Final Report Summary - BETA-JUDO (Beta-cell function in juvenile diabetes and obesity)
Executive Summary:
The project “Beta-cell function in Juvenile Diabetes and Obesity” (Beta-JUDO) aimed at investigating the role of the insulin-producing beta-cell in development of obesity and type 2 diabetes mellitus (T2DM) in children. The rationale behind the project was that accentuated insulin response (insulin hypersecretion) is an early phenomenon promoting development of overweight and insulin resistance, since high insulin levels favours storage rather than combustion of energy. Individuals with this secretory phenotype are therefore at risk of developing obesity and T2DM. The project was organised in a translational approach involving studies of obese children and beta-cells from isolated human islets of Langerhans. In Beta-JUDO, paediatric obesity clinics and academic centres with focus on treatment of obese children, beta-cell biology, protein and transcript profiling, genetics, and bioinformatics formed a consortium together with a large drug company and small companies (SMEs) specialized on biomarker discovery and clinical trials.
The overall aim of the project was to identify new pharmacological approaches to normalize the high insulin levels observed in obese children and children of risk to become obese. The aim was addressed by, firstly, defining factors and molecular mechanisms that contribute to insulin hypersecretion, and secondly, to address ways to counteract insulin hypersecretion by investigating pharmacological principles potentially targeting insulin hypersecretion in both isolated islets and in obese children.
Factors contributing to insulin hypersecretion included high circulating levels of the free fatty acid (FFA) palmitate, which were high in obese children with accentuated insulin hypersecretion. Correspondently, in isolated islets such elevated palmitate levels caused accentuated glucose-stimulated insulin secretion. The findings supported a role of elevated levels of palmitate as an underlying factor contributing to insulin hypersecretion. We found that palmitate enhanced islet respiration, which was closely linked with islet insulin hypersecretion. In obese children raised glucagon levels were compatible with a role of altered glucagon response in obese children contributing to hyperinsulinemia. In genetic studies using large obesity cohorts we identified a number of genes that are associated with insulin hypersecretion, such as the gene PCSK1 that is involved in insulin processing. Further underlying factors for islet insulin hypersecretion were found by generating extensive protein and transcript expression patterns.
Strategies to counteract insulin hypersecretion included free fatty acid receptor 1 (FFAR1) antagonism, which normalized insulin hypersecretion from isolated islets exposed to palmitate. FFAR1 may represent a prevention strategy/therapy in subjects with high palmitate levels, but no FFAR1 antagonists are currently available for clinical use. GLP-1 receptor agonism showed promise in normalizing insulin hypersecretion. There are several available clinical intervention options using GLP-1 agonism. Indeed, we have conducted the first extended placebo-controlled, double-blinded intervention trial, where the GLP-1 analogue exenatide was administered to obese adolescents. Significant reduction in weight was observed with no rise in adverse events compared to placebo. The trial will potentially supply this patient group a novel pharmacological alternative.
The Beta-JUDO project has developed a cohort counting to about 700 obese children with detailed phenotype and blood samples collected over time including subjects that develop T2DM. The cohort provides unique possibilities to delineate early events in obesity-related T2DM and identify prognostic markers for subjects at risk. A definition of insulin hypersecretion in obese children has been proposed, which was used when genetic markers for insulin hypersecretion were identified.
Project Context and Objectives:
In Europe more than 20% of children are overweight and 5% even obese. Obesity is a strong risk factor for developing complications including type 2 diabetes mellitus (T2DM), which is now observed in children in increasing numbers. These high obesity prevalence numbers in children, which do not decline, pose a major health threat for Europe. Balancing caloric uptake and expenditure by changes in life-style including enhanced physical activity and modifying eating habits form the basis of any intervention. Regrettably, such interventions are not effective in many of these children, especially the older ones, and additional approaches including pharmacology-based are desperately needed.
High insulin levels are observed in most obese children both at fasting and during oral glucose tolerance test (OGTT). The hyperinsulinemia has traditionally been interpreted as a result of decreased insulin sensitivity in insulin target tissue. Although this is probably correct in many obese adults, the situation may be different in children. In a considerable portion of the obese children hyperinsulinemia is observed despite essentially normal insulin sensitivity. Insulin is secreted from the insulin- producing beta-cells in the pancreas and is the prime hormone promoting storage of calories. Evidence is accumulating and supporting that high insulin levels in certain subjects could be a primary event leading to development of overweight and obesity. Based on these observations we proposed insulin hypersecretion as an early etiological factor preceding insulin resistance and promoting lipid deposition and T2DM. This is in contrast to the general view that hyperinsulinemia is a result, and not a cause, of insulin resistance in the obese state. From the proposal follows that reducing insulin hypersecretion would be a strategy to intervene with the development of childhood obesity and obesity-related T2DM. Mechanisms responsible for insulin hypersecretion are not well understood, however, and potential targets for intervention remain undefined.
The project “Beta-cell function in Juvenile Diabetes and Obesity” (Beta-JUDO) aimed at investigating the role of the insulin-producing beta-cell in development of obesity and T2DM in children. The hypothesis was that accentuated insulin response was an early phenomenon preceding development of overweight and insulin resistance since high insulin levels favours storage rather than combustion of energy. Individuals with this secretory phenotype were therefore at risk of developing obesity and T2DM. The project aims were addressed in a translational approach involving measuring insulin secretory responses in obese children and lean controls and from isolated human islets of Langerhans. In Beta-JUDO, paediatric obesity clinics and academic centres with focus on beta-cell biology, protein and transcript profiling, genetics, and bioinformatics formed a consortium together with a large drug company and small companies (SMEs) specialized on biomarker discovery and clinical trials.
The main aim of the Beta-JUDO project was to identify ways to reduce and normalize the high insulin levels observed in obese children by novel pharmacological approaches, which were expected to lead to weight reduction. The following two objectives were addressed:
The first objective was to define factors and molecular mechanisms that contribute to insulin hypersecretion. The objective was addressed by investigating obese children belonging to childhood obesity cohorts and isolated islets. In the obese children special focus was on insulin secretory response during an OGTT. In addition to measuring insulin during the OGTT, also C-peptide, glucose, free fatty acids (FFAs) and incretins were determined. Magnetic resonance (MR) images were obtained from the subjects in order to determine percentages of fat in different body parts. Genetic analyses were conducted, where special focus was on identifying genetic variants correlating with the insulin hypersecreting phenotype. Questionnaires addressing eating, exercise and sleep patterns were also part of the characterization. Lean children were recruited as controls subjects to go through the same procedures as the obese children. In islets isolated from human donors insulin hypersecretion in response to elevated glucose concentrations was observed when the islets were exposed to elevated levels of the saturated fatty acid palmitate. Mechanisms for this augmented glucose-stimulated insulin secretion (GSIS) were delineated by generating expression patterns at the transcript (transcriptomics), protein (proteomics) and lipid (lipidomics) levels. These omics data sets were analyzed with the aim of identifying novel molecular players and pathways contributing to insulin hypersecretion.
The second objective was to define ways to counteract insulin hypersecretion with the aim of normalizing the secretory pattern. The objective was addressed by investigating pharmacological principles in the isolated islets and in the obese children. Isolated islets hypersecreting insulin were exposed to different pharmacological principles. Subsequently, GSIS was measured and the influence of the drugs on insulin hypersecretion determined. In drugs capable of normalizing insulin hypersecretion molecular mechanisms in the islets were investigated by generating expression patterns at the transcript (transcriptomics), protein (proteomics) and lipid (lipidomics) levels. In the obese children novel ways of intervening pharmacologically with the aim of normalizing insulin hypersecretion were explored. A study drug was selected based on the results from the ability of drugs to normalize insulin hypersecretion in the isolated islets, and a randomized, two-arm, double-blinded, placebo-controlled interventional study was conducted.
Project Results:
The first objective of the project was to define mechanisms for insulin hypersecretion using the translational approach with obese and lean control children and isolated islets exposed or not to elevated levels of fatty acid palmitate. To address the first objective of the project we established clinical examination and assessment procedures as outlined in Table 1 (see attached pdf document) obtained from reference (1). In the Beta-JUDO project four clinical centers participated. To make it possible to compare results between the centers all examinations and procedures were harmonized.
Free fatty acid palmitate was examined as potential factor causing insulin hypersecretion. Whereas glucose levels were generally normal in young obese children, lipid levels including circulating FFAs were elevated already in these young subjects. We proposed elevated levels of FFAs to be an underlying factor for insulin hypersecretion. To more accurately measure levels of multiple FFA species, we developed a method for measuring these levels (2). The GC-MS based method was able to resolve and quantify palmitate (Fig 1 in attached pdf document) and several additional FFA species. Among the FFA species we focused on the saturated fatty acid palmitate. In obese children and adolescents with normal glucose tolerance (NGT) levels of free fatty acid palmitate and insulin at fasting and during OGTT were measured (3). Fasting palmitate levels varied 3-fold in the obese children and adolescents (Fig 2). To investigate if high palmitate levels were associated with high insulin levels the obese children and adolescents were divided based on fasting palmitate levels. Insulin levels at fasting and during OGTT were significantly higher in children with high fasting palmitate levels compared with obese children with low palmitate levels (Fig 3). In obese adolescents with high palmitate levels insulin levels during OGTT were similar to those in obese with showed a delayed peak, which is observed in subjects with risk of developing T2DM.
Islets of Langerhans isolated from human organ-donors were used to further investigate the potential causality between high palmitate levels and insulin hypersecretion. In human islets exposed for 2 days to palmitate concentrations observed in the obese children (Fig 2), subsequent GSIS was accentuated (3). When the culture period was extended to 7 days impaired insulin secretion and apoptosis was observed (Fig 4). We concluded that that elevated palmitate levels provoked insulin hypersecretion from human islets after 2 days and impaired insulin secretion after 7 days. Also, we proposed that these time-dependent changes islet GSIS may be reflected in the obese children and adolescents. In this way the high insulin levels in obese children with high palmitate levels may reflect insulin hypersecretion and the delayed insulin response and lower insulin levels in adolescents with high palmitate levels reflect impaired insulin secretion. The results were compatible with a role of elevated palmitate levels causing insulin hypersecretion.
Free fatty acid palmitate is modulated by other fatty acids. In vivo multiple FFAs are present in the circulation. To further delineate mechanisms by which palmitate specifically affected insulin secretion we exposed beta-cells to either palmitate alone or in combination with the mono-unsaturated fatty acid oleate (4). Oleate is one of the most abundant FFAs in the circulation. Whereas palmitate alone caused impaired GSIS and increased apoptosis, GSIS and apoptosis measurements were essentially normalized when the cells were exposed to the combination of palmitate and oleate (Fig 5). The mono-unsaturated fatty acid achieved the reversal of the negative effects of palmitate by activating pro-survival pathways of the ER stress response (Fig 6). To further investigate the different effects of saturated and mono-unsaturated fatty acids we tested the effects of FFAs on insulin secretion by investigating how different fatty acids present in the circulation affected the acute insulin secretory response (5). Saturated fatty acids palmitate and stearate were less potent insulin secretagogues compared to mono-unsaturated fatty acids palmitoleate and oleate in producing about half the amount of secreted insulin compared to the mono-unsaturated fatty acids (Fig 7). These differences in potency of the fatty acids to induce insulin secretion were correlated with similar variations in islet respiration (Fig 8). Ceramide formation as consequence of elevated palmitate levels: When palmitate levels are elevated for extended time periods, which are observed in obese subjects, the fatty acid can be shunted into the sphingolipid pathway generating increased amounts of ceramide. High ceramide levels are connected with increased rates of apoptosis. We examined to what extent chronically high palmitate levels affected the ceramide levels in beta-cells (6). Indeed, ceramide levels were doubled after 2 days exposure to elevated palmitate levels (Fig 9). Reduced ceramide levels were observed when ceramide production was inhibited by either inhibiting the ceramide synthases or the enzyme serine palmitoyltransferase. We concluded that there was a close correlation between islet secretion and respiration, that the effects of palmitate were modulated by other FFAs and that increased ceramide production was a likely player in palmitate-mediated effects on insulin secretion.
Beta-cell lipotoxicity was also studied in vivo. Obese adolescents show impaired insulin secretion during OGTT, especially those with high fasting palmate levels (Fig 3). It has been proposed that elevated circulating levels of lipids and FFAs cause such impaired beta-cell function via lipid deposition in the beta-cells i.e. lipotoxicity. To examine this we correlated pancreatic fat with degree of overweight or obesity in obese adolescents (7). No relation was observed between pancreatic fat and BMI in obese, i.e. with BMI SDS > 2, adolescents, however (Fig 10). We concluded that evidence for beta-cell lipid deposition in obese children could not be obtained with MRI.
Altered glucagon secretion was studied as a factor causing insulin hypersecretion. Elevated glucagon levels have been observed in obese subjects. Based on these observations we proposed that hyperglucagonemia could contribute to hypersinsulinemia. To address this question insulin and glucagon levels were measured at fasting and during OGTT in obese children with normal glucose tolerance (NGT), impaired fasting glucose (IFG), impaired glucose tolerance (IGT) or T2DM (8). Insulin levels at fasting and during OGTT rose with deterioration of glucose tolerance (Fig 11). When glucagon was measured increased fasting levels were measured in the glucose intolerant obese subjects. Also, reduction in glucagon during OGTT was delayed in the glucose intolerant obese subjects. We concluded that the altered glucagon response in obese children was compatible with a role of the glucagon response contributing to hyperinsulinemia.
Genetic variants were studied as potential causes of insulin hypersecretion. Genome-wide association studies (GWAS) were performed to identify genes implicated in insulin hypersecretion using selected indices of insulin hypersecretion. We identified 45 independent genetic variants (Figure 12). In addition to the identification of candidates from new GWAS, we selected candidates from previous GWAS on obesity and related traits, which were matched with hits from proteomic experiments from Beta-JUDO partners (PCSK1 and LEP). The PCSK1 gene is a plausible candidate since it encodes for an important enzymatic step in insulin processing and maturation. However, so far it has been associated with monogenic (and polygenic) risk for obesity. We sequenced the gene in patients with childhood obesity regarded prone to carry variants in PCSK1. We identified two novel variants in this PCSK1 gene and have subsequently assessed their clinical-metabolic and functional relevance (9). One variant leads to complete loss of function of the enzymatic function and increased ER stress. Moreover, the SNP rs725522 was associated with elevated insulin secretion during OGTT in obese children independent from BMI. These results suggested that PCSK1 could be a candidate for insulin hypersecretion and obesity, and not only for (monogenic) obesity as expected from previous published reports. We concluded that insulin hypersecretion was associated with specific genetic variants of which one (PCSK1) was further investigated (9).
The second objective of the project was to define mechanisms for normalizing insulin hypersecretion using the translational approach with isolated islets and obese children.
FFAR1 antagonism was investigated as a factor potentially normalizing insulin hypersecretion. Based on our conclusion that high FFA levels were implicated in insulin hypersecretion, antagonizing the effects of FFAs would be a potential avenue to normalize insulin hypersecretion. FFAs including palmitate affect cells by entering the cell traversing the cell membrane and exerting intracellular actions and via binding to the FFA receptor 1 (FFAR1) with post-receptor signaling events. In isolated islets and beta-cell lines we investigated the role of FFAR1 in how palmitate modulated insulin secretion. FFAs have both acute and chronic effects and FFAR1 is involved in both (10). Whereas acutely exposing the islet to palmitate caused 3-fold rise in release of insulin, such acute potentiation was reduced to 2-fold in the presence of FFAR1 antagonist (Fig 13). FFAR1 also affected the long-term effects of fatty acid palmitate on insulin secretion (10). In particular, insulin hypersecretion evoked by exposing human islets to palmitate for 2 days was attenuated (Fig 14). The attenuation of insulin hypersecretion induced by FFAR1 antagonism was accompanied by reduction in islet metabolism (11). Indeed, in the presence of FFAR1 antagonist 203 mitochondrial oxygen consumption rate (OCR) was normalized (Fig 15). We concluded that FFAR1 signaling and metabolism contribute to palmitate-induced insulin hypersecretion and that counteracting FFAR1 may represent a prevention strategy/therapy in subjects with high palmitate levels. Today, no FFAR1 antagonists are available for clinical use, however.
GLP-1 receptor agonism was studied as a factor potentially normalizing insulin hypersecretion. In addition to FFAR1 the beta-cell expresses numerous other GPCRs, including the glucagon like peptide-1 (GLP-1) receptor. GLP-1 is a gut hormone, which plays an important role in beta-cell function, proliferation and survival. When we exposed human islets for 7 days with palmitate and also included the stable GLP-1 analogue exendin-4 during culture, insulin secretory response was no longer lowered but essentially normalized by including the GLP-1 analogue (Fig 16) (12).
To delineate mechanisms for this action different GPCR signaling pathways were analyzed. Palmitate was found to up-regulate inflammatory pathway components including Il-1b, and SOCS2, which were effectively reversed by including exendin-4 during culture (Fig 17). We concluded that GLP-1 receptor agonism shows promise in normalizing insulin hypersecretion. In contrast to FFAR1-antagonism there are several clinical intervention options using GLP-1 agonism. Indeed, we have conducted the first 6-month, placebo-controlled, double-blinded study, where GLP-1 analogue exenatide was administered to obese adolescents (ClinicalTrials.gov Identifier: NCT02794402). Significant reduction in weight was observed with no rise in adverse events compared to placebo. The trial will potentially supply this patient group a novel pharmacological alternative.
References, which are publications from the Beta-JUDO research effort:
1. Forslund, A., Staaf, J., Kullberg, J., Ciba, I., Dahlbom, M., and Bergsten, P. (2014) Uppsala longitudinal study of childhood obesity: protocol description. Pediatrics 133, e386-393.
2. Ubhayasekera, S. J., Staaf, J., Forslund, A., Bergsten, P., and Bergquist, J. (2013) Free fatty acid determination in plasma by GC-MS after conversion to Weinreb amides. Analytical and bioanalytical chemistry 405, 1929-1935.
3. Staaf, J., Ubhayasekera, S. J., Sargsyan, E., Chowdhury, A., Kristinsson, H., Manell, H., Bergquist, J., Forslund, A., and Bergsten, P. (2016) Initial hyperinsulinemia and subsequent beta-cell dysfunction is associated with elevated palmitate levels. Pediatric research 80, 267-274.
4. Sargsyan, E., Artemenko, K., Manukyan, L., Bergquist, J., and Bergsten, P. (2016) Oleate protects beta-cells from the toxic effect of palmitate by activating pro-survival pathways of the ER stress response. Biochim Biophys Acta 1861, 1151-1160.
5. Cen, J., Sargsyan, E., and Bergsten, P. (2016) Fatty acids stimulate insulin secretion from human pancreatic islets at fasting glucose concentrations via mitochondria-dependent and -independent mechanisms. Nutrition & metabolism 13, 59.
6. Manukyan, L., Ubhayasekera, S. J., Bergquist, J., Sargsyan, E., and Bergsten, P. (2015) Palmitate-induced impairments of beta-cell function are linked with generation of specific ceramide species via acylation of sphingosine. Endocrinology 156, 802-812.
7. Staaf, J., Labmayr, V., Paulmichl, K., Manell, H., Cen, J., Ciba, I., Dahlbom, M., Roomp, K., Anderwald, C. H., Meissnitzer, M., Schneider, R., Forslund, A., Widhalm, K., Bergquist, J., Ahlstrom, H., Bergsten, P., Weghuber, D., and Kullberg, J. (2017) Pancreatic Fat Is Associated With Metabolic Syndrome and Visceral Fat but Not Beta-Cell Function or Body Mass Index in Pediatric Obesity. Pancreas 46, 358-365.
8. Manell, H., Staaf, J., Manukyan, L., Kristinsson, H., Cen, J., Stenlid, R., Ciba, I., Forslund, A., and Bergsten, P. (2016) Altered Plasma Levels of Glucagon, GLP-1 and Glicentin During OGTT in Adolescents With Obesity and Type 2 Diabetes. The Journal of clinical endocrinology and metabolism 101, 1181-1189.
9. Loffler, D., Behrendt, S., Creemers, J. W., Klammt, J., Aust, G., Stanik, J., Kiess, W., Kovacs, P., and Korner, A. (2017) Functional and clinical relevance of novel and known PCSK1 variants for childhood obesity and glucose metabolism. Mol Metab 6, 295-305.
10. Kristinsson, H., Smith, D. M., Bergsten, P., and Sargsyan, E. (2013) FFAR1 is involved in both the acute and chronic effects of palmitate on insulin secretion. Endocrinology 154, 4078-4088.
11. Kristinsson, H., Bergsten, P., and Sargsyan, E. (2015) Free fatty acid receptor 1 (FFAR1/GPR40) signaling affects insulin secretion by enhancing mitochondrial respiration during palmitate exposure. Biochim Biophys Acta 1853, 3248-3257.
12. Chowdhury, A. I., and Bergsten, P. (2017) GLP-1 analogue recovers impaired insulin secretion from human islets treated with palmitate via down-regulation of SOCS2. Molecular and cellular endocrinology 439, 194-202.
Potential Impact:
The translational project Beta-JUDO spanned both clinical and pre-clinical work and addressed the role of insulin hypesecretion from the pancreatic beta-cells in obesity-related development of type 2 diabetes mellitus in children. The project has generated results, which have had impact in the following areas:
Clinical area: The characterization of the obesity cohorts has addressed the question to what extent insulin hypersecretion is an early phenomenon in the development of childhood obesity. Definition of international standards and OGTT reference values have been generated for obese and normal weight boys and girls of different ages. From these values opportunities to developing standards for insulin secretion and insulin resistance for different age groups were explored. Characterization of the obese children supplied evidence supporting the importance of maintaining normal insulin levels and also identified situations and foods that elicit accentuated insulin secretion, which are to be avoided. Also, the characterization yielded information on how insulin secretory changes observed in children that develop T2DM change over time, i.e. give knowledge about early prognostic signs of subjects at risk. The genetic work conducted in obesity childhood cohorts identified genes associated with insulin hypersecretion. Importantly, experience with one compound and its efficacy in childhood obesity was obtained through the interventional study conducted within the project. The successful intervention with exenatide paves the way for further clinical studies, where the drug will be further tested in larger cohorts for its safety and efficacy in combatting childhood obesity. To give this patient group a pharmacological alternative would be very important not only for the patients but also for the European health providers.
Pre-clinical area: Cellular and molecular mechanisms underlying insulin hypersecretion from the pancreatic beta-cell were delineated by applying systems biology analysis on data sets generated by multiple omics approaches including transcriptomics, proteomics and lipidomics on human isolated islets. The analyses of transcripts, proteins and lipids, associated with the distinct islet secretory phenotypes, defined mechanism responsible for insulin hypersecretion but also how different compounds associated with the treatment of obesity and T2DM acted at the islet level and affected these mechanisms. Several novel, not yet proposed drug candidates coming form the Beta-JUDO work and operative in the islets of Langerhans were validated for a role in insulin hypersecretion. Such candidates may be of interest for further development into pharmacological strategies for intervention in childhood obesity.
Societal and economical area: Childhood obesity has reached levels never seen before and Europe faces a major challenge with regard to finding ways to reverse this trend. The project aimed at attenuating the rise in in childhood obesity and T2DM. By reducing obesity rates, both major health benefits, and economic benefits are achieved. This project raised awareness and knowledge among citizens in Europe, in health care, industry and funding agencies. Scientific benefits were based on the integrated efforts in elucidating the impact of insulin early in childhood obesity development and developing strategies for the prevention. It is imperative to meet this challenge and to find new ways to attenuate obesity and obesity-related T2DM in the young population, and to identify and evaluate new drugs for this rapidly growing patient group. The very limited pharmacology-based alternatives for the treatment of young obese individuals pose a problem not only for the afflicted individuals at risk of developing T2DM and other related diseases but also for European health providers. The results of the project are expected to lead to a reduced number of young individuals with overweight or obesity. The involvement of SMEs in the project contributed to that therapeutic strategies proposed by the project led to new opportunities for European industry strengthening European health economy.
The multi-disciplinary consortium was providing the framework for building capacity by providing the educational basis, where both postdoctoral persons and PhD-students were trained to tackle the complex health issue of childhood obesity in a translational way involving training from industry, health care and academia.
The project’s main dissemination activities include publication of the results in scientific journals and at meetings. At the time of writing this report close to 60 articles had been published. The results of the Beta-JUDO project have been presented at close to 130 dissemination events. Many of these events have had a scientific medical context but with attendance of stakeholders from other areas of society.
List of Websites:
https://betajudo.org.
The project “Beta-cell function in Juvenile Diabetes and Obesity” (Beta-JUDO) aimed at investigating the role of the insulin-producing beta-cell in development of obesity and type 2 diabetes mellitus (T2DM) in children. The rationale behind the project was that accentuated insulin response (insulin hypersecretion) is an early phenomenon promoting development of overweight and insulin resistance, since high insulin levels favours storage rather than combustion of energy. Individuals with this secretory phenotype are therefore at risk of developing obesity and T2DM. The project was organised in a translational approach involving studies of obese children and beta-cells from isolated human islets of Langerhans. In Beta-JUDO, paediatric obesity clinics and academic centres with focus on treatment of obese children, beta-cell biology, protein and transcript profiling, genetics, and bioinformatics formed a consortium together with a large drug company and small companies (SMEs) specialized on biomarker discovery and clinical trials.
The overall aim of the project was to identify new pharmacological approaches to normalize the high insulin levels observed in obese children and children of risk to become obese. The aim was addressed by, firstly, defining factors and molecular mechanisms that contribute to insulin hypersecretion, and secondly, to address ways to counteract insulin hypersecretion by investigating pharmacological principles potentially targeting insulin hypersecretion in both isolated islets and in obese children.
Factors contributing to insulin hypersecretion included high circulating levels of the free fatty acid (FFA) palmitate, which were high in obese children with accentuated insulin hypersecretion. Correspondently, in isolated islets such elevated palmitate levels caused accentuated glucose-stimulated insulin secretion. The findings supported a role of elevated levels of palmitate as an underlying factor contributing to insulin hypersecretion. We found that palmitate enhanced islet respiration, which was closely linked with islet insulin hypersecretion. In obese children raised glucagon levels were compatible with a role of altered glucagon response in obese children contributing to hyperinsulinemia. In genetic studies using large obesity cohorts we identified a number of genes that are associated with insulin hypersecretion, such as the gene PCSK1 that is involved in insulin processing. Further underlying factors for islet insulin hypersecretion were found by generating extensive protein and transcript expression patterns.
Strategies to counteract insulin hypersecretion included free fatty acid receptor 1 (FFAR1) antagonism, which normalized insulin hypersecretion from isolated islets exposed to palmitate. FFAR1 may represent a prevention strategy/therapy in subjects with high palmitate levels, but no FFAR1 antagonists are currently available for clinical use. GLP-1 receptor agonism showed promise in normalizing insulin hypersecretion. There are several available clinical intervention options using GLP-1 agonism. Indeed, we have conducted the first extended placebo-controlled, double-blinded intervention trial, where the GLP-1 analogue exenatide was administered to obese adolescents. Significant reduction in weight was observed with no rise in adverse events compared to placebo. The trial will potentially supply this patient group a novel pharmacological alternative.
The Beta-JUDO project has developed a cohort counting to about 700 obese children with detailed phenotype and blood samples collected over time including subjects that develop T2DM. The cohort provides unique possibilities to delineate early events in obesity-related T2DM and identify prognostic markers for subjects at risk. A definition of insulin hypersecretion in obese children has been proposed, which was used when genetic markers for insulin hypersecretion were identified.
Project Context and Objectives:
In Europe more than 20% of children are overweight and 5% even obese. Obesity is a strong risk factor for developing complications including type 2 diabetes mellitus (T2DM), which is now observed in children in increasing numbers. These high obesity prevalence numbers in children, which do not decline, pose a major health threat for Europe. Balancing caloric uptake and expenditure by changes in life-style including enhanced physical activity and modifying eating habits form the basis of any intervention. Regrettably, such interventions are not effective in many of these children, especially the older ones, and additional approaches including pharmacology-based are desperately needed.
High insulin levels are observed in most obese children both at fasting and during oral glucose tolerance test (OGTT). The hyperinsulinemia has traditionally been interpreted as a result of decreased insulin sensitivity in insulin target tissue. Although this is probably correct in many obese adults, the situation may be different in children. In a considerable portion of the obese children hyperinsulinemia is observed despite essentially normal insulin sensitivity. Insulin is secreted from the insulin- producing beta-cells in the pancreas and is the prime hormone promoting storage of calories. Evidence is accumulating and supporting that high insulin levels in certain subjects could be a primary event leading to development of overweight and obesity. Based on these observations we proposed insulin hypersecretion as an early etiological factor preceding insulin resistance and promoting lipid deposition and T2DM. This is in contrast to the general view that hyperinsulinemia is a result, and not a cause, of insulin resistance in the obese state. From the proposal follows that reducing insulin hypersecretion would be a strategy to intervene with the development of childhood obesity and obesity-related T2DM. Mechanisms responsible for insulin hypersecretion are not well understood, however, and potential targets for intervention remain undefined.
The project “Beta-cell function in Juvenile Diabetes and Obesity” (Beta-JUDO) aimed at investigating the role of the insulin-producing beta-cell in development of obesity and T2DM in children. The hypothesis was that accentuated insulin response was an early phenomenon preceding development of overweight and insulin resistance since high insulin levels favours storage rather than combustion of energy. Individuals with this secretory phenotype were therefore at risk of developing obesity and T2DM. The project aims were addressed in a translational approach involving measuring insulin secretory responses in obese children and lean controls and from isolated human islets of Langerhans. In Beta-JUDO, paediatric obesity clinics and academic centres with focus on beta-cell biology, protein and transcript profiling, genetics, and bioinformatics formed a consortium together with a large drug company and small companies (SMEs) specialized on biomarker discovery and clinical trials.
The main aim of the Beta-JUDO project was to identify ways to reduce and normalize the high insulin levels observed in obese children by novel pharmacological approaches, which were expected to lead to weight reduction. The following two objectives were addressed:
The first objective was to define factors and molecular mechanisms that contribute to insulin hypersecretion. The objective was addressed by investigating obese children belonging to childhood obesity cohorts and isolated islets. In the obese children special focus was on insulin secretory response during an OGTT. In addition to measuring insulin during the OGTT, also C-peptide, glucose, free fatty acids (FFAs) and incretins were determined. Magnetic resonance (MR) images were obtained from the subjects in order to determine percentages of fat in different body parts. Genetic analyses were conducted, where special focus was on identifying genetic variants correlating with the insulin hypersecreting phenotype. Questionnaires addressing eating, exercise and sleep patterns were also part of the characterization. Lean children were recruited as controls subjects to go through the same procedures as the obese children. In islets isolated from human donors insulin hypersecretion in response to elevated glucose concentrations was observed when the islets were exposed to elevated levels of the saturated fatty acid palmitate. Mechanisms for this augmented glucose-stimulated insulin secretion (GSIS) were delineated by generating expression patterns at the transcript (transcriptomics), protein (proteomics) and lipid (lipidomics) levels. These omics data sets were analyzed with the aim of identifying novel molecular players and pathways contributing to insulin hypersecretion.
The second objective was to define ways to counteract insulin hypersecretion with the aim of normalizing the secretory pattern. The objective was addressed by investigating pharmacological principles in the isolated islets and in the obese children. Isolated islets hypersecreting insulin were exposed to different pharmacological principles. Subsequently, GSIS was measured and the influence of the drugs on insulin hypersecretion determined. In drugs capable of normalizing insulin hypersecretion molecular mechanisms in the islets were investigated by generating expression patterns at the transcript (transcriptomics), protein (proteomics) and lipid (lipidomics) levels. In the obese children novel ways of intervening pharmacologically with the aim of normalizing insulin hypersecretion were explored. A study drug was selected based on the results from the ability of drugs to normalize insulin hypersecretion in the isolated islets, and a randomized, two-arm, double-blinded, placebo-controlled interventional study was conducted.
Project Results:
The first objective of the project was to define mechanisms for insulin hypersecretion using the translational approach with obese and lean control children and isolated islets exposed or not to elevated levels of fatty acid palmitate. To address the first objective of the project we established clinical examination and assessment procedures as outlined in Table 1 (see attached pdf document) obtained from reference (1). In the Beta-JUDO project four clinical centers participated. To make it possible to compare results between the centers all examinations and procedures were harmonized.
Free fatty acid palmitate was examined as potential factor causing insulin hypersecretion. Whereas glucose levels were generally normal in young obese children, lipid levels including circulating FFAs were elevated already in these young subjects. We proposed elevated levels of FFAs to be an underlying factor for insulin hypersecretion. To more accurately measure levels of multiple FFA species, we developed a method for measuring these levels (2). The GC-MS based method was able to resolve and quantify palmitate (Fig 1 in attached pdf document) and several additional FFA species. Among the FFA species we focused on the saturated fatty acid palmitate. In obese children and adolescents with normal glucose tolerance (NGT) levels of free fatty acid palmitate and insulin at fasting and during OGTT were measured (3). Fasting palmitate levels varied 3-fold in the obese children and adolescents (Fig 2). To investigate if high palmitate levels were associated with high insulin levels the obese children and adolescents were divided based on fasting palmitate levels. Insulin levels at fasting and during OGTT were significantly higher in children with high fasting palmitate levels compared with obese children with low palmitate levels (Fig 3). In obese adolescents with high palmitate levels insulin levels during OGTT were similar to those in obese with showed a delayed peak, which is observed in subjects with risk of developing T2DM.
Islets of Langerhans isolated from human organ-donors were used to further investigate the potential causality between high palmitate levels and insulin hypersecretion. In human islets exposed for 2 days to palmitate concentrations observed in the obese children (Fig 2), subsequent GSIS was accentuated (3). When the culture period was extended to 7 days impaired insulin secretion and apoptosis was observed (Fig 4). We concluded that that elevated palmitate levels provoked insulin hypersecretion from human islets after 2 days and impaired insulin secretion after 7 days. Also, we proposed that these time-dependent changes islet GSIS may be reflected in the obese children and adolescents. In this way the high insulin levels in obese children with high palmitate levels may reflect insulin hypersecretion and the delayed insulin response and lower insulin levels in adolescents with high palmitate levels reflect impaired insulin secretion. The results were compatible with a role of elevated palmitate levels causing insulin hypersecretion.
Free fatty acid palmitate is modulated by other fatty acids. In vivo multiple FFAs are present in the circulation. To further delineate mechanisms by which palmitate specifically affected insulin secretion we exposed beta-cells to either palmitate alone or in combination with the mono-unsaturated fatty acid oleate (4). Oleate is one of the most abundant FFAs in the circulation. Whereas palmitate alone caused impaired GSIS and increased apoptosis, GSIS and apoptosis measurements were essentially normalized when the cells were exposed to the combination of palmitate and oleate (Fig 5). The mono-unsaturated fatty acid achieved the reversal of the negative effects of palmitate by activating pro-survival pathways of the ER stress response (Fig 6). To further investigate the different effects of saturated and mono-unsaturated fatty acids we tested the effects of FFAs on insulin secretion by investigating how different fatty acids present in the circulation affected the acute insulin secretory response (5). Saturated fatty acids palmitate and stearate were less potent insulin secretagogues compared to mono-unsaturated fatty acids palmitoleate and oleate in producing about half the amount of secreted insulin compared to the mono-unsaturated fatty acids (Fig 7). These differences in potency of the fatty acids to induce insulin secretion were correlated with similar variations in islet respiration (Fig 8). Ceramide formation as consequence of elevated palmitate levels: When palmitate levels are elevated for extended time periods, which are observed in obese subjects, the fatty acid can be shunted into the sphingolipid pathway generating increased amounts of ceramide. High ceramide levels are connected with increased rates of apoptosis. We examined to what extent chronically high palmitate levels affected the ceramide levels in beta-cells (6). Indeed, ceramide levels were doubled after 2 days exposure to elevated palmitate levels (Fig 9). Reduced ceramide levels were observed when ceramide production was inhibited by either inhibiting the ceramide synthases or the enzyme serine palmitoyltransferase. We concluded that there was a close correlation between islet secretion and respiration, that the effects of palmitate were modulated by other FFAs and that increased ceramide production was a likely player in palmitate-mediated effects on insulin secretion.
Beta-cell lipotoxicity was also studied in vivo. Obese adolescents show impaired insulin secretion during OGTT, especially those with high fasting palmate levels (Fig 3). It has been proposed that elevated circulating levels of lipids and FFAs cause such impaired beta-cell function via lipid deposition in the beta-cells i.e. lipotoxicity. To examine this we correlated pancreatic fat with degree of overweight or obesity in obese adolescents (7). No relation was observed between pancreatic fat and BMI in obese, i.e. with BMI SDS > 2, adolescents, however (Fig 10). We concluded that evidence for beta-cell lipid deposition in obese children could not be obtained with MRI.
Altered glucagon secretion was studied as a factor causing insulin hypersecretion. Elevated glucagon levels have been observed in obese subjects. Based on these observations we proposed that hyperglucagonemia could contribute to hypersinsulinemia. To address this question insulin and glucagon levels were measured at fasting and during OGTT in obese children with normal glucose tolerance (NGT), impaired fasting glucose (IFG), impaired glucose tolerance (IGT) or T2DM (8). Insulin levels at fasting and during OGTT rose with deterioration of glucose tolerance (Fig 11). When glucagon was measured increased fasting levels were measured in the glucose intolerant obese subjects. Also, reduction in glucagon during OGTT was delayed in the glucose intolerant obese subjects. We concluded that the altered glucagon response in obese children was compatible with a role of the glucagon response contributing to hyperinsulinemia.
Genetic variants were studied as potential causes of insulin hypersecretion. Genome-wide association studies (GWAS) were performed to identify genes implicated in insulin hypersecretion using selected indices of insulin hypersecretion. We identified 45 independent genetic variants (Figure 12). In addition to the identification of candidates from new GWAS, we selected candidates from previous GWAS on obesity and related traits, which were matched with hits from proteomic experiments from Beta-JUDO partners (PCSK1 and LEP). The PCSK1 gene is a plausible candidate since it encodes for an important enzymatic step in insulin processing and maturation. However, so far it has been associated with monogenic (and polygenic) risk for obesity. We sequenced the gene in patients with childhood obesity regarded prone to carry variants in PCSK1. We identified two novel variants in this PCSK1 gene and have subsequently assessed their clinical-metabolic and functional relevance (9). One variant leads to complete loss of function of the enzymatic function and increased ER stress. Moreover, the SNP rs725522 was associated with elevated insulin secretion during OGTT in obese children independent from BMI. These results suggested that PCSK1 could be a candidate for insulin hypersecretion and obesity, and not only for (monogenic) obesity as expected from previous published reports. We concluded that insulin hypersecretion was associated with specific genetic variants of which one (PCSK1) was further investigated (9).
The second objective of the project was to define mechanisms for normalizing insulin hypersecretion using the translational approach with isolated islets and obese children.
FFAR1 antagonism was investigated as a factor potentially normalizing insulin hypersecretion. Based on our conclusion that high FFA levels were implicated in insulin hypersecretion, antagonizing the effects of FFAs would be a potential avenue to normalize insulin hypersecretion. FFAs including palmitate affect cells by entering the cell traversing the cell membrane and exerting intracellular actions and via binding to the FFA receptor 1 (FFAR1) with post-receptor signaling events. In isolated islets and beta-cell lines we investigated the role of FFAR1 in how palmitate modulated insulin secretion. FFAs have both acute and chronic effects and FFAR1 is involved in both (10). Whereas acutely exposing the islet to palmitate caused 3-fold rise in release of insulin, such acute potentiation was reduced to 2-fold in the presence of FFAR1 antagonist (Fig 13). FFAR1 also affected the long-term effects of fatty acid palmitate on insulin secretion (10). In particular, insulin hypersecretion evoked by exposing human islets to palmitate for 2 days was attenuated (Fig 14). The attenuation of insulin hypersecretion induced by FFAR1 antagonism was accompanied by reduction in islet metabolism (11). Indeed, in the presence of FFAR1 antagonist 203 mitochondrial oxygen consumption rate (OCR) was normalized (Fig 15). We concluded that FFAR1 signaling and metabolism contribute to palmitate-induced insulin hypersecretion and that counteracting FFAR1 may represent a prevention strategy/therapy in subjects with high palmitate levels. Today, no FFAR1 antagonists are available for clinical use, however.
GLP-1 receptor agonism was studied as a factor potentially normalizing insulin hypersecretion. In addition to FFAR1 the beta-cell expresses numerous other GPCRs, including the glucagon like peptide-1 (GLP-1) receptor. GLP-1 is a gut hormone, which plays an important role in beta-cell function, proliferation and survival. When we exposed human islets for 7 days with palmitate and also included the stable GLP-1 analogue exendin-4 during culture, insulin secretory response was no longer lowered but essentially normalized by including the GLP-1 analogue (Fig 16) (12).
To delineate mechanisms for this action different GPCR signaling pathways were analyzed. Palmitate was found to up-regulate inflammatory pathway components including Il-1b, and SOCS2, which were effectively reversed by including exendin-4 during culture (Fig 17). We concluded that GLP-1 receptor agonism shows promise in normalizing insulin hypersecretion. In contrast to FFAR1-antagonism there are several clinical intervention options using GLP-1 agonism. Indeed, we have conducted the first 6-month, placebo-controlled, double-blinded study, where GLP-1 analogue exenatide was administered to obese adolescents (ClinicalTrials.gov Identifier: NCT02794402). Significant reduction in weight was observed with no rise in adverse events compared to placebo. The trial will potentially supply this patient group a novel pharmacological alternative.
References, which are publications from the Beta-JUDO research effort:
1. Forslund, A., Staaf, J., Kullberg, J., Ciba, I., Dahlbom, M., and Bergsten, P. (2014) Uppsala longitudinal study of childhood obesity: protocol description. Pediatrics 133, e386-393.
2. Ubhayasekera, S. J., Staaf, J., Forslund, A., Bergsten, P., and Bergquist, J. (2013) Free fatty acid determination in plasma by GC-MS after conversion to Weinreb amides. Analytical and bioanalytical chemistry 405, 1929-1935.
3. Staaf, J., Ubhayasekera, S. J., Sargsyan, E., Chowdhury, A., Kristinsson, H., Manell, H., Bergquist, J., Forslund, A., and Bergsten, P. (2016) Initial hyperinsulinemia and subsequent beta-cell dysfunction is associated with elevated palmitate levels. Pediatric research 80, 267-274.
4. Sargsyan, E., Artemenko, K., Manukyan, L., Bergquist, J., and Bergsten, P. (2016) Oleate protects beta-cells from the toxic effect of palmitate by activating pro-survival pathways of the ER stress response. Biochim Biophys Acta 1861, 1151-1160.
5. Cen, J., Sargsyan, E., and Bergsten, P. (2016) Fatty acids stimulate insulin secretion from human pancreatic islets at fasting glucose concentrations via mitochondria-dependent and -independent mechanisms. Nutrition & metabolism 13, 59.
6. Manukyan, L., Ubhayasekera, S. J., Bergquist, J., Sargsyan, E., and Bergsten, P. (2015) Palmitate-induced impairments of beta-cell function are linked with generation of specific ceramide species via acylation of sphingosine. Endocrinology 156, 802-812.
7. Staaf, J., Labmayr, V., Paulmichl, K., Manell, H., Cen, J., Ciba, I., Dahlbom, M., Roomp, K., Anderwald, C. H., Meissnitzer, M., Schneider, R., Forslund, A., Widhalm, K., Bergquist, J., Ahlstrom, H., Bergsten, P., Weghuber, D., and Kullberg, J. (2017) Pancreatic Fat Is Associated With Metabolic Syndrome and Visceral Fat but Not Beta-Cell Function or Body Mass Index in Pediatric Obesity. Pancreas 46, 358-365.
8. Manell, H., Staaf, J., Manukyan, L., Kristinsson, H., Cen, J., Stenlid, R., Ciba, I., Forslund, A., and Bergsten, P. (2016) Altered Plasma Levels of Glucagon, GLP-1 and Glicentin During OGTT in Adolescents With Obesity and Type 2 Diabetes. The Journal of clinical endocrinology and metabolism 101, 1181-1189.
9. Loffler, D., Behrendt, S., Creemers, J. W., Klammt, J., Aust, G., Stanik, J., Kiess, W., Kovacs, P., and Korner, A. (2017) Functional and clinical relevance of novel and known PCSK1 variants for childhood obesity and glucose metabolism. Mol Metab 6, 295-305.
10. Kristinsson, H., Smith, D. M., Bergsten, P., and Sargsyan, E. (2013) FFAR1 is involved in both the acute and chronic effects of palmitate on insulin secretion. Endocrinology 154, 4078-4088.
11. Kristinsson, H., Bergsten, P., and Sargsyan, E. (2015) Free fatty acid receptor 1 (FFAR1/GPR40) signaling affects insulin secretion by enhancing mitochondrial respiration during palmitate exposure. Biochim Biophys Acta 1853, 3248-3257.
12. Chowdhury, A. I., and Bergsten, P. (2017) GLP-1 analogue recovers impaired insulin secretion from human islets treated with palmitate via down-regulation of SOCS2. Molecular and cellular endocrinology 439, 194-202.
Potential Impact:
The translational project Beta-JUDO spanned both clinical and pre-clinical work and addressed the role of insulin hypesecretion from the pancreatic beta-cells in obesity-related development of type 2 diabetes mellitus in children. The project has generated results, which have had impact in the following areas:
Clinical area: The characterization of the obesity cohorts has addressed the question to what extent insulin hypersecretion is an early phenomenon in the development of childhood obesity. Definition of international standards and OGTT reference values have been generated for obese and normal weight boys and girls of different ages. From these values opportunities to developing standards for insulin secretion and insulin resistance for different age groups were explored. Characterization of the obese children supplied evidence supporting the importance of maintaining normal insulin levels and also identified situations and foods that elicit accentuated insulin secretion, which are to be avoided. Also, the characterization yielded information on how insulin secretory changes observed in children that develop T2DM change over time, i.e. give knowledge about early prognostic signs of subjects at risk. The genetic work conducted in obesity childhood cohorts identified genes associated with insulin hypersecretion. Importantly, experience with one compound and its efficacy in childhood obesity was obtained through the interventional study conducted within the project. The successful intervention with exenatide paves the way for further clinical studies, where the drug will be further tested in larger cohorts for its safety and efficacy in combatting childhood obesity. To give this patient group a pharmacological alternative would be very important not only for the patients but also for the European health providers.
Pre-clinical area: Cellular and molecular mechanisms underlying insulin hypersecretion from the pancreatic beta-cell were delineated by applying systems biology analysis on data sets generated by multiple omics approaches including transcriptomics, proteomics and lipidomics on human isolated islets. The analyses of transcripts, proteins and lipids, associated with the distinct islet secretory phenotypes, defined mechanism responsible for insulin hypersecretion but also how different compounds associated with the treatment of obesity and T2DM acted at the islet level and affected these mechanisms. Several novel, not yet proposed drug candidates coming form the Beta-JUDO work and operative in the islets of Langerhans were validated for a role in insulin hypersecretion. Such candidates may be of interest for further development into pharmacological strategies for intervention in childhood obesity.
Societal and economical area: Childhood obesity has reached levels never seen before and Europe faces a major challenge with regard to finding ways to reverse this trend. The project aimed at attenuating the rise in in childhood obesity and T2DM. By reducing obesity rates, both major health benefits, and economic benefits are achieved. This project raised awareness and knowledge among citizens in Europe, in health care, industry and funding agencies. Scientific benefits were based on the integrated efforts in elucidating the impact of insulin early in childhood obesity development and developing strategies for the prevention. It is imperative to meet this challenge and to find new ways to attenuate obesity and obesity-related T2DM in the young population, and to identify and evaluate new drugs for this rapidly growing patient group. The very limited pharmacology-based alternatives for the treatment of young obese individuals pose a problem not only for the afflicted individuals at risk of developing T2DM and other related diseases but also for European health providers. The results of the project are expected to lead to a reduced number of young individuals with overweight or obesity. The involvement of SMEs in the project contributed to that therapeutic strategies proposed by the project led to new opportunities for European industry strengthening European health economy.
The multi-disciplinary consortium was providing the framework for building capacity by providing the educational basis, where both postdoctoral persons and PhD-students were trained to tackle the complex health issue of childhood obesity in a translational way involving training from industry, health care and academia.
The project’s main dissemination activities include publication of the results in scientific journals and at meetings. At the time of writing this report close to 60 articles had been published. The results of the Beta-JUDO project have been presented at close to 130 dissemination events. Many of these events have had a scientific medical context but with attendance of stakeholders from other areas of society.
List of Websites:
https://betajudo.org.
