Regulatory dendritic cells in celiac disease: biology and potential cell-based approaches

Celiac Disease (CD) is a T cell mediated inflammatory disorder caused by an altered immune response to gluten in genetically predisposed individuals. At present, a lifelong gluten-free diet (GFD) is the only available treatment. Compelling evidence indicate the central role of IL-10 and IL-10-secreting regulatory cells, including T cells and tolerogenic dendritic cells (tolDC), in the regulation of inflammatory responses to dietary antigens (Ags). The goal of the present project was i) to determine the presence and functions of IL-10 and IL-10-related regulatory cells in CD patients at different stage of the disease, including untreated patients with atrophic or normal mucosa and patients in remission because of gluten-free diet (GFD); ii) to evaluate whether a tolDC-based approach has the potential to control unwanted immune response to gluten in CD patients. To this aim, we enrolled 70 children with CD who were classified after histological diagnosis in CD patients at onset with intestinal villous atrophy (acute-CD, n=38), patients with positive serology and normal intestinal mucosa (potential-CD, n=21), CD patients in remission on GFD (GFD-CD, n=11), and age-matched healthy subjects (n=13). To evaluate whether regulatory cells are orchestrating the divergent evolution of disease we analyzed peripheral blood (PB) and single cell suspensions from duodenal biopsies to assess the frequency of lymphoid and myeloid cell subsets, levels of plasmatic cytokines, and peripheral T-cell response to gliadin. Flow cytometric analysis showed that the composition of the lymphoid/myeloid compartments in PB was similar in CD patients and healthy subjects, although T cells in CD patients were more activated than in healthy controls and the frequency of inflammatory DC in untreated-CD (acute-CD at diagnosis and potential-CD) was higher than in GFD-CD patients. In line with these results, measurement of cytokines in plasma confirmed a pro-inflammatory profile in untreated patients (e.g., IL-1b, TNFa and IL-12p70) and an overall lower IFN-g/IL-10 ratio in potential-CD subjects. Regardless of dietary treatment or mucosal damage degree, peripheral T cells secreted IFN-g in response to gliadin in several patients, while IL-10 secretion was prominent only in potential-CD. The overall inflammatory gut-infiltrate was lower in potential-CD and GFDCD biopsies than in patients with atrophic mucosa (acute-CD). T cells were the most abundant cell subset at all stages of disease, although they were less activated in potential-CD than in acute-CD or GFD-CD. B cells and TCRgd T cells were also abundant in the infiltrates, with reduction of CD19+ cells and maintenance of high frequency of TCRgd T cells in patients on GFD. Importantly, we observed increased frequency of a subset of tolerogenic DC (DC-10, characterized by IL-10 production and ability to promote the induction of Ag-specific anergic and regulatory T cells) and higher percentage of IL-10-producing type 1 regulatory T (Tr1) cells in patients with normal mucosa (potential-CD and GFD-CD, respectively), as compared to patients with active disease (acute-CD). Overall, these results show a reduction of effector lymphocytes in the gut mucosa of patients undergoing GFD and a prevalence of IL-10-producing regulatory cells (either DC-10 or Tr1) in the intestinal mucosa of non-atrophic CD patients (potential-CD and GFD-CD). The second part of our study was dedicated to the development of a tolDC based approach for the treatment of CD. To this aim, we designed lentiviral vectors (LV) which enforce presentation of the deamidated 17-mer peptide, including DQ2.5-glia-α1a/DQ2.5-glia-α2 epitopes of α-gliadin, in the context of HLA-classII-DQ2 molecules, thanks to the fusion to sequences encoding for the human invariant chain (Iip33), in the absence or presence of human IL-10. We engineered monocytes isolated from the peripheral blood of CD patients with LVs encoding for the gliadin peptide alone (DCglia) or in combination with IL-10 (DCIL-10/glia) and differentiated transduced monocytes in vitro into DC. When the gliadin peptide and IL-10 were co-expressed, the resulting cell population acquired a tolDC phenotype, (DC-10-like: CD16+CD14+ILT4hiCD141+CD163+), and constitutively produced high amounts of IL-10. DCglia were able to induce proliferation and IFN-g release by Ag-specific CD4+ T cells from untreated CD patients, while DC coencoding for IL-10 induced a hypo-proliferative response and promoted the expansion of gliadin-specific CD49b+LAG-3+ Tr1 cells. These results demonstrate that patient-derived DCIL-10/glia are tolDC with the potential to control pathogenic gliadin-specific T cell responses, via induction of Agspecific regulatory cells. Our study provides novel insights into the mechanisms underlying CD progression and response to GFD and confirms the central role of IL-10-producing cells in maintaining and promoting tolerance in the gut mucosa of CD patients. Furthermore, our results indicate that engineered tolDC modulate gliadin-specific T cell responses in vitro and pave the way for designing a DC-based cell therapy for CD.