Review
Understanding Type 1 Diabetes: Etiology and Models

https://doi.org/10.1016/j.jcjd.2013.05.001Get rights and content

Abstract

Type 1 diabetes is a complex disease involving a combination of factors, such as genetic susceptibility, immunologic dysregulation and exposure to environmental triggers. Animal models serve an important function both in elucidating the pathophysiology and preliminary screening of antidiabetic molecules. Hence, the development of models for type 1 diabetes can be broadly divided into 3 categories, namely: identification of spontaneously developing type 1 diabetes mellitus strains, creating diabetes-prone species through gene transfer techniques and forced destruction of islet cells through chemical or surgical means. This review discusses the models used to study type 1 diabetes with special emphasis on genetics.

Résumé

Le diabète de type 1 est une maladie complexe se caractérisant par une combinaison de facteurs tels que la prédisposition génétique, la dysrégulation du système immunitaire et l’exposition aux facteurs environnementaux déclencheurs. Les modèles animaux jouent un rôle important tant dans l’élucidation de la physiopathologie que dans le dépistage préliminaire des molécules antidiabétiques. En conséquence, l’élaboration de modèles liés au diabète de type 1 peut généralement être divisée en 3 catégories, à savoir l’identification des souches de diabète sucré de type 1 apparaissant spontanément, la création des espèces sujettes au diabète par les techniques de transfert génétique et la destruction forcée des îlots de Langerhans par des moyens chimiques ou chirurgicaux. Cette revue discute des modèles utilisés pour étudier le diabète de type 1, en accordant une attention particulière à la génétique.

Introduction

The past few decades have seen an unprecedented increase in the occurrence of diabetes mellitus throughout the world 1, 2. Broadly, the disease can be classified into 2 major types: type 1 and type 2. Research has been able to shed light on the multifactorial nature of type 2 diabetes, which is believed to be influenced by genetic, environmental and lifestyle factors (3). However, the etiology of type 1 diabetes mellitus has not yet been elucidated fully.

Numerous studies have been performed to identify the causative factors involved in the disease 4, 5. Although a complete picture has not yet emerged, certain genes that play a significant role have been identified and mapped. On the basis of this knowledge, type 1 diabetes mellitus is now considered an autoimmune disease involving genetic, immunologic and environmental factors. Nevertheless, the ultimate aim of diabetic research is to find a cure for diabetes or at least find moieties that would be able to prevent the symptoms of the disease. The preliminary screening for antidiabetic molecules usually is performed in animal models; however, because of the complicated etiology, finding the right animal model for type 1 diabetes mellitus remains elusive. Extensive research in this field for the past 4 decades has resulted in a number of animal models suitable for type 1 diabetes mellitus, but a perfect model has yet to be found. The development of models for type 1 diabetes mellitus is still a high-priority research area. In this article, we have explored the existing animal models used for type 1 diabetes mellitus research with special emphasis on genetic regulation of the disease.

To understand the basis of type 1 diabetes mellitus, one needs to understand the events that lead to the development of this autoimmune disease. So far, the causes leading to diabetes have been categorized as 3 types, namely: genetic predisposition/hereditary factors, viral infections and environmental factors. The basics of these 3 causes are described later.

Diabetes runs in families. Epidemiologic studies have shown a higher incidence (6% in siblings vs. 0.4% in the general population) of the disease among the relatives of type 1 diabetes mellitus patients, underlying the role of genetic factors as a cause of type 1 diabetes mellitus (5). The immunologic model of type 1 diabetes mellitus was developed on a simple hypothesis based on the assumption that insulin is recognized as a non-self-substance in populations suffering from type 1 diabetes mellitus. Hence, it is logical to assume that either a defect in the structure of insulin or an imperfection in the recognition process is responsible for the disease.

In the mammalian system, each and every nucleated cell has special marker molecules, expressed at the cell surface, which help in identifying them as constituents of its own system. A group of genes known as the major histocompatibility complex (MHC) is responsible for the production of these marker molecules (6). In human beings, this system is known as the human leukocyte antigen complex (HLA). Two chromosomal regions in the human genome have emerged with consistent and significant evidence of an association with type 1 diabetes mellitus. These are the HLA at the short arm of chromosome 6 (locus 6p21.3) and the insulin gene region at chromosome 11 (locus 11p15). HLA genes are grouped into 3 classes (classes I, II and III), which produce 3 distinct types of molecules.

Class I molecules are found on the surface of all types of nucleated body cells and these cells are identified as self. Class II molecules appear only on cells that actively take part in body defense mechanisms, namely macrophages, dendritic cells, T cells and epithelial cells of the islets of Langerhans. In type 1 diabetes mellitus, class II genes are of prime importance and can be divided into 3 subclasses, namely: HLA-DQ, HLA-DP and HLA-DR. Studies have shown that some variants of HLA-DQ and DR genes (HLA-DQA1, DQB1 and DRB1) are involved primarily in the genetic predisposition to type 1 diabetes mellitus. Among these, the HLA-DQ locus is the strongest susceptibility candidate (4). The insulin gene region on chromosome 11 (locus 11p15) is the second most important genetic susceptibility factor identified for type 1 diabetes mellitus. Further, the insulin is a product of one gene located in the insulin gene region. Studies in human beings and mice have shown the presence of insulin as an autoantigen along with insulin antibodies in the beginning stage of diabetes (7). Some other regions of the human genome also were shown to have a plausible role in the development of type 1 diabetes mellitus, namely, cytotoxic T-lymphocyte antigen-4, protein tyrosine phosphatase nonreceptor type 22 (PTPN22) and interleukin-2 alpha chain receptor (IL-2RA) (8).

Recently, genome-wide association studies have been used to identify genetic loci-associated type 1 diabetes mellitus. In contrast to the traditional methods of studying a candidate chromosome, genome-wide association studies scan the whole genome for single nucleotide polymorphisms (SNPs). SNPs that occur more frequently in people suffering from type 1 diabetes mellitus are said to be associated with the disease. The associated SNPs then are used to mark the susceptibility loci. By using the SNP typing technology, a number of additional susceptibility loci were discovered for type 1 diabetes mellitus, namely: CLEC16A, CI1QTNF6, UBASH3A, CD226, PTPN2, CTSH, SH2B3, ERBB3, PRKCQ, TAGAP, IL-2RA, TNFAIP3, BACH2, IL-7R, IL-2, CCR5, IFIH1, IL-18RAP, RGS1, IL-10, IL-19, IL-20, GLIS3, CD69 and IL-27 9, 10, 11.

The complex cascade of events that lead to type 1 diabetes mellitus have been simplified in a model scheme developed by Mahaffy and Edelstein-Keshet (Figure 1) (12). The model indicates that any damage of insulin-producing beta cells can lead to the activation of T cells against self-antigens of the human system. Briefly, the damaged beta cells undergo apoptosis, which produces self-antigen peptides. In the pancreatic lymph nodes, this peptide is presented on the antigen-presenting dendritic cells (p-MHC). The native T cells coming in contact with these antigens fail to recognize this as a self-protein and get differentiated to recognize these proteins as foreign antigens. A fraction of differentiated T cells remain as memory cells, while the other fraction take part in active killing (cytotoxic T cells), leading to type 1 diabetes mellitus.

Some investigations have indicated an association of type 1 diabetes mellitus with certain types of enteroviral infections. Coxsackie virus B4 contains a protein 2C(P2C) that is similar to the enzyme glutamic acid decarboxylase, which is present on the islets of Langerhans. Because of the molecular mimicry, P2C mistakenly is taken as a self-molecule and is not attacked by the T lymphocytes. Normally, T cells target the envelope proteins (VP1, VP2 and VP3) of Coxsackie virus B4, but the T cell proliferative response was reduced markedly in type 1 diabetes mellitus patients compared with control subjects, which eventually results in destruction of beta cells (13).

Epidemiologic studies have been performed to pinpoint the role of environmental factors in type 1 diabetes mellitus. Studies on people from the same ethnic group but located in different geographic areas (e.g. Finland vs Estonia) have shown a different prevalence of diabetes. The probability of developing type 1 diabetes mellitus in Estonia (Baltic region) is only one third of that in Finland (14). Among other environmental factors, exposure to antigenic substances early in life also is thought to contribute to disease development. Certain ingredients of diet, such as bovine serum albumin, beta-casein and gluten, are implicated as causative factors of type 1 diabetes mellitus. Beta-casein and bovine serum albumin of cow's milk seems to act via generation of T lymphocytes that specifically attack the beta cell-specific glucose transporter molecule GLUT-2 (Glucose Transporter 2) (15). Undissolved gluten causes subclinical inflammation of intestinal mucosa, which raises the proportion of aggressive T cells. The functional state of beta cells also plays a role in the pathogenesis of type 1 diabetes mellitus, and food intake with a high glycemic index increases the insulin demand and forces the beta cell to produce more insulin, which accelerates its destruction. This observation has inspired the accelerator hypothesis, which states that increased weight gain in youngsters might accelerate type 1 diabetes mellitus development (16). Another hypothesis formulated by Strachan (17) states that autoimmunity is more common in clean surroundings and exposure to infectious diseases in early childhood can reduce the incidence of type 1 diabetes mellitus. It has been suggested that the protection from type 1 diabetes mellitus comes from innate immunity, specifically the activation of natural killer T (NKT) cells. This theory gets support from the observation that NKT cells are reduced in human beings affected by type 1 diabetes mellitus (18).

Although a complete cure of type 1 diabetes mellitus still evades us, there has been significant advancement in the understanding of the disease. Every year a number of plausible anti-type 1 diabetes mellitus molecules are developed and their evaluation in terms of therapeutic efficacy has become a prime research area. Because type 1 diabetes mellitus is a long drawn-out process with heterogeneous genetic and environmental influences, evaluation of these molecules is performed in animals. The guiding principle of animal research is to use the lowest possible animal in the phylogenetic order, therefore rodents are used extensively in the study of type 1 diabetes mellitus (19). In rodent models, genetic and environmental factors can be controlled easily and hence primarily are used for evaluating the therapeutic value as well as the mechanism of action of candidate drug moieties.

Similarities in the etiology and pathogenesis of type 1 diabetes mellitus in human beings and rodents have led to the development of rodent models specific for type 1 diabetes mellitus research. Rodents are favored in type 1 diabetes mellitus for convenience and economic reasons, although extrapolating the outcome of such studies to human beings often is fraught with difficulties. The life span of rodents is much shorter compared with human beings, hence only preclinical studies can be justified in rodents. Certain strains of rodents develop type 1 diabetes mellitus spontaneously. Such animals, although expensive and require special maintenance, are of immense importance in research. However, studies also are performed in rodent strains that do not have any genetic predisposition. Induction of diabetes in such animals is performed by forced destruction of beta cells by chemical/surgical methods. The advanced-level studies usually are performed on larger mammals, although presently the ethical question of using companion animals for experimentation has resulted in banning such experiments in many countries. Animal models can be categorized into 2 types: spontaneous models and models developed through various manipulations. The particulars of various rodent models used in type 1 diabetes mellitus are summarized in Table 1.

Although human data are necessary to study the etiology of type 1 diabetes mellitus, spontaneous rodent models are the principal animals used in diabetes research. Prominent among these are nonobese diabetic (NOD) mice and the Biobreeding (BB) rat (Biobreeding Laboratories, Ontario, Canada). These spontaneous strains have evolved from the pool of normal animals by careful selection of type 1 diabetes mellitus-prone individuals. Similar to human beings, some animals in rodent strains may be diabetic. When these diabetic animals are bred selectively among each other (inbreeding), a greater percentage develop diabetes. These overtly diabetic animals are bred repeatedly, causing a gradual enrichment of genes until the animals achieve uniformity in the genetic traits. These models show significant similarities in genetic loci, environmental influences, as well as disease pathogenesis with respect to human type 1 diabetes mellitus. Some of the commonly used spontaneous rodent models are listed later.

NOD mice are the most favoured model for studying the etiopathogenesis of type 1 diabetes mellitus. NOD mice show the same clinical symptoms of diabetes (hyperglycemia, polyuria and polydipsia) as observed in human beings. Similar to human beings, NOD mice undergo subclinical beta cell destruction before overt diabetes presents, and show a similarity in genetic level. In both cases, type 1 diabetes mellitus susceptibility is determined largely by MHC complex. For example, genes encoding MHC class II analog (H2g7 in mice) showed the same diabetogenic amino acid substitution (20). Defects in antigen-presenting cell maturation have been noted in both cases and both produce autoantibodies to insulin, glutamic acid decarboxylase and islet cell antibodies (5). Experiments in congenic strains of NOD mice have shown that higher expression of the cytotoxic T-lymphocyte antigen-4 gene can render protection from type 1 diabetes mellitus. Another type 1 diabetes mellitus gene shared by both human beings and NOD mice is PTPN22. It has been observed that congenic intervals having the mouse orthologue of PTPN22 is associated with the occurrence of type 1 diabetes mellitus. The role of the IL-2 receptor alpha gene also has been confirmed through this experimental model. IL-2 plays a significant role in the generation of regulatory T cells. In the NOD mice, the gene encoding IL-2 is the major candidate gene for insulin-dependent diabetes susceptibility loci. Experiments in which anti-CD3 antibody was injected to suppress T cell-dependent immunity, caused reduction in the expression of IL-2, which reduced the frequency of type 1 diabetes mellitus (21). This finding suggests that the alteration in the function of the IL-2-receptor complex can be used as a strategy to control type 1 diabetes mellitus in human beings too. In both human beings and NOD mice, insulitis that initiates the disease process is destructive T cell driven. The role of T cells in beta cell destruction has been studied extensively and it has been suggested that IL-1beta, probably together with IFN gamma and TNF alpha, induce necrosis. Sulphated beta-galactosyl ceramide (sulfatide), an anti-inflammatory molecule associated with insulin, seems to offer protection against this destruction. In NOD mice as well, sulfatide was shown to inhibit diabetes development (16). Similarities between NOD and human beings also extend to environmental factors. Similar to human beings, NOD mice have defects in NKT cell development (22) and overexpression of NKT cells prevents transgenic NOD mice from developing diabetes (16). Both type 1 diabetes mellitus-prone human beings and NOD mice show subclinical inflammation of the intestine with exposure to a gluten diet. The subclinical inflammation is thought to increase the level of effector T cells and expedite the process of type 1 diabetes mellitus (16). The other similarities between the two are rapid onset of type 1 diabetes mellitus, responsiveness to immunomodulation and susceptibility to clinical thyroiditis.

NOD mice have severe limitations too. For example, susceptibility to autoimmunity in the NOD strain is influenced by many genes (23). A slight variation in the MHC complex can change its autoimmunity profile (24). NOD H2g7, which is linked to the development of diabetes, cannot cause diabetes if it is transferred to a different species. To produce the disease, it needs a specific genetic environment (i.e. a specific MHC with a large set of permissive background genes). The chance of having the same combination in human beings is almost nil (25). Furthermore, in human beings the distribution of disease in both sexes seems to be equal, whereas in the NOD strain females (90%) are more affected than males (50%). A significant dissimilarity that criticizes the use of NOD mice as a type 1 diabetes mellitus model is an unpredictable response to the environment. It is observed that virus infection can reduce the frequency of diabetes and often prevent it entirely. Compared with human beings, insulitis is initiated 4 to 5 weeks earlier, and in contrast to human beings the infiltration of the islets by the aggressive leukocytes (CD4+) starts at the perimeter. Moreover, because of a higher resistance to ketoacidosis, the overtly diabetic NOD mouse (in which around 90% of pancreatic beta cells are destroyed) can survive longer than human beings with a similar level of pancreatic dysfunction (26).

As the name indicates, a Biobreeding rat is a product of Biobreeding Laboratories. It is available in 2 varieties: Biobreeding–diabetes prone (BB-DP) and Biobreeding–diabetes resistant (BB-DR). The BB-DP rat is considered to be the best rat model to study type 1 diabetes mellitus. Development of type 1 diabetes mellitus is spontaneous as well as virus induced. Remarkable similarities exist between the BB rat and human beings in terms of disease development. Overt diabetes is preceded by insulitis (2 to 3 weeks before) with a predominance of Th1 lymphocytes. At about 8 to 16 weeks (adolescence), BB rats become overtly diabetic and manifest insulinopenia, polyuria and polydipsia, as observed in human beings. Males and females are affected equally, similar to human beings. In the BB-DP rat, the expression of diabetes requires the presence of at least 1 MHC class II RT1B/Du allele (insulin-dependent diabetes mellitus [type 1 diabetes] susceptibility loci in rats). The BB-DP rat develops T cell-dependent autoimmune diabetes, which is characterized by islet autoantibodies as well as glutamic acid decarboxylase antibodies. Similar to human beings, ketoacidosis is very severe in the BB rat and cannot survive without insulin (26). High insulin demand seems to expedite the process of type 1 diabetes mellitus. It is observed in BB rat litters and it is the heaviest rat that develops type 1 diabetes mellitus first. Similar to human beings, BB rats also have impaired intestinal function (16). The only significant disadvantage against this model is the development of severe T cell lymphopoenia, which is absent in both human beings and NOD mice (5).

Apart from the BB-DP rat, the other rat model that seems to mimic human for type 1 diabetes mellitus is the Komeda diabetes-prone (KDP) rat. It is the substrain developed from the Long Evan Tokushima Lean rat, which was the first rat observed to undergo spontaneous destruction of islet cells. The KDP rat develops diabetes slightly later compared with the BB-DP rat. Severe insulitis is observed at approximately 17 to 31 weeks (5). Both MHC and non-MHC genes are involved in type 1 diabetes mellitus susceptibility in this rat. The KDP rat shares the Rt1B/Du haplotype with the BB-DP rat, however, the contribution of a non-MHC gene (Cblb) gene also is important for the development of type 1 diabetes mellitus. Cblb codes a ubiquitin ligase that acts as a costimulator of CD28 during T cell activation and is the major susceptibility gene for type 1 diabetes mellitus. However, this gene has no linkage with diabetes in human beings (27). The frequency of the disease in this model is 70% to 80%, and is distributed equally between both the sexes.

LEW.1AR1/Ztm-type 1 diabetes is another rat model used for type 1 diabetes mellitus research. The advantage with this model is that development of type 1 diabetes mellitus is rapid (8 weeks). Compared with the BB-DP and KDP rats, this model shows greater similarity with human disease with the occurrence of pronounced insulitis, B and T lymphocytes (CD4+, CD8+), as well as innate immune components, such as macrophages and NK cells (28).

The second category (i.e. models developed through manipulation) can be divided into several subtypes.

The chemical approach of inducing type 1 diabetes mellitus into rodents involves injecting chemicals to specifically destroy the beta cells of Langerhans (beta-cytotoxic agents). Initially, alloxan was used as a beta-cytotoxic agent to induce symptoms of diabetes in both mice and rats. Alloxan-treated animals expressed the classic symptoms of human diabetes (hyperglycemia, glycosuria polyuria, polydipsia and so forth) and were used extensively in the initial phase of type 1 diabetes mellitus research for inducing diabetes. However, the renal toxicity of this compound was an undesirable factor. Alloxan-treated mice can show spontaneous recovery from a chronic diabetic condition, which makes the interpretation of therapeutic efficacy of the candidate drug extremely difficult.

Streptozotocin (STZ), another beta-cytotoxic agent, was more specific in the destruction of beta cells. It is a powerful alkylating agent that induces deoxyribonucleic acid breaks in the beta cells. In susceptible rodents, this induces insulinopoenic diabetes, in which immune destruction plays a role (29). The added advantage of STZ is that the degree of beta cell damage is dose dependent and it can be used to create subclinical conditions of diabetes. Of the 2 agents, STZ specifically is suited for rats. In mice, the substance should be used with caution because the window between the diabetogenic and general toxic dose is narrow.

One of the prominent theories suggests that type 1 diabetes mellitus happens when genetically susceptible hosts are exposed to strong nongenetic environmental factors such as viral infections (30). Rat models for type 1 diabetes mellitus have been developed based on this rationale. The beta cell destruction can be caused by either direct infection or through initiation of an autoimmune response (antigenic challenge) against the beta cell (6). It is well established that the entry of the viral proteins into a biological system can initiate a signalling cascade, which results in expression of some cytokine genes. Once cytotoxic T lymphocyte proliferates and differentiates into an activated effector cell, it can cause lysis of the virus-infected cell. Type 1 diabetes mellitus-like syndrome can be induced in BB-DR by using certain viruses, such as Coxsackie B virus, encephalomyocarditis virus and Kilham rat virus (19). Studies on Kilham rat virus-infected DR-BB rats have suggested that the destruction of beta cells was caused by selective activation of beta-cell cytotoxic effector T cells (31).

Type 1 diabetes mellitus also can be developed in rat models by manipulation of innate immunity. Macrophages have a special type of surface molecules called Toll-like receptors that can specifically recognize and bind unique pathogen-associated molecular patterns of different classes of pathogens. This can induce a cascade of intracellular signalling events, culminating in the up-regulation of the proinflammatory pathway (32). Based on this rationale, type 1 diabetes mellitus has been developed in the BB-DR rat by transfection of viral pathogen-associated molecular pattern-polyinosinic-polycytidylic acid (poly I:C) (33). Poly I:C binds to Toll-like receptor 3 and induces proinflammatory cytokines that ultimately cause destruction of beta cells. At low doses, poly I:C induces diabetes in a limited number of BB-DR rats, whereas at higher doses nearly all animals are affected (31).

Transgenic technology is used widely to manipulate the genome of rodents for the creation of specific models for both type 1 and type 2 diabetes. Research is on to cure diabetes through stem cell transplantation. Human stem cells are self-renewing, clonogenic and pluripotent cells with a potential to renew pancreatic beta cells and ensure a permanent supply of insulin (34). However, to be effective, the transplanted beta cells must survive, differentiate and function in the presence of the human immune system. To test this survivability, experiments are conducted in animal models. Immunodeficient strains of mice usually are used and conditions similar to the human immune system are created in them through engraftment of peripheral blood mononuclear cells/hemopoietic stem cells. A significant number of immunodeficient and gene-deficient knockout models have been developed to act as a preclinical bridge for understanding human immunity in vivo. The prominent models among these are described later 35, 36, 37, 38.

Nude mice are naturally athymic, hairless mice with a single gene mutation.

A mutation at Prkdc/severe combined immunodeficiency mice (scid) (protein kinase DNA activated catalyst polypeptide) gene needed for joining the homologous ends of double-stranded DNA was found on a C.B17 inbred strain that causes impaired T cell and B cell development. However, low levels of engraftment of human peripheral blood mononuclear cells (PBMCs) and hepatic stellate cells (HSCs) have been reported as a result of immunoglobulin leakiness in older animals because unhampered innate immunity in this strain leads to the generation of functional T and B cells with age.

This hybrid xenotransplantation model generated by back-crossing scid mutation to a NOD/LtSz background is an alternative model for human bone marrow transplantation. NOD/LtSz mice have multiple defects in antigen-presenting cell function along with innate immunity components of NK cells, macrophages and so forth. Transferring scid mutation in a NOD/LtSz background strain for 10 generations generates a model lacking T and B cell function, NK cell dysfunction and cytokine dysregulation. NOD/LtSz-scid mice have been used to transplant a variety of normal and malignant human cell populations and tissues successfully. However, the usefulness of these models is limited because they are predisposed to formation of thymic lymphomas at an accelerated rate and display a poor engraftment rate.

The immunogenicity of an antigen depends on its ability to bind with a MHC molecule. Hence, understanding the property of autoantigens that leads to insulitis is of paramount importance in type 1 diabetes research. HLA-A2 is a MHC class I molecule that presents antigenic peptides to the CD8+ T cell, which leads to a cascade of events causing destruction of beta cells. Hence, humanized HLA transgenic immune-competent mice have been developed for the identification of islet autoantigens. Studies in NOD mice have indicated that HLA-A2-restricted T cell responses are essential to type 1 diabetes mellitus development. It is found that the incorporation of HLA-A2 into the genome HLA-Lt accelerates the rate of diabetes onset when compared with HLA/Lt controls, resulting in a fast model in terms of type 1 diabetes mellitus development (39).

The application of gene targeting has facilitated the creation of knockout models in which genes involved in immune regulation are deleted selectively. Some of the models created by this method are listed later 35, 36, 37, 38.

Mice strains for type 1 diabetes mellitus were generated by deletion of Rag1/Rag2 (recombination activating) genes on background strains of Balb/C, C57BL/6J and NOD. Rag1 and Rag2 are 2 proteins necessary for immunoglobulin and T cell receptor gene recombination. Hence, deletion of either Rag1/Rag2 leads to an absence of mature T and B cells. The model shows a phenotype similar to that of the scid model, but free of immunoglobin leakiness.

Further genetic manipulation have created immunodeficient mice more permissive to human lymphohematopoietic engraftment-like expression of IL-2rg, beta2 microglobulin, perforins (pore forming proteins) and so forth.

This strain is developed by deletion of the IL-2R gamma chain gene, impairing multiple cytokine signalling and causing a complete block of T and B cells as well as NK cell development. This model is quite effective in accepting human cells and tissue transplants.

This strain was created by back-crossing Rag1null mice gene mutations into a NOD/LtSz strain background. NOD/Ltsz-Rag1null mice are devoid of T and B cells, have low NK activity and low serum immunoglobulin levels. The advantage with this model is a longer life span compared with NOD-scid mice.

This is a radioresistant hybrid model, generated by crossing NOD-scid mice with IL-2rgnull mice. The model lacks functional B and T cells and is free of immunoglobulin leakiness. Lymphoma resistant and lacking NK cell activity, the strain lives approximately 16 months and can be used for long-term experimentation.

This is a radioresistant double-mutant model created by breeding NOD-Rag1null mice with NOD-scid IL-2rgnull mice. Absence of mature T and B cells and cells functioning in innate immunity are severely reduced in this model.

Compared with parental stains of NOD-scid or NOD-Rag1null, both NOD-scid-IL-2rg null and NOD-Rag1null IL-2rg null strains show greater effectiveness of engraftment and differentiation of human tissues, PBMCs and HSCs.

Recently, immunodeficient transgenic mice containing human gene constructs (NOD-Rag1null IL-2rγnull Ins2Akita) were developed, which showed a high engraftment ratio of human PBMCs and HSCs (40). This newly developed mouse is useful for studying the engraftment potential of human islet cells in the presence of the robust human immune system. The schematic representation of producing transgenic animal models is presented in Figure 2.

Surgical removal of the pancreas is one of the most reliable methods to develop insulin-dependent type 1 diabetes mellitus. However, the usefulness of this method is restricted in small animals because of the stress factors associated with surgery. Despite this limitation, a large number of studies are conducted on this model because of the simplicity of the procedure as well as affordable cost.

As a new molecule starts its journey toward the status of drug, its efficacy needs to be proved in larger animals. Current drug product safety regulations mandate that the pharmacokinetic and pharmacodynamic of new chemical entities are to be tested in both rodent and nonrodent laboratory animals before human administration. Unlike the rodent models, the probability of spontaneous diabetes in larger animals is rare and erratic in onset (19). That is why the most widely used techniques to induce type 1 diabetes mellitus in nonrodents are either surgical or chemical (19). The choice of animals bigger than rodents is a complex issue that takes into account a number of factors, such as physiological/metabolic similarity, disease profile, as well as the cost and availability of the animals. Animal species currently used in type 1 diabetes mellitus research include dogs, primates, pigs and mini pigs 5, 41.

Convenient size, moderate life span (5 to 8 years), mild temperament, ease of breeding and low maintenance has made the rabbit one of the primary choices in diabetes research. In rabbits, diabetes is induced chemically with multiple dose of alloxan, and animals were maintained throughout the diabetic period with a regular dose of insulin (42). However, alloxan is a beta-cytotoxic agent, which injures cellular DNA and is associated with a high death rate. Recently, a long-term rabbit model was developed, opening the vista to study the long-term complications of diabetes (42).

Dogs happen to be the first animal model used in diabetic research. In 1921, insulin was first administered to a diabetic dog, paving the way for human insulin therapy. The incidence of spontaneous diabetes in pet dogs is quite common (43), although there is no specific indication whether it is of autoimmune type. Evidence of juvenile diabetes is rare in dogs, but immune-mediated beta cell damage in middle-aged and older dogs indicate that canine diabetes can be a model for latent autoimmune diabetes in adults. Keeshond dogs can be spontaneously diabetic because they develop diabetes within a period of 2 to 6 months and can survive without insulin therapy for 2 to 5 months. Moreover, diabetic Keeshond dogs show histopathologic and biological changes similar to that of human beings (44).

The biological system of pigs seems to be very close to human beings in terms of structure and functions. Similarities exist in the morphology of pancreas, structure and function of the gastrointestinal tract and overall metabolic status. Hence, the pig can serve as a valuable animal model for mechanistic study as well as therapeutic intervention. In general, the induction of diabetes is performed chemically and surgically. Pancreatic islet transplantation is an advanced strategy to prevent diabetes in patients whose pancreas is removed surgically. In pancreatic islet transplantation, healthy islets isolated from the removed pancreas are transplanted into the patients. Pig islets are considered to be very good candidates (45).

A chemically induced model for type 1 diabetes mellitus also has been developed in nonhuman primates, pig tailed macaques (Macaca nemestrina). Beta cell impairment was achieved after intravenous injection of STZ at a dose of 10 to 40 mg/kg body weight and the macaques became frankly diabetic in 7 to 10 days (46).

An important large nonhuman primate used in type 1 diabetes mellitus research is the baboon from the National Health & Medical Research Council (NHMRC) baboon colony of Australia, which have been used to develop a pig-to-baboon model of xenotransplantation (47).

Section snippets

Conclusion

Type 1 diabetes, being a complex and poorly understood disease, results from an unfortunate combination of genetic susceptibility, immunologic deregulation and exposure to environmental triggers. It is hypothesized that T cell activity, beta cell stress, NKT cell activity and dietary components act in concert to some degree for the development of type 1 diabetes mellitus, and intervention of any of these factors can obstruct the type 1 diabetes mellitus process (16). However, no single animal

References (55)

  • F. Pociot et al.

    Genetics of type 1 diabetes mellitus

    Genes Immun

    (2002)
  • J.M. Willey et al.

    Prescott's principles of microbiology

    (2008)
  • C. Pihoker et al.

    Autoantibodies in diabetes

    Diabetes

    (2005)
  • L.E. Paisan et al.

    An insight into the genetics of type 1 diabetes

    Immunologia

    (2009)
  • P. Concannon et al.

    Genetics of type 1A diabetes

    N Engl J Med

    (2009)
  • J.C. Barrett et al.

    Genome-wide association study and meta-analysis find that over 40 loci affect risk of type 1 diabetes

    Nat Genet

    (2009)
  • J.D. Cooper et al.

    Confirmation of novel type 1 diabetes risk loci in families

    Diabetologia

    (2012)
  • J.M. Mahaffy et al.

    Modeling cyclic waves of circulating T cells in autoimmune diabetes

    SIAM J Appl Math

    (2007)
  • P. Zimmet et al.

    Global and societal implications of the diabetes epidemic

    Nature

    (2001)
  • P. Pozzilli

    Prevention of insulin-dependent diabetes mellitus

    Diabetes Metab Rev

    (1998)
  • K. Buschard

    What causes type 1 diabetes? Lessons from animal models

    APMIS Suppl

    (2011)
  • D.P. Strachan

    Hay fever, hygiene, and household size

    BMJ

    (1989)
  • A.J. King

    The use of animal models in diabetes research

    Br J Pharmacol

    (2012)
  • M.A. Atkinson et al.

    The NOD mouse model of type 1 diabetes: as good as it gets?

    Nat Med

    (1999)
  • J.M. Gombert et al.

    Early quantitative and functional deficiency of NK1+-like thymocytes in the NOD mouse

    Eur J Immunol

    (1996)
  • K.T. Coppieters et al.

    Histopathology of type 1 diabetes: old paradigms and new insights

    Rev Diabet Stud

    (2009)
  • D.V. Serreze et al.

    Genes and pathways underlying autoimmune diabetes in NOD mice

  • Cited by (0)

    View full text