Glucose inhibits angiogenesis of isolated human pancreatic islets.

Owing to strong interactions between pancreatic islets and the surrounding capillary network, we hypothesized that high glucose concentrations might affect key angiogenesis factors from isolated human islets, thus contributing to beta-cell failure in diabetes. Human islets from eight distinct donors were studied following 96 h in culture in the presence of normal (5.5 mmol/l) or high (16.7 mmol/l) glucose concentrations. Similar studies were performed with HUVECs. Human angiogenesis-related genes and corresponding proteins were studied by real-time quantitative PCR (RT-qPCR) and protein arrays respectively. Angiogenesis and proliferation assays were also performed with HUVECs under the same culture conditions. RT-qPCR and proteome analysis of human islets incubated with 16.7 mM/l glucose revealed a significant decrease in pro-angiogenic factors including vascular endothelial growth factor (VEGF) mRNA by 20% and VEGF protein levels by 42% as well as additional proteins such as fibroblast growth factor-4 by 41%, MMP9 by 18%, monocyte chemoattractant protein-1 by 21%, and prolactin by 25%. In contrast, we observed a 17% increase in thrombospondin-1 (TSP-1, listed as THBS1 in the HUGO database) and a 37% increase in angiotensinogen gene expression levels, but neither angiotensin-converting enzyme nor angiotensin II type 1 receptor gene expression was affected. The amounts of anti-angiogenic proteins such as TSP-1 and serpin B5/maspin were also increased by 70 and 98% respectively as well as endostatin by 63%. Angiogenesis assays of HUVECS in the presence of high glucose concentrations revealed a 30% decrease in tree-like tubular network formation. These data suggest that glucose reduces key factors of islet angiogenesis, which might exacerbate beta-cell failure.

Mesenchymal stem cells protect NOD mice from diabetes by inducing regulatory T cells.

AIMS/HYPOTHESIS: Displaying immunomodulatory capacities, mesenchymal stem cells (MSCs) are considered as beneficial agents for autoimmune diseases. The aim of this study was to examine the ability of MSCs to prevent autoimmune diabetes in the NOD mouse model. METHODS: Prevention of spontaneous insulitis or of diabetes was evaluated after a single i.v. injection of MSCs in 4-week-old female NOD mice, or following the co-injection of MSCs and diabetogenic T cells in irradiated male NOD recipients, respectively. The frequency of CD4(+)FOXP3(+) cells and Foxp3 mRNA levels in the spleen of male NOD recipients were also quantified. In vivo cell homing was assessed by monitoring 5,6-carboxyfluorescein diacetate succinimidyl ester (CFSE)-labelled T cells or MSCs. In vitro, cell proliferation and cytokine production were assessed by adding graded doses of irradiated MSCs to insulin B9-23 peptide-specific T cell lines in the presence of irradiated splenocytes pulsed with the peptide. RESULTS: MSCs reduced the capacity of diabetogenic T cells to infiltrate pancreatic islets and to transfer diabetes. This protective effect was not associated with the modification of diabetogenic T cell homing, but correlated with a preferential migration of MSCs to pancreatic lymph nodes. While injection of diabetogenic T cells resulted in a decrease in levels of FOXP3(+) regulatory T cells, this decrease was inhibited by MSC co-transfer. Moreover, MSCs were able to suppress both allogeneic and insulin-specific proliferative responses in vitro. This suppressive effect was associated with the induction of IL10-secreting FOXP3(+) T cells. CONCLUSIONS/INTERPRETATION: MSCs prevent autoimmune beta cell destruction and subsequent diabetes by inducing regulatory T cells. MSCs may thus offer a novel cell-based approach for the prevention of autoimmune diabetes and for islet cell transplantation.