Peptide Immunotherapies in Type 1 Diabetes

7 downloads 0 Views 410KB Size Report
prevent or treat T1D by injection of either the self proteins or peptides derived from them. Peptide .... breakthroughs in progress toward a cure for diabetes, in ad-.

Current Medicinal Chemistry, 2011, 18, ????-????

1

Peptide Immunotherapies in Type 1 Diabetes: Lessons from Animal Models A. Fierabracci* Department of Immunology, Ospedale Pediatrico Bambino Gesu’ Research Institute, Rome, Italy Abstract: Insulin dependent diabetes mellitus (Type 1 diabetes, T1D) is a chronic autoimmune disorder characterized by the destruction of insulin-producing pancreatic beta cells by proinflammatory autoreactive T cells. In the past, several therapeutic approaches have been exploited by immunologists aiming to regulate the autoimmune response; this can occur by deleting lymphocyte subsets and/or re-establishing immune tolerance via activation of regulatory T cells. The use of broad immunosuppressive drugs was the first approach to be explored. Subsequently, antibody-based immunotherapies failed to discriminate between autoreactive versus non-autoimmune effectors. Antigen-based immunotherapy is a third approach developed to manipulate beta cell autoimmunity. This approach allows the selective targeting of disease-relevant T cells, while leaving the remainder of the immune system intact. Animal models have been successfully employed to prevent or treat T1D by injection of either the self proteins or peptides derived from them. Peptide immunotherapies have been mainly experimented in the NOD mouse spontaneous model of disease. In this review we therefore report the main approaches that rely on the use of peptides obtained from relevant autoantigens such as glutamic acid decarboxylase, isoform 65 (GAD65), insulin, proinsulin and islet-specific glucose 6 phosphatase catalytic subunit-related protein (IGRP). Protective peptides have proven to be effective in treating or delaying the diabetic process. We also highlight the main difficulties encountered in extrapolating data to guide clinical translational investigations in humans.

Keywords: ????????????????????????. INTRODUCTION T1D is an autoimmune disease which occurs in HLA genetically predisposed individuals [1] as a consequence of the organ-specific immune destruction of the insulin-producing  cells in the islets of Langherans within the pancreas. The disorder derives from a combination of genetic predisposition, still-unknown environmental factors and stochastic events [2-4]. From a pathogenetic point of view the disease is the result of a breakdown in immune regulation that leads to the expansion of autoreactive CD4+ and CD8+ T cells, autoantibody-producing B lymphocytes and the activation of the innate immune system. For many years immunologists aimed to develop methods for predicting, preventing and/or treating T1D by identifying those self antigens that are the target of the autoimmune process, resulting in tissue damage and clinical symptomatology. To this end, islet-related autoantibodies were identified that proved to be good predictors for future onset of the disease; nevertheless these are not directly pathogenetic, while T cells play a dominant role in disease initiation and progression. Major autoantigens in T1D include insulin, glutamic acid decarboxylase (GAD), isoforms 65 and 67 (GAD65, GAD67), the insulinoma-associated antigen (IA2)/ tyrosine phosphatase-like molecule (IA-2), IA-2  or phogrin and proinsulin [5]. Notably, immunologists also sought to exploit immunotherapeutic strategies. These can be applied in two distinct stages of the disease: the prediabetic period and the overt disease [6, 7]. When used in the prediabetic period the treatment is aimed at suppressing the diabetogenic process and preventing the clinical onset of disease in ‘high risk’ prediabetic individuals. These subjects, in which  cell autoimmunity is ongoing, can be monitored by detection of serum autoantibodies and metabolic markers. Indeed, the onset of *Address correspondence to this author at the Ospedale Pediatrico Bambino Gesu’, Research Institute, Piazza S. Onofrio 4, 00165 Rome, Italy; Tel: +39 06 6859 2656; Fax: +39 06 6859 2904; E-mail: [email protected] 0929-8673/11 $58.00+.00

symptoms, including hyperglycemia may occur even several years after the initiating event of the autoimmune process. When immunotherapies are applied after the disease onset, these are aimed at preserving the residual  cells thereby controlling blood glucose and possibly ensuring  cell regeneration. Blocking autoimmunity may also in some instances lead to disease remission. Cyclosporine A was one of the first immunosuppressive drugs to be used; however, this requires continuous administration and can produce problematic adverse effects. A second approach employed antibodies such as anti-CD3 antibodies to deplete or modify the function of effector cells such as T and B cells and antigenpresenting cells, antagonizing effector molecules, cytokines and co-stimulatory molecules. Antibody-based immunotherapy, however, has the limitation of not being able to specifically discriminate autoreactive versus non-autoimmune effectors. Antigen-based immunotherapy has also been exploited to halt beta-cell autoimmunity. The use of such ‘vaccines’ for the prevention and treatment of T1D has been developed for many years to allow the selective targeting of disease-specific autoreactive T cells, without affecting the remainder of the immune system. Antigen-specific immunotherapy can achieve specific T cell ablation (by activationinduced cell-death,) and T cell anergy [8]. This usually requires the injection of high doses of soluble antigen; it has proven effective when the range of pathogenic effector T cells is limited and the specificity of the dominant clonotypes is defined. The other mechanism of action is the induction/expansion of immunoregulatory T cells, primarily CD4+ CD25+ regulatory T cells (nTregs) [9]; these originate from the thymus, constitutively express the transcription factor FOXP3 and suppress immune responses via cell-to-cell contact. Expansion of immunoregulatory T cells is the preferred mechanism when multiple unknown autoantigens are targeted, as in the case of T1D [6]. Antigen-specific peptide immunotherapy can also cause a shift in the predominant phenotype from a T helper (Th1) to a Th2 phenotype. The treatment can be proposed with the protein self-antigen target; alternatively, peptide epitopes of the self antigen can be administered. © 2011 Bentham Science Publishers Ltd.

2

Current Medicinal Chemistry, 2011 Vol. 18, No. 1

Apart from epidemiological studies in human populations, limited series of experiments can be carried out in human subjects at risk of developing diabetes. Since the initial studies by Claude Bernard and the pancreatectomy experiments on dogs performed by von Mering and Minkowski [10] and Banting’s discovery of insulin [11], several animal models of disease have been implemented, resulting in major breakthroughs in progress toward a cure for diabetes, in addition to the discovery of molecular pathways, genes and environmental factors involved in the disease etiopathogenesis. In this review we particularly emphasize the main discoveries achieved through the use of animal models in the field of current peptide-based immunotherapies in T1D. It should be pointed out that special attention is required when trying to extrapolate data obtained from animal models, since they always give an incomplete portrait of the more complex human situation, given the fact that significant differences distinguish human and spontaneous from experimentallyinduced disease [12]. Nevertheless, results obtained from animal studies have always opened the pathway to relevant translational investigations in humans. MOST FREQUENTLY USED MODELS OF TYPE 1 DIABETES An initial consideration needs to be addressed when trying to choose the best animal model to answer specific questions in T1D. As for the more complex human model, the distinction between Type 1 and Type 2 (non-insulin- dependent) diabetes is not always clear, because clinical features may not be completely separated and similarities in presentation can occur [13]. Knowledge regarding the pathophysiology of T1D has mainly been acquired through the study of animal models where the disease develops ‘spontaneously’ or is induced by either environmental insult or genetic manipulation (transgene and knockout animals). The most interesting and useful data were obtained by the analysis of the NOD (non obese diabetic) mouse and the BB (Bio breeding) rat. The NOD mouse model was discovered in Osaka, Japan in 1974 [12]. In addition to diabetes, the mouse also has sialoadenitis and is deaf; in the majority of strains, association with thyroid infiltration seldom occurs [12]. As has been extensively reviewed elsewhere [10], the clinical syndrome is similar to that of human T1D, including glycosuria, hyperglycemia and hypo-insulinemia; insulin-dependence starts at about 14 weeks of age. Almost all diabetes-prone animals show pancreatic insulitis at histological examination from about 5 weeks of age, but only in 80% of females and 20% of males. Cellular immunological abnormalities are detected in NOD mice; these include antibody-dependent cell-mediated cytotoxicity and a decreased number of T lymphocytes and NK cells. Serum autoantibodies are detectable in the prediabetic period. These include islet cell antibodies (ICA), islet cell surface antibodies (ICSA) and insulin autoantibodies (IAA). Insulitis shows the presence of helper, cytotoxic T cells and NK cells, together with other immunological cell players. The genetics of T1D in NOD mice has greatly helped to clarify some aspects of the genetics of the disease in humans [12]. Unlike the human disease and the BB-DP rat syndrome,

A. Fierabracci

the animals can survive without insulin administration and ketoacidosis rarely occurs. Typically female mice are more severely affected. In the NOD mouse over 15 genetic loci located on different chromosomes contribute to diabetes and/or insulitis and/or sialoadenitis. Linked to the major histocompatibility complex (MHC) are the most important diabetic loci (idd1 loci) related to the diabetic syndrome. A unique I-A locus is expressed, i.e., I-Ag7 (histidine as residue 56 and serine as residue 57). This is homologous to the human allele containing ‘diabetogenic’ HLA-DQ Beta nonaspartic acid 57; there is no expression of I-Ea (homologous to DR alpha in humans). Idd1 is not related to sialoadenitis development; the putative contribution of idd3 and idd5 loci is not supported by sufficient experimental data [12]. The BB-DP rat was discovered in a commercial breeding company in Canada in 1974 [12]. This animal develops a spontaneous T cell dependent, ketosis prone, diabetic syndrome at 8-12 weeks of age. In the course of the disease, ketoacidosis may develop and administration of exogenous insulin is fundamental. Both sexes are equally affected. The mouse has profound lymphopenia [12]. Insulitis is more similar to the human T1D, with the presence of T helper and cytotoxic T lymphocytes, B lymphocytes, macrophages, and NK cells. Humoral immunity also underlies the development of the disorder. Autoantibodies, except ICA, were reported in the BB rat, and also GAD-reactive T cells. The disease susceptibility is associated with MHC, as it is for humans and the NOD mouse. The RT1u MHC haplotype is necessary for induction of the disease; the contribution of one MHC and 2 other non-MHC linked genes is required [10]. CONTRIBUTION OF THE NOD MOUSE MODEL TO DISCOVERING PEPTIDE IMMUNOTHERAPIES IN T1D As stated above, strategies based on the administration of immunosuppressant drugs and antibodies specific to T cells have both been used in experimental animal models and, in some instances, in the clinical setting (vide infra). However, these approaches fail to discriminate between T cells specific for self- and foreign antigens and variably compromise the normal functioning of the immune system [6]. Several methods are being developed and are also currently being assessed in T1D trials, aiming to target multiple diabetogenic antigens. These strategies are supposed to act via cell intrinsic and/or extrinsic mechanisms [7]. Peptide-based immunotherapies have been explored as an alternative approach to selectively targeting autoreactive T cells, without affecting the remainder of the immune system. At the moment there are limitations to the use of antigenspecific and antigenic peptide immunotherapies in the clinical setting, because of limited knowledge regarding beta cell autoantigens that are recognized by pathogenic and immunoregulatory T cells in the human diabetogenic response. Therefore, the search to identify the autoantigens that are most relevant in the disease process must absolutely continue [1, 6]. Most of our knowledge regarding peptide immunotherapies has been obtained from analysis of the NOD mouse spontaneous model of disease. In the sections that follow we

Peptide Vaccination to Treat Type 1 Diabetes

report the most significant recent approaches that have been implemented in the last decade based on the use of peptides from relevant autoantigens: GAD65, insulin and proinsulin, and islet-specific glucose 6 phosphatase catalytic subunitrelated protein (IGRP) (Fig. 1). GAD65 Peptide Immunotherapies There is a general consensus, derived from both human epidemiological studies and animal models of disease that GAD65 autoantigen is an early target of autoantibodies in the initiating events of T1D pathogenesis [14, 15]. Interestingly, it has been demonstrated that the initial antigenic motif is limited to the C terminus of GAD protein and consists of only a few epitopes. Later in the disease process, intramolecular spreading occurs and subsequently other autoantigens are involved [6]. Tisch et al. [14] first reported experimental evidence for the critical role of GAD in the diabetogenic process, demonstrating early cellular and antibody responses to this antigen. T cell responses in unimmunised NOD mice were evaluated against a panel of  cell autoantigens, including GAD (isoforms 65 and 67), peripherin, carboxypeptidase H and heat shock protein 60 (HSP60). Previously, only GAD65 and GAD67 responses were detected in splenic cell cultures obtained from 4-weekold females. In cultures from 6-week-old females, T cell responses for the other antigens were also reported. Responses to the entire antigenic panel were detected by 8 weeks of age. Diabetic and 24-week-old non diabetic NOD females also showed responses to the antigenic panel, while no response was detected at any age to other  cell antigens such as amylin, thus indicating that not all  cell proteins

Current Medicinal Chemistry, 2011 Vol. 18, No. 1

3

become target of the autoimmune response. By 4 weeks of age NOD mice showed a significant antibody response to GAD65 and GAD67. In 6-week-old females, antibodies specific for carboxypeptidase H and peripherin were also demonstrated. Antigen tolerance strategies were also investigated by intravenous or intrathymic injections of GAD molecule in 3-week-old NOD females that eliminate anti-GAD T cell responses resulting in the animals remaining diabetes-free [14]. In initial attempts with peptide immunotherapy, NOD female mice were treated at two time intervals: at 4 weeks of age, when beta cell-specific T cell reactivity is first detected, and at 12 weeks of age, a late preclinical stage with evident insulitis. GAD65-specific peptides GAD65 217-236 and GAD65 290-309 (50μg each) were administered intraperitonally in incomplete Freund’s adjuvant (IFA) [6, 16, 17]. The majority of 4-week-old female mice remained diabetes-free, but the disease continued to develop when ovoalbumin peptide OVA 321-339 was administered. In mice that did not develop diabetes, there was a correlation with reduced development of insulitis and induction of IL 4 secreting Th2 cells in cultures prepared from pancreatic draining lymph nodes (PLN). In parallel, there was also a reduction in the number of GAD65 and InsB peptide 9-23-specific IFN- secreting CD4+ T cells. The results obtained from 12-weekold mice were different; development of diabetes was halted only in cases where a mixture of the 2 GAD65 peptides was administered, producing a sufficient frequency of immunoregulatory T effector cells in PNL. It has already been shown that single immunization with the GAD65 p500-585 in young NOD mice expanded the

Fig. (1). Overview of recent immunotherapies with relevant autoantigenic peptides in NOD mice in different stages of disease.

4

Current Medicinal Chemistry, 2011 Vol. 18, No. 1

CD4+ CD25+ Foxp3+ Tregs in the peripheral blood; as a result, the pancreatic insulitis process was halted and the onset of diabetes clinical symptoms was effectively prevented. Furthermore, dendritic cells (DCs) bearing -cell autoantigenic peptides have been shown to have the ability to expand Tregs and consequently suppress diabetes development [18]. In the light of this knowledge, Chen et al. [19] recently aimed to investigate whether known autoantigenic epitopes of GAD65 500-585 p509 (GAD 509-528), p524 (GAD 524-538) and p530 (GAD 530-543) could expand suppressive Tregs. Consequently, splenic lymphocytes from NOD mice were exposed to stimulation with the 3 different GAD65-derived epitopes for 7-10 days. Following peptide stimulation the frequency and function of Tregs were analyzed. CD4+ CD25+ Foxp3+ T cells expanded in vitro after challenge with the above peptide epitopes, as compared to controls. In vitro, all expanded Tregs were able to suppress CD4+ T effector cells from NOD mice. Authors further evaluated the ability of the in vitro epitope-expanded Tregs to halt diabetes whenever co-transfected with diabetogenic T cells into 4-6 week-old NOD.SCID mice. Interestingly, although all expanded Tregs showed suppressive function in vitro, only p524 (GAD65 524-538)-expanded CD4+ CD25+ T cells inhibited the diabetes process in the co-transfer model. On the contrary, p509 (GAD 509-528) or p530 (GAD 530-543) expanded CD4+CD25+ T cells did not produce a protective effect. Computer-guided molecular modeling and docking methods were also employed to estimate whether differences in the Treg-expanding function of the examined epitopes could be related to their specific structural characteristics and their binding ability to MHC class II I Ag7. Epitope p524, which induced protective Tregs, possessed negative surface-electrostatic potential and bound two chains of MHC class II I Ag7 (chains A and B), while the other 2 epitopes had positive surface-electrostatic potential and bound one chain of I Ag7 (chain B). In addition, binding of p524 was more stable. It was also experimentally proven that epitope GAD570-585, p570, having similarities to p524 in sequence, electrostatic potential and interaction ability with I Ag7, was protective; indeed p570-expanded CD4+ CD25+ T cells halted the onset of diabetes symptoms in NOD mice. The efficacy of mature myeloid DC pulsed with autoantigenic  cell epitopes to elicit tolerance in NOD mice has been known for several years [20]. Using this information, Lo et al. [21] demonstrated that down-regulation of the diabetic process can also be induced with the use of antigenpulsed immature dendritic cells (iDC). DCs were isolated from bone marrow, grown and pulsed with 3μg of peptides. These included dominant insulin 9-23 chain (InsB9-23), proinsulin C19-A3 or ignored GAD65 79-97 peptides. NOD mice received footpath administrations of either unpulsed or peptide-pulsed myeloid iDC beginning at 9 weeks of age for 3 consecutive weeks. Diabetes incidence in mice treated with unpulsed iDC was comparable to unmanipulated animals (~80%) while GAD65 78-97 pulsed iDC treated animals were protected from the disease. No significant protection was obtained in mice receiving inulin and proinsulin-pulsed DCs. Proliferation of splenic T cells after challenge with a panel of peptides was also evaluated in diabetic and non diabetic mice. Non- diabetic mice receiving insulin or proinsulin peptide-pulsed iDCs showed a significant reduction in T

A. Fierabracci

cell response as compared to diabetic unpulsed recipients. A low reduction in response to  chain insulin and proinsulin was detected in diabetic mice receiving pulsed iDCs. No enhanced response to any immunizing peptide including the ignored GAD peptide was observed. There was a nonspecific reduction in T cell responses to other beta cell antigens; this is probably due to the induction of regulatory T cells. The authors concluded that protective iDC-based therapies require target antigen presentation [21]. Mucosal administration of autoantigens is considered a promising ‘tolerogenic’ strategy, based on the evidence derived from the successful treatment of several animal models of autoimmunity [22]. In particular, the cholera toxin B subunit (CTB) has been used in vaccine strategies as an immunopotentiating carrier for autoantigenic peptides. A fusion protein CTB-GADIII was constructed with 3 tandem epitopes from GAD65 p217-236, p524-538 and p290-306 [22]. These were shown to effectively prevent T1D when administered early in the NOD mouse model of disease, although not effective when administered alone [16]. Eight-week-old NOD females from 3 groups of animals were immunized intranasally with CTB-GADIII, CTB or phosphate-buffered saline (PBS) at 8, 10, 12 week time intervals. Administration inhibited spontaneous Th1 autoimmunity, but did not produce Th2 responses to  cell autoantigens. Authors also investigated the effect of CTB-GADIII immunization on the progression of insulitis. Approximately 70% of the islets from CTB-GADIII-treated mice had little or absent infiltration, while severe insulitis was detected on pancreatic sections of control mice either PBS or CTB treated. Generally the efficacy of any peptide immunotherapy approach is strictly dependent on several factors [23]. These include the stage of the disease process in which the therapy is started; usually the efficacy is enhanced at late preclinical stage of disease, when there is a high number of pathogenic circulating T cells. Other factors that are known to influence the efficacy of peptide immunotherapy are the dose, the route of administration of the therapy and whether an adjuvant is employed. The efficacy also relies on the binding affinity of peptides to MHC molecules and the stability of peptides in vivo. For example, sometimes the efficacy of peptide immunotherapy is halted when a rapid clearance of the peptide from circulation or an inappropriate presentation by antigen-presenting cells (APC) occurs [23]. As reported in the paper by Lin [23] special recombinants were engineered, called soluble (s) IA g7-Ig dimers, with the aim of enhancing the efficacy of peptide immunotherapy. These are composed of extracellular domains of the MHCII  and  chains linked to an Ig scaffold. In these constructs an immunogenic peptide is exposed by the soluble MHCII-  chain. Thus, each bivalent fusion molecule presents a peptide which binds to T cells directly, independently of APC. Dimers were covalently linked to beta cell autoantigen-derived peptides from GAD65, i.e. spanning amino acid residues 217 to 236 (GADp217) and residues 290 to 309 (GADp290) and tested for their capacity to suppress late preclinical T1D. In order to verify their improved efficacy, 12-week-old NOD female mice with established beta cell autoimmunity (late preclinical disease) were vaccinated i.v. with a short course of these (s) IA g7-Ig dimers. Previous work from the same authors had shown that the administration of GADp217 and GADp290

Peptide Vaccination to Treat Type 1 Diabetes

prepared in IFA was able to suppress late preclinical T1D in 12- week-old NOD female mice. The treatment was efficient if multiple injections were applied and even at high dose (approximately 200μg) [16]. Prevention of diabetes did not occur when the single peptide was administered alone. Mice received 3 i.v. injections of 50μg of (s)IA g7-Ig in PBS over three days, followed by another 3 injections 3 days later; clinical disease was monitored up to 35 weeks of age. Each 50μg injection of (s)IA g7-Ig was equivalent to 1.1μg of native peptide. Treatment with (s)IAg7-Ig dimers tethered with only approximately 7μg of native peptide effectively blocked the progression of insulitis and the development of diabetes. No significant difference in the time of onset and frequency of diabetes was observed in NOD mice without treatment versus those treated with the control (s)IAg7 dimer constructed with hen egg lysozyme (HEL) epitope 12-26. In vaccinated NOD mice, suppression of the disease was dependent on beta cell-specific IL-10 secreting CD4+ T cells, although the frequency of GAD65-specific FoxP3expressing CD4+ T cells was also increased. These results demonstrate that MHC class II-Ig dimer vaccination is a robust approach to suppress ongoing T cell mediated autoimmunity and may provide a superior strategy of adjuvant-free, peptide-based immunotherapy to induce immunoregulatory T cells [23]. In another study a soluble I-A g7/GAD65 217230/ Fc2a dimer (DEF-GAD65) was able to activate a silent, monoclonal T regulatory cell population, to prevent and reverse T1D in double transgenic mice [24], where the disease is induced by a monoclonal diabetogenic T cell population. In a recent study, a soluble dimeric I-A g7/GAD65 217-230/ Fc2a (DEF-GAD65) chimera was even able to suppress the polyclonal diabetogenic T-cell process in NOD adult mice by a ‘single epitope bystander mechanism.’ This was done by inducing a GAD65 217-230 specific CD4 T regulatory cell population [25]. Hyperglycemia was reversed and the diabetogenic process was stabilized as long as 2 months post-therapy. The chimera was able to induce IL 10 secretion by antigen-specific CD4+ T cells in vitro. Insulin and Proinsulin Peptide Immunotherapies Experimental evidence suggests that insulin is a primary -cell specific autoantigen in T1D; indeed NOD mice deficient in proinsulin I and II, but maintaining the expression of an altered insulin molecule, fail to develop diabetes [26]. Furthermore, insulin-reactive T cell clones have been isolated from the PNL of long lasting T1D patients [27]. Autoantigenic peptides of insulin and proinsulin showed a difference in efficacy in inducing a sufficient frequency of immunoregulatory T cells, and thus T cell tolerance, than GAD65 epitopes. The insulin peptide InsB 9-23 prepared in IFA, injected intraperitoneally at 50μg dose, was found to effectively prevent diabetes and block insulitis in NOD females at 4 weeks of age [28] in agreement with Tisch’s studies [16]. Diabetes prevention coincided with a significant increase in the frequency of InsB9-23-specific IL 4 secreting CD4+ T cells in the PNL. However, when 12-week-old NOD females were vaccinated with 4 injections of 200μg of InsB9-23 peptide, these animals developed diabetes similarly to control groups and failed to induce InsB9-23-specific immunoregulatory T cells.

Current Medicinal Chemistry, 2011 Vol. 18, No. 1

5

Proinsulin has been considered a candidate for an early autoantigen, because it is found to be expressed in the thymus and at highest concentrations in  cells. Thus authors investigated whether the proinsulin p24-33 epitope that spans the B chain/C peptide junction in proinsulin could have a tolerogenic effect in the NOD mouse in the perinatal period [29]. Intraperitoneal immunization with p24-33 proinsulin epitope, beginning perinatally at 18 days of age did indeed delay the onset and reduced the incidence of T1D. Peripheral T cell responses of draining lymph node T cells to the recombinant mouse intact proinsulin II and proinsulin B2433 emulsified in complete Freund adjuvant (CFA) were investigated in NOD female mice (4-5 week old) immunized in the footpad. Experiments in vitro confirmed that perinatal NOD proinsulin-specific T cells appear to react to the immunodominant B24-33 peptide. Peripheral T cells from perinatal NOD mice are polarized toward a Th1-type cytokine secretion profile, following stimulation by mouse proinsulin and its peptides. In contrast, when the proinsulin treatment was applied after the development of insulitis (5 weeks of age), there was acceleration of the clinical onset of disease in NOD mice. These findings supported the notion that proinsulin p24-33 may be a primary autoantigenic epitope in the pathogenesis of T1D in NOD mouse. Delayed onset of T1D was obtained by treating 12-week-old NOD female mice with B: 9-23 insulin peptide (40μg) by the intranasal route [30]. Mice were also treated at 4 weeks of age and every 4 weeks thereafter. A more complete protection from diabetes was obtained with multiple treatments rather than a single treatment. In a study by Martinez (2003) [31], 8-week-old female NOD mice were treated with 3 intranasal doses of proinsulin B24-C36 peptide binding to I-Ag7, the MHC class II molecule of a NOD mouse. Splenocytes from NOD mice, treated with proinsulin peptide B24-C36, when co-transfected with diabetogenic splenocytes, were able to moderately reduce the incidence of diabetes. The disease was hampered instead by co-transfer of either purified CD4+ T cells or splenocytes depleted of CD8+ T cells from mice treated with B24-C36 peptide. However, in mice treated with a single dose of intranasal peptide, there was only a small effect of suppression of spontaneous diabetes. This was probably due to the fact that not only CD4+ regulatory T cells but also CD8+ autoreactive T cells were induced. Homology modeling studies demonstrated that B24-C36 and its core sequence B25-C34 were able to bind with high affinity to Kd and to elicit CD8+ cytotoxic T lymphocytes in vitro. When 50μg of the B24C32/33 were administered in IFA intraperitoneally to NOD mice, treatment was able to delay diabetes onset. Remarkably, the peptide B24-C32/33, ‘unmasked’ by the integral CD8+ T cell epitope by truncation at C34 residue, was revealed to be a potential CD4+ tolerogenic peptide of proinsulin. Indeed, a single intranasal administration of the peptide (40μg) significantly reduced diabetes development. Arai et al. [32] selected a determinant in the leader sequence of preproinsulin, the preproinsulin 1 L7-24; its therapeutic potential was investigated because of its binding affinity to the MHC I-A g7 molecule. The peptide (100μg), emulsified in IFA, was administered subcutaneously to 14-weekold NOD mice. Two subcutaneous administrations were applied at 1-week intervals. Administration of L7-24 induced

6

Current Medicinal Chemistry, 2011 Vol. 18, No. 1

CD4+ CD25+ FoxP3+ regulatory T cells in the spleen. Splenocytes of NOD mice immunized with this peptide secreted IL 4 and IL 10 in response to the peptide. Subcutaneous administration of the peptide into 12-week-old mice significantly prevented the development of diabetes. However, the L7-24 peptide could not prevent diabetes in insulin 2 knockout mice; this represents an accelerated model of T1D in which the disease develops around 10 weeks of age and diabetes incidence is approximately 100%. Authors investigated the therapeutic efficacy of the L7-B24 peptide on newlydiagnosed diabetic mice. They were scheduled to receive injected L7-B24, B9-23 peptides or PBS alone. Normoglycemia was restored 2-5 weeks after the last injection in 5 out of 13 mice treated with L7-B24, but in only 1 out of 18 B923-treated mice. After 16 weeks, normoglycemia was restored and, on histological examination, some islets were still preserved. Authors concluded that, since epitope spreading occurs at advanced stages of the disease, combined peptide immunotherapy might be the best option, including the combination of peptide therapy and CD3 antibody [32]. It has been shown that systemic immunosuppression, for example with cyclosporin, can halt beta cell destruction. However the protection only lasted as long as the drug was present; long-term immunological tolerance to beta cell antigens was not achieved and extended therapy was not feasible due to the side effects of cyclosporin. Two animal models of disease, NOD and H-2d RIP-LCMV mice (transgenic mice expressing the LCMV viral protein [12]), were administered a combination therapy that combined anti-CD3 specific antibody with various peptides derived from islet antigens, at the stage of diabetes onset [33]. The best synergistic effect of enhanced remission of diabetes was produced by combining anti-CD3 and intranasal proinsulin II B24-C36 peptide (hpIIp). Expansion of Tregs and hpIIp-specific T cells was obtained after 5-6 months of treatment in pooled cells from spleens and PLN of treated animals. HpIIp specific T reg cells also produced ‘regulatory’ cytokines such as IL 10, TGF beta. A lower increase of IL 4 was also detected. The authors also produced evidence that, after administration of the combined therapy to newly diagnosed NOD and RIPLCMV mice, there was indeed induction of Tregs able to suppress CD8+ autoaggressive T cells. A reduction in the percentage of these cells was verified in single-cell suspensions from spleens and PNL by staining with NRP-7 and NP118 tetramers. It was also envisaged that the combined treatment, being able to induce Tregs that act sitespecifically, may carry a reduced risk of systemic side effects [33]. As already mentioned (vide supra) several studies have shown that DCs cultured in the presence of fetal bovine serum (FBS) may induce a powerful T helper type 2 (Th2) immune response towards FBS-related antigens halting an ongoing immune response [21]. This may interfere with diabetes development in the NOD mouse by induction of immune deviation rather than by antigen-specific tolerance. Haase et al. [34] also recently further investigated whether antigen-specific tolerance induction by DC therapy is feasible in the NOD mouse. After their isolation from bonemarrow, immature DCs were generated using autologous serum [normal mouse serum (NMS)-supplemented cultures] instead of FBS. Cells were cultured in the presence of IL 10,

A. Fierabracci

which is known to prevent their maturation. Cells were pulsed with 10 μg of InsB9-23 and InsB15-23 peptide on day 7, harvested on day 8 and washed extensively in PBS. Prediabetic NOD mice were receiving weekly pulsed or unpulsed NMS-IL 10 treated DCs, administered intraperitoneally in PBS. It was shown that these DCs can protect NOD mice from diabetes, if pulsed with insulin-peptide antigens InsB9-23 and InsB15-23 before adoptive transfer [34]. Unpulsed DCs had no effect on disease development. Other Peptide Immunotherapy Approaches The use of heat shock proteins in the pathogenesis of T1D has been questioned in literature [35]. Extensive preclinical studies using the hsp60 peptide p277 (contained between aminoacids 437 and 460) have proven the efficacy of peptide vaccination in blocking diabetes progression in NOD mice. p277 administration was able to induce Th2 cell type response with upregulation of IL 10 and IL 13 and downregulation of IFN- [36]. In a more recent study [37] the effect of the intranasal vaccination with p277 peptide carried by HSP65 in the absence of adjuvants on autoimmune diabetes in 4-week-old NOD mice was investigated. The role of chaperone molecules as facilitators of immune responses to proteins and peptides is well documented both in vivo and in vitro [reviewed in 37]. In Liang’s paper [37], HSP65 was used because it also works as vaccine antigen in T1D as well as an adjuvant itself. Again a significant decrease in the incidence of disease [37] was obtained by inducing an antiinflammatory immune response due to IL 10 and IL 4 secretion. Tarbell et al. [38] exploited the possibility of using DCs in the development of Ag-specific immunotherapies based on the expansion of CD25+ CD4+ suppressor T cells. In this paper diabetes was rapidly induced in NOD BDC2.5 transgenic mice [12, reviewed in 38] by administering one dose of cyclophosphamide. CD25+ CD4+ suppressive T cells from these mice were first expanded in vitro with DCs obtained from normoglycemic NOD mice and BDC mimotope (termed 1040-p31) for 5-7 days. The expanded CD25+ CD4+ BDC2.5 T cells were more suppressive in vitro than CD25+ CD4+ T cells expanded with only DCs and antiCD3. Authors also proved that CD25+ CD4+ BDC2.5 T cells expanded with DCs and antigen could halt the diabetic process in NOD BDC2.5 mice if administered 3 days after cyclophosphamide treatment. This indicates that the cell population was able to suppress the autoimmune process even when the disease developed rapidly. In a second model, spleen cells from diabetic NOD mice were injected into NOD.SCID female mice. It is well known that this model is due to pathogenic T cells having different TCR specificities. Diabetes development was halted when DC-expanded CD25+ CD4+ BDC2.5 T cells from BDC2.5 mice were injected with spleen cells. On the contrary, when mice received only spleen cells from diabetic mice, the disease developed. Authors also raised the relevant issue that CD25+ CD4+ T cells expanded with DCs and a single autoantigen specificity can suppress diabetes directed to multiple specificities [38]. Effective tolerance induction for treatment of autoimmune diseases can be achieved with the administration of autoantigenic peptides covalently cross-linked to cellular

Peptide Vaccination to Treat Type 1 Diabetes

vehicles via ethylene carbodiimide (ECDI, reviewed in [39]). This method has been shown to inhibit pathogenic T cell proliferation, decrease cytokine production and induce anergy. In the NOD mouse model, diabetes was first induced by the adoptive transfer of activated CD4+ BDC2.5 transgenic T cells. These T cells recognize an islet antigen that can be mimicked by the peptide 1040-p31. The next day newly diagnosed diabetic mice were treated with 1040-p31 peptide-coupled ECDI fixed splenocytes (SP). A control was established by administration of irrelevant peptide-coupled ECDI-fixed splenocytes. In these control animals severe diabetes develops within 7 days of transfer. P31-splenocyte treatment provided complete and long-lasting protection. The model replicated the results obtained with insulin-SP in spontaneous diabetic NOD mice, representing a useful tool to dissect the basis for tolerance [40]. Robust long-term tolerance depended on the programmed death 1 (PD-1) and the programmed death ligand pathway (PD-L), but not on the distinct cytotoxic T lymphocyte associated antigen 4 pathway (CTLA-4). CD8+ T cells are certainly important contributors to the initiation and progression of T1D [1, 2]. A significant fraction of CD8+ T cells attacking the NOD islets recognize epitopes of islet-specific glucose-6-phosphatase catalyticsubunit related protein (IGRP), a known autoantigen protein of unknown function. The pathogenetic activity of this T cell subset is controlled by genetic elements associated with diabetes susceptibility and resistance. Nevertheless, some IGRP-reactive CD8 T cell clones are not pathogenic but antidiabetogenic. Altered peptide ligands (APL) have been exploited as peptide variants of the agonist peptide that are substituted at T Cell receptor (TCR) contact sites. It has been shown that administration of APLs targeting IGRP (206214) reactive CD8+ T cells to female NOD mice transgenic for IGRP-reactive NY8.3 TCR resulted in diabetes protection only at doses that did not delete low-avidity clones, suggesting a protective role for these clonotypes [41, 42]. The use of MHC tetramers, in particular class I tetramers, has revolutionized the study of CD8+ T cells [43]. Experimental data demonstrate that the ability of peptide-MHC class I tetramers to bind Ag-specific T cells can be used to specifically identify those populations that are relevant to the pathogenesis of autoimmune disease and therefore ‘unwanted’ in the peripheral blood. However, peptide-MHC tetramers (‘suicide’ MHC tetramers) can also carry toxins and radionuclides to these cell populations for either their imaging or deletion. Saporin-coupled MHC tetramers were used to delete IGRP-reactive T cells in vitro and in vivo in NOD mice transgenic for IGRP-reactive NY8.3 TCR [44]. The treatment was able to delay development and progression of the disease nearly as effective as low affinity altered peptides administration in earlier studies [42, 45-46]. Peptide treatment was only effective when initiated at 4 weeks of age and disease progression was unaltered when peptide treatment was initiated in 10-week-old female NOD mice. In the Vincent study [44], the therapy was markedly effective on T1D when 8-week-old female mice were treated with the toxin tetramer, when islet infiltrates are more evident. Mice received only 3 administrations of toxic MHC tetramer over a 10-day period, while in earlier studies the effect was only observed with repeated administrations of peptide. Therefore

Current Medicinal Chemistry, 2011 Vol. 18, No. 1

7

it has been envisaged that peptide-MHC tetramers are an effective novel strategy for in vivo immunomodulation whenever repeated administration of peptide is not recommended or a direct rather than an APC-mediated depletion of antigenic specificities is desirable. Glatimer acetate, also known as Copaxone, is a random copolymer of alanine, lysine, glutamate and tyrosine that has been successfully used to treat multiple sclerosis and other autoimmune conditions such as uveoretinitis, inflammatory bowel disease, graft rejection and hepatic fibrosis. Regarding T1D [47], it can alter the clinical course of diabetes in cyclophosphamide (CY)-potentiated non-obese diabetic (CYNOD) mice. GA was administered subcutaneously (2mg per mouse) to 11-week-old NOD mice every other day for a total of 7 days. Treatment with GA significantly reduced the diabetic rate in mice and ameliorated insulitis, which coincided with increased CD4+CD25+ Foxp3+ T cell response (Tregs) in treated mice. The treatment led to increased expression of transcription factor Foxp3 mediated partially through IL 4. This experimental evidence suggests the potential success of GA for T1D treatment. PEPTIDE IMMUNOTHERAPIES IN OTHER ANIMAL MODELS OF T1D As stated above (vide infra), knowledge related to peptide immunotherapies has mainly been obtained through studies on the NOD mouse model of disease. As reported by Roep and Atkinson [48], extensive investigation on more animal models could provide relevant and more complete information, considering the fact T1D in humans has a heterogeneous phenotype. In addition to what has already been reported for transgenic NOD animal models of disease (vide supra), here below we review the limited number of studies employing other animal models of disease, such as BB rats and mice with low-dose streptozotocin-induced diabetes. Indirect evidence is produced that GAD65 and insulin B chain (9-23) may not be immunodominant antigens in this model. Bieg et al. [49] proved that when rats were injected with GAD65 intrathymically and intraperitoneally they developed high anti-GAD65 antibody titers without altering diabetes development. A form of autoimmune diabetes can be induced in the C57BL/ksj strain of mice by injection of low dose streptozotocin (STZ). In the course of the disease, spontaneous autoimmunity to hsp60, its p277 peptide and insulin develops. It has been shown [50] that STZ-induced autoimmune disease could be blocked by the administration of p277 peptide in mineral oil (100μg). The treatment was effective both when administered early in the disease course, before appreciable insulitis (1 week after toxin induction), or late, after the initiation of autoimmunity (85 days later). The peptide caused a T cell proliferative response to p277, following the production of anti-p277 antibodies (Th2-type). CONCLUDING REMARKS The characteristics in the immunopathogenesis of autoimmune diabetes have profound consequences for immunotherapy.  cell destruction is a T cell mediated process in both preclinical animal models and humans.

8

Current Medicinal Chemistry, 2011 Vol. 18, No. 1

A. Fierabracci

both preclinical animal models and humans. Differences exist between mice and humans – discrepancies in both innate and adaptive immunity, contribution of B cells, autoantibodies and primary antigenic targets. In humans there is a limited infiltration of pancreatic islets; peri-insulitis is absent, while it is well-represented in NOD mice [51]. As previously mentioned, in contrast, the insulitis lesion in BB rat is most similar to the human model [48].

various animal models. In addition to what has been already discussed, this could also result from an inappropriate dose of antigen being administered. Another observation is related to the difficulty of investigating the prediabetic period in humans. By the time prediabetics are identified on the basis of autoantibody screening, the autoreactive repertoire is likely to be well-expanded and it is then difficult to halt the diabetogenic process.

Investigations into the NOD mouse model have provided relevant information suggesting routes of clinical research in humans in the field of immunotherapy, in particular, peptide immunotherapy. Peptide immunotherapy will be crucial in defining the exact therapeutic window for intervention to successfully prevent T1D, including the dose of treatment and adjuvant to be administered and the frequency of treatment. On the other hand, results obtained in animal research have also often generated erroneous expectations in guiding clinical translation. Designing therapies for humans, based on the outcome of experimental models of T1D, is a very difficult task. In other words, it is difficult to extrapolate conclusions and predictions on the possible effect that an approach could have when applied to humans. This is due to the considerable disease heterogeneity associated with disease genetic and non-genetic determinants that underlies T1D in humans, in comparison with the ‘simple’ and ‘well standardized’ condition in the NOD mouse. The results obtained in NOD mice are limited to providing information only for a subpopulation of diabetic patients. In future, improved screening techniques for detecting autoantibodies, their target antigens and autoreactive T cells [52] using microarray technologies will probably allow classification of proteins according to their individual immune cell phenotype. This would make it possible to design more-specific patient-tailored vaccines. Furthermore, protective treatments may produce different effects at the molecular level in different species. The molecular pathways that are altered in the course of peptide immunization are the object of intense investigations even in animal models [54]. These studies are relevant in order to interpret the differences from the human model in order to design improved strategies for intervention.

Immune interventions in humans are currently being explored in particular, focusing on diagnosis in T1D patients, with the aim of blocking the autoimmune process, thereby preserving and restoring  cell mass and function [56]. In addition to non-antigen specific immunoregulatory approaches with anti-CD3 antibody and anti-inflammatory agents (anti-IL 1 receptor agonist), antigen specific therapies are also being proposed, including the peptide DiaPep277 and GAD65 alum-formulated vaccine Diamyd. Large international multicenter trials are currently underway and are already showing similar beneficial effects, aiming to preserve insulin-secreting cells [56, 57]. Regarding p277, in contrast to the experience in NOD mice treated with p277 by intranasal vaccination [37], Diapep277 treatment given as a subcutaneous vaccination in humans appears to have only slight beneficial effects in adult T1D patients; this effect has not been confirmed since it is not proven in children with T1D [58, 59]. Citing other examples, isoform GAD65 in NOD mice prevented autoimmune destruction of pancreatic  cells and reduced the subsequent need for exogenous insulin replacement [reviewed in 57]. This prevention was characterized by induction of a  cell-specific Th2-like cytokine response. In the NOD mouse, disease progression is highly synchronized and can be staged on the basis of the progression from preinsulitis to destructive insulitis. The efficacy of Diamyd has been proven in Phase I and II human trials: the treatment was successfully administered (subcutaneously) to recent onset diabetic patients [60] and the specific group of LADA (latent autoimmune diabetes) patients [61], without adverse effects. There was attenuation of decline in stimulated C peptide in GAD treated groups compared to controls, indicating its effectiveness in limiting the disease process. In particular, the protective effect of GAD alum was verified within 6 months of diagnosis. It must be underlined that, in contrast with the NOD model, the efficacy of this particular vaccine has not been explored thus far as applied in the prediabetic period in humans. Nevertheless, with the promising data obtained with recent onset T1D and LADA patients, further studies are on their way to verifying its preventive effect on the disease onset as well as its therapeutic effect.

As extensively discussed by Luo [7], interventions that provide valid results when applied at disease onset are postulated to be effective in the prediabetic period, when the molecular repertoire of autoreactive T cells is more restricted and epitope spreading has not yet occurred. Contrary to the mouse model [55], when moving to humans we should expect interindividual variations in the autoreactive repertoire of T cells. Some clones may already be activated at the time of immunization and their avidities can vary on the basis of central and peripheral mechanisms of tolerance, thereby influencing the individual characteristics of the immune response. Generally, as in the animal model, human trials have been applied both in prediabetic (prevention trials) and in recently-diagnosed diabetics (therapeutic trials). In providing examples, some trials using intensive insulin therapy (i.e. the Diabetes Prevention Trial 1), either orally- or parentallyadministered in the prediabetic state [reviewed in 7], failed to produce prevention of disease, or even produced disease acceleration, showing no clear effect on the autoimmune process [28], even though they were proven to be effective in

In future, the need to develop combination therapies with various immunoregulatory or suppressive agents used in combination with antigen or antigenic-peptide-specific tolerance strategies could arise. In particular, it may be necessary to exploit alternative methods for down-regulating the diabetic process, such as the use of antigen-pulsed iDCs or methods to selectively delete populations of autoreactive T cells. This is in agreement with data already existing in literature [7]. These strategies would be ideal in translational research not only to treat patients in the preclinical stage or at early onset of disease but also in long-standing disease. This is because pancreatic autoantigenic-specific processes require targeting even prior to transplanting stem cells that

Peptide Vaccination to Treat Type 1 Diabetes

could be capable of differentiation into insulin-producing beta cells. As a concluding remark, anaphylactic sensitization is a real risk in using GAD65 immunization to prevent T1D, as revealed by animal models of the disease [7]. The phenomenon is not limited by the nature of the immunizing peptide mixture (whether peptides emulsified in adjuvants or aqueous solution) or by the genetic background of the recipient strain. These undesirable side effects have also been reported when using altered peptide immunization strategies to treat patients with multiple sclerosis [62]. These observations are clearly of great relevance in pursuing the possibility of specific antigen/ peptide immunotherapies in T1D [57].

Current Medicinal Chemistry, 2011 Vol. 18, No. 1 [18]

[19]

[20]

[21]

[22]

REFERENCES [23] [1]

[2]

[3]

[4]

[5] [6]

[7]

[8] [9]

[10] [11] [12]

[13]

[14]

[15]

[16]

[17]

Fierabracci, A.; Bottazzo, GF. The continuous discovery of autoantigens in endocrine organ-specific autoimmunity: do they help us to understand pathogenesis? Springer Semin. Immunopathol., 2002, 24, 243-259. Fierabracci, A. Unravelling autoimmune pathogenesis by screening random peptide libraries with human sera. Immunol. Lett., 2009, 124, 35-43. Fierabracci, A. Unravelling the role of infectious agents in the pathogenesis of human autoimmunity: the hypothesis of the retroviral involvement revisited. Curr. Mol. Med., 2009, 9, 1024-1033. Bluestone, J.A.; Herold, K.; Eisenbarth, G. Genetics, pathogenesis and clinical interventions in Type 1 diabetes. Nature, 2010, 464, 1293-1300. Taplin, C.E.; Barker, J.M. Autoantibodies in Type 1 diabetes. Autoimmunity, 2008, 41, 11-18. Wang, B.; Tisch, R. Parameters influencing antigen-specific immunotherapy for Type 1 diabetes. Immunol. Res., 2008, 41, 175187. Luo, X.; Herold, K.C.; Miller, S.D. Immunotherapy of Type 1 diabetes: where are we and where should be going ? Immunity, 2010, 32, 488-499. McDevitt, H. Specific antigen vaccination to treat autoimmune disease. Proc. Natl. Acad. USA, 2004, 101, 14627-14630. Gregori, S.; Battaglia, M.; Roncarolo, M.G. Re-establishing immune tolerance in type 1 diabetes via regulatory T cells. Novartis Found. Symp., 2008, 292, 174-183 Chea, D. Animal models of Type 1 (Insulin-Dependent) Diabetes mellitus. J. Pediatr. Endocrinol. Metab., 1998, 11, 11-19. Banting, F.G; Best C.H. The internal secretion of the pancreas. Lab. Clin. Med., 1922, 7, 251-266. Lam-Tse, W.K.; Lernmark, A.; Drexhage, H.A. Animal models of endocrine/organ-specific autoimmune diseases: do they really help us to understand human autoimmunity? Springer Semin. Immunopathol., 2002, 24, 297-321. Stone M.A.; Camosso-Stefinovic, J.; Wilkinson, J.; de Lusignan, S.; Hattersley, A.T.; Khunti, K. Incorrect and incomplete coding and classification of diabetes: a systematic review. Diabet. Med., 2010, 27, 491-497. Tisch, R.; Yang, X.D.; Singer, S.M.; Liblau, R.S.; Fugger, L.; McDevitt, H.O. Immune response to glutamic acid decarboxylase correlates with insulitis in non-obese diabetic mice. Nature, 1993, 366, 72-75. Yoon, J.W.; Yoon, C.S.; Lim, H.W.; Huang, Q.Q.; Kang, Y.; Pyun K.H.; Hirasawa, K.; Sherwin, R.S.; Jun, H.S. Control of autoimmune diabetes in NOD mice by GAD expression or suppression in beta cells. Science, 1999, 284, 1183-1187. Tisch, R.; Wang, B.; Serreze, D.V. Induction of glutamic acid decarboxylase 65-specific Th2 cells and suppression of autoimmune diabetes at late stages of disease is epitope dependent. J. Immunol., 1999, 163, 1178-1187. Tisch, R.; Wang, B.; Atkinson M.A.; Serreze, D.V.; Friedline, R. A glutamic acid decarboxylase 65-specific Th2 cell clone immunoregulates autoimmune diabetes in nonobese diabetic mice. J. Immunol., 2001, 166, 6925-6936.

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

9

Tarbell, K.V.; Yamazaki, S.; Olson, K.; Toy, P.; Steinman, R.M. CD5+ CD4+T cells, expanded with dendritic cells presenting a single autoantigenic peptide, suppress autoimmune diabetes. J. Exp. Med., 2004, 199, 1467-1477. Chen, G.; Han, G.; Feng, J.; Wang, J.; Wang, R.; Xu, R.; Shen, B.; Qian, J.; Li, Y. Glutamic acid decarboxylase-derived epitopes with specific domains expand CD4+ CD25+ regulatory T cells. PloS One, 2009, 4, e7034. Clare-Saltzer, M.J.; Brooks, J.; Chai, A.; Van Herle, K.; Anderson, C. Prevention of diabetes in nonobese diabetic mice by dendritic cell transfer. J. Clin. Invest., 1992, 90, 741-748. Lo, J.; Peng, R.H.; Barker, T.; Xia, C.-Q.; Clare-Salzler, M.J. Peptide-pulsed immature dendritic cells reduce response to  cell target antigens and protect NOD recipients from Type 1 diabetes. Ann. N.Y. Acad. Sci., 2006, 1079, 153-156. Wang, H.; Yang, J.; Jin, L.; Feng, J.; Lu, Y.; Sun, Y.; Li, T.; Cao, R.; Wu, J.; Fan, H.; Liu, J. Immunotherapy of autoimmune diabetes by nasal administration of tandem glutamic acid decarboxylase 65 peptides. Immunol. Invest., 2009, 38, 690-703. Li, L.; Yi, Z.; Wang, B.; Tisch, R. Suppression of ongoing T cellmediated autoimmunity by peptide-MHC class II dimer vaccination. J. Immunol., 2009, 183, 4809-4816. Casares, S.; Hurtado, A.; McEvoy, R.C.; Sarukhan, A.; von Boehmer, H.; Brumeanu, T.D. Down-regulation of diabetogenic CD4+ T cells by a soluble dimeric peptide-MHC class II chimera. Nat. Immunol., 2002, 3, 383-391. Lin, M.; Stoica-Nazarov, C.; Surfs, J.; Kehl, M.; Bona, C.; Olsen, C.; Brumeanu, T.D.; Casares, S. Reversal of type 1 diabetes by a new MHC II-peptide chimera: “Single-epitope-mediated suppression” to stabilize a polyclonal autoimmune T-cell process. Eur. J. Immunol., 2010, 40, 2277-2288. Nakayama, M.; Abiru, N.; Moriyama, H.; Babaya, N.; Liu, E.; Miao, D.; Yu, L.; Wegmann, D.R.; Hutton, J.C.; Elliott, J.F.; Eisenbarth, G.S. Prime rile for an insulin epitope in the development of type 1 diabetes in NOD mice. Nature, 2005, 435, 220-223. Kent, S.C.; Chen, Y.; Bregoli, L.; Clemmings, S.M.; Kenyon, N.S.; Ricordi, C.; Hering, B.J.; Hafler, D.A. Expanded T cells from pancreatic lymph nodes of type 1 diabetic subjects recognize an insulin epitope. Nature, 2005, 435, 224-228. Muir, A.; Peck, A.; Clare-Salzler, M.; Song, Y.H.; Cornelius, J.; Luchetta, R.; Krischer, J.; Maclaren, N. Insulin immunization of nonobese diabetic mice induces a protective insulitis characterized by diminished intraislets interferon-gamma transcription. J. Clin. Invest., 1995, 95, 628-634. Chen, W.; Bergerot, I.; Elliott, J.F.; Harrison, L.C.; Abiru, N.; Eisenbarth, G.S.; Delovitch, T.L. Evidence that a peptide spanning the B-C Junction of proinsulin is an early autoantigen epitope in the pathogenesis of type 1 diabetes. J. Immunol., 2001, 167, 49264935. Daniel, D.; Wegmann, D.R. Intranasal administration of insulin peptide B: 9-23 protects NOD mice from diabetes. Ann. N.Y. Acad. Sci., 1996, 778, 371-372. Martinez, N.R.; Augstein, P.; Moustakas, A.K.; Papadopoulos, G.K.; Gregori, S.; Adorini, L.; Jackson, D.C.; Harrison, L.C. Disabling an integral CTL epitope allows suppression of autoimmune diabetes by intranasal proinsulin peptide. J. Clin Invest., 2003, 111, 1365-1371. Arai, T. ; Moriyama, H.; Shimizu, M. ; Sasaki, H. ; Kishi, M. ; Okumachi, Y. ; Yasuda, H. ; Hara, K. ; Yokono, K. ; Nagata, M. Administration of a determinant of preproinsulin can induce regulatory T cells and suppress anti-islet autoimmunity in NOD mice. Clin. Immunol., 2010, 136, 74-82. Bresson, D.; Togher, L.; Rodrigo, E.; Chen, Y.; Bluestone, J.A.; Herold, K.C.; von Herrath, M. Anti-CD3 and nasal proinsulin combination therapy enhances remission from recent-onset autoimmune diabetes by inducing Tregs. J. Clin. Invest., 2006, 116, 1371-1381. Haase, C.; Yu, L.; Eisenbarth, G.; Markholst, H. Antigendependent immunotherapy of non-obese diabetic mice with immature dendritic cells. Clin. Exp. Immunol., 2010, 160, 331-339. Atkinson, M.A.; Holmes, L.A.; Scharp, D.W.; Lacy, P.E.; Maclaren, N.K. No evidence for serological autoimmunity to islet cell heat shock proteins in insulin-dependent diabetes. J. Clin. Invest, 1991, 87, 721-724. Elias, D.; Cohen, I.R. Treatment of autoimmune diabetes and insu-

10 Current Medicinal Chemistry, 2011 Vol. 18, No. 1

[37]

[38]

[39]

[40]

[41] [42]

[43]

[44]

[45]

[46]

[47]

[48]

[49]

litis in NOD mice with heat shock protein 60 peptide p277. Diabetes, 1995, 44, 1132-1138. Liang, J.; Aihua, Z.; Yu, W.; Yong, L.; Jingjing, L. HSP65 serves as an immunogenic carrier for a diabetogenic peptide p277 inducing anti-inflammatory immune response in NOD mice by nasal administration. Vaccine, 2010, 28, 3312-3317. Tarbell, K.V.; Yamazaki, S.; Olson, K.; Toy, P.; Steinman, R.M. CD25+ CD4+ T cells, expanded with dendritic cells presenting a single autoantigenic peptide, suppress autoimmune diabetes. J. Exp. Med., 2004, 199, 1467-1477. Miller, S.D.; Turley, D.M.; Podojil, J.R. Antigen-specific tolerance strategies for the prevention and treatment of autoimmune disease. Nat. Rev. Immunol., 2007, 7, 665-677. Fife, B.T.; Guleria, I.; Gubbels Bupp, M.; Eagar, T.N.; Tang, Q., Bour-Jordan, H.; Yagita, H.; Azuma, M.; Sayegh, M.H.; Bluestone, J.A. Insulin-induced remission in new-onset NOD mice is maintained by the PD-1-PD-L1 pathway. J. Exp. Med., 2006, 203, 27322747. Santamaria, P. Genetic and therapeutic control of diabetogenic CD8+ T cells. Novartis Found. Symp., 2008, 292, 130-136. Han, B.; Serra, P.; Amrani, A., Yamanouchi, J.; Marée, A.F.; Edelstein-Keshet, L.; Santamaria, P. Prevention of diabetes by manipulation of anti-IGRP autoimmunity: high efficiency of a low affinity peptide. Nat. Med., 2005, 11, 645-652. Altman, J.D.; Moss, P.A.; Goulder, P.J.; Barouch, D.H.; McHeyzer-Williams, M.G.; Bell, J.I.; McMichael, A.J.; Davis, M.M. Phenotypic analysis of antigen-specific T lymphocytes. Science,1996, 274 , 94-96. Vincent, B.G.; Young, E.F.; Buntzman, A.S.; Stevens, R.; Kepler, T.B.; Tisch, R.M.; Frelinger, J.A.; Hess, P.R. Toxin-coupled MHC class I tetramers can specifically ablate autoreactive CD8+ T cells and delay diabetes in NOD Diabetic Mice. J. Immunol, 2010, 184, 4196-4204. Amrani, A.; Verdaguer, J.; Serra, P.; Tafuro, S.; Tan, R.; Santamaria, P. Progression of autoimmune diabetes driven by avidity maturation of a T cell population. Nature, 2000, 406, 739-742. Gebe, J.A.; Masewicz, S.A.; Kochik, S.A.; Reijonen, H.; Nepom, G.T. Inhibition of altered-peptide ligand-mediated antagonism of human GAD65-responsive CD4+ T cells by non-antagonizable T cells. Eur. J. Immunol., 2004, 34 , 3337-3345. Cui, G.; Zhang, Y.; Gong, Z.; Zhang , J.Z.; Zang, Y.Q. Induction of CD4+CD25+Foxp3+ regulatory T cell response by glatiramer acetate in type 1 diabetes. Cell Res, 2009, 19, 574-583. Roep, B.O.; Atkinson, M. Animal models have little to teach us about Type 1 diabetes: 1. In support of this proposal. Diabetologia, 2004, 47, 1650-1656. Bieg, S.; Hanlon, C.; Hampe, C.S.; Benjamin, D.; Mahoney, C.P. GAD65 and insulin B chain peptide (9-23) are not primary autoantigens in the type 1 diabetes syndrome of the BB rat. Autoimmunity, 1999, 31, 15-24.

A. Fierabracci [50]

[51]

[52]

[53] [54]

[55]

[56]

[57] [58]

[59]

[60]

[61]

[62]

Elias, D.; Cohen, I.R. The hsp60 peptide p277 arrests the autoimmune diabetes induced by the toxin streptozotocin. Diabetes, 1996, 45, 1168-1172. Roep, B.O. Are insights gained from NOD mice sufficient to guide clinical translation? Another inconvenient truth. Ann. N.Y. Acad. Sci., 2007, 1103, 1-10. Giuliani, L.; Mele, R.; Di Giovine, M.; Altieri, L.; Crinò, A.; Ravà, L; Fierabracci, A. Detection of GAD65 autoreactive T-cells by HLA class I tetramers in type 1 diabetic patients. J. Biomed. Biotechnol., 2009, 2009:576219. Raz, I; Eldor, R.; Naparstek, Y. Immune modulation for prevention of type 1 diabetes mellitus. Trends Biotechnol., 2005, 23, 128-134. Eckenrode, S.E.; Ruan, Q.-G.; Collins, C.D.; Yang, P.; McIndoe, R.A.; Muir, A.; She, J.X. Molecular pathways altered by insulin B9-23 immunisation. Ann. N.Y. Acad. Sci., 2004, 1037, 175-185. Peakman, M.; von Herrath, M. Antigen-specific immunotherapy for Type 1 diabetes: maximizing the potential. Diabetes, 2010, 59, 2087-2093. Pozzilli, P.; Leslie, R.D. New prospects for immunotherapy at diagnosis of type 1 diabetes. Diabetes Metab. Res. Rev., 2009, 25, 299-301. Ludviggson, J. Therapy with GAD in diabetes. Diabetes Metab. Res. Rev., 2009, 25, 307-315. Raz, I.; Elias, D.; Avron, A.; Tamir, M. ; Metzger, M. ; Cohen, I.R. Beta-cell function in new-onset type 1 diabetes and immunomodulation with a heat-shock protein peptide (DiaPep277) : a randomised, double-blind, phase II trial. Lancet, 2001, 358, 1749-1753. Schloot, N.C.; Meierhoff, G.; Lengyel, C.; Vándorfi, G.; Takács, J.; Pánczél, P.; Barkai, L.; Oroszlán, T.; Kovács, P.; St, G.; Battelino, T.; Hosszufalusi, N.; Jermendy, G. Effect of heat shock protein peptide DiaPep277 on beta-cell function in paediatric and adult patients with recent-onset diabetes mellitus type 1: two prospective, randomized, double-blind phase II trials. Diabetes Metab. Res. Rev., 2007, 23, 276-285. Ludvigsson, J.; Faresj, M.; Hjorth, M.; Axelsson, S.; Chérami, M.; Pihl, M.; Vaarala, O. ; Forsander, G. ; Ivarsson, S. ; Johansson, C. ; Lindh, A. ; Nilsson, N.O. ; Aman, J. ; Ortqvist, E. ; Zerhouni, P. ; Casas, R. GAD treatment and insulin secretion in recent-onset type 1 diabetes. N. Engl. J. Med., 2008, 359, 1909-1920. Agardh, C.-D.; Cilio, C.M.; Lethagen, A.; Lynch, K.; Leslie, R.D.G.; Palmér, M.; Harris, R.A.; Robertson, J.A.; Lernmark, Å. Clinical evidence for the safety of GAD65 immunomodulation in adult-onset autoimmune diabetes. J.Diabetes Complications, 2005, 19, 238-246. Kappos, L. ; Comi, G.; Panitch, H.; Oger, J.; Antel, J.; Conlon, P.; Steinman, L. Induction of a non-encephalitogenic type 2 T helpercell autoimmune response in multiple sclerosis after administration of an altered peptide ligand in placebo-controlled, randomized phase II trial. The altered peptide ligand in relapsing MS study group. Nat. Med., 2000, 10, 1176-1182.