Interleukin-13 Pathway Alterations Impair Invariant Natural Killer T-Cell–Mediated Regulation of Effector T Cells in Type 1 Diabetes

Type 1 diabetes (T1D) results from an autoimmune attack that leads to the destruction of insulin-secreting pancreatic β-cells. Infiltration of islets by T cells, B cells, and macrophages (insulitis) is a pathognomonic sign for T1D at disease onset (1,2) as is that of autoantibodies toward islet proteins (3). β-cell destruction is suggested to be mediated by CD4⁺ and CD8⁺ T cells (4), but the mechanisms (likely multifactorial) are not completely understood. Despite a strong emphasis placed on the role of effector T cells (Teffs) in disease development and progression, deficiencies in multiple immune pathways have also been associated with T1D, such as decreased regulation by regulatory T cells (Tregs) or defects in the invariant natural killer T (iNKT)–cell population (5).

Type I or iNKT cells express markers for natural killer and T-cell lineages with different characteristics from conventional T cells: 1) iNKT cells specifically recognize lipid-based antigens presented by the MHC-I–like CD1d molecule; 2) virtually all human iNKT cells express a T-cell receptor (TCR), with the Vα24Jα18 chain mostly paired with Vβ11; and 3) iNKT cells express an activated/memory phenotype so that they can rapidly exert effector functions without further TCR activation (6). iNKT cells from human peripheral blood mononuclear cells (PBMCs) in healthy donors (HDs) typically range from 0.04% to 1.3% of the lymphocytes (7). In addition, human iNKT cells have been subdivided by their expression of CD4 or CD8, defining CD4⁺, CD8⁺, and CD4⁻CD8⁻ double-negative iNKT cell subsets constituting 27%, 24%, and 49% of peripheral iNKT cells, respectively (8). These populations show different cytokine profiles (9).

Although many studies have shown that iNKT cells promote immunity to pathogens, their capacity to secrete interleukin (IL)-4 suggests a potential regulatory function of some autoimmune responses. In the NOD mouse model of T1D, iNKT cells provide a protective effect by releasing IL-4, inducing T-helper 2–type responses to islet autoantigens (10,11). Adoptive transfer experiments revealed that iNKT cells potently inhibit the differentiation of islet-specific transferred Teffs in the pancreatic lymph node (12). This was explained by the ability of iNKT cells to recruit tolerogenic dendritic cells favoring the presence of Tregs (13). These were necessary for iNKT cells to transfer protection from T1D (14). In contrast, data from human T1D are scarce. Clinical studies have shown low frequencies and functional defects of iNKT cells that might contribute to disease development, although no consensus has been reached about the nature of iNKT cell defects (7). Some studies pointed to similar defects of iNKT cells in patients with T1D and at-risk groups (15,16).

Following activation, iNKT cells promptly produce large amounts of various cytokines, thereby providing signals to other immune cells, including dendritic cells, natural killer cells, and conventional T and B cells (17,18). The strength of TCR downstream signaling, the integration of cytokine receptor signaling, and the cellular qualities of the lipid-presenting antigen-presenting cells (APCs) have been proposed to determine the expression of particular transcription factors favoring specific iNKT effector phenotypes (19–21).

The composition of the T-cell cytokine profile is an important factor in the outcome of the autoimmune response. Proinflammatory cytokines are abundantly secreted by in situ Teff lymphocytes and macrophages in T1D, whereas the secretion of anti-inflammatory cytokines is not predominant (22). IL-13 is a cytokine primarily produced by T-helper 2 lymphocytes, with powerful anti-inflammatory properties that retard or prevent T1D in NOD mice (23). IL-13 binds to two cell surface receptors: type 1 (IL13Rα1/IL4Rα) and type 2 (IL13Rα2). The IL13Rα1 monomer binds IL-13 before dimerizing with the IL-4Rα chain to transduce intracellular signals through STAT6 phosphorylation (24). IL13Rα2 binds IL-13 with higher affinity than IL13Rα1, but its signaling properties are less known. Reports have shown that IL13Rα2 binds to several intracellular signaling molecules, suggesting specific signaling functions (25–27).

We hypothesized that iNKT cells play a regulatory role in human T1D and investigated the cellular requirements and molecular mechanisms underlying iNKT-cell–mediated regulation in HDs and patients with T1D. We demonstrate that 1) iNKT cells have regulatory effects on Teffs, 2) the regulatory mechanism is mediated by their ability to secrete IL-13, 3) suppression by iNKT cells is impaired in patients with T1D compared with control subjects, and 4) the reduced regulatory capacity of iNKT cells from patients with T1D correlates with defective IL-13 production. The results show that a diminished IL-13 secretion by T1D-derived iNKT cells may explain the impaired regulation of the autoimmune response described in human T1D.

Teffs and iNKT cells were purified from PBMCs from HDs (n = 14) (used as control samples) and patients with T1D (n = 18) (Table 1). Heparinized blood samples were taken after participants signed an informed consent approved by the ethical committees of the relevant institutions. Blood was drawn from patients at disease clinical onset before receiving any treatment.

iNKT cells were obtained from PBMCs by Ficoll-Paque density gradient centrifugation. iNKT cells were selected based on the expression of the TCR α-chain Vα24Jα18 (anti-iNKT MicroBeads, human; Miltenyi Biotec) using the autoMACS system (Miltenyi Biotec). A fraction of the iNKT purified cells was cultured in Iscove’s modified Dulbecco’s medium supplemented with l-glutamine (2 mmol/L), penicillin (100 units/mL), streptomycin (100 μg/mL) (all from Sigma-Aldrich), and human serum (8%). iNKT cells were expanded in the presence of 30-Gy irradiated autologous PBMCs (3 × 10⁶ cells/well), recombinant human (rh) IL-2 (20 units/mL; National Institutes of Health), and 3 × 10⁵ cells/well of irradiated C1Rd cells (lymphoblastoid B-cell line C1R stably expressing CD1d) previously pulsed with 100 ng/mL of α-galactosylceramide (αGC) for 1 h at 37°C (Enzo Life Sciences). After 3 days of culture, the medium was replaced with fresh medium containing rhIL-2 (20 units/mL). Cells were expanded for a minimum of 2 weeks before being used in the assays. Cells were phenotyped by using specific monoclonal antibodies purified by cell sorting and cryopreserved until use. Teffs (CD4⁺ CD25⁻) were isolated in a two-step negative selection process: first based on CD4 expression and then on the absence of CD25 by using the CD4⁺ T Cell Isolation Kit II, human, and the CD25 MicroBeads, human, kit (Miltenyi Biotec).

Cell phenotype was analyzed by flow cytometry (FACSCanto II; BD Biosciences) by using the following specific monoclonal antibodies for iNKT cells: CD3-PE-Cy7 (UCHT1; BioLegend), TCR α-chain Vα24Jα18-PE (6B11; Miltenyi Biotec), CD4α-APC (RPA-T4; BioLegend), CD8β-PE-Cy5 (2ST8.SH7; Beckman Coulter), and TCRVβ11-FITC (C21; Beckman Coulter). Purity of the Vα24Jα18CD3Vβ11⁺ cells before and after in vitro expansion was typically >98%. Two combinations of monoclonal antibodies were used for Teff characterization, following manufacturers’ protocol: 1) CD3-Alexa Fluor 488 (UCHT1; BD Pharmingen), CD4-PERCP (SK3; BD Biosciences), and CD25-PE (3G10; Miltenyi Biotec) and 2) CD4-PERCP, CD25-PE, and Foxp3-Alexa Fluor 488 (PCH101; eBioscience). The purity of CD3⁺CD4⁺CD25⁻Foxp3⁻ Teffs was 80% on average. IL-13 secretion was measured by using the anti-IL-13 antibody (JES10-5A2.2; BioLegend). Expression of IL-13 receptors was measured using specific antibodies to IL13Rα1 (SS12B; BioLegend) and IL13Rα2 (SHM38; BioLegend).

iNKT-cell function was assayed by measuring tumor necrosis factor-α (TNF-α) secretion in cultures with 30-Gy–irradiated autologous PBMCs (1 × 10⁵ cells/well), 45-Gy–irradiated C1R⁻ (mock transfected C1R cells), or C1Rd cells (2 × 10⁴ cells/well) as APCs pulsed for 1 h at 37°C with αGC. As positive controls, cells were stimulated with αCD3/CD28-coated beads (Dynabeads; Life Technologies). Culture supernatants were collected after 48 h. TNF-α production was measured by ELISA (TNF-α Human Antibody Pair; Invitrogen, Life Technologies). Concentration was quantified in a VICTOR3 Multilabel Plate Reader system (PerkinElmer).

iNKT-cell regulatory properties were analyzed by in vitro suppression assays. Various ratios of iNKT:Teffs (0.25:1, 0.5:1, 1:1, 2:1 [where 1 = 10⁴ cells/well]) were seeded in triplicate in 96-well plates (Nunc; Labclinics) and stimulated with anti-CD3/CD28–coated beads (cell:bead ratio 2:1) (Dynabeads; Life Technologies) and autologous irradiated PBMCs (5 × 10⁴ cells/well). Cultures of each cell type (1 × 10⁴ cells/well) were used as controls. Plates were incubated 4 days at 37°C in 5% CO₂. Teff proliferation was measured by ³H-thymidine incorporation pulsing with 1 μCi/well of [methyl-³H] thymidine (PerkinElmer) 16 h before harvesting and analysis on a β-counter (Wallac TriLux MicroBeta 1450 liquid scintillation luminescence counter). Percent suppression was calculated as follows: 100 − [(100 × cpm ratio iNKT:Teff)/cpm Teffs], where cpm is counts per minute. All cultures were done in triplicate.

The requirement for cell-cell contact was assayed by using a Transwell system. Autologous 30-Gy–irradiated PBMCs (4 × 10⁵ cells/well) were cultured in Transwell 24-well plates (0.4-μm pore; Labclinics) by using a 1:1 iNKT:Teffs ratio, where 1 = 10⁵ cells. Teffs were seeded on the lower chamber, whereas iNKT cells were added to the lower or the upper chamber together with feeder cells. As controls, each cell type was cultured alone. After the aforementioned stimulation, 24-h and 48-h supernatants were collected to analyze cytokine secretion.

To confirm the role of IL-13, an anti-IL-13 neutralizing antibody (JES10-5A2) was added to the suppression assays. iNKT cells and Teffs alone or mixed at various ratios were incubated in the presence of increasing concentrations of the antibody or 30 μg/mL control IgG₁ (Rat IgG₁, κ; BioLegend). Culture supernatants were collected after 48 h. For some experiments, various concentrations of rhIL-13 (BioLegend) were added to the Teff cultures. In all cases, αCD3CD28-coated beads were used as stimulus. Teff proliferation was measured by ³H-thymidine incorporation pulsing after 4-day culture in all conditions.

Cytokine secretion was measured by Cytometric Bead Array (CBA) (BD Biosciences) on 24-h culture supernatants (IL-2) or 48-h culture supernatants (TNF-α, interferon-γ [IFN-γ], IL-4, IL-10, IL-13, and IL-17A). Data were analyzed with FCAP Array Software Version 3.0 (BD Biosciences). For intracellular staining, iNKT cells were incubated with 20 ng/mL phorbol myristic acid and 1 μg/mL ionomycin for 6 h at 37°C, and 1 μg/mL brefeldin A was added the last 4 h. Cells were collected, fixed with 4% paraformaldehyde, and permeabilized with 0.1% saponin. Cells were blocked with human serum plus 0.3% saponin before an anti-IL-13 antibody was added. IL-13 production was measured by flow cytometry.

RNA was extracted from iNKT-cell and Teff subpopulations from PBMCs of HDs and patients with T1D. RNA was extracted by using the RNeasy Plus Mini Kit (QIAGEN) and subsequently quantified with a NanoDrop 1000 spectrophotometer (Thermo Scientific). Integrity of RNA was assessed with the Agilent 2100 Bioanalyzer and LabChip (Agilent Technologies). Samples with integrity of RNA >7 were considered for the experiments. cDNA was synthesized from total RNA using Superscript III (200 units/μL; Invitrogen) according to manufacturer protocol.

IL-13, IL13Rα1, and IL13Rα2 expression was analyzed by RT-PCR with previously described primers (28–30). Primers specific for CD3γ were forward 5′-CCCAATGACCAGCTCTACCA-3′ and reverse 5′-GGAACTGAATAGGAGGAGAACAC-3′. PCRs were performed at an annealing temperature of 59°C for IL13Rα1 and IL-13, 57°C for IL13Rα2, and 58°C for CD3γ. Primers to characterize iNKT cell TCR were 1) Vα24 (forward 5′-CTGGATGCAGAACAAAGCAGAGC-3′) and the Jα18 TCR segment (reverse 5′-AGGCCAGACAGTCAACTGAGTTCC-3′) and 2) Vβ11 (forward 5′-ACAGTCTCCAGAATAAGGACG-3′) and the Cβ TCR segment (reverse 5′-CTTCTGATGGCTCAAACA-3′). PCRs were performed at an annealing temperature of 65°C for the TCRα chain and 55°C for the TCRβ chain (31). PCR products were electrophoresed on 2% agarose gel, purified by using the illustra GFX PCR DNA and Gel Band Purification Kit (GE Healthcare Life Sciences) and sequenced at the Genomics and Bioinformatics Service at Universitat Autònoma de Barcelona (Applied Biosystems 3130xl genetic analyzer) using the BigDye Terminator v1.1 or v3.1 Cycle Sequencing Kit, depending on the base pair numbers for each fragment.

To quantify the expression of IL13Rα1 and IL13Rα2, gels were analyzed by densitometry using GelAnalyzer 2010 software. Each band was quantified by extrapolating from a standard concentration curve of known molecular weight fragments (molecular weight and mass of 100-bp DNA ladder; Genegraft). Relative receptor expression by T cells was normalized with the CD3γ expression.

IL-13 receptor sequences were analyzed with Basic Local Alignment Search Tool and ClustalW2 alignment software. Sequence assignment for the V and J segments of the TCRα and TCRβ chains was performed with IMGT V-QUEST software (ImMunoGeneTics).

Statistical analysis was done with Student t or one-way ANOVA tests by using GraphPad Prism 4 software. P < 0.05 was considered statistically significant.

The capacity of human iNKT cells to regulate Teff proliferation was tested. PBMC-derived iNKT cells were cultured with CD4⁺CD25⁻ autologous Teff at various ratios. Regulation was assessed on a cell proliferation assay upon stimulation with anti-CD3/CD28–coated beads. Teff proliferation was inhibited when iNKT cells were added to the culture (Fig. 1A, left). At a high iNKT:Teff ratio, inhibition reached 40%, and the difference was statistically significant (Fig. 1A, right). Proliferation of iNKT cells alone was similar to that of feeder cells.

We next analyzed the suppression function of in vitro expanded iNKT cells. Expression of TCRVα24Jα18/Vβ11 was measured by flow cytometry (Supplementary Fig. 1A) and αGC specificity and CD1d restriction by TNF-α production (Supplementary Fig. 1B). iNKT cells obtained and expanded from control PBMCs (n = 9) maintained their capacity to suppress Teff proliferation. A statistically significant higher suppression activity was observed at a 2:1 iNKT:Teff ratio compared with proliferation by Teffs alone (Fig. 1B, left). Therefore, we demonstrate that human iNKT cells suppress the proliferation of Teffs.

To define the mechanism behind iNKT cell–mediated suppression, production of IL-2 was measured in iNKT:Teff coculture supernatants. As shown in Fig. 2, IL-2 secretion depended on the activation of Teffs, and the addition of iNKT cells to the cultures significantly reduced the secretion. Irradiation of Teffs abrogated IL-2 production; however, irradiated iNKT cells did not suppress IL-2 production by Teffs. The correlation between a decreased Teff proliferation in the presence of iNKT cells and the reduction on IL-2 secretion in the cocultures indicates that one mechanism of iNKT cell suppression is the inhibition of IL-2 secretion by Teffs.

To assess cell-cell contact requirements for suppression, proliferation was assayed on a Transwell system. iNKT cells were alternatively loaded on the upper or the lower chamber of the Transwells where Teffs were seeded. IL-2 secretion was inhibited by iNKT cells in a cell-cell contact–independent manner, that is, IL-2 reduction was equivalent whether iNKT cells were loaded on the upper or the lower chamber (Fig. 2). This indicates that suppression of Teff proliferation by iNKT cells is mediated by a soluble factor secreted by iNKT cells.

To find out whether cytokines mediate suppression, we measured cytokine secretion in iNKT:Teff cocultures and single-cell type cultures. Individual stimulation of Teffs and iNKT cells produced an array of cytokines, including IFN-γ and TNF-α (Fig. 3A) but not IL-17A or IL-10 (data not shown). Of note, iNKT cells secreted high levels of immunomodulatory cytokines IL-4 and IL-13, which increased in iNKT:Teff cocultures (Fig. 3B) in a cell-cell contact–independent manner. The most prominent increase was on IL-13 production. No IL-13 was produced by unstimulated ex vivo or in vitro expanded iNKT cells (data not shown).

To assess the importance of IL-13 involvement on suppression, we measured proliferation in iNKT:Teff cocultures, adding increasing concentrations of an IL-13–blocking antibody (Fig. 4A). The addition of the antibody partially restored proliferation of Teffs, whereas no recovery was observed if an irrelevant antibody was added instead. The same pattern was observed for IL-2 production by the same Teffs (Fig. 4B). These results strongly indicate that IL-13 is responsible for the suppression. Further experiments showed that the addition of anti-IL-4 to these cocultures does not modify the effect of anti-IL-13 antibody (Fig. 4C).

We next analyzed the suppressor capacity of iNKT cells isolated at disease onset in patients with T1D. In vitro expanded iNKT cells from patients showed an impaired ability to suppress the proliferation of autologous Teffs as measured by a negative percentage of cell suppression (Fig. 5A and B). Of note, at a 2:1 iNKT:Teff ratio, at which HD-derived iNKT cells reached a mean Teff suppression of 66%, T1D-derived iNKT cells did not have an effect on the proliferation of autologous Teffs, and the differences were statistically significant (Fig. 5C). No correlation was found between the suppression capacity of each patient’s iNKT cells and age of disease onset.

We then investigated whether the lack of suppression was due to an impairment of patient iNKT cells or Teffs. The control iNKT cells inhibited the proliferation of Teffs from patients with T1D at similar levels as they inhibited control Teffs. In addition, T1D-derived iNKT cells were unable to suppress the proliferation of allogeneic control Teffs. Therefore, suppression impairment is not due to a defect on patients’ Teffs but to an intrinsic defect of iNKT cells from patients with T1D (Fig. 6A and B).

Because regulation by iNKT cells depended on IL-13 secretion, we measured the T1D-derived iNKT-cell IL-13 secretion in culture supernatants from stimulated HD and patient iNKT cells. The results showed that iNKT cells from patients with T1D secreted significantly lower levels of IL-13 (Fig. 7A, left). Analysis by flow cytometry by intracellular staining clearly illustrates the differences observed (Fig. 7A, right). Furthermore, IL-13 secretion did not increase when patient iNKT cells and Teffs were cultured together as shown previously for HD cells (data not shown). All this correlated with the mRNA analysis for IL-13 in human pancreas, showing that IL-13 expression is substantially lower in T1D pancreas samples than in control samples (Supplementary Fig. 2).

To assess whether human IL-13 would restore the suppression function of T1D-derived iNKT cells, rhIL-13 was added to the cocultures. This enhanced the suppression of Teff proliferation by iNKT cells of patients with T1D in a dose-dependent manner (Fig. 7B), further confirming the involvement of IL-13 as a major mechanism of iNKT-mediated suppression.

If IL-13 has a direct effect on Teffs, these should express the IL-13 receptor. Signaling through type 1 IL13Rα1 is essential for type 2 IL13Rα2 because it induces its expression (26). mRNA expression for IL13R chain genes was analyzed by RT-PCR on Teffs, showing that both IL13Rα1 and IL13Rα2 are expressed but with a completely different expression pattern between patients and control subjects. Teffs from patients with T1D showed a higher expression of IL13Rα1 compared with control Teffs. On the contrary, they expressed a significantly lower level of IL13Rα2 than controls (Fig. 8A). The same differences were observed when the receptor expression was analyzed by flow cytometry (Fig. 8B and C). These differences were not detected in iNKT cells from HDs or patients with T1D (data not shown). Altogether, the data show that defects related to the IL-13 pathway are responsible for the impaired Teff regulation by T1D-derived iNKT cells.

We investigated the function of iNKT cells in the context of human T1D onset to identify the mechanisms underlying their ability to suppress autoreactive CD4⁺ T cells. iNKT cells activated by glycolipids produce a wide variety of cytokines and interact with other immune cells so that they can regulate the response in disease conditions. iNKT-cell–mediated protection has been demonstrated in NOD mouse T1D, although the mechanisms are undefined. For the first time in our knowledge, the present study 1) demonstrates the functional suppression activity of human iNKT cells and provides novel information on the direct regulatory effect of human αGC-activated iNKT cells on CD4⁺ effector cells; 2) identifies the IL-13 pathway as an essential component of the mechanism underlying regulation by iNKT cells; and 3) more importantly, demonstrates the impairment of iNKT-cell–mediated regulation in patients with T1D at disease onset and proposes that this defect is related to deficiencies in the IL-13 pathway.

Ex vivo isolated CD4⁺ iNKT cells from the peripheral blood of HDs suppressed the proliferation of autologous CD4⁺ Teffs cells irrespective of their antigenic specificity. Furthermore, iNKT cells expanded in vitro with αGC maintained their dose-dependent suppressor capacity. This expansion was required because iNKT cell frequency in human peripheral blood is low (7). Of note, suppression by iNKT cells worked better on resting naive and memory T cells than on in vitro expanded Teffs (data not shown), which agrees with published data on mouse iNKT cells that propose that these cells preferentially act at the level of activation and/or differentiation of naive T cells (12).

The present data show that Teff suppression by human iNKT cells did not require cell-cell contact, in contrast to contact-dependent Treg-mediated regulation (32,33). We also demonstrate that IL-13 secreted by activated iNKT cells is the main effector of suppression, reducing both Teff proliferation and IL-2 secretion.

The most relevant finding is that the suppressor function was impaired in iNKT cells of patients with T1D at disease onset. Little information is available on the potential role of human iNKT cells in T1D pathogenesis. Studies have pointed to a decreased frequency of CD4⁺ iNKT cells playing a role in the development of T1D, but no real consensus has been reached on the existence of frequency alterations in these patients (34). For instance, an analysis using CD1d tetramers as a readout concluded that no alteration of the number of iNKT cells or in the production of IL-4 by PBMCs from patients with T1D compared with control subjects exists (35). In contrast, an earlier study showed both diminished iNKT cell frequency and IL-4 production in patients with T1D (36). Studies in the NOD mouse have shown that functional defects and a reduced frequency of iNKT cells contribute to T1D development (37,38). Besides, restoring iNKT cell numbers reduces T1D incidence or slows down its progression (39,40).

The present study demonstrates that human iNKT cells are an important source of IL-13 but that iNKT cells from patients with T1D have a remarkably reduced IL-13 production that could account for their suppression disability. Altered IL-13 secretion has been reported in children at T1D onset as well as in high-risk relatives of affected children (22,41). The origin of such deficiency has not been found, but the present findings suggest that defective iNKT cells may be a main contributor to the diminished serum levels of IL-13 at T1D onset.

iNKT cells are found in the pancreas of patients with T1D (L. Usero, unpublished observations); therefore, they can be influenced by the cytokine milieu in the tissue. The enterovirus Coxsackievirus B4 was considered of etiological significance in T1D because it leads to functional impairments and β-cell damage (42). Evidence suggests that human islet cells can sustain an enteroviral infection in patients with T1D. IFN-α, a cytokine secreted in response to such viral infections, is known to increase IL-4 synthesis while strongly downregulating IL-13 production (43). Together with the finding that infection of human islets with Coxsackievirus B4 causes downregulation of IL-13 (44), this suggests that enterovirus alteration of the islets inflammatory milieu in T1D may favor β-cell loss by impairing suppression of Teffs by iNKT cells. Furthermore, Coxsackievirus B4 infection may abrogate IL-13–mediated β-cell protection. IL-13 is a cytoprotective agent of pancreatic β-cells because it maintains β-cell viability after treatment with a range of cytotoxic insults (45). Thus, we can speculate that low levels of IL-13 at onset may contribute to the final deregulation of the autoimmune attack leading to the destruction of the β-cells. In this context, it was shown that early administration of rhIL-13 to NOD mice prevents insulitis and diabetes development, whereas later prophylaxis confers long-lasting protection against T1D and blocks progression of insulitis (23). IL-13 is thus likely to be of primary importance in diabetes pathogenesis, and any reduction in IL-13 level may exacerbate the disease effects on islet cell viability and function. Future studies are warranted to define the IL-13 signaling pathway as a potential therapeutic target of early intervention in patients with T1D.

The differences in IL-13 secretion by iNKT cells in patients and control subjects could depend on factors such as the APCs activating iNKT cells in vivo, which may prevent the expansion of regulatory iNKT cell subsets in T1D. The quality of the TCR signal alone or the integration of signals from various receptors could be part of the explanation (21). In this context, we observed that suppression by iNKT cells is abrogated when expanded in vitro with phytohemagglutinin instead of αGC. This correlated with lower secretion of IL-13 (L. Usero, unpublished observations). Thus, alterations in the TCR signaling events in iNKT cells from patients with T1D may be involved in the absence of IL-13–mediated regulatory function.

Activation of IL13Rα1 by IL-13 phosphorylates STAT6, which induces IL13Rα2 transcription (46,47). The present data and that of others support a role for IL-13 signaling in Teff alterations in T1D (48,49). Indeed, Teffs of patients with T1D expressed high levels of IL13Rα1 and low levels of IL13Rα2 compared with control subjects, and this can modulate IL-13 signaling in vivo (25,26). Albeit initially considered a decoy receptor, IL13Rα2-mediated signaling has been involved in the activation of the transforming growth factor-β1 (TGF-β1) promoter, underscoring its role as a signaling receptor (26). Furthermore, IL13Rα2 receptor can act as a negative regulator of IL-13 effects. IL13Rα2, expressed after exposure to IL-13, exits rapidly to the cell membrane from intracellular locations to capture and internalize extracellular IL-13 (25). Because IL13Rα2 is a high-affinity receptor, this process reduces the IL-13 available to bind IL13Rα1 in a negative feedback loop. Thus, a control pattern with high IL13Rα2 expression would promote regulation, whereas predominant IL13Rα1 signaling in patients with T1D would prevent the regulatory action of IL13Rα2. Of note, we have preliminary data suggesting that Teffs from control subjects and patients increase the expression of IL13Rα2 when cultured in the presence of rhIL-13.

Mechanisms proposed for iNKT-mediated regulation of Teffs in the NOD mouse model have involved the action of intermediate cells but not a direct effect of iNKT-derived IL-13 on Teffs (50,51). These differences could be explained by the differentiation state of NOD Teffs and iNKT cells compared with that of humans. However, NOD mice show high basal production of IL-13, so the mechanism proposed for human iNKT cell–mediated regulation could go undetected in this model (52).

Overall, this study describes an IL-13–mediated suppression mechanism in human iNKT cells. In the context of T1D onset, a decreased secretion of IL-13 by iNKT cells from patients with T1D leads to an impaired regulation of Teffs that favors the development of the autoimmune disease. Although the use of αGC for treatment or prevention of human T1D has not been conclusively tested, the use of αGC analogs favoring IL-13 secretion may be a useful approach to boost the suppressive function of iNKT cells to reach T1D prevention.

Acknowledgments. The authors thank the study volunteers for participating and the staff at Hospital de Mataró for assistance in subject recruitment. They also thank Marta Catalfamo (National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD) for help with manuscript preparation.

Funding. This work was supported by Spanish Ministry of Education project RYC-2006-002996; JDRF project JDRF25-2010-687 to M.M., D.J., and C.R.-M.; Spanish Ministry of Science (MINECO) project SAF2012-35344 to D.J.; and Spanish Ministry of Science (MICINN) project SAF2008-01629 to C.R.-M.

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. L.U. contributed to the data research and analysis and writing of the manuscript. A.S. contributed to the data analysis. E.P. recruited and monitored patients. C.X. contributed to the experimental design of the suppression assays. M.M. contributed to the review and editing of the manuscript. D.J. contributed to discussion and review and editing of the manuscript. C.R.-M. contributed to the data analysis and writing of the manuscript. C.R.-M. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.