Type 1 Diabetes

Type I diabetes mellitus - otherwise known as juvenile diabetes or insulin-dependent diabetes mellitus (IDDM)- is considered to be an autoimmune disease. This disease usually begins in childhood or in the young adult years and tends to be more prevalent among females than males (InteliHealth 1999). Accounting for only 5% or less of diabetes in the United States, IDDM is not the most common form of diabetes. However, the physiological effects of IDDM tend to have a much greater impact upon patients' lives than the more common adult-onset form of diabetes known as non-insulin-dependent diabetes (NIDDM) (InteliHealth 1999). Thus, a thorough understanding IDDM and the possible methods of prevention and treatment of this disease is of utmost importance. Simply put, IDDM results when the immune system attacks and destroys the insulin-producing ß cells of the pancreas. The results of this attack are a pancreas that produces little or no insulin (Fig 1) and an inability to regulate the level of sugar in the blood (Scriver et al., 1995). Research shows that a great majority of people with IDDM inherit traits that put them at risk for this disease. However, studies also show that not everyone who inherits these traits develops type 1 diabetes. Therefore, it is hypothesized that environmental factors trigger the immune system to destroy the insulin-producing cells. In a few cases, researchers have been able to link the onset of diabetes with a viral infection (InteliHealth 1999). However, in most cases, the trigger for diabetes and the exact cause and mechanism of this autoimmune attack on the ß cells is not completely understood. The immunological basis of these topics will be the primary focus of this website.

Fig 1. Structure of insulin. Type 1 diabetes is characterized by the absence of insulin from the body. This lack of insulin presents a problem because insulin is needed by the body in order to allow sugar to pass into the cells for energy (Vallence-Owen 1975). Figure courtesy of the Protein Data Bank. See references for a link to the page where this image can be viewed at the Protein Data Bank.

Use of animal models: Several animal models of the human autoimmune disease IDDM have helped us understand the mechanism of this disease and need to be explained at this time. The NOD (non-obese diabetic) strain of mice spontaneously develops IDDM and is a useful tool for studying this disease. Diabetes in these mice involves a progressive mononuclear cell infiltration in the pancreatic islets of the mice. The infiltration begins at about four weeks of age and leads to ß cell destruction and hyperglycemia (Trembleau et al., 1999). The pattern of disease in these mice is very similar to that in humans except for the degree of incidence between the sexes. In the NOD mouse, 70 - 80% of females develop diabetes while only 10 - 20% of males develop this disease (Trembleau et al., 1999). The BB (BioBreeding) rat also spontaneously develops diabetes and the pattern of this disease is very similar to IDDM in humans with the exception of the association of T cell lymphopenia with the BB rat (Delves et al., 1998).

The general finding that endocrine tissue is especially vulnerable to autoimmune mechanisms led researchers to hypothesize that autoimmunity plays a pathogenic role in the development of type 1 diabetes (IDDM) (Andersen 1980). Indeed, research has demonstrated that IDDM is an autoimmune disease in which specific T cells selectively destroy the insulin-producing ß cells of the pancreatic islets (Janeway et al., 1999). Several lines of evidence support this important determination. First, it has been shown in human patients with IDDM that lymphocytic infiltrates (known as insulitis) containing activated T lymphocytes are located around pancreatic islets in patients who die shortly after being diagnosed with type 1 diabetes. In addition to studies in humans, animal models have also confirmed the finding of insulitis surrounding islets of diabetic animals. Additionally, in the BB rat and the NOD mouse, islet cell surface autoantibodies (ICSA) have been identified. Research has also indicated that immunosuppressive drugs are readily able to delay or prevent the development of disease in BB rats or NOD mice (Andersen 1980). Lastly, a strong association exists between type 1 diabetes and other autoimmune diseases such as vitiligo and pernicious anemia, and many patients with type 1 diabetes have family histories characterized by the prevalence of autoimmune disease (Scriver et al, 1995). The mechanism by which the ß cells are selectively destroyed is of utmost interest and has been highly researched; however the immunological basis of this disease is not yet fully understood. IDDM has been described to proceed in four distinct stages: (1) pre-clinical ß cell autoimmunity, (2) onset of clinical diabetes, (3) transient remission, and (4) established diabetes associated with acute and chronic complications and premature death (Rewers et al., 1997). The first of these steps will be the main focus of this website because it is during this stage of diabetes that the pancreatic ß cells are selectively destroyed by the immune system.

Researchers have determined that during the first stage of IDDM, antibodies are synthesized that act against the insulin-producing cells of the pancreas (Scriver et al., 1995). The consequence of these autoantibodies is a destruction of the insulin-producing beta cells of the islets of Langerhans’ cells and an absence or deficiency of circulating insulin (Slavkin 1999). This destruction probably occurs in genetically susceptible individuals in response to a particular environmental agent. It appears that ß cell-autoimmunity develops in less than 5% of the population and progresses into full diabetes in less than 1% of the general population (Slavkin 1999). The following specific autoantibodies have been discovered:

Islet Cell Cytoplasmic Antibodies: Indirect immunofluorescence stains of human pancreas sections demonstrate that nearly 90% of recently diagnosed diabetics have islet cell cytoplasmic antibodies (ICCA). ICCA are present in only 0.5 - 4% of non-diabetic individuals (Scriver et al., 1995). Although research indicates that ICCA are not specific for ß cells (and thus recognize antigens present in other cell types of the islet), the autoimmune attack of these antibodies appears to destroy ß cells selectively. It is interesting to note that the concentration of these autoantibodies decreases with time; research indicates that approximately two years after an initial diagnosis of IDDM, about 90% of patients no longer have detectable levels of anti-islet antibodies. The few individuals with persistent anti-islet antibodies may represent a subpopulation of persons with a more generalized autoimmune defect (Scriver et al., 1995).

Islet Cell surface Antibodies: Eighty percent of diabetic individuals also have autoantibodies directed against islet cell surface antigens (ICSA) at the time of diagnosis (Scriver et al., 1995). These ICSAs can lead to lysis of the islet cells in the presence of complement and tend to be IgM and IgG antibodies (Andersen 1980). As with ICCA discussed previously, ICSA also decrease considerably after the initial diagnosis. A recent interesting finding is that ICSA has also been found in some patients with non-insulin dependent diabetes mellitus (type 2 diabetes or NIDDM). Thus, it is probable that autoimmune mechanisms can cause either IDDM or NIDDM. IDDM involves total or almost total destruction of ß cells, while NIDDM results in partial destruction of the ß cells.

Currently, research is aimed at identifying the multiple antigens within the islet cells with which these anti-islet antibodies react. As reflected in the categories below, some progress has been made toward identifying these antigenic targets.

Antibodies to glutamic acid and decarboxylase: The enzyme glutamic acid decarboxylase (GAD) has been identified within the human islet cells. Further research determined that 80% of patients with newly diagnosed IDDM have anti-GAD antibodies while only 2% of normal individuals have these anti-GAD antibodies. The presence of anti-GAD antibodies has become a predictor of future development of IDDM in high risk populations (Scriver et al., 1995).

Autoantibodies to insulin receptors: Anti-insulin receptor autoantibodies have also been identified in individuals diagnosed with IDDM and in their relatives who are at risk for IDDM (Scriver et al., 1995). These antibodies are predominantly IgG with some IgM activity as well. In a study investigating insulin receptors on diabetic patients' circulating monocytes, researchers observed severe insulin resistance due to antibodies to insulin receptors. It was also found that when normal insulin receptors were exposed to serum from the diabetic patients in the previous study the insulin-binding defect was reproduced (Andersen 1980).

Antibodies to bovine serum: Interestingly, many patients recently diagnosed with IDDM have elevated levels of antibodies to bovine serum albumin. Most patients also have an increased number of antibodies directed toward a 17-amino-acid epitope in the albumin molecule. Antibodies to this peptide cross react with a protein located on the surface of pancreatic ß cells. These findings led researchers to the interesting hypothesis that ingestion of cow's milk leads to an immune response to bovine serum albumin in susceptible persons. The epitope in the bovine serum albumin molecule appears to mimic the structure of the surface protein of the pancreatic ß cells, and thus the anti-bovine serum albumin antibody is also an islet cell surface autoantibody. Therefore, a rather controversial hypothesis is that through a process known as molecular mimicry, intake of cow's milk, which contains bovine serum albumin, is an environmental element that precipitates the development of IDDM in genetically susceptible persons (Scriver et al., 1995).

The autoantibodies described above provide excellent markers for the destruction of pancreatic ß cells and are useful in predicting the future development of type 1 diabetes mellitus. It is possible that these autoantibodies may have a primary causal role in the ß cell destruction observed in IDDM. However, it is also equally likely that as a result of ß cell destruction, ß cell antigens are released into circulation, thereby inducing anti-islet antibodies. Most evidence supports the latter of these two hypotheses and suggests a primary role for cell-mediated autoimmunity (Scriver et al., 1995). As discussed below, it is probable that pancreatic ß cells are specifically targeted and destroyed by CD8 T cells (killer T cells) (Janeway et al., 1999).

Distinct cell types including alpha, beta, and delta cells are found within the islets of Langerhans in the pancreas. Each of these cell types secretes distinct hormones and expresses various tissue-specific proteins (Janeway et al., 1999). Alpha cells secrete glucagon, beta cells secrete insulin, and the delta cells are responsible for secretion of somatostatin. Research suggests that during the disease course of type 1 diabetes, self antigens unique to the ß cells are presented via MHC class I molecules. These autoantigens are then recognized by effector CD8 T cells and selectively destroyed. Thus because only the ß cells are eliminated by the CD8 T cells, alpha and delta cells still produce glucagon and somatostatin, respectively. However, the body's ability to make insulin is severely impaired (Janeway et al., 1999). Research has also determined that CD4 T cells (also known as helper T cells) may play a role in IDDM. This suggestion would be consistent with the previous findings that particular MHC class II alleles render some individuals more susceptible to the disease (see the section: Genetic Causes of ß cell Autoimmunity and Diabetes). Again, identifying the autoantigens recognized by CD8 T cells is of utmost importance. Identifying these antigens will allow scientists to gain a greater understanding of the disease, thereby increasing their ability to discover effective treatments or preventative measures for IDDM (Janeway et al., 1999).

Figure 2 shows the islets of Langerhans from non-diabetic (left) and diabetic (right) individuals. These pancreatic cells have been stained for insulin (brown) and glucagon (black). As can be seen in the photos, in the normal islets of Langerhans, the ß cells are able to produce insulin and the cells are highly stained with brown. However, in the diabetic individual on the right, it can be seen that little insulin is produced due to the selective loss of the ß cells, and thus significantly decreased amounts of brown staining can be detected. It should also be noted that the islets of the diabetic patient on the right are stained in black, thereby confirming the notion that the alpha cells are spared while the ß cells are selectively destroyed (Janeway et al., 1999).


Fig. 2. Islets of Langerhans cells from a normal individual. Brown staining confirms presence of insulin from ß cells. Black staining indicates the presence of glucagon made by alpha cells (Janeway et al., 1998).	Islets of Langerhans cells from a type 1 diabetic. Black staining indicates the presence of glucagon made by alpha cells. Extreme decrease in the level of brown staining demonstrates lowered levels of insulin due selective destruction of ß cells (Janeway et al., 1998).

Before approximately 1995 it was thought that IDDM was a direct result of a defect in CD8 T cells (killer T cells). However, research under the leadership of Denise Faustman of Harvard University demonstrated that the defect could be found not within the CD8 cells, but within the MHC class I molecules (Faustman et al., 1995). The defect in these MHC class I molecules causes the T cells to act inappropriately and respond to self antigens presented on the ß cells of the islets of Langerhans (Faustman et al., 1995). As discussed in the section below, it has been determined that genetic alterations on chromosome six in MHC class II molecule genes have been identified in a significant number of diabetic individuals. This information initially appeared contradictory to Faustman's finding that MHC class I molecules were instrumental in causing an autoimmune response. However, the researchers then discovered that the genes for the Tap-1 and Tap-2 proteins were located within the region encoding the MHC class II molecule. The Tap-1 and Tap-2 proteins are critical in the assembly of the MHC class I molecule. Thus, Faustman concluded that the deficiency in the MHC molecule leading to a CD8 T cell autoimmune attack could be due to a defect in the Tap-1 and Tap-2 genes (Faustman et al, 1998). However, subsequent research has not been able to identify any point mutations or deletions in the Tap genes. Faustman and her co-workers argue that the genetic defect in these Tap genes may be more subtle and complicated than a mutation or deletion. She suggests that simple variations in these genes may make them incompatible with other genes in the region. This suggestion implies that variations in MHC class I molecules will have unpredictable effects, and this idea is exactly what is observed in the realm of autoimmune diseases. For example, often a variety of autoimmune diseases are prevalent in a family. Variations in MHC class I alleles may explain why several autoimmune diseases such as rheumatoid arthritis, multiple sclerosis, or diabetes can be found within a single family (Faustman et al., 1995).

Additionally, it is becoming widely apparent that cytokines play an integral role in the development of type 1 diabetes. Research indicates that both IFN-alpha and IFN-gamma are closely associated with IDDM in both human and in NOD mice (Baldeon et al., 1998). Recent studies have demonstrated that IFN-gamma acts directly to decrease insulin production and upregulate cell surface expression of MHC class I molecules in pancreatic ß cells, thereby amplifying the insulitic process. However, studies involving IFN-alpha do not suggest similar activity for this cytokine. The action of IFN-alpha is hypothesized to occur during the stages before the onset of diabetes. It is proposed that the expression of IFN-alpha in response to potential diabetogenic stimuli (such as viruses as discussed below) may trigger insulitis to begin. Because IFN-alpha is also known to be associated with the stimulation of natural killer cells and T helper 1 cells, early IFN-alpha expression by ß cells may play a critical role in the development of IDDM (Baldeon et al., 1998). Interestingly other cytokines such as IL-10 and IL-12 are also intimately associated with the development of IDDM (Balasa et al., 1998). Research in this area has shown that IL-10 is essential for the early phase of diabetes in NOD mice, but later protects against the development of this disease. The mechanism of its action is not yet clear, however, scientists have shown that IL-10 is able to affect the disease process of NOD mice via the CD8+ T cell pathway without B cells present as antigen presenting cells and without the need for the CD40-CD40 ligand pathway. These studies suggest that it is possible that during the early stages of diabetes, IL-10 may act as a chemoattractant and a differentiation factor for CD8⁺T cells that are specific for ß cell peptides. However, during the late stages of diabetes, IL-10 may inhibit the generation of pathogenic CD4⁺ T helper 1 cells (Balasa et al., 1998). Additionally, the lack of IL-12 also appears to play a role in the onset of diabetes (Trembleau et al., 1999). Studies indicate that IL-12 deficient NOD mice are unable to develop a regulatory pathway that is able to counteract diabetogenic T helper 1 cells and readily develop IDDM (Trembleau et al., 1999). Research in the area of cytokines is currently very intense and may provide invaluable information regarding the mechanism of type 1 diabetes development.

As mentioned previously, the development of IDDM is controlled by several genetic loci, of which the most significant contributor may be the major histocompatibility complex (MHC) located on chromosome six. Specifically, it has been found that the primary locus of susceptibility to IDDM includes the, HLA-DR and HLA-DQ genes, but new possible loci for IDDM outside of the HLA region are currently being identified (Rewers et al., 1997). Studies involving the BB rat demonstrate that susceptibility to diabetes is strongly linked to genetic markers for the MHC and the inheritance of phenotypic markers like T lmyphopenia (Delves et al., 1998). It is not yet known which of these genetic markers associated with diabetes are important for development of ß cell autoimmunity and which determine progress to full scale diabetes (Rewers et al., 1997). For example, three out of the fifteen loci linked to type 1 diabetes in NOD mice lead to autoimmunity without the progression of diabetes. Similar experiments have not yet been performed in humans (Rewers et al., 1997). Researchers have been unable to identify a particular HLA genotype that is associated with the initiation of ß cell autoimmunity in humans. However, it has been determined that the HLA-DRß1*0301/04, HLA-DQß1*0201/0302 genotype promotes autoimmunity persistence and progression to the disease state of diabetes. Although the DRß1*0301/04, DQß1*0201/0302 heterozygotes make up only 2% of the population, this genotype is found in 30-40% of IDDM patients. The genetic nature of diabetes continues to perplex scientists as both susceptibility and protection from IDDM are associated with changes in the sequence of amino acids even within one locus (Janeway et al., 1999). For example, the human HLA-DQß1 gene, the genes *0302 and *0201 are found to be linked to progression of autoimmunity, while other genes such as *0602 inhibit progression from autoimmunity to diabetes (Rewers et al., 1997).

In most non-diabetic persons, position 57 of the HLA-DQß1 chain contains an aspartic acid residue (Fig. 3) (Janeway et al., 1999). However, Caucasians patients with IDDM show an increased likelihood of having valine, serine, or alanine at this position. In non-diabetic individuals with an aspartic acid residue at position 57, a salt bridge is able to form to an arginine residue in the adjacent alpha chain of the MHC class II molecule (Fig. 4). In patients with IDDM, a substituted amino acid at position 57 to an uncharged residue such as alanine disrupts the stability of the MHC class II molecule and interferes with the salt bridge formation (Fig. 5) (Janeway et al., 1998). Additionally, as discussed in the previous section, some research has found that slight variations in the genes for the Tap transporter protein, which are located within the region of the MHC class II molecule may play an integral role in causing an autoimmune reaction by CD8 T cells to self peptides on ß cells (Faustman et al., 1995). Obviously, further research in this area needs to be conducted in order to determine the genotype of many other individuals with ß cell autoimmunity. This knowledge may then allow scientists to determine the role of HLA and additional IDDM candidate genes in the induction of autoimmunity and progression to diabetes (Rewers et al., 1997).

Other research indicates that there must be an interaction of genes and environment to initiate the disease process (Slavkin et al., 1999). Gene-environment interactions are thought to operate in early childhood to initiate the disease process. The exact nature of these multiple gene-environment interactions and the possible involvement of infectious and noninfectious agents is not yet completely understood (Slavkin et al., 1999). However, we do know that several environmental factors are associated with the onset of diabetes.

Viruses: Viral infections appear to initiate autoimmunity in some circumstances. Islet cell antibodies and antibodies against insulin have been detected after mumps, rubella, measles, chickenpox, Coxsackie, and ECHO4 infections. Additionally, fetuses and newborn babies may be at an increased risk because of their heightened likelihood of developing persistent infections (Rewers et al., 1997). One report indicates that ß cell autoimmunity and diabetes may be related to expression of human endogenous retrovirus and that the mechanism may function via superantigen activation of autoreactive T cells. However, these results remain unconfirmed (Rewers et al., 1997)

Factors in utero: Recent evidence also points to the probability that ß cell autoimmunity and diabetes may be a result of enteroviral infections, which may be acquired in utero. Researchers hypothesize that molecular mimicry may exist between the P2-C protein of Coxsackie virus and the GAD protein and that this relationship may be responsible for ß cell autoimmunity (Rewers et al., 1997).

Cows' milk: As discussed previously, a controversial hypothesis that cows' milk may lead to the initiation of ß cell autoimmunity and diabetes. A recent study, The Diabetes Autoimmunity Study in the Young (DAISY), found that there was no association between early exposure to cow's milk and ß cell autoimmunity in young children (Rewers et al., 1997). However, other research has indicated that children with diabetes are 60% more likely to have been exposed early to cow's milk than children without diabetes (Rewers et al., 1997).

While the complications associated with diabetes are severe and possibly life-threatening, improved understanding of this IDDM has provided researchers with much optimism that this autoimmune disease will soon be either preventable or curable (Delves et al., 1998). Our current situation provides hope for the development of preventative vaccines. Some researchers suggest that it may be possible to design recombinant vaccines that would result in long-term protection against diabetogenic strains while preventing adverse effects. Other researchers suggest vaccinations involving antiviral agents (Rewers et al., 1997). As evidenced by the vagueness of these possible vaccinations, much work is still yet to be done in this area before a definitive preventative measure is developed.

Other researchers are targeting the relatives of IDDM patients in order to prevent the onset of type 1 diabetes. A recent pilot trial gave rise to a widespread unmasked trial in which first and second degree relatives who test positive for islet cell antibodies receive low doses of oral insulin. The results of this trial will be known in 2002 (Rewers et al., 1997).

Faustman, researcher at Harvard University, has discovered a method of correcting the MHC class I defect and thereby preventing the T cells from reacting against autoantigens within the ß cells (Faustman et al., 1995). Her technique includes transplantation of one of the two Tap genes (Tap 1 or Tap 2) from normal MHC class 1 molecules into the defective MHC class I molecules of patients with IDDM. In vitro studies revealed that the gene-altered MHC class I molecules were, indeed, able to elicit a normal response from CD8 T cells (Faustman et al., 1995). Her ideas and techniques have not yet been applied to human studies.

A less invasive method of preventing type 1 diabetes is the administration of vitamin D. A recent European study has demonstrated that treatment of NOD mice with the active form of vitamin D prevented the development of insulitis (Dahlquist et al., 1999). Additionally, in a questionnaire and interview study that they conducted, these researchers found that children taking vitamin D supplements were significantly less likely to develop IDDM. They hypothesize that vitamin D played a role in increasing tolerance to ß cells and improving sensitivity to apoptosis. These aspects lead to a better elimination of self-reactive effector cells (Dahlquist et al., 1999)

The usual treatment for type 1 diabetes is insulin injections, which provide the diabetic individual with temporary insulin that will then allow sugar to pass into their cells. However, for those that have suffered the consequences of type 1 diabetes for many years, full pancreas transplants may be a more permanent solution by providing the diabetic individual with functioning ß cells. But, as with any transplant, the body's natural immune system may attack this foreign organ. Thus in pancreas transplants immunosuppressive drugs must be given. These drugs are generally accompanied by several adverse side effects including increased susceptibility to infections and even cancer (McCarren 1996). As an alternative treatment, the Diabetes Research Institute (DRI) has recently made significant progress in the development of the islet transplant. Recently, Dr. Camillo Ricordi of the DRI invented an automated method of islet cell isolation, which made it possible for scientists to obtain large numbers of islets from a human pancreas (Diabetes 1999). Further improvements on his isolation technique made it possible for scientists to isolate enough islets from one pancreas to transplant into one recipient patient. Through an islet cell transplant, type 1 diabetic patients are able to reclaim the insulin-producing ß cells that were mistakenly destroyed by their own immune system (Diabetes 1999). This procedure is currently entering large scale clinical trials and may lend much encouragement to researchers looking for a treatment for IDDM. However, with this new procedure, as with full pancreas transplants, immunosuppressive drugs must be given and therefore the risk of subsequent infection and cancer are still present. Click here to link to the Diabetes Research Institute home page and view an amazing interactive animation of the islet isolation process: http://www.drinet.org/html/ricordi_method_of_islet_isolat.htm. Also from this page, download the appropriate plug-ins and view actual footage from an islet cell transplant in progress.

In order to provide treatment for the diabetic without life-long immunosuppression, researchers at the University of Miami have invented a technique by which they propose to give recipients islet cells and bone marrow from the same donor (McCarren 1996). In this procedure islet cells and bone marrow will be given separately. The islet cells will originate in a cadaver donor and will drip through a catheter into the liver of the recipient. It is hoped that once these islet cells reach the liver they will settle in and begin to secrete insulin. Immunosuppressive drugs will be given only initially in order to prevent rejection of these islet cells. Five days and then eleven days after the introduction of the islet cells, the patient will receive a bone marrow infusion from the same donor. In this procedure the researchers are hoping that the donated bone marrow will decrease the body's tendency to attack the donor islet cells. Their technique is based on the theory that the donor bone marrow will "educate" the recipient's immune system to accept islet cells from the donor. It has been questioned whether the donor's immune system would then attack the rest of the recipient's organs. However, researchers at Miami suggest that although the mechanism whereby this procedure works is unknown, the two immune systems do, indeed, learn to co-exist (McCarren 1996). What is not yet known is whether the islet cells will continue to survive after the stopping the use of immunosuppressive drugs.

Thus, the scientific community is hopeful that better preventative measures or a treatment for diabetes can and will be successfully developed in the future. Continuing studies in the area of IDDM will focus even more on the genetics behind this disorder, the mechanism of autoimmunity, and the cytokines involved in this process. The physiological effects of IDDM are severe and the better that scientists can understand the mechanism of this disease, the better they will be at intervening in the disease process and finding a way to prevent or treat type 1 diabetes mellitus.

Andersen O. Secondary Diabetes: The Spectrum of the Diabetic Syndromes. New York: Raven Press; 1980. p 409 - 419.

Balasa B, Davies J, Lee J, Good A, Yeung B, Sarvetnick N. 1998. IL-10 impacts autoimmune diabetes via a CD8+ T cell pathway circumventing the requirement for CD4+ T and B lymphocytes. Journal of Immunology 161: 6963 - 6969.

Baldeon M, Chun T, Gaskins R. 1998 July. Interferon alpha and interferon gamma differentially affect pancreatic ß cell phenotype and function. American Journal of Cellular Physiology 275(1): C25-C32.

Dahlquist, G. 1999. Vitamin D supplement in early childhood and risk for Type 1 (insulin-dependent) diabetes mellitus. Diabetologia 42: 51-54.

Delves P, Roitt I (eds). 1998. Encyclopedia of Immunology. 2nd Ed. San Diego: Academic Press.

Janeway CA, Travers P, Walport M, Capra JD. Immunobiology: the Immune System in Health and Disease. 4th ed. London: Current
Biology Publication; 1999. p 493 - 494, 508.

Scriver C, Beaudet A, Sly W, Valle D. The Metabolic and Molecular Bases of Inherited Disease. 7th ed. New York: McGraw-Hill Inc.; 1995. p 859 - 863.

Slavkin, H. 1999 May 11. Insights on Human Health: Towards a Common Theme in Autoimmunity. National Institute for Dental and Cranofacial Research. <http://www.nidcr.nih.gov/slavkin/slav0499.htm> Accessed 2000 Apr 17.

Trembleau S, Penna G, Gregori S, Chapman H, Serreze D, Magram J, Adorini L. 1999. Pancreas-infiltrating Th1 cells and diabetes develop in IL-12-deficient nonobese diabetic mice. Journal of Immunology 163: 2960 - 2968.

Vallance-Owen J (ed). Diabetes: Its Physiological and Biochemical Basis. Baltimore: University Park Press; 1975. p. 1.


Fig. 3. Position 57 (shown in red) of the HLA-DQß chain (blue) plays an integral role in susceptibility to IDDM (Janeway et al., 1998).	Fig. 4. An MHC class II molecule (from a non-diabetic individual) with an aspartic acid residue at position 57 is able to form a stable salt bridge (green) between an aspartic acid residue (red) and an arginine residue (pink) of the adjacent alpha chain (gray). An aspartic acid residue at this position is associated with resistance to IDDM (Janeway et al., 1998).	Fig. 5. An MHC class II molecule from a type 1 diabetic has a substituted amino acid at position 57. This photo shows alanine (yellow) at position 57. The presence of alanine at this position disrupts the formation of the salt bridge and is associated with increased susceptibility to IDDM (Janeway et al., 1998).
Images courtesy of Janeway C, Travers P, Walport M, Capra J.Immunobiology:the Immune System in Health and Disease.4th ed. London Current Biology Publication;1999. p 493.