The NOD mouse is an inbred strain that originated in the Shionogi Laboratories, Fukushima, Japan as a sub-strain of the cataract-prone mouse (CTS) that was noted for its spontaneous development of hyperglycemia. It has been extensively utilized in scientific research as a polygenic model of human type 1 diabetes (T1D). NOD mice develop diabetes spontaneously between 12-30 weeks of age and share many symptomatic and pathophysiological features of T1D including; inflammation of the pancreatic islet’s of Langerhans or “insulitis”, apoptosis of insulin-producing beta cells, hyperglycemia, polyuria and weight loss eventually leading to death. The discovery of the NOD mouse in the late 1970s was a breakthrough for T1D research, and preceded the development of engineered animal models of diabetes by about 10 years. The NOD mouse has since been an invaluable tool for the study of T1D aetiology, pathogenesis and prevention. Female NOD mice experience a higher incidence (60-80%) of type 1 diabetes compared to males (10-30%) and therefore are most commonly utilized in scientific research. Studies of the autoimmune inflammatory mechanisms that precede beta cell death in NOD mice have improved scientific understanding of the pathophysiology of T1D, however their usefulness in terms of developing intervention therapies that modulate the immune system remains to be seen.
NOD mice develop insulitis by the age of approximately 3 weeks, shortly after being weaned from their mothers. Mononuclear cell infiltrates can be found histologically within and surrounding pancreatic islets, resulting in death of beta cells. On average, 60-80% of female and 10-20% of male NOD mice will develop T1D. Female NOD mice tend to develop overt hyperglycemia indicating diabetes onset between 12-20 weeks of age (peak = 16-18 weeks). They subsequently display the hallmarks of T1D; polyuria, polydipsia, glycosuria, hypercholesterolemia and weight loss. Males not only have a lower incidence of the disease but also a delayed onset, often reaching 30 weeks of age before developing hyperglycemia and associated symptoms.
Insulitis and beta cell death
The current paradigm stipulates that the autoimmune destruction of pancreatic islets in NOD and T1D humans is initiated by an infiltration of dendritic cells (DCs) and macrophages (MØ) which is later followed by both T and B lymphocytes. The presence of autoreactive T and B cells is explained by defective negative selection of these populations in the thymus during early development. More recent studies have further implicated that the recruitment and activation of innate immune cells such as neutrophils, B-1a cells and plasmacytoid DCs occurs in the early stages of insulitis and is required for subsequent activation of autoreactive lymphocytes, proving that an innate-adaptive cell cross-talk occurs. Insulin-specific CD4+ T cells are thought to be central to the development of autoimmune diabetes and have often been identified posthumously in human subjects as well as NOD mice. CD4+ T cells activate cytotoxic CD8+ T cells which are able to directly induce apoptosis in beta cells by activation of the Fas receptor and release of granular toxins perforin and granzyme. Infiltrating leukocytes also indirectly promote apoptosis through stress pathways by the release of reactive oxygen species (ROS). While it is widely accepted that diabetes in NOD mice is primarily T cell mediated, they also develop a number of islet-specific autoantibodies indicating B cell involvement. While these antibodies alone cannot initiate significant beta cell damage, they have been shown to significantly increase the rate of disease progression1. Additionally, B cells have been implicated in T1D via their actions as antigen-presenting cells (APCs) and have been proven to be resistant to peripheral tolerance mechanisms (anergy) in NOD mice. Investigation of the pre-diabetic inflammatory mechanisms in the NOD mouse have implicated many cellular and non-cellular components of the immune system in the destruction of pancreatic beta cells and shown that defects in both central and peripheral tolerance mechanisms coincide to cause T1D.
To date, more than 30 genetic loci (termed Idd for Insulin dependent diabetes) related to diabetes susceptibility or resistance have been identified in the NOD mouse, many of which have helped to identify individual genes that drive the pathogenic process. The first and perhaps most important loci identified was designated Idd1 and is linked to the H-2 complex on mouse chromosome 17, indicating that the disease is associated with a particular major-histocompatibility (MHC) complex haplotype.
Unravelling the genetic code
Congenic lines of NOD mice, which contain sections of chromosomes from healthy mouse strains such as the closely related NON (non-obese normal) have been utilized in various linkage studies to identify genes and genomic regions relating to immune tolerance and self-reactivity. This method has also been useful to identify epistatic interactions between multiple loci and genes. Additionally, transgenic and knockout NOD mice have displayed the importance of individual genes in the pathogenic process.
The NCBI, MGI and Sanger databases contain genetic and mapping data from the NOD and various other mouse strains and are useful resources for finding existing Idd loci and genes. Online, the entire NOD genome with extensive sequencing information can be found in addition to a catalogue of human IDDM (Insulin-Dependent-Diabetes-Mellitus) loci and genes.
The incomplete penetrance of diabetes in NOD mice suggests a strong role for environmental factors in the pathogenesis of the disease. Many researchers have repeatedly shown that if you transfer a NOD colony to a germ-free environment you can increase diabetes incidence drastically. Specific pathogen free (SPF) holding rooms can also lead to defective maturation of T suppressor cells and antigen presenting cells (APCs). These observations together with epidemiological studies of autoimmune disease and allergies in humans have contributed to the development of the hygiene hypothesis, which stipulates that exposure to various pathogens early in life positively influences maturation of the immune system and aids self-tolerance mechanisms. In the late 1980’s it was shown that the effects of microbial immunomodulation can be reproduced by infusing various cytokines such as IL-1, IL-2, or synthetic toll-like receptor (TLR) agonists such as Poly(I:C), which are all protective against diabetes development. Furthermore, factors influencing the intestinal microbiome such as diet have been shown to alter the pathogenesis of T1D. Studies on the NOD mouse have implicated that gluten is one dietary component that has direct pro-diabetogenic effects and that its presence alters the levels of several microbial species in the gut including Akkermansia, Bifidobacterium, Tannerella, and Barnesiella.
Additionally, environmental influences such as temperature and sleep/circadian rhythm can modulate diabetes incidence in NOD mice. The gender imbalance that we observe would also indicate that the hormonal environment (including sex steroids) influences development of the disease – indeed it was demonstrated in early work that testosterone has some protective effects.
Similarities and differences: human T1D
On both a physiological and genetic basis, NOD mice share many features with human T1D patients. Characterisation of NOD diabetes pathogenesis has in many ways exceeded what we are able to study in humans due to the difficulty of detecting the disease in its early stages and inaccessibility of the organs key to studying the disease process. Cadaveric human islets and circulating lymphocytes in the blood have provided many clues to the pathogenesis of T1D, however mouse studies have been far more extensive and invasive.
Symptoms and co-morbidity
The NOD mouse shares virtually all the hallmarks of human T1D including polydipsia, polyuria, glucosuria, ketonuria, hypercholesterolemia and polyphagia. NOD mice and human T1D are also more susceptible to other autoimmune diseases such as Grave’s disease, Hashimoto’s thyroiditis, systemic lupus erythamatosus and celiac disease.
Image 1* Islet size and cell composition in NOD mice. (a) Islets from NOD mouse that did not develop hyperglycemia were immunostained for β-cells (green), α-cells (red), and δ-cells (blue). Scale bar—100 μm for all images. (b) Islet from an NOD mouse diabetic for 1 week. (c) Islet from an NOD mouse diabetic for 3 weeks. (d) Because the onset of hyperglycemia was extremely variable in the NOD mice, islet diameter was plotted according to the duration of diabetes. Islet diameter decreased with increasing duration of diabetes. (e) Cell composition was also plotted according to the duration of hyperglycemia. With increasing duration of diabetes, β-cells were lost and α-cell numbers increased. N for both graphs = 11.803 cells from 116 islets from 6 mice.Under the microscope, NOD and human T1D pancreatic islets share many similarities in their gross morphology. Leukocyte infiltration into the islets appears in the early stages of the disease. At later time points, the islets appear to shrink with massive beta cell loss. Alpha cells appear to increase and make up the majority of the remaining islet mass while delta cells remain more or less unchanged. NOD mice, however, tend to have a more distinct pattern of immune infiltrate around the islets (“peri-insulitis”) which is apparent for an extended period before they develop hyperglycemia. Humans, on the hand, tend to have a more “delicate” infiltrate within the islets that more immediately precedes the onset of overt symptoms.
*Image 1:  Variations in Rodent Models of Type 1 Diabetes: Islet Morphology, Novikova et al J Diabetes Res. 2013
NOD mice share many of the genetic elements associated with T1D, including polymorphisms in genes affecting IL-2 production, CTLA4 expression and MHC class I and II molecular recognition. It is of relevance that the H2g7 MHC haplotype of NOD allele shares homology with the human T1D susceptibility HLA-DQB1 locus. In particular, the recognition of autoantigens by MHC is essential for disease development in both human T1D and NOD and there are surprising similarities in the antigens that exact a response in both species, reflecting common genetic components. As these genes are central to immune regulation, mutations could explain the increased susceptibility observed in T1D patients and NOD mice to other types of autoimmune disease.
Autoreactive T cells
T cells with specific receptors for islet antigens such as insulin, glutamic acid decarboxylase 65 (GAD65), insulinoma-associated protein 2 (IA2), zinc transporter 8 (ZnT8) and islet-specific glucose-6-phosphatase catalytic subunit related protein (IGRP) have been identified in both NOD mice and T1D patients. In both humans and mice, CD4+ helper T cells are activated by APCs and consequently activate cytotoxic CD8+ effector T cells, which induce apoptosis in islets.
Role of autoantibodies
Autoantibodies against various islet-associated antigens including insulin, GAD65 and IA2 are present in T1D humans and are strong predictors for disease development. In contrast, NOD mice have no detectable islet-specific autoantibodies aside from those against insulin. Autoantibodies in NOD mice have not been shown to play a strong role in the pathogenesis of diabetes, however they can be predictive of the rate of disease progression. In humans, the presence of islet autoantibodies are highly predictive of T1D. Additionally, complement fixing antibodies are present in T1D patients and have been demonstrated to be cytolytic to beta cells.
Unlike humans, NOD mice show a clear female bias in terms of disease incidence and age of onset. Although both male and female NOD mice have lymphocytic infiltrates in the pancreatic islets from a young age, >80% of males are protected from developing diabetes. In humans, the sex ratio is roughly equal below the age of 15 while above age 15 males have a higher incidence (approximately 3:2 male:female ratio). Depending the on the specific population one might look at (eg. developed vs developing nations), the sex bias can shift, although a satisfactory explanation is yet to be offered for this trend.
Similarities and differences: BB rat
The BioBreeding (BB) rat colony was established in Canada from outbred Wistar rats in the 1970’s. They were selected on the basis that they spontaneously developed hyperglycemia and ketoacidosis, and therefore are one of the few existing models for polygenic, naturally occurring autoimmune diabetes. Their pathogenesis involves insulitis leading to beta cell death between 50-90 days of age, with a similar frequency of occurrence in both males and females. Unlike NOD mice, they also produce islet-cell surface autoantibodies. The BB rat is therefore useful in many ways as a complementary model of T1D with additional similarities to the human condition.
Development of T1D therapies: is the NOD mouse a false friend?
In NOD mice, more than 450 interventions have been trialed with the aim of preventing or retarding the onset of overt diabetes. The targets of these interventions fall into several categories including; cells of the immune system, costimulation, antigens, pathogens, cytokines / hormones and environmental triggers. Many of these therapeutic interventions have been successful in the NOD mouse, but unsuccessful in humans. Perhaps one possible reason for this discrepancy is that NOD mice are very much on the “knife’s edge” in terms of disease development, and therefore numerous non-specific therapeutic or environmental interventions are successful at tipping the balance and protecting the mice against diabetes. This is not to say that the NOD mice are not useful tools in medical research, however novel immunotherapies could be better validated by applying them to at least one other complementary animal model such as the BB rat.
^ Thayer, T et al. Use of NOD mice to understand human type 1 diabetes. Endocrinol. Metab. 39, 541–561 (2011).
^ Diana, J. et al. Crosstalk between neutrophils, B-1a cells and plasmacytoid dendritic cells initiates autoimmune diabetes. Nat. Med. 19, 65–73 (2013).
^ Stolp, J. et al. Subcongenic analyses reveal complex interactions between distal chromosome 4 genes controlling diabetogenic B cells and CD4 T cells in nonobese diabetic mice. J. Immunol. 189, 1406–17 (2012).
^ Wicker, L. S et al. B. GENETIC CONTROL OF NOD MOUSEl Natural History of Disease. Analysis 179–200 (1995).
^ Alam, C. et al. Effects of a germ-free environment on gut immune regulation and diabetes progression in non-obese diabetic (NOD) mice. Diabetologia 54, 1398–406 (2011).
^ Serreze, D. V et al. Immunostimulation circumvents diabetes in NOD/Lt mice. J. Autoimmun. 2, 759–76 (1989).
^ Williams, A. J. et al. Raised temperature reduces the incidence of diabetes in the NOD mouse. Diabetologia 33, 635–7 (1990).
^ Ruiz, F. S et al. Sleep deprivation reduces the lymphocyte count in a non-obese mouse model of type 1 diabetes mellitus. Braz. J. Med. Biol. Res. 40, 633–7 (2007).
^ Makino, S et al. Effect of castration on the appearance of diabetes in NOD mouse. Jikken Dobutsu. 30, 137–40 (1981).
^ Aoki, C. A. et al. NOD mice and autoimmunity. Autoimmun. Rev. 4, 373–9 (2005).
^ Novikova, L. et al. Variations in rodent models of type 1 diabetes: islet morphology. J. Diabetes Res. 2013, 965832 (2013).
^ Polychronakos, C. Animal models of spontaneous autoimmune diabetes: notes on their relevance to the human disease. Curr. Diab. Rep. 4, 151–4 (2004).
^ Lieberman, S. M. & DiLorenzo, T. P. A comprehensive guide to antibody and T-cell responses in type 1 diabetes. Tissue Antigens 62, 359–377 (2003).
^ Roep, B. O et al. Satisfaction (not) guaranteed: re-evaluating the use of animal models of type 1 diabetes. Nat. Rev. Immunol. 4, 989–97 (2004).
^ Gale, E. a & Gillespie, K. M. Diabetes and gender. Diabetologia 44, 3–15 (2001).
^ Bortell, R. & Yang, C. The BB rat as a model of human type 1 diabetes. Methods Mol. Biol. 933, 31–44 (2012).
^ Kachapati, K et al. The non-obese diabetic (NOD) mouse as a model of human type 1 diabetes. Methods Mol. Biol. 933, 3–16 (2012).