Type 1 Diabetes: Pathophysiology, Molecular Mechanisms, Genetic Insights

Type 1 diabetes mellitus (T1DM) is one of the most prevalent chronic diseases of childhood, affecting more than 1.4 million people in the United States, of whom 150,000 are children. ^, Over the past 50 years, the incidence in children has been increasing at a rapid rate of up to 5% per year worldwide—that is, doubling every 20 years. The lifetime risk of developing T1DM now exceeds 1% in North America and Europe. Whereas T1DM accounts for only approximately 5% of diabetes, it is associated with higher per-person morbidity, mortality, and health care costs than type 2 diabetes (T2DM).

T1DM incidence is trimodal, peaking at the ages of 2, 4 to 6, and 10 to 14 years. This pattern may reflect age-specific infections and increased insulin resistance of puberty. Although children are most visibly affected, half of T1DM patients are diagnosed after age 20. There is generally an equal male-to-female distribution of T1DM; however, a slight male predominance has been reported in high-risk populations, and the opposite in low-risk ethnic groups.

The ongoing pandemic of T1DM cannot be attributed to genetics or increasing survival and fecundity of adults with T1DM; a powerful environmental factor or factors must be at play. In this chapter we review the pathophysiology underlying T1DM, and discuss possible mechanisms linking genetic predisposition to environmental triggers in the development of T1DM.

Pathogenesis

T1DM is characterized by a long preclinical period of autoimmune attack on the beta cells, carried out by autoreactive T cells and marked by the emergence of autoantibodies against beta cell autoantigens. The process appears to result from loss of tolerance to beta cell autoantigens in genetically susceptible individuals. Several environmental triggers have been implicated, but none have been definitely proven. Figure 3-1 illustrates the complexity of the pathogenesis underlying T1DM.

The normal pancreas has a large reserve capacity; at least 70% of the functional capacity of the beta cells must be lost before clinical T1DM develops. Studies of human pancreata in patients with established T1DM suggest that a number of beta cells are able to survive the autoimmune insult, but are unable to secrete sufficient amounts of insulin to prevent hyperglycemia. Rodents may generate new beta cell progenitor cells, but there is no evidence for beta cell regeneration in humans with diabetes.

Selective destruction of pancreatic beta cells results in insulinopenia. The impairment in insulin secretion is also partially functional and caused by the inhibition of insulin secretion by cytokines interleukin 1 (IL-1), tumor necrosis factor alpha (TNF-α), TNF-β, and interferon gamma (IFN-γ). Insulin resistance may also play a role in T1DM pathogenesis and cannot be explained simply by obesity or puberty.

After diagnosis, T1DM patients are more insulin resistant than nondiabetic controls despite similar adiposity, body fat composition, and high-density lipoprotein (HDL) cholesterol. Significant insulin resistance has been documented in T1DM patients at or near hemoglobin A1c (HbA1c) targets, ^, suggesting that resistance to insulin action on glucose and nonesterified fatty acid suppression are not mediated by prevailing glycemia. In insulin-treated patients, insulin resistance is secondary to prolonged exposure to supraphysiologic levels of exogenous insulin that increase ectopic fat accumulation in liver and skeletal muscles and increase oxidative stress. The ectopically accumulated fat and its catabolites are thought to induce insulin resistance via various signaling pathways including mitogen-activated protein kinases (MAPKs), protein kinase C, IκB kinases, S6 kinases, and endoplasmic reticulum stress. Chronic insulin resistance increases the risk of diabetic macrovascular and microvascular complications. ^,

Autoimmunity

The nonobese diabetic (NOD) strain of mouse is an essential model of autoimmune T1DM. The advantage of this murine strain is that it develops spontaneous autoimmune diabetes, which shares many similarities with autoimmune type 1 diabetes in human patients. Recent research on this model has provided a wealth of insight into mechanisms likely involved in pathogenesis of T1DM. ^,

It is now generally accepted that T1DM arises from a breakdown in self-tolerance to beta cell autoantigens. Chronic T cell–mediated inflammation of the islets results in selective destruction of beta cells and sparing of the alpha, delta, and pancreatic polypeptide cells. Alternative scenarios are possible—for example, an adaptive immune response to persistent infection of the islets where beta cells are particularly sensitive to cytokine IL-1β–mediated killing, and increased expression of class I molecules during local infections may enhance their susceptibility. ^,

Autopsy data have shown that destruction is caused by infiltration of the islets by macrophages, dendritic cells, natural killer cells, and lymphocytes. The T cells are the key players in the autoimmune attack of beta cells, including helper T cells, cytotoxic T cells, and regulatory T cells. Humoral response and autoantibody production do not cause direct beta cell damage, but develop secondary to beta cell damage, and are useful disease markers.

Loss of Tolerance

Immune tolerance is essential to achieve immune homeostasis and self-tolerance. The loss of self-tolerance is the hallmark of T1DM pathogenesis. The establishment of tolerance starts in fetal life and includes both a central and a peripheral arm. Central tolerance is the process whereby immature T and β cells acquire tolerance to self-antigens during maturation within the thymus and bone marrow, respectively. It consists of positive and negative selection. Positive selection is the process of testing T cells for major histocompatibility complex (MHC) restriction. T cells with receptors with weak binding to MHC class I and II are allowed to survive (positively selected). This process is important in that it sets up a system in which all mature T cells will have T cell receptors (TCRs) that recognize antigens presented by MHC. Negative selection is the process whereby T cells that bind with high affinity to MHC class I and II, alone or carrying self-peptides, are eliminated by apoptosis. For central tolerance to be efficient, the negatively selecting stromal elements in the thymus medulla will have to express a large diversity of tissue-restricted antigens (TRAs) that represent as many self-antigens expressed outside of the thymus as necessary to establish and maintain self-tolerance. This is possible by promiscuous gene expression, which is the expression of a highly diverse set of genes in the medullary thymic epithelial cells, otherwise expressed in a strictly tissue-restricted fashion. Except for the involvement of the autoimmune regulator (AIRE), the molecular and cellular regulation of this gene expression pattern is poorly understood. The absence of a single TRA is sufficient to elicit spontaneous autoimmunity.

Under this model, thymus dysfunction could lead to a decrease in the expression of T1DM-related antigens promoting a continuous enrichment of the peripheral T cell repertoire with self-reactive T cells, as well as a decrease in the selection of specific T regulatory cells (Tregs). Thymus transplantation from diabetes-resistant to diabetes-prone rats can prevent insulitis and diabetes; conversely, transplantation of thymus from nonobese diabetic to diabetes-resistant mice induced insulitis. ^, All the members of the insulin gene family are expressed in the thymus. In mice, in which two genes code for (pro)insulin ( Ins1 and Ins2 ), Ins2 is predominantly expressed in the thymus, whereas Ins1 is dominant in the islet beta cells, which leads to a higher immune tolerance to Ins2. Ins2 ^−/− cogenic NOD mice have a significantly higher rate of insulitis and diabetes than Ins1 ^−/− cogenic NOD mice. ^,

Negative thymic selection is not entirely efficient and inadvertently permits the efflux of some autoreactive T cells with low-affinity TCR for self-antigens. To avoid the development of autoimmunity, additional mechanisms are in place. One of these mechanisms is the thymic generation of Tregs, which maintain homeostasis of the immune system and tolerance to self-antigens by controlling self-reactive T cells. The depletion of naturally occurring Tregs elicits multiorgan autoimmune disease, for example, the immunodysregulation polyendocrinopathy enteropathy X-linked (IPEX) syndrome—a rare congenital deficiency of forkhead box P3 (FOXP3) expression in humans. Tregs constitute 10% of CD4 ⁺ T cells in the thymus and the periphery and express the IL-2 receptor alpha chain (CD25) and FOXP3 protein. They are initially anergic, but when activated they suppress proliferation and IL-2 production of naïve and memory T cells.

Another mechanism allowing peripheral tolerance to self- antigens involves anergy. When a self-reactive lymphocyte recognizes its cognate antigen on a cell but does not receive the required co-stimulatory signal, it becomes anergized. The cell-surface glycoproteins CD80 (B7-1) and CD86 (B7-2) are essential co-stimulatory molecules, found almost exclusively on professional antigen-presenting cells (APCs). Interaction of these B7 molecules on APCs with CD28 on T cells is required for T cell activation. Moreover, if naïve T cells do become activated, they express an additional receptor called cytotoxic T lymphocyte–associated 4 (CTLA-4), which has a greater binding affinity for the B7 molecules than CD28. Binding of CTLA-4 to B7 results in a negative signal to the T cells, resulting in inhibition of T cell activity. CTLA-4 is also an important co-stimulatory molecule expressed by T-regulatory cells.

β cell tolerance occurs as a result of clonal deletion through apoptosis of immature β cells reactive to self-antigens. Immature β cells expressing surface IgM that reacts with self-antigens are rendered unresponsive or anergic. Thus, only those β cells that do not react with self-antigens in the bone marrow are allowed to mature and migrate to the periphery where further maturation occurs.

Autoantigens

Several autoantigens have been identified in T1DM and may play an important role in the initiation and progression of the autoimmune injury ( Table 3-1 ). Most of the autoantigens are human leukocyte antigen (HLA) A2 restricted, CD8 ⁺ T cell epitopes such as proinsulin, glutamic acid decarboxylase (GAD), islet-specific glucose-6-phosphatase catalytic subunit–related protein (IGRP), and islet amyloid polypeptide (IAPP). ^, The study of antigens in the development of T1DM is more complicated than initially thought and includes the following concepts: (1) intermolecular spreading of antigenicity; (2) tissue-specific cleavage producing antigenic peptides specific to beta cells; and (3) synergy of multiple islet antigens.

Table 3-1

Autoantigens and Autoantibodies

Autoantigen	Description of Antigen	Antibody
Insulin	Protein secreted by β cells	IAA
Glutamic acid decarboxylase	Enzyme catalyzing decarboxylation of glutamate to GABA	GADA
Insulinoma-associated protein 2	Neuroendocrine protein	IA-2A
Islet-specific glucose-6-phosphatase catalytic subunit–related protein	Catalytic subunit of glucose-6-phosphatase	IGRPA
Chromogranin A	Protein found in secretory granules in β cells	ChgAA
Zinc transporter 8	β cell–specific cation efflux zinc transporter	ZnT8A

GABA = γ-aminobutyric acid.

With regard to intermolecular spreading of antigenicity, we know that the initial antibody response in T1DM occurs primarily against insulin or GAD, spreading over time to other antigens. Tissue-specific cleavage appears critical to generation of diabetogenic autoantigens. A good example is the cleavage product of chromogranin A (ChgA)—WE-14, which is specifically recognized by the pathogenic BDC2.5 TCR (see later). It may also be necessary for T cells to target multiple beta cell antigens for the development of T1DM to occur (synergy of multiple antigens). For example, the targeting of IGRP by the CD8 ⁺ T cells is very diabetogenic, but only in the context of the T cells also targeting insulin peptide B:9-23.

Insulin

Insulin is composed of two peptide chains referred to as the A chain and B chain, which are linked together by two disulfide bonds, and an additional disulfide is formed within the A chain. The A chain consists of 21 amino acids, and the B chain of 30 amino acids. Proinsulin is the prohormone precursor to insulin. C peptide, a 31–amino acid peptide, is cleaved from proinsulin as it is enzymatically converted to insulin. One current leading hypothesis is that insulin itself may be the crucial autoantigen in T1DM. In NOD mice, a single amino acid mutation of insulin peptide 9-23 prevents development of diabetes. Recently, NOD studies have also shown that only APCs from islets are able to stimulate anti-B:9-23 T cells. Furthermore, knockouts of the insulin genes in NOD mice greatly influence progression to disease. In addition, the administration of insulin or its B chain can prevent or delay diabetes in susceptible mice during the prediabetic phase. Prospective studies, including the German BABYDIAB and the Finnish Diabetes Prediction and Prevention Study (DIPP), also indicate that autoantibodies against insulin usually emerge before any other antibodies, including anti-GAD65, anti–IA-2, and anti–zinc transporter 8 (ZnT8) ( Table 3-2 ).

Table 3-2

Prospective Cohort Studies of the Natural History of Type 1 Diabetes

	BABYDIAB (Germany)	DAISY (Colorado)	DIPP (Finland)	TEDDY (Four countries)
Year started	1989	1993	1994	2004
First-degree relatives (n)	1650 offspring	1120 offspring siblings	8150	923
General population (n)	–	1422		7754
Persistent islet Ab + (n)	149	183	537	450 *
Diabetes ( n )	47	71	320	126 ^†

Note: The BABYDIAB consists of offspring of parents with T1DM. DAISY has two groups: first-degree relatives of T1DM and high-risk individuals from the general population. DIPP screened infants in the general population, including first-degree relatives, for HLA types. Finally, the TEDDY cohort consists of newborns with a first-degree relative with T1DM as well as those from the general population enrolled from six clinical centers in four countries (personal communication from Ziegler, Simell, and Rewers, October 2011).

Ab+ = Autoantibody positive.

* As of October 2012, 800 cases expected by 15 years of follow-up.

† As of October 2012, 400 cases expected by 15 years of follow-up.

It is astonishing that proinsulin, a protein of only 86 amino acids, contains so many epitopes for a spectrum of HLA class I and class II alleles. T cell reacting epitopes have been demonstrated within insulin A and B chains, ^, the C peptide and B-C chain junction, and the C-A chain junction region. Specific CD4 ⁺ and CD8 ⁺ T cells targeting insulin and precursor epitopes has been reported both in newly diagnosed and in chronic T1DM. CD4 ⁺ T cell reactivity has been noted mostly in connection with susceptibility alleles HLA-DR3 and HLA-DR4. CD8 ⁺ T cell reactivity, on the other hand, is associated in particular with HLA-A2.

Glutamic Acid Decarboxylase

GAD is an enzyme that catalyzes the decarboxylation of glutamate to γ-aminobutyric acid (GABA) and CO ₂ . GAD isoforms GAD67 and GAD65 are encoded by two different genes, GAD1 and GAD2. GAD2 is expressed in the pancreas and the brain. The association of GAD65 autoantibodies with T1DM is well known, ^, and these autoantibodies were found in 52% of newly diagnosed children in the U.S. SEARCH for Diabetes in Youth study. More recently, GAD has been used as a tolerogenic vaccine to preserve functional beta cells. Unfortunately, GAD65 antigen therapy did not significantly improve clinical outcomes over a 15-month period in a recent randomized controlled trial (RCT).

Insulinoma-Associated Protein 2

Insulinoma-associated protein 2 (IA-2) is a neuroendocrine protein and a member of the tyrosine phosphatase family, with its gene located on chromosome 14. Antibodies against IA-2 appear later than antibodies against insulin or GAD and are associated with progression to diabetes. Ellis and coworkers found anti–IA-2 antibodies in the sera of 58% of newly diagnosed T1DM patients ; and, similarly, Dabelea and colleagues identified IA-2A in 60% of newly diagnosed children in the SEARCH cohort.

ZnT8

ZnT8 is a novel autoantigen in T1DM. It is a beta cell–specific cation efflux zinc transporter with its gene located on chromosome 8. The Diabetes Autoimmunity Study in the Young (DAISY) cohort consists of two groups, including first-degree relatives of individuals with T1DM and individuals from the general population who underwent HLA typing of cord blood, and has shown that ZnT8 antibodies are present in 60% to 80% of newly diagnosed T1DM patients (see Table 3-2 ). In addition, 25% of T1DM patients who are negative for autoantibodies to insulin, GAD, IA-2, and islet cells (ICA) tested positive for ZnT8 autoantibodies. Moreover, the DAISY study showed that ZnT8 autoantibodies emerge later than the other insulin autoantibodies. The antibodies have also been shown to decline quickly after diagnosis of T1DM. Howson and colleagues found an association between anti-ZnT8 antibodies and the single-nucleotide polymorphisms (SNPs) rs7522061 and rs9258750A > G in, respectively, the Fc-receptor-like–3 (FCRL3) gene and HLA class I locus.

Islet-Specific Glucose-6-Phosphatase Catalytic Subunit–Related Protein

IGRP is an important autoantigen that is selectively expressed in beta cells. IGRP is recognized as an antigen by the CD8 ⁺ T cell clone NY8.3. IGRP is not expressed in the thymus in NOD mice, thereby allowing IGRP-reactive T cells to escape into the periphery. For this reason, peripheral tolerance independently confers protection against autoimmunity. ^, Krishnamurthy and colleagues successfully showed that peripheral tolerance alone is sufficient to protect NOD8.3 mice from autoimmune diabetes.

Chromogranin A

The chromogranin A (ChgA) gene is located on chromosome 14 and encodes a protein found in secretory granules of many different secretory cell types, including beta cells. It is a precursor pro-protein that is proteolytically processed within the granule to form a variety of peptides, including vasostatin 1 (VS-1; ChgA 1-76, ChgA 29-42), VS-2 (ChgA 1-113), and WE-14 (ChgA 358-371). The functions of these peptides are still not clearly understood. As previously mentioned, ChgA illustrates the importance of tissue-specific cleavage. In fact, a specific cleave of ChgA within the beta cells is essential for T cell binding. The cleavage product WE-14 is tissue-specific, meaning it is specifically produced in the islet cells and recognized by pathogenic BDC2.5 TCR of NOD mice via pockets 5 through 9 of the I-Ag7 MHC class II molecule. Nikoopour and colleagues identified another cleavage product, ChgA 29-42 peptide, as the natural epitope of BDC2.5 CD4 ⁺ T cells, and demonstrated induction of diabetes after transfer of ChgA 29-42 activated BDC2.5 splenocytes into NOD/severe combined immunodeficiency (SCID) mice.

In contrast, cells expressing noncleaved ChgA do not effectively produce these autoantigenic peptides and are thus not recognized by T cells. For example, production of WE-4 cleaves four N-terminal amino acids that, if present, would occupy pockets 1 to 4 on the I-Ag7 cMHC class II molecule and thereby block BDC2.5 TCR stimulation.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here