harrison, L. C., M. C. Honeyman, S. Trembleau, F. Gallazzi, P. Augstein, V. Brusic, J. Hammer, L. Adorini. 1997. A peptide-binding motif for I-A^g7, the class II MHC molecule of NOD and Biozzi AB/H mice. J. Exp. Med. 185:1013.[Abstract/Free Full Text]
32. Gregori, S., S. Trembleau, G. Penna, F. Gallazi, J. Hammer, G. K. Papadopoulos, L. Adorini. 1999. A peptide binding motif for I-E^g7, the class II molecule that protects E-transgenic nonobese diabetic mice from autoimmune diabetes. J. Immunol. 162:6630.[Abstract/Free Full Text]
33. Openshaw, P., E. E. Murphy, N. A. Hosken, V. Maino, K. Davis, K. Murphy, A. O’Garra. 1995. Heterogeneity of intracellular cytokine synthesis at the single-cell level in polarized T helper 1 and T helper 2 populations. J. Exp. Med. 182:1357.[Abstract/Free Full Text]
34. Adorini, L., E. Appella, G. Doria, Z. A. Nagy. 1988. Mechanisms influencing the immunodominance of T cell determinants. J. Exp. Med. 168:2091.[Abstract/Free Full Text]
35. Baum, H., V. Brusic, K. Choudhuri, P. Cunningham, D. Vergani, M. Peakman. 1995. MHC molecular mimicry in diabetes. Nat. Med. 1:388.[Medline]
36. Chaturvedi, P., B. Agrawal, M. Zechel, E. Lee-Chan, B. Singh. 2000. A self MHC class II -chain peptide prevents diabetes in nonobese diabetic mice. J. Immunol. 164:6610.[Abstract/Free Full Text]
37. Merino, R., M. Iwamoto, L. Fossati, P. Muniesa, K. Araki, S. Takahashi, J. Huarte, K.-I. Yamamura, J.-D. Vassali, S. Izui. 1993. Prevention of systemic lupus erythematosus in autoimmune BXSB mice by a transgene encoding I-E chain. J. Exp. Med. 178:1189.[Abstract/Free Full Text]
38. Iwamoto, M., N. Ibnou-Zekri, K. Araki, S. Izui. 1996. Prevention of murine lupus by an I-E chain transgene: protective role of I-E chain-derived peptides with a high affinity to I-A^b molecules. Eur. J. Immunol. 26:307.[Medline]
39. Guéry, J.-C., F. Ria, L. Adorini. 1996. Dendritic cells but not B cells present antigenic complexes to class II-restricted T cells after administration of protein in adjuvant. J. Exp. Med. 183:751.[Abstract/Free Full Text]
40. Guéry, J.-C., L. Adorini. 1995. Dendritic cells are the most efficient in presenting endogenous naturally processed self epitopes to class-II restricted T cells. J. Immunol. 154:536.[Abstract]
41. Corper, A. L., T. Stratmann, V. Apostolopoulos, C. A. Scott, K. C. Garcia, A. S. Kang, I. A. Wilson, L. Teyton. 2000. A structural framework for deciphering the link between I-A^g7 and autoimmune diabetes. Science 288:505.[Abstract/Free Full Text]
42. Brown, J. H., T. S. Jardetzky, J. C. Gorga, L. J. Stern, R. G. Urban, J. L. Strominger, D. C. Wiley. 1993. Three-dimensional structure of the human class II histocompatibility antigen HLA-DR1. Nature 364:33.[Medline]
43. Reich, E.-P., R. S. Sherwin, O. Kanagawa, Jr C. A. Janeway. 1991. An explanation for the protective effect of the MHC class II I-E molecule in murine diabetes. Nature 352:88.
44. Galbiati, F., L. Rogge, J. Guéry, S. Smiroldo, L. Adorini. 1998. Regulation of the IL-12 receptor 2 subunit by soluble antigen and IL-12 in vivo. Eur. J. Immunol. 28:209.[Medline]
45. Sercarz, E. E., P. V. Lehmann, A. Ametani, G. Benichou, A. Miller, K. Moudgil. 1993. Dominance and crypticity of the cell antigenic determinants. Annu. Rev. Immunol. 11:729.[Medline]
46. Murphy, D. B., D. Lo, S. Rath, R. L. Brinster, R. A. Flavell, A. Slanetz, Jr C. A. Janeway. 1989. A novel MHC class II epitope expressed in thymic medulla but not cortex. Nature 338:765.[Medline]
47. Humblet, C., A. Y. Rudensky, B. Keyewski. 1994. Presentation and intercellular transfer of self antigen within the thymic microenvironment: expression of the E peptide I-Ab complex by isolated thymic stroma cells. Int. Immunol. 6:1949.[Abstract/Free Full Text]
48. Arnold, B., G. Schoenrich, G. Haemmerling. 1993. Multiple levels of peripheral tolerance. Immunol. Today 14:12.[Medline]
49. Adorini, L., S. Muller, F. Cardinaux, P. V. Lehmann, F. Falcioni, Z. A. Nagy. 1988. In vivo competition between self peptides and foreign antigens in T-cell activation. Nature 334:623.[Medline]
50. Hurtenbach, U., E. Lier, L. Adorini, Z. A. Nagy. 1993. Prevention of autoimmune diabetes in non-obese diabetic mice by treatment with a class II major histocompatibility complex-blocking peptide. J. Exp. Med. 177:1499.[Abstract/Free Full Text]
51. Murphy, B., C. C. Magee, S. I. Alexander, A. M. Waaga, H. W. Snoeck, J. P. Vella, C. B. Carpenter, M. H. Sayegh. 1999. Inhibition of allorecognition by a human class II MHC-derived peptide through the induction of apoptosis. J. Clin. Invest. 103:859.[Medline]
52. Boytim, M. L., S.-C. Lyu, R. Jung, A. M. Krensky, C. Clayberger. 1998. Inhibition of cell cycle progression by a synthetic peptide corresponding to residues 65–79 of an HLA class II sequence: functional similarities by mechanistic differences with the immunosuppressive drug rapamycin. J. Immunol. 160:2215.[Abstract/Free Full Text]
53. Basu, D., S. Horvath, I. Matsumoto, D. H. Fremont, P. M. Allen. 2000. Molecular basis for recognition of an arthritic peptide and a foreign epitope on distinct MHC molecules by a single TCR. J. Immunol. 164:5788.[Abstract/Free Full Text]
54. Wicker, L. S., M. C. Appel, F. Dotta, A. Pressey, B. J. Miller, N. H. DeLarato, P. A. Fischer, Jr R. C. Boltz, L. B. Peterson. 1992. Autoimmune syndromes in major histocompatibility complex (MHC) congenic strains of nonobese diabetic (NOD) mice: the NOD MHC is dominant for insulitis and cyclophosphamide-induced diabetes. J. Exp. Med. 176:67.[Abstract/Free Full Text]
55. Miyazaki, T., M. Uno, M. Uehira, H. Kikutani, T. Kishimoto, M. Kimoto, H. Nishimoto, J. Miyazaki, K. 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.[Medline]
56. Slattery, R. M., L. Kjer-Nielsen, J. Allison, B. Charlton, T. E. Mandel, J. F. Miller. 1990. Prevention of diabetes in non-obese diabetic I-A^k transgenic mice. Nature 345:724.[Medline]
57. Singer, S. M., R. Tisch, X. D. Yang, H. O. McDevitt. 1993. An Ab^d transgene prevents diabetes in nonobese diabetic mice by inducing regulatory T cells. Proc. Natl. Acad. Sci. USA 90:9566.[Abstract/Free Full Text]
58. Quartey-Papafio, R., T. Lund, P. Chandler, J. Picard, P. Ozegbe, S. Day, P. R. Hutchings, L. O’Reilly, D. Kioussis, E. Simpson, A. Cooke. 1995. Aspartate at position 57 of nonobese diabetic I-Ag7 -chain diminishes the spontaneous incidence of insulin-dependent diabetes mellitus. J. Immunol. 154:5567.[Abstract]
59. Podolin, P. L., A. Pressey, N. H. DeLarato, P. A. Fisher, L. B. Peterson, L. S. Wicker. 1993. I-E⁺ nonobese diabetic mice develop insulitis and diabetes. J. Exp. Med. 178:793.[Abstract/Free Full Text]
60. Trembleau, S., G. Penna, S. Gregori, M. K. Gately, L. Adorini. 1997. Deviation of pancreas-infiltrating cells to Th2 by interleukin-12 antagonist administration inhibits autoimmune diabetes. Eur. J. Immunol. 27:2330.[Medline]
61. Rothe, H., R. M. O’Hara, S. Martin, H. Kolb. 1997. Suppression of cyclophosphamide induced diabetes development and pancreatic Th1 reactivity in NOD mice treated with the interleukin (IL)-12 antagonist IL-12(p40)₂. Diabetologia 40:641.[Medline]
62. Singer, S. M., R. Tisch, X. D. Yang, H. K. Sytwu, R. Liblau, H. O. McDevitt. 1998. Prevention of diabetes in NOD mice by a mutated I-Ab transgene. Diabetes 47:1570.[Abstract]
63. Serreze, D. V., H. D. Chapman, C. M. Post, E. A. Johnson, W. L. Suarez-Pinzon, A. Rabinovitch. 2001. Th1 to Th2 cytokine shifts in nonobese diabetic mice: sometimes an outcome, rather than the cause, of diabetes resistance elicited by immunostimulation. J. Immunol. 166:1352.[Abstract/Free Full Text]
64. Takacs, K., D. C. Douek, D. M. Altmann. 1995. Exacerbated autoimmunity associated with a T helper-1 cytokine profile shift in H-2E-transgenic mice. Eur. J. Immunol. 25:3134.[Medline]

This article has been cited by other articles:

				N. K. Brown, D. J. McCormick, C. S. David, and Y.-c. M. Kong H2E-Derived E{alpha}52-68 Peptide Presented by H2Ab Interferes with Clonal Deletion of Autoreactive T Cells in Autoimmune Thyroiditis J. Immunol., May 15, 2008; 180(10): 7039 - 7046. [Abstract] [Full Text] [PDF]

				S. Trembleau, G. Penna, S. Gregori, N. Giarratana, and L. Adorini IL-12 Administration Accelerates Autoimmune Diabetes in Both Wild-Type and IFN-{gamma}-Deficient Nonobese Diabetic Mice, Revealing Pathogenic and Protective Effects of IL-12-Induced IFN-{gamma} J. Immunol., June 1, 2003; 170(11): 5491 - 5501. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Trembleau, S. Articles by Adorini, L. Search for Related Content PubMed PubMed Citation Articles by Trembleau, S. Articles by Adorini, L.