Type 1 Diabetes: an investigation of its causes and pathology.
Introduction
1.1 Background to T1D
1.1.1 – Epidemiology and background
Type 1 diabetes (T1D), formerly known as insulin-dependent diabetes, is a common autoimmune disease that comprises five to ten per cent (ca. 11–22 million people) of all diabetes cases worldwide (World Health Organisation, WHO). Worryingly, incidences of the disease appear to be escalating rapidly in Western countries (1), increasing by three per cent per year (2). If current trends continue, it is predicted that more than 50/100,000 people in Sardinia and Finland will develop the disease (3, 4), making T1D a larger socioeconomic burden than it already is. Although estimates vary with regards to the number of adult T1D cases diagnosed (5), new cases typically present between the ages of five and seven, with a second peak in diagnoses observed during adolescence (6). Incidence is seasonal with more cases diagnosed in winter (7), and there is a pronounced north-south gradient in prevalence (4, 8).
Although T1D most likely comprises a group of heterogeneous disorders with different underlying causes (5), it uniformly involves the autoimmune-mediated destruction of pancreatic beta (β) cells. Accumulating evidence in the 1970s that T1D was an autoimmune disease came from clinical, genetic and immunological observations (9-11): the immunosuppressive drugs cyclosporin A and anti-thymocyte globulin can positively impact T1D progression (12, 13). In addition, when T1D patients received a pancreatic isograft from a genetically identical twin, T-cell infiltration into the donor tissue correlated with a decline in graft function (14), suggesting that T-cells contributed to T1D pathology. Furthermore, the presence of anti-islet cell autoantibodies added support to the notion that inappropriate immune responses contributed to β-cell destruction, even though autoantibodies may be present for extended periods before clinical T1D manifests (15, 16). These and several subsequent observations led to a model for the natural history of T1D (Figure 1.1), which suggests that genetically at-risk individuals, exposed to unknown environmental factors such as viral infections, proceed to autoimmunity (5, 9). Importantly, however, although the immune system is clearly implicated in the development and progression of T1D, and many molecular mechanisms have been characterised, the precise initiating events of β-cell dysfunction and their progressive destruction are variable and complex: genetic, environmental and stochastic factors contribute to T1D (17).
1.1.2 – The targets of the autoimmune response: pancreatic β-cells
β-cells, the major source of insulin in the body, are located in the endocrine islets of Langerhans, where they constitute 70–80 per cent of the cells: islets themselves make up approximately one to two per cent of total pancreatic mass (18) and it is estimated that a single human islet contains approximately 1,140 β-cells (19).
Insulin is a hormone with wide-ranging effects, and a lack of β-cell function leads to insulin insufficiency. Insulin stimulates the uptake of glucose from the blood by skeletal muscle and fat cells, as well as the synthesis of glycogen by the liver: glucose is important in multiple metabolic paths, and the regulation of extracellular glucose levels by insulin and glucagon is critical (20). Glucose uptake is followed by its conversion to glycogen and triglycerides, which allows the body to stop using fat reserves for energy (20). Glucagon has opposing effects and stimulates the breakdown of glycogen to glucose: insulin and glucagon work in tandem to control the flow of fuels in the body (20). In the absence of insulin, glucose uptake ceases, allowing it to reach toxic levels in the blood (hyperglycaemia), which can cause acute ketoacidosis, as well as neuropathy, nephropathy and cardiovascular problems if left untreated (www.mayoclinic.com). In fact, until insulin was discovered to be essential for blood glucose homeostasis, T1D was a lethal disease (21). Ultimately, a lack of β-cells and the resulting hyperglycaemic state is used to diagnose T1D.
In addition to insulin, β-cells produce C-peptide (a short polypeptide that connects the insulin A and B chains before it is removed during insulin synthesis) and amylin. C-peptide is secreted into the blood in equimolar quantities to insulin and may have several physiological functions (22), whilst amylin works synergistically to insulin and slows the rate at which glucose enters the blood (23). Importantly, C-peptide levels can be monitored to evaluate β-cell function (24).
1.1.3 – The clinical diagnosis of T1D
Patients can present with various symptoms prior to diagnosis, including polyuria, polydipsia, polyphagia, and weight loss, which normally accompany overt hyperglycaemia (5). Furthermore, numbness in the extremities, recurrent infections and ketoacidosis are hallmarks of the disease (WHO). T1D is diagnosed by combining the classic symptoms of hyperglycaemia with an abnormal blood glucose test: a plasma glucose concentration of ≥ 7 mmol/L or ≤ 11.1 mmol/L, two hours after a 75 g glucose drink (WHO). In addition to blood glucose levels, HbA1c and C-peptide levels can be monitored to help identify metabolic abnormalities during diagnosis (WHO).
Intriguingly, some patients may be diagnosed before complete β-cell destruction and insulin deficiency, and some newly diagnosed patients enter the ‘honeymoon period’ shortly after diagnosis (25). During this time of metabolic stabilisation, some patients are able to discontinue the use of exogenous insulin as endogenous insulin secretion improves, perhaps due to compensatory mechanisms. However, this period commonly reverts to re-dependence on exogenous insulin (6). Indeed, the amount of β-cell function at the time of diagnosis is thought to vary considerably (5), which complicates the examination of the molecular causes of β-cell destruction.
In addition to abnormal blood-glucose levels and the clinical manifestations mentioned, more than 90 per cent of T1D patients test positive for at least one autoantibody, and the presence of two or more autoantibodies helps identify relatives of T1D patients at high risk for the disease (6, 16). To date, several islet autoantigens have been identified as autoantibody targets (26, 27), and they still represent the best biomarker for disease development, as discussed later.
Although T1D detection and treatment has improved in recent decades, there is no known prevention or cure. Treatment is primarily based on treating the symptoms and on the prevention or delay of associated complications. Various regimens of life-long insulin treatment are available and essential for maintaining blood-glucose control (20). In industrialised countries, self-monitoring and insulin administration is common and well established: insulin infusion pumps are now commonly used to monitor blood glucose levels and assist the patient with insulin administration. Modern healthcare systems have also improved the early detection of associated complications, such as peripheral neuropathy (28) and cardiovascular complications (29), reducing the toll on both the individual and social welfare systems. The more complicated treatment options of islet or pancreas transplantation exist to varying degrees, although the complications associated with these procedures makes them less favourable.
The numerous acute and chronic complications associated with T1D emphasise the need to develop a more complete understanding of the disease in order to establish preventative treatments and cures. With this aim, detailed mechanistic studies in human and murine models as well as efforts to elucidate biomarkers, will guide our understanding of T1D immunopathology, aid patient stratification and tailor treatments. Some potential therapeutic approaches in T1D are outlined in Section 1.4.
In the remainder of this chapter, I will summarise our current understanding of the causes and pathology of T1D, to provide a rational for the research performed during my PhD. I will present specific introductions to both PTPN22 and the IL-2 pathway at the beginning of the respective results chapters.
1.2 Environment, genetics and stochastic factors contribute to T1D
1.2.1 – What is autoimmunity?
The Merck Manual describes autoimmunity as ‘a malfunction of the body’s immune system that causes the body to attack its own tissue’. In T1D, β-cell destruction likely involves antigen-specific T-cells and their antigen-presenting cells (APCs), as well as a diverse range of other immune lineages and their varying effector modalities: a complex process.
1.2.1.1 – An overview of the immune system
The immune system is a defence mechanism, able to employ soluble, cellular and physical mechanisms to defend the host (30). The first defences are the physical barriers; the skin and mucous membranes (31). The cellular and soluble vertebrate immune system has been conceptually divided into the innate and adaptive immune systems (32), which carry out different, yet overlapping functions. When a pathogen breaches the physical barriers, the innate immune system is generally engaged to control infection, whilst the adaptive immune system may remain quiescent (30). In the event of pathogen evasion of the innate immune system (30), or under direct instruction from the innate immune system (33), the adaptive immune system will become engaged.
The innate, often mislabelled the non-specific, immune system is phylogenetically older: all multicellular organisms have some form of innate immune system that likely evolved under the selective pressure imposed by infectious organisms (34). It is the primary mode of defence in plants, fungi and insects (30). Said to be innate, it is poised to respond rapidly to certain evolutionarily-conserved signatures of microbes (31). In vertebrates, it exists primarily as two branches: the complement and cellular branches. The complement system consists of a group of proteins (more than 25 soluble and cell surface receptors), so named for their ability to aid antibody neutralisation and activate the cells required for pathogen clearance: it induces/aids chemotaxis, opsonisation, cell lysis and antibody aggregation (30). It has recently been shown that components of the complement system in serum can be used to discriminate between T1D patients and healthy controls (35), suggesting that activation of the innate immune system may contribute to pathology.
The cellular arm of the innate immune system consists of: natural killer cells (NK-cells), mast cells, macrophages, neutrophils, dendritic cells (DCs), basophils, eosinophils, and possibly γδT-cells and NKT-cells, although some of these lineages display properties characteristic of adaptive immune system cells (30).
In general terms, the cells of the innate immune system utilise pattern recognition receptors (PRRs) to recognise and process foreign material before it can mediate damage to the host: many cells directly phagocytose foreign matter (30). Many innate PRRs recognise the conserved moieties of pathogens, such as those typically found in the bacterial cell wall – lipopolysaccharides (LPS), for example (31). Various families of PRRs exist (36) and can be membrane bound or cytoplasmic. For example, various Toll-like and C-type lectin receptors recognise specific pathogen-associated molecular patterns (PAMPs), such as mannose-containing carbohydrates or bacterial flagellin, on the cell surface or inside endosomal compartments; leading to the activation of the NFκB and MAPK pathways and the secretion of pro-inflammatory cytokines (31). In addition to cell surface receptors, cytoplasmic PRRs such as NOD-like receptors (NLRs), RIG-I-like receptors (RLRs) and RNA helicases, are able to recognise PAMPs present in the cytoplasm, such as viral RNA and other bacterial components (36). Interestingly, genetic variation within PRR genes has been associated with risk for autoimmunity: NOD2 (37) and IFIH1 (38) are excellent examples.
However, these characteristics directly contribute to the major limitation of the innate immune system. As the PRRs are genetically hard-wired into the germ line of the organism, evolving pathogens can circumvent them. It is postulated that the adaptive immune system evolved for this reason approximately 500 million years ago (39).
The adaptive immune system is built upon the variable use of germ-line encoded gene segments to form countless novel antigen receptors, theoretically able to recognise any epitope (30). In this way, the adaptive immune system can mount a response against a pathogen that has evolved to avoid detection by the innate PRRs, for example. These variable antigen receptors are expressed on T- and B-cells, as the T- or B-cell antigen receptor, TCR or BCR, respectively: each lymphocyte clone expresses a single receptor specificity (30). The TCR recognises processed peptide antigen in the context of either MHC class I (CD8+ T-cells), or class II (CD4+ T-cells) expressed on other cells, including the antigen-presenting cells of the innate immune system (30). In contrast, the BCR recognises whole antigen, as it is the membrane-bound form of the antibody molecule produced and secreted from mature B-lymphocytes during humoral immune responses (30). It is important to note that all cells express MHC class I, allowing T lymphocytes to monitor for virally infected or tumour cells. Furthermore, an important feature of the adaptive immune system is immunological memory: antigen-specific T and B lymphocytes are able to clonally expand and persist after the initial infection has resolved, allowing the adaptive immune system to quickly recognise previously encountered antigens on subsequent challenge (30).
However, a balance exists: as the antigen receptors of the adaptive immune system are randomly generated and are limitless in their target specificity, they are able to recognise and target self-peptides. Although numerous mechanisms exist to avoid autoreactivity, self-reactive T- and B-lymphocytes exist in the periphery of healthy adults (30, 40). Consequently, several regulatory systems are in place to prevent self-directed immunity that in the context of autoimmunity, break down.
1.2.1.2 – Mechanisms of tolerance and autoimmunity
Autoimmunity is established after a break in tolerance, the point at which reactivity to self overcomes regulatory mechanisms. How self-directed immunity is established, and why certain leukocyte populations are so destructive in some individuals and not others, is not fully understood.
Joshua Lederberg proposed in 1959 that ‘tolerance to self’ was a property acquired by the developing lymphocyte upon antigen encounter (41), implicating early developmental check-points in lymphocytes as essential to a functional adult immune system (42). Since then, sophisticated work has shown that lymphocytes reactive to self-peptide are deleted during their development (43, 44). In the case of T-cells, once a T-cell has matured to the point of expressing a newly generated TCR on its surface, it undergoes a screening process in the thymus to determine whether it is likely to react against self-antigen in peripheral tissues (30): TCRs with too high an affinity for self-peptides, presented on medullary epithelial cells under the control of the autoimmune regulator (AIRE) (45), are deleted in a process called clonal deletion, whilst non-reactive cells exit into the peripheral tissues (30). Importantly, a population of T lymphocytes with intermediate-affinity for self-antigens, FOXP3+ regulatory T-cells (Tregs), are not induced to undergo apoptosis during thymic selection but egress into the periphery to mediate peripheral tolerance (46), as discussed later.
In contrast to T-cells, B-cells develop and mature in the bone marrow, where they are screened for self-reactivity and deleted accordingly, although receptor editing allows B-cells to try again and form a non-self-reactive BCR (30).
As mentioned, all individuals possess a certain number of autoreactive T-cell clones (47) – thymic selection is not perfect in the removal of T-cells with an affinity to self (48) – and it is clear that mutations in AIRE and other genes involved downstream of the TCR, such as ZAP70, can influence the success of thymic selection (42, 45, 49, 50). Therefore, as autoreactive cells exist in the periphery, regulatory mechanisms are in place to constrain self-directed immunity.
Peripheral tolerance is achieved through a variety of mechanisms (51), including the suppression of autoreactivity and on-going immune responses by Tregs, discussed in Chapters Three and Four. In addition, anergy can be induced in lymphocytes recognising an antigen in the absence of appropriate co-stimulatory signals, or in the presence of co-inhibitory signals when inflammatory stimuli are present (30), and extrathymic AIRE-expressing cells can induce CD4+ T-cell inactivation in mice (52).
The further characterisation of two disease-relevant pathways that may impinge upon a break in tolerance and/or T1D progression, have been the focus of my research.