Variation of diabetes incidence over time in countries with different incidence levels and socio-economic conditions, and in an age span beyond the childhood years, may give clues for diabetes causes. Materials: Data from prospective type 1 diabetes registers in Sweden and Lithuania in children (0-14 years) and young adults (15-34 and 15-39 years, respectively). Number of infections recorded in health care booklets (117 cases; 270 controls); interview about the dietary intake one-year before the diagnosis and routinely recorded growth data (99 cases; 180 controls). Results: The incidence of type 1 diabetes in Sweden and Lithuania differed most in the younger age groups, 28.9 and 7.5/100,000/year in 0-14-year group, respectively. During 1983-2000 incidence increased in 0-14-year old children in both countries, but the pattern of change differed. During 1983-1998 the incidence increased in Swedish children, but tended to decrease in young adults, with no increase in the age group below 35 years, indicating that the increase of childhood diabetes may be due to a shift towards a younger age at diagnosis. Within a low-incidence country Lithuania there was an urban-rural gradient of incidence, especially in the younger age groups, that seemed to follow poverty distribution: incidence in the 0-39-year group was 7.1, 9.0 and 8.8/100,000/year in rural areas, towns and cities, respectively, p<0.001. Exposure to one or more non-specific infection during the first half-year of life reduced diabetes risk: odds ratios (95%-CI) in 0-14 and 5-14-year groups were (0.60; 0.37-0.98) and (0.47; 0.26-0.87), respectively. Higher energy intake and weight-for-age were independent diabetes risk factors: odds ratios for medium and high levels of energy were 1.33 (0.52-3.42) and 5.23 (1.67-16.38), and for weight-for-age 3.20 (1.30-7.88) and 3.09 (1.16-8.22), respectively. High intake of carbohydrates, disaccharides and sucrose in particular, increased diabetes risk independently of the high intake of energy. Conclusion: Environmental factors associated with socio-economic conditions in childhood may be important for the occurrence of type 1 diabetes. Lack of exposure to microbial antigens early in life, higher intake of energy and more rapid growth may contribute to the increase of childhood-onset diabetes observed in many countries.

Type 1 diabetes is resulting from the selective destruction of insulin-producing beta-cells within the pancreatic islets. Somatostatin acts as an inhibitor of hormone secretion through specific receptors (sst1-5). All ssts were expressed in normal rat and mouse pancreatic islets, although the expression intensity and the co-expression pattern varied between ssts as well as between species. This may reflect a difference in response to somatostatin in islet cells of the two species. The Non-Obese Diabetic (NOD) mouse model is an experimental model of type 1 diabetes, with insulitis accompanied by spontaneous hyperglycaemia. Pancreatic specimens from NOD mice at different age and stage of disease were stained for ssts. The islet cells of diabetic NOD mice showed increased islet expression of sst2-5 compared to normoglycemic NOD mice. The increase in sst2-5 expression in the islets cells may suggest either a contributing factor in the process leading to diabetes, or a defense response against ongoing beta-cell destruction. Somatostatin analogues were tested on a human endocrine pancreatic tumour cell line and cultured pancreatic islets. Somatostatin analogues had an effect on cAMP accumulation, chromogranin A secretion and MAP kinase activity in the cell line. Treatment of rat pancreatic islets with somatostatin analogues with selective receptor affinity was not sufficient to induce an inhibition of insulin and glucagon secretion. However, a combination of selective analogues or non-selective analogues via co-stimulation of receptors can cause inhibition of hormone production. For insulin and glucagon, combinations of sst2 + sst5 and sst1 + sst2, respectively, showed a biological effect. In summary, knowledge of islet cell ssts expression and the effect of somatostatin analogues with high affinity to ssts may be valuable in the future attempts to influence beta-cell function in type 1 diabetes mellitus, since down-regulation of beta-cell function may promote survival of these cells during the autoimmune attack.

Transplanting pancreatic islets into the liver through the portal vein is currently the most common procedure in clinical islet transplantations for treating patients with brittle type 1 diabetes. However, most islet grafts fail within a 5-year period necessitating retransplantation. The vascular connections are disrupted at islet isolation and implanted islets depend on diffusion of oxygen and nutrients in the immediate posttransplantation period. Rapid and efficient revascularization is of utmost importance for the survival and long-term function of transplanted islets. In this thesis, the influence of the implantation microenvironment for islet engraftment and function was studied. Islets were transplanted into the liver, the renal subcapsular site or the pancreas. Islets implanted into the liver contained fewer glucagon-positive cells than islets implanted to the kidney and endogenous islets. Intraportally transplanted islets responded with insulin and glucagon release to secretagogues, but only when stimulated through the hepatic artery. Thus, the intrahepatic grafts were selectively revascularized from the hepatic artery. The vascular density in human islets transplanted into the liver of athymic mice was markedly lower when compared to human islets grafted to the kidney. Islets implanted into their physiological environment, the pancreas, were markedly better revascularized. Insulin content, glucose-stimulated insulin release, (pro)insulin biosynthesis and glucose oxidation rate were markedly decreased in transplanted islets retrieved from the liver, both when compared to endogenous and transplanted islets retrieved from the pancreas. Only minor changes in metabolic functions were observed in islets implanted into the pancreas when compared to endogenous islets. The present findings demonstrate that the microenvironment has a major impact on the engraftment of transplanted islets. Elucidating the beneficial factors that promote engraftment would improve the survival and long-term function of transplanted islets. Ultimately, islet transplantation may be provided to an increased number of patients with type 1 diabetes.

Type 1 diabetes (T1D) is a chronic autoimmune disease targeted against the pancreatic β-cells. Proinflammatory cytokines are considered to play a major role in the destruction of the insulin-producing β-cells. This thesis studied strategies to counteract proinflammatory cytokines in experimental T1D. Both animal models for T1D as well as β-cell preparations exposed in vitro to putative noxious conditions were examined. In the first study we observed that cytokine treatment of mouse pancreatic islets lacking inducible nitric oxide synthase (iNOS) induced a prolongation of the early stimulatory phase of glucose stimulated insulin secretion. Various experiments led to the conclusion that this prolonged stimulatory effect may involve the DAG/PLD/PKC pathway. Next, we transplanted mouse islets deficient in iNOS to spontaneously diabetic NOD mice. We observed a normalization of hyperglycemia but not a delayed allograft rejection compared to transplanted wild type islets. Thus, absence of iNOS in the graft was not sufficient to prolong allograft survival. In paper III we found that sustained glucose stimulation of rat pancreatic islets was coupled to a decreased conversion of proinsulin to insulin. Islet treatment with IL-1β was also coupled to a decreased proinsulin conversion. Islet proconvertase activity may be a target in islet damage. In paper IV prolactin (PRL) was administered to mice in the multiple low dose streptozotocin model and we observed that PRL enhanced a Th2 response. This may contribute to the protective action by PRL in this model of autoimmune T1D. Finally, by examining β-cells overexpressing Suppressor of cytokine signalling 3 (SOCS-3) it was found that this could inhibit IL-1β induced signalling through the NF-κB and MAPK pathways. SOCS-3 overexpression also inhibited apoptosis induced by cytokines in primary β-cells. Lastly, we demonstrated that SOCS-3 transgenic islets were protected in an allogeneic transplantation model.

The nonobese diabetic mouse (NOD) is an excellent animal model to study type 1 diabetes. As with some humans, disease in the NOD mouse is effected by a combination of genetic and environmental factors. At least 20 insulin dependent diabetes (Idd) susceptibility loci have been identified so far, both in humans and in the NOD mouse. In this thesis, the overall aim has been to understand the genetic basis of diabetes in the NOD mouse by assessing immunogically-related phenotypes. As lymphocytes are the main players in the onset and progression to overt diabetes, we searched for physiological abnormalities in T and B cells, which could contribute to the breakdown of tolerance to pancreatic antigens. Ultimately, we postulate that abnormalities in the T or B cell compartments, under the genetic control of a previously defined diabetes susceptibility regions (Idds) could unravel the biological mechanisms underlying diabetes susceptibility and facilitate the identification of etiological polymorphisms involved in the disease. NOD T cells are defective in upregulating CTLA-4 upon in vitro activation. Previous studies have shown that this defect is, at least in part, controlled by gene(s) in the Idd5 region on chromosome 1. In paper I, we provide evidence that defective upregulation of the CTLA-4 in NOD T cells is not controlled by the Idd5.1 and Idd5.2 regions, but rather by genes linked to the telomeric region of chromosome 1 and to the Idd3 locus, for which the prime candidate gene is Il-2. Interestingly, we could restore some of the defective CTLA-4 expression in NOD T cells by the addition of exogenous IL-2 during T cell activation in vitro. In paper II, we show that NOD thymocytes are resistant to superantigen-mediated negative selection and that this trait is under control of the Idd5.2 region. Interestingly, it appears to operate in a T cell non-autonomous manner. In paper III, we describe a competitive advantage of NOD thymocytes to mature when they co-develop with B6 thymocytes in embryo aggregation chimeras. These results imply that defects exist in the positive/negative selection mechanisms in the NOD thymus. Apart from T cells, B cells also play an important role in the initiation of diabetes in NOD mice, probably as antigen presenting cells. In paper IV, we report that the genetic basis of an enlarged marginal zone (MZ) B cell population observed in the NOD mice is linked to the Idd9/Idd11 region. Together, these findings contribute to the dissection of the molecular mechanisms underlying diabetes pathogenesis, and shed light on the contribution of central and peripheral tolerance mechanisms to this process.

Insulin pump treatment (CSII) is considered the most physiological way to imitate the healthy body’s insulin profile in adolescents with diabetes. However, despite the use of CSII, achieving the recommended disease control is difficult for adolescents. The aim of this thesis was to explore aspects of self-management of diabetes in adolescents using insulin pumps in order to describe conditions contributing to the recommended disease control. Three methods of bolusing (normal, dual-wave and square-wave) in connection with pasta meals were tested in a crossover study among 15 adolescents with diabetes to assess whether one method was superior in managing glucose levels. A cross-sectional study among 90 adolescents being treated with CSII was conducted to investigate the management of CSII, including the administration of bolus doses. Two qualitative interview studies, based on the grounded theory method, were performed to gain insight into the processes involved in taking bolus doses and to investigate reasons for missed bolus doses and strategies for avoiding missing them. Twelve adolescents, four parents and one diabetes specialist nurse were interviewed. No method of bolusing was found to be superior in managing the glucose levels after these meals. The post-prandial glucose peaks were <10 mmol/L, in 48% of the cases, regardless of bolus methods. This indicates that adolescents can be encouraged to individually test which bolus method gives them the most normal post-prandial glucose levels. The cross-sectional study showed that adolescents were satisfied with CSII, but that 38% had missed more than 15% of the bolus doses the day under study. The frequency of bolus doses correlated with the disease control. Findings from the interview study revealed the need to clarify the responsibility for diabetes self-management in continuous negotiation between adolescents and parents to avoid insulin omission. The main reason for missed boluses was lost focus, and the strategies for remembering them were agreements involving reminders. The thesis describes that individual dose testing, clarification of responsibility and agreements involving reminders are conditions contributing to the recommended disease control. The thesis also describes that lost focus and a lack of responsibility can lead to insulin omission and be a hindrance to achieving disease control. / Anna Kernell avled maj 2010.

Type 1 diabetes (T1D) is an auto-immune disease characterized by insulin-deficiency. Insulin is a metabolic hormone that is involved in lowering blood glucose (BG) levels in order to control BG level to a tight range. In T1D this glycemic control is lost, causing chronic hyperglycemia (excess glucose in blood stream). Chronic hyperglycemia damages vital tissues. Therefore, glycemic control must be restored. A common therapy for restoring glycemic control is intensive insulin therapy, where the missing insulin is replaced with regular insulin injections. When dosing this compensatory insulin many factors that affect glucose metabolism must be considered. Linkura is a company that has developed tools for monitoring the most important factors, which are meals and exercise. In the Linkura meal and exercise tools, the nutrition content in meals and the calorie consumption during exercise are estimated. Another tool designed to aid control of BG is the bolus calculator. Bolus calculators use input of BG level, carbohydrate intake, and insulin history to estimate insulin need. The accuracy of these insulin bolus calculations suffer from two problems. First, errors occur when users inaccurately estimate the carbohydrate content in meals. Second, exercise is not included in bolus calculations. To reduce these problems, it was suggested that the Linkura web tools could be utilized in combination with a bolus calculator. For this purpose, a bolus calculator was developed. The bolus calculator was based on existing models that utilize clinical parameters to relate changes in BG levels to meals, insulin, and exercise stimulations. The bolus calculator was evaluated using data collected from Linkura's web tools. The collected data showed some inconsistencies which cannot be explained by any model.  The performance of the bolus calculator in predicting BG levels using general equations to derive the clinical parameters was inadequate. Performance was increased by adopting an update-algorithm where the clinical parameters were updated daily using previous data. Still, better model performance is prefered for use in a bolus calculator.   The results show potential in developing bolus calculator tools combined with the Linkura tools. For such bolus calculator, further evaluation on modeling long-term exercise and additional safety features minimizing risk of hypoglycemia are required.

Studies demonstrate localized neuro-retinal dysfunction in patients with diabetes and no visible diabetic retinopathy (DR). Poor glycemic control is a strong risk factor for DR. We hypothesized that poor glycemic control predicts increased areas of localized neuro-retinal dysfunction in patients with diabetes. Forty-eight adolescents with diabetes and 45 controls were tested using the standard (103 hexagons) multifocal electroretinogram (mfERG). Negative binomial regression analysis was conducted with number of abnormal hexagons (delayed responses) as the dependent variable and glycated hemoglobin (HbA1c), disease duration, age and sex as covariates. Results indicate that a one-unit increase in HbA1c predicts an 80% (p = 0.002) increase in the number of abnormal hexagons when controlling for age. Increased areas of neuro-retinal dysfunction are predicted by worsening glycemic control in patients with no visible DR. Standard mfERG may be useful in monitoring patients with diabetes and identifying those who may be at risk of developing DR.

Studies demonstrate short-wavelength cone pathway dysfunction in patients with diabetes and no clinically visible DR. Poor glycemic control, as measured by hemoglobin A1c (HbA1c), is a strong risk factor for DR. We hypothesized that raised HbA1c was associated with short-wavelength cone sensitive visual evoked potential (S-VEP) and electroretinogram (sERG) dysfunction. Forty adolescents with diabetes and 39 controls were tested using the S-VEP. Latencies to a short-wavelength stimulus were delayed in patients at low contrasts. Patient S-VEP latencies were not associated with HbA1c when controlling for age and time since diagnosis. Twenty-one adolescents with diabetes and 19 controls were tested using the sERG. Implicit times of the b-wave were delayed but not associated with HbA1c when controlling for time since diagnosis.Patient PhNR amplitudes were reduced. A one-unit increase in HbA1c was associated with a 15% sERG PhNR amplitude reduction (p=0.004). The sERG PhNR may be a potential biomarker for DR.

Studies demonstrate localized neuro-retinal dysfunction in patients with diabetes and no visible diabetic retinopathy (DR). Poor glycemic control is a strong risk factor for DR. We hypothesized that poor glycemic control predicts increased areas of localized neuro-retinal dysfunction in patients with diabetes. Forty-eight adolescents with diabetes and 45 controls were tested using the standard (103 hexagons) multifocal electroretinogram (mfERG). Negative binomial regression analysis was conducted with number of abnormal hexagons (delayed responses) as the dependent variable and glycated hemoglobin (HbA1c), disease duration, age and sex as covariates. Results indicate that a one-unit increase in HbA1c predicts an 80% (p = 0.002) increase in the number of abnormal hexagons when controlling for age. Increased areas of neuro-retinal dysfunction are predicted by worsening glycemic control in patients with no visible DR. Standard mfERG may be useful in monitoring patients with diabetes and identifying those who may be at risk of developing DR.