Autoantigen-Specific Immunotherapy

1. Introduction

The incidence of type 1 diabetes (T1D) is increasing dramatically and new treatment modalities are needed urgently. Arising from autoimmune destruction of the pancreatic β-cells, T1D manifests itself when less than 10-20% of functional β-cells remain in the islets and is characterized as a disorder of glucose regulation due to the insufficient production of insulin. Insulin is a hormone secreted by pancreatic islet β-cells and its main function is to move blood sugar into cells where it is stored and later used for energy. In T1D, the β-cells are destroyed by the patients’ immune cells leading to limited or no insulin production. Without enough insulin, glucose accumulates in the bloodstream instead of entering the cells, causing symptoms of hyperglycemia. Chronic hyperglycemia can lead to many serious complications. Insulin replacement therapy is the current standard for treatment involving injection of recombinant insulin. Normal pancreatic insulin secretion is exquisitely sensitive to the minute-to-minute changes in blood glucose and glucose-stimulated insulin secretion (GSIS) and cannot be mimicked precisely by exogenous insulin injections.Thus, while insulin treatment successfully prolongs the life of affected individuals, it fails to prevent some of the serious complications of T1D that adversely affect quality of life and ultimately lead to significant morbidity and mortality.An alternative promising therapy is islet cell transplantation, but this suffers from several drawbacks including limited pancreatic islet tissue from cadaveric donors, the requirement for lifelong immunosuppression, and the increased risk of infection due to non-specific suppression of the immune system.

To cure T1D, researchers are pursuing combined immunological and biological strategies for restoring β-cell mass. Strategies for increasing β-cell mass include promoting endogenous pancreatic β-cell regeneration, reprogramming non-pancreatic cells into insulin-producing cells (IPCs), and generating unlimited autologous pancreatic β-cells from induced pluripotent stem (iPS) cells of the T1D patients.At the same time, there is considerable interest in approaches for modulating and suppressing autoimmunity to pancreatic β-cells and surrogates of IPCs.An individual’s pancreatic β-cell mass is tightlyregulated according to insulin demand, reflecting a balancebetween the rate of β-cell replication, regeneration/reprogramming, and the rate of β-cell apoptosis. Research efforts are focusing on how best to expand, reprogram, or generate β-cells or their surrogates from pancreatic stem/precursor cells, non-pancreatic adult/stem cells, or the iPS cells of T1D patients for β-cell replacement therapy.However, without preventing ongoing autoimmune destruction of β-cells, the regenerated, reprogrammed, or replaced β-cells or their surrogates may suffer the same fate as the endogenous pancreatic β-cells.

The loss of pancreatic β-cells in T1D is a T cell-mediated autoimmune process orchestrated by CD4+ and CD8+ T cells specific for β-cell antigens. Pathogenic diabetogenic T cells and autoantibodies to β-cell specific autoantigens have been found in T1D patients and non-obese diabetic (NOD) mice, a well-studied animal model of T1D.These T cells employ a variety of mechanisms to destroy β-cells. Moreover, genetic susceptibility factors for T1D mainly affect the ontogeny and survival of immune cell subsets at the level of the thymic selection and in the peripheral tolerance. T1D is an organ-specific autoimmune disease where the immune tolerance to the β-cells is broken and the immune system attacks and eliminates these β-cells. So, restoration of self-tolerance to β-cells and suppression of β-cell autoimmunity combined with increasing β-cell mass are the key components for β-cell replacement therapy.

Experimental evidence indicates that the immune system can be modulated and immune tolerance to β-cells can be restored in experimental animal models of T1D.Self-tolerance can be re-established centrally or peripherally by broad-based immunoregulatory strategies, including the inhibition or deletion of lymphocyte subsets and/or the use of agents proposed to induce or re-establish immune tolerance via activation of regulatory T (Treg) cells.An example is the use of nonmitogenic anti-CD3 or antithymocyte globulin. However, the induction of antigen-specific tolerance to β-cell antigens is a highly desirable objective because only the pathogenic autoreactive T cells are inactivated whereas the remainder of immune system remains unperturbed.Many autoantigens have been identified and implicated in relation to T1D.Well-established autoantigens include insulin (Palmer et al., 1983), glutamic acid decarboxylase 65 (GAD65) (Baekkeskov et al., 1990), insulinoma associated-2 antigen (IA-2) (Bonifacio et al., 1995), and heat shock protein (HSP) (Birk et al., 1996), as well as newly discovered β-cell-specific autoantigens zinc transporter-8 (ZnT8) (Wenzlau et al., 2007), pancreatic duodenal homeobox factor 1(PDX1) (Li et al., 2010), and chromogranin A (CHGA) (Stadinski et al., 2010).These markers of T1D are useful for diagnosis as well as for prediction prior to disease onset and may prove useful for intervention by restoring immune tolerance[See table-1].

Antigen-specific immunoregulatory therapies are currently being evaluated for preventing and treating T1D and they include autoantigen therapy, dendritic cell (DC)-based strategies, and adoptive transfer of Treg cells. This chapter will review recent work on the antigen specific immunotherapy of T1D including strategies for tolerance induction.

2. The importance and feasibility of antigen specific immunotherapy

Nonspecific and specific approaches have been exploited by immunologists to control the autoimmune response in T1D. They include deleting pathogenic T cells and activating regulatory T cells to induce immune tolerance against β-cells. Non specific approaches included using immunosuppressive drugs such as steroids or cyclosporin A as well as some monoclonal or polyclonal antibodies. Polyclonal anti-lymphocyte antisera and anti-CD3 or anti-CD4 monoclonal antibodies have been used to cure early diabetic disease in murine modelsfor T1D(Maki et al., 1992, Chatenoud et al., 1994, Makhlouf et al., 2004). Anti-lymphocyte antisera and anti-CD4 antibodies are strong immune suppressors and principally act on depleting the autoaggressive T cells. In contrast, anti-CD3 antibodies function as immune modulators (Chatenoud et al., 2007) since they can maintain longterm

Table 1.

Islet autoantigens in type 1 diabetes

immune tolerancethrough inducing adaptive Tregs, interfere with T cell activation, and induce anergy/apoptosis in activated cells. Based on results from the mouse model, the humanized anti-CD3 antibodies (teplizumab and otelixizumab) were explored in phase II trials. It was found that anti-CD3 antibody treatment could preserve endogenous insulin secretion and reduce HBA1c levels and insulin requirement in the treated group during a two year period in recent onset T1D. Recent follow-up studies indicate that the beneficial effects extend over a period of up to 5 years following a single treatment (Cernea et al., 2010).However, some patients develop “flu-like” symptoms due to significant cytokine release by antibody-bound T cells, and recurrent Epstein-Barr viral infections (Keymeulen et al., 2010). It is well known that steroid and and cyclosporin A therapy have very serious side effects. The common problem of antibody-based immunotherapies is that they can’t discriminate between normal versus autoimmune effector T cells. The safety concerns and adverse effects of antigen nonspecific interventions, as well as the lack of permanent remission of disease with any agent tested to date have prompted interest in antigen specific interventions that might modulate the autoimmune aspect of the disease. Compared with nonspecific immune-approach, antigen-based immunotherapy allows selective targeting of disease-relevant T cells, while leaving the remainder of the immune system intact. Numerous studies have demonstrated that antigen specific immunotherapy can be highly effective in preventing and suppressing autoimmunity in rodent models of T1D and other autoimmune diseases (Harrison et al., 2000).

The concept of antigen-specific tolerance has been apparent for some years. The presence of self-reactive lymphocytes in the blood of healthy individuals (Burns et al., 1983) implies that self-antigen-specific regulatory mechanismsare physiological and prevent pathological autoimmunity.Antigen-specific tolerance operatesdirectly on effector T cells (via apoptosis or anergy) andvia Treg cells that secrete anti-inflammatory cytokines orcompete with effector T cells at the level of the antigenpresentingcell (APC). Many factors determine the induction of antigen specific tolerance including: 1) characters/properties of antigen,dose/concentration, physical form (soluble or aggregated, intact proteinor peptide), chemical composition, affinity for MHC molecule,content of CD4+ and CD8+ T cell epitopes and‘immunodominance’, purityand contaminants (e.g. endotoxin), and adjuvants; 2) route of administration; 3) level of co-stimulation; and 4) antigen-specific precursor cell frequency and duration of treatment(Harrison et al. 2000). High-dose soluble peptide or monomeric protein delivered by the intraperitoneal (i.p.), subcutaneous (s.c.), or intravenous (i.v.) route induces clonal deletion or clonal anergy (Liblau et al., 1997).Delivering antigen through mucosal routes, for example oraland naso-respiratory routes, could induce mucosal immune tolerance that suppresses subsequent systemicpriming to the antigen (Faria et al., 1999).Oral tolerance means the specific suppression of immune responses to an antigen by prior administration of the antigen by the oral route.Oral tolerance is an active immunologic process that is mediated by more than one mechanism and is dose-dependent. Low doseshave been shown to induce differentiation of regulatory T cellscapable of downregulating T cells specific for distinct antigenswithin a given tissue. This phenomenon of bystander suppressionin oral tolerance is associated with the induction ofTh2 and TGF-β-secreting Tregs (Anderton et al., 1999) whereas higher doses favor the induction of clonal anergy or deletion (Chen et al., 1995).

T cells recognize antigen peptide via the T cell receptor (TCR) and produce a spectrum of responses that range from activation to anergy or cell death. In the late 1980s, peptides were modified to interfere with the MHC/peptide/TCR complex to prevent autoaggressive T cell activation (Adorini et al., 1988).Later on, this type of peptide was termed an altered peptide ligand (APL). In APLs, TCR contact residues in an immunodominant self-epitope are altered to impair the T cell strength of signal. So, APLs could induce anergy by engaging less than optimal numbers of TCRs.APLs have been applied therapeutically in animal models of autoimmune diseases and in the clinical trial (Harrison et al. 2000). One recent paper reported that proinsulin-derived APLs (prAPLs) modulate the cytokine signature of proinsulin-reactive T cells at a micromolar range by increasing anti-inflammatory cytokines, including TGF-β1 (van et al., 2010).The results suggest that APLs do not act as antagonists in vivo but mediate bystander suppression, probably by the generation of Treg cells.

Treg cells have an ability to regulate immune response by non specific bystander suppression in response to specific antigen that is recognized locally at the site of the lesion or in the draining lymph node. During the disease process of animal models and human T1D, T cell autoimmunity progressively spreads intra- and inter-molecularly among β-cell autoantigens such asinsulin, GAD65, HSP, and IGRP (van Belle et al., 2011). Bystander suppressionresolves the dilemma of multiple autoantigens in human autoimmune diseases by obviating the need to know ifthe antigen used to induce tolerance is the major or primarypathogenic autoantigen(Tang et al., 2008). Therefore, antigen-specific approaches use only one antigen to tolerize against multiple autoreactivities.In the long term, Treg cellscreate a regulatory environment by producing TGF-β and/or by substantiallychanging DC functions to promote new Treg cell production.These processes promote bystander suppression and are critical to thecreation of infectious tolerance that can spread beyond the local tissueand exert long-lasting influence over the immune system.In contrast, tolerance-based onanergy or deletion is limited to each individual antigen.

Antigen-specific therapy is logically appealing and demonstrablyeffective in experimental autoimmune disease animal models. This proof-of-concept and its safety make it an attractive approach for treatment of humanautoimmune disease including T1D.

3. Antigen specific immune tolerance induction strategies

Immunotherapy forT1D has been attempted at two main stages of the diseaseprocess - prior to clinical onset but after the appearance of islet autoantibodies (secondary prevention) and immediatelyafter diagnosis (intervention)(Staeva-Vieira, T. et al.2007).

Prevention of T1D, meaning intervening early during the development of disease,holds greater promise for success because, in this stage, organ damage has not progressed very far and β-cells are still capable of self regeneration. Successful prevention of T1D depends on:1) a good prediction/identification of at-risk individuals and 2) a safeintervention that causes no harm in those individuals whowould have never developed T1D.In T1D, identifyingthe individuals at risk for developing disease can be obtained through screening islet autoantibodies, combined with genetic markers in the HLA region, and additional risk factors (Siljander et al., 2009). Moreover, metabolic assessments can also be included to identify late-stages of disease development, at which there is already some loss of glycemic control. Thus, it is now feasible to ‘stage’ individuals according to various modeled rates of progression to T1D (Staeva-Vieira et al., 2007). Currently, preventive trials for T1D include two main classes, antigen- or non-antigen-specific. Compared with a non-antigen specific-approach, an antigen-specific immune tolerance holds more promise and advantage. Many antigen-specific prevention trials have been conducted in recentyears. Of note is the oral insulin trial used in relatives of T1D patients following screening of 103,391 first- and second-degree relatives of patients with T1D. Of them, 97,273 samples were analyzed for islet cell antibodies and 2,523 underwent genetic, immunological, and metabolic staging to quantify risk of developing T1D.This trialconcluded that it is possible to identify individuals at high risk for T1D and enroll them in a large, multisite, randomized, and controlled clinical trial (Skyler et al., 2005).

Immune intervention for recent onset T1D aims to prevent or reverse the disease by blocking autoimmunity, thereby preserving/restoring β-cell mass and function and decreasing the likelihood of both hypoglycemia and long-term complications. It is a clinically feasible approach because potential subjects are readily identified and efficacy can be evaluated within a much shorter time frame. Currently, large quantities of evidence show that it is possible to reverse clinically established T1D. Firstly, evidence shows that loss of β-cells is a gradual process and that newly diagnosed patients still have 10-20% of functional β-cells remaining within the islets. Therefore, the reversal of hyperglycemia could be obtained through (1) promoting endogenous regeneration and replication of remaining β-cells, (2) reducing metabolic stress on the remaining β-cells, (In support, it is known that early and intensive insulin therapy in patients with recent onset T1D can clearly reduce the loss of remaining β-cell capacity (Shah et al., 1989).), (3) changing immune environment from producing proinflammatory (IFNγ) to anti-inflammatory (IL-10) cytokines (Alizadeh et al., 2006), and (4) suppressing recurrent β-cell autoimmunity. In support, it was found that anti-CD3 antibody could induce durable regression of recently diagnosed T1D in NOD mice by restoring self-tolerance following short-term treatment (Chatenoud 2003). According to this evidence, it is very clear that interventions in recent onset T1D have to be operational within a small and clearly defined window, which might also vary on an individual basis. In this stage patients still have sufficient remaining β-cell mass that can be assessed by C-peptide measurements(Staeva-Vieira, T. et al.2007). In order to maintain or restore critical mass of viable and functional β-cells in T1D patients for developing immune intervention, many approached are being explored. They include (1) stimulation of β-cell regeneration/proliferation in vivo through hormone and growth factor treatment, (2) direct conversion by in vivo reprogramming of adult pancreatic exocrine cells to transform into β-cells, (3) isolation and in vitro differentiation of pancreatic progenitors, embryonic stem cells, or extrapancreatic adult stem cells, and (4) induction of iPS cells from adult-differentiated cells through targeted reprogramming with soluble factors (5) islet transplantation (Guo T et al. 2009). As soon as these approaches for increasing viable and functional β-cells become readily available, immune intervention will be feasible for recent onset T1D patients and even for late stage patients.

4. Antigen-specific immunotherapy approaches

4.1. Autoantigen therapy

Antigens may behave as immunogens that provoke an adaptive immune response or as tolerogens that induce a state of specific immunological unresponsiveness (immune tolerance) to subsequent challenging doses of the antigen. Severalfactors determine the immunogenic versus tolerogenic capacity of an antigen, including its molecular form, the nature of APCs, the dose, the route of administration and local response environment (Anderton et al., 1999). Many attempts have been made to induce or restore tolerance in T1D by the use of candidate autoantigens.

Insulin

Insulin and proinsulin moleculeshave been identified to play a primary role in the initiation ofthe autoimmune process that ultimately leads to destruction of β-cells and onset of clinical T1Dfor both humans and the NOD mouse.Insulin autoantibodies usually precede the development of T1D and can beutilized to assist in disease prediction. Transplanted T cell clonesrecognizing insulin, both CD4+ and CD8+, can transferdisease to young mice or immunodeficient animals.Specific insulin peptides reacting with these clones have been identified(Gottlieb et al., 2002). Based on knowledge about insulin immunity in T1D, insulin specific tolerance induction approaches for prevention and immune intervention have been explored in T1D animal modelsand patients[See table-2].

Table 2.

Reference: Gottlieb PA, Eisenbarth GS. Insulin-specific tolerance in diabetes. Clin Immunol. 2002, 102(1):2-11

Table 2. Insulin Immunotherapy of Type 1 Diabetes [Gottlieb PA, 2002-p2]

In NOD mice, reports demonstrated that delivery of insulin (intravenously, subcutaneously, orally, or intranasally), proinsulin, or insulin peptides could delay the onset and reduce the incidence of T1D (Chatenoud 2010). Further study investigating the protection mechanism indicated that administration of insulin could induce active tolerance against T1D (Aspord et al., 2002). Insulin-derived mutated proinsulin peptide B24-C36 was shown to induce anti-diabetogenic regulatory T cells that block the adoptive transfer of T1D and protect the NOD mice from T1D (Fierabracci 2011).

Based on the success in animal models, clinical trials of oral or nasal insulin have been conducted in humans. These trials can be divided into prevention trials and intervention trials. One human prevention trial has included a double-blinded crossover safety study conducted in 38 individuals with antibodies to one or more islet antigens (insulin, GAD65, or IA-2) and the results showed intranasal insulin does not accelerate loss of β-cell function in individuals at risk for T1D and induces immune changes consistent with mucosal tolerance to insulin (Harrison et al., 2004). This is thefirst report that showed intranasal insulin was immunotherapeutic and retarded progression to clinical T1D.However, another double blind prevention trial with nasal insulin obtained contradictory findings.In this trial,224 infants and 40 siblings positive for two or more autoantibodies received short-acting human insulin or placebo once a day intranasally. Median duration of the intervention was 1.8 years. The results showed that, in children with HLA-conferred susceptibility to T1D, administration of nasal insulin immediately following detection of autoantibodies could not be shown to prevent or delay T1D. The prevention efficacy of oral insulin was also tested in 388 prediabetic patients who were first- and second-degree relatives of T1D patients and were also classified as having increased risk for developing T1D by genetic, immunological, and metabolic staging (Skyler et al., 2005, Sosenko et al., 2006). It was concluded that oral insulin did not delay or prevent T1D.However, according to subgroup analysis, it appeared that there might be a potential benefit in T1D prevention in those subjects with higher autoantibody levels. Reasons for failure may include insufficient dosing as well as the fact that by the time an individual is identified with autoantibodies, the disease process is well established and difficult to reverse.Moreover, the dose, frequency of mucosal antigen administration, and timing of antigen delivery are critical determinants of tolerance induction and subsequent suppression of T cell mediated autoimmune disease.

Intervention trials were also performed in patients with recent onset T1D with insulin. In one trial, 52 adults with recent onset T1D were given nasal insulin. It was found that nasal insulin did not retard loss of residual β-cell function, but the evidence showed that it could induce immune tolerance to insulin and it provided a rationale for its application to prevent T1D in at-risk individuals (Fourlanos et al., 2011).Two studies including about 100patients each tested the use of oral insulin at a limited doserange and they didn’t observe clinical efficacy (Chaillous et al., 2000, Pozzilli et al., 2000). A large phase II clinicaltrial with NBI 6,024 (an altered peptideligand of the 9–23 insulin B chain peptide) showed there is no efficacy of APL treatment in patientsat the stage of overt disease (Walter et al., 2009).It did not cause significant changes ininsulin requirement, metabolic control, hypoglycemic and hyperglycemic events, autoantibody concentrations, or CD4+ and CD8+ T cell numbers. The lack of response may reflect a fundamental defect in the proposed mechanism of action or inadequacy of exposure (dose), frequency,or timing of injected peptide.

GAD65

GAD65 is one of the major T1D-related autoantigens recognized by self-reactive Tcells. T cells specific for GAD65 are among the first to enter inflamed pancreatic islets and may be important for theinitiation of autoimmunity (Kaufman et al., 1993). Moreover, GAD65 is also an early target of autoantibodiesduring the initiationof T1D. Immune therapies targeting GAD65 have been tested in both animal models and in human T1D patients [See table-3].

Table 3.

GAD65 Immunotherapy of Type 1 Diabetes

In NOD mice, administration of GAD65 can prevent and block autoimmune destruction of pancreatic β-cells and subsequent need for exogenous insulin replacement. It was found thatoral administration of a fusion protein composed of cholera toxin B subunit (CTB) and GAD65 peptide could significantly reduce pancreatic islet inflammation and delay the development of T1D through generating regulatory T cells and induction of immunological tolerance (Gong et al., 2010).Intravenousinjections of GAD during the later stages of disease still effectivelyblocked disease progression in prediabetic mice and protectedsyngeneic islet graft survival in diabetic NOD mice (Tian et al., 1996). The identification of CD4+ Treg cells in GAD-treatedmice suggests a major role for bystander suppression in theinduction of tolerance by treatment with this autoantigen (Tisch et al., 1998).

Based on the above convincing data from NOD mice, autoantigen immunotherapy with GAD65 was tried in human.GAD-Alum (Diamyd Therapeutics) is an adjuvant-formulated vaccine incorporating recombinant human GAD65 which is the specific isoform of GAD expressed in human pancreatic β-cells and a major antigen targeted by autoreactive T lymphocytes in T1D. It was shown to be safe and to preserve residual insulin secretion in patients with late onset T1D of adulthood (Agardh et al., 2005). A subsequent phase II trial in recent onset T1D showed significant preservation of residual insulin secretion and a GAD-specific immune response, both humoral and cell mediated immune response (Agardh et al., 2009). Moreover, Alum-formulated GAD65 has been given subcutaneously in two injections with one month apart to recent onset T1D patients with positive GAD65 autoantibodies. The injections were found to preserve residual β-cell function without treatment related serious adverse events (Agardh et al., 2005).Further study showed that two injections of GAD-Alum to children and adolescents with T1D could induce GAD specific CD4+CD25high forkhead box P3 (Foxp3)+ regulatory T cells and secretion of Th2 and regulatory cytokines (Hjorth et al., 2011).These data support a long-lasting immunomodulatory effect of GAD-Alum treatment.Follow-up after 5 years completed in2008 still show a significantly beneficial effect of the Diamyd (Morales et al., 2011). According to these promising clinical data, Phase III studies in children with recent onset T1D are ongoing along with a study (DIAPREV-IT) aimed at testing whether Diamyd^® may prevent the clinical onset of T1D in non-diabetic children with GAD65 autoantibodies and at least one more islet autoantibody (Larsson et al., 2011).Also, Diamyd Medical is currently conducting two clinical Phase III studies on T1D: one in Europe and one in the United States. Here,the formulation is crucial to Dyamid's GAD drug because 1) adjuvant reduces the required quantity of antigen by maximizing its immunogenicity and 2) aluminum salts preferentially induce a humoral rather than cellular immune response. Immune readouts show an increase in FoxP3 and transforming growth factor-β in cells from GAD-Alum-treated patients compared with placebo after 15 months(Ludvigsson 2009). In summary, GAD 65 vaccination hasshown encouraging results with preservation of residualinsulin secretion in 10- to 18-year-old type 1 diabeticpatients with recent onset. In patients with short T1D duration, the effect was quite pronounced. These effectswere reached with a very well tolerated treatment. Future studieswill show if the good effect seen so far can be confirmed.There is a hope that GAD vaccination will cause remission or even cure and prevention of T1D seem to be arealistic possibility (Morales et al., 2011).

DiaPep277

HSP60 is a ubiquitous protein that is part of a highly conserved family of intracellular chaperones and is also located in the mitochondria and mature insulin–secretory granules of pancreatic β-cells with an important regulatory role in the innate immune system and considered an important autoantigen in T1D(Birk et al., 1996). Although HSP60 autoantibodies are not yet considered a diagnostic tool for T1D, the evidence suggests this autoantigen may be important in disease development. Therefore, vaccine strategies that are based on HSP60 as a T1D autoantigen have been tested in mouse models and human T1D patients [See table-4].

Table 4.

DiaPep277 Immunotherapy of Type 1 Diabetes

DiaPep277 is a 24 amino acid synthetic peptide derived from the C 31 terminus of the human HSP60.Compared with HSP60, DiaPep277 activates T cell TLR2 receptors but has no effect on macrophage TLR4 receptors.So, DiaPep277 only mediated anti-inflammatory effect on T cells and lacked pro-inflammatory effect on innate immune cells. In NOD mice, nasal administration of a diabetogenic peptide Pep277 showed significant anti-inflammatory immune response (Liang et al., 2010). Moreover, neonatal oral administration of DiaPep277combined with hydrolysed casein dietcould protect against T1D in BB-DP rats. All the results showed that DiaPep277 could significantly prevent T1Ddevelopment and arrest the progression of insulitis in mice that have already become hyperglycemic (Fierabracci 2011). These promising experimental results led to clinical trials with DiaPep277. It was found that DiaPep277 treatmentfor patients with recent onset T1Dcouldprevent further β-cell loss (Raz et al., 2001, Raz et al., 2007). Specific interventionwith DiaPep277 re-establishes the balance in theimmune system by enhancing the Th2-type cytokine productionand inducing a strong signal for the developmentof Th2 cells. Phase II clinical trials in adult subjects withT1D have shown suggestive evidence of better preservationof C-peptide and cessation of destruction of β-cellswith DiaPep277 (Raz et al., 2001). Moreover, the adult trials showed significantlybetter preservation of insulin synthesis as measured byC-peptide production in the treated groups compared withplacebo, but this effect was not seen in the pediatric trial. Similarresults were observed in one other trial performed in pediatricpatients (Schloot et al., 2007). Diapep277 was more pronounced in patients with a high β-cell reserve at the start of the treatment. The phase III study has began in 40 medical centers worldwide and the results are expected in 2011 (Eldor et al., 2009).If DiaPep277 proves to be efficacious, it will causea paradigm shift in the treatment of T1D from treating thesubsequent insulin deficiency to addressing the initial autoimmune processthat is at the core of the disease (Eldor et al., 2009).

In summary, efficacy of autoantigen therapy has been disappointing.Based on preclinical models, several factors are emerging as critical in determining the outcome of Autoantigen immunization most notably dose, route, adjuvant, and frequency of administration. Studies have shown that too frequent antigen administrations, as well as very high dosages, do not result in optimal induction of immune regulation and tolerance(Peakman M et al. 2010).Another challenge for developing effective autoantigen therapy is to know the role that β-cell autoantigens play in islet inflammation and restoration of immunological tolerance, for example, how DCs recognize autoantigens and how autoantigen exposed DCs guide T cell development into immunosuppressive or autoreactive T cell subsets. The other important questions that need tobe answered include precisely which autoantigen epitopesshould be targeted and whether the use of a single antigenis sufficient in order to induce immune tolerance andstop an ongoing T cell response to multiple autoantigens. Moreover, only blood samples are readily available from patients to evaluate efficiency of antigen immunotherapy. The question is if peripheral T cell populations reflect the T cell population infiltratingthe target organ. Another important concern is if immune deviation and functional dominance of regulatory responses arereflected by changes in peripheral antigen-specific T cells. More-sensitive assays are necessary todevelopto evaluate the impact of immunotherapy ondisease progression through analysis of peripheral cells.All these issues need to be addressed by performing autoantigen specific immunotherapy studies in future clinical trials.

4.2. Dendritic cell (DC)-based immunotherapy

4.2.1. Brief overview of DC biology

Dendritic cells (DCs) are a special subset of leukocytes able to alert the immune system of the presence of infections andare responsible for the activation and the control of adaptive immune responses. There is also mounting evidence that DCs establishand maintain immunological tolerance. Indeed, DCs can prevent, inhibit, or modulate T cellmediated effector responses through a variety of mechanismsincluding T cell anergy, T cell deletion, immune deviation (i.e. polarization of T cell cytokine profiles), and the expansion or induction of Tregs.The induction/expansion of FoxP3+ Tregs is one ofthe major mechanisms by which DCs maintain immune tolerance (Steinman et al., 2003).

DCs are a heterogeneous mix of the distinct leukocyte subsets that have different functions and homing patterns. In humans, there are two major and intrinsically different subpopulations of DCs:myeloid DCs (myDCs) and plasmacytoid DCs (pDCs). These two DC subsets recognize different microbial pathogens by expressing distinct repertoires of toll-like receptors (TLR) and induce different types of innate and adaptive immune responses depending on environmental factors. Much evidencesupportsthepresence of DCs with tolerogenic properties in both myDCs and pDC subsets (Gregori 2010). Both myDC and pDCsdevelop from hematopoietic bone marrow progenitor cells that turn into DC precursors (pre-pDCs) upon encounter of Flt3-ligand (Flt-3L). Pre-DCs migrate fromthe bone marrow to the bloodstream where theycirculate as immature DCs characterized by low expression ofHLA class II and co-stimulatorymolecules, high endocytic activity, and low T cell activationpotential. Immature myDCs and pDCs encounter pathogens and become activated mature DCs to prime CD4+ and CD8+ T cells (Liu et al., 2010). In addition to their potential to become activated DCs, immature myDCs and pDCs in the steady state are also criticallyinvolved in homeostasis and in promoting and maintainingperipheral tolerance, primarily via the induction of CD4+ Tregs. Of note, the DCs with tolerogenic function not only compriseimmature DCs, but also include other DCs covering a spectrum of different maturation states. The induced CD4+ Tregs include the naturally occurring Tregs (nTregs) and the adaptive Tregs.Thymic-derived nTregs are defined by the constitutivehigh expression of FoxP3 (Sakaguchi et al., 2010). Adaptive Tregs composed of different Treg subsets are generated in the peripheryin tolerogenic microenvironments and suppress T cellresponses via immunomodulatory cytokines. The best characterized are the IL-10-producing type 1 regulatoryT (Treg1) cells that depend on IL-10 for their generation and functions (Allan et al., 2008).

According to current understanding of DC development, protocols have been set up for in vitro propagationof large numbers ofactivated and tolerogenic DCs using defined growthfactors as well as biological, genetic, and pharmacological modifications (Morelli et al., 2007).Depending on the agent used for tolerogenic DC induction,the resulting DCs are equipped with defined tolerogenicmolecules which determine their ability to promote differentsubsets of Treg cells.Tolerogenic DCs secreting high levels ofIL-10 in the absence of IL-12 are primarily involved in theinduction of adaptive IL-10-producing Treg1 cells.FoxP3-inducer DCs express low levels of co-stimulatory molecules and express indoleamine 2,3-dioxygenase which allows them to secrete TGF-β and retinoic acid and to promote FoxP3+ Treg cells (Gregori 2010). According to these prerequisites, it can be postulated that depending on the type of Tregs required for promoting tolerance toward a given antigen, different protocols for in vitro generation of tolerogenic myDC with the desired phenotypic and functional characteristics could be designed.In general, tolerogenicpotential of DCs were determined by low-expressionof co-stimulatory molecules,microenvironmental factors (in particular, immunosuppressivecytokines),expression of death-inducing ligands, (in particular Fas ligand(FasL)), and inhibition of gene transcriptionregulatory proteins (e.g. NF-kB)(Morelli et al., 2007). The development of culture methods for generating large numbersof DCs that can be pharmacologically or biologicallymanipulatedopens important new perspectives for the developmentof DC-based cell therapy protocols.

4.2.2. DC-based therapeutic approach

DCs are critical for both preventing and perpetuating autoimmunity,reflecting their dual roles in regulating the immune system, i.e., maintenance of T cell tolerance and induction of pro-inflammatory immuneresponses. The defects in immune regulation by DCs may play a central role in T1D (Anderson et al., 2005). In T1D, abnormalities in DC phenotype and function in both NOD mice and humans mayresult in the skewing response towardpathogenic Th1 cells (Feili-Hariri et al., 2006). The purpose of using tolerogenic DC therapy for T1D is to re-balance dysregulated immune responses that have alreadybeen ongoing for some time.A number ofstudies have shown that therapy with various forms of DCs can protect mice from T1D development. For example, DCs generated from the bonemarrow of NOD mice by culturing DCs in granulocyte–macrophage colony-stimulating factor (GM-CSF), IL-4, andfetal bovine serum (FBS) could prevent T1D in somerecipients when administered to young (5-week-old) NOD mice (Feili-Hariri et al., 1999).Transfer of DCsex vivo stimulated with IFN-γ down-modulates T1D in NOD mice (Shinomiya et al., 1999).However, one study showed that only DCs from pancreatic lymph nodes (panLNs), but not DCs from other anatomical sites, could protect from disease, presumably becausepanLN-derived DCs would present β-cell-derived antigensafter adoptive transfer (Clare-Salzler et al., 1992). This study indicated that autoantigen exposed DCs have better protection effects.

4.2.2.1.Exposure of therapeutic DCs to antigen ex vivo

The original observation that DCs from the panLNs could prevent T1D when transferred toNOD mice, while those from other sites could not, suggests the potential importance of the incorporation of β-cell antigens into DC-based therapeutics for thisdisease (Mukherjeeet al. 2010). The von Herrath group explored this approach in a rat model (Haase et al., 2005). They utilized the rat insulin promoter (RIP)-LCMV model of T1D in which disease is induced upon LCMV infection. BM-DCs were generated and pulsed with a viral peptide recognized by CD8+ T cells. It was found that theantigen pulsed DCs have stronger ability to inhibit development of T1D than the unpulsed DCs. This study suggests that disease relevant antigen exposed tolerogenic DCs have therapeutic benefit for T1D.Next, autoantigen pulsed DCs were explored for treating T1D in NOD mice. It was found that adoptive transfer of insulin peptide- pulsed, immature DCs could protect NOD mice from T1D whereas unpulsed DCs had no effect on T1D development (Haase et al., 2010). In another study, immature DCs derived from BM were pulsed with antigen-specific apoptoticbodies from the β-cell line NIT-1 cells and the administration of these cells to NOD could decreaseT1D incidence significantlyand correlated positively with insulitis reduction (Marin-Gallen et al., 2010).All these results show that DC therapy is capable of preventingT1D in an antigen-specific manner in animal model and DCs expressing disease-associated antigens constitutes a promising strategy to prevent T1D. The mechanism for preventing and delaying development of T1D with autoantigen exposed DCs is inducing expansion of regulatory T cells (Cheatem et al., 2009).

4.2.2.2. Targeting of antigens to DCs in vivo

Loading antigen ex vivo to DCs is challenging and faces many problemsincluding labor-intensive procedure for each patient, high cost, and limited DC number (Tacken et al., 2007). Thus, many studies have begun to explore the utility of in vivo delivery of β-cell antigens to DCs for the prevention and treatment of T1D.

Many receptors used in targeting DC studies belong to theC-type lectin receptor (CLR) family, a family of calcium-dependent lectins that shareprimary structural homology in their carbohydraterecognitiondomain (CRD). Through their CRD, theCLRs bind to specific self or non-self sugar residues.Many of these lectins were shown to be associated with antigen uptake by DCs.Approachesto targeting CLRs fall into two categories: based on the binding of natural receptor ligands andexploited antibodies against the receptor. CD205 belongs to type I CLR family and study indicated that itis a suitable target for in vivo antigen-targeting studies. CD205 recycles through late endosomal or lysosomal compartments and mediates antigen presentation. (Tacken et al. 2007).

Much evidence shows that DEC-205 is involved in antigen processing and presentation.In mice, DEC205 expression is relatively DC restricted: it is highly expressed by mature DCs (mDCs), thymic epithelium, at low levelsby B cells, and at very low levels by T cells and granulocytes.Therefore, DEC205 is an excellent target to studyin vivo DC targeting. The therapeutic potential of DEC-205-mediated antigendelivery has begun to be explored in mouse models of T1D (Mukhopadhaya et al., 2008, Bruder et al., 2005).In Mukhopadhaya A et al’s study, autoantigen peptide-linked anti-DEC-205could lead to deletion of autoreactiveCD8+ T cells even in the context of ongoing autoimmunity in NOD mice and induce subsequent immune tolerance to β-cells. Thisstudy provides support for the development of DC targeting of self antigens fortreatment of chronic T cell-mediated autoimmune diseases. Bruder D et al.utilizedtheINS-HA/TCR-HA transgenic mouse model for T1D.It was found that T1D was prevented in most animals when treated with HA peptidelinkedanti-DEC-205. Further analysis showed that these HA-specific CD4+ T cells from anti–DEC-HA–treated mice exhibited increased expression of FoxP3, cytotoxic T-lymphocyte–associated antigen-4, and the immunosuppressive cytokines interleukin-10 and TGF-β. The findings indicate that targeting of the HA antigen to immature DCs in vivo leads to a relative increase of antigen-specific FoxP3⁺regulatory T cells that suppress the development of T1D. Anotherrecent related study showed thattargeting a mimotope peptide to the endocytic receptor DEC-205 on DCs in NOD mice induces efficient conversion of pancreatic β-cell-reactive BDC2.5 CD4+ T cells into long-lived FoxP3+ Treg cells (Petzold et al., 2010). These results suggest that promoting antigen-specific Treg cells in vivo might be a feasible approach in T1D with DC in vivo targeting.

4.2.3.Future challenges for DC based approaches

4.2.3.1. Challenge for exposure of DCs to antigen ex vivo

Adoptive cellular therapy with tolerogenic DCshas potential applications not only in human autoimmune diseasesbut also forallogeneic tissue transplantation and hypersensitivity. One big challenge is to generatetolerogenic DCswith a stable phenotype which are resistantto maturation mediated by pro-inflammatory mediators.

Currently, DCs for clinical use are generated from peripheral blood monocytes of the patients. However, the number of monocytes obtained from the patients is limited and the potential of monocytes to differentiate into DCsvaries depending on the blood donor. Thus, the issue of limited cells is a serious obstacle for DC therapy. ES cells and iPS cells have pluripotency and unlimited propagation capacity and may be an ideal cell source for DC-therapy. Several groups have developed methods to generate DCsfrom ES cells or iPS cells (Menendez et al., 2005).

Developing tolerogenic DC-based therapies also requires optimizing the following conditions. (1) To define the best source of DC precursorsfor the in vitro induction of tolerogenic DCs, i.e. peripheralblood monocytes vs bone marrow CD34+ cells. Studies performedin non-human primates indicate that bone marrow cellsare a good source of tolerogenic DCs(Steinman et al., 2003). (2) To determine the disease stage, route of delivery, and dose(Lo J et al., 2006) of DC transfer.Because the autoimmune response leading to TID in humans takes place for several years before hyperglycemiadevelops, it may be important to determine a critical stage for initiation of DC-based therapy. The study from NOD mice indicates that prevention initialized at early stage has a positive effect. Future studies are needed to establish whether DC-based therapies block disease onset at advancedstages of islet destruction or even reverse acute onset T1D.For the route of DCs administration, until now, there is no solid evidence to showwhich administration route is the best.It was noted DCs have different migration in different administration route. For example,subcutaneous injected DCshomemainly to the draining lymph nodes whereas intravenousadministration mainly goes to the spleen, lungs, and liver. Concerning the injection dose, the effect of the number per treatment hasnot been comparatively evaluated.Of note, the lifespan of DCsin vivois limited to about 2-3 weeks. Therefore, it could be envisagedto perform repetitive injections unless a single injectionwould be sufficient to generate a microenvironment suitablefor sustaining de novo in vivo induction of tolerogenic DCs (Morelli et al., 2007).

4.2.3.2. Challenge for targeting DC in vivo

When developing DEC-205-mediated therapeutic strategies for T1D, the choice of antigen is a major consideration. Until now, the antibody can be modified to include an antigen and it proves difficult to combine multiple antigens. As mentioned before,multiple antigens are targeted by Tcells in both NOD mice and T1D patients (DiLorenzo et al., 2002).Particularly in humans, it is unclear which of these are themost ‘important’, i.e. critical for disease initiation and/orprogression.However, accumulating evidence suggests thatinsulin and GAD65 are key targets of pathogenic T cells in T1D in both NOD mice and humans and that they may be candidates for targeting to DEC205 in T1D.

DCs have two functions, maintaining tolerance to self and inducing immunity. In the context of activation stimuli such as those found during inflammation, infection, or tissue destruction, DCs induce prolonged T cell activation and in steady state, they induce tolerance. This character of DCs was also shown in the Hawiger D study. It was found that HEL linked anti-DEC 205 could induce immune tolerance under physiological conditions but induce prolonged T cell activation and immunity against HEL when performed in conjunction with an agonistic anti-CD40 antibody. (Hawiger et al., 2001). It indicatedantigendelivery to DCs in the context of an inflammatory environment may lead to exacerbation of a pathogenic autoimmuneresponse rather than tolerance induction. So, when considering the translation ofsuch a strategy to humans, the potential remedy should be considered. (Mukherjeeet al. 2010)

Another potential concern is that CD205 expression in humans is less DC restricted than in mice, and targeting constructs might also be endocytosed by several other cell types for example B cells, T cells, monocytes, macrophages, and natural killer (NK) cells (Erbacher et al., 2009). It still needs to be confirmed if targeting these cells has any side effects. Onthe whole, many issues need to be resolved before translating DC based approachesinto human T1D prevention and treatment.

4.3. Adoptive transfer ofTregs

4.3.1. Brief overview ofTregs

Tregs play a crucial role in the maintenance of immune homeostasis. The two best studied types of CD4+ regulatory T cells are the FoxP3+ Tregs and Th1 cells secreting IL-10. Large quantities of evidence show that CD4+ regulatory T cells play a protective role in autoimmune disease (Spoerl et al., 2011).

CD4+FoxP3+ Tregs account for approximately 5-10% of the total CD4+ T cells in the periphery, most prominently in the lymphoid organs such as the spleen and lymph nodes, with a small subset found in the non-lymphoid tissues. FoxP3+ Tregs suppress both innate and adaptive immune cells through several mechanisms and maintain tolerance and immune homeostasis.They exert suppression by cell contact-dependent mechanisms (for example, functional modulation by means of CD39, CD73, and LAG-3, or killing of APCs or responder T cells by means of granzyme and perforin) as well as mechanisms mediated by soluble factors (for example, secretion of immunosuppressive cytokines such as IL-10, TGF-β, IL-35, and galectin-1), or deprivation of cytokines (for example, IL-2) necessary for the expansion and/or survival of responder T cells (Zhang et al., 2009). Moreover, Tregs have two key characters that are very important for their translational application; (1) Bystander suppression:When Tregs are first activated through their respective TCRs to inducesuppressive activity, but once activated, the cells are capable ofsuppressing responses in a non-specific manner. (2) Infectious tolerance:The transferred Tregs generate animmunoregulatory milieu that persists long after the removalof the eliciting cell population.This occurs as new regulatory cell populations are activated by antigen in the context of tolerogenic APCs and Tregs. These two characteristics allow Tregs of single or limited-antigen specificity to establish a broad and stable immunoregulatory effect and constitute the quintessential Treg phenotype (Brusko et al., 2008b). Phenotypically, CD4+FoxP3+ Tregs express multiple cell surface markers such as CD25, CTLA-4, GITR (glucocorticoid-induced tumor necrosis factor receptor), OX40, CD103, CD39, and CD73, many of which are also expressed by activated CD4+ T effector cells (Kang et al., 2007). Thus, these markers are not Treg specific.

CD4+FoxP3+ Tregs are not a uniform cell population, but consisted of those that developed in the thymus (nTregs) and those that are derived from T effector cells in the periphery (adaptive or induced Tregs, iTregs). nTregs are shown to arise asa committed lineage in the thymusand bear a diverse TCRrepertoire against a broad range of self antigens.iTreg cells are induced under particular conditions of antigenic stimulation both in vitro and in vivo.Furthermore, iTregs are particularly unstable and tend to lose FoxP3 expression more easily than nTregs, which is likely related to epigenetic differences in the FoxP3 gene in both subsets. Indeed, certain non-coding regions in the FoxP3 gene in nTregs have been shown to be completely demethylated while such regions are often methylated in iTregs (Curotto de Lafaille et al., 2009). Such differences may be important in therapeutic manipulation of nTregs and iTregs (Spoerl et al., 2011).

Another important development about Tregs is that many protocols were developed for isolating and expanding nTregs for therapeutic applications from healthy individuals and patients (Brusko et al., 2008a). This provides important tools and principles for translating Treg therapies into clinical treatments for patients with T1D and other autoimmune disorders.

4.3.2. Tregs in the pathophysiology of T1D

A growing body of evidence suggests that an imbalance inthe immune system plays a major role in the pathogenesisof autoimmune disease (Brusko et al., 2008b). Tregs playa major role in controlling the activity of self-aggressiveT cells that were not deleted in the thymus and are one of the key factors formaintaining homeostatic balance in the immune system (Chatenoud et al., 2001). Defects in the number andfunction of Tregsand a resistance of effectorT cells to Treg-mediated suppression could each contributeto failed T cell regulation. Each of these defectscontributes to the development ofautoimmune disease including T1D (Buckner 2010).

A role for nTreg cells in T1D is evident in individuals with IPEX, in which the absence of nTreg cells results in enhanced susceptibility to T1D. In NOD mice, reducedCD4+ Treg cell frequencies or function represent a primary predisposing factor to T1D (Sgouroudis et al., 2009). The adoptive transfer of Tregs could suppress disease through reducing effector Th1 T cells and macrophages and inhibiting effector T cell cytokine and chemokine production (Tonkin et al., 2009).However, Tritt M et al. have contradicting reports as they couldn’t detect the defect of cellular frequency of FoxP3+nTreg cells in the lymphoid tissues. However, the nTreg cell functional potency and intra-pancreatic proliferative potentialdeclines with age, in turn augmenting diabetogenic responses and disease susceptibility (Tritt et al., 2008). On the whole, strong evidence shows that complete absence of Tregs precipitates T1D and the diseasecan be prevented by the adoptive transfer of additional Tregs (Sgouroudis et al., 2009).

Most studies investigating the role of Tregs inhuman T1D showed contradictory results in the peripheral bloodfrequency and suppression function of these cells between T1D and control subjectswhen the expression of transcription factor FoxP3 was used for their identification (Alonso et al., 2009).One reason is that FoxP3 is an unreliable markerfor Tregs in humans, as it is also expressed in activated effector Tcells. Moreover, unlike mice, the frequency of Tregs in islet couldn’t be obtained from human. The results obtained from blood samples might not indicate the involvement of Tregs in the pathophysiology of T1D.Moreover, the method used to isolate Tregsmay influence the degree of suppression observed in vitro. This effectmay also account for some of the discrepancies in the reportedfunctions of these cells in T1D (Brusko et al., 2008b). The above data from mice and human indicated central importance of Tregs inthepathogenesis and potentially the treatmentof T1D.

4.3.3. Therapy of T1D withTregs

In the mouseanimal model, the adoptive transfer of CD4+ CD25+ Treg cells has been shown to protect from T1D (Salomon et al., 2000, Tonkin et al., 2008, Tonkin et al., 2009). According to published results, the adoptive Treg cells include polyspecific and autoantigen specific. Now the evidence is very clear that antigen-specific Tregs are more effective in several autoimmune syndromes when compared with polyclonal populations (Brusko et al., 2008b). According to a study by Wu AJ, polyspecific Tregs hadto be transferred regularly (twice weekly) in high doses andmice developed T1D after Treg transfer had been stopped (Wu et al., 2002). This makes this approach difficult to be performed in human patients.The other concerns about polyspecific Tregs include the efficacy and safety of these Tregs related to their polyspecificity and a paucity of information about their stability, function, and fate in vivo. So, autoantigen-specific Tregs would have more advantages interms of safety and efficacy because they could specifically recognize antigens from thetarget tissue to achieve tissue specific immunoregulation. Several studiesalready showed natural CD4+ CD25+ Tregs from the NOD mouse specific for an islet antigen were more protective than their polyspecific Tregs(Tang et al., 2004, Tarbell et al., 2004, Tarbell et al., 2007). Compared with poly specific Tregs, relativelysmall numbers of islet autoantigen-specific Tregsprevent T1D in NOD mice. Moreover, it was found that adoptive transfer of islet specific Tregs showed considerable efficacy in ameliorating ongoing T1D in NOD mice (Tarbell et al., 2007). Until now, not many treatments were shown to interfere withdisease pathophysiology at later disease stages in the NOD mouse. Therefore, adoptive autoantigen specific regulatory T cellsshould have great potential therapeutic benefit in humans.

For developing adoptive autoantigen specific regulatory T celltherapy, in vitro expansion of this unique cell population becomes very important. Several research groups already developed the expansion protocol of autoantigen specific Tregs from peripheral blood in vitro (Dromey et al., 2011, Tang et al., 2004, Tarbell et al., 2004).Because proinsulin and GAD65 usually aretargets of autoreactive T cells in human T1D(Harrison 2008).Tregs specific for these autoantigens are being attempted to expand in vitro and they will have potential application for adoptiveimmunotherapy of T1D.

4.3.4. Future challenges forTregtherapy

Translating Tregs into the clinical setting still faces several challenges. Firstly, until now, no unique cell surface markers were identified for human Tregs. Because activated effector T cells also express CD25, there exits high frequency of activated effector T cells contamination present in the CD4+ CD25+ fraction with present Treg isolating kits.So, for the isolation of human Tregs, the combination of the surface markers CD4+ CD25highCD127low results in a populationwhich is enriched in Tregs with fewer recently activatedeffector T cells (Brusko et al., 2008b).Second, another major challenge in antigen-specific Treg therapy lies in the isolation of sufficient quantities of antigen-specific T cells. In order to obtain enough cell number for adoptive transfer, in vitro expansion of this population is necessary. Although several groups already developed protocolsfor expanding autoantigen specific Treg cells, the current technology may notprovide a robust method for isolating and expanding a sufficientquantity of these cells for use in the clinic. The next most important factor will be the stability of the Treg differentiation phenotype.Recent animal studies provide evidence for functional heterogeneity and lineage plasticity within the Treg compartment (Addey et al., 2011). In the healthy immune system, the majority of natural Tregs are relatively stable, but the study showed that 10-15% of “stable” Treg cells were foundto lose FoxP3 expression after adoptive transfer into lymphopenic hosts (Zhou et al., 2009). There are twofates for Tregs that lose FoxP3 in lymphopenic hosts: death or de-differentiation. One study showed that half of the Tregs transferred into lymphopenic hosts did not die but ratherbegan producing IL-2 and IFNγ and became effector T cells with the same antigen specificity(Komatsu et al., 2009).These reverted autoreactive effector cells will aggravate the autoimmune damage.

The studies described above suggest a critical challenge inthe transition of Treg-based cell therapies from animal studiesto human clinical trials. Knowledge of the purity, stability, andphenotypic characteristics of cell therapy products will be essentialprior to their introduction into patients.

5. The importance of combination therapy

After several attempts to develop potent immunotherapy to cure T1D, the results are disappointing.Because of the multifaceted nature of this disease, a monotherapy will not be sufficient to combat T1D.The best approach is to use a combination therapy. Two major goals have to be accomplished: (1) the rapid blockade of the immune system and dampening of any auto-reactive responses without strong side-effects and (2) the regeneration of a critical β-cell mass in order to maintain euglycemia without repetitive insulin injections (Bresson et al., 2007). Each of these goals can be achieved by different means, but the combination of different therapies is very necessary for successful prevention and treatment for T1D.It should be noted that the ideal combination therapy would utilize two or more agents whose mechanisms of action are complementary and that have already been demonstrated as safe T1D monotherapies in humans.

5.1. Combination of anti-CD3 Ab with antigen-specific approach

The first and key step for treatment of T1D is dampening the auto-reactive responses and maintaining permanent tolerance to islet cells. In order to obtain this goal, the combination of immunotherapy approaches including nonspecific immunoregulatory and antigen-specific tolerancestrategies are necessary. It was found that efficacy of antigen-specific therapy can be greatly enhanced when combined with anti-CD3 in recent onset T1D.A previous study (Bresson et al., 2006) showed that combination treatment with anti-CD3 specific antibody and intranasal administration of proinsulinpeptide can reverse recent onset T1D inmurine T1D models with muchhigher efficacy than with monotherapy using anti-CD3 or antigen alone. Further in vivo analysis showed that expansion of insulin-specific CD25+Foxp3+ Tregs producing IL-10, TGF-β, and IL-4 was strongly enhanced. These cells could transferdominant tolerance to immunocompetent recent onset T1D recipients and suppressed autoaggressive CD8 responses.Moreover, the combination using anti-CD3 antibody and GAD65-expressing plasmid was shown to reverse recent onset T1D in RIP-LCMV-GP mice models (Bresson et al., 2010).Thus, the efficacy of antigen-specific therapy to induceTregs and long-term tolerance can be greatly enhanced by addition of anti-CD3 treatment. The potential reasons include:(1) It was found that Tregs were not able to suppress activated autoaggressive T cells within the islets and PLNs (Nagler-Anderson et al., 2004; Tarbell et al., 2004). That is thereason why antigen-specific therapies were only effective duringthe prediabetic phase in animal models. Anti-CD3 antibody could directly decrease numbers of autoaggressive T cells in mice as described in the study (Herold et al., 2002) and create a systemic immunomodulatorymilieu to facilitate the islet antigen–specific inductionof Tregs. (2) Anti-CD3 has been shown in some settings to directly promote the generation of Tregs. In NOD studies by Chatenoud and others, anti-CD3 increased the number of CD4+CD25+ cells as well as systemic TGF-βproduction,which was required for therapeutic efficacy (Chatenoud et al., 2007). According to these results, an ideal combination therapy will consist of an islet antigen-specific immune tolerance induction, such as autoantigen immunization, coupled with a systemic drug that dampens islet destructive immunity but does not affect Tregs. Anti-CD3 treatment is suggested as a good candidatefor this.In summary, the combination therapies with anti-CD3 Ab and antigen specific tolerance induction constitutes a great hope towards the development of safer andmore effective immuno-interventions and would be excellent candidates for clinical studies in humans.

5.2. Combination of immunotherapy and β-cell regeneration

Since β-cell mass decreases over time, patients with long-standing T1D are very likely to have lost most of their insulin producingcells. Thus,treatments that only target the autoreactive immunesystem would not obviate the need for insulin injections in thesepatients. So, improving β-cell function at the time of diagnosisis an important determinant of response to treatment with immunetherapy, and therefore strategies that improve β-cell massand/or function may have a positive effect on response to immunetreatment. Much evidence showed that reduced β-cell mass can eventually be restored endogenously. It has beendemonstrated that β-cellscan replicate, differentiate, or transdifferentiatefrom various endocrine or non-endocrine cells. Specifically, certain peptidesand growth factors such as gastrin andglucagon-like peptide 1 (GLP1) can increaseβ-cell mass and restore normoglycaemiain animal models of T1D in the absence ofimmunosuppressants (Suarez-Pinzon, WL et al.2008). So, it is necessary and feasible to develop combination therapyincluding improving β-cell regeneration and immunotherapy. Sherry NA et al. reported the efficiency of such combination therapy. They found that treatment with glucagon-like peptide-1 receptoragonist enhances remission of T1D in NOD mice treated withanti-CD3 Abby enhancing the recovery of the residual islets (Sherry et al., 2007). Also, it was demonstrated that exendin-4 synergistically augmentsthe remission-inducing effect of anti-lymphocyte serum in NOD mice (Ogawa et al., 2004).These results suggested that this combinatorial approach may be useful in treatment of patientswith recent onset T1D. Accordingly, a new phase II study about combination of regenerative agents (lansoprazole and sitagliptin) and Diamydare recruiting patients. Currently, this is the only clinical trial combining a treatment to control the autoimmune attack with autoantigen vaccine and a treatment to help the body generate more insulin. It isbelieved that combinationslike this represent a very promising approach to the futurecure for T1D

6. Conclusion remark

Antigen specific tolerance induction is the targeted therapy for prevention or early reversal of T1D. Ideally the tolerance therapy would specifically target β-cell antigens and create a local tolerance environment for inhibiting ongoing autoimmune attack against β-cells and contributing to β-cell regeneration.

Development of safe and effective prevention of T1D is becoming a major worldwide public health goal. While many preventive modalities have succeeded in animal models of T1D, prevention of human T1D remains elusive. With further understanding of genetic and environmental factors that determine the relapsing-remitting course of β-cell destruction and autoimmune mechanisms of β-cell destruction, primary prevention with antigen specific immune tolerance induction will likely be the optimal approach for the prevention of T1D in high-risk groups.

As patients develop autoimmunity, β-cell function declines. At this stage, efficacy of antigen-specific therapies is limited.The experience accumulated from previous clinical trials indicated that combination therapies may enhance efficacy while lowering risk. It is likely that one day combination therapy including systemic immune modulators, antigen specific immune tolerance induction, and a compound with positive effectson β-cell proliferation will become the standard of care for newly diagnosed T1D.