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
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Forestier, C.
Right arrow Articles by Porcelli, S. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Forestier, C.
Right arrow Articles by Porcelli, S. A.
The Journal of Immunology, 2007, 178: 1415-1425.
Copyright © 2007 by The American Association of Immunologists, Inc.

Improved Outcomes in NOD Mice Treated with a Novel Th2 Cytokine-Biasing NKT Cell Activator1

Claire Forestier2,*, Toshiyuki Takaki2,*, Alberto Molano*, Jin S. Im*, Ian Baine*, Elliot S. Jerud*, Petr Illarionov{dagger}, Rachel Ndonye{ddagger}, Amy R. Howell{ddagger}, Pere Santamaria§, Gurdyal S. Besra{dagger}, Teresa P. DiLorenzo*,¶ and Steven A. Porcelli3,*,¶

* Department of Microbiology and Immunology, Albert Einstein College of Medicine, Bronx, NY 10461; {dagger} School of Biosciences, University of Birmingham, Birmingham, United Kingdom; {ddagger} Department of Chemistry, University of Connecticut, Storrs, CT 06269; § Department of Microbiology and Infectious Diseases and Julia McFarlane Diabetes Research Centre, University of Calgary, Calgary, Canada; and Department of Medicine, Albert Einstein College of Medicine, Bronx, NY 10461


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Activation of CD1d-restricted invariant NKT (iNKT) cells by {alpha}-galactosylceramide ({alpha}GalCer) significantly suppresses development of diabetes in NOD mice. The mechanisms of this protective effect are complex, involving both Th1 and Th2 cytokines and a network of regulatory cells including tolerogenic dendritic cells. In the current study, we evaluated a newly described synthetic {alpha}GalCer analog (C20:2) that elicits a Th2-biased cytokine response for its impact on disease progression and immunopathology in NOD mice. Treatment of NOD mice with {alpha}GalCer C20:2 significantly delayed and reduced the incidence of diabetes. This was associated with significant suppression of the late progression of insulitis, reduced infiltration of islets by autoreactive CD8+ T cells, and prevention of progressive disease-related changes in relative proportions of different subsets of dendritic cells in the draining pancreatic lymph nodes. Multiple favorable effects observed with {alpha}GalCer C20:2 were significantly more pronounced than those seen in direct comparisons with a closely related analog of {alpha}GalCer that stimulated a more mixed pattern of Th1 and Th2 cytokine secretion. Unlike a previously reported Th2-skewing murine iNKT cell agonist, the {alpha}GalCer C20:2 analog was strongly stimulatory for human iNKT cells and thus warrants further examination as a potential immunomodulatory agent for human disease.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Type 1 diabetes is an organ-specific autoimmune disease mediated by pathogenic Th1-type T cells that specifically destroy the pancreatic islet beta cells. The NOD mouse is susceptible to the spontaneous development of type 1 diabetes and many studies support its usefulness as a model for this disease and its prevention or treatment in humans (1). Type 1 or invariant NKT (iNKT)4 cells are unconventional regulatory and effector T cells that recognize glycolipid Ags presented by CD1d (2). Upon activation, iNKT cells release copious amounts of both Th1 and Th2 cytokines such as IFN-{gamma} and IL-4 and significantly influence the outcome of developing or ongoing adaptive immune responses. Multiple studies show that iNKT cells play a regulatory role in autoimmune diabetes in NOD mice (reviewed in Ref. 2), and treatment with synthetic {alpha}-galactosylceramides ({alpha}GalCer) that activate iNKT cells can block or attenuate the development of clinical diabetes in these animals (3, 4, 5). However, the mechanisms by which iNKT cell activation protects NOD mice from diabetes are still not well-understood and are likely to be multifactorial.

Previous work has shown that alterations of the basic lipid structure of the prototypical iNKT cell activator KRN7000 ([(2S, 3S, 4R)-1-O-({alpha}-D-galactopyranosyl)-N-hexacosanoyl-2-amino-1,3,4-octadecanetriol], in this study referred to as {alpha}GalCer C26:0) can lead to striking changes in the patterns of cytokine production elicited following iNKT cell activation (6, 7, 8, 9, 10). One of the most intensively studied examples of this is an {alpha}GalCer analog designated OCH, in which the fatty acid chain has been shortened to C24:0 and the sphingoid base truncated to C9. This analog is a relatively selective stimulator of IL-4 secretion and has pronounced anti-inflammatory effects in NOD mice and other mouse models of autoimmune disease (6, 11, 12, 13, 14, 15). Our earlier studies identified a novel derivative of KRN7000 containing an 11,14-cis-diunsaturated C20 fatty acid ({alpha}GalCer C20:2) as a potent iNKT cell activator that powerfully stimulates production of IL-4 in association with reduced production of IFN-{gamma}. In this study, we have directly compared the iNKT cell-activating properties of the C20:2 and OCH compounds and find that the former is a substantially more potent iNKT cell agonist, particularly for human iNKT cells. In studies of diabetes prevention in NOD mice, we observed that C20:2 produced significant improvements in clinical and immunological outcomes compared with a closely related analog ({alpha}GalCer C24:0), which has similar potency but elicits a more mixed cytokine response similar to KRN7000. These results point to important advantages of C20:2 for prevention of autoimmunity in NOD mice compared with other previously studied analogs of {alpha}GalCer and suggest that C20:2 warrants further investigation as a potential agent for prevention or treatment of human type 1 diabetes.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Mice

Female NOD/LtJ, C57BL/6, and BALB/c mice at 3–4 wk of age were purchased from The Jackson Laboratory. NOD/Mrk mice at 3–4 wk of age were purchased from Taconic Farms. Mice were maintained in the animal facilities at Albert Einstein College of Medicine under specific pathogen-free conditions and were allowed to adapt to these housing conditions for 1–3 wk before initiating experimental protocols. The institutional animal welfare committee approved all experimental procedures.

Cell lines

Murine iNKT hybridomas DN3A4-1.2 (V{alpha}14/Vbeta8.2) and DN3A4-1.4 (V{alpha}14/Vbeta10), and murine CD1d-transfected A20 B lymphoma, were provided by Dr. M. Kronenberg (La Jolla Institute for Allergy and Immunology, San Diego, CA). Murine iNKT hybridoma N38.2H4 (V{alpha}14/Vbeta7) was provided by Dr. K. Hayakawa (Fox Chase Cancer Center, Philadelphia, PA). The human iNKT cell clone DN2.D5 has been previously described (16) and other clones used for experiments referred to but not shown in the current manuscript were unpublished human iNKT cell clones derived in our laboratory from human peripheral blood (clones DN2.C4, DN2.C5, HDD.11, HDA.A7, and HDA.A8). H-2Kd-transfected murine RMA-S lymphoma cells (RMA-S/Kd cells) were provided by Dr. M. Bevan (University of Washington, Seattle, WA).

Glycolipid preparation

All glycolipids used in this study were synthesized according to previously reported methods (8, 17). For in vivo administration, glycolipids were dissolved at 200 µg/ml in PBS containing 0.5% Tween 20, heated to 80°C for 10 min, and sonicated for 5 min in a water bath sonicator (model 2510; Branson Ultrasonic) before use. The resulting suspensions were then diluted in prewarmed PBS to a glycolipid concentration of 16 µg/ml (0.04% final (Tween 20)), sonicated, heated to 80°C, and immediately used for injection. For in vitro assays, glycolipids were dissolved in DMSO at a concentration of 500 µM, briefly sonicated, and then diluted directly into medium.

Diabetes treatment and monitoring

Starting at 4–6 wk of age, female NOD mice received seven weekly i.p. injections of glycolipid (4 µg/mouse in 250 µl of vehicle consisting of PBS plus 0.04% Tween 20) or vehicle. Diabetes was assessed starting at 10 wk of age by monitoring mice weekly for glucosuria with Diastix reagent strips (Bayer). Mice were considered diabetic when two consecutive positive measurements were obtained and the time of onset of diabetes was recorded as the date of the first of the two consecutive readings.

Assessment of insulitis by histopathology

Pancreata were fixed in Bouin’s solution and sectioned at three nonoverlapping levels. Granulated beta cells were stained with aldehyde fuchsin and leukocytes with an H&E counterstain. Islets (at least 20/mouse in most cases) were individually scored as follows: 0, no lesions; 1, peri-insular leukocytic aggregates, usually periductal infiltrates but no islet destruction; 2, ≤25% islet destruction; 3, >25–75% islet destruction; and 4, >75% islet destruction. An insulitis index was calculated by the formula: insulitis index = ((total score for all islets)/(4 x number of islets examined)). Diabetic mice or animals that died after becoming diabetic were assigned an insulitis index of 1, because these mice usually had <20 remaining islets and all examined islets were score 4.

Peptides and tetramers

IGRP206–214 (VYLKTNVFL), Ins-I9 (LYLVCGERI), NRP-V7 (KYNKANVFL), and MimA2 (YAIENYLEL) peptides were synthesized by standard solid-phase methods using F-moc chemistry in an automated peptide synthesizer (model 433A; Applied Biosystems) and their identities were confirmed by mass spectrometry. Concentrated stocks (10 mM) were prepared in DMSO and 10 µM working stocks were obtained by dilution in PBS. PE-labeled H-2Kd tetramers loaded with NRP-V7 or InsI9 peptides were produced as described previously (18). PE-labeled H-2Db tetramers loaded with MimA2 peptide were produced by the National Institutes of Health/National Institute of Allergy and Infectious Diseases Tetramer Core Facility. Soluble native murine and human CD1d proteins were prepared using baculovirus and mammalian cell expression systems and enzymatically biotinylated at a specific C-terminal biotin ligase site as described in previous reports from our laboratories (8, 19). Biotinylated CD1d proteins at 2 µM were incubated with a 20-fold excess of the indicated glycolipid for 24 h and subsequently assembled into tetramers by addition of 0.5 µM allophycocyanin- or PE-conjugated streptavidin. To produce fluorescent glycolipid-loaded CD1d dimers or monomers, 2 µM biotinylated murine CD1d/glycolipid complexes were mixed with biotinylated human CD1b protein (produced by expression in Chinese hamster ovary cells as previously published (19)) at a ratio of 1:1 (for dimers) or 3:1 (for monomers), and then incubated with 0.5 µM PE-conjugated streptavidin.

Propagation of islet-infiltrating T cells

Islet isolation by collagenase perfusion of the common bile duct was modified from a previously described protocol (18). Briefly, the bile duct was cannulated and the pancreas perfused with collagenase P (Roche). The inflated pancreas was removed and incubated at 37°C to digest exocrine tissue. Following dispersion of digested tissue and three washes with HBSS, islets were resuspended in HBSS containing DNase I (Worthington Biochemical) and handpicked using a siliconized micropipet under a dissecting microscope (all islets that could be identified were collected and enumerated separately for each mouse). Isolated islets were washed with 2% FBS in HBSS, resuspended in RPMI 1640 medium supplemented with 10% FBS (HyClone), and 50 U/ml recombinant human IL-2 (PeproTech), and cultured in 24-well tissue culture plates (~50 islets/well) at 37°C, 5% CO2. Cultures were harvested after 7 days, and cells were cryopreserved in aliquots for subsequent FACS and ELISPOT analyses.

Preparation of cell suspensions from mouse tissues

For thymocyte suspensions, aseptically removed thymuses from 6- to 8-wk-old C57BL/6 mice were mechanically forced through stainless steel mesh, and tissue fragments were removed by settling. Cell suspensions were prepared from spleens and pancreatic lymph nodes (PLN) by gently grinding the tissues with a syringe plunger on the surface of a 70-µm nylon mesh, and cells were washed through and collected with PBS. For dendritic cell (DC) isolation, spleen and PLN were first injected with a solution of collagenase at 400 U/ml (Sigma-Aldrich), cut into small pieces, and suspended in 100 U/ml collagenase, then incubated for 4 h at 37°C. Tissue fragments were harvested and disaggregated into single-cell suspensions by pressing through nylon mesh.

Measurement of cytokine secretion by in vitro- and in vivo-activated iNKT cells

For in vitro activation of murine splenic iNKT cells, splenocytes from BALB/c mice were plated at 5 x 105 cells/well in 96-well flat-bottom tissue culture plates and stimulated with {alpha}GalCer analogs at a range of concentrations. Culture supernatants were harvested after 48 h at 37°C and levels of IL-4, IL-13, and IFN-{gamma} were quantified by ELISA. For in vitro activation, murine iNKT hybridomas or human iNKT cell clones at 5 x 104/well were stimulated with 5 x 104 murine CD1d-transfected A20 cells or human DCs (derived by in vitro culture of peripheral blood monocytes for 3 days in medium supplemented with GM-CSF and IL-4) in the presence of the indicated concentrations of {alpha}GalCer analogs. Levels of murine IL-2 were measured at 12 h, and levels of human IL-4, IL-13, IFN-{gamma}, and TNF-{alpha} were measured at 48 h in culture supernatants by capture ELISA. For in vivo iNKT cell activation, female NOD or C57BL/6 mice, 5–8 wk of age, were injected i.p. with vehicle or {alpha}GalCer analogs (4 µg/mouse). Blood was collected 2, 6, and 21 h postinjection by retro-orbital puncture and serum samples were diluted 1/5 with PBS and tested for IL-4, IFN-{gamma}, and IL-12p70 by ELISA. Cytokine production by iNKT cells was assessed 2 h postinjection by intracellular IL-4 and IFN-{gamma} cytokine staining followed by FACS analysis. All ELISA used specific capture and biotinylated detection mAb pairs (BD Pharmingen), streptavidin-alkaline phosphatase (Zymed Laboratories) and 4-nitrophenyl phosphate (Sigma-Aldrich) as substrate. Recombinant cytokines (PeproTech) were used to generate standard curves. Measurements were performed in duplicate or triplicate.

IFN-{gamma} ELISPOT assays

ELISPOT plates (MAHA S45 10; Millipore) were precoated with anti-murine IFN-{gamma} mAb (R4-6A2; BD Pharmingen) and blocked with 1% BSA (Fraction V; Sigma-Aldrich) in PBS. Mitomycin C-treated RMA-S/Kd cells were added at 2 x 104 cells/well and pulsed with 1 µM peptide. Cryopreserved aliquots of cultured islet-infiltrating T cells were thawed and added at 2 x 104 cells/well, and plates were incubated at 37°C for 40 h. IFN-{gamma} secretion was detected with a second, biotinylated anti-murine IFN-{gamma} mAb (XMG1.2; BD Pharmingen). Spots were developed using streptavidin-alkaline phosphatase (Zymed Laboratories) and 5-bromo-4-chloro-3-indolyl-phosphate/NBT chloride substrate (Sigma-Aldrich) and counted using an automated ELISPOT reader system (Autoimmun Diagnostika).

Flow cytometry

Single-cell suspensions were stained in FACS buffer (1% BSA plus 0.05% NaN3 in PBS) with PE- or allophycocyanin-conjugated glycolipid-loaded CD1d tetramers for 1 h at room temperature, followed by staining when indicated with fluorescent-labeled Abs for 30 min at 4°C. For equilibrium binding assays using CD1d tetramers, dimers, or monomers, C57BL/6 thymocytes or iNKT hybridoma cell lines were incubated with the indicated concentrations of PE-labeled CD1d complexes in 50 µl of FACS buffer for 1 h at room temperature. For intracellular cytokine staining, cells were stained with allophycocyanin-conjugated {alpha}GalCer-CD1d tetramers for 3 h on ice before being permeabilized using a BD Pharmingen Cytofix/Cytoperm kit, and then stained for 30 min at 4°C with the relevant PE-conjugated anti-cytokine Ab. Nonspecific staining was blocked using 10 µg/ml rat anti-mouse CD16/32 (2.4G2; BD Pharmingen). For analysis of in vitro-expanded islet-derived T cell cultures, aliquots of cryopreserved cells were thawed and incubated with FITC-conjugated anti-murine CD8{alpha} mAb, allophycocyanin-conjugated anti-murine CD4 mAb and PE-labeled peptide-loaded H-2Kd or H-2Db tetramers in the presence of mAb 2.4G2. FITC-, PE-, PerCP- or allophycocyanin-conjugated mAbs to TCR{alpha}beta, B220, CD44, CD69, CD25, CD4, CD8{alpha}, CD11c, CD80, CD86, I-Ak (cross-reactive with I-Ag7), IL-4, and IFN-{gamma} were all purchased from BD Pharmingen. Samples were analyzed by flow cytometry using a FACSCalibur instrument and CellQuest software (BD Immunocytometry Systems).

Statistics

GraphPad Prism software was used for statistical calculations and p values <0.05 were considered significant. One-way ANOVA was used for analysis of experiments with three or more data sets, and posttest comparisons were analyzed by either unpaired Student’s t test or Mann-Whitney nonparametric test as indicated. Differences in Kaplan-Meier curves for onset of glycosuria and death were analyzed by log-rank test. For analysis of insulitis index (see Fig. 4B), a weighted least squares linear regression was used to test for significance. To prevent overrepresentation by animals that survived longer as a result of random background resistance to development of diabetes (e.g., not all NOD mice develop diabetes), data from surviving animals was supplemented with imputed insulitis scores of 1.0 for all animals that died before an analysis time point. With the supplemented data, linear regression analysis was conducted with the insulitis score as the outcome variable. The predictor variables were indicators for week, treatment group, and their interaction effects. To reduce heteroscedasticity, the regression analysis weighted each observation in proportion to the number of islets assessed in that animal. Because islets of deceased animals could not be assessed, the imputed insulitis scores for these animals were weighted as if three islets had been observed, which was the median number of islets assessed per animal for all experimental groups combined.


Figure 4
View larger version (32K):
[in this window]
[in a new window]

  		FIGURE 4. Improved outcomes in NOD mice treated with C20:2. A, Groups of female NOD mice were injected i.p. weekly with 4 µg of C24:0, C20:2 or vehicle starting at 4–6 wk of age for 7 wk. The percentages of mice negative for glucosuria over time are shown on the left. Differences between incidence curves were tested for significance by log-rank test, which gave p values of 0.0204 for C24:0 vs vehicle, 0.0007 for C20:0 vs vehicle, and 0.0424 for C20:2 vs C24:0. The percentages of mice surviving over time are shown on the right (p = 0.0318 and 0.0006 vs vehicle control group for C24:0 and C20:2, respectively). The difference in survival between C24:0- and C20:2-treated groups was statistically significant at 30 wk (p = 0.0049), whereas the differences in glycosuria at this time point did not achieve significance (p = 0.0939). Pooled data are shown from three separate experiments (n = 29 for each group). B, Histopathologic analysis of insulitis in {alpha}GalCer-treated NOD mice. Female NOD mice were injected i.p. with 200 µg/kg {alpha}GalCer analogs (C24:0 or C20:2) or vehicle once per week for 7 wk starting at 4–5 wk of age. Means for histologic sections of pancreata from five to nine mice per group are plotted. Both the cumulative insulitis index (left) and the percentage of islets at each insulitis stage (right) are displayed. The values for insulitis index were significantly different compared with vehicle-treated animals at 17 wk for C24:0 (p = 0.004) and at 53 wk for C20:2-treated animals (p = 0.002).

 

   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Stimulation of altered cytokine responses by {alpha}GalCer analogs in vitro and in vivo

Analogs of {alpha}GalCer KRN7000 with fatty acid chain substitutions (C20:2 and C24:0) were identified by our earlier studies as promising candidates for further development as in vivo iNKT cell agonists (Fig. 1A) (8). To assess further the potential applications of these novel compounds, we compared their iNKT cell-activating properties to those of two other well-studied forms of {alpha}GalCer, KRN7000 and OCH. Using a panel of mouse iNKT cell hybridomas expressing different TCR Vbeta chain segments, we confirmed the stimulatory properties of all four {alpha}GalCer analogs. However, there were substantial variations in potency for the four compounds, with C26:0 consistently showing the most potent activity while OCH had the weakest (Fig. 1A). The OCH and C20:2 compounds have previously been described as stimulators of a Th2-biased iNKT cell response (6, 8) and we confirmed and extended this observation in studies of in vitro mouse splenocyte stimulation that compared the production of two Th2-associated cytokines (IL-4 and IL-13) with the concurrent production of IFN-{gamma} (Fig. 1B). This analysis revealed that C20:2 was a particularly strong inducer of both IL-4 and IL-13 (Fig. 1B, left). By considering the ratios of IL-4 or IL-13 to IFN-{gamma} production, it was apparent that these were substantially higher over the entire range of glycolipid concentrations for both C20:2 and OCH compared with C24:0 and C26:0 (Fig. 1B, right). These findings confirmed the previous characterization of C20:2 and OCH as Th2-biasing {alpha}GalCer analogs and the direct comparison of these two analogs indicated that C20:2 had the potential to stimulate substantially greater quantities of Th2 cytokines.


Figure 1
View larger version (41K):
[in this window]
[in a new window]

  		FIGURE 1. Distinct cytokine responses of mouse iNKT cells to {alpha}GalCer analogs. A, Structures of synthetic {alpha}GalCer analogs used in this study are illustrated (left). Shown on the right are IL-2 secretion by a representative iNKT cell hybridoma (DN3A4-1.2) over a range of glycolipid concentrations (top), and relative potencies for each of the glycolipids using murine V{alpha}14i NKT cell hybridomas expressing either Vbeta8.2, Vbeta7, or Vbeta10 (bottom). APCs were CD1d-transfected RMA-S cells. Relative potencies are expressed as 1/ED50, (reciprocal of glycolipid concentration giving half maximal IL-2 production). B, Supernatant levels of IL-4, IL-13, and IFN-{gamma} from cultured BALB/c splenocytes were determined by capture ELISAs after 48 h of culture with the indicated glycolipids at concentrations of 31.6 nM ({square}), 10 nM (Figure 1), or 3.16 nM ({blacksquare}). Bars show mean and range for duplicate measurements. Calculated ratios of IL-4 and IL-13 levels to IFN-{gamma} levels in supernatant samples are shown in scatter plot to the right. C, C57BL/6 mice received one i.p. injection of 4 µg of each glycolipid, and blood samples were obtained at the indicated times for measurement of serum IL-4, IFN-{gamma}, and IL-12 p70 by ELISA. Means plus SDs for groups of five mice are shown. *, Below limit of detection (<0.02 ng/ml for IL-4 and IFN-{gamma}, and <0.31 ng/ml for IL-12 p70). Results shown are representative of two (C) or three (A and B) independent experiments.

 
To compare in vivo the relative cytokine stimulation by the four {alpha}GalCer analogs, we assessed the serum levels of IL-4, IFN-{gamma}, and IL-12p70 after i.p. injection of a single dose of each glycolipid. IL-4 levels are known to peak at 2 h after {alpha}GalCer stimulation in this assay, whereas IFN-{gamma} peaks much later and reaches a plateau between 20 and 24 h (6). The levels of IL-4 and IFN-{gamma} observed at their expected peak time points were supportive of the previous characterization of C20:2 and OCH as Th2-biasing analogs (Fig. 1C). Again, the direct comparison suggested a more potent stimulation of IL-4 by C20:2, although this trend did not reach statistical significance (p = 0.06, one-way ANOVA). Interestingly, IL-12p70 levels did not show any significant variation between the four compounds, as all of them showed a peak of similar magnitude at 6 h (Fig. 1C), followed by a return to undetectable levels by 20 h (data not shown). This finding underscores the complexity of the iNKT cell cytokine response, which even following stimulation with C20:2 or OCH shows certain features that are typically linked with Th1-type inflammatory responses. Furthermore, these findings indicate that the failure of C20:2 and OCH to stimulate a strong and sustained IFN-{gamma} response cannot be attributed to the failure of these analogs to evoke IL-12p70 release, which is in contradiction to a previous report describing OCH as a less potent inducer of IL-12p70 than KRN7000 (15).

High-avidity interaction of {alpha}GalCer C20:2 with TCRs of murine and human iNKT cells

Because avidity of the TCRs of iNKT cells with their specific ligands has been proposed as an important determinant of the outcome of activation (20, 21), we assessed this parameter for each of the four {alpha}GalCer analogs. First, we used costaining of murine splenocytes with allophycocyanin-labeled murine CD1d tetramers loaded with C26:0 plus PE-labeled murine CD1d tetramers loaded with each of the different analogs to determine whether all of these were recognized by the same iNKT cell populations. This proved to be the case, as each analog-loaded tetramer bound to all C26:0-loaded tetramer-positive T cells, albeit with varying intensities (Fig. 2A, left). We then measured the equilibrium-binding levels of murine CD1d tetramers loaded with each of the {alpha}GalCer analogs over a range of tetramer concentrations to determine the equilibrium dissociation constant for tetramer binding (KD), which is inversely related to TCR avidity (Fig. 2A, center). Analogous binding curves and KD values were similarly generated using {alpha}GalCer analog-loaded dimers and monomers (Fig. 2A, right). The KD values for iNKT cell TCR interactions with either multivalent or monovalent ligands were low for C20:2-loaded CD1d complexes, indicating a binding strength similar or even greater than that for complexes of CD1d with the non-Th2-skewing analogs, C26:0 and C24:0. In contrast, OCH-loaded CD1d complexes showed much higher KD values, consistent with relatively weak binding to the TCRs of mouse iNKT cells.


Figure 2
View larger version (41K):
[in this window]
[in a new window]

  		FIGURE 2. Molecular interactions of {alpha}GalCer analogs with TCRs of murine and human iNKT cells. A, Mouse CD1d tetramers labeled with allophycocyanin were loaded with each of the glycolipid analogs (x-axis; glycolipid used for loading is indicated in box surrounding each dot plot), and used together with PE-labeled C26:0-loaded mCD1d tetramers (y-axis) to costain C57BL/6 thymocytes (left). Tetramers loaded with each of the analogs were titrated for staining of mouse iNKT cell hybridoma cells (DN3A4-1.2) to generate equilibrium binding curves based on measurement of mean fluorescence intensity (MFI) by FACS (middle). This procedure was also conducted with fluorescent dimers and monomers of mCD1d loaded with each glycolipid, and equilibrium dissociation constants (KD) were determined as the concentrations of CD1d/glycolipid complexes required to give 50% of maximal binding (right). B, Similar analyses as described for murine iNKT cell binding of glycolipid tetramers in A were conducted using human CD1d tetramers to stain human PBL (dot plots), or a human V{alpha}24+ iNKT cell clone, DN2.D5 (histograms; heavy line is staining with tetramer loaded with indicated glycolipid, and light line is unloaded tetramer control) (left). Cytokine secretion by iNKT cell clone DN2.D5 cultured for 48 h with human monocyte-derived DCs and each of the glycolipids was determined using capture ELISAs for IL-4, IL-13, IFN-{gamma}, and TNF-{alpha} (right). Stimulation with immobilized anti-CD3 mAb (OKT3) was used as a positive control for T cell activation. *, Below limits of detection (<0.05 ng/ml).

 
Previous studies have emphasized the strong conservation of CD1d and iNKT cell responses between mice and humans, which supports the use of mouse models in preclinical exploration of iNKT cell-based therapies (22). Nevertheless, the surprisingly weak interaction of OCH with TCRs of mouse iNKT cells prompted us to examine whether this analog could be recognized significantly by human iNKT cells. This was analyzed initially by using human CD1d tetramers loaded with each of the four {alpha}GalCer analogs to stain human peripheral blood iNKT cells. Whereas all cells that stained with C26:0-loaded human CD1d tetramers were strongly costained by C24:0- or C20:2-loaded human CD1d tetramers, we could not detect any significant staining with OCH-loaded human CD1d tetramers in peripheral blood of normal human subjects (Fig. 2B, left). We conducted similar tetramer staining studies on a panel of in vitro-expanded human iNKT cell clones (clones DN2.D5, DN2.C4, DN2.C5, HDD.11, HDA.A7, and HDA.A8) and again failed to detect significant staining with OCH-loaded human CD1d tetramers while tetramers loaded with the other {alpha}GalCer analogs stained strongly (Fig. 2B, left, and data not shown). The tetramer binding results were confirmed by functional responses of human iNKT cell clone DN2.D5 (Fig. 2B, right) showing that C20:2, C26:0, and C24:0 were all strong activators of both Th2 and Th1 associated cytokine production, whereas OCH failed to stimulate any detectable responses.

Induction of Th2-biased cytokine response by C20:2 in NOD mice

Although a previous study found that repeated administration of OCH was effective for prevention of type 1 diabetes in NOD mice (12), our finding that OCH had no stimulatory activity for human iNKT cells indicated that this compound may not be suitable for development as an immunomodulator for use in humans. Given the strong activity of the C20:2 analog on human iNKT cells, it was of interest to know whether this analog would also behave as a strong iNKT cell activator and give preferential Th2-cytokine induction in NOD mice. Examining this question first in vitro, we found that C20:2 induced NOD splenocyte cultures to produce significantly more IL-4 and less IFN-{gamma} compared with C24:0 (Fig. 3A, top panel). In vivo, we found that the rapid burst of serum IL-4 occurring 2 h after injection was significantly higher in NOD mice receiving C20:2 compared with C24:0, whereas the delayed production of IFN-{gamma} usually observed at 20–24 h postinjection was absent in C20:2-treated mice (Fig. 3A, bottom panel). The iNKT cell responses of NOD mice were overall significantly weaker than those previously observed in C57BL/6 mice (8), consistent with the well-described quantitative and qualitative deficiencies of iNKT cells characteristic of NOD mice (2). We also used intracellular cytokine staining and FACS to examine more directly the impact of C20:2 stimulation on iNKT cell cytokine production (Fig. 3B). This showed that in splenocytes, C20:2 was not able to induce a Th2 bias at the iNKT cell level because similar percentages of iNKT cells secreting IL-4 and IFN-{gamma} were obtained in C20:2- and C24:0-treated mice, respectively. However, in PLN, C20:2 triggered a significant increase in the percentage of IL-4-positive iNKT cells compared with either vehicle or C24:0-treated mice, and this was associated with low percentages of IFN-{gamma}-positive iNKT cells (Fig. 3B).


Figure 3
View larger version (20K):
[in this window]
[in a new window]

  		FIGURE 3. Cytokine responses of NOD iNKT cells to {alpha}GalCer analogs. A, Splenocytes from female NOD mice (5 wk of age) were stimulated with 100 nM C20:2 or C24:0 {alpha}GalCer for 72 h and supernatants were tested for IL-4 and IFN-{gamma} content by ELISA (top). Bars represent means of triplicates cultures. Serum levels of IL-4 and IFN-{gamma} in mice given 4 µg i.p. of C24:0 or C20:2 were measured by ELISA at 2 and 23 h postinjection (bottom). Bars show means and SDs for groups of five mice. One representative example of two separate experiments is shown. B, Female NOD mice (5 wk old) received a single i.p. injection of either inert vehicle or 4 µg of glycolipid (C20:2 or C24:0), and 2 h later the spleens and PLNs were harvested for intracellular staining of IL-4 or IFN-{gamma}. Means and SDs of triplicate measurements are shown. *, p < 0.01 (upper panel) or p < 0.05 (lower panel) compared with vehicle control (unpaired Student’s t test).

 
Efficacy of C20:2 for diabetes prevention and inhibition of progressive insulitis

Having established that C20:2 evoked a Th2-biased cytokine profile in NOD mice, we investigated the effect of this compound on diabetes incidence by comparing its effects directly to those of the C24:0 analog which showed similar potency for iNKT cell activation but a more mixed cytokine response. Female NOD mice were treated starting from 4 to 6 wk of age with weekly i.p. injections of {alpha}GalCer analogs for 7 wk and progression of diabetes was monitored by development of glucosuria and time to death (Fig. 4A). Although both iNKT cell activators gave a significant delay in diabetes onset and death compared with vehicle-treated animals, the C20:2 analog showed significant improvement in both outcomes compared with C24:0. This was evident both in the median time to development of glucosuria (36 wk for C20:2 vs 27 wk for C24:0) and in the median survival (>50 wk for C20:2 vs 40 wk for C24:0).

Because the two {alpha}GalCer analogs had a different impact on diabetes development in NOD mice, we next compared their effect on the progression of insulitis with age. As shown in Fig. 4B, mice treated with C24:0 showed significantly reduced insulitis at 17 wk of age (i.e., 4–6 wk after cessation of treatment) compared with vehicle- and C20:2-treated animals. However, insulitis progressed with age in C24:0-treated groups, ultimately reaching the same severity as the vehicle-treated control animals at 53 wk of age. In contrast, treatment with C20:2 did not give detectable inhibition of insulitis at 17 wk of age, but did show a significant reduction of insulitis in animals at 53 wk of age (Fig. 4B). This observation suggested that at least part of the protective effect with C20:2 involved different mechanisms than with C24:0 treatment, leading to more pronounced suppressive effects on the long-term progression of late insulitic lesions.

Systemic and local alterations in T cell subsets following treatment with {alpha}GalCer analogs

Our regimen of seven weekly injections of {alpha}GalCer was based on earlier published observations suggesting that effective prevention of diabetes in NOD mice requires multiple injections of the iNKT cell agonist (3, 4, 5, 12). Although in vivo dynamics of iNKT cells after a single injection of {alpha}GalCer are well-described (23), the outcome of responses to serial injections of iNKT cell ligands are more complex and may lead to an exaggerated contraction of cell numbers and anergy (24, 25, 26). To determine the impact of our multiple injection regimen on iNKT cells in NOD mice, we monitored their levels and phenotype in the spleen and PLN at different intervals after receiving a course of seven weekly injections of either C20:2 or C24:0 (Fig. 5). In spleen, the percentages of iNKT cells in both C20:2- and C24:0-treated animals were significantly reduced compared with vehicle-treated animals, from 1 wk up to 30 wk after the last injection (Fig. 5A, left). Although the kinetics were slower, a similar phenomenon was detected in PLN of treated mice (Fig. 5A, right). No expansion of iNKT cells was observed in either of the {alpha}GalCer-treated groups, and iNKT cell levels remained significantly reduced even at 30 wk after the cessation of treatment, indicating a long-lasting depletion of iNKT cells rather than a temporary down-regulation of their TCRs. Phenotypic studies showed that the minor fraction of iNKT cells that was resistant to depletion in the spleen at 1 wk following the last injection of {alpha}GalCer analogs had down-regulated the activation markers CD25, CD44, and CD69 (Fig. 5B), suggesting a hyporesponsive or partially anergic state. Similar results were observed in spleen and PLN at 1, 5, and 27 wk posttreatment (data not shown), indicating that this was a durable and possibly even irreversible phenomenon.


Figure 5
View larger version (28K):
[in this window]
[in a new window]

  		FIGURE 5. iNKT cell populations in {alpha}GalCer-treated NOD mice. A, The percentages of iNKT cells in the spleens (left) and PLNs (right) of treated NOD female mice were determined by FACS using staining with {alpha}GalCer C26:0-loaded mCD1d tetramers. Groups of five mice (4–5 wk of age) were treated with seven weekly injections of vehicle or glycolipids, and sacrificed 1, 5, or 30 wk after the last injection. Bars show means and SDs for groups of five mice injected with vehicle only ({square}), C24:0 (Figure 5), or C20:2 ({blacksquare}). *, Significant reductions compared with vehicle (unpaired Student’s t test, p < 0.01). B, FACS analysis of activation markers on tetramer+ splenic iNKT cells from mice in (A) sacrificed 1 wk after the last injection. Each symbol represents one animal, and horizontal lines show means for the groups. Reductions were statistically significant (unpaired Student’s t test) for both C24:0- and C20:2-treated mice compared with vehicle-treated control mice for CD25 (p < 0.001), CD44 (p < 0.01), and CD69 (p < 0.01).

 
The slowing of the late progression of insulitis that we observed in C20:2-treated mice (Fig. 4B) suggested that treatment with this analog might have reduced the expansion or homing of islet beta cell autoantigen-reactive CD4+ or CD8+ T cells to the pancreatic islets. To explore this possibility, we isolated pancreatic islets from vehicle-treated and C20:2-treated NOD mice 4 wk after cessation of treatment (i.e., 15 wk of age), and expanded the islet-infiltrating T cells by short-term culture in medium containing IL-2. Total numbers of both CD4+ and CD8+ T cells recovered were markedly reduced in the C20:2-treated animals compared with controls, whether this was calculated on a per mouse or per islet basis (Fig. 6A). To analyze specifically the beta cell-reactive T cells, we took advantage of some of the known specificities of islet-reactive CD8+ T cells corresponding to peptides derived from insulin B chain, islet-specific glucose-6-phosphatase catalytic subunit-related protein (IGRP), and dystrophia myotonica kinase) (18, 27, 28). Both ELISPOT and MHC class I tetramer analyses were used to examine the frequencies of these autoreactive CD8+ T cells in islet-infiltrating T cell populations derived from vehicle-treated and C20:2-treated NOD mice (Fig. 6, C and D). Consistent with other studies, we found that T cells reactive with IGRP peptide 206–214, or its mimotope NRP-V7, were the most prevalent among the defined populations of autoantigen-specific CD8+ T cells (18, 28). Although the relative frequencies of CD8+ T cells reactive with the three autoantigenic peptides did not differ significantly between islet-derived T cell populations from control and C20:2-treated animals, there was a substantial reduction in the total number of CD8+ T cells reactive with the three defined islet autoantigens (Fig. 6, C and D). This reduction in autoantigen-specific CD8+ T cells was demonstrated using both tetramer staining and IFN-{gamma} ELISPOT and was of similar magnitude whether quantitated on a per mouse or per islet basis.


Figure 6
View larger version (35K):
[in this window]
[in a new window]

  		FIGURE 6. Reduction of islet-infiltrating autoreactive T cells by C20:2 treatment. A, Islet-infiltrating T cells were isolated by short-term culture of islets from 15-wk-old female NOD mice (n = 7) treated with either C20:2 or vehicle (seven weekly injections beginning at 3–4 wk of age). Means and SDs for total CD4+ and CD8+ T cells obtained from each mouse (left), and per islet for each mouse (right) are displayed as stacked bar graphs. B, Representative examples of FACS and ELISPOT analyses of CD8+ T cells reactive with specific H-2Kd-restricted (Ins-I9 and NRP-V7, corresponding to epitopes from insulin B chain and IGRP, respectively) or H-2Db-restricted (MimA2, corresponding to an epitope from dystrophia myotonica kinase) self peptide epitopes. Shown are results from one representative vehicle-treated animal sacrificed at 15–16 wk of age. Numbers in dot plots are percentages of CD8+ T cells staining with peptide-loaded tetramers (circled regions). Representative images from paired analyses of the same samples by IFN-{gamma} ELISPOT are shown to the right. The same peptides were used for ELISPOT as for tetramers, except that mimotope NRP-V7 was replaced with the natural IGRP206–214 peptide for ELISPOT. C, Left, Percentages of CD8+ T cells from each animal that stained with the H-2Kd or H-2Db tetramers loaded with the indicated peptides are shown ({circ}, vehicle-treated mice; •, C20:2 treated). Right, Total number of tetramer-positive cells (sum for all three epitopes) isolated from each mouse, and the number isolated per islet from each mouse. D, Left, Frequencies of IFN-{gamma} spot-forming cells (SFC) detected using the indicated peptides (symbols as in C). Right, Total numbers of IFN-{gamma} SFC (sum for all three peptides) from each mouse, or obtained per islet from each mouse. The p values indicated in C and D were calculated by Mann-Whitney U test.

 
Effects of {alpha}GalCer treatment on DC subset and maturation anomalies in NOD mice

Previous studies have suggested that activation of iNKT cells may suppress autoimmunity in NOD mice at least partly through effects on frequency and functions of tolerogenic DCs in the PLNs (29, 30). To assess the effects of {alpha}GalCer analog treatment on DCs during diabetes progression, we monitored the levels in spleen and PLN of various DC subsets, including conventional myeloid CD11chighCD8{alpha}–, CD11chighCD8{alpha}+, and plasmacytoid CD11clow DC, during the course of the disease in treated and untreated NOD mice. As shown in Fig. 7A, we observed an alteration in the relative proportions of these three DC subsets in both spleen and PLN of control (vehicle-treated) NOD mice as these animals aged from 16 to 30 wk. Specifically, we observed that diabetes progression was accompanied by a significant augmentation of the proportion of plasmacytoid CD11clow DC, most of which were B220low (data not shown), and a significant reduction of the proportion of myeloid CD11chighCD8{alpha}– DCs (Fig. 7A). Interestingly, treatment of NOD mice during the prediabetic phase with {alpha}GalCer analogs C24:0 or C20:2 significantly reduced the evolution of DC subset alterations observed in the vehicle-treated animals at 30 wk, and restored the relative proportions of the three DC subsets to levels that approximated those present in 16 wk old prediabetic mice (Fig. 7B). This effect was observed in the PLNs of both C24:0- and C20:2-treated mice, although a trend toward greater correction of the DC imbalance was observed in the C20:2-treated animals (Fig. 7B). No posttreatment modifications were observed in splenic DCs from the same mice, indicating that {alpha}GalCer treatment specifically targeted the DC abnormality in PLNs.


Figure 7
View larger version (29K):
[in this window]
[in a new window]

  		FIGURE 7. Reversal of DC subset abnormalities in PLN of NOD mice by C20:2. A, DC subsets were identified in suspensions of spleen and PLN using FACS to gate on all CD11c+ cells, and subsequently dividing this population into three subsets based on level of CD11c expression and the presence of CD8{alpha} (left). Stacked bar graphs on the right show relative proportions of these three subsets in spleen or PLN of 16- and 30-wk-old NOD mice that were treated with seven injections of inert vehicle only from 4 to 10 wk of age. Mean values for groups of five mice are shown, and SDs (data not shown) were <5% of the means throughout. B, Long-term effects of {alpha}GalCer treatment on relative proportions of DC subsets in PLN of NOD mice. Female NOD mice received seven weekly i.p. injections of vehicle, {alpha}GalCer C24:0 or C20:2 starting at 4 wk of age. Mice were sacrificed at 30 wk of age, and relative proportions of DC subsets were analyzed by FACS as described above in A for cell suspensions from spleens (top row) and PLNs (bottom row). Each symbol represents one mouse, and statistically significant differences for {alpha}GalCer-treated groups vs vehicle control group are indicated with brackets and p values (Mann-Whitney U test).

 
Knowing that NOD mice display defective maturation of myeloid DCs (29, 31), we also evaluated the alteration of DC surface phenotype after treatment with C20:2 and C24:0. Using a single i.p. injection, we found that both forms of {alpha}GalCer were able to induce similar maturation of spleen and PLN CD11chigh DCs in prediabetic NOD mice within 24 h, as characterized by up-regulation of MHC class II and costimulatory molecules (CD86, CD80, and CD40) (Fig. 8A and additional data not shown), as has been described previously for C57BL/6 mice (32). The DC maturation was transient, as demonstrated by a return of all maturation markers to their baseline levels by 3 days after stimulation (data not shown). We next compared the maturation phenotype of CD11chigh DCs in spleen and PLN at different time points after seven weekly injections of either vehicle, C20:2 or C24:0. One day after the last of the seven injections, quantitation of MHC class II, CD86, and CD80 expression by DCs indicated the full acquisition of a maturation phenotype both in spleen and PLN (data not shown) that was not appreciably different from the DC profile obtained after a single injection (Fig. 8A). In contrast, the DCs of 30-wk-old animals (i.e., 17–19 wk after the last injection) showed persistent alterations in maturation phenotype, including up-regulation of MHC class II expression in the spleens of both C24:0- and C20:2-treated mice (Fig. 8B, left), and significant down-regulation of MHC class II and CD80 levels in the PLNs of C20:2-treated mice (Fig. 8B, right).


Figure 8
View larger version (32K):
[in this window]
[in a new window]

  		FIGURE 8. Acute and chronic effects of iNKT cell activators on DC maturation. A, Cell surface phenotype of CD11c+ DC in PLN of NOD mice sacrificed 24 h following one i.p. injection of either inert vehicle or 4 µg of C20:2. Representative FACS histograms show cell surface levels of the indicated proteins (x-axis) vs cell number (y-axis) on DCs gated according to forward and side light scatter and for high CD11c expression. Shaded histograms are DCs from PLN of vehicle-treated control mice, and open histograms with bold outline DCs from PLN of C20:2-treated mice. Results shown are representative of those obtained with five mice in each group. Similar results were obtained for splenic DCs (data not shown). B, Expression of DC maturation markers on CD11chigh cells in spleen (left) or PLN (right) of 30-wk-old NOD female mice treated from age 4 wk with seven weekly i.p. injections of either vehicle ({square}), or 4 µg of {alpha}GalCer analogs (C24:0 (Figure 8) or C20:2 ({blacksquare})). Bars represent means and SDs of the MFI values determined by FACS for five mice in each treatment group. *, Significant differences (p < 0.05, Mann-Whitney U test) compared with vehicle control group.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
Recent studies have revealed the ability to elicit more selective expression of iNKT cell functions using chemically modified forms of {alpha}GalCer (6, 8), although little information is currently available on potential applications of such compounds to the prevention or treatment of specific diseases. In the current study, we have analyzed multiple clinical, pathological, and immunological outcomes in NOD mice to evaluate the impact of {alpha}GalCer C20:2, a novel synthetic Th2-cytokine skewing iNKT cell agonist developed in our laboratories, on autoimmune diabetes. Our initial analysis of the iNKT cell-activating properties of C20:2 demonstrated several important potential advantages over other previously studied {alpha}GalCer analogs. Although ex vivo splenocyte stimulation confirmed the Th2 cytokine bias evoked by OCH, we found that C20:2 stimulated a similar cytokine bias based on ratios of IL-4 or IL-13 to IFN-{gamma}, while also leading to the production of substantially higher absolute levels of Th2-associated cytokines. In addition, C20:2 was clearly a potent ligand for murine iNKT cell TCRs, based on both functional stimulation assays and also on direct binding measurements using glycolipid-loaded CD1d tetramers. In this regard, C20:2 exhibited much greater potency for iNKT cell TCRs than OCH, which appeared to be an extremely low-avidity ligand as suggested by one previous study (20).

Most strikingly, we found that OCH showed no detectable interaction with the TCRs of human iNKT cells by any of the criteria examined, while C20:2 was a potent ligand for human iNKT cells. This leads to the conclusion that OCH and possibly other truncated sphingosine analogs of {alpha}GalCer may not be useful as immunomodulators in humans, whereas N-acyl derivatives such as C20:2 could prove to be much more promising. It should be noted that the absence of Th2 cytokine-biased responses to C20:2 by human iNKT cell clones that we observed (Fig. 2B) is expected, because studies in mice show that this cytokine bias is mainly due to the failure of C20:2 to induce the ability of iNKT cells to transactivate Th1 cytokine production by other leukocytes such as NK cells (8). Thus, the potential for inducing a Th2 cytokine bias can only be evaluated using in vitro systems that allow stimulation of complex cell mixtures such as whole splenocytes, or by in vivo activation of iNKT cells within the context of an intact host.

We also showed that the C20:2 analog was more effective for prevention of diabetes in NOD mice when compared with a related {alpha}GalCer analog (C24:0) which had similar potency for iNKT cell activation but did not stimulate a Th2 cytokine bias. The analysis of diabetes prevention reported in this study differs from previous studies in a number of important ways. First, we extended the period of observation beyond that which has been previously reported (32 wk or less), and observed animals through >45 wk to demonstrate evidence of sustained, long-term benefit. Second, we initiated therapy in most of our experiments at 4–6 wk of age, as opposed to nearly all previous studies of this type which initiated therapy with KRN7000 or OCH in 3- to 4-wk-old female NOD mice (3, 4, 5, 12, 30). Finally, our treatment protocol was also significantly less intensive in terms of frequency and duration of {alpha}GalCer administration, consisting of injections once per week for 7 wk as compared with daily or twice weekly injections for much longer than 7 wk in most published studies (3, 4, 5, 12, 30). In our view, this later initiation of treatment and less intensive regimen may more closely resemble an intervention that could be feasible in actual clinical settings, following as a model the encouraging example of beneficial sustained immunomodulation in humans with new onset diabetes using short course i.v. administration of modified anti-CD3 mAb (33).

A further point that supports the use of a less intensive regimen for iNKT cell activation is the known ability of repeated injections of KRN7000 in mice to induce a form of hyporesponsiveness or partial anergy in these T cells (24, 25). This has been associated with reduced proliferative capacity to subsequent stimulation and also with diminished cytokine secretion. In our study, we observed a very substantial reduction in iNKT cell numbers following seven weekly injections of {alpha}GalCer. This contraction of the iNKT cell population was sustained for at least 30 wk and preliminary studies suggest that it may never completely resolve during the lifespan of the animals (C. Forestier and S. A. Porcelli, unpublished data). Interestingly, in vehicle-treated NOD mice, we observed a reduction in the percentage of iNKT cells in the spleen between 12 and 17 wk of age, which was accompanied by a reciprocal rise in the percentage of iNKT cells in the PLNs of these animals during the same time period. This may have reflected an increased migration of iNKT cells during this period into the PLNs and possibly other tissues involved in the autoimmune inflammatory process. These time-dependent changes in iNKT cell numbers were significantly blunted in C20:2- and C24:0-treated animals, possibly because of a partial depletion of iNKT cells resulting from the treatment. These findings raise issues about the potential long-term implications of aggressive treatment with {alpha}GalCer and suggests that further investigations will be needed to identify the optimal approach to activating the regulatory properties of iNKT cells without leading to their prolonged or permanent depletion.

The primary mechanism by which {alpha}GalCer treatment suppresses autoimmunity in NOD mice remains a contentious issue. Initial work strongly suggested that the creation of a Th2-promoting environment, presumably due to the secretion of IL-4 and possibly other Th2-biasing cytokines by activated iNKT cells, was responsible for the beneficial effects on disease. Indeed, it has been reported that KRN7000 treatment of NOD mice leads to a Th2 cytokine-enriched environment in the spleen and PLN, and CD4+ T cells specific for islet autoantigens show enhanced Th2 cytokine production in these mice (3, 4). The relevance of this mechanism is supported by a report showing that suppression of diabetes by {alpha}GalCer is largely dependent on IL-4 (34), although this appears not to have been confirmed by all investigators (29). Our observations in the current study of improved diabetes suppression with the Th2-skewing C20:2 analog are consistent with an important role for Th2 cytokines such as IL-4 or IL-13 in the mechanism of disease prevention, as is the finding of improved outcomes in this model following treatment with the OCH analog (12).

An alternative mechanism for the action of {alpha}GalCer in diabetes prevention is the indirect effect of this compound in inducing the maturation and migration of DCs (29, 30). It is well-known that activation of iNKT cells leads to DC maturation through the actions of costimulatory molecules and cytokines, such as CD40L and IL-12 (32, 35). This may potentially overcome the demonstrable defect in DC maturation in NOD mice (36) and in humans with type 1 diabetes (37). A recent publication has highlighted the induction of regulatory or tolerogenic DCs following repeated injection of {alpha}GalCer into normal mice (38) and these appear to have predominantly a CD11chighCD8{alpha}– phenotype. This is consistent with the DC subset that has been previously found to be tolerogenic in NOD mice (39), and that was previously shown to be recruited to the PLN by {alpha}GalCer treatment (30). This latter finding was confirmed in the current study, which demonstrated the effect of {alpha}GalCer in maintaining the relative proportions of DC subsets in the PLNs of 30-wk-old NOD mice at levels similar to those seen in younger animals that lack clinically apparent autoimmune disease. Another interesting finding from our analyses was that DCs in the PLN of C20:2-treated mice showed persistent down-regulation of MHC class II and CD80, raising the possibility that {alpha}GalCer-mediated protection might be associated with the establishment of an anergic state in DCs at this key location.

A third mechanism to explain the effect of {alpha}GalCer on ameliorating autoimmunity in NOD mice is the impact of activated iNKT cells on other T cell subsets that either mediate or regulate autoimmune insulitis. For example, the CD4+CD25+ FoxP3-dependent regulatory T cell (Treg) population has recently been found to engage in significant cross-talk with iNKT cells (40). In preliminary studies, we have not observed any gross alterations in the levels of the total CD4+CD25+ T cell populations in either spleen or PLN following {alpha}GalCer treatment of NOD mice (data not shown). However, we have not studied the levels of Foxp3+ T cells in treated vs untreated animals and it remains unclear whether numbers or functions of Tregs specific for islet Ags are altered by treatment with iNKT cell activators. In contrast, the current study did provide clear evidence for a marked influence of iNKT cell activation on the autoreactive CD4+ and CD8+ effector T cell populations in the pancreatic islets of NOD mice. Although we did not evaluate in detail the effects on CD4+ autoreactive effector T cells in our studies, we observed a marked reduction in total CD4+ T cells in the T cell populations expanded from pancreatic islets of C20:2-treated NOD mice. More notably, our detailed analysis of CD8+ T cells specific for defined autoantigens showed these to be markedly reduced in frequency in C20:2-treated mice. To our knowledge, this is the first study to assess the impact of iNKT cell activation with {alpha}GalCer on the frequency of natural nontransgenic CD8+ T cells with defined beta cell autoantigen specificities in the pancreatic islet infiltrates. Our studies have not yet compared the relative efficacy of C20:2 with non-Th2-biasing {alpha}GalCer analogs with regard to the impact on autoreactive CD8+ T cells in the islets, which will be an interesting focus for detailed future studies in this area.

In summary, we have confirmed the potential regulatory role of iNKT cells in type 1 diabetes and provide further support for the development of novel {alpha}GalCer compounds as therapeutic immunomodulators for prevention and treatment of this disease. Our findings also emphasize the beneficial influence of a Th2 cytokine environment in diabetes-prone animals, which was apparent from the improved clinical and immunological outcomes achieved with the Th2 cytokine-skewing C20:2 analog. Based on the results that we have obtained to date comparing C20:2 with OCH, the other well-studied {alpha}GalCer analog known to generate Th2 cytokine-biased responses, we propose that C20:2 is a more promising potential therapeutic agent because of its high avidity for the TCRs of human iNKT cells. Our studies have not yet assessed whether the Th2-skewing property of C20:2 is conserved in other mammals besides mice, particularly nonhuman primates and humans, which will be an important next step in determining the potential of C20:2 as an agent for prevention or therapy of diabetes or other human autoimmune diseases.


   Acknowledgments
 
We thank Dr. C. Schechter (Department of Epidemiology and Population Health, Albert Einstein College of Medicine (AECOM), Bronx, NY) for expert assistance with statistical analysis of Fig. 4B. MimA2-loaded H-2Db tetramers used in this study were provided by the National Institutes of Health/National Institute of Allergy and Infectious Diseases Tetramer Core Facility. We acknowledge the assistance of the Flow Cytometry Core Facilities supported by the AECOM Center for AIDS Research and the AECOM Cancer Center.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
S. A. Porcelli is a consultant for Vaccinex and is the inventor on a patent application for the development of iNKT cell activators. The patent is the property of Albert Einstein College of Medicine (AECOM). Certain aspects of the patent rights have been licensed by AECOM to Vaccinex.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 This work was supported by National Institutes of Health Grants RO1 AI45889 and RO1 AI064424 (to S.A.P.), RO1 AI064422 and RO1 DK064315 (to T.P.D.), RO1 AI057519 (to A.R.H.), and a Pilot and Feasibility Award (to S.A.P. and T.P.D.) from the Albert Einstein College of Medicine Diabetes Research and Training Center (DK20541). S.A.P. is the recipient of an Irma T. Hirschl Career Scientist Award. C.F. is the recipient of a Human Frontier Science Program long-term fellowship. G.S.B. is a former Lister Institute-Jenner Research Fellow and acknowledges support from the Medical Research Council (United Kingdom), the Wellcome Trust, and from the James Bardrick Research Chair. P.S. is supported by the Canadian Institutes of Health Research, and is a Scientist of the Alberta Heritage Foundation for Medical Research. Flow cytometry studies were supported by the FACS Core Facilities of the Albert Einstein College of Medicine Center for AIDS Research (National Institutes of Health/National Institute of Allergy and Infectious Diseases AI51519) and the Albert Einstein College of Medicine Cancer Center (National Institutes of Health/National Cancer Institute P30 CA13330). Back

2 C.F. and T.T. contributed equally to this work. Back

3 Address correspondence and reprint requests to Dr. Steven A. Porcelli, Department of Microbiology and Immunology, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, NY 10461. E-mail address: porcelli{at}aecom.yu.edu Back

4 Abbreviations used in this paper: iNKT, invariant NKT; {alpha}GalCer, {alpha}-galactosylceramide; PLN, pancreatic lymph node; DC, dendritic cell; IGRP, islet-specific glucose-6-phosphatase catalytic subunit-related protein; Treg, regulatory T cell. Back

Received for publication September 22, 2006. Accepted for publication November 14, 2006.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Anderson, M. S., J. A. Bluestone. 2005. The NOD mouse: a model of immune dysregulation. Annu. Rev. Immunol. 23: 447-485. [Medline]
  2. Yu, K. O., S. A. Porcelli. 2005. The diverse functions of CD1d-restricted NKT cells and their potential for immunotherapy. Immunol. Lett. 100: 42-55. [Medline]
  3. Hong, S., M. T. Wilson, I. Serizawa, L. Wu, N. Singh, O. V. Naidenko, T. Miura, T. Haba, D. C. Scherer, J. Wei, et al 2001. The natural killer T-cell ligand {alpha}-galactosylceramide prevents autoimmune diabetes in non-obese diabetic mice. Nat. Med. 7: 1052-1056. [Medline]
  4. Sharif, S., G. A. Arreaza, P. Zucker, Q. S. Mi, J. Sondhi, O. V. Naidenko, M. Kronenberg, Y. Koezuka, T. L. Delovitch, J. M. Gombert, et al 2001. Activation of natural killer T cells by {alpha}-galactosylceramide treatment prevents the onset and recurrence of autoimmune type 1 diabetes. Nat. Med. 7: 1057-1062. [Medline]
  5. Wang, B., Y. B. Geng, C. R. Wang. 2001. CD1-restricted NK T cells protect nonobese diabetic mice from developing diabetes. J. Exp. Med. 194: 313-320. [Abstract/Free Full Text]
  6. Miyamoto, K., S. Miyake, T. Yamamura. 2001. A synthetic glycolipid prevents autoimmune encephalomyelitis by inducing TH2 bias of natural killer T cells. Nature 413: 531-534. [Medline]
  7. Goff, R. D., Y. Gao, J. Mattner, D. Zhou, N. Yin, C. Cantu, III, L. Teyton, A. Bendelac, P. B. Savage. 2004. Effects of lipid chain lengths in {alpha}-galactosylceramides on cytokine release by natural killer T cells. J. Am. Chem. Soc. 126: 13602-13603. [Medline]
  8. Yu, K. O., J. S. Im, A. Molano, Y. Dutronc, P. A. Illarionov, C. Forestier, N. Fujiwara, I. Arias, S. Miyake, T. Yamamura, et al 2005. Modulation of CD1d-restricted NKT cell responses by using N-acyl variants of {alpha}-galactosylceramides. Proc. Natl. Acad. Sci. USA 102: 3383-3388. [Abstract/Free Full Text]
  9. Schmieg, J., G. Yang, R. W. Franck, M. Tsuji. 2003. Superior protection against malaria and melanoma metastases by a C-glycoside analogue of the natural killer T cell ligand {alpha}-galactosylceramide. J. Exp. Med. 198: 1631-1641. [Abstract/Free Full Text]
  10. Fujio, M., D. Wu, R. Garcia-Navarro, D. D. Ho, M. Tsuji, C. H. Wong. 2006. Structure-based discovery of glycolipids for CD1d-mediated NKT cell activation: tuning the adjuvant versus immunosuppression activity. J. Am. Chem. Soc. 128: 9022-9023. [Medline]
  11. Chiba, A., S. Oki, K. Miyamoto, H. Hashimoto, T. Yamamura, S. Miyake. 2004. Suppression of collagen-induced arthritis by natural killer T cell activation with OCH, a sphingosine-truncated analog of {alpha}-galactosylceramide. Arthritis Rheum. 50: 305-313. [Medline]
  12. Mizuno, M., M. Masumura, C. Tomi, A. Chiba, S. Oki, T. Yamamura, S. Miyake. 2004. Synthetic glycolipid OCH prevents insulitis and diabetes in NOD mice. J. Autoimmun. 23: 293-300. [Medline]
  13. Ueno, Y., S. Tanaka, M. Sumii, S. Miyake, S. Tazuma, M. Taniguchi, T. Yamamura, K. Chayama. 2005. Single dose of OCH improves mucosal T helper type 1/T helper type 2 cytokine balance and prevents experimental colitis in the presence of valpha14 natural killer T cells in mice. Inflamm. Bowel Dis. 11: 35-41. [Medline]
  14. Oki, S., A. Chiba, T. Yamamura, S. Miyake. 2004. The clinical implication and molecular mechanism of preferential IL-4 production by modified glycolipid-stimulated NKT cells. J. Clin. Invest. 113: 1631-1640. [Medline]
  15. Oki, S., C. Tomi, T. Yamamura, S. Miyake. 2005. Preferential Th2 polarization by OCH is supported by incompetent NKT cell induction of CD40L and following production of inflammatory cytokines by bystander cells in vivo. Int. Immunol. 17: 1619-1629. [Abstract/Free Full Text]
  16. Exley, M., J. Garcia, S. P. Balk, S. Porcelli. 1997. Requirements for CD1d recognition by human invariant V{alpha}24+ CD4–CD8– T cells. J. Exp. Med. 186: 109-120. [Abstract/Free Full Text]
  17. Ndonye, R. M., D. P. Izmirian, M. F. Dunn, K. O. Yu, S. A. Porcelli, A. Khurana, M. Kronenberg, S. K. Richardson, A. R. Howell. 2005. Synthesis and evaluation of sphinganine analogues of KRN7000 and OCH. J. Org. Chem. 70: 10260-10270. [Medline]
  18. Lieberman, S. M., A. M. Evans, B. Han, T. Takaki, Y. Vinnitskaya, J. A. Caldwell, D. V. Serreze, J. Shabanowitz, D. F. Hunt, S. G. Nathenson, et al 2003. Identification of the beta cell antigen targeted by a prevalent population of pathogenic CD8+ T cells in autoimmune diabetes. Proc. Natl. Acad. Sci. USA 100: 8384-8388. [Abstract/Free Full Text]
  19. Im, J. S., K. O. Yu, P. A. Illarionov, K. P. LeClair, J. R. Storey, M. W. Kennedy, G. S. Besra, S. A. Porcelli. 2004. Direct measurement of antigen binding properties of CD1 proteins using fluorescent lipid probes. J. Biol. Chem. 279: 299-310. [Abstract/Free Full Text]
  20. Stanic, A. K., R. Shashidharamurthy, J. S. Bezbradica, N. Matsuki, Y. Yoshimura, S. Miyake, E. Y. Choi, T. D. Schell, L. Van Kaer, S. S. Tevethia, et al 2003. Another view of T cell antigen recognition: cooperative engagement of glycolipid antigens by Va14Ja18 natural T (iNKT) cell receptor. [Published erratum appears in 2004 J. Immunol. 172: 717.]. J. Immunol. 171: 4539-4551. [Abstract/Free Full Text]
  21. Sidobre, S., K. J. Hammond, L. Benazet-Sidobre, S. D. Maltsev, S. K. Richardson, R. M. Ndonye, A. R. Howell, T. Sakai, G. S. Besra, S. A. Porcelli, M. Kronenberg. 2004. The T cell antigen receptor expressed by V{alpha}14i NKT cells has a unique mode of glycosphingolipid antigen recognition. Proc. Natl. Acad. Sci. USA 101: 12254-12259. [Abstract/Free Full Text]
  22. Brossay, L., M. Kronenberg. 1999. Highly conserved antigen-presenting function of CD1d molecules. Immunogenetics 50: 146-151. [Medline]
  23. Wilson, M. T., C. Johansson, D. Olivares-Villagomez, A. K. Singh, A. K. Stanic, C. R. Wang, S. Joyce, M. J. Wick, L. Van Kaer. 2003. The response of natural killer T cells to glycolipid antigens is characterized by surface receptor down-modulation and expansion. Proc. Natl. Acad. Sci. USA 100: 10913-10918. [Abstract/Free Full Text]
  24. Parekh, V. V., M. T. Wilson, D. Olivares-Villagomez, A. K. Singh, L. Wu, C. R. Wang, S. Joyce, L. Van Kaer. 2005. Glycolipid antigen induces long-term natural killer T cell anergy in mice. J. Clin. Invest. 115: 2572-2583. [Medline]
  25. Uldrich, A. P., N. Y. Crowe, K. Kyparissoudis, D. G. Pellicci, Y. Zhan, A. M. Lew, P. Bouillet, A. Strasser, M. J. Smyth, D. I. Godfrey. 2005. NKT cell stimulation with glycolipid antigen in vivo: costimulation-dependent expansion, Bim-dependent contraction, and hyporesponsiveness to further antigenic challenge. J. Immunol. 175: 3092-3101. [Abstract/Free Full Text]
  26. Ikarashi, Y., A. Iizuka, Y. Koshidaka, Y. Heike, Y. Takaue, M. Yoshida, M. Kronenberg, H. Wakasugi. 2005. Phenotypical and functional alterations during the expansion phase of invariant V{alpha}14 natural killer T (V{alpha}14i NKT) cells in mice primed with {alpha}-galactosylceramide. Immunology 116: 30-37. [Medline]
  27. Lieberman, S. M., T. P. DiLorenzo. 2003. A comprehensive guide to antibody and T-cell responses in type 1 diabetes. Tissue Antigens 62: 359-377. [Medline]
  28. Lieberman, S. M., T. Takaki, B. Han, P. Santamaria, D. V. Serreze, T. P. DiLorenzo. 2004. 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. 173: 6727-6734. [Abstract/Free Full Text]
  29. Chen, Y. G., C. M. Choisy-Rossi, T. M. Holl, H. D. Chapman, G. S. Besra, S. A. Porcelli, D. J. Shaffer, D. Roopenian, S. B. Wilson, D. V. Serreze. 2005. Activated NKT cells inhibit autoimmune diabetes through tolerogenic recruitment of dendritic cells to pancreatic lymph nodes. J. Immunol. 174: 1196-1204. [Abstract/Free Full Text]
  30. Naumov, Y. N., K. S. Bahjat, R. Gausling, R. Abraham, M. A. Exley, Y. Koezuka, S. B. Balk, J. L. Strominger, M. Clare-Salzer, S. B. Wilson. 2001. Activation of CD1d-restricted T cells protects NOD mice from developing diabetes by regulating dendritic cell subsets. Proc. Natl. Acad. Sci. USA 98: 13838-13843. [Abstract/Free Full Text]
  31. Dahlen, E., G. Hedlund, K. Dawe. 2000. Low CD86 expression in the nonobese diabetic mouse results in the impairment of both T cell activation and CTLA-4 up-regulation. J. Immunol. 164: 2444-2456. [Abstract/Free Full Text]
  32. Fujii, S., K. Shimizu, H. Hemmi, M. Fukui, A. J. Bonito, G. Chen, R. W. Franck, M. Tsuji, R. M. Steinman. 2006. Glycolipid {alpha}-C-galactosylceramide is a distinct inducer of dendritic cell function during innate and adaptive immune responses of mice. Proc. Natl. Acad. Sci. USA 103: 11252-11257. [Abstract/Free Full Text]
  33. Herold, K. C., S. E. Gitelman, U. Masharani, W. Hagopian, B. Bisikirska, D. Donaldson, K. Rother, B. Diamond, D. M. Harlan, J. A. Bluestone. 2005. A single course of anti-CD3 monoclonal antibody hOKT3{gamma}1(Ala-Ala) results in improvement in C-peptide responses and clinical parameters for at least 2 years after onset of type 1 diabetes. Diabetes 54: 1763-1769. [Abstract/Free Full Text]
  34. Mi, Q. S., D. Ly, P. Zucker, M. McGarry, T. L. Delovitch. 2004. Interleukin-4 but not interleukin-10 protects against spontaneous and recurrent type 1 diabetes by activated CD1d-restricted invariant natural killer T-cells. Diabetes 53: 1303-1310. [Abstract/Free Full Text]
  35. Brigl, M., L. Bry, S. C. Kent, J. E. Gumperz, M. B. Brenner. 2003. Mechanism of CD1d-restricted natural killer T cell activation during microbial infection. Nat. Immunol. 4: 1230-1237. [Medline]
  36. Boudaly, S., J. Morin, R. Berthier, P. Marche, C. Boitard. 2002. Altered dendritic cells (DC) might be responsible for regulatory T cell imbalance and autoimmunity in nonobese diabetic (NOD) mice. Eur. Cytokine Netw. 13: 29-37. [Medline]
  37. Jansen, A., M. van Hagen, H. A. Drexhage. 1995. Defective maturation and function of antigen-presenting cells in type 1 diabetes. Lancet 345: 491-492. [Medline]
  38. Kojo, S., K. Seino, M. Harada, H. Watarai, H. Wakao, T. Uchida, T. Nakayama, M. Taniguchi. 2005. Induction of regulatory properties in dendritic cells by V{alpha}14 NKT cells. J. Immunol. 175: 3648-3655. [Abstract/Free Full Text]
  39. Feili-Hariri, M., X. Dong, S. M. Alber, S. C. Watkins, R. D. Salter, P. A. Morel. 1999. Immunotherapy of NOD mice with bone marrow-derived dendritic cells. Diabetes 48: 2300-2308. [Abstract]
  40. La Cava, A., L. Van Kaer, D. S. Fu. 2006. CD4+CD25+ Tregs and NKT cells: regulators regulating regulators. Trends Immunol. 27: 322-327. [Medline]



This article has been cited by other articles:


Home page
 J. Immunol.Home page
J. A. Berzofsky and M. Terabe
NKT Cells in Tumor Immunity: Opposing Subsets Define a New Immunoregulatory Axis
J. Immunol., March 15, 2008; 180(6): 3627 - 3635.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
D. Torres, C. Paget, J. Fontaine, T. Mallevaey, T. Matsuoka, T. Maruyama, S. Narumiya, M. Capron, P. Gosset, C. Faveeuw, et al.
Prostaglandin D2 Inhibits the Production of IFN-{gamma} by Invariant NK T Cells: Consequences in the Control of B16 Melanoma
J. Immunol., January 15, 2008; 180(2): 783 - 792.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Gastrointest. Liver Physiol.Home page
L. Wu and L. V. Kaer
Role of NKT Cells in the Digestive System. II. NKT cells and diabetes
Am J Physiol Gastrointest Liver Physiol, November 1, 2007; 293(5): G919 - G922.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
M. W.L. Teng, J. A. Westwood, P. K. Darcy, J. Sharkey, M. Tsuji, R. W. Franck, S. A. Porcelli, G. S. Besra, K. Takeda, H. Yagita, et al.
Combined Natural Killer T-Cell Based Immunotherapy Eradicates Established Tumors in Mice
Cancer Res., August 1, 2007; 67(15): 7495 - 7504.
[Abstract] [Full Text] [PDF]

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
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Forestier, C.
Right arrow Articles by Porcelli, S. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Forestier, C.
Right arrow Articles by Porcelli, S. A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS