Selection for CD26- and CD49A+ Cells From Pluripotent Stem Cells-Derived Islet-Like Clusters Improves Therapeutic Activity in Diabetic Mice.

Background: Cell therapy of diabetes aims at restoring the physiological control of blood glucose by transplantation of functional pancreatic islet cells. A potentially unlimited source of cells for such transplantations would be islet cells derived from an in vitro differentiation of human pluripotent stem cells (hESC/hiPSC). The islet-like clusters (ILC) produced by the known differentiation protocols contain various cell populations. Among these, the ß-cells that express both insulin and the transcription factor Nkx6.1 seem to be the most efficient to restore normoglycemia in diabetes animal models. Our aim was to find markers allowing selection of these efficient cells. Methods: Functional Cell-Capture Screening (FCCS) was used to identify markers that preferentially capture the cells expressing both insulin and Nkx6.1, from hESC-derived ILC cells. In order to test whether selection for such markers could improve cell therapy in diabetic mouse models, we used ILC produced from a clinical-grade line of hESC by a refined differentiation protocol adapted to up-scalable bioreactors. Re-aggregated MACS sorted cells were encapsulated in microspheres made of alginate modified to reduce foreign body reaction. Implantation was done intraperitoneally in STZ-treated C57BL/6 immuno-competent mice. Results: CD49A (integrin alpha1) was identified by FCCS as a marker for cells that express insulin (or C-peptide) as well as Nkx6.1 in ILC derived by hESC differentiation. The ILC fraction enriched in CD49A + cells rapidly reduced glycemia when implanted in diabetic mice, whereas mice receiving the CD49A depleted population remained highly diabetic. CD49A-enriched ILC cells also produced higher levels of human C-peptide in the blood of transplanted mice. However, the difference between CD49A-enriched and total ILC cells remained small. Another marker, CD26 (DPP4), was identified by FCCS as binding insulin-expressing cells which are Nkx6.1 negative. Depletion of CD26 + cells followed by enrichment for CD49A + cells increased insulin+/Nkx6.1+ cells fraction to ~70%. The CD26 - /CD49A + enriched ILC exhibited improved function over non-sorted ILC or CD49A + cells in diabetic mice and maintain prolonged blood C-peptide levels. Conclusions: Refining the composition of ILC differentiated from hPSC by negative selection to remove cells expressing CD26 and positive selection for CD49A expressing cells could enable more effective cell therapy of diabetes.

Nedd4 family interacting protein 1 (Ndfip1) promotes death of pancreatic beta cells.

High-throughput siRNA screening was employed to identify novel genes that regulate cytokine-induced death of pancreatic ß-cells. One of the 'hits' was Nedd4 family interacting protein 1 (Ndfip1), an adaptor and activator of Nedd4-family ubiquitin ligases. Silencing of Ndfip1 inhibited cytokine-induced apoptosis of mouse and human pancreatic islets and promoted glucose-stimulated insulin secretion. These effects were associated with an increase in the cellular content of JunB, a potent inhibitor of ER stress and apoptosis. Silencing of Ndfip1 also increased the expression of ATF4, IRE-1α, and the spliced form of XBP that govern the unfolded protein response (UPR) and relieve cytokine-induced ER stress, while overexpression of Ndfip1 exerted opposite effects. These findings implicate Ndfip1 in the degradation of JunB; inhibition of the UPR and insulin secretion; and promotion of cytokine-induced death of pancreatic ß-cells.

A novel domain mediates insulin-induced proteasomal degradation of insulin receptor substrate 1 (IRS-1).

Insulin receptor substrate-1 (IRS-1) plays a pivotal role in insulin signaling, therefore its degradation is exquisitely regulated. Here, we show that insulin-stimulated degradation of IRS-1 requires the presence of a highly conserved Ser/Thr-rich domain that we named domain involved in degradation of IRS-1 (DIDI). DIDI (amino acids 386-430 of IRS-1) was identified by comparing the intracellular degradation rate of several truncated forms of IRS-1 transfected into CHO cells. The isolated DIDI domain underwent insulin-stimulated Ser/Thr phosphorylation, suggesting that it serves as a target for IRS-1 kinases. The effects of deletion of DIDI were studied in Fao rat hepatoma and in CHO cells expressing Myc-IRS-1(WT) or Myc-IRS-1(Δ386-430). Deletion of DIDI maintained the ability of IRS-1(Δ386-434) to undergo ubiquitination while rendering it insensitive to insulin-induced proteasomal degradation, which affected IRS-1(WT) (80% at 8 h). Consequently, IRS-1(Δ386-434) mediated insulin signaling (activation of Akt and glycogen synthesis) better than IRS-1(WT). IRS-1(Δ386-434) exhibited a significant greater preference for nuclear localization, compared with IRS-1(WT). Higher nuclear localization was also observed when cells expressing IRS-1(WT) were incubated with the proteasome inhibitor MG-132. The sequence of DIDI is conserved more than 93% across species, from fish to mammals, as opposed to approximately 40% homology of the entire IRS-1. These findings implicate DIDI as a novel, highly conserved domain of IRS-1, which mediates its cellular localization, rate of degradation, and biological activity, with a direct impact on insulin signal transduction.