Ethylenecarbodiimide-fixed splenocytes carrying whole islet antigens decrease the incidence of diabetes in NOD mice via down-regulation of effector memory T cells and autoantibodies

Abstract

Type 1 diabetes mellitus (T1DM) is a syndrome of loss of glucose homeostasis caused by the loss of β cell chronic autoimmunity against islet cells. Islet-specific epitopes coupled antigen presenting cells by Ethylenecarbodiimide (ECDI) is a promising strategy to induce antigen-specific tolerance. However, single epitope induced tolerance is insufficient to prevent the onset of T1DM. The aim of this study is to evaluate the efficacy of whole islet antigens in preventing the onset and progression of T1DM and identify the underlying immune mechanism in NOD mice. In this study, the whole islet antigens, derived from islet lysate isolated from BALB/c mice, were coupled to splenocytes of BALB/c mice by ECDI fixation (SP-Islet lysate), and then intravenously administrated to NOD mice. The results showed that, compared with control group, SP-Islet lysate group significantly decreased T1DM incidence and improved the survival of NOD mice. SP-Islet lysate treated mice had reduced insulitis score and autoantibody levels, and improved glucose tolerance and insulin/glucagon production. Furthermore, the effector memory T cells (T_EMs) were downregulated and regulatory T cells (Tregs) were upregulated by the SP-Islet lysate treatment, with reduced populations of Th1&Th17 cells. In conclusion, ECDI-fixed splenocytes carrying whole islet antigens effectively prevented the onset of T1DM in NOD mice, via suppressing the production of autoantibodies and inducing anergy of autoreactive T cells.

TYPE 1 DIABETES MELLITUS (T1DM) is a chronic autoimmune disease, mainly characterized with cell-mediated autoimmune destruction of the pancreatic β-cells in predisposed individuals [1]. The autoimmunity in T1DM is predisposed by the genetic susceptibility, which is known to lead to loss of immunological tolerance to certain β-cell antigens. Autoreactive T cells specifically attack the insulin producing β cells, and both NOD mice and T1DM patients exhibit mostly an effector/memory phenotype consistent with ongoing β cell autoimmunity due to antigen-experienced in vivo proliferation history which contributes to the difficulty in preventing disease progression of T1DM [2, 3]. Regulatory T cells (Tregs) have an essential role in the maintenance of immunological tolerance and participate in regulating the onset and progression of T1DM [4]. Autoantibodies such as anti-insulin autoantibodies (IAA), anti-glutamic acid decarboxylase autoantibodies (GADA), anti-islet cell antibodies (ICA), anti-islet antigen-2 autobodies (IA-2A), and anti-zinc transporter autoantibodies (ZnT8A) are likely detectable for months to years before the clinical onset of this disease, and the detection of these autoantibodies significantly increases the risk for T1DM [5-7].

The prevalence and incidence of T1DM have kept rising in many countries [8-10]. The treatment targeting T cells, B cells, cytokines or co-stimulators has shown some efficacy in alleviating T1DM [11], but the use of immunosuppressive agents failed to achieve long-term protective effects on the disease, and could cause a variety of side effects such as increasing risks of infection and tumorigenesis. New therapeutic strategies to induce β-cell antigens-specific tolerance are urgently needed for T1DM treatment.

Ethylenecarbodiimide (ECDI) is a chemical crosslinker [12] and can be used to couple splenocytes with antigens. The ECDI fixed splenocytes undergo rapid apotosis [13]. The antigens carried by the apoptotic splenocytes can be processed by host antigen-presenting cells (APCs) and presented to host T cells in tolerogenic fashion, resulting in indirect or cross-tolerance [13]. The ECDI fixed splenocyte coupled with antigens has been reported as a potent and safe method to induce antigen-specific immunological tolerance through targeting recipient APCs and the PD-1/PD-L1-dependent inhibition of effector T cell activity as well as the expansion of Tregs [14-17]. Female non-obese diabetic (NOD) mice will develop T1DM spontaneously, and are broadly used for T1DM studies. Recent studies using NOD mice have found that intravenous infusion of Ins B_9–23-pulsed APCs fixed by ECDI to young NOD mice can effectively induce tolerance and eventually inhibit the onset of T1DM, but Ins B_9–23-SP treatment failed to decrease T1DM incidence in old NOD mice, suggesting that insulin epitope(s) to initiate autoimmune responses are different from those dominating the disease progression [12]. Destructed β-cells expose multiple antigen epitopes concurrently around 10 weeks in NOD mice, and hence various autoimmune responses are activated targeting distinct epitopes [12], thus the single antigen tolerance is insufficient for immune suppression ten weeks later. Therefore, therapies containing multiple islet-specific epitopes, coupled to APCs by ECDI fixation, are needed to be developed and evaluated for the prevention and treatment of autoimmune T1DM.

In this context, we proposed that ECDI-fixed splenocytes carrying whole islet antigens might effectively inhibit the autoimmunity. In this study, splenocytes were fixed with islet lysate containing the whole islet antigens by ECDI and then intravenously infused to NOD mice, aiming to evaluate the efficacy of this treatment in inducing immunological tolerance and preventing the onset of T1DM.

Methods

Mice

Female NOD mice aged 8 weeks and male BALB/c mice aged 7 weeks were purchased from Beijing Huafukang Biosciences (Beijing, China), and housed in a specific pathogen-free facility at a cycle of 12:12 light/dark, constant temperature and humidity at Key Laboratory for Critical Care Medicine of the Ministry of Health. The female NOD strain in this study derived from the outbred Jcl:ICR line of mice through repetitive brother-sister mating [18]. This sub-strain of NOD mice spontaneously presents a diabetic incidence of around 60%. Animal bodyweight was measured at least every two days. NOD mice were considered diabetic when random blood glucose levels were ≥250 mg/dL for two consecutive days measured from tail bleeding using Roche glucometer. All animal studies were approved by the Institutional Animal Care and Ethics Committee of Tianjin First Center Hospital.

Tolerance induction

Spleen tissues from BALB/c mice were grinded with a frosted surface of a slide in RPIM 1640 (Solarbio, Beijing, China), filtrated through a 70 μm-pore size mesh Cellstrainer^TM (Falcon, BD Biosciences, Heidelberg, Germany) and washed in phosphate buffered saline (PBS) with centrifuge (400 g, 5 min) to remove tissue clumps. Tris-NH₄Cl (Solarbio, Beijing, China) was used to eliminate the red blood cells. The cells were washed in PBS with centrifuge (400 g, 5 min) and resuspended in PBS to create a single cell suspension. Pancreases of BALB/c mice were digested by collagenase P (0.5 mg/mL, Roche, Indiana, USA) at 37°C for 15 min, and purified by density gradient (Histopaque 1077, Sigma, Louis, USA). Islet lysate was prepared by ten freeze-thaw cycles of islets from BABL/c mice. Splenocytes (3.2 × 10⁸ cells/mL) were incubated with islet lysates (lysate of 1,600 islets/mL) in the presence of ECDI (150 mg/mL) on ice for 1 hour with intermittent shaking. The splenocytes coupled with cell-free islet lysate were washed with PBS and filtrated to remove cell clumps, and then resuspended in PBS. Unconjugated lysates and free ECDI had been eliminated by above washing and filtration. A total of 1 × 10⁸ splenocytes coupled with islet lysate from about 500 islets were injected intravenously to female NOD mice firstly at the age of 10 weeks and secondly at 11 weeks, and sham injection with equal volume of saline was performed in the control group.

Oral glucose tolerance test

Prior to sacrifice, oral glucose tolerance test (OGTT) was performed. NOD mice were fasted for 16 hours, and then administrated with glucose (50% W/V, 2 g/Kg) by intragastric gavage. Blood glucose levels were determined at 0, 15, 30, 45, 60, 90, and 120 minutes after glucose administration. By using the fasting blood glucose level as baseline, area under the curve (AUC) was calculated.

H&E and immunohistochemical staining

Pancreatic tissue was fixed in formaldehyde, embedded in paraffin blocks, and sectioned into 3 μm slices. Tissue slides were deparaffinized, rehydrated and stained with Hematoxylin and eosin (H&E), anti-mouse insulin (1:200, Abcam, Cambridge, UK) and anti-mouse glucagon (1:200, Abcam, Cambridge, UK), as previously described [19, 20]. At least 10 nonadjacent sections were analyzed per pancreas. At least 20 islets from each mouse pancreas were scored according to the following criteria: 0, no infiltration; 1, intact islets but with immune cells surrounding the islets (peri-isulitis); 2, immune cells infiltration in less than 50% of the area; 3, immune cells infiltration in more than 50% of the area; 4, islets that are completely destroyed.

Flow cytometry (FCM)

The NOD mice were scarified at the 10 weeks after first injection of SP-Islet lysate. At the same time, pancreatic draining lymph node cells and splenocytes were collected and assessed immediately. For analysis of memory T cells, after being washed with ice-cold PBS, cells were surface labeled with CD3-PE-Cyanine7, CD4-FITC, CD44-APC and CD62L-PE. For analysis of Treg cells, the samples were surface labeled with CD3-PE-Cyanine7, CD4-FITC and CD25-PE, and intracellularly stained with Foxp3-APC, following the manufacturer’s instruction (eBioscience^TM Mouse Regulatory T cell Staining kit, eBioscience, Califormia, USA). For analysis of Th1 and Th17 cells, the samples were suspended at a density of 5 × 10⁶ cells/mL in complete culture medium (RPMI 1640 supplemented with 1% Penicillin-Streptomycin Solution, and 10% FBS), and were stimulated for 5 hours using 50 ng/mL of phorbol myristate acetate (PMA, Sigma, St. Louis, Missouri, USA) and 1 μg/mL ionomycin (Solarbio, Beijing, China) at 37°C and 5% CO₂. Washed with ice-cold PBS, the cells were surface labeled with CD3-PE-Cyanine7, CD4-FITC. After fixation and permeabilization, the cells were stained with IL-17-PE and IFN-γ-APC. CD4-FITC, CD3-PE-Cyanine7, CD44-APC, CD62L-PE antibodies were purchased from eBioscience (San Diego, California, USA), and IFN-γ, and IL-17 antibodies from Biolegend (Dedham, Massachusetts, USA). All samples were analyzed using a CANTOII flow cytometer (BD Biosciences, Franklin, New Jersey, USA) and FlowJo cytometry analysis software (FlowJo, San Francisco, California, USA).

Autoantibody detection

IAA, ICA, GADA, IA-2A in sera from different treatment groups were measured with mouse autoantibodies ELISA kit (IAA, ICA, GADA, IA-2A) (Neoscientific, Boston, Massachusetts, USA) according to the manufacturer’s instructions.

Statistical analysis

T1DM incidence and survival rate were analyzed by Kaplan-Meier plot with a log-rank test. Differential autoantibodies levels, and percentage of lymphocyte subsets were analyzed by Student t test, and represented as mean ± SEM. AUC in the OGTT was calculated by SPSS 17 (Armonk, New York, USA). Other statistical analyses were performed using Graphpad Prism 7.2 (GraphPad, La Jolla, California, USA). p values <0.05 were considered as statistically significant.

Results

SP-Islet lysate treatment prevents the onset and progression of T1DM

Female NOD mice aged ≥10 weeks are in the effector phase of autoimmunity, and generally assumed as a clinically relevant model for the investigation of the potential efficacy of novel treatments for T1DM [21]. To assess the efficacy of SP-Islet lysate in preventing T1DM, BALB/c islet lysate was used as a soluble form of whole islet antigens, and coupled to the splenocytes isolated from BALB/c mice by ECDI fixation (SP-Islet lysate). SP-Islet lysate was intravenously injected into the recipient female NOD mice firstly at the age of 10 weeks and secondly at 11 weeks (SP-Islet lysate group, n = 18, Fig. 1A). Our preliminary study has compared the effect of ECDI-SP with saline on NOD mice. The results showed that ECDI-SP and saline had no obvious difference on the protection of diabetes incidence (Fig. S1), as well as tolerance induction (Fig. S2) in NOD mice. Thus, the present study adopted saline injection as control treatment (n = 10) to avoid non-necessary animal sacrifice. In our study, 60% of the mice became diabetic at 20 weeks of age in the control group, while only 16.7% of mice developed T1DM in the SP-Islet lysate group (p < 0.05, log rank test, Fig. 1B), with better survival rate (p < 0.01, log-rank test, Fig. 1C). These results indicate that SP-Islet lysate treatment decreased the incidence of T1DM in NOD mice.

Fig. 1

SP-Islet lysate treatment prevents the onset and progression of T1DM. (A) Schematic diagram for SP-Islet lysate treatment. SP-Islet lysate was intravenously (i.v.) injected to NOD mice aged 10 and 11 weeks, respectively. (B) The diabetes incidence was analyzed by Kaplan Meier curve with a log rank test. (C) The survival rate of NOD mice was analyzed by Kaplan Meier curve with a log rank test. (D) The levels of GADA, ICA, IAA, IA-2A in the serum of NOD mice aged 20 weeks were detected by ELISA and repeated at least for three times independently. (E) Representative photomicrographs of pancreas from NOD mice aged 20 weeks stained with hematoxylin and eosin (H & E). (F) Percentage distribution of insulitis levels: score 0 (white), score 1 (light grey), score 2 (grey), score 3 (dark grey), and score 4 (black). (G) Islet area in pancreas sections. Scale bar: 50 μm. *p < 0.05, **p < 0.01.

Autoantibodies, such as GADA, IAA, ICA, IA-2A, are considered as the typical markers for autoimmunity in T1DM [22]. To determine the effect of SP-Islet lysate on the autoimmunity development, the levels of GADA, IAA, ICA, IA-2A in mice blood serum were detected by ELISA in 20-week-old mice (10 weeks after treatment). The level of GADA, IAA, IA-2A of the mice in treatment group were significantly lower than those in control group, with a relative decrease of 80.2% (p < 0.01), 53.1% (p < 0.01), and 63.3% (p < 0.01), respectively (Fig. 1D). However, there was no significant decrease in ICA level after SP-Islet lysate treatment (p = 0.49, Fig. 1D).

The immune cells specifically damage pancreatic β-cells and the presence of inflammation peri-islet predicts the progression of T1DM disease in NOD mice. To further elucidate the effect of SP-Islet lysate treatment, we analyzed the immune cells infiltration by H&E staining and insulitis scoring. Results showed that insulitis score was reduced in the SP-Islet lysate treated group (Fig. 1E & F). This was clearly seen from islets with insulitis score ≥2, which was 35.7% in the SP-Islet lysate treated group and 56.3% in control group (Fig. 1F). Concomitantly, islet area in pancreas was estimated to be 150% greater in the SP-Islet lysate treated group than in the control (14,251 ± 2,768 vs. 5,703 ± 1,262 μm²/islet, p < 0.01) (Fig. 1G). This indicates the SP-Islet treatment prevented the occurrence of harmful insulitis. Taken together, the results above imply that SP-Islet lysate treatment effectively prevented the onset and progression of T1DM.

SP-Islet lysate treatment maintains islet function

Insulitis seriously influences islet function. The impairment of glucose tolerance in T1DM predicts the impairment of islet function. OGTT results demonstrated that the SP-Islet lysate treated mice had better glucose tolerance, evidenced by the greater AUC (p < 0.05, Fig. 2A & B). This implies that SP-Islet lysate treatment prevented impairment of glucose tolerance. Concomitantly, the IHC result showed that the SP-Islet lysate treatment significantly ameliorated the islet function, with significant increase of insulin positive cells (72.92 ± 5.014 for the treated group, 37.76 ± 8.804 for the control group, p < 0.05), as well as glucagon positive cells (27.54 ± 2.926 for the treated group, 14.44 ± 2.642 for the control group, p < 0.05), while islet function was well maintained in the treated mice (Fig. 2C & D). These results indicate that SP-Islet lysate treatment maintained the islet function of NOD mice.

Fig. 2

SP-Islet lysate treatment improves glucose homeostasis and maintains insulin and glucagon secretion in the pancreatic islets. (A) Oral glucose tolerance test (OGTT) was performed on NOD mice aged 20 weeks, which were administrated with 50% W/V glucose 2 g/kg body weight after 16-hour fasting. Blood glucose was measured at the indicated time afer glucose administration. (B) Area under the curve (AUC) was calculated for the OGTT by using the fasting blood glucose level as baseline. (C) Immunohistochemistry against insulin (up layer) and glucagon (down layer) and (D) Quantitative analysis of the percent of insulin and glucagon in pancreatic islet of NOD mice from 20 weeks. Scale bar: 50 μm. *p < 0.05.

SP-Islet lysate treatment decreases the frequency of autoreactive effector memory T-cells and Th1&Th17 cells

The percentage of memory T cells was reported to be increased in NOD mice when developed T1DM [23], and autoreactive memory T cells contributes to the difficulty in preventing disease progression in new-onset T1DM [2]. As shown in Fig. 3A, effector memory T cells (T_EMs) were represented by CD4⁺CD44^highCD62L^– T cell populations. SP-Islet lysate treatment downregulated T_EMs both in the pancreatic draining lymph nodes (11.70 ± 1.93% for the treatment group, 34.06 ± 3.63% for the control group, p < 0.001, Fig. 3B) and in the spleens (13.7 ± 0.96% for the treatment group, 17.95 ± 1.09% for the control group, p < 0.05, Fig. 3B). However, there is no obvious change for the central memory T cells (p = 0.52 for the T_EMs in pancreatic draining lymph nodes, and p = 0.53 for the splenic T_EMS, Fig. 3C & D). These results suggest that SP-Islet lysate treatment can inhibit the stimulation of autoreactive T_EMs but not the T_CMs.

Fig. 3

SP-Islet lysate treatment decreases the frequency of autoreactive T_EMs and Th1/Th17. (A) Representative FCM dot plots of memory T-cells in pancreatic draining lymph nodes and spleens from NOD mice in control group and SP-Islet lysate treated group aged 20 weeks. Each sample was stained with CD3, CD4, CD44, and CD62L antibodies. CD3 and CD4 were used to gate CD4⁺T cells. CD44 and CD62L were used to gate memory T-cells. (B) Quantitative analysis of the frequency of T_EMs (CD4⁺ CD44^high CD62L^–) and T_CMs (CD4⁺ CD44^high CD62L⁺). (C) FCM analysis of Th1 and Th17 lymphocyte subsets in pancreatic draining lymph nodes and spleens from 20-week NOD mice treated by SP-Islet lysate and control. Each sample was stained with CD3, CD4, IFN-γ and IL-17 antibodies, CD3 and CD4 was used to gate CD4⁺ T cells. IFN-γ and IL-17 were used to gate Th1 and Th17. (D) The proportions of Th1 cells (CD4⁺ IFN-γ⁺ IL17^–), Th17 cells (CD4⁺ IL17⁺ IFN-γ^–) in each group. *p < 0.05, **p < 0.01, ***p < 0.001. Experiments were repeated three times independently.

Th1 and Th17 have been implicated in the pathogenesis of T1D in mice and humans due to secreting pro-inflammatory cytokines such as IL-17, IL-21, IL-27, IFN-γ and TGF-β [24]. In our study, the IFN- γ⁺CD4⁺ T cells (Th1) and IL-17⁺CD4⁺ T cells (Th17) were both downregulated by SP-Islet treatment in both pancreatic draining lymph nodes (p < 0.05) and spleens (p < 0.01, Fig. 3B).

SP-Islet lysate treatment upregulates the frequency of Tregs

Previous studies displayed that an increasing loss of Tregs function conduces to autoimmunity in the NOD-mouse models of T1DM [25, 26]. Tregs have been shown to suppress the functions of effector T cells through multiple mechanisms including (but not limited to) production of inhibitory cytokines (IL-10, TGF-β), IL-2 sequestration, and blockade of the co-stimulatory molecule [27]. In our study, FCM was used to analyze the frequency of CD4⁺CD25⁺Foxp3⁺ Treg cells in pancreatic draining lymph nodes and spleens (Fig. 4). The results showed that the percentage of CD4⁺CD25⁺Foxp3⁺ Tregs of the pancreatic lymph nodes in the SP-Islet lysate group was significantly higher than that in control group (8.04 ± 0.43% for treatment group, 5.48 ± 0.55% for the control group, p < 0.01, as shown in Fig. 4B). Similarly, the splenic Tregs population also presents a significant upregulation (4.78 ± 1.18% for treatment group, 2.42 ± 0.35% for the control group, p < 0.05, as shown in Fig. 4B). These results indicated that SP-Islet lysate treatment upregulated the frequency of CD4⁺CD25⁺Foxp3⁺ Tregs in NOD mice.

Fig. 4

SP-Islet lysate treatment upregulates the Tregs frequency. (A) Representative FCM panels of Tregs in pancreatic draining lymph nodes and spleens harvested from NOD mice in control group and SP-Islet lysate treated group aged 20 weeks. Each sample was stained with CD3, CD4, CD25, and Foxp3 antibodies. CD3 and CD4 were used to gate CD4⁺ T cells. CD25 and Foxp3 were used to gate Tregs. (B) Quantitative analysis of the frequency of Tregs (CD4⁺CD25⁺Foxp3⁺). *p < 0.05, **p < 0.01. Experiments were repeated three times independently.

Discussion

T1DM occurs as a result of an immune regulation disorder, including the formation of T1DM-associated autoantibodies, the expansion of autoreactive CD4⁺ and CD8⁺ T cells, and the imbalance of Tregs and Th1&Th17 cells. These factors collaborate to destroy β-cells in islets [28]. ECDI-fixed splenocytes coupled with special antigens can induce antigen-specific tolerance [13], and this method is currently receiving much attention for its potential to serve as a clinical therapeutic strategy for autoimmune diseases. In this study, we detected the effect of SP-Islet lysate on the incidence of T1DM in NOD mice. The results revealed that this treatment effectively provided protection against the onset and progression of T1DM in NOD mice.

Islet autoantibodies are considered as the markers of T1DM-related autoimmunity, and are associated with the onset of T1DM [29]. The present study found that SP-Islet lysate treatment significantly reduced the levels of GADA, IA-2A, IAA in serum, indicating a general inhibition on humoral autoimmunity. Interestingly, no significant decrease of ICA levels was observed in the SP-Islet lysate treatment group in this study. This may be due to the amount of islet antigens in SP-Islet lysate system exceeded the proper antigen amount for tolerance induction.

Accumulated studies in NOD mice have demonstrated that T1DM occurred as a consequence of the β-cell destruction by the β-cell-autoreactive CD4⁺ and CD8⁺ T cells [4, 27, 30-32]. In our study, similar to the changing trend of autoantibodies, the frequency of T_EMs also presented a significant decrease in the SP-Islet lysate group compared with control group. These results indicate that the ECDI fixed splenocytes carrying whole islet antigens can potently induce antigen-specific tolerance. Two possible reasons might collectively explain this phenomenon. At first, β-cell antigens existed in ill-defined environmental cues can not only directly stimulate the production of autoantibodies, but also govern the islets infiltration and the activation of naive CD4⁺ T cells to antigen-responsive memory T cells [27, 33]. Secondly, autoantibodies and autoreactive memory T cells can mutually stimulate each other and further amplify the autoimmunity. Heninger et al. found that the activation of memory CD4⁺ T cells could further promote the maturation of B cells and hence stimulate the secretion of autoantibodies [34]. The existence of autoantibodies can, in turn, induce the activation and proliferation of antigen-responsive CD4⁺ T cells, reported by two recent studies [35, 36]. Taken together, ECDI-fixed splenocytes carrying whole islet antigens can effectively inhibit both humoral and cellular autoimmunity involved in T1DM, and hence confer protection against the onset and progression of T1DM in NOD mice. Nonetheless, in the tolerance induced by SP-Islet lysate treatment, the relationship of humoral and cellular autoimmunities still needs to be elucidated in future.

Imbalance of Th1&Th17 and Tregs is crucial for generation of increasing T_EMs in T1DM [37]. While some studies have shown that tolerance-inducing therapies can lead to an increase in Tregs in lymphoid and non-lymphoid target tissue, conversely, it is difficult to inhibit T_EMs formation in autoimmunity diseases [38]. In our study, the treatment of SP-Islet lysate had an anticipated effect in promoting Tregs, what’s more important, the frequency of T_EMs and Th1/Th17 also presented a significant decrease in the SP-Islet lysate group compared with control group. Therefore, another reason for the reduction of T_EMs by SP-Islet lysate treatment might be partially attributed to the preservation of the balance between Th1&Th17 and Tregs.

In summary, the treatment of ECDI-fixed splenocytes carrying whole islet antigens effectively cut down the morbidity of T1DM in NOD mice, and maintained the islet functions. This protective effect is mediated by reduced response of memory T cells and secretion of autoantibodies in SP-Islet lysate treated NOD mice. Our findings imply that this treatment has a potential to be employed as an immunotherapy to prevent the clinical onset of T1DM in genetically predisposed individuals. In future, elegant studies are needed for the clinical translation of this treatment.

Acknowledgement

This work was supported by grants from the National key research and development program (2016YFC1305104), Tianjin Clinical Research Center for Organ Transplantation Project (15ZXLCSY00070), Scientific Research Foundation for Returned Scholars, Ministry of Education of China (730-BE034251), and Foundation of State Key Laboratory of Medicinal Chemical Biology (Nankai University) (2018016).

Disclosure

The authors declare that they have no conflict of interest to disclose.

Reference

• 1   Katsarou  A,  Gudbjornsdottir  S,  Rawshani  A,  Dabelea  D,  Bonifacio  E, et al. (2017) Type 1 diabetes mellitus. Nat Rev Dis Primers 3: 17016.
• 2   Ehlers  MR,  Rigby  MR (2015) Targeting memory T cells in type 1 diabetes. Curr Diab Rep 15: 84.
• 3   Monti  P,  Scirpoli  M,  Rigamonti  A,  Mayr  A,  Jaeger  A, et al. (2007) Evidence for in vivo primed and expanded autoreactive T cells as a specific feature of patients with type 1 diabetes. J Immunol 179: 5785–5792.
• 4   Masteller  EL,  Warner  MR,  Tang  Q,  Tarbell  KV,  McDevitt  H, et al. (2005) Expansion of functional endogenous antigen-specific CD4+CD25+ regulatory T cells from nonobese diabetic mice. J Immunol 175: 3053–3059.
• 5   Ziegler  AG,  Rewers  M,  Simell  O,  Simell  T,  Lempainen  J, et al. (2013) Seroconversion to multiple islet autoantibodies and risk of progression to diabetes in children. JAMA 309: 2473–2479.
• 6   Paschou  SA,  Papadopoulou-Marketou  N,  Chrousos  GP,  Kanaka-Gantenbein  C (2018) On type 1 diabetes mellitus pathogenesis. Endocr Connect 7: R38–R46.
• 7   Ilonen  J,  Lempainen  J,  Hammais  A,  Laine  AP,  Harkonen  T, et al. (2018) Primary islet autoantibody at initial seroconversion and autoantibodies at diagnosis of type 1 diabetes as markers of disease heterogeneity. Pediatr Diabetes 19: 284–292.
• 8   Forlenza  GP,  Rewers  M (2011) The epidemic of type 1 diabetes: what is it telling us? Curr Opin Endocrinol Diabetes Obes 18: 248–251.
• 9   Lam  DW,  LeRoith  D (2012) The worldwide diabetes epidemic. Curr Opin Endocrinol Diabetes Obes 19: 93–96.
• 10   Dabelea  D (2009) The accelerating epidemic of childhood diabetes. Lancet 373: 1999–2000.
• 11   Haller  MJ,  Atkinson  MA,  Schatz  D (2005) Type 1 diabetes mellitus: etiology, presentation, and management. Pediatr Clin North Am 52: 1553–1578.
• 12   Prasad  S,  Kohm  AP,  McMahon  JS,  Luo  X,  Miller  SD (2012) Pathogenesis of NOD diabetes is initiated by reactivity to the insulin b chain 9–23 epitope and involves functional epitope spreading. J Autoimmun 39: 347–353.
• 13   Turley  DM,  Miller  SD (2007) Peripheral tolerance induction using ethylenecarbodiimide-fixed APCs uses both direct and indirect mechanisms of antigen presentation for prevention of experimental autoimmune encephalomyelitis. J Immunol 178: 2212–2220.
• 14   Smith  CE,  Eagar  TN,  Strominger  JL,  Miller  SD (2005) Differential induction of ige-mediated anaphylaxis after soluble vs. Cell-bound tolerogenic peptide therapy of autoimmune encephalomyelitis. Proc Natl Acad Sci U S A 102: 9595–9600.
• 15   Chen  G,  Kheradmand  T,  Bryant  J,  Wang  S,  Tasch  J, et al. (2012) Intragraft CD11b(+) IDO(+) cells mediate cardiac allograft tolerance by ecdi-fixed donor splenocyte infusions. Am J Transplant 12: 2920–2929.
• 16   Wang  S,  Zhang  X,  Zhang  L,  Bryant  J,  Kheradmand  T, et al. (2015) Preemptive tolerogenic delivery of donor antigens for permanent allogeneic islet graft protection. Cell Transplant 24: 1155–1165.
• 17   Kheradmand  T,  Wang  S,  Bryant  J,  Tasch  JJ,  Lerret  N, et al. (2012) Ethylenecarbodiimide-fixed donor splenocyte infusions differentially target direct and indirect pathways of allorecognition for induction of transplant tolerance. J Immunol 189: 804–812.
• 18   Anderson  MS,  Bluestone  JA (2005) The NOD mouse: a model of immune dysregulation. Annu Rev Immunol 23: 447–485.
• 19   Emamaullee  JA,  Davis  J,  Merani  S,  Toso  C,  Elliott  JF, et al. (2009) Inhibition of Th17 cells regulates autoimmune diabetes in NOD mice. Diabetes 58: 1302–1311.
• 20   Girard  CA,  Wunderlich  FT,  Shimomura  K,  Collins  S,  Kaizik  S, et al. (2009) Expression of an activating mutation in the gene encoding the KATP channel subunit Kir6.2 in mouse pancreatic beta cells recapitulates neonatal diabetes. J Clin Invest 119: 80–90.
• 21   Shoda  LK,  Young  DL,  Ramanujan  S,  Whiting  CC,  Atkinson  MA, et al. (2005) A comprehensive review of interventions in the NOD mouse and implications for translation. Immunity 23: 115–126.
• 22   Barker  JM,  Barriga  KJ,  Yu  L,  Miao  D,  Erlich  HA, et al. (2004) Prediction of autoantibody positivity and progression to type 1 diabetes: diabetes autoimmunity study in the young (daisy). J Clin Endocrinol Metab 89: 3896–3902.
• 23   Sheu  TT,  Chiang  BL,  Yen  JH,  Lin  WC (2014) Premature CD4+ T cell aging and its contribution to lymphopenia-induced proliferation of memory cells in autoimmune-prone non-obese diabetic mice. PLoS One 9: e89379.
• 24   Baharlou  R,  Ahmadi-Vasmehjani  A,  Davami  MH,  Faraji  F,  Atashzar  MR, et al. (2016) Elevated levels of T-helper 17-associated cytokines in diabetes type i patients: indicators for following the course of disease. Immunol Invest 45: 641–651.
• 25   Kornete  M,  Mason  E,  Istomine  R,  Piccirillo  CA (2017) KLRG1 expression identifies short-lived Foxp3+ Treg effector cells with functional plasticity in islets of NOD mice. Autoimmunity 50: 354–362.
• 26   Li  CR,  Mueller  EE,  Bradley  LM (2015) Targeting CD44 augments the efficacy of tregs in autoimmune diabetes. Immunol Lett 163: 199–205.
• 27   Burton  AR,  Vincent  E,  Arnold  PY,  Lennon  GP,  Smeltzer  M, et al. (2008) On the pathogenicity of autoantigen-specific T-cell receptors. Diabetes 57: 1321–1330.
• 28   Bluestone  JA,  Herold  K,  Eisenbarth  G (2010) Genetics, pathogenesis and clinical interventions in type 1 diabetes. Nature 464: 1293–1300.
• 29   Steimle  A,  Frick  JS (2016) Molecular mechanisms of induction of tolerant and tolerogenic intestinal dendritic cells in mice. J Immunol Res 2016: 1958650.
• 30   DiLorenzo  TP,  Serreze  DV (2005) The good turned ugly: immunopathogenic basis for diabetogenic CD8+ T cells in NOD mice. Immunol Rev 204: 250–263.
• 31   Zipris  D,  Lien  E,  Nair  A,  Xie  JX,  Greiner  DL, et al. (2007) TLR9-signaling pathways are involved in kilham rat virus-induced autoimmune diabetes in the biobreeding diabetes-resistant rat. J Immunol 178: 693–701.
• 32   Devendra  D,  Jasinski  J,  Melanitou  E,  Nakayama  M,  Li  M, et al. (2005) Interferon-alpha as a mediator of polyinosinic: polycytidylic acid-induced type 1 diabetes. Diabetes 54: 2549–2556.
• 33   Arif  S,  Tree  TI,  Astill  TP,  Tremble  JM,  Bishop  AJ, et al. (2004) Autoreactive T cell responses show proinflammatory polarization in diabetes but a regulatory phenotype in health. J Clin Invest 113: 451–463.
• 34   Heninger  AK,  Eugster  A,  Kuehn  D,  Buettner  F,  Kuhn  M, et al. (2017) A divergent population of autoantigen-responsive CD4+ T cells in infants prior to beta cell autoimmunity. Sci Transl Med 9: eaaf8848.
• 35   Burns  AM,  Chong  AS (2011) Alloantibodies prevent the induction of transplantation tolerance by enhancing alloreactive T cell priming. J Immunol 186: 214–221.
• 36   Phan  TG,  Grigorova  I,  Okada  T,  Cyster  JG (2007) Subcapsular encounter and complement-dependent transport of immune complexes by lymph node B cells. Nat Immunol 8: 992–1000.
• 37   Zhang  C,  Shao  Y,  Yang  S,  Bi  X,  Li  L, et al. (2017) T-cell tolerance and exhaustion in the clearance of Echinococcus multilocularis: role of inoculum size in a quantitative hepatic experimental model. Sci Rep 7: 11153.
• 38   Devarajan  P,  Miska  J,  Lui  JB,  Swieboda  D,  Chen  Z (2014) Opposing effects of CTLA4 insufficiency on regulatory versus conventional T cells in autoimmunity converge on effector memory in target tissue. J Immunol 193: 4368–4380.