Yoshiaki Okubo

Acceleration of diabetes development in CXC chemokine receptor 3 (CXCR3)-deficient NOD mice

Aims/hypothesisThe aim of this study was to understand the role of CXC chemokine receptor 3 (CXCR3), a T-helper 1(Th1) type chemokine receptor, in the pathogenesis of type 1 diabetes.MethodsWe observed the incidence of diabetes in Cxcr3 homozygous knockout mice. We compared the expression pattern of various cytokines and chemokines and the frequency of FOXP3+ cells in the pancreas and pancreatic lymph nodes from Cxcr3−/− NOD mice and wild-type NOD mice. In addition, we observed the migration ability of CXCR3+CD4+ cells to pancreatic islets upon adoptive transfer. Finally, we examined whether Cxcr3+ regulatory T cells (Tregs) actually suppressed the onset of diabetes in vivo.ResultsCxcr3−/− NOD mice developed spontaneous diabetes earlier than did wild-type NOD mice. In Cxcr3−/− NOD mice, Tregs were more frequent in pancreatic lymph nodes and less frequent in pancreatic islets than in wild-type NOD mice. While transferred CXCR3−CD4+ cells from wild-type NOD mice did not infiltrate pancreatic islets of NOD–severe combined immunodeficiency (SCID) mice, CXCR3+CD4+ cells from the same mice migrated into the recipient islets and contained Forkhead box P3 (FOXP3) upon adoptive transfer. Moreover, CD4+CD25+ cells from wild-type NOD mice suppressed and delayed the onset of diabetes compared with those from Cxcr3−/− NOD mice in a cyclophosphamide-induced diabetes model system.Conclusions/interpretationThe mechanism of accelerated diabetes onset in Cxcr3−/− NOD mice was considered to be due to the lack of hybrid Tregs (CXCR3+FOXP3+CD4+ cells), which could effectively migrate into and regulate Th1 inflammation in local lesions under Cxcr3 knockout conditions.

CXCL10 DNA Vaccination Prevents Spontaneous Diabetes through Enhanced β Cell Proliferation in NOD Mice

CXCL10, a chemokine for Th1 cells, is involved in the pathogenesis of various Th1-dominant autoimmune diseases. Type 1 diabetes is considered to be a Th1-dominant autoimmune disease, and a suppressive effect of CXCL10 neutralization on diabetes development has been reported in a cyclophosphamide-induced accelerated diabetes model through induction of β cell proliferation. However, intervention in a diabetes model might bring about opposite effects, depending on the timing, amount, or method of treatment. In the present study, we examined the effect of CXCL10 neutralization in a “spontaneous diabetes” model of NOD mice, using CXCL10 DNA vaccination (pCAGGS-CXCL10). pCAGGS-CXCL10 treatment in young NOD mice induced the production of anti-CXCL10 Ab in vivo and suppressed the incidence of spontaneous diabetes, although this treatment did not inhibit insulitis or alter the immunological response. pCAGGS-CXCL10 treatment enhanced the proliferation of pancreatic β cells, resulting in an increase of β cell mass in this spontaneous diabetes model as well. Therefore, CXCL10 neutralization is suggested to be useful for maintaining β cell mass at any stage of autoimmune diabetes.

Induction of anti-whole GAD65 reactivity in vivo results in disease suppression in type 1 diabetes.

Most type 1 diabetes mellitus is caused by autoimmune pancreatic beta-cell destruction. Several antigens such as insulin, glutamic acid decarboxylase (GAD) and islet-specific glucose-6-phosphatase catalytic subunit related protein (IGRP) are considered to take part in the autoimmune destructive process. Because the role of GAD in the disease process of type 1 diabetes is still controversial, we investigated the disease phenotype upon in vivo induction of whole GAD65 reactivity using a GAD65 homo knockout NOD splenocytes to NOD-scid transfer system. Splenocytes from 8 to 10-week-old female GAD65 homo knockout (=KOT splenocytes) or age-matched wild type (=WTT splenocytes) NOD mice were transferred into female NOD-scid recipients. As compared to recipients of WTT splenocytes, the onset of diabetes in recipients of KOT splenocytes was significantly delayed (p<0.001). Moreover, TGF-beta expression was enhanced in the pancreas from recipients of KOT splenocytes. Splenocytes from recipients of KOT splenocytes produced IL-10 (/IFN-gamma) upon GAD65 stimulation, whereas those from recipients of WTT splenocytes did not. Based upon these results, we propose that anti-whole GAD65-reactive T cells have the ability to regulate the development of type 1 diabetes.

FOXP3/Scurfin gene polymorphism is associated with adult onset type 1 diabetes in Japanese, especially in women and slowly progressive-type patients

Purpose: Some reports have shown that the cause of type 1 diabetes is associated with dysfunction of regulatory T cells (Tregs). The FOXP3/Scurfin gene is known to be a master gene of Tregs. Therefore, we tried to analyze the relation between the gene polymorphism and adult onset type 1 diabetes and its subtype in the Japanese population. Methods: In this study, we recruited 316 Japanese patients with type 1 diabetes (155 male and 161 female, mean onset age 35.4 years) and 432 healthy Japanese controls (263 male and 169 female, mean age 44.4 years). Then we subdivided the patients by onset type, sex, and islet-associated autoantibody positivity. Results: The genotype frequency of (GT)16/(GT)16 in female patients with overall type 1 diabetes was especially lower than that in controls (19.9% vs. 38.5%, p = 0.0002). Moreover, the genotype frequency of (GT)16/(GT)16 in female patients with slowly progressive type 1 diabetes was significantly lower than that in controls (15.4% vs. 38.5%, p = 0.002). Conclusion: Our data showed that the (GT)n microsatelloite polymorphism in the FOXP3/Scrufin gene was associated with Japanese adult onset type 1 diabetes, especially in females, and slowly progressive type 1 diabetes.

Pioglitazone may accelerate disease course of slowly progressive type 1 diabetes.

It has been reported that intervention with insulin in slowly progressive type 1 diabetic (SPIDDM) patients delays the progression to an insulin‐dependent state compared to that with sulfonylureas. However, the rate of progression to SPIDDM with the use of insulin‐sensitizing agents is unknown. The aim of this study was to determine the effect of insulin‐sensitizing agents on SPIDDM patients.

A Mimic of Viral Double-Stranded RNA Triggers Fulminant Type 1 Diabetes-like Syndrome in Regulatory T Cell-Deficient Autoimmune Diabetic Mouse

Human fulminant type 1 diabetes (FT1D) is an extremely aggressive disease. The delay of proper diagnosis results in high mortality. However, the pathophysiology of this disease remains unclear. We took advantage of CD28-deficient NOD (CD28−/− NOD) mice, which have limited numbers of regulatory T cells and develop aggressive autoimmune diabetes, to create a FT1D model that mimicked the disease in humans. Young CD28−/− NOD mice were injected with polyinosinic-polycytidylic acid to activate innate immunity in an effort to induce diabetes onset. In this model, innate immune cell activation precedes the onset of diabetes similar to ∼70% of FT1D patients. Eighty-three percent of CD28−/− NOD mice developed diabetes within 1–6 d after injection of polyinosinic-polycytidylic acid. Moreover, T cells infiltrated the pancreatic exocrine tissue and destroyed α cells, an observation characteristic of human FT1D. We conclude that an FT1D-like phenotype can be induced in the background of autoimmune diabetes by a mimic of viral dsRNA, and this model is useful for understanding human FT1D.

GAD‐Reactive T Cells Were Mainly Detected in Autoimmune‐Related Type 1 Diabetic Patients with HLA DR9

Abstract: Type 1 diabetes mellitus (T1DM) is considered to be a T cell‐mediated disease, and many reports suggest that some HLA types, especially HLA DR4 and DR9, convey susceptibility to T1DM in Japanese. We investigated the association between T cell reactivity against GAD and HLA types in “islet‐associated autoantibody‐positive” T1DM in Japanese. Blood samples were obtained from 36 “autoantibody‐positive” type 1 diabetic patients with HLA DR4 or DR9 and 23 type 2 diabetic patients with HLA DR4 or DR9 as controls. They were divided into three groups, DR4/9, DR4/X, and DR9/X groups. In each HLA type group, GAD‐reactive IFN‐γ‐producing CD4+ cells were assessed by means of intracellular cytokine staining for flow cytometry. Type 1 diabetic patients with HLA DR9/X had significantly higher numbers of GAD‐reactive IFN‐γ‐producing CD4+ cells as compared to type 1 diabetic patients with DR4/X or DR4/9 (P < 0.05) and all type 2 diabetic patients. There was no significant difference in the number of GAD‐reactive IFN‐γ‐producing CD4+ cells between type 1 diabetic and type 2 diabetic patients belonging to the DR4/X and DR4/9 groups. There was an association between T cell reactivity against GAD and HLA DR9 in Japanese type 1 diabetes.

Thiazolidinediones May Not Reduce Diabetes Incidence in Type 1 Diabetes

Abstract:  Thiazolidinediones improve glycemic control by reducing insulin resistance. Some studies have demonstrated that troglitazone had a preventative effect on diabetes in NOD (non‐obese diabetic) mice. One of the mechanisms proposed for the prevention of diabetes by thiazolidinediones is an effect on T‐helper 1/T‐helper 2 (Th1/Th2) balance. In this article, we attempted to clarify whether pioglitazone is also effective in preventing diabetes as compared to metformin, which has no immunological effect. Female NOD mice were administered pioglitazone or metformin orally, and the insulitis score, cytokines secreted from splenocytes, cytokine expression levels in the pancreas, and the incidence of diabetes after acceleration by cyclophosphamide were analyzed. We could not find any advantage of pioglitazone in preventing Th1 skewing and the development of diabetes over metformin. Therefore, further research should take place before the application of thiazolidinediones to human slowly progressive insulin‐dependent diabetes mellitus (SPIDDM) patients.

Frequency of CTLA-4 Gene CT60 Polymorphism May Not Be Affected by Vitamin D Receptor Gene Bsm I Polymorphism or HLA DR9 in Autoimmune-Related Type 1 Diabetes in the Japanese

Abstract:  One of the CTLA‐4 SNPs, +6230G>A (CT60), has recently been reported to be related to susceptibility to type 1 diabetes and autoimmune thyroid disease. We have previously reported an association between acute‐onset type 1 diabetes in Japanese and the Vitamin D receptor (VDR) gene Bsm I large B polymorphism, which is related to the Th1‐type response. Moreover, we found a significant correlation between autoimmune‐related type 1 diabetes with HLA DR9 and detection of GAD‐reactive Th1 (T helper 1)‐type cells. In the present article, we tried to clarify whether the frequency of one of the CTLA‐4 SNPs, +6230G>A (CT60), is affected by the VDR gene Bsm I polymorphism or by HLA DR9 in Japanese type 1 diabetics. The frequency of the CT60 GG genotype did not appear to be affected by either the VDR gene Bsm I large B polymorphism or HLA DR9.