Le diabète de type 2 (DT2) est caractérisé par une résistance à l’action de l’insuline et une dysfonction des cellules β pancréatiques. Il apparait lorsque la cellule β devient incapable d’augmenter sa masse fonctionnelle afin de compenser la résistance périphérique à l’action de l’insuline. L’identification de molécules capables de stimuler la réplication des cellules β et ainsi de préserver leur masse fonctionnelle aurait donc un intérêt thérapeutique majeur. Nous avons établi un modèle d’excès de nutriments in vivo chez le rat, dans lequel nous avons observé qu’une augmentation de la prolifération des cellules β associée à une augmentation de l’expression du facteur de croissance « heparin-binding EGF-like growth factor » (HB-EGF). L’objectif de cette thèse était de valider l’effet mitogène du HB-EGF sur les cellules β de rats et humaines, puis d’identifier le mécanisme d’activation de la voie HB-EGF-EGFR. Dans une première étude, nous avons démontré ex vivo que le facteur croissance HB-EGF stimule la prolifération des cellules β pancréatiques d’îlots isolés de rats et humains via l’activation de son récepteur EGFR. Nous avons également observé que la stimulation de la prolifération des cellules β de rats par le glucose nécessite l’activation de la voie de signalisation HB-EGF-EGFR par un mécanisme qui implique à la fois une augmentation de l’expression du gène codant pour HB-EGF via le facteur de transcription ChREBP, et l’activation du récepteur EGFR via une protéine de la famille des protéines Src tyrosine kinase. Les cellules β des îlots humains étant réfractaires à la prolifération, il est essentiel de confirmer les résultats obtenus chez les rongeurs dans des tissus humains. Nous avons observé un effet mitogène d’HB-EGF sur les cellules β humaines. En revanche, nous n’avons pas pu détecter de manière reproductible un effet stimulant du glucose sur la prolifération des cellules β humaines. Notre deuxième étude a donc consisté à identifier la technique la plus appropriée pour mesurer la prolifération des cellules β humaines. Nous avons comparé systématiquement la mesure de la prolifération en réponse à divers stimuli par cytométrie en flux ou par immunohistochimie sur des îlots intacts ou dispersés. Nous avons testé trois facteurs mitogènes soit le glucose, l’HB-EGF et l’harmine. Nous avons observé que l’HB-EGF et l’harmine stimulent la prolifération des cellules β et non β indépendamment de la méthode utilisée. En revanche, l’action mitogène du glucose semble être dépendante de la méthode. En conclusion, nous avons d’abord démontré que l’effet mitogène du glucose nécessite l’activation de la voie de signalisation HB-EGF-EGFR. Ensuite, nous avons observé que la mesure de la prolifération des cellules β humaines par cytométrie en flux offre plusieurs avantages par rapport à l’immunohistochimie. / Type 2 diabetes (T2D) is characterized by peripheral insulin resistance and pancreatic β-cell dysfunction. T2D occurs when β cells become unable to increase their functional mass in order to compensate for insulin resistance. The identification of molecules capable of stimulating β-cell replication to preserve their functional mass would therefore be of major therapeutic interest. We previously established a model of nutrient excess in which we observed an increase in β-cell proliferation associated with enhanced expression of the growth factor "heparin-binding EGF-like growth factor" (HB-EGF). The objective of the work presented in this thesis was to test the hypothesis that HB-EGF stimulates both rodent and human β-cell proliferation and to identify the underlying mechanisms. In a first study, we demonstrated ex vivo that HB-EGF stimulates pancreatic β-cell proliferation of isolated rat and human islets by activating EGFR. We also demonstrated that glucose, an important mitogen of the β cells, requires the activation of this HB-EGF-EGFR signaling pathway, ex vivo and in vivo in an infused rat model, to stimulate β-cell replication. Mechanistically, we demonstrate that glucose promotes HB-EGF gene expression via the ChREBP transcription factor and EGFR activation via a protein from the Src kinase family. Since adult human β cells tend to be refractory to proliferation, it is essential to confirm the findings obtained in rodents in human tissues. In isolated human islets, we confirmed the mitogenic action of HB-EGF but we were unable to detect a consistent stimulation of human β-cell proliferation in response to glucose. Our second study therefore consisted in identifying the most appropriate technique to measure human β-cell proliferation. We systematically compared proliferation levels measured by flow cytometry or immunohistochemistry in intact and dispersed human islets. We tested three mitogenic factors: glucose, HB-EGF and harmine. We observed that HB-EGF and harmine stimulate non-β cells and β-cell proliferation regardless of the method used. In contrast, the mitogenic action of glucose is variable depending on the method used. In conclusion, we first demonstrated that the mitogenic effect of glucose in β cells requires the activation of the HB-EGF-EGFR signaling pathway. Then we demonstrated that assessment of human β cell proliferation by flow cytometry offers several advantages over the use of immunohistochemical methods.

Pancreatic β-cells have a central function in the regulation of glucose homeostasis by releasing the blood sugar-lowering hormone insulin. Disruption of this process results in diabetes, which has a tremendous impact on the quality of life and requires lifelong treatment. Elucidating the mechanisms of pancreatic progenitor cell differentiation into fully functional β-cells will contribute to identifying the underlying reasons for β-cell dysfunction and to finding a cure for diabetes. Aldh1b1 was identified by our research group as a regulator of pancreas development and β-cell functionality. Aldh1b1 is a mitochondrial enzyme, expressed in all embryonic pancreas progenitors. Its expression is switched off during the process of differentiation and is undetectable in differentiated cells. Functional inactivation of Aldh1b1 in the mouse leads to premature differentiation of progenitor cells in the embryo and dysfunctional β-cells in the adult. However, the enzymatic function of Aldh1b1 in pancreas progenitors and how it ultimately affects β-cell functionality remained to be elucidated. In this study, I analyzed the role of Aldh1b1 in the metabolism of embryonic pancreas progenitor cells and its impact on chromatin structure and gene expression in both, progenitors and postnatal β-cells. Flow cytometry analysis of freshly isolated Aldh1b1 null embryonic pancreas progenitors showed a significant increase in ROS levels as well as a significant decrease in mitochondrial mass, whereas the mitochondrial membrane potential was not affected. To elucidate the impact of Aldh1b1 on cellular metabolism, I conducted metabolic flux experiments and untargeted metabolomics studies using FACS-isolated embryonic pancreas progenitors expanded in a 3D spheroid culture. Analyses following metabolic labeling with either 13C6-Glucose or 13C2-Glutamine showed that the absence of Aldh1b1 lead to an increase of the reductive glutamine metabolism towards citrate, a reaction that channels carbon units into the acetyl-CoA biosynthesis. However, the ACLy-dependent flux towards acetyl-coA synthesis was reduced and this was consistent with reduced expression of ACLy as well as the citrate transporter SLC25a1. A decrease in cellular acetyl-CoA would reduce histone acetylation. Untargeted metabolomics showed an increase in the concentration of S-adenosyl-methionine, suggesting increased DNA and histone methylation. Consistent with these findings, ATAC-Seq analyses on freshly isolated pancreatic progenitors showed reduced chromatin accessibility at genes implicated in chromatin organization, protein acetylation and histone modification. Transcription motif analysis showed that the affected genomic sites were mainly associated with the binding of Klf/Sp and Nrf1 transcription factors. Transcriptome analyses displayed that the expression of genes implicated in progenitor differentiation, ECM organization and transcriptional regulation was affected. Furthermore, transcriptome analyses of early postnatal β-cells uncovered early signs of oxidative stress and increased proliferation, thus providing the basis to explain the β-cell phenotype in Aldh1b1 null mice. I then used organotypic cultures of embryonic pancreata to investigate the connection between high ROS levels and aberrant differentiation in the Aldh1b1 null pancreata. Reducing ROS levels using NAC enabled the reversal of the aberrant transcription factor expression and increased viability of Aldh1b1 null explants, thus identifying high ROS levels as a driving force in this process. To investigate how persisting Aldh1b1 expression would affect progenitor differentiation, I generated ROSA26LSLAldh1b1, an inducible constitutive Aldh1b1 expression line. Progenitors with continuous Aldh1b1 expression avoided the endocrine cell fate, underscoring the importance of timely Aldh1b1 downregulation in the course of β-cell differentiation. Altogether, my work provides strong evidence for the role of Aldh1b1 as a metabolic regulator in the process of progenitor cell differentiation and identifies a link between metabolism and gene regulation through chromatin accessibility during development. Aldh1b1 inactivity causes defects in embryonic progenitor cells as well as postnatal β-cells and could therefore contribute, as genetic risk factor, to the development of hyperglycemia and diabetes later in life. Comprehending the mechanisms underlying the process of pancreas progenitor differentiation as well as the origins of β cell dysfunction should assist in the design of novel therapeutic interventions for diabetes.

Background and aims: Type 2 Diabetes Mellitus (T2DM) presents a significant global health challenge, characterized by impaired insulin secretion and/or action. A critical aspect of managing T2DM involves understanding the regulatory mechanisms of insulin secretion in pancreatic β-cells. Pancreatic β-cells play a pivotal role in maintaining glucose homeostasis. Although glucose is the primary stimulator of insulin secretion, certain amino acids also have regulatory roles. Traditional views have held that while glutamine contributes to insulin secretion, it does not directly influence this process in the absence of glutamate dehydrogenase (GDH) activation. We found that glutamine increases insulin secretion independently of GDH activation in INS-1 832/13 cells. Therefore, the aim of the thesis is to elucidate the role of glutamine in insulin secretion and examining its regulatory effects on glucose metabolism in pancreatic β-cells. To achieve this, we leverage advanced methodologies, including metabolomics and network analysis, to provide a comprehensive understanding of these complex mechanisms. Methods and results: Our initial findings presented a surprising challenge to the conventional belief that glutamine induces insulin secretion only in the presence of leucine. We discovered that glutamine (independent of leucine) could increase insulin secretion in a dose-dependent manner in INS-1 832/13 cells. To delve further into this phenomenon, we employed inhibitors of key enzymes in glutamine metabolism - GDH (responsible for glutamate oxidation) and glutaminase (converts glutamine to glutamate). Our results highlighted that while inhibiting GDH did not alter insulin secretion, inhibiting glutaminase significantly reduced the insulin-secretory response to glutamine in INS-1 832/13 cells. This finding indicated that the effect of glutamine on insulin secretion operates independently of glutamate oxidation. Our study also investigated the regulatory role of glutamine in insulin secretion and on the rate of glucokinase (GK) in response to glucose levels. We observed that increasing concentrations of glutamine affected both the dynamics of insulin secretion and the kinetic parameters of GK in INS-1 832/13 cells, suggesting a regulatory relationship between glutamine and glucose phosphorylation that had not been previously observed. To deepen our understanding of the intricate relationship, we developed a novel analytical approach that combined network analysis with metabolomics. This innovative method provided an unbiased assessment of the interrelationships between various metabolites, enabling a more comprehensive understanding of the metabolic pathways and their interactions. A striking outcome of our network analysis was the identification of proline as a key metabolite in the glutamine-glucose crosstalk. To validate this link, we conducted siRNA knockdown experiments targeting proline synthesis in INS-1 832/13 cells. Knockdown of these genes resulted in a significant reduction in insulin secretion in response to glutamine. Further, this effect could be rescued by the addition of proline, thereby underscoring the essential role of proline in glutamine-mediated insulin secretion. Furthermore, in vitro enzymatic assays using INS-1 832/13 cell extracts and purified rat GK revealed proline- mediated changes in kinetic parameters consistent with glutamine-mediated alterations in GK activity in live INS-1 832/13 cells. Additionally, a thermal stability assay demonstrated that the melting temperature of purified rat GK varied with proline concentration, suggesting a direct interaction of proline with GK. This effect of glutamine on insulin secretion was also observed in isolated rat islets, thereby affirming the physiological relevance of our results. Moreover, the thermal stability assay using purified human GK confirmed that this interaction is conserved in humans as well. Conclusion and outlook: This study reveals a novel mechanism by which glutamine metabolism, through proline synthesis, regulates GK activity and thereby influences insulin secretion in pancreatic β-cells. The outlook of this thesis opens promising avenues for future research and potential clinical applications, particularly in the context of T2DM management. Key areas for future exploration include translating these findings to in vivo models and clinical settings could open new therapeutic avenues for T2DM, emphasizing the importance of modulating glutamine and proline metabolism for more effective regulation of insulin secretion. Investigating the direct causal relationship between plasma proline levels and diabetic conditions could not only deepen our understanding of diabetes but also provide a potential biomarker for early risk assessment. Understanding the precise molecular interactions between proline and GK could allow the identification of potential novel binding sites for therapeutic intervention to enhance GK activity and improve glucose regulation. Extending this research to human cells and examining its implications in diabetes and other metabolic disorders is a vital next step, offering potential for significant advancements in diabetes treatment and understanding of metabolic diseases.:Table of Contents List of abbreviations List of figures List of tables 1. Introduction 1.1. Type 2 Diabetes 1.1.1. Definition, epidemiology, and risk factors 1.1.2. Pathophysiology of T2DM 1.1.3. Preserving or enhancing β-cell function 1.2. Physiology of pancreatic β-cells 1.2.1. Overview of glucose-stimulated insulin secretion 1.2.2. Regulation of glucose entry into the β-cells 1.2.3. Role of glucokinase as a glucose sensor 1.2.4. Regulation of mitochondrial metabolism in insulin secretion 1.2.5. Regulation of amino acid mediated insulin secretion 1.3. Metabolomics approach in studying β-cell function 1.4. Network analysis in metabolomics data analysis and interpretation 2. Aims of the study 3. Materials and Methods 3.1. Materials 3.1.1. INS-1 832/13 cells 3.1.2. Chemicals, solutions, and buffers for cell culture 3.1.3. Chemicals, solutions, and buffers for molecular and metabolic experiments 3.1.4. Software 3.2. Methods 3.2.1. Cell culture 3.2.1.1. Culturing INS-1 832/13 cells 3.2.1.2. Cryopreservation and thawing of INS1 832/13 cells 3.2.1.3. Isolation of rat islets 3.2.2. Expression and Purification of GST-fusion GK Proteins in E. coli. 3.2.3. Insulin secretion studies in INS1 832/13 cells 3.2.3.1. Effect of Glutamine on insulin secretion 3.2.3.2. Effect of chronic and acute exposure of glutamine on insulin secretion 3.2.3.3. Glutamine-responsive insulin secretion 3.2.3.4. Effect of glutamate oxidation in glutamine-mediated insulin secretion 3.2.3.5. Effect of glutamine on glucose-responsive insulin secretion 3.2.3.6. Effect of 2DG on glucose stimulated insulin secretion 3.2.3.7. Insulin and total protein quantification 3.2.4. Metabolomic experiments in INS-1 832/13 cells 3.2.4.1. Effect of specific perturbations on metabolomic profile 3.2.4.2. Effect of glutamine on metabolomic profile 3.2.5. Metabolomic analyses 3.2.5.1. LC-MS/MS method for characterization of metabolites 3.2.5.2. Metabolite concentration calculation 3.2.6. Network analysis 3.2.6.1. Metabolite network construction 3.2.6.2. Comparative metabolite analysis with weighted network metrics 3.2.7. GK kinetic studies 3.2.7.1. GK activity with GK activator in INS-1 832/13 cells 3.2.7.2. GK activity with glutamine in INS1 cells & rat islets 3.2.7.3. GK kinetics measurement 3.2.7.4. In vitro GK kinetic studies using cell extracts & purified GK enzyme 3.2.8. Gene expression analysis 3.2.8.1. RNA isolation 3.2.8.2. cDNA synthesis 3.2.8.3. qPCR 4. Results 4.1. Glutamine mediated insulin secretion in INS-1 832/13 cells 4.1.1. Glutamine alone stimulates insulin secretion 4.1.2. Glutamine amplifies insulin secretion independently of glutamate oxidation 4.2. Glutamine mediated insulin secretion and its impact on glucose responsiveness 4.2.1. Glutamine modulates the regulation of insulin secretion in INS-1 832/13 cells 4.2.2. Live cell GK activity measurement using 2DG uptake in INS-1 832/13 cells 4.2.3. Glutamine modulates GK activity in INS-1 832/13 cells 4.3. Identifying the glutamine-derived factor regulating GK activity 4.3.1. Network analysis to identify key metabolites associated with specific perturbations 4.3.2. Glutamine-induced insulin secretion is mediated by proline 4.3.3. Proline modulates GK activity in INS-1 832/13 cell extracts 4.3.4. Proline modulates activity of purified rat GK 4.3.5. Thermal stability assays in rat GK 4.3.6. siRNA knockdown of proline synthesis 4.4. Glutamine modulates insulin secretion and GK activity in rat islets 4.5. Proline interacts and modulate GK in human 5. Discussion 5.1. Reevaluating glutamine-mediated insulin secretion in pancreatic β-cells 5.2. Novel role of glutamine-mediated modulation of GK activity and insulin secretion in pancreatic β-cells 5.3. Network analysis as a tool to unravel complex interactions in metabolic research 5.4. Proline as a novel modulator of GK 5.5. Contrasting role of glutamine in pancreatic and liver metabolism 6. References 7. Summary 8. Zussammenfassung 9. Acknowledgements 10. Declaration / Hintergrund und Ziele: Typ-2-Diabetes mellitus (T2DM) stellt eine bedeutende globale Herausforderung für die Gesundheit dar und ist durch eine gestörte Insulinsekretion und/oder -wirkung gekennzeichnet. Ein entscheidender Aspekt bei der Behandlung von T2DM ist das Verstehen von Regulationsmechanismen der Insulinsekretion in den β-Zellen der Pankreas. Die β-Zellen der Bauchspeicheldrüse spielen eine zentrale Rolle bei der Aufrechterhaltung der Glukosehomöostase. Obwohl Glukose der primäre Stimulator der Insulinsekretion ist, spielen bestimmte Aminosäuren auch eine regulierende Rolle. Nach traditioneller Auffassung trägt Glutamin zwar zur Insulinsekretion bei, hat aber keinen direkten Einfluss auf diesen Prozess, es sei denn, er wird durch Glutamatdehydrogenase (GDH) aktiviert. Wir fanden heraus, dass Glutamin die Insulinsekretion unabhängig von der GDH-Aktivierung in INS-1 832/13-Zellen erhöht. Ziel dieser Arbeit war es daher, die Rolle von Glutamin bei der Insulinsekretion aufzuklären und seine regulierenden Effekte auf den Glukosestoffwechsel in β-Zellen der Pankreas zu untersuchen. Um dies zu erreichen, nutzen wir fortschrittliche Methoden, einschließlich Metabolomik- und Netzwerkanalysen, um ein umfassendes Verständnis dieser komplexen Mechanismen zu erlangen. Methoden und Ergebnisse: Unsere anfänglichen Ergebnisse stellten eine überraschende Inhomogenität zur herkömmlichen Annahme dar, dass Glutamin die Insulinsekretion nur in der Anwesenheit von Leucin induziert. Wir entdeckten, dass Glutamin (unabhängig von Leucin) die Insulinsekretion in INS-1 832/13-Zellen dosisabhängig steigern kann. Um dieses Phänomen näher zu untersuchen, setzten wir Hemmstoffe von Schlüsselenzymen des Glutaminstoffwechsels ein - GDH (verantwortlich für die Glutamatoxidation) und Glutaminase (konvertiert Glutamin zu Glutamat). Unsere Ergebnisse zeigten, dass die Hemmung der GDH die Insulinsekretion nicht modifizierte, während die Hemmung der Glutaminase die Insulinsekretionsantwort auf Glutamin in INS-1 832/13-Zellen deutlich verringerte. Diese Erkenntnis deutet darauf hin, dass die Wirkung von Glutamin auf die Insulinsekretion unabhängig von der Glutamatoxidation ist. In dieser Studie untersuchten wir weiterhin die regulatorische Rolle von Glutamin bei der Insulinsekretion und für die GK-Rate in Abhängigkeit vom Glukosespiegel. Wir stellten fest, dass steigende Glutaminkonzentrationen sowohl die Dynamik der Insulinsekretion als auch die kinetischen Parameter der Glucokinase (GK) in INS-1 832/13-Zellen beeinflussten, was auf eine bisher nicht erkannte regulatorische Beziehung zwischen Glutamin und Glukosephosphorylierung schließen lässt. Um unser Verständnis dieser komplexen Beziehung zu vertiefen, entwickelten wir einen neuartigen analytischen Ansatz, der die Netzwerkanalyse mit der Metabolomforschung kombinierte. Diese innovative Methode ermöglichte eine unvoreingenommene Bewertung der Wechselbeziehungen zwischen verschiedenen Metaboliten und damit ein umfassenderes Verständnis der Stoffwechselwege und ihrer Wechselwirkungen. Ein bemerkenswertes Ergebnis unserer Netzwerkanalyse war die Identifizierung von Prolin als Schlüsselmetabolit im Glutamin-Glukose-Crosstalk. Um diese Verbindung zu bestätigen, führten wir siRNA-Knockdown-Experimente durch, die auf die Prolinsynthese in INS-1 832/13-Zellen abzielten. Die Ausschaltung dieser Gene führte zu einer deutlichen Verringerung der Insulinsekretion als Reaktion auf Glutamin. Bemerkenswerterweise konnte dieser Effekt durch die Zugabe von Prolin wiederhergestellt werden, was die wesentliche Rolle von Prolin bei der Glutamin-vermittelten Insulinsekretion unterstreicht. Darüber hinaus ergaben in vitro Enzymassays mit INS-1 832/13-Zellextrakten und gereinigter Ratten-GK Prolin-vermittelte Veränderungen der kinetischen Parameter, die mit Glutamin-vermittelten Veränderungen der GK-Aktivität in lebenden INS-1 832/13-Zellen übereinstimmen. Darüber hinaus zeigte ein Thermal Stability Assay, dass die Schmelztemperatur von gereinigtem Ratten-GK mit der Prolin-Konzentration variierte, was auf eine direkte Interaktion von Prolin mit der GK hindeutet. Dieser Effekt von Glutamin auf die Insulinsekretion wurde auch in aus Ratten isolierten Langerhansschen Inseln beobachtet, was die physiologische Relevanz unserer Ergebnisse bestätigt. Darüber hinaus bestätigte der Thermal Stability Assay mit gereinigter menschlichen GK, dass diese Interaktion auch beim Menschen konserviert ist. Schlussfolgerung und Ausblick: Diese Studie enthüllt einen neuartigen Mechanismus, durch den der Glutamin-Stoffwechsel über die Prolin-Synthese die GK-Aktivität reguliert und dadurch die Insulinsekretion in den β-Zellen der Bauchspeicheldrüse beeinflusst, was bestehende Paradigmen in Frage stellt. Perspektivisch ermöglichen die Erkenntnisse dieser Arbeit vielversprechende Wege für die zukünftige Forschung und potenzielle klinische Anwendungen, insbesondere im Zusammenhang mit T2DM-Management. Zu den Schlüsselbereichen der zukünftigen Forschung gehören die Übertragung dieser Ergebnisse auf in vivo Modelle und klinische Studien, die neue therapeutische Wege für T2DM eröffnen könnten und die Bedeutung der Modulation des Glutamin- und Prolin-Stoffwechsels für eine effektivere Regulierung der Insulinsekretion unterstreichen. Die Untersuchung des direkten kausalen Zusammenhangs zwischen Plasmaprolinspiegeln und diabetischen Erkrankungen könnte nicht nur unser Verständnis von Diabetes vertiefen, sondern auch einen potenziellen Biomarker für eine frühzeitige Risikobewertung liefern. Die Entschlüsselung der genauen molekularen Wechselwirkungen zwischen Prolin und GK könnte die Identifizierung potenzieller neuer Bindungsstellen für therapeutische Eingriffe zur Steigerung der GK- Aktivität und zur Verbesserung der Glukoseregulierung ermöglichen. Die Erweiterung dieser Forschung auf menschliche Zellen und die Untersuchung ihrer Auswirkungen auf Diabetes und andere Stoffwechselstörungen ist ein wichtiger nächster Schritt, der das Potenzial für bedeutende Fortschritte bei der Behandlung von Diabetes und dem Verständnis von Stoffwechselkrankheiten bietet.:Table of Contents List of abbreviations List of figures List of tables 1. Introduction 1.1. Type 2 Diabetes 1.1.1. Definition, epidemiology, and risk factors 1.1.2. Pathophysiology of T2DM 1.1.3. Preserving or enhancing β-cell function 1.2. Physiology of pancreatic β-cells 1.2.1. Overview of glucose-stimulated insulin secretion 1.2.2. Regulation of glucose entry into the β-cells 1.2.3. Role of glucokinase as a glucose sensor 1.2.4. Regulation of mitochondrial metabolism in insulin secretion 1.2.5. Regulation of amino acid mediated insulin secretion 1.3. Metabolomics approach in studying β-cell function 1.4. Network analysis in metabolomics data analysis and interpretation 2. Aims of the study 3. Materials and Methods 3.1. Materials 3.1.1. INS-1 832/13 cells 3.1.2. Chemicals, solutions, and buffers for cell culture 3.1.3. Chemicals, solutions, and buffers for molecular and metabolic experiments 3.1.4. Software 3.2. Methods 3.2.1. Cell culture 3.2.1.1. Culturing INS-1 832/13 cells 3.2.1.2. Cryopreservation and thawing of INS1 832/13 cells 3.2.1.3. Isolation of rat islets 3.2.2. Expression and Purification of GST-fusion GK Proteins in E. coli. 3.2.3. Insulin secretion studies in INS1 832/13 cells 3.2.3.1. Effect of Glutamine on insulin secretion 3.2.3.2. Effect of chronic and acute exposure of glutamine on insulin secretion 3.2.3.3. Glutamine-responsive insulin secretion 3.2.3.4. Effect of glutamate oxidation in glutamine-mediated insulin secretion 3.2.3.5. Effect of glutamine on glucose-responsive insulin secretion 3.2.3.6. Effect of 2DG on glucose stimulated insulin secretion 3.2.3.7. Insulin and total protein quantification 3.2.4. Metabolomic experiments in INS-1 832/13 cells 3.2.4.1. Effect of specific perturbations on metabolomic profile 3.2.4.2. Effect of glutamine on metabolomic profile 3.2.5. Metabolomic analyses 3.2.5.1. LC-MS/MS method for characterization of metabolites 3.2.5.2. Metabolite concentration calculation 3.2.6. Network analysis 3.2.6.1. Metabolite network construction 3.2.6.2. Comparative metabolite analysis with weighted network metrics 3.2.7. GK kinetic studies 3.2.7.1. GK activity with GK activator in INS-1 832/13 cells 3.2.7.2. GK activity with glutamine in INS1 cells & rat islets 3.2.7.3. GK kinetics measurement 3.2.7.4. In vitro GK kinetic studies using cell extracts & purified GK enzyme 3.2.8. Gene expression analysis 3.2.8.1. RNA isolation 3.2.8.2. cDNA synthesis 3.2.8.3. qPCR 4. Results 4.1. Glutamine mediated insulin secretion in INS-1 832/13 cells 4.1.1. Glutamine alone stimulates insulin secretion 4.1.2. Glutamine amplifies insulin secretion independently of glutamate oxidation 4.2. Glutamine mediated insulin secretion and its impact on glucose responsiveness 4.2.1. Glutamine modulates the regulation of insulin secretion in INS-1 832/13 cells 4.2.2. Live cell GK activity measurement using 2DG uptake in INS-1 832/13 cells 4.2.3. Glutamine modulates GK activity in INS-1 832/13 cells 4.3. Identifying the glutamine-derived factor regulating GK activity 4.3.1. Network analysis to identify key metabolites associated with specific perturbations 4.3.2. Glutamine-induced insulin secretion is mediated by proline 4.3.3. Proline modulates GK activity in INS-1 832/13 cell extracts 4.3.4. Proline modulates activity of purified rat GK 4.3.5. Thermal stability assays in rat GK 4.3.6. siRNA knockdown of proline synthesis 4.4. Glutamine modulates insulin secretion and GK activity in rat islets 4.5. Proline interacts and modulate GK in human 5. Discussion 5.1. Reevaluating glutamine-mediated insulin secretion in pancreatic β-cells 5.2. Novel role of glutamine-mediated modulation of GK activity and insulin secretion in pancreatic β-cells 5.3. Network analysis as a tool to unravel complex interactions in metabolic research 5.4. Proline as a novel modulator of GK 5.5. Contrasting role of glutamine in pancreatic and liver metabolism 6. References 7. Summary 8. Zussammenfassung 9. Acknowledgements 10. Declaration

Le diabète de type 2 (DT2) résulte d’une résistance à l’insuline par les tissus périphériques et par un défaut de sécrétion de l’insuline par les cellules β-pancréatiques. Au fil du temps, la compensation des îlots de cellules β pour la résistance à l’insuline échoue et entraine par conséquent une baisse progressive de la fonction des cellules β. Plusieurs facteurs peuvent contribuer à la compensation de la cellule β. Toutefois, la compréhension des mécanismes cellulaires et moléculaires sous-jacents à la compensation de la masse de la cellule β reste à ce jour inconnue. Le but de ce mémoire était d’identifier précisément quel mécanisme pouvait amener à la compensation de la cellule β en réponse à un excès de nutriments et plus précisément à l’augmentation de sa prolifération et de sa masse. Ainsi, avec l’augmentation de la résistance à l’insuline et des facteurs circulants chez les rats de six mois perfusés avec du glucose et de l’intralipide, l’hypothèse a été émise et confirmée lors de notre étude que le facteur de croissance HB-EGF active le récepteur de l’EGF et des voies de signalisations subséquentes telles que mTOR et FoxM1 impliquées dans la prolifération de la cellule β-pancréatique. Collectivement, ces résultats nous permettent de mieux comprendre les mécanismes moléculaires impliqués dans la compensation de la masse de la cellule β dans un état de résistance à l’insuline et peuvent servir de nouvelles approches thérapeutiques pour prévenir ou ralentir le développement du DT2. / Type 2 diabetes (T2D) results from insulin resistance in peripheral tissues and impaired insulin secretion from the pancreatic β-cell. Over the time, compensation of the β cell islets for insulin resistance fails, and therefore leads to a gradual decline in β-cell function. Several factors may contribute to β-cell compensation. However, the cellular and molecular mechanisms underlying β-cell compensation remain unknown. The purpose of this thesis was to identify what mechanism could lead to β cell compensation in response to nutrients excess and specifically the increase in proliferation and β-cell mass. Thus, with increasing insulin resistance and circulating factors in the 6 month rats infused with glucose + intralipid, the hypothesis was made and confirmed in our study that the growth factor HB-EGF would activate the EGF receptor, and subsequent signaling pathways such as mTOR and FoxM1, both involved in the proliferation of the pancreatic beta-cell. Collectively, these results allow us to understand better the molecular mechanisms involved in the β cell compensation in the insulin resistance state and may serve as a potential new therapeutic approach to prevent or delay T2D development.

La greffe d’îlots pancréatiques est caractérisée par une réponse inflammatoire et procoagulante précoce, connue sous le nom d’IBMIR (Instant Blood Mediated Inflammatory Reaction). Les microparticules (MPs) porteuses de facteur tissulaire (TF) sont le témoin d’un important remodelage membranaire et constituent des acteurs centraux dans la dissémination du stress de l’IBMIR. Nous avons exploré l’effet d’un stress inflammatoire sur la survie et la fonction de la cellule β dans un modèle de communication cellulaire médiée par les MPs. La modulation pharmacologique par les analogues du GLP-1 a été évaluée, la par la mesure de la sécrétion d’insuline, de l’activité TF et l'analyse du remodelage de la membrane plasmique. Nos résultats décrivent les MPs comme des effecteurs autocrines et indiquent que les MPs sont des cibles potentielles pour les analogues du GLP-1 au cours de l'IBMIR. Les données apportent de nouvelles pistes sur les mécanismes cellulaires mis en jeu lors des phénomènes d’ischémie reperfusion durant l’IBMIR. / Pancreatic islets graft is characterized by early inflammatory and procoagulant events known as Instant Blood Mediated Inflammatory Reaction (IBMIR). Tissue factor (TF) bearing microparticles (MPs) are surrogates of important membrane remodeling and key players in the systemic and local dissemination of such stress.We investigated the effect of inflammatory stress on β cell survival and function in a MP-mediated cell crosstalk model. Pharmacological modulation by GLP-1 analogues was evaluated by measurement of insulin secretion, TF activity and assessment of plasma membrane remodeling. Our data evidenced MPs as autocrine effectors and possible new target for GLP-1 analogues. They bring new hints on the cellular mechanisms prompted by ischemia reperfusion during IBMIR.

The effect of gluten-free diet on β-cell residual capacity, immune function and gut microbiome in children with newly diagnosed type 1. diabetes Abstract The pathophysiology of the onset and progression of type 1 diabetes (T1D) is not fully understood. Gluten has a proinflammatory effect on the immune system and is therefore considered as one of the factors affecting the onset and progression of T1D. The aim of the thesis is to allow a complex insight into the role of the GFD on the residual β-cell capacity, T1D control, gut microbiome, gut permeability, subtypes of immune cells and the effect of gut microbiome transfer into germ-free non-obese diabetic (NOD) mice on the incidence of diabetes. On the group of 45 children with T1D (26 intervention group, 19 control group) we proved the association of the GFD with slower decrease of β-cell residual capacity (the difference in the trend of C-peptide decrease 409 pmol/l/year; p = 0,04) and lower HbA1c (by 7,8 mmol/mol; p=0,02). We also described the changes in the gut bacteria that were differentially abundant after the administration of the GFD and the changes in abundance of the regulatory and effector immune cells. We showed there was no change in the gut permeability with respect to the study group. We also proved that the transfer of human gut microbiota...

Background: The prevalence of diabetes and obesity has been on the rise for many years and the search for new and effective dietetic solutions aiming at reducing calories, reducing body mass and improving diabetes has been ongoing. Currently, the intermittent fasting diet - the practice of alternating periods of eating and fasting - is gaining popularity. One of them is Time-restricted feeding (TRF), which time-limits energy intake within a defined window of time up to 10 hours per day without necessarily altering diet quality or quantity. A reduction in calorie intake, bodyweight, blood pressure, oxidative stress, inflammation biomarkers and triglycerides are evident with TRF studies conducted so far. Aim: The aim of the thesis is to investigate the effects of time-restricted feeding on glycemic biomarkers in human studies. Methods: A literature study is conducted with six chosen experimental studies which are primarily randomized controlled trials or randomized crossover trials with a TRF window of maximum 10 hours per day and predominantly with participants with overweight/obesity, prediabetes, type 2 diabetes and metabolic syndrome. Results: Compared to either baseline and/or control group, fasting glucose was reduced in 3 out of 6 TRF studies, while fasting insulin was reduced in 3 out of 5 TRF studies and HbA1C was decreased in 1 out of 2 TRF studies. For postprandial response, 1 out of 2 TRF studies found a reduction in glucose and likewise for insulin. Mean glucose levels were reduced in 1 out of 3 TRF studies. Insulin resistance was reduced in 3 out of 4 TRF studies while insulin sensitivity was reduced in the one study measuring this. Beta cell function improved in 2 out of 2 TRF studies compared to the control group or baseline. Conclusion: There are indications that TRF has an effect on glycemic biomarkers and thus potentially being able to reduce the risk and/or improve the treatment of type 2 diabetes. But in order to give a more definite answer more studies need to be conducted. In general, these studies should preferably have more participants and be methodologically stronger when it e.g. comes to the control of the dietary regimen.

Le diabète de type 2 (DT2) est caractérisé par une résistance des tissus périphériques à l’action de l’insuline et par une insuffisance de la sécrétion d’insuline par les cellules β du pancréas. Différents facteurs tels que le stress du réticulum endoplasmique (RE) et l’immunité innée affectent la fonction de la cellule β-pancréatique. Toutefois, leur implication dans la régulation de la transcription du gène de l’insuline demeure imprécise. Le but de cette thèse était d’identifier et de caractériser le rôle du stress du RE et de l’immunité innée dans la régulation de la transcription du gène de l’insuline. Les cellules β-pancréatiques ont un RE très développé, conséquence de leur fonction spécialisée de biosynthèse et de sécrétion d’insuline. Cette particularité les rend très susceptible au stress du RE qui se met en place lors de l’accumulation de protéines mal repliées dans la lumière du RE. Nous avons montré qu’ATF6 (de l’anglais, activating transcription factor 6), un facteur de transcription impliqué dans la réponse au stress du RE, lie directement la boîte A5 de la région promotrice du gène de l’insuline dans les îlots de Langerhans isolés de rat. Nous avons également montré que la surexpression de la forme active d’ATF6α, mais pas ATF6β, réprime l’activité du promoteur de l’insuline. Toutefois, la mutation ou l’absence de la boîte A5 ne préviennent pas l’inhibition de l’activité promotrice du gène de l’insuline par ATF6. Ces résultats montrent qu’ATF6 se lie directement au promoteur du gène de l’insuline, mais que cette liaison ne semble pas contribuer à son activité répressive. Il a été suggéré que le microbiome intestinal joue un rôle dans le développement du DT2. Les patients diabétiques présentent des concentrations plasmatiques élevées de lipopolysaccharides (LPS) qui affectent la fonction de la cellule β-pancréatique. Nous avons montré que l’exposition aux LPS entraîne une réduction de la transcription du gène de l’insuline dans les îlots de Langerhans de rats, de souris et humains. Cette répression du gène de l’insuline par les LPS est associée à une diminution des niveaux d’ARNms de gènes clés de la cellule β-pancréatique, soit PDX-1 (de l’anglais, pancreatic duodenal homeobox 1) et MafA (de l’anglais, mammalian homologue of avian MafA/L-Maf). En utilisant un modèle de souris déficientes pour le récepteur TLR4 (de l’anglais, Toll-like receptor), nous avons montré que les effets délétères des LPS sur l’expression du gène de l’insuline sollicitent le récepteur de TLR4. Nous avons également montré que l’inhibition de la voie NF-kB entraîne une restauration des niveaux messagers de l’insuline en réponse à une exposition aux LPS dans les îlots de Langerhans de rat. Ainsi, nos résultats montrent que les LPS inhibent le gène de l’insuline dans les cellules β-pancréatiques via un mécanisme moléculaire dépendant du récepteur TLR4 et de la voie NF-kB. Ces observations suggèrent ainsi un rôle pour le microbiome intestinal dans la fonction de la cellule β du pancréas. Collectivement, ces résultats nous permettent de mieux comprendre les mécanismes moléculaires impliqués dans la répression du gène de l'insuline en réponse aux divers changements survenant de façon précoce dans l’évolution du diabète de type 2 et d'identifier des cibles thérapeutiques potentielles qui permettraient de prévenir ou ralentir la détérioration de l'homéostasie glycémique au cours de cette maladie, qui affecte plus de deux millions de Canadiens. / Type 2 diabetes is characterized by insulin resistance and impaired insulin secretion from the pancreatic β-cell. Endoplasmic reticulum (ER) stress and innate immunity have both been reported to alter pancreatic β-cell function. However, it is not clear whether these factors can affect the transcription of the insulin gene. The aim of this thesis was to assess the role of ER stress and innate immunity in the regulation of the insulin gene. Pancreatic β-cells have a well-developed endoplasmic reticulum (ER) due to their highly specialized secretory function to produce insulin in response to glucose and nutrients. In a first study, using several approaches we showed that ATF6 (activating transcription factor 6), a protein implicated in the ER stress response, directly binds to the A5/Core of the insulin gene promoter in isolated rat islets. We also showed that overexpression of the active (cleaved) fragment of ATF6α, but not ATF6β, inhibits the activity of an insulin promoter-reporter construct. However, the inhibitory effect of ATF6α was insensitive to mutational inactivation or deletion of the A5/Core. Therefore, although ATF6 binds directly to the A5/Core of the rat insulin II gene promoter, this direct binding does not appear to contribute to its repressive activity. In recent years, the gut microbiota was proposed has an environmental factor increasing the risk of type 2 diabetes. Subjects with diabetes have higher circulating levels of lipopolysaccharides (LPS) than non-diabetic patients. Recent observations suggest that the signalling cascade activated by LPS binding to Toll-Like Receptor 4 (TLR4) exerts deleterious effects on pancreatic β-cell function; however, the molecular mechanisms of these effects are incompletely understood. We showed that exposure of isolated human, rat and mouse islets of Langerhans to LPS dose-dependently reduced insulin gene expression. This was associated in mouse and rat islets with decreased mRNA expression of two key transcription factors of the insulin gene, PDX-1 (pancreatic duodenal homeobox 1) and MafA (mammalian homologue of avian MafA/L-Maf). LPS repression of insulin, PDX-1 and MafA expression was not observed in islets from TLR4-deficient mice and was completely prevented in rat islets by inhibition of the NF-kB signalling pathway. These results demonstrate that LPS inhibits β-cell gene expression in a TLR4-dependent manner and via NF-kB signaling in pancreatic islets, suggesting a novel mechanism by which the gut microbiota might affect pancreatic β-cell function. Our findings provide a better understanding of the molecular mechanisms underlying insulin gene repression in type 2 diabetes, and suggest potential therapeutic targets that might prevent or delay the decline of β-cell function in the course of type 2 diabetes, which affects more than two million Canadians.

Le diabète de type 2 (DT2) est caractérisé par une résistance des tissus périphériques à l’action de l’insuline et par une insuffisance de la sécrétion d’insuline par les cellules β du pancréas. Différents facteurs tels que le stress du réticulum endoplasmique (RE) et l’immunité innée affectent la fonction de la cellule β-pancréatique. Toutefois, leur implication dans la régulation de la transcription du gène de l’insuline demeure imprécise. Le but de cette thèse était d’identifier et de caractériser le rôle du stress du RE et de l’immunité innée dans la régulation de la transcription du gène de l’insuline. Les cellules β-pancréatiques ont un RE très développé, conséquence de leur fonction spécialisée de biosynthèse et de sécrétion d’insuline. Cette particularité les rend très susceptible au stress du RE qui se met en place lors de l’accumulation de protéines mal repliées dans la lumière du RE. Nous avons montré qu’ATF6 (de l’anglais, activating transcription factor 6), un facteur de transcription impliqué dans la réponse au stress du RE, lie directement la boîte A5 de la région promotrice du gène de l’insuline dans les îlots de Langerhans isolés de rat. Nous avons également montré que la surexpression de la forme active d’ATF6α, mais pas ATF6β, réprime l’activité du promoteur de l’insuline. Toutefois, la mutation ou l’absence de la boîte A5 ne préviennent pas l’inhibition de l’activité promotrice du gène de l’insuline par ATF6. Ces résultats montrent qu’ATF6 se lie directement au promoteur du gène de l’insuline, mais que cette liaison ne semble pas contribuer à son activité répressive. Il a été suggéré que le microbiome intestinal joue un rôle dans le développement du DT2. Les patients diabétiques présentent des concentrations plasmatiques élevées de lipopolysaccharides (LPS) qui affectent la fonction de la cellule β-pancréatique. Nous avons montré que l’exposition aux LPS entraîne une réduction de la transcription du gène de l’insuline dans les îlots de Langerhans de rats, de souris et humains. Cette répression du gène de l’insuline par les LPS est associée à une diminution des niveaux d’ARNms de gènes clés de la cellule β-pancréatique, soit PDX-1 (de l’anglais, pancreatic duodenal homeobox 1) et MafA (de l’anglais, mammalian homologue of avian MafA/L-Maf). En utilisant un modèle de souris déficientes pour le récepteur TLR4 (de l’anglais, Toll-like receptor), nous avons montré que les effets délétères des LPS sur l’expression du gène de l’insuline sollicitent le récepteur de TLR4. Nous avons également montré que l’inhibition de la voie NF-kB entraîne une restauration des niveaux messagers de l’insuline en réponse à une exposition aux LPS dans les îlots de Langerhans de rat. Ainsi, nos résultats montrent que les LPS inhibent le gène de l’insuline dans les cellules β-pancréatiques via un mécanisme moléculaire dépendant du récepteur TLR4 et de la voie NF-kB. Ces observations suggèrent ainsi un rôle pour le microbiome intestinal dans la fonction de la cellule β du pancréas. Collectivement, ces résultats nous permettent de mieux comprendre les mécanismes moléculaires impliqués dans la répression du gène de l'insuline en réponse aux divers changements survenant de façon précoce dans l’évolution du diabète de type 2 et d'identifier des cibles thérapeutiques potentielles qui permettraient de prévenir ou ralentir la détérioration de l'homéostasie glycémique au cours de cette maladie, qui affecte plus de deux millions de Canadiens. / Type 2 diabetes is characterized by insulin resistance and impaired insulin secretion from the pancreatic β-cell. Endoplasmic reticulum (ER) stress and innate immunity have both been reported to alter pancreatic β-cell function. However, it is not clear whether these factors can affect the transcription of the insulin gene. The aim of this thesis was to assess the role of ER stress and innate immunity in the regulation of the insulin gene. Pancreatic β-cells have a well-developed endoplasmic reticulum (ER) due to their highly specialized secretory function to produce insulin in response to glucose and nutrients. In a first study, using several approaches we showed that ATF6 (activating transcription factor 6), a protein implicated in the ER stress response, directly binds to the A5/Core of the insulin gene promoter in isolated rat islets. We also showed that overexpression of the active (cleaved) fragment of ATF6α, but not ATF6β, inhibits the activity of an insulin promoter-reporter construct. However, the inhibitory effect of ATF6α was insensitive to mutational inactivation or deletion of the A5/Core. Therefore, although ATF6 binds directly to the A5/Core of the rat insulin II gene promoter, this direct binding does not appear to contribute to its repressive activity. In recent years, the gut microbiota was proposed has an environmental factor increasing the risk of type 2 diabetes. Subjects with diabetes have higher circulating levels of lipopolysaccharides (LPS) than non-diabetic patients. Recent observations suggest that the signalling cascade activated by LPS binding to Toll-Like Receptor 4 (TLR4) exerts deleterious effects on pancreatic β-cell function; however, the molecular mechanisms of these effects are incompletely understood. We showed that exposure of isolated human, rat and mouse islets of Langerhans to LPS dose-dependently reduced insulin gene expression. This was associated in mouse and rat islets with decreased mRNA expression of two key transcription factors of the insulin gene, PDX-1 (pancreatic duodenal homeobox 1) and MafA (mammalian homologue of avian MafA/L-Maf). LPS repression of insulin, PDX-1 and MafA expression was not observed in islets from TLR4-deficient mice and was completely prevented in rat islets by inhibition of the NF-kB signalling pathway. These results demonstrate that LPS inhibits β-cell gene expression in a TLR4-dependent manner and via NF-kB signaling in pancreatic islets, suggesting a novel mechanism by which the gut microbiota might affect pancreatic β-cell function. Our findings provide a better understanding of the molecular mechanisms underlying insulin gene repression in type 2 diabetes, and suggest potential therapeutic targets that might prevent or delay the decline of β-cell function in the course of type 2 diabetes, which affects more than two million Canadians.

No description available.