Research Highlights

Human Pluripotent Stem-cell-derived Islets Ameliorate Diabetes in Non-human Primates Du Y, Liang Z, Wang S, et al. Nat Med. 2022;28(2):272–282. Type 1 diabetes (T1DM) is an autoimmune disease in which the immune system destroys insulin-producing β-cells within the pancreatic islets. Islet transplantation is a treatment strategy that has been used to restore endogenous insulin production in humans with T1DM.1 However, this strategy is limited by a lack of sufficient quantities of islets and the persistent immune attack of the transplanted islets.2 Inducible pluripotent stem cells (iPSCs) have revolutionized the field of regenerative medicine and expanded the potential for many cell therapies. These iPSCs represent a powerful technology in which somatic cells are induced by either gene transfer or chemical methods to become iPSCs that can differentiate into a variety of cell types by modifying culture conditions.3 Specific methods for producing large quantities of functional human pancreatic β-cells from iPSCs have been established,4 but the lack of appropriate preclinical models in which to test these iPSC-derived islets has limited their translation for the treatment of T1DM. In a recent study published in Nature Medicine, Du et al5 describe a novel method for generating large quantities of human iPSC-derived islets and test their function by transplanting them in a nonhuman primate (NHP) model of diabetes. iPSCs were generated from human adipose-derived fibroblasts and used to produce islets. During islet differentiation, the authors discovered that the formation of dense 3-dimensional cell aggregates at early stages and the addition of NeuroD1 inducer (ISX9) and wingless-related integration site inhibitor (Wnt-C59) were critical elements to the optimal generation of functional islets from iPSCs. Transcriptional and morphologic characterization of these iPSC-derived islets demonstrated the presence of all 3 types of pancreatic endocrine cells (insulin-producing β-cells, glucagon positive α-cells, and somatostatin positive δ-cells), similar to native human pancreatic islets. Additionally, the iPSC-derived islets produced insulin in response to glucose stimulation after transplantation into immunodeficient diabetic mice. These findings laid the groundwork for the authors to test the iPSC-derived islets in diabetic NHPs. Streptozotocin was used to induce diabetes in 5 NHPs that have been transplanted with the human iPSC-derived islets injected intraportally. Immunosuppression consisted of an induction therapy with rituximab, thymoglobulin, and basiliximab and maintenance immunosuppression with tacrolimus, sirolimus, and belatacept. There was a reduction in exogenous insulin requirements in all the NHPs after islet transplantation; however, 2 NHPs developed complications related to immunosuppression (posttransplant lymphoproliferative disease and gastrointestinal bleeding). The remaining NHPs demonstrated islet function up to 6 mo before immune rejection destroyed the islets. These findings represent a significant leap forward, as iPSC-derived islets represent a theoretically unlimited supply of islets for transplantation. However, there are several challenges remaining. The NHP model described relies on chemical induction of diabetes and lacks the autoimmune component. Furthermore, the model relies on immunosuppression that is necessary to overcome a xenogeneic immune response rather than the combined alloimmune and autoimmune response in T1DM islet recipients. A detailed characterization of the immune response to the iPSC-derived islets and the effects of that immune response on the function and maturation of the iPSC-derived islets is lacking. Additionally, an in depth understanding of the necessary immunosuppression and its effects on the function and maturation of the human iPSC-derived islets will be necessary to translate this into a successful therapy for T1DM. Despite these limitations this work opens a plethora of new avenues for islet transplant investigators. Soon, we may see these human iPSC-derived islets used in combination with gene editing technologies such as CRISPR to generate functional islets that are less immunogenic, potentially obviating the need for toxic immunosuppressants in islet transplant recipients. Interferon-β Acts Directly on T Cells to Prolong Allograft Survival by Enhancing Regulatory T Cell Induction Through Foxp3 Acetylation Fueyo-Gonzalez F, McGinty M, Ningoo M, et al. Immunity. 2022;55:459–474.e7. Regulatory T cells (Tregs) have the potential to effectively inhibit alloimmune responses. The notion of using Tregs as a form of cell therapy for controlling transplant rejection and possibly inducing transplant tolerance has gained traction and is being tested in various clinical trials.1 In 2003, a landmark publication by Chen et al2 demonstrated the critical role of transforming growth factor β in inducing the production of the key transcription factor forkhead box P3 (Foxp3), a master regulator of Treg function and stability. Since then, it has been known that a number of posttranslational modifications of Foxp3, including dimerization, cleavage, ubiquitination, and acetylation, play an additional critical role. How these processes are regulated is less clear. Fueyo-González et al3 examined the role of a pleiotropic cytokine interferon β (IFN-β) in the acetylation of Foxp3. Using a murine allogeneic heart transplant model, the authors demonstrated that exogenous IFN-β synergizes with cytotoxic T-lymphocyte associated protein, significantly prolonging allograft survival. Next, they showed that this graft-protective effect was mediated by the expression of IFN-β receptor interferon alpha/beta receptor (IFNAR) on CD4 T cells. To show this, the authors used mice with CD4 T cell-specific deletion of IFNAR as heart allograft recipients. In an in vitro Treg culture assay with transforming growth factor β, the addition of IFN-β led to a dose-dependent augmentation of the frequency of Foxp3+ cells. Additionally, the expression level of Foxp3 on a per cell basis was more prominent during the initial stage of Treg induction and diminished as IFN-β was introduced into the culture at later time points. Using neuropilin 1 as a marker for natural Tregs, they further demonstrated that the effect of IFNβ on Tregs was selective toward induced Tregs (iTregs) but not natural Tregs. Furthermore, iTregs functionally exhibited a potent suppression of naive CD4 T cell proliferation. The authors further demonstrated this chain of events in vivo using adoptive transfer of naive T cells in Rag1−/− mice treated with IFN-β. Next, the investigators took an unconventional computational approach to simulate IFN-β-IFNAR signaling in Tregs and hypothesized that the increase of Foxp3+ cells was primarily driven by STAT1 phosphorylation, leading to a direct increase in Foxp3 acetylation and hence its stability. They then returned to cellular models to examine the effect of IFN-β on various acetyltransferases. Gene expression analysis of IFN-β–treated cultures allowed them to identify the acetyltransferase P300 as the key enzyme that was responsible for Foxp3 acetylation induced by IFN-β, an effect that was abrogated by the P300 inhibitor C646. Lastly, they demonstrated in human donor peripheral blood mononuclear cells that IFN-β-mediated Foxp3 induction also occurred in human iTregs generated from human naive CD4 T cells. By providing a detailed mechanistic understanding of the effect of IFNβ on Foxp3 acetylation in CD4 T cells, the current study adds to our armamentarium for generating efficient Treg-based cell therapies for transplantation. This work also provides a rational option of immunosuppression in setting of certain transplant-relevant opportunistic viral infections for which type 1 interferons are the only effective treatment options. Most intriguingly, the authors provide an example of how in silico experiments can be utilized to delineate mechanisms that can then be confirmed in conventional cellular and molecular experiments. In summary, the current study identifies a mechanistic link between the immunosuppressive effects of IFN-β and Tregs and demonstrates the utility of computational tools in streamlining experiments to test the most likely hypothesis.