Autoimmunity is a Protective Response to β Cell Damage in Type 1 Diabetes

6/1/2019

Introduction

In 1984, Eisenbarth and Rabinowe reviewed the emerging case for considering Type 1 diabetes to be an autoimmune disease. They suggested that this hypothesis “will probably be proven only if immunotherapy is successful in preventing or “curing” the illness”[1]. They also explained that the current trials with immunosuppressants had shown no clinical benefit. But, as a new theoretical approach with convincing immunological research to back it, the hopes were that some variation of immunosuppressive treatment would cure Type 1 diabetes[2]. 35 years later, immunosuppressive drugs have not proven beneficial[3] and in many cases are diabetogenic, causing insulin resistance, decreased insulin secretion, and β cell damage[4]. Alternatively, various immunostimulant therapies have protected mice from diabetes[5] and the diabetes protective effects of some infections[6] are leading towards an implication for immunostimulation in humans[7]. Eisenbarth and Rabinowe noted that the diabetes-prone Biobreeding (DPBB) rats and non-obese diabetic (NOD) mice have severe lymphopenia, a necessary factor in their diabetes development[8][9], but that this particular characteristic is not seen in human diabetics[1]. The presence of immunodeficiency in human diabetes must still be considered in light of the findings that diabetics have significantly lower serum IgG levels[10], the in vitro behaviors of immune cells from diabetics show similarities to those in immunodeficiency syndromes[11], and that Vitamin D, which is needed for innate[12] and adaptive[13] immune function, is often deficient in high-risk individuals[14] and at the onset of the disease[15].

In the current model that suggests that the immune system discriminates between self and non-self to protect against foreign organisms[16], autoimmunity appears to be a fundamental mistake. Indeed, autoimmunity can be mistaken when triggered by a molecular mimicry between an immune-targeted antigen and a self-antigen, but in multiple animal models of Type 1 diabetes, molecular mimicry has shown the capacity to enhance but not initiate the autoimmunity[17][18]. Autoimmunity can otherwise be triggered by tissue damage[19]. For instance, heart specific autoantibodies are detected after myocardial infarction[20], and in acute burn patients, autoantibodies reactive to multiple skin antigens are found[21]. Polly Matzinger has suggested that autoimmunity may regularly occur after any tissue trauma and resolve once the tissue is healed[22][19]. Therefore, autoimmunity appears to be more of a commonality than an aberration, and as such, it would likely prove to have a conserved purpose. Accumulating evidence suggests that autoimmunity can promote healing after trauma. Detection of autoantibodies to the wounded tissue is associated with enhanced cutaneous wound healing[23]. Autoimmunity has been seen to support repair from glutamate damage to the central nervous system[24] and from traumatic injury to the spinal cord[25]. Jamie Cunliffe has pressed for renaming the immune system as the “morphostatic system”, a system that responds to disturbance of tissue organization and works to restore and maintain order[26]. Polly Matzinger’s “danger theory” similarly poses that the immune system does not react to foreignness but to danger and to the antigens associated with that danger[27]. In these new models, autoimmunity is an understandable reaction to dangerous destruction of tissues.

β Cell Damage Triggers Autoimmunity

The innate immune system is the first line of defense and the link to the adaptive system. During normal conditions, the islet resident antigen presenting cells (APCs) are mainly macrophages, with fewer dendritic cells[28]. Activation of nuclear factor-kB (NF-kB), a mediator of cellular stress[29], causes β cells to produce monocyte chemoattractant protein-1 (MCP-1)[30], which recruits more monocytes to the islets[31]. Macrophages and dendritic cells are the first cells to infiltrate the islets of DPBB rats and NOD mice[32]. In humans, where it is unfeasible to study the pancreas at this stage, innate immune activity is detected in blood RNA samples before seroconversion to autoantibody positivity[33]. The pre-autoimmune state is also reflected in the blood as raised levels of pro-inflammatory lysophosphatidylcholine, glutamate, and branched chain amino acids, and a decrease in citric acid cycle metabolites several months before seroconversion[34]. Interestingly, this metabolic profile normalizes upon the appearance of autoimmunity, perhaps indicating a resolution of some stress. These findings support the notion that β cell stress precedes and is not caused by autoimmunity.

Due to their ability to enhance inflammation and activate autoreactive lymphocytes, macrophages have been considered pathogenic in diabetes, but in a mouse model of pancreatitis-induced diabetes, macrophages are essential for β cell preservation and regeneration[35]. Selective elimination of macrophages using Cl2MPD liposomes does not protect against streptozotocin-induced diabetes, and the diabetes protective effect of silica, originally thought to deactivate macrophages, appears instead to be the result of an irritating activation of the macrophages[36]. Transgenic expression of MCP-1 on β cells via the insulin promoter, which leads to an increased monocyte infiltration, suppresses diabetes development[37].

In order for β cell autoreactive lymphocytes to become activated, they must come in contact with APCs that are presenting β cell antigens on major histocompatibility complex (MHC), signal one, with costimulatory molecules, signal two[38]. Phagocytes engulf dying cells and present their antigens in both MHC I and II[39]. When autoreactive lymphocytes encounter their cognate antigen without costimulation they are induced to anergy[40]. APCs will present costimulatory molecules and produce inflammatory cytokines when their pattern recognition receptors (PRRs) are stimulated by pathogen or damage associated molecular patterns (PAMPs or DAMPs), collectively known as danger signals[22][41]. These danger signals are intracellular molecules that are released into the interstitium when microbes or host cells rupture, releasing PAMPs or DAMPs respectively[42].

Apoptosing cells die in a controlled way to prevent the release and neutralize the immunostimulatory potential of DAMPS[43]. Exposure to apoptosing β cells causes APCs to adopt an anti-inflammatory cytokine profile (TGFβ and IL-10)[44]. These APCs, presenting only signal one, will tolerize β cell specific lymphocytes. Conversely, necrotic cells spill their inflammatory DAMPs[45]. A number of DAMPs seen in human islet cell necrosis are high mobility group box 1 (HMGB1), uric acid, and double stranded DNA fragments, which mainly signal through the PRR class, toll like receptors (TLRs)[46]. In DPBB rats, necrosis is seen to be the predominant form of β cell death in the prediabetes period[47], and the β cells of prediabetic NOD mice release HMGB1, indicating necrotic death[48].

PAMPs may participate in amplifying the immune response in Type 1 diabetes, as high levels of the bacterial PAMP, lipopolysaccharide (LPS), has been associated with the disease[49], but PAMPs alone cannot trigger autoimmunity. If macrophages have accumulated the auto-antigens from apoptosing cells, they will maintain an anti-inflammatory profile even when exposed to LPS[50][51][52]. Necrotic cells, however, enhance LPS stimulated release of pro-inflammatory cytokines[53]. Therefore, if microbes participate in the onset of autoimmunity in diabetes, it would be either through directly causing β cell necrosis or by amplifying the immune response to necrotic cells.

Necrosis can occur via two pathways: primary or secondary necrosis. Primary necrosis occurs when an extreme insult causes cells to rupture immediately and release immunostimulatory DAMPs[45]. This results in a potent innate mediated inflammatory response but it is not seen to trigger adaptive immunity[45]. The selective β cell toxin, streptozotocin, administered in single-high doses causes extensive β cell primary necrosis and diabetes without triggering autoimmunity[54][55]. Secondary necrosis occurs when apoptotic cells are not effectively cleared by phagocytes and eventually lose their membrane integrity, thus also releasing DAMPs[45]. Secondary necrosis is morphologically identical to primary necrosis but secondary necrotic cells will have gone to some length to modify intracellular components along the apoptotic pathway[45]. As a result, DAMPs released from secondary necrotic cells are sometimes less immunostimulatory, but persisting or massive secondary necrosis is able to induce adaptive immunity and autoimmunity[45]. A single low dose of streptozotocin causes low amounts of β cell apoptosis and tolerization of lymphocytes[56]. However, a low dose repeated over several days can induce autoimmune diabetes in mice[57] and primates[58]. Streptozotocin has a half-life of fifteen minutes so the daily dosages do not accumulate to higher concentrations[59], but the multiple apoptotic events overwhelm the phagocytic capacity and result in secondary necrosis[60][61].

The different immune responses to primary and secondary necrosis may be explained by their unique forms of DAMPs. HMGB1 may be the key inflammatory DAMP since necrotic cells that lack HMGB1 do not elicit an inflammatory response[41]. HMGB1 that is freely released from primary necrotic cells is quickly oxidized to a state that allows it to agonize the TLR4 receptor[62][63]. Further oxidation in the extracellular environment then inactivates it, meaning that its inflammatory activity is naturally limited[62][63]. During apoptosis HMGB1 becomes sequestered in the nucleus and tightly bound to chromatin on the nucleosomes to prevent its release[64]. When apoptosis proceeds to secondary necrosis, HMGB1-nucleosome complexes are released and signal through TLR2[64]. Secondary necrotic β cells have been seen to activate autoreactive lymphocytes through a TLR2 dependent stimulation of APCs[44]. Interestingly, an agonist for TLR2 was seen to protect against diabetes twice as quickly as agonists for other TLRs[65].

Macrophage engulfment of apoptotic cells in NOD mice is 5.5 times slower than that of controls[66]. Vitamin D deficiency causes an increased tendency towards apoptosis of β cells in the presence of inflammatory cytokines and a decreased rate of phagocytosis[67]. This imbalance leads to a higher tendency towards secondary necrosis and the release of DAMPs. When phagocytosis is insufficient to clean up the damage, the APCs must call upon the next shell of the immune system, the lymphocytes.

Absolving T Cells

Despite the popular insistence that autoreactivity is a pathological mistake, autoreactive CD4 T cells can be found in most healthy individuals[68]. CD4 and CD8 T cells reactive to the diabetes-related β cell antigen glutamic acid decarboxylase (GAD) are detected in healthy subjects but only show a memory phenotype in diabetic subjects, indicating their prior activation[69]. This shows that these autoreactive T cells are part of the normal T cell repertoire. Autoreactive lymphocytes are capable of escaping central tolerance in the thymus. If the self-antigen to which they are reactive is absent from the thymus, they will survive[70]. If they have a low affinity for their self-peptide, and especially if that self-peptide is presented in low amounts, they will also survive[70]. As many as 25-40% of self-specific T cells escape clonal deletion in the presence of their antigen due to a low affinity phenotype[71]. Although low in affinity, these autoreactive lymphocytes can be activated when their antigen is presented by APCs in relatively greater abundance than found in the thymus and along with costimulation[70][72]. Autoreactive T cells with dual T cell receptors (TCRs) can also escape negative selection when the self-reactive TCR is relatively hidden by the other non-self-reactive TCR[73]. Some autoreactive lymphocytes escaping central tolerance will then undergo peripheral tolerization when they interact with their peptide in the absence of costimulation[74]. Low-affinity autoreactive T cells, however, are also more elusive to peripheral tolerization[72]. Autoreactive T cells that have recently escaped central tolerance and are yet to experience peripheral tolerance can be activated if they meet their antigen for the first time in a dangerous context.

As Type 1 diabetes is believed to be caused by T cell mediated destruction of the β cells, the presence of T cell infiltration of the islets (insulitis) would be expected in every case of diabetes. However, this is not the case[75][76]. Insulitis was detected in 73% of young (≤14 years) Type 1 diabetics within a month of the disease onset, and in only 29% of older (15-40 years) Type 1 diabetics within the first month[76]. In the young population with detectable insulitis and with recent onset diabetes, 34% of the islets contained β cells and only 33.6% of the islets showed insulitis[76]. In the older population with detectable insulitis and recent onset, 63% of the islets contained β cells and only 18.3% of the islets showed insulitis[76]. This data came from diabetics that often died in ketoacidosis and thus potentially represents a more severe version of the disease[76]. Therefore, insulitis, the supposed hallmark of diabetes, is not present in all Type 1 diabetic pancreata, and in those where it is present, it has an apparent protective effect. In the pancreata of Type 1 diabetic donors, β cell area and mass were significantly higher in donors with insulitis compared to those without insulitis[77].

Transferring diabetes with lymphocytes from diabetic animals has been considered proof for the diabetogenic effect of autoreactive T cells. Transfer of splenocytes from diabetic NOD mice to young non-diabetic NOD mice accelerated the onset of diabetes in 95% of the 82 recipients within 28 days of the treatment only if they had been immunosuppressed by 775rad of γ-radiation from cesium-137[78]. Ten control mice treated with the radiation alone did not develop diabetes within 28 days but did show significant insulitis[78]. In another study, 100% of thymectomized, diabetes-resistant BB rats developed diabetes in around 28 +/-6 days after being irradiated with 950rad of γ-irradiation from cesium-137[79]. An increase in Type 1 diabetes incidence was observed in the radioactively contaminated areas around the Chernobyl disaster[80], which released cesium-137[81]. Another study relied on lethal doses of the immunosuppressant cyclophosphamide that killed half of the rats in order to render the remaining ones susceptible to diabetes transfer[82]. Cyclophosphamide at a similar dosage is itself used to induce overt diabetes in young NOD mice[83]. Therefore, most successful transfer studies rely on near diabetogenic doses of immunosuppressive poisons. One notable study did not use any preliminary treatment before a successful transfer of diabetes to younger NOD mice using isolated islet autoreactive CD4 T cell clones (BDC2.5 and BDC6.9)[84]. However, transgenic NOD mice expressing the BDC2.5 TCR on a large fraction of their T cells do not develop diabetes[85]. Infiltration occurs early but active lesions are not associated with β cell destruction, and instead appear to be protective against diabetes[86]. In these mice, the introduction of a mutation (RAG-1–/–) that prevented the rearrangement of genes encoding non-BDC2.5 antigen-specific receptors accelerated the onset of diabetes[86]. In the BDC2.5 RAG-1–/– mice, the transfer of splenocytes from young NOD donors significantly protected the recipients from diabetes[86]. The protective splenocytes were narrowed down to two distinct cell populations of CD4 αβ T cells, a DX5+ and DX5-[86]. These cells did not show a suppressor function as they did not interrupt the infiltration and activation of the autoreactive BDC2.5 cells[86]. The protective cells lacked the expression of both IL-4 (Th2) and IL-10 (Treg) and the introduction of IL-10 mildly inhibited a protective effect[86]. The donor cells expanded, developed activated or memory characteristics and concentrated mainly in the pancreas in both the BDC2.5 RAG-1–/– mice and other non-transgenic RAG-1–/– mice[86]. Neither CD4CD25 nor CD45RBlo Treg cells were seen to exert a protective effect[86]. The protective CD4 populations are apparently needed to ensure a protective autoimmune response. Therefore, the successful transfer of diabetes with BDC2.5 cells likely depended on the lymphopenic background of the NOD mice being unable to support a protective phenotype from the unphysiological bolus of autoreactive and activated BDC2.5 cells. When the background immune system is replete, the BDC2.5 cells adopt a harmless and likely protective autoimmune role. Radiation[87] and cyclophosphamide[88] worsen lymphopenia, thereby further diminishing the proper support for the incoming cell transfer.

Despite the inability of immunosuppressants to meet clinical goals, a few trials have demonstrated a prolonging of the honeymoon remission period using drugs that are believed to selectively inhibit lymphocytes. The two more successful substances used in these trials have been cyclosporin and anti-CD3 monoclonal antibody. Bougneres et al. (1990) used low-dose cyclosporin to significantly prolong insulin production in recent onset children for at most 2 years when used continually[89]. Cyclosporin inhibits the NFAT mediated transcription of a number of cytokines, importantly IL-2, which is considered an essential mediator of T cell activation and proliferation[90]. However, costimulation through CD28 induces T cell proliferation in a cyclosporin-resistant and IL-2-independent way[91], and cyclosporin significantly increases interferon-gamma (IFNγ) production by costimulated T cells starting at concentrations below and continuing into the therapeutic range used by Bougneres[92][93]. Therefore, in the context of β cell damage, cyclosporin may play an immunostimulatory role, increasing a Th1-like, IFNγ-based response. It should be noted, however, that cyclosporin can also lead to β cell dysfunction and diabetes in transplant patients[94]. The non-fcr binding anti-CD3 antibody, teplizumab, similarly caused a 75% improvement in C-peptide at 2 years after onset in humans[95]. The mechanism of action for the anti-CD3 antibody was purported to be through T cell suppression, but responders to the drug showed higher memory and effector CD8 cells and some suffered an over-expression of cytokines, signifying an immune activating response[95]. It has been seen in vitro that in human T cells, a non-fcr binding anti-CD3 antibody causes activation, proliferation and release of cytokines in the presence of costimulatory signals (Il-2 or CD28)[96]. An anti-cd3 antibody prevented diabetes in normal NOD mice, but in immunodeficient NOD BDC2.5 mice it had no protective effect, further suggesting a T cell stimulatory role[97].

Interventions that more reliably suppress lymphocyte activity have diabetogenic effects in mice. NOD mice made deficient in B7-1 and B7-2, removing major costimulatory signals, have exacerbated diabetes[98]. In these mice Th1, Th2 and CD4CD25 function were markedly reduced[98]. Thymectomy at the time of weaning, which results in depletion of T cell numbers, a loss of responsiveness to T cell specific mitogens and a decrease in immune reactivity[99], causes an acceleration of the onset of diabetes in NOD mice[100].

Protective Autoimmunity

There is an ongoing debate about which T cell subsets participate in diabetes. While earlier studies suggested that diabetes is mediated by a Th1-dependent attack, with an apparent deficiency of Th2 cells, more modern studies show that both subsets are involved[101]. Because cytokines from each subset suppress those of the other subset, it is likely that Th1 and Th2 activities differentiate temporally[102]. A likely time sequence drawn from mice and human studies is an early Th2 response[103][104], followed by long-term pre-diabetic Th1 activity[105], a temporary shift to Th3 at onset[106], and a return to Th2 as the disease progresses[103][107]. Th1 driven insulitic lesions are characterized by a rapid, aggressive and sustained activity of CD8 and CD4 T cells, where β cells die by apoptosis[104]. During this response, phagocytes lower their threshold of what they consider unhealthy cells and work alongside CD8 cells to cause apoptosis of those cells[26]. Th2 driven insulitic lesions are non-aggressive, short-lived, consist of an accumulation of eosinophils, macrophages, and fibroblasts, and are associated with β cell death by necrosis[104]. The Th2 response is focused on cleaning up necrotic death rather than promoting inflammation and apoptosis. Th2 cells cooperate more closely with B cells[108] that produce antibodies against the cellular debris in order to enhance the phagocytosis of the necrotic cells[26]. The polarization to Th1 or Th2 may be driven by the oxidized or reduced state of macrophage glutathione. Macrophages with a greater abundance of reduced glutathione induce a Th1 response while those with oxidized glutathione induce a Th2 response[109]. The oxidation status of macrophage glutathione reflects exposure to oxidative stress [110]. Higher exposure to oxidative stress, and in particular hypoxia-induced oxidative stress[103], will induce Th2 type insulitic lesions. In the diabetic pancreas, it has been seen that Th1 and Th2 driven insulitis are associated with reduced macrophages and oxidized macrophages respectively[109]. This may explain why Th2 activity is seen early on, perhaps in response to the initiating insult, and after symptoms of metabolic derangement have developed, both of which are likely times of high oxidative stress[111].

Both immune responses have protective intentions, limiting the spread of danger in different ways, but in particular, the pro-inflammatory Th1 activity may have the more regenerative effect. Th1-derived IFNγ causes β cell apoptosis[112], but also accelerates phagocytosis[113] and enhances β cell proliferation from ductal precursors[114][115][116], the sum of which amounts to tissue renovation, removing the old cells to make way for new. IFNγ has been seen to mediate the immunostimulant induction of diabetes resistance[117] and transgenic Balb/c mice expressing IFNγ in their pancreatic β cells are resistant to streptozotocin induced diabetes[118]. The inability of diabetic CD4CD25 Treg cells to suppress IFNγ secretion has been considered a sign of regulatory T cell dysfunction[119]. But activation of TLRs by danger signals releases effector T cells from Treg suppression[120][121][122][123], allowing them to respond adequately to damage. IFNγ levels gradually increase before overt diabetes onset, a time in which high β cell proliferation is observed[124], and decrease around onset[114][125], when proliferation is seen to be low[126]. This suggests that diabetes onset may be caused in part by decreased β cell proliferation in the absence of a sufficient IFNγ based immune response. Very few samples have been available for study from human prediabetic pancreata, during the time of high Th1 activity and high β cell proliferation. Amongst 62 samples from high-risk human pancreata, only two were detected to have insulitis in 3-9% of the islets[127]. The insulitis was not observed to be associated with atrophy of the islets and instead, in one patient, high levels of β cell proliferation were limited to the inflamed islets[127].

Th1 driven inflammation does have a regenerative potential, but in order for optimal recovery, inflammation must not be exaggerated or prolonged unnecessarily. Hauben and Schwartz (2003) saw that pro-inflammatory Th1 activity was necessary to supported recovery from spinal cord injury but proposed that either an over- or under-active response would respectively worsen or enable the damage to persist[128]. However, in both cases, vaccination with a relevant auto-antigen could benefit the organism. The neuroprotective effects of the T cells were thought to be mediated through the activation of microglia to clean up dead cells and cell debris, buffer toxic concentrations of nitric oxide and glutamate, and produce protective neurotrophins and cytokines. In the heart, repair of cardiomyocytes after damage requires the Th1 pro-inflammatory cytokine, oncostatin m[129], which also enhances IFNγ production[130], to cause dedifferentiation by activating fetal genes in order to promote proliferation of the cells[131][132]. However, the inability to resolve this inflammation, causing chronic dedifferentiation and functional deterioration, leads to a vulnerability to heart failure[131]. IFNγ similarly causes dedifferentiation of β cells, as seen by reduced insulin secretion and intracellular concentration[133], and rapidly proliferating β cells in vitro are seen to be dedifferentiated[134]. Resolution of inflammation, which enables redifferentiation of the newly proliferated cells, involves a number of factors[135], but most importantly it requires that the original threat to the target tissue be resolved.

Insulin-positive β cells are observed in 88% of Type 1 diabetics with a duration of diabetes between 4-67 years[136], and 80% of diabetics with a duration of more than five years have measurable, low levels of C-peptide[137]. Although capable of some insulin production, the residual β cells often exhibit degenerative changes like condensed nuclei, cellular edema, loss of cell to cell organization, and a higher frequency of apoptosis[136]. Together, this indicates a persisting regenerative effort that is undermined by an ongoing destructive insult. This ongoing regenerative effort is likely to be driven by the same immune factors involved since before onset. The frequency of residual β cells in diabetics that had undergone immunosuppression for renal transplantation prior to death was lower than those that had not had immunosuppression (5.3± 2.2 vs. 104.8±43.8 cells/cm2, p=0.19)[136].

Various immunostimulatory treatments have proven protective against diabetes in mice[5]. These include various infections, vaccinations with islet cell antigens, TLR agonists[65], transfer of islet-autoreactive T cells, and supplemental Vitamin D3[5]. In human diabetes, immunosuppressants have been deleterious and the few that have delayed full insulin dependence likely worked through unintended immunostimulation. But even today, after 35 years of immunosuppression without a cure, some are calling for intensified immunosuppression[138]. The BCR vaccine, which has classically been considered a non-specific immunostimulant, has been shown to transiently increase c-peptide in long-standing human diabetics[139]. The authors of this study claimed that the results are due to the induction of Tregs, but in another study the production of IFNγ from stimulated T cells remained elevated even a year after patients received the BCG vaccine[140]. In NOD mice, the BCR vaccine was shown to inhibit diabetes progression in an IFNγ dependent fashion[117] and without interrupting autoimmunity[141].

Conclusion

Autoimmunity appears to be the natural protective response to β cell damage and not the precipitator of it, but this response is unsuccessful in those who progress to diabetes. The inability to mount a fully protective autoimmune response due to immunodeficiency and the presence of continual β cell damage leads to chronic autoimmunity, as a perpetual attempt to resolve the problem, and eventual diabetes. Immunosuppression worsens diabetes outcomes by decreasing or deranging the autoimmune response while immunostimulation supports protective autoimmunity. Overt diabetes onset occurs when there is a drop in Th1 activity. Multiple factors could conceivably interrupt the protective autoimmune function, such as a vitamin D deficiency. Insulin deficiency itself may interrupt the protective response since a deficiency in insulin signaling reduces T cell driven inflammation by decreasing antigen specific proliferation and production of pro-inflammatory cytokines, including IFNγ[142]. Future articles will deal with uncovering the initiating and perpetuating agents of β cell destruction as well as the factors involved in ensuring protective autoimmunity and the proper differentiation[134][143] of newly regenerated β cells.

Bibliography

1. Rabinowe, S.L. and G.S. Eisenbarth, Type I diabetes mellitus: a chronic autoimmune disease? Pediatr Clin North Am, 1984. 31(3): p. 531-43.

2. Diem, P., [Will type I diabetes mellitus be treated with immunosuppressive drugs in the future?]. Ther Umsch, 1993. 50(2): p. 114-8.

3. Chatenoud, L., K. Warncke, and A.G. Ziegler, Clinical immunologic interventions for the treatment of type 1 diabetes. Cold Spring Harb Perspect Med, 2012. 2(8).

4. Jindal, R.M., R.A. Sidner, and M.L. Milgrom, Post-transplant diabetes mellitus. The role of immunosuppression. Drug Saf, 1997. 16(4): p. 242-57.

5. Hauben, E., et al., Beneficial autoimmunity in Type 1 diabetes mellitus. Trends Immunol, 2005. 26(5): p. 248-53.

6. Cooke, A., et al., Infection and autoimmunity: are we winning the war, only to lose the peace? Trends Parasitol, 2004. 20(7): p. 316-21.

7. Kukreja, A. and N.K. Maclaren, NKT cells and type-1 diabetes and the "hygiene hypothesis" to explain the rising incidence rates. Diabetes Technol Ther, 2002. 4(3): p. 323-33.

8. MacMurray, A.J., et al., Lymphopenia in the BB rat model of type 1 diabetes is due to a mutation in a novel immune-associated nucleotide (Ian)-related gene. Genome Res, 2002. 12(7): p. 1029-39.

9. King, C., et al., Homeostatic expansion of T cells during immune insufficiency generates autoimmunity. Cell, 2004. 117(2): p. 265-77.

10. Hoddinott, S., et al., Immunoglobulin levels, immunodeficiency and HLA in type 1 (insulin-dependent) diabetes mellitus. Diabetologia, 1982. 23(4): p. 326-9.

11. Moutschen, M.P., A.J. Scheen, and P.J. Lefebvre, Impaired immune responses in diabetes mellitus: analysis of the factors and mechanisms involved. Relevance to the increased susceptibility of diabetic patients to specific infections. Diabete Metab, 1992. 18(3): p. 187-201.

12. Lin, R., Crosstalk between Vitamin D Metabolism, VDR Signalling, and Innate Immunity. Biomed Res Int, 2016. 2016: p. 1375858.

13. von Essen, M.R., et al., Vitamin D controls T cell antigen receptor signaling and activation of human T cells. Nat Immunol, 2010. 11(4): p. 344-9.

14. Raab, J., et al., Prevalence of vitamin D deficiency in pre-type 1 diabetes and its association with disease progression. Diabetologia, 2014. 57(5): p. 902-8.

15. Littorin, B., et al., Lower levels of plasma 25-hydroxyvitamin D among young adults at diagnosis of autoimmune type 1 diabetes compared with control subjects: results from the nationwide Diabetes Incidence Study in Sweden (DISS). Diabetologia, 2006. 49(12): p. 2847-52.

16. Marshall, J.S., et al., An introduction to immunology and immunopathology. Allergy Asthma Clin Immunol, 2018. 14(Suppl 2): p. 49.

17. Filippi, C.M. and M.G. von Herrath, Viral trigger for type 1 diabetes: pros and cons. Diabetes, 2008. 57(11): p. 2863-71.

18. Coppieters, K.T., A. Wiberg, and M.G. von Herrath, Viral infections and molecular mimicry in type 1 diabetes. APMIS, 2012. 120(12): p. 941-9.

19. Mackay, I.R., N.V. Leskovsek, and N.R. Rose, Cell damage and autoimmunity: a critical appraisal. J Autoimmun, 2008. 30(1-2): p. 5-11.

20. Elahi, A.W., et al., Interplay of antibody and T cell responses in acute myocardial infarction. J Lab Clin Med, 2001. 138(2): p. 112-8.

21. Ablin, R.J. and I.A. Holder, Immunologic sequelae of thermal injury autoantibodies in acute adult burn patients. Arch Dermatol Res, 1982. 273(3-4): p. 267-71.

22. Matzinger, P., The danger model: a renewed sense of self. Science, 2002. 296(5566): p. 301-5.

23. Nishio, N., et al., Antibodies to wounded tissue enhance cutaneous wound healing. Immunology, 2009. 128(3): p. 369-80.

24. Schwartz, M., et al., Protective autoimmunity against the enemy within: fighting glutamate toxicity. Trends Neurosci, 2003. 26(6): p. 297-302.

25. Hauben, E., et al., Autoimmune T cells as potential neuroprotective therapy for spinal cord injury. Lancet, 2000. 355(9200): p. 286-7.

26. Cunliffe, J., Morphostasis and immunity. Med Hypotheses, 1995. 44(2): p. 89-96.

27. Matzinger, P., The evolution of the danger theory. Interview by Lauren Constable, Commissioning Editor. Expert Rev Clin Immunol, 2012. 8(4): p. 311-7.

28. Calderon, B., J.A. Carrero, and E.R. Unanue, The central role of antigen presentation in islets of Langerhans in autoimmune diabetes. Curr Opin Immunol, 2014. 26: p. 32-40.

29. Mercurio, F. and A.M. Manning, NF-kappaB as a primary regulator of the stress response. Oncogene, 1999. 18(45): p. 6163-71.

30. Kutlu, B., et al., Molecular regulation of monocyte chemoattractant protein-1 expression in pancreatic beta-cells. Diabetes, 2003. 52(2): p. 348-55.

31. Grewal, I.S., et al., Transgenic monocyte chemoattractant protein-1 (MCP-1) in pancreatic islets produces monocyte-rich insulitis without diabetes: abrogation by a second transgene expressing systemic MCP-1. J Immunol, 1997. 159(1): p. 401-8.

32. Mandrup-Poulsen, T., The role of interleukin-1 in the pathogenesis of IDDM. Diabetologia, 1996. 39(9): p. 1005-29.

33. Kallionpaa, H., et al., Innate immune activity is detected prior to seroconversion in children with HLA-conferred type 1 diabetes susceptibility. Diabetes, 2014. 63(7): p. 2402-14.

34. Oresic, M., et al., Dysregulation of lipid and amino acid metabolism precedes islet autoimmunity in children who later progress to type 1 diabetes. J Exp Med, 2008. 205(13): p. 2975-84.

35. Tessem, J.S., et al., Critical roles for macrophages in islet angiogenesis and maintenance during pancreatic degeneration. Diabetes, 2008. 57(6): p. 1605-17.

36. Kraal, G., G. Martens, and K. Kuystermans, Systemic elimination of macrophages using liposomes does not prevent the induction of type I diabetes. Int Arch Allergy Immunol, 1993. 100(2): p. 115-20.

37. Kriegel, M.A., C. Rathinam, and R.A. Flavell, Pancreatic islet expression of chemokine CCL2 suppresses autoimmune diabetes via tolerogenic CD11c+ CD11b+ dendritic cells. Proc Natl Acad Sci U S A, 2012. 109(9): p. 3457-62.

38. Bernard, A., Lamy, and I. Alberti, The two-signal model of T-cell activation after 30 years. Transplantation, 2002. 73(1 Suppl): p. S31-5.

39. Viorritto, I.C., N.P. Nikolov, and R.M. Siegel, Autoimmunity versus tolerance: can dying cells tip the balance? Clin Immunol, 2007. 122(2): p. 125-34.

40. Schwartz, R.H., Models of T cell anergy: is there a common molecular mechanism? J Exp Med, 1996. 184(1): p. 1-8.

41. Miyake, Y. and S. Yamasaki, Sensing necrotic cells. Adv Exp Med Biol, 2012. 738: p. 144-52.

42. Seong, S.Y. and P. Matzinger, Hydrophobicity: an ancient damage-associated molecular pattern that initiates innate immune responses. Nat Rev Immunol, 2004. 4(6): p. 469-78.

43. Szondy, Z., et al., Anti-inflammatory Mechanisms Triggered by Apoptotic Cells during Their Clearance. Front Immunol, 2017. 8: p. 909.

44. Kim, H.S., et al., Toll-like receptor 2 senses beta-cell death and contributes to the initiation of autoimmune diabetes. Immunity, 2007. 27(2): p. 321-33.

45. Sachet, M., Y.Y. Liang, and R. Oehler, The immune response to secondary necrotic cells. Apoptosis, 2017. 22(10): p. 1189-1204.

46. Paredes-Juarez, G.A., et al., DAMP production by human islets under low oxygen and nutrients in the presence or absence of an immunoisolating-capsule and necrostatin-1. Sci Rep, 2015. 5: p. 14623.

47. Fehsel, K., V. Kolb-Bachofen, and K.D. Kroncke, Necrosis is the predominant type of islet cell death during development of insulin-dependent diabetes mellitus in BB rats. Lab Invest, 2003. 83(4): p. 549-59.

48. Li, M., et al., Toll-like receptor 4 on islet beta cells senses expression changes in high-mobility group box 1 and contributes to the initiation of type 1 diabetes. Exp Mol Med, 2012. 44(4): p. 260-7.

49. Okorokov, P.L., et al., [Intestinal endotoxin in induction of type 1 diabetes]. Fiziol Cheloveka, 2011. 37(2): p. 138-41.

50. Cvetanovic, M. and D.S. Ucker, Innate immune discrimination of apoptotic cells: repression of proinflammatory macrophage transcription is coupled directly to specific recognition. J Immunol, 2004. 172(2): p. 880-9.

51. Voll, R.E., et al., Immunosuppressive effects of apoptotic cells. Nature, 1997. 390(6658): p. 350-1.

52. Fadok, V.A., et al., Macrophages that have ingested apoptotic cells in vitro inhibit proinflammatory cytokine production through autocrine/paracrine mechanisms involving TGF-beta, PGE2, and PAF. J Clin Invest, 1998. 101(4): p. 890-8.

53. Cocco, R.E. and D.S. Ucker, Distinct modes of macrophage recognition for apoptotic and necrotic cells are not specified exclusively by phosphatidylserine exposure. Mol Biol Cell, 2001. 12(4): p. 919-30.

54. Brosky, G. and J. Logothetopoulos, Streptozotocin diabetes in the mouse and guinea pig. Diabetes, 1969. 18(9): p. 606-11.

55. Saini, K.S., et al., Streptozotocin at low doses induces apoptosis and at high doses causes necrosis in a murine pancreatic beta cell line, INS-1. Biochem Mol Biol Int, 1996. 39(6): p. 1229-36.

56. Eizirik, D.L., M.L. Colli, and F. Ortis, The role of inflammation in insulitis and beta-cell loss in type 1 diabetes. Nat Rev Endocrinol, 2009. 5(4): p. 219-26.

57. Bonnevie-Nielsen, V., M.W. Steffes, and A. Lernmark, A major loss in islet mass and B-cell function precedes hyperglycemia in mice given multiple low doses of streptozotocin. Diabetes, 1981. 30(5): p. 424-9.

58. Wei, L., et al., Induction of diabetes with signs of autoimmunity in primates by the injection of multiple-low-dose streptozotocin. Biochem Biophys Res Commun, 2011. 412(2): p. 373-8.

59. Lee, J.H., et al., Pharmacokinetics of drugs in rats with diabetes mellitus induced by alloxan or streptozocin: comparison with those in patients with type I diabetes mellitus. J Pharm Pharmacol, 2010. 62(1): p. 1-23.

60. Leiter, E.H., Multiple low-dose streptozotocin-induced hyperglycemia and insulitis in C57BL mice: influence of inbred background, sex, and thymus. Proc Natl Acad Sci U S A, 1982. 79(2): p. 630-4.

61. Alam, M.U., et al., Transgenic expression of Hsc70 in pancreatic islets enhances autoimmune diabetes in response to beta cell damage. J Immunol, 2009. 183(9): p. 5728-37.

62. Broggi, A. and F. Granucci, Microbe- and danger-induced inflammation. Mol Immunol, 2015. 63(2): p. 127-33.

63. Janko, C., et al., Redox modulation of HMGB1-related signaling. Antioxid Redox Signal, 2014. 20(7): p. 1075-85.

64. Urbonaviciute, V., et al., Induction of inflammatory and immune responses by HMGB1-nucleosome complexes: implications for the pathogenesis of SLE. J Exp Med, 2008. 205(13): p. 3007-18.

65. Aumeunier, A., et al., Systemic Toll-like receptor stimulation suppresses experimental allergic asthma and autoimmune diabetes in NOD mice. PLoS One, 2010. 5(7): p. e11484.

66. Maree, A.F., et al., A quantitative comparison of rates of phagocytosis and digestion of apoptotic cells by macrophages from normal (BALB/c) and diabetes-prone (NOD) mice. J Appl Physiol (1985), 2008. 104(1): p. 157-69.

67. Munoz, L.E., et al., Do low vitamin D levels cause problems of waste removal in patients with SLE? Rheumatology (Oxford), 2012. 51(4): p. 585-7.

68. Danke, N.A., et al., Autoreactive T cells in healthy individuals. J Immunol, 2004. 172(10): p. 5967-72.

69. Monti, P., et al., Evidence for in vivo primed and expanded autoreactive T cells as a specific feature of patients with type 1 diabetes. J Immunol, 2007. 179(9): p. 5785-92.

70. Goverman, J.M., Immune tolerance in multiple sclerosis. Immunol Rev, 2011. 241(1): p. 228-40.

71. Bouneaud, C., P. Kourilsky, and P. Bousso, Impact of negative selection on the T cell repertoire reactive to a self-peptide: a large fraction of T cell clones escapes clonal deletion. Immunity, 2000. 13(6): p. 829-40.

72. Enouz, S., et al., Autoreactive T cells bypass negative selection and respond to self-antigen stimulation during infection. J Exp Med, 2012. 209(10): p. 1769-79.

73. Gavin, M.A. and A.Y. Rudensky, Dual TCR T cells: gaining entry into the periphery. Nat Immunol, 2002. 3(2): p. 109-10.

74. Xing, Y. and K.A. Hogquist, T-cell tolerance: central and peripheral. Cold Spring Harb Perspect Biol, 2012. 4(6).

75. Foulis, A.K., et al., The histopathology of the pancreas in type 1 (insulin-dependent) diabetes mellitus: a 25-year review of deaths in patients under 20 years of age in the United Kingdom. Diabetologia, 1986. 29(5): p. 267-74.

76. In't Veld, P., Insulitis in human type 1 diabetes: The quest for an elusive lesion. Islets, 2011. 3(4): p. 131-8.

77. Campbell-Thompson, M., et al., Insulitis and beta-Cell Mass in the Natural History of Type 1 Diabetes. Diabetes, 2016. 65(3): p. 719-31.

78. Wicker, L.S., B.J. Miller, and Y. Mullen, Transfer of autoimmune diabetes mellitus with splenocytes from nonobese diabetic (NOD) mice. Diabetes, 1986. 35(8): p. 855-60.

79. Ramanathan, S., et al., Thymectomy and radiation-induced type 1 diabetes in nonlymphopenic BB rats. Diabetes, 2002. 51(10): p. 2975-81.

80. Zalutskaya, A., et al., Did the Chernobyl incident cause an increase in Type 1 diabetes mellitus incidence in children and adolescents? Diabetologia, 2004. 47(1): p. 147-8.

81. Svendsen, E.R., et al., 137Cesium exposure and spirometry measures in Ukrainian children affected by the Chernobyl nuclear incident. Environ Health Perspect, 2010. 118(5): p. 720-5.

82. Koevary, S.B., et al., Passive transfer of diabetes from BB/W to Wistar-Furth rats. J Clin Invest, 1985. 75(6): p. 1904-7.

83. Harada, M. and S. Makino, Promotion of spontaneous diabetes in non-obese diabetes-prone mice by cyclophosphamide. Diabetologia, 1984. 27(6): p. 604-6.

84. Haskins, K. and M. McDuffie, Acceleration of diabetes in young NOD mice with a CD4+ islet-specific T cell clone. Science, 1990. 249(4975): p. 1433-6.

85. Ramirez, L. and A.R. Hamad, Status of autoimmune diabetes 20-year after generation of BDC2.5-TCR transgenic non-obese diabetic mouse. World J Diabetes, 2013. 4(4): p. 88-91.

86. Gonzalez, A., et al., Damage control, rather than unresponsiveness, effected by protective DX5+ T cells in autoimmune diabetes. Nat Immunol, 2001. 2(12): p. 1117-25.

87. Cui, Y.F., et al., Apoptosis of circulating lymphocytes induced by whole body gamma-irradiation and its mechanism. J Environ Pathol Toxicol Oncol, 1999. 18(3): p. 185-9.

88. Hurd, E.R. and V.J. Giuliano, The effect of cyclophosphamide on B and T lymphocytes in patients with connective tissue diseases. Arthritis Rheum, 1975. 18(1): p. 67-75.

89. Bougneres, P.F., et al., Limited duration of remission of insulin dependency in children with recent overt type I diabetes treated with low-dose cyclosporin. Diabetes, 1990. 39(10): p. 1264-72.

90. Schreiber, S.L. and G.R. Crabtree, The mechanism of action of cyclosporin A and FK506. Immunol Today, 1992. 13(4): p. 136-42.

91. Boulougouris, G., et al., IL-2-independent activation and proliferation in human T cells induced by CD28. J Immunol, 1999. 163(4): p. 1809-16.

92. Rafiq, K., et al., Cyclosporin A increases IFN-gamma production by T cells when co-stimulated through CD28. Eur J Immunol, 1998. 28(5): p. 1481-91.

93. Wittmann, M., et al., Critical involvement of IL-12 in IFN-gamma induction by calcineurin antagonists in activated human lymphocytes. J Leukoc Biol, 2006. 80(1): p. 75-86.

94. Jindal, R.M., et al., Diabetogenicity of FK506 versus cyclosporine in liver transplant recipients. Transplantation, 1994. 58(3): p. 370-2.

95. Herold, K.C., et al., Teplizumab (anti-CD3 mAb) treatment preserves C-peptide responses in patients with new-onset type 1 diabetes in a randomized controlled trial: metabolic and immunologic features at baseline identify a subgroup of responders. Diabetes, 2013. 62(11): p. 3766-74.

96. Meijer, R.T., et al., Non FcR-binding murine antihuman CD3 monoclonal antibody is capable of productive TCR signalling and induces proliferation in the presence of costimulation. Clin Exp Immunol, 2001. 123(3): p. 511-9.

97. You, S., et al., Unique role of CD4+CD62L+ regulatory T cells in the control of autoimmune diabetes in T cell receptor transgenic mice. Proc Natl Acad Sci U S A, 2004. 101 Suppl 2: p. 14580-5.

98. Salomon, B., et al., B7/CD28 costimulation is essential for the homeostasis of the CD4+CD25+ immunoregulatory T cells that control autoimmune diabetes. Immunity, 2000. 12(4): p. 431-40.

99. Doenhoff, M.J., et al., Adult and pre-adult thymectomy of mice: contrasting effects on immune responsiveness, and on numbers of mitogen-responsive and Thy-1+ lymphocytes. Immunology, 1979. 37(2): p. 397-405.

100. Dardenne, M., et al., Acceleration of the onset of diabetes in NOD mice by thymectomy at weaning. Eur J Immunol, 1989. 19(5): p. 889-95.

101. Walker, L.S. and M. von Herrath, CD4 T cell differentiation in type 1 diabetes. Clin Exp Immunol, 2016. 183(1): p. 16-29.

102. Paludan, S.R., Interleukin-4 and interferon-gamma: the quintessence of a mutual antagonistic relationship. Scand J Immunol, 1998. 48(5): p. 459-68.

103. Murata, Y., M. Amao, and J. Hamuro, Sequential conversion of the redox status of macrophages dictates the pathological progression of autoimmune diabetes. Eur J Immunol, 2003. 33(4): p. 1001-11.

104. Azar, S.T., et al., Type I (insulin-dependent) diabetes is a Th1- and Th2-mediated autoimmune disease. Clin Diagn Lab Immunol, 1999. 6(3): p. 306-10.

105. Milicic, T., et al., High Risk First Degree Relatives of Type 1 Diabetics: An Association with Increases in CXCR3(+) T Memory Cells Reflecting an Enhanced Activity of Th1 Autoimmune Response. Int J Endocrinol, 2014. 2014: p. 589360.

106. Ryden, A., et al., Switch from a dominant Th1-associated immune profile during the pre-diabetic phase in favour of a temporary increase of a Th3-associated and inflammatory immune profile at the onset of type 1 diabetes. Diabetes Metab Res Rev, 2009. 25(4): p. 335-43.

107. Vaseghi, H. and Z. Jadali, Th1/Th2 cytokines in Type 1 diabetes: Relation to duration of disease and gender. Indian J Endocrinol Metab, 2016. 20(3): p. 312-6.

108. Rothermel, A.L., K.M. Gilbert, and W.O. Weigle, Differential abilities of Th1 and Th2 to induce polyclonal B cell proliferation. Cell Immunol, 1991. 135(1): p. 1-15.

109. Murata, Y., T. Shimamura, and J. Hamuro, The polarization of T(h)1/T(h)2 balance is dependent on the intracellular thiol redox status of macrophages due to the distinctive cytokine production. Int Immunol, 2002. 14(2): p. 201-12.

110. Ullevig, S., H.S. Kim, and R. Asmis, S-glutathionylation in monocyte and macrophage (dys)function. Int J Mol Sci, 2013. 14(8): p. 15212-32.

111. Dominguez, C., et al., Oxidative stress at onset and in early stages of type 1 diabetes in children and adolescents. Diabetes Care, 1998. 21(10): p. 1736-42.

112. Burrack, A.L., T. Martinov, and B.T. Fife, T Cell-Mediated Beta Cell Destruction: Autoimmunity and Alloimmunity in the Context of Type 1 Diabetes. Front Endocrinol (Lausanne), 2017. 8: p. 343.

113. Fernandez-Boyanapalli, R., et al., Impaired phagocytosis of apoptotic cells by macrophages in chronic granulomatous disease is reversed by IFN-gamma in a nitric oxide-dependent manner. J Immunol, 2010. 185(7): p. 4030-41.

114. Valdez, I.A., et al., Proinflammatory Cytokines Induce Endocrine Differentiation in Pancreatic Ductal Cells via STAT3-Dependent NGN3 Activation. Cell Rep, 2016. 15(3): p. 460-470.

115. Gu, D., M. Arnush, and N. Sarvetnick, Endocrine/exocrine intermediate cells in streptozotocin-treated Ins-IFN-gamma transgenic mice. Pancreas, 1997. 15(3): p. 246-50.

116. Sarvetnick, N.E. and D. Gu, Regeneration of pancreatic endocrine cells in interferon-gamma transgenic mice. Adv Exp Med Biol, 1992. 321: p. 85-9; discussion 91-3.

117. Serreze, D.V., et al., Th1 to Th2 cytokine shifts in nonobese diabetic mice: sometimes an outcome, rather than the cause, of diabetes resistance elicited by immunostimulation. J Immunol, 2001. 166(2): p. 1352-9.

118. Gu, D., et al., Transgenic mice expressing IFN-gamma in pancreatic beta-cells are resistant to streptozotocin-induced diabetes. Am J Physiol, 1995. 269(6 Pt 1): p. E1089-94.

119. Lindley, S., et al., Defective suppressor function in CD4(+)CD25(+) T-cells from patients with type 1 diabetes. Diabetes, 2005. 54(1): p. 92-9.

120. Pasare, C. and R. Medzhitov, Toll pathway-dependent blockade of CD4+CD25+ T cell-mediated suppression by dendritic cells. Science, 2003. 299(5609): p. 1033-6.

121. Yang, Y., et al., Persistent Toll-like receptor signals are required for reversal of regulatory T cell-mediated CD8 tolerance. Nat Immunol, 2004. 5(5): p. 508-15.

122. Peng, G., et al., Toll-like receptor 8-mediated reversal of CD4+ regulatory T cell function. Science, 2005. 309(5739): p. 1380-4.

123. Schenten, D., et al., Signaling through the adaptor molecule MyD88 in CD4+ T cells is required to overcome suppression by regulatory T cells. Immunity, 2014. 40(1): p. 78-90.

124. Sreenan, S., et al., Increased beta-cell proliferation and reduced mass before diabetes onset in the nonobese diabetic mouse. Diabetes, 1999. 48(5): p. 989-96.

125. Halminen, M., et al., Cytokine expression in unstimulated PBMC of children with type 1 diabetes and subjects positive for diabetes-associated autoantibodies. Scand J Immunol, 2001. 53(5): p. 510-3.

126. Butler, A.E., et al., Modestly increased beta cell apoptosis but no increased beta cell replication in recent-onset type 1 diabetic patients who died of diabetic ketoacidosis. Diabetologia, 2007. 50(11): p. 2323-31.

127. In't Veld, P., et al., Screening for insulitis in adult autoantibody-positive organ donors. Diabetes, 2007. 56(9): p. 2400-4.

128. Hauben, E. and M. Schwartz, Therapeutic vaccination for spinal cord injury: helping the body to cure itself. Trends Pharmacol Sci, 2003. 24(1): p. 7-12.

129. Broxmeyer, H.E., et al., Th1 cells regulate hematopoietic progenitor cell homeostasis by production of oncostatin M. Immunity, 2002. 16(6): p. 815-25.

130. Jung, I.D., et al., Oncostatin M induces dendritic cell maturation and Th1 polarization. Biochem Biophys Res Commun, 2010. 394(2): p. 272-8.

131. Kubin, T., et al., Oncostatin M is a major mediator of cardiomyocyte dedifferentiation and remodeling. Cell Stem Cell, 2011. 9(5): p. 420-32.

132. Morrisey, E.E., Rewind to recover: dedifferentiation after cardiac injury. Cell Stem Cell, 2011. 9(5): p. 387-8.

133. Baldeon, M.E., T. Chun, and H.R. Gaskins, Interferon-alpha and interferon-gamma differentially affect pancreatic beta-cell phenotype and function. Am J Physiol, 1998. 275(1): p. C25-32.

134. Ouziel-Yahalom, L., et al., Expansion and redifferentiation of adult human pancreatic islet cells. Biochem Biophys Res Commun, 2006. 341(2): p. 291-8.

135. Ortega-Gomez, A., M. Perretti, and O. Soehnlein, Resolution of inflammation: an integrated view. EMBO Mol Med, 2013. 5(5): p. 661-74.

136. Meier, J.J., et al., Sustained beta cell apoptosis in patients with long-standing type 1 diabetes: indirect evidence for islet regeneration? Diabetologia, 2005. 48(11): p. 2221-8.

137. Oram, R.A., et al., Most people with long-duration type 1 diabetes in a large population-based study are insulin microsecretors. Diabetes Care, 2015. 38(2): p. 323-8.

138. Couri, C.E.B., K.C.R. Malmegrim, and M.C. Oliveira, New Horizons in the Treatment of Type 1 Diabetes: More Intense Immunosuppression and Beta Cell Replacement. Front Immunol, 2018. 9: p. 1086.

139. Faustman, D.L., et al., Proof-of-concept, randomized, controlled clinical trial of Bacillus-Calmette-Guerin for treatment of long-term type 1 diabetes. PLoS One, 2012. 7(8): p. e41756.

140. Kleinnijenhuis, J., et al., Long-lasting effects of BCG vaccination on both heterologous Th1/Th17 responses and innate trained immunity. J Innate Immun, 2014. 6(2): p. 152-8.

141. Shehadeh, N., et al., Effect of adjuvant therapy on development of diabetes in mouse and man. Lancet, 1994. 343(8899): p. 706-7.

142. Tsai, S., et al., Insulin Receptor-Mediated Stimulation Boosts T Cell Immunity during Inflammation and Infection. Cell Metab, 2018. 28(6): p. 922-934 e4.

143. Friedman-Mazursky, O., R. Elkon, and S. Efrat, Redifferentiation of expanded human islet beta cells by inhibition of ARX. Sci Rep, 2016. 6: p. 20698.