Defective Invariant Natural Killer T-Cell Suppression in Patients With Type 1 Diabetes

Type 1 diabetes (T1D) results from gradual alteration of insulin-secreting pancreatic β-cells. When most β-cells are destroyed and/or nonfunctional, patients are unable to regulate hyperglycemia and clinical signs appear (1). Although T cells play a key role in β-cell destruction, other immune cells are also involved in T1D physiopathology (2). Both Foxp3⁺ regulatory T cells (Tregs) and invariant natural killer T (iNKT) cells exert a major role in maintaining peripheral tolerance. iNKT cells are unconventional T lymphocytes expressing the semi-invariant T-cell receptor (Vα14-Jα18 in mice, Vα24-Jα18 in humans). They recognize glycolipid ligands, presented by the highly conserved CD1d molecule. iNKT cells are considered as “innate-like” T cells because they express a memory phenotype and rapidly respond upon activation. NOD mice, which spontaneously develop T1D, present abnormalities in iNKT cell frequency and function. Studies based on this mouse model have demonstrated the protective role of iNKT cells in T1D. Increasing iNKT cell frequency by adoptive transfer or Vα14-Jα18 transgenesis decreases T1D incidence (3,4). Conversely, CD1d-deficient NOD mice lacking iNKT cells present an acceleration of T1D development (5). Ligand-specific stimulation of iNKT cells suggests that the protective and regulatory role of these cells could be used as a therapeutic strategy in patients (6,7). Although numerous studies using mouse models have highlighted the regulatory role of iNKT cells in T1D, the regulatory role of iNKT cells in T1D patients remains unclear.

In this issue, Usero et al. (8) show for the first time in vitro that iNKT cell suppression of T-cell response is defective in T1D patients. iNKT cells from healthy control subjects ex vivo and after expansion in vitro suppress interleukin (IL)-2 synthesis and the proliferation of effector T cells (CD4⁺CD25⁻) via a mechanism independent of cell contact (Fig. 1A). This result is reminiscent of the mouse iNKT cell suppressive effect observed in cocultures with autoreactive BDC2.5 T cells from NOD mice (9). However, this suppressive effect of mouse iNKT cells is dependent of cell contact and independent of IL-4, IL-10, IL-13, and transforming growth factor-β secretion (Fig. 1B). In contrast, Usero et al. (8) describe the key role of IL-13 in the suppressive effects of human iNKT cells, using a specific blocking antibody. Interestingly, iNKT cells from T1D patients at diagnosis are unable to exert the suppressive function observed with cells from healthy donors due to decreased IL-13 production. Moreover, the addition of IL-13 restores the suppressive function of iNKT cells from T1D patients. Usero et al. also report a decrease of IL-13 mRNA levels in the pancreata of T1D patients compared with control individuals and an altered IL-13 receptor expression on effector T cells from peripheral blood of T1D patients.

Altogether these data point out for the first time the main role of IL-13 in the suppressive and protective function of iNKT cells in T1D physiopathology. This finding is in agreement with studies in NOD mice describing an association between T1D protection and a Th2 shift of effector T-cell responses by iNKT cells, involving IL-4 and IL-10 (Fig. 1C) (6,7,10). Previous studies in T1D patients have reported a decrease of IL-4 production by iNKT cells from the peripheral blood and pancreatic lymph nodes (11,12). It would have been interesting if Usero et al. (8) had measured IL-4 production by iNKT cells from recent-onset T1D patients. Defective Th2 cytokine production could be due to iNKT cell repolarization and/or the depletion of CD4⁺ iNKT cell subset preferentially producing anti-inflammatory cytokines, as reported by Kis et al. (13). Although Usero et al. (8) describe a direct effect of iNKT cells on effector T cells, it remains to be addressed whether iNKT cells from T1D patients could modify dendritic cell (DC) function. It has been well established in mouse models that iNKT cells can induce tolerogenic DCs generating Treg cells in the pancreatic lymph nodes (14–16). This protective effect of iNKT cells results from both cell-contact–dependent and –independent mechanisms (Fig. 1D). In humans, an in vitro study has shown the induction of tolerogenic DCs by iNKT cells (17), and it still has to be investigated in T1D patients (Fig. 1E). Of note, Usero et al. (8) do not analyze iNKT frequency in T1D patients and therefore do not shed a new light on this controversial issue (18,19). Moreover, as T1D particularly affects children and is diagnosed more and more early in childhood, it would be interesting to perform a similar study in children with recent-onset diabetes.

The current study (8) and others have described a decrease of IL-13 and IL-4 production by iNKT cells in T1D patients. It would be important to determine whether these abnormalities are restricted to the production of these cytokines or rather reflect a more generally altered status. To address this issue, it might be interesting to analyze fresh cells and not only iNKT cells expanded in vitro for 2 weeks of culture, which might have selected the most robust iNKT cells. Even though the global role of iNKT cells appears to be beneficial against T1D, one study in NOD mice has revealed that iNKT17 cells are enriched in the pancreas and promote T1D development (20). Thus the complexity and ambivalence of iNKT cell functions prompt further studies in patients to determine whether iNKT cell defects are a cause and/or a consequence of T1D onset. Indeed hyperglycemia in T1D patients appears at the onset of the disease, whereas the autoimmune reaction begins earlier. So, is the altered phenotype of iNKT consecutive to hyperglycemia or is it an earlier mechanism promoting the development of the disease? Depending on the stages of the disease development, is iNKT cell cytokine profile altered and/or are specific iNKT cell subsets selected? In the physiopathology of T1D, environmental factors such as viral infection, alimentation, and intestinal microbiota have been involved. Interestingly Kostic et al. (21) have observed an increase of some sphingomyelin moieties, inhibitory ligands of iNKT cells, in the stools of at-risk children before developing diabetes. Further studies should determine how environmental factors modify the iNKT cell ligands and the cytokines locally present at the sites of inflammation. There is no therapy to cure or prevent T1D, and the main hope is to develop strategies to avoid the pancreatic β-cell alterations before the onset of the disease. Determining the origin of iNKT cell alteration in T1D could represent a new path to intervene before T1D onset.