Institute of Physiological Chemistry, Ulm University Head of Institute: Prof. Dr. Thomas Wirth

Functional analysis of hepatic FOXO3 signalling in glucose and lipid metabolism

DISSERTATION Dissertation submitted in partial fulfillment of the requirements for the degree of PhD (Dr. rer. nat) Faculty of Natural Sciences University of Ulm

Presented by Sarah Gul From Pakistan

2014 Dean: Prof. Dr. Joachim Ankerhold

First reviewer: Prof. Dr. Thomas Wirth Institute of Physiological Chemistry Ulm University

Second reviewer: Prof. Dr. Martin Wagner Centre for Internal Medicine Ulm University Medical Center Ulm University

Day doctorate awarded: 19th May 2014 Table of Contents

Table of Contents

Table of contents 1 Summary 4 Abbreviations 6 1 Introduction 12 1.1 Forkhead transcription factors 12 1.1.1 Forkhead box subclass O 12 1.1.2 Regulation of FoxO signalling 14 1.2 Role of FoxO transcription factors in metabolism 16 1.2.1 Role of FoxO transcription factors in the liver regulating glucose homeostasis 17 1.2.2 Role of FoxO transcription factors in lipid metabolism 20 1.2.3 Role of FoxO transcription factors in pancreatic β-cells 21 1.3 Liver is a “complex metabolic” organ 21 1.3.1 Regulation of hepatic glucose production 24 1.3.2 Regulation of transcription factors controlling gluconeogenesis 25 1.3.3 Regulation of hepatic lipid metabolism 28 1.4 Diabetes mellitus 29 1.4.1 Type 2 diabetes mellitus 30 1.4.2 Obesity and type 2 diabetes 32 1.4.3 Factors controlling β- mass and type 2 diabetes 34 1.4.4 Molecular mechanism of insulin resistance 34 1.4.5 Genetics of type 2 diabetes 36 1.4.6 Animal models of type 2 diabetes 37 1.5 Metformin 39 1.5.1 Metformin as an anti-hyperglycemic agent 40 1.5.2 Mechanism of metformin action 41 1.6 Aims of the study 45 2 Materials and Methods 46 2.1. Mice 46 2.1.1 Animal studies 46 2.1.2 Doxycycline administration 46 2.1.3 Metformin administration 46

1 Table of Contents

2.2 Genotyping 47 2.2.1 Isolation of genomic DNA 47 2.2.2 Polymerase chain reaction (PCR) 47 2.3 In Vivo Bioluminescence imaging 48 2.4 Metabolic studies 48 2.4.1 Glucose Tolerance Test (GTT) and Insulin Tolerance Test (ITT) 48 2.4.2 Animal studies 49 2.4.3 -linked immunosorbent assay (ELISA) 49 2.5 biochemistry 49 2.5.1 Protein extraction with dignam C buffer 49 2.5.2 Protein concentration measurement by Bradford method 50 2.5.3 In vitro Luciferase reporter activity measurement 50 2.5.4 Western blotting 50 2.6 RNA Analysis 52 2.6.1 RNA extraction and cDNA synthesis 52 2.6.2 analysis 54 2.7 Histology 54 2.7.1 Liver tissue preparation 54 2.7.2 Immunofluorescence staining of paraffin embedded sections 55 2.7.3 Immunofluorescence image aquisition and processing 56 2.7.4 Quantification of TUNEL staining, Cd45 and Insulin staining 56 2.7.5 Immunohistochemical staining of paraffin embedded sections 56 2.7.6 PAS staining for Glycogen 57 2.8 Statistical analysis 57 2.9 Reagents and Materials 58 2.9.1 General chemicals 58 2.9.2 Buffers and solutions 59 2.9.3 Protein biochemistry materials 60 2.9.4 Histology and microscopy materials 61 2.9.5 Molecular biology materials 61 2.9.6 General laboratory materials 62 3 Laboratoryequipment 62 3.1 General laboratory equipment 62 3.2 Histology and microscopy equipment 63

2 Table of Contents

4 Software 64 3 Results 65 3.1 Generation of transgenic mice with inducible liver-specific expression of constitutively active FOXO3 65 3.1.1 Tetracycline regulated expression of a constitutively active FOXO3 allele driven by the LAP promotor 65 3.2 Perinatal induction of FOXO3CA results in lethal 66 3.2.1 liver-specific expression of FOXO3CA 66 3.2.2 FOXO3CA activation affects size and structure of hepatocytes and induces liverdysfunction and metabolic abnormalities 69 3.2.3 FOXO3CA activation in younger animals induces autophagy 72 3.3 FOXO3CA activation in adult animals 73 3.3.1 Expression pattern of FOXO3CA transgene 73 3.3.2 FOXO3CA activation in adult animals is not associated with growth defect 76 3.3.3 FOXO3CA activation results in altered hepatic morphology 76 3.3.4 FOXO3CA activation induces moderate hepatocyte damage 77 3.3.5 Hepatocellular FOXO3 activation induces metabolic abnormalities 80 3.3.6 Activation of catabolic pathways by expression of constitutively active FOXO3 82 3.3.7 Continuous expression of constitutively active FOXO3 in the liver induces loss of gluconeogenic potential 88 3.3.8 FOXO3CA activation in the liver induces subordinate pancreatic hyperplasia 89 3.4 Metformin normalizes FOXO3-induced distorted glucose and insulin metabolism 91 3.5. FOXO3-induced activation of global catabolism, hepatocyte atrophy, or hepatocyte damage were reversed after Metformin administration 96 4 Discussion 102 5 References 116 6 Appendix 135 Declaration 154 Acknowledgement 155 Curriculum Vitae 156

3 Summary

Summary

Type 2 diabetes mellitus is a complex metabolic disorder characterized by high blood glucose levels and insulin resistance that occur as a consequence of uncontrolled gluconeogenesis that is failed to be suppressed by insulin. It is a major worldwide health care problem affecting the quality of life. The physiological processes contributing to insulin resistance in type 2 diabetes remain enigmatic. Therefore, there is a continuous need to identify molecular events responsible for type 2 diabetes and insulin resistance and this could help to discover new targets for the improvement of anti-diabetic drugs. Currently metformin is the most widely used oral anti-diabetic drug that inhibits hepatic glucose production, although the underlying molecular mechanism is not completely understood. FoxO transcription factors represent key downstream targets of insulin and growth factors regulating energy metabolism. However the interplay between different FoxO family members in metabolism is far from being understood. Although it had been suggested that FoxO1 is the major executer for regulation of insulin signalling mediated glucose. In addition to FoxO1, FoxO3 is also significantly expressed in the liver its contribution to metabolic regulation has however not been thoroughly analysed. Interestingly FoxO3 genotypes are associated with insulin sensitivity and longevity in humans suggesting that FoxO3 is critical for metabolic control. Therefore in order to understand the importance of FoxO3 function in the liver in a time- and cell-type specific manner, we generated transgenic (FoxO3CAHep) mice that allow the conditional expression of a constitutively active FoxO3 allele. We crossed transgenic mice carrying the tetracyclin-regulated transactivator (tTA) under the control of liver-activator protein promoter with transgenic mice bearing a constitutively active form of the FoxO3 allele (FoxO3CA) controlled by tTA- responsive promoter. Here we demonstrate that FoxO3 is a key regulator of hepatic glucose and lipid metabolism in transgenic mice. FoxO3 activation led to progressive hepatocyte atrophy in the absence of significant and apoptosis with moderate hepatocyte damage. Perinatal activation of FoxO3 results in lethal phenotype and the transgenic animals survived showed hypoglycemic phenotype with moderate upregulation of autophagy-associated . However, FoxO3 activation in adult animals showed hyperglycemic, hyperinsulinemic phenotype followed by up-

4 Summary regulation of gluconeogenesis-associated genes. The elevated blood glucose levels in transgenic mice are coupled with impaired glucose tolerance and impaired insulin sensitivity. Furthermore expression of FoxO3CA results in activation of catabolic pathways in transgenic mice including increased glycogenolysis, inhibition of lipid synthesis and activation of lipid β-oxidation. Consistently the results from gene expression profiles indicated up-regulation of genes involved in lipid and glucose catabolic pathways and down-regulation of anabolic genes as shown previously in type 2 diabetic subjects. This suggests that FoxO3 activates catabolic pathways to provide substrates and energy for enhanced gluconeogenesis. For our surprise the disease progression in transgenic mice showed hypoglycemia, might be as a consequence of beginning of liver failure or this could be due to high insulin production by pancreatic β-cells. Indeed the transgenic mice showed significantly higher insulin levels followed by pancreatic hyperplasia. It is plausible that the pancreatic hyperplasia in transgenic mice could arise might be due to the normal compensatory response of pancreas to enhanced insulin secretion or several other reasons. Strikingly FoxO3CA-induced metabolic alterations in glucose and lipid metabolism were completely normalized after metformin treatment in mice expressing constitutively active FoxO3. Taken together our findings suggest that dysbalanced FoxO3 activity in the liver results in insulin resistance. Therefore tight regulation of FoxO3 activity would be very crucial and agents controlling hepatic FoxO3 system may represent therapeutic tools to prevent type 2 diabetes.

5 Abbreviations

Abbreviations

AMPK Adenosine mono-phosphate (AMP)-activated protein AMPKα1α2 AMPK alpha1 and alpha 2 Ab APS Ammonium persulfate ADA American Diabetes Association ApoCIII Apolipoprotein C3 ALT Alanine aminotransferase ATG1 Autophagy related gene 1 ATG5 Autophagy related gene 5 ATG12 Autophagy related gene 12 AST Aspartate aminotransferase ALP Alkaline phosphatase AdipoR1/2 Adiponectin receptor 1 and 2 AMP Adenosine mono-phosphate ATP Adenosine tri-phosphate Acc Acetyl-CoA carboxylase BSA Bovine serum albumin CA Constitutively active CD45 cluster of differentiation 45 cDNA Complementary DNA CPT1α Carnitine palmitoyltransferase 1α CREB cAMP response element-binding protein CRTC2 CREB-regulated transcriptional coactivator2 CRE cAMP-response element CNS Central nervous system DOX Doxycycline DNL De novo lipogenesis DNA Deoxyribonucleicacid DKO Double knockout DLKO Akt1 Akt2 knockout dNTP Deoxynucleotide triphosphate

6 Abbreviations

DTT Dithiothreitol Dil Dilution DAPI 4´,6-diamidino-2-phenylindole ER Endoplasmicreticullum ERK Extracellular regulated kinase EDTA Ethylenediamine tetraacetic acid ELISA Enzyme-linked immunosorbent assay Echs Enoyl-CoA hydratase ERK Extracellular signal-regulated FOXO3CA Constitutively active FOXO3 FTO Fat mass and obesity related gene FFA Free fatty acid Fabp5 Fatty acid binding protein 5 Fas Fatty acid synthase Fbpase Fructose-1,6-biphosphatase FOXO1ADA Constitutively nuclear mutant FOXO1 G6pc Glucose-6-phosphatase GTT Glucose tolerance test Glut2 Glucose transporter Gck Glucokianse GEO Gene expression Omnibus GCKR Glucokinase regulatory protein GWAS Genome wide association and linkage studies GLP-1 Glucagon-like-peptide 1 GR Glucocorticoid receptor GRE Glucocorticoid receptor binding element Gckr Glucokinase regulatory protein HPRT Hypoxanthine-guanine phosphoribosyltransferase gene HbA1c Glycated haemoglobin A1c HNF Hepatic nuclear factor HNF4α Hepatic nuclear factor-4α HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid HGF Hepatocyte growth factor HFD High fat diet

7 Abbreviations

IR Insulin receptor ITT Insulin tolerance test IB Immunoblot IF Immunofluorscence IHC Immunohistochemistry i.p. Intraperitoneally IKK IκB kinase Irs1/Irs2KO Insulin receptor substrate1and 2 knockout IKK2CA IKK2 constitutively active IKKβ IκB kinase subunit β IFG Impaired fasting glucose InsR Insulin receptor IGT Impaired glucose tolerance IGFBP1 Insulin like growth factor binding protein 1 IRS1 Insulin receptor substrate-1 IRS2 Insulin receptor substrate-2 JNK c-Jun N-terminal kinases KLF15 Krüpple-like factor 15 K8/18 Keratin 8/18 LKB1 Liver kinase B1 LKB1-KO Liver kinase B1 knockout LPCKO Liver parenchymal knockout Lepr Leptin receptor LPL Lipoprotein lipase L-p110αKO Liver-specific deletion of catlytic subunit of PI3K (p110α) LD Lipid droplets LAP Liver activator protein LIRKO Liver-specific insulin receptor knockout mice LW/BW Liver weight/ Body weight MTP Microsomal triglyceride transfer protein MEK1 Mitogen activated protein kinase1 NADPH Nicotinamide adenine dinucleotide phosphate NFAT Nuclear factor of activated T-cells OGTT Oral glucose tolerance test

8 Abbreviations

OCT1 Oragnic cation transporter 1 OCT2 Oragnic cation transporter 2 PKB B P53 Cellular tumor antigen/ Tumor suppressor PBS Phosphate buffer saline PCR Polymerase chain reaction PAGE Polyacryamide gel electrophoresis rpm Revolutions per minute Pygl Glycogen phosphorylase Pepck Phosphoenolpyruvate carboxykinase Pdk4 Pyruvate dehydogenase kinase 4 PKA PI3K Phosphatidylinsositol 3-kinase PAS Periodic Acid Schiffs Pdx-1 Pancreatic and duodenal homeobox 1 PIP3 Phosphatidylinositol (3,4,5)-triphosphate PTEN Phosphatase and tension homologue Pklr Pyruvate kinase liver PGC-1α Peroxisome proliferator activated receptor γ coactivator-1α q-PCR Quantitative realtime-PCR RLU Relative light units RNA Ribonucleic acid RIPA Radio immunoprecipitation assay RBCs Erthrocytes/ Red blood cells Rb Rabbit Rb and P53 Tumor suppressor genes RT Room temperature SD Standard deviation Srebp1 Sterol regulatory element-binding protein 1c STK11 Serine/Threonine kinase 11 SDS Sodiumdodecylsulfate SAT Subcutaneous adipose tissue SOCS Supressors of signalling SD Standard deviation

9 Abbreviations

SIRT1 NAD-dependent deacetylase 1 STZ Streptozotocin SHP Small heterodimer partner SHIP SH2-containing inositol phosphatase 1 SPF Specific pathogen-free S253 Serine253 S315 Serine315 tTA Tetracycline-responsive transactivator tetO Tet operator/ tetracyclin response element T2D Type 2 Diabetes mellitus TUNEL Terminal deoxynucleotidyl dUTP nick end labeling TBS Tris buffer saline TEMED Tetramethyethylenediamine TNT buffer Tris-Nacl-Tween buffer TKO Triple knockout Trb3 Psuedokianse tribble homologue 3 TG Triglycerides T32 Threonine 32 TAK1 Transforming growth factor (TGF)-β-activated kinase 1 TE Tris-EDTA TBE Tris-buffer saline TCA Tricarboxylicacid cycle UAE United Arab Emirates UPL Universal probe libraray UV Ultravoilet irradiation VLDL Very low density lipoprotein WHO World Health Organization WT Wild type XBP1 X-box binding protein 1 ZFR Zucker fatty rat ZDR Zucker diabetic rat µg Microgram µl Microlitre 7Wk 7-week

10 Abbreviations

9Wk 9-week 6Wk 6-week 5Wk 5-week (-)Met Without Metformin (+)Met3D With Metformin 3-days (+)Met3W With Metformin 3-weeks

11 Introduction

1 Introduction

1.1 Forkhead transcription factors

The family of forkhead box proteins belongs to a major group of transcription factors present in all eukaryotes. Thus are transcriptional regulators characterized by conserved DNA-binding domain called “Forkhead box”. Forkhead box is a 110 amino-acid region located in the central portion of molecules and is classified as winged helix structure. The family of forkhead transcription factors has many subgroups ranging from (Forkhead Box A-S). An important subgroup of forkhead box proteins is the FoxO subgroup (Greer and Brunet 2005; Calnan and Brunet 2008).

1.1.1 Forkhead box subclass O

The O branch of the large Forkhead family of transcription factors is highly conserved evolutionarily and ubiquitously expressed (Huang and Tindall 2007; Calnan and Brunet 2008). The prototypes of the FoxO family were first described in Caenorhabditis elegans and drosophila demonstrating the task of FOXO proteins as the downstream targets of insulin-like signalling involved in the modulation of longevity and metabolism. Mammals contain four members of the Forkhead family such as FOXO1, FOXO3, FOXO4 and FOXO6. The FOXO family was initially described in human tumors as three members of Forkhead family; FOXO1, FOXO3 and FOXO4 were identified at chromosomal translocations in human cancers (Galili, Davis et al. 1993; Davis, D'Cruz et al. 1994; Parry, Wei et al. 1994; Borkhardt, Repp et al. 1997). FOXO1, FOXO3, FOXO4 are widely distributed and are expressed in almost all tissues whereas FOXO6 is largely specific to neurons. FOXO transcription factors attaches to DNA in the form of monomers on conserved consensus core recognition motif TTGTTTAC in the nucleus. FOXO transcription factors perform diversified function by binding to the consensus core recognition motif and promote the transcriptional activities of genes regulating glucose metabolism, cell cycle arrest, apoptosis, stress resistance, autophagy, cellular atrophy in response to oxidative stress and DNA damage (Figure 1) (Hwangbo, Gershman et al. 2004; Kenyon 2010; Eijkelenboom and Burgering 2013).

12 Introduction

Figure 1: Functions of FoxO transcription factors (Eijkelenboom and Burgering

2013). FOXO transcription factors perform diversified functions and hence function as the regulators for homeostasis during the process of fasting and feeding. FOXO transcription factors get activated in response to any metabolic stress e.g starvation, growth factor deprivation and oxidative stress. FOXO transcription factors perform various functions by up-regulating a series of target genes involved in the regulation of gluconeogenesis, cell-cycle arrest, differentiation, DNA repair, autophagy and apoptosis.

FoxO transcription factors perform overlapping functions and it is still unknown whether these factors share similar target genes or have different subsets of target genes. The complexity of FoxO transcription factors is supported by the notion that mice missing three main FoxO isoforms exhibit different phenotypes. Homozygous FoxO1 knockout mice are embryonically lethal (Hosaka, Biggs et al. 2004) due to defective angiogenesis. FoxO3 knockout mice produce premature ovarian failure and FoxO4 knockout mice are with no obvious phenotype (Paik, Kollipara et al. 2007). It is also evidenced that FoxO transcription factors perform redundant functions and

13 Introduction each of these can compensate to some degree for the loss of the other as triplet conditional knockout leads to severe abnormalities in hematopoietic systems which was not observed in double or single knockout combinations (Nakae, Oki et al. 2008; Tikhanovich, Cox et al. 2013). However some specificity is contributed among FoxO members as FoxO1 performs a major role in the regulation of glucose homeostasis and FoxO3 is an important factor for longevity in invertebrates, mice and humans (Brunet, Bonni et al. 1999; van der Horst and Burgering 2007; Sedding 2008; Willcox, Donlon et al. 2008; Flachsbart, Caliebe et al. 2009) and it is strongly correlated with antioxidant stress response and tumor suppressor activity (Huang and Tindall 2007; Myatt and Lam 2007; Paik, Kollipara et al. 2007).

1.1.2 Regulation of FoxO signalling

FoxO transcription factors are the downstream targets of insulin/PI3K/PKB/AKT signalling pathway. During fed/unstressed conditions binding of insulin or growthfactors to their tyrosine kinase receptors prompt the recruitment and activation of the serine/threonine kinase Akt and phosphoinositide kinase (PI3K) which results in inactivation of FoxO transcription factors through AKT induced of FoxO transcription factors at conserved three amino acids (FoxO3: T32, S253, S315). This subsequently results in rapid relocalization of FoxO transcription factors from the nucleus to cytoplasm. Phosphorylated FoxO transcription factors respectively interact with 14-3-3 proteins, which function as molecular chaperons to remove FoxO proteins out of the nucleus (Brunet, Bonni et al. 1999; Brunet, Kanai et al. 2002) (Figure 2). Conversely during fasted/stressed conditions in the absence of growth factors and insulin FoxO transcription factors localizes in the nucleus to upregulate a series of target genes (Greer and Brunet 2005).

14 Introduction

Figure 2: FoxO transcription factors are activated in response to Insulin signalling pathway adapted from (Greer and Brunet 2005). During feeding/unstressed conditions activated insulin-signalling results in phophorylation/inactivation and proteosomal degradation of FOXO transcription factors. FOXO trancription factors are no longer in the nucleus to regulate their metabolic target genes. In contrast during starved/stressed conditions due to absence of insulin and growth factors FOXO transcription factors gets activated, move into the nucleus, bind to their promoters to upregulate the series of metabolic target genes.

Additionally cellular stress conditions can also cause activation of FoxO transcription factors via AMPK (Greer, Banko et al. 2009), or under oxidative stress conditions via JNK mediated phosphorylation (Eijkelenboom and Burgering 2013). FoxO transcription factors are also modulated in response to several post translational modifications acetylation, ubiquitination and phosphorylation which results in altered subcellular localization, a modulated transcriptional-regulatory activity and change in protein stability (Greer and Brunet 2005).

15 Introduction

1.2 Role of FoxO transcription factors in metabolism

FoxO transcription factors are involved in the metabolism regulation through their effects in various organs such as liver, pancreas, adipose tissue and skeletal muscle. Alterations in FoxO function in simple organisms affects the process of aging and lead to metabolic disorders including diabetes. This can be hypothesized from the list of known target genes which play an important role in metabolism (Table 1) (Barthel, Schmoll et al. 2005). The role of FoxO proteins in the context of regulating glucose homeostasis in the liver and pancreas will be discussed in the following sections.

Table 1: Metabolic selected FoxO target genes (Barthel, Schmoll et al. 2005)

Genes FoxO-effect Organ or cell type Metabolic effects References G6Pase Upregulation Liver, Kidney Increased (Schmoll, Walker gluconeogenesis et al. 2000), (Nakae, Kitamura et al. 2001) G6Pase Upregulation Liver Increased (Kallwellis-Opara, transporter gluconeogenesis Zaho et al. 2003) PEPCK Upregulation Liver, Kidney Increased (Nakae, Kitamura gluconeogenesis et al. 2001; Altomonte, Richter et al. 2003), (Hall, Yamasaki et al. 2000) IGFBP-1 Upregulation Liver Inhibition of IGF-1 (Guo, Rena et al. 1999; Yeagley, Guo et al. 2001) (Hall, Yamasaki et al. 2000) PGC-1α Upregulation Liver Increased (Daitoku, gluconeogenesis Yamagata et al. 2003) PDK4 Upregulation Muscle, Liver Glucose saving (Furuyama, Kitayama et al. 2003).(Kwon, Huang et al. 2004) LPL Upregulation Muscle Triglyceride (Kwon, Huang et clearance, fatty al. 2004) acid metabolism HMG-CoA Upregulation Liver Ketone body (Nadal, Marrero et synthase production al. 2002) PDX-1 Downregulation Pancreatic β-cell Inhibition of β-cell Kitamura, T. et al, differentiation 2002, P21 Upregulation Adipocyte Inhibition of fat-cell (Nakae, Kitamura differentiation et al. 2003) AdipoR1/2 Upregulation Liver, Muscle Fatty acid (Tsuchida, oxidation, glucose Yamauchi et al. uptake 2004) 16 Introduction

1.2.1 Role of FoxO trancription factors in the liver regulating Glucose homeostasis

As outlined above the major role of hepatic FoxO proteins includes the regulation of gluconeogenesis and among the FoxO subgroup FoxO1 was shown as the major insulin-regulated involved in the modulation of hepatic glucose production (Nakae, Kitamura et al. 2001; Accili and Arden 2004). It was shown in mouse models of insulin resistance that FoxO1 transcription factors are deregulated (Cheng, Guo et al. 2009) consistently gluconeogenic genes glucose-6-phosphatase (G6pc), phosphoenolpyruvate carboxykinase (Pepck) and pyruvate-dehydrogenase kinase 4 (Pdk4) are up-regulated during insulin resistance, fasting, type 2 diabetes, insulin deficiency state and downregulated at postprandial state by insulin (O'Brien and Granner 1996; O'Brien, Streeper et al. 2001). Accordingly mice with liver-specific inactivation of FoxO1 showed hypoglycemia, decreased gluconeogenic gene expression and enhanced glucose tolerance (Matsumoto, Pocai et al. 2007). Furthermore (Altomonte, Richter et al. 2003) showed that hepatic inhibition of FoxO1 in diabetic mice rescues the diabetic phenotype of insulin resistance by lowering the gluconeogenic gene expression. Consistently it was shown that targeting FoxO1 in diet-induced obese mice using antisense oligoneucleotides reduced G-6pc and Pepck gene expression with improved glucose tolerance and decreased hepatic glucose production in the liver (Samuel, Choi et al. 2006). In line with these studies haploinsufficiency of FoxO1 (Nakae, Biggs et al. 2002) in InsR haploinsufficient mice rescued the loss of insulin sensitivity by reducing the hepatic gluconeogenic gene expression. Additionally ablation of hepatic FoxO1 in mice lacking the insulin receptor substrates IRS1 and IRS2 rescues the insulin resistance and hepatic lipid abnormalities (Matsumoto, Pocai et al. 2007; Dong, Copps et al. 2008). It was further reported that hepatic deletion of all three FoxO isoforms (FoxO1, FoxO3, FoxO4) in mice showed more pronounced fasting hypoglycemia, improved glucose tolerance and enhanced insulin sensitivity (Haeusler, Kaestner et al. 2010; Zhang, Li et al. 2012; Xiong, Tao et al. 2013) (Table 2).

17 Introduction

Table 2: Characteristic metabolic features of hepatic FoxO-null mouse models (Estall 2012)

Genotype Background Glycemia Insulin Serum Hepatic Ref (liver lipids lipids specific) Foxo1-/- WT Decreased –– –– –– Zhang et al. IRSR-/- Decreased Resistant Decreased Decreased 2012 IRS1/2-/- Decreased Resistant Increaseda –– Mutsomoto WT + STZ Decreased Depleted Increased –– etal.2007 db/db Decreased Resistantb –– –– Dong et al. 2008 Haeusler et al. 2010 Zhang et al. 2012 Foxo3-/- WT –– –– –– –– Zhang et al. db/db –– Resistantb –– –– 2012 Zhang et al. 2012 Foxo4-/- WT –– –– –– Zhang et al. 2012 Foxo1/3-/- WT Decreased Decreased NR NR Haeusler et al. WT Decreased Decreased Increased Increased 2010 db/db Decreased Resistantb Increased –– Zhang et al. 2012 Zhang et al. 2012 Foxo1/4-/- WT Decreased NR –– NR Zhang et al. 2012 Foxo3/4-/- WT –– NR –– NR Zhang et al. 2012 Foxo1/3/4-/- WT Decreased Decreased NR NR Haeusler et al. WT NR NR –– Increased 2010 WT + HFD NR Resistantb Increased Increased Tao et al. 2011 WT Decreased NR Increased NR Tao et al. 2011 Zhang et al. 2012 HFD, High fat diet; STZ, Streptozotocin; WT, Wild type;__, no significant change; NR, not reported. a with age b db/db mouse insulin resistance due to obesity.

However, it is important to know that in triple knockout studies the effects achieved are redundant or in a synergistic manner still remains unknown, as the relative contributions of individual FoxO proteins were not addressed. In contrast constitutive activation of FoxO1 in the liver (Zhang, Patil et al. 2006) results in impaired hepatic glucose production through increased expression of gluconeogenic genes G-6-pc, Pepck. Additionally, gain-of-function of FoxO1 in the liver paradoxically increases insulin sensitivity and leads to steatosis (Matsumoto,

18 Introduction

Han et al. 2006). Of interest under certain conditions FoxO1 gain of functions in the liver results in insulin sensitivity through its increased ability to phophorylate Akt in Trb3 (Tribbles homologue 3)-dependent or -independent manner (Matsumoto, Han et al. 2006; Naimi, Gautier et al. 2007). Activation of FoxO1 in insulin resistance states may improve insulin sensitivity through induction of Irs2 (Ide, Shimano et al. 2004). Interestingly several mouse models demonstrated that hepatic deletion of insulin receptor (IR) (Michael, Kulkarni et al. 2000), the p110 a form of PI3K(Sopasakis, Liu et al. 2010), AKT2 (Cho, Mu et al. 2001) or AKT1 and AKT2 (Lu, Wan et al. 2012) and IRS1/IRS2 (Dong, Copps et al. 2008) all result in insulin resistance and diabetes. The ablation of IR, Akt1/Akt2, IRS1/IRS2 lead to constitutive activation of FoxO transcription factors and as a consequence aberrant gluconeogenesis is observed (Figure 3).

Figure 3: models that leads to constitutive activation of FoxO transcription factors.

FoxO transcription factors also interact with other coactivators and transcription factors for regulating gucose homeostasis e.g PGC1α, HNF4 and CREB (Herzig, Long et al. 2001; Yoon, Puigserver et al. 2001; Puigserver, Rhee et al. 2003; Schilling, Oeser et al. 2006) and deactylases SIRT1 (Frescas, Valenti et al. 2005).

19 Introduction

Several novel mechanisms have also been shown to regulate glucose metabolism such that interaction of FoxO1 with XBP-1. XBP-1 is the transcription factor involved in insulin sensitivity and it was evidenced that direct binding of XBP-1 to FoxO1 directs it to the proteosomal degradation (Zhou, Lee et al. 2011). Similarly glycosylation modifications promote the activation of FoxO transcription factors and hence cause the upregulation of gluconeogenic target genes (Housley, Udeshi et al. 2009).

1.2.2 Role of FoxO transcription factors in lipid metabolism

FoxO transcription factors play an important role in regulating lipid metabolism. As shown later the role of glucokinase (GCK) (Figure 5) in assisting lipogenesis it was shown that constitutive activation of FoxO1 in the liver results in down-regulation of the glycolytic pyruvate kinase and glucokinase subsequently and decreased sterol/lipid synthesis results in reduced concentration of triglycerides in transgenic mice (Zhang, Patil et al. 2006). In a similar mouse model (Matsumoto, Han et al. 2006; Qu, Altomonte et al. 2006) enhanced lipogenesis and liversteatosis was observed. However inhibition of FoxO1 in the liver does not change glucokinase expression (Altomonte, Richter et al. 2003). It was evidenced that downstream of insulin signaling SREBP-1c, HNF4 and hypoxia-inducible factor 1 regulate glucokinase expression in addition to FoxO proteins (Roth, Curth et al. 2004). The disparity observed in accordance with insulin sensitivity and lipogenesis results from the feedback loop that increases insulin signalling thereby regulating lipid metabolism in a FoxO1-independent manner through SREBP-1c. Additionally these differences could be caused due to variation in experimental models, extent of overexpression, and the time of judgement on the subject of fasting and feeding (Gross, van den Heuvel et al. 2008). In addition FoxO1 plays an important insulin-dependent role in assembly and perseverance of very low-density lipoproteins (VLDL) in circulation. This effect is attained via transcriptional regulation of ApoCIII and microsomal triglyceride transfer protein (MTP) that plays a pivotal role in regulating the circulation of triglycerides during the periods of starvation (Kamagate, Qu et al. 2008). ApoCIII function as lipoprotein lipase (LPL) inhibitor the enzyme involved in hydrolysis of TG in VLDL and chylomicrons. FoxO1 increases the transcriptional activity and secretion of ApoCIII during fasting and thus extends the endurance of VLDL in circulation

20 Introduction

(Altomonte, Cong et al. 2004). Importantly during insulin resistance states, due to continous activation of FoxO1 results in hypertriglyceridemia and hyperglycemia (Kim, Zhang et al. 2011). Consistently elevated ApoCIII levels are linked with impaired hydrolysis and defective clearence of TG culminating in accumulation of TG and augmenting hypertriglyceridemia (Ito, Azrolan et al. 1990; Shachter 2001).

1.2.3 Role of FoxO transcription factors in pancreatic β-cells

Glucose homeostasis is dependent on genuine beta cell function and on pancreatic beta cell mass. Among the forkhead transcription factors FoxO1 is predominently expressed in the pancreas and regulates pancreatic and duodenal homeobox factor- 1 (Pdx-1) (Kitamura, Nakae et al. 2002). It was shown that haploinsufficiency of FoxO1 restores the insulin sensitivity in insulin resistant Irs2 knockout mice by increasing β-cell proliferation and restoring Pdx-1 expression (Nakae, Biggs et al. 2002). On the other hand gain-of-function of FoxO1 in the liver and pancreatic β-cells leads to diabetes owing to combined effect of elevated hepatic glucose production and altered β-cell compensation due to reduced Pdx-1 expression (Nakae, Biggs et al. 2002). FoxO1 by inhibiting FoxA2 decreases Pdx-1 expression and β-cell proliferation (Kitamura, Nakae et al. 2002), FoxA2 is another member of the fork head family function as positive regulator of Pdx-1 expression (Lee, Sund et al. 2002). During metabolic stress conditions FoxO1 attenuate the replication of β-cells and neogenesis but maintains the identity and function of β-cells. Conversely in Type 2 diabetes FoxO1 participates in diverse mechanisms including ER, oxidative, hypoxic stress, inflammation, raised apoptosis, inadequate proliferation, unconstrained autophagy and dedifferentiation results in β-cell dysfunction (Kitamura 2013). The significance of fine-tuning FoxO1 action in β-cells may provide therapeutic targets against type 2 diabetes (Buteau and Accili 2007; Kluth, Mirhashemi et al. 2011).

1.3 Liver is a " complex metabolic " organ

Multicellular organisms have developed systematic nutrient storage strategies in evolution. Controlled by hormones as well as energy sensors and nutrient, the metabolic system reacts to environmental conditions by storing or utilizing nutrients.

21 Introduction

This system optimizes metabolic balance, e.g. lack of metabolites in glucose metabolism will stimulate catabolic pathways like glycogenolysis or lipid β-oxidation (Sun, Miller et al. 2012). The liver is a central organ for metabolic regulation including glucose homeostasis regulating glucose transport, gluconeogenesis, glycolysis, glycogenesis, glycogenolysis and integrates both cell-nonautonomous and cell- autonomous mechanisms to command glucose release into the blood stream (Lin and Accili 2012). During the postprandial period the elevated levels of plasma glucose stimulate insulin secretion from pancreatic β-cells. Insulin, after binding to its receptors on the plasma membrane of liver, muscle, and adipose tissues activates glycogenesis and glycolysis in liver and muscle simultaneously inhibiting hepatic glycogenolysis and gluconeogenesis to avert hyperglycemia (Saltiel and Kahn 2001). Additionally, insulin stimulates the synthesis of triglycerides through de novo lipogenesis and prompting the mobilization of triglycerides from liver to adipose tissue. As a consequence blood glucose levels are reduced in the body by the action of insulin (Jitrapakdee 2012). During starvation, pancreatic α-cells secrete glucagon in response to low blood glucose levels (Jiang and Zhang 2003) which triggers fat mobilization from adipose tissue to skeletal muscle for β-oxidation. As a result many substrates are released which serve as precursor for hepatic gluconeogenesis. In short-term glycogenolysis and in long-term gluconeogenesis are the sources for glucose production. The glucose released serves as a fuel for and red blood cells / erthrocytes (RBCs) (Figure 4).

22 Introduction

Figure 4: Effects of Insulin and Glucagon during starvation and feeding

(Jitrapakdee 2012) A. During feeding conditions in response to glucose insulin released from pancreatic beta cells acts on its target tissues liver and muscles to promote glycolysis and glycogenesis. Insulin also promotes de novo lipogenesis and mobilization of triglycerides in liver and adipose tissue. B. In fasting glucagon released from pancreatic alpha cells stimulate fat mobilization and β -oxidation from adipose tissue to skeletal muscle to provide increased substrates for glycogenolysis and gluconeogenesis in the liver. The glucose released serves as a fuel for brain and RBCs. RBC, red blood cell; TG, triglycetides; FFA, free fatty acid; yellow hexagons, glucose; pink circles, insulin; white circles, triglycerides; green circles, glucagon.

Gluconeogenesis occurs predominantly in the liver and is regulated at the transcriptional level of key enzymes, like phosphoenolpyruvate carboxykinase (Pepck) (Hanson and Reshef 1997), glucose-6-phosphatase (G6pc) (Zhang, Patil et al. 2006), and pyruvate-dehydrogenase kinase 4 (Pdk4). Pdk4 inhibits pyruvate dehydrogenase by phosphorylation providing pyruvate as gluconeogenic substrates (Wei, Tao et al. 2011). Two other critical enzymes are pyruvate carboxylase (Pc) (Utter and Keech 1960) and fructose-1,6-biphosphatase (Fbpase) (O'Brien and Granner 1996). These enzymes are involved in the direct rapid regulation of hepatic glucose production.

23 Introduction

1.3.1 Regulation of hepatic glucose production

Glucose after intestinal absorption enters the hepatocytes (Figure 5) (Bechmann, Hannivoort et al. 2012). Glucose transporter 2 (Glut2) internalises glucose through the hepatocytes membrane. In hepatocytes, liver glucokinase (GCK) converts glucose to glucose-6-phosphate (Agius 2008). Mutations in GCK have been correlated with the pathogenesis of insulin resistance in diabetes (Miller, Anand et al. 1999; Cuesta-Munoz, Tuomi et al. 2010). Glucose-6-phophate is the most critical intermediate in the regulation of hepatic glucose metabolism, it is degraded during glycolysis to provide energy and the final product pyruvate is decraboxylated to acetyle-CoA which enters into intramitochondrial TCA cycle finally linking glucose and lipid metabolism as acetyle-CoA is the substrate for de novo lipogenesis (DNL). Furthermore, glucose-6-phosphate can be degraded in the pentose phosphate pathway shunt further providing the reduction equivalents (NADPH) for DNL.

Figure 5: Regulation on hepatic glucose metabolism (Bechmann, Hannivoort et

24 Introduction al. 2012). A. The insulin-independent glucose transporter 2 (GLUT-2) pump glucose over the membrane. Glucokinase regulatory protein (GCKR) induces a confirmational change to export glucokinase (GCK) in the cytoplasm. Glucokinase then phosphorylate glucose to glucose-6-phosphate (Glu-6-P). Nuclear receptor signalling and insulin transcriptionally regulate Glucokinase. B. The central intermediate in hepatic glucose metabolism is glucose-6-phosphate it is degraded in glycolysis to provide energy in the form of ATP and results in production of pyruvate, which enters into tricarboxylicacid (TCA) cycle. Glucose-6-phosphate is degraded in pentose phosphate pathway shunt to provide NADPH a substrate for de novo lipogenesis (DNL). Acetyl-CoA also links glucose and lipid metabolism and serve as a substrate for DNL.

1.3.2 Regulation of transcription factors controlling gluconeogenesis

In response to hypoglycemic conditions during fasting pancreatic α-cells secrete glucagon. Activation of glucagon signalling increases cAMP levels, which stimulate protein kinase A (PKA) activity, and eventually promotes CREB phosphorylation (Authier and Desbuquois 2008). CREB is critical regulator of the hepatic gluconeogenic programme in mice (Koo, Flechner et al. 2005). CREB-regulated transcriptional coactivator 2 (CRTC2 also reffered to as TORC2) (Shaw, Lamia et al. 2005) enters into the nucleus to interact with phospho-CREB and binds to cAMP- response element (CRE) to regulate the expression of gluconeogenic target genes like phosphoenolpyruvatecarboxykinase (PEPCK) and glucose-6-phosphatase (G6Pase) (Mayr and Montminy 2001). During this period another coactivator, the NAD-dependent deacetylase sirtuin-1 (SIRT1) is activated, which deacetylates peroxisome proliferator activated receptor γ-1α (PGC-1α) (Nemoto, Fergusson et al. 2005) (Brunet, Sweeney et al. 2004; Rodgers and Puigserver 2007) and FoxO transcription factors (Frescas, Valenti et al. 2005). PGC-1α interacts with FoxO1 and with hepatic nuclear factor-4 to regulate the expression of gluconeogenic target genes (Li, Monks et al. 2007). Additionaly, increased glucocorticoid concentrations during fasting allows binding of glucocorticoid receptors (GR) to their respective binding elements (GRE) in the nucleus to regulate the transcription of gluconeogenic target genes (Schoneveld, Gaemers et al. 2004). On the other hand during feeding states, insulin signalling gets activated and results in phosphorylation of FoxO and CRTC2 hence transcriptional inactivation and cytoplasmic exclusion (Brunet 2004). High levels of insulin also limit SIRT1 activity, which results in PGC- 1α transcriptional inactivation. Transcriptional activities of several hepatic nuclear

25 Introduction factors (HNFs) are modulated by interaction with the corepressor “small heterodimer partner (SHP)” (Yamagata, Daitoku et al. 2004) and DAX1 the dosage-sensitive sex reversal adrenal hypoplasia congenital critical region on the X gene 1 (DAX-1). Both DAX-1 and SHP function as corepressors and expressed in the liver in response to fed conditions (Nedumaran, Hong et al. 2009). All these events lead to transcriptional inactivation and attenuation of hepatic glucose production during feeding conditions (Figure 6).

26 Introduction

Figure 6: Hepatic transcriptional modulation of gluconeogenic enzymes G6Pase and PEPCK during fasting (A) and Feeding (B) conditions (Jitrapakdee 2012)

A. During fasting conditions due to low glucose level pancreatic α-cells release glucagon and glucocorticoids. The resulting activation of glucagon signalling signals several transcription factors involved in the regulation of gluconeogesis such as PGC-1a, CREB, CRTC2, FOXO1 to move into the nucleus and bind to their respective promoters to regulate gluconeogenic enzymes such as glucose-6 -phosphatase (G6Pase) and phosphoenolpyruvatecarboxykinase (PEPCK). B. However, during feeding conditions insulin released from pancreatic β-cells in response to high glucose levels results in activation of insulin signalling that inactivate and cause nuclear exclusion of transcription factors such as FOXO1 and CRTC2 critical for gluconeogenesis. In addition certain other transcription factors important for glconeogenesis such as hepatic nuclear factors (HNFs) and PGC-1α are modulated by corepressors such as SHP and DAX-1 resulting in overall inhibition of gluconeogenesis during feeding conditions.

1.3.3 Regulation of hepatic Lipid metabolism

Fattyacids are metabolized with in the hepatocytes to yield energy serving as a feul during the process of starvation. Under normal physiologic conditions free fatty acids (FFA) derived from lipid breakdown in the adipose tissue are taken up by fatty acid transporters (FATP2 and FATP5) into the hepatocytes under the control of insulin 27 Introduction signalling and nuclear receptor (Figure 7). Fatty acids are oxidized intramitochondrially thereby generating acetyl-CoA that enters into the tricarboxylic acid cycle (TCA). Triglycerides (TAGs) synthesized from denovo lipogenesis are stored in the form of lipid droplets (LP) or packaged into very low-density lipoprotein (VLDL) and thus released into blood stream. Depending upon the nutritional status fatty acids are again released from the lipid droplet by the process of autophagy to provode energy. (Shibata, Yoshimura et al. 2009; Singh, Kaushik et al. 2009). An impaired hepatic glucose and lipid metabolism leads to the development of type2 diabetes (Bouche, Serdy et al. 2004).

Figure 7: Hepatic lipid metabolism (Bechmann, Hannivoort et al. 2012)

Under normal physiological conditions fatty acids enter into the hepatocytes and oxidized intramitochondrially to release energy and acetyl-CoA that enter into TCA cycle. Triglycerides synthesized are either stored as lipid droplets LP or package into very low-density lipoprotein VLDL. The LD also serves as a source of feul during caloric restriction and is released by the process of autophagy. 28 Introduction

1.4 Diabetes mellitus

Diabetes mellitus is represented by high blood glucose levels (Table 3) as a result of insufficient insulin action for the body´s need. It is a major healthcare problem because it increases the risk for several other diseases e.g cardiovascular disease, renal failure, and peripheral neuropathy. As a consequence it places a severe economic burden on healthcare systems and governments. It is a heterogeneous disorder with insulin resistance and pancreatic β-cell dysfuntion. (Nolan, Damm et al. 2011; Ashcroft and Rorsman 2012).

Table 3: Diagnostic criteria for diabetes, Imapired fasting glucose (IFG), Impaired glucose tolerance (IGT) (Nolan, Damm et al. 2011)

Diabetes IFG and IGT Prediabetes

(WHO and ADA) (WHO) (ADA)

HbA 1c ≥6.5% NA ≥ 5.7% and < 6.5% Fasting plasma ≥ 7.0 (≥ 126 mg/dl) IFG ≥ 6.1 and < 7.0 ≥ 5.6 and < 7.0 glucose (mmol/L) 75g OGTT post- 2h,≥11.1(≥ 200 mg/dl) IGT 2h ≥ 7.8 and < 11.1 2h ≥ 7.8 and < 11.1 load plasma glucose (mmol/L) Random glucose ≥11.1 (≥ 200 mg/dl) with NA NA (mmol/L) classic symptoms

Although the diagnostic criteria of World health organization (WHO) and American diabetes association (ADA) for diabetes are identical, the measures to assess intermediate hyperglycemia or prediabetes are different. IFG: impaired fasting glucose, IGT: impaired glucose tolerance, HbA1c: glycated haemoglobin A1c. NA: not applicable, OGTT: oral glucose tolerance testing.

1.4.1 Type 2 Diabetes mellitus

Type 2 diabetes mellitus is a complex disorder characterized by hyperglycemia, and altered lipid metabolism caused by β-cell dysfunction (mainly impaired insulin 29 Introduction secretion) and inactivity, overnutrition, obesity and peripheral insulin resistance (Nolan, Damm et al. 2011; Ashcroft and Rorsman 2012). Type 2 diabetes accounts for almost 90% of diabetes cases worldwide. The prevelance of diabetes is higher in middle-income and upper-middle countries. When people become richer they often eat more fat-rich and sugar-rich diet and are less physically active. The International Diabetes Federation in 2011 announced that 336 million people across the world are suffering from type 2 diabetes and the disease leads to 4.6 million deaths each year. The highest death rate (more than one million) occurs in China and less than a million are in India. The highest prevelance of diabetes occurs in Middle East countries due to their rapid economic development with Qatar having (23%) the highest prevelance in the world followed by (19%) in United arab emirates (UAE), Egypt and Saudi Arabia (Scully 2012) (Ashcroft and Rorsman 2012). Hyperglycemia is an increase of blood sugar levels it occurs as a result of insufficient insulin production or enough insulin is being produced from pancreatic β-cells but the body is not making an effective use of insulin followed by insulin resistance and type 2 diabetes. Under insulin resistance conditions insulin production by pancreatic β- cells increases to meet the high demand resulting in an increase of β-cell mass and compensatory hyperinsulinemia. The major contributors of insulin resistance and impaired glucose tolerance are obesity and overweight. Abormalities in hormones such as impaired insulin secretion, impaired glucagon secretion, reduced secretion of the incretin glucagon-like-peptide1 (GLP-1), reduced glucose uptake by skeletal muscle and adipose tissue, and increased concentration of other counter regulatory hormones promotes insulin resistance and hyperglycemia in type 2 diabetes (Tahrani, Bailey et al. 2011) (Figure 8).

30 Introduction

Figure 8: Pathogenesis of hyperglycemia in type 2 diabetes (Tahrani, Bailey et al. 2011) Impaired action of multiorgan system resulting in hyperglycemia in type2 diabetes. Impaired secretion of hormones e.g reduced secretion of incretin hormones glucagon like peptide1 (GLP-1), increased glucagon secretion hyperglucagonemia, impaired insulin secretion, uncontrolled hepatic glucose production by the liver, Impairment of neurotransmitter function in the brain, decreased uptake of glucose in skeletal muscle, abnormalities in the reabsorption of glucose in the kidney, elevated lipolysis and impaired glucose uptake by adipose tissue, all these abnormalities culminates in hyperglycemia in type 2 diabetes.

1.4.2 Obesity and type2 diabetes

The catastrophic increase in T2D over the last 50 years is due to the concomitant increase in Obesity. The major risk factor in type 2 diabetes is obesity as it reduces insulin sensitivity in peripheral tissues by directly affecting the β-cell function (Donath, Dalmas et al. 2013). From the very first studies it was shown that increased adipose tissue, in the upper body is correlated with diabetes (Vague 1996). In obesity primarily muscle and liver insulin resistance are involved due to enhanced adipose tissue derived free fatty acids stimulating triglyceride accumulation in liver and muscle (Perseghin, Petersen et al. 2003). Interestingly most obese people with insulin-resistance states do not develop type 2 diabetes because their β-cell are able

31 Introduction to compensate for increase demand of insulin secretion and this is influenced by their strong genetic constitution (Bergman 2005; Ashcroft and Rorsman 2012). Many healthy overnourished obese individuals do not develop diabetes they safely deposit the excess calories to subcutaneous adipose tissue (SAT) rather then to skeletal muscle, liver, heart and islet β-cell as a result the key organs of the body are not effected due to following processes including compensation of islet β-cell, development of minimal insulin resistance and less increase in liver fat (Wajchenberg 2000; Stefan, Kantartzis et al. 2008) (Figure 9). Conversely susceptible overnourished people develop diverse metabolic defects: inability of β-cell to compensate for fuel surfeit, hyperadiponectinemia, adipose tissue inflammation, impaired hepatic glucose production and devlopment of peripheral insulin resistance (Kahn 2003; Defronzo 2009). A strong interplay between genetics, diabetes and obesity is evidenced by a report showing that triglyceride emulsion in a non-diabetic people who had first-degree relatives with T2D showed decrease in insulin secretion but not in others (Kashyap, Belfort et al. 2003). This showed that genetic differences become obvious only in the presence of elevated fat.

32 Introduction

Figure 9: Pathways and related complications of type 2 diabetes (Nolan, Damm et al. 2011) .

Additionally several mechanisms such as: tissue inflammation, lipotoxicity (caused by ectopic lipid accumulation or elevated free fatty acids (FFA), glucotoxicity (caused by sustained hyperglycemia), oxidative stress (caused by FFA or glucose) and endoplasmic reticulum stress (ER) have been shown to impair insulin secretion, induce pancreatic β-cell dysfunction and insulin resistance in type 2 Diabetes (Wilcox 2005) (Donath and Shoelson 2011; Samuel and Shulman 2012).

33 Introduction

1.4.3 Factors controlling β-cell mass and T2D

Several studies points to the reduction of β-cell mass in T2D although there is a continuous ongoing debate about this controversy and it is also unclear whether T2D occurs as a result of reduced β-cell mass or it developed as a consequence of sustained hyperglycemia (Ashcroft and Rorsman 2012). It was also evidenced that reduced number of β-cell rather than impaired β-cell function are the major contributors of T2D (Butler, Janson et al. 2003). β-cell mass is also influenced by long term insulin secretion and certain growth hormones prolactin, placental lactogen and GLP1 increases β-cell proliferation in addition to glucose stimulated insulin secretion and insulin gene expression (Nielsen, Galsgaard et al. 2001; MacDonald, El-Kholy et al. 2002). Importantly on the other hand β-cell hyperplasia in insulin resistance states driven by liver-derived systemic factors has also been reported (El Ouaamari, Kawamori et al. 2013). Furthermore the role of betatrophin hormone, which is primarily expressed in liver and fat induces a dramatic increase in β-cell proliferation, β-cell expansion and improves glycemic control in mouse models of insulin resistance (Yi, Park et al. 2013). It is uncertain that how much β-cell mass is needed for the proper glycemic control nevertheless it was suggested that 40% of β- cell mass is sufficient for normal glucose homeostasis and it is supported by the notion that removal of almost half of pancreas has only a small effect on glucose tolerance (Menge, Tannapfel et al. 2008). This suggests that although β-cell mass plays some important role in T2D but this is not the only, or even the most pivotal, factor in T2D (Ashcroft and Rorsman 2012).

1.4.4 Molecular mechanism of Insulin resistance

Disruption of insulin signaling, at different levels of insulin signalling pathways are the major characteristic of insulin resistance states (Kwon and Pessin 2013). Insulin resistance at the molecular level can be acquired by either decreased or increased degradation or decreased transcriptional activities of critical components of insulin signalling cascade e.g IRS1, IRS2 (Bar, Harrison et al. 1979; Shimomura, Matsuda et al. 2000; Zhande, Mitchell et al. 2002). It was reported previously that increased stimulation of SREBP1-c could inhibit transcription of IRS-2 levels (Ide, Shimano et

34 Introduction al. 2004) suggesting that interaction between signalling components regulate one another. Serine phosphorylation of insulin receptor and IRS proteins by ERK, PKB and the oxidative stress kinases JNK and IKKβ can impair their tyrosine kinase (Liu, Paz et al. 2001) (Gao, Hwang et al. 2002; Hirosumi, Tuncman et al. 2002) activity and futher association with 14-3-3 proteins cause the removal from the insulin signalling pathways (Craparo, Freund et al. 1997). Several inhibitory proteins can interact with components of insulin signalling cascade to produce insulin resistance such as several inflammatory induce the suppressors of cytokine signalling proteins (SOCS) which binds to insulin receptor and attenuate insulin signalling pathway (Emanuelli, Peraldi et al. 2000; Rui, Yuan et al. 2002) (Figure 10). Additionally insulin resistance can be due to increase in amount and activities of phosphotyrosine phosphatases enzymes that normally reverse insulin action e.g PIP3 phosphatases, PTEN and SHIP (Nakashima, Sharma et al. 2000; Clement, Krause et al. 2001). Finally the phenotype of insulin resistance depends on the exact components of affected insulin signalling pathway and the exact tissues in which the components of insulin signalling are affected (Biddinger and Kahn 2006).

Figure 10: Molecular events that can result in insulin resistance (Biddinger and Kahn 2006)

35 Introduction

1.4.5 Genetics of Type 2 Diabetes mellitus (T2DM)

Genome wide association (GWAS) and linkage studies have discovered more than 60 genes correlated with increased risk of type2 diabetes mellitus (Morris, Voight et al. 2012) (Bonnefond, Froguel et al. 2010; McCarthy 2010; Taneera, Lang et al. 2012). Several studies have shown that many of these genes are important for β-cell function, β-cell development, regulation of β-cell mass and some are associated with hepatic glucose production, insulin resistance and predisposition to obesity such as FTO (fat mass and obesity-related gene). Although it is uncertain how these genes predispose to full blown diabetes (Bonnefond, Froguel et al. 2010; McCarthy 2010). The important diabetes suscebtible gene identified is TCF7L2 a transcription factor (T-cell factor 7 like-2) involved in WNT signalling. It has been shown that the risk allele for TCF7L2 was associated with altered insulin secretion and improved hepatic glucose production rate (Lyssenko, Lupi et al. 2007; McCarthy, Rorsman et al. 2013). Other genes involved in impaired β-cell function include (KCNJ11, TCF7L2, WFS1, HNF1B, SLC30A8, CDKAL1, IGF2BP2, CDKN2A, CDKN2B, NOTCH2, CAMK1D, THADA, KCNQ1, MTNR1B, GCKR, GCK, PROX1, SLC2A2, G6PC2, GLIS3, ADRA2A, and GIPR) and some important genes involved in impaired insulin senstivity include (PPARG, IRS1, IGF1, FTO, and KLF14) (Nolan, Damm et al. 2011) (Figure 11).

36 Introduction

Figure 11: Genetic risk factors associated with type 2 diabetes (McCarthy 2010)

1.4.6 Animal models of type 2 diabetes

Several invivo genetic models are greatly useful as they provide new awareness into the functions of human diseases. Obesity, insulin resistance and type 2 diabetes can be induced by various diet/nutritional, chemical, surgical agents and knockout or transgenic approaches (Table 4) (Chatzigeorgiou, Halapas et al. 2009) . The most characteristic examples of T2DM model include db/db, ob/ob and Zucker fatty rat (ZFR). The diabetic models db/db and ob/ob mice develop obesity due to autosomal recessive mutations in leptin and leptin receptors genes that finally leads to diabetes. While Zucker fatty rats (ZFR) phenotype is associated with hypothalamic defect in leptin receptor signalling, and particular inbreeding of zucker fatty rat for hyperglycemia gave birth to zucker diabetic rats (ZDF). Of these models the ob/ob mouse and zucker fatty rat (ZFR) is characterized by mild to moderate hyperglycemia, hyperinsulinemia, hyperphagia, obesity and insulin resistance. These mice do not develop diabetes because of compensatory action of `sturdy´ pancreatic β-cells capable of maintaining lifelong insulin secreting capacity (Durham and Truett 2006; Srinivasan and Ramarao 2007; Chatzigeorgiou, Halapas et al. 2009). On the other hand zucker diabetic fatty rats (ZDF) and db/db mouse have pancreatic β-cells

37 Introduction which are `labile or brittle´ enabling only for a short-term hyperinsulin secretion for obesity. Eventually as a result of other environmental and genetic predisposition factors it induces pressure on pancreatic β-cells for insulin secretion and which ultimaltely leads to apoptosis, degranulation and overt hyperglycemia. At this stage, the animals lose their accumulated fat, suffer from ketosis and require lifelong insulin therapy. This ZDF rat and db/db mouse resemble nearly to type 2 diabetes of humans (Srinivasan and Ramarao 2007).

Table 4: Animal models of type 2 diabetes (T2D) (Srinivasan and Ramarao 2007)

Model category Type 2 diabetic models Obese Non obese

Spontaneous or genetically ob/ob mouse Cohen diabetic rat derived diabetic animals db/db mouse GK rat KK mouse Torri rat Non obese C57BL/6 KK/Ay mouse (Akita) mutant mouse NZO mouse NONcNZO10 mouse TSOD mouse M16 mouse ALS/Lt mouse

ZDF rat SHR/N-cp rat JCR/LA-cp rat OLETF rat Obese rhesus monkey

II. Diet/nutrition induced diabetic Sand rat animals C57/BL 6J mouse Spiny mouse

III. Chemically induced diabetic GTG treated obese mice animals

IV. surgical diabetic animals VMH lesioned dietary obese diabetic rat

38 Introduction

V. Transgenic/knock-out β3 receptor knockout mouse Low dose ALX or STZ adult diabetic animals Uncoupling protein (UCP1) rats, mice, etc. knock-out mouse Neonatal STZ rat

Partial pancreatectomized animals e.g. dog, primate, pig & rats

Transgenic or knockout mice involving genes of insulin and insulin receptor and its components of downstream insulin signalling e.g.

IRS-1, IRS-2, GLUT-4, PTP-1B and others PPAR-C tissue specific knockout Mouse

Glucokinase or GLUT 2 gene knockout mice

Human islet amyloid

1.5 Metformin

1.5.1 Metformin as an anti-hyperglycemic agent

The most widely recommended drug to treat hyperglycemia in patients with type 2 diabetes is Metformin (1,1-dimethylbiguanide), a derivative of biguanide. According to the guidelines of American Diabetes Association and European Association of the Study of Diabetes, Metformin is used as the first line of oral therapy (Adler, Shaw et al. 2009; Nathan, Buse et al. 2009). Additionally Metformin is also used for treating several other diseases e.g as cancer preventive drug (Marchesini, Brizi et al. 2001; Evans, Donnelly et al. 2005; Ben Sahra, Le Marchand-Brustel et al. 2010), in cases with fatty liver diseases (Nair, Diehl et al. 2004), in pregnancy during gestational diabetes (Rowan, Hague et al. 2008; Balani, Hyer et al. 2009), and to reduce macrovascular and microvascular complication in T2D (Hurst and Lee 2003). In cases with type 2 diabetes metformin is recently being prescribed to at least 120 million people worldwide and the antihyperglycemic effect of metformin works by lowering the blood glucose concentrations without causing overt hypoglycemia (Viollet, Guigas et al. 2012). It was shown that metformin is the best insulin sensitizer in cases of insulin resistance where metformin reduces the fasting plasma insulin

39 Introduction levels. Metformin improves insulin sensitivity through positive effects on activation of insulin signalling by increased tyrosine kinase activity and increased insulin receptor expression (Gunton, Delhanty et al. 2003). It has been shown from animal models and clinical studies that metformin reduces hepatic glucose production principally by inhibiting hepatic gluconeogenesis (Cusi, Consoli et al. 1996; Hundal, Krssak et al. 2000; Natali and Ferrannini 2006). Metformin-induced attenuation of hepatic glucose production is mediated by changes in enzymatic activities and decreased uptake of gluconeogenic substrates to liver. It is shown in high-fat diet insulin resistant rats that metformin lowers hepatic gluconeoenesis by inhibiting the activity of glucose-6- phosphatase (G6pc) (Argaud, Roth et al. 1993; Radziuk, Zhang et al. 1997; Mithieux, Guignot et al. 2002).

1.5.2 Mechanism of Metformin action

The exact and precise mechanism of hepatic metformin action is a matter of debate and still unknown. The explanation of such a mechanism is pivotal issue as it may unveil new targets for the treatment of type 2 diabetes. Metformin at the physiological pH exist as organic cation acts as a hydrophilic base and can only pass through the plasma membrane by passive diffusion (Detaille, Guigas et al. 2002; Viollet, Guigas et al. 2012). The cationic transporters, which are involved in the intracellular internalization of metformin, are (OCTs). OCT1 transport metformin to the liver and OCT2 transports metformin to the kidney (Graham, Punt et al. ; Shu, Sheardown et al. 2007). Metformin is mostly acculmulated in the liver than in any other tissues (Wilcock and Bailey 1994). The molecular targets and mechanism of Metformin action was difficult to achieve for several years. It has been shown that multiple actions of Metformin are associated with activation of AMPK (AMP-activated protein kinase) (Zhou, Myers et al. 2001). AMPK is serine/threonine protein kinase, which is phylogenetically conserved among eukaryotes. Under stress condition activation of AMPK protect the cellular functions. AMPK exists as heterodimeric complexes comprising a catalytic-α subunit, two regulatory subunits β and γ (Corton, Gillespie et al. 1994; Viollet, Guigas et al. 2012) (Hardie, Scott et al. 2003; Kahn, Alquier et al. 2005). AMPK is activated under cellular stress conditions, an increased cellular AMP/ATP ratio or when ATP is depleted (Hardie, Scott et al. 2003; Carling 2004; Hardie 2004; Viollet, Guigas et al. 2012). AMP promotes the AMPK activation, by

40 Introduction stimulating phosphorylation at a critical threonine residue (Thr172) of the kinase domain at catalytic-α subunit. AMPK is activated by upstream kinases such as serine/threonine kinase 11 (STK11/LKB1) identified as tumor suppressor (Viollet, Guigas et al. 2009). When AMPK is activated it turn off the anabolic pathways and triggers the catabolic pathways as a result production of glucose, lipid synthesis and protein synthesis are inhibited and utilization of glucose and oxidation of fatty acid get started.

Of notice it was shown that metformin does not directly activate AMPK, the primary target for metformin are the mitochondria where metformin blocks mitochondrial respiratory chain at the complex 1, resulting in lower oxidation of NADH, decreased pumping of protons across the inner mitochondrial membrane leading to lowering of proton gradient that results in reduced proton-driven synthesis of ATP (Viollet, Guigas et al. 2012) (Figure 10). Inhibitory effect of metformin through blocking mitochondrial respiratory chain complex 1 was shown in isolated hepatocytes and perfused livers from rodents (El-Mir, Nogueira et al. 2000; Owen, Doran et al. 2000). It was also shown in human primary hepatocytes that metformin firstly targets complex 1 of mitochondrial respiratory chain resulting in decreased cellular energy status that activates AMPK (Stephenne, Foretz et al. 2011).

41 Introduction

Figure 12: Metformin target mitochondrial respiratory-chain complex 1 (Viollet,

Guigas et al. 2012). Metformin has a positive charge and can pass across the cell membrane by passive diffusion and internalized with in the hepatocytes with the help of OCT1 transporters in the liver. Once the drug is with in the liver mitochondria are the primary targets. Metformin selectively inhibits the complex-1 of the respiratory chain without affecting the other complexes this causes a decrease in oxidation of NADH, consumption of oxygen rate and resulting in decreasing proton gradient across the mitochondrial membranes and subsequently it results in lowering proton-driven synthesis of ATP.

Importance of AMPK activation in antidiabetic effect by metformin was firstly supported from the study showing that mice lacking hepatic LKB1 were unable to lower the blood glucose and this also showed that metformin require LKB1 in the liver to reduce the levels of blood glucose (Shaw, Lamia et al. 2005). Importantly it was evidenced that LKB1/AMPK signalling regulates phosphorylation and cytoplasmic translocation of CREB-regulated transcriptional coactivator 2 (CRTC2 also reffered to as TORC2) (Shaw, Lamia et al. 2005). CREB is critical regulator of hepatic gluconeogenic programme in mice (Koo, Flechner et al. 2005). TORC2 interact with CREB to regulate the expression of PGC-1α and gluconeogenic target genes phosphoenolpyruvatecarboxykinase (PEPCK) and glucose-6-phosphatase (G6Pase).

42 Introduction

AMPK or AMPK related kinases are involved in the phosphorylation of CRTC2 that results in degradation of complex and hence inhibiting the gluconeogenic programme (Koo, Flechner et al. 2005; Shaw, Lamia et al. 2005) (Figure 11). Furthermore metformin induced activation of AMPK results in the upregulation of an orphan nuclear receptor small heterodimer partner (SHP), which works as a transcriptional repressor (Kim, Park et al. 2008). SHP represses the transcriptional activities of FoxO1, hepatocyte nuclear factor 4-α and a number of nuclear receptors that are involved in regulation of gluconeogenesis (Lu, Makishima et al. 2000; Lee and Moore 2002; Kim, Kim et al. 2004; Yamagata, Daitoku et al. 2004). In addition SHP inhibits hepatic gluconeogenic gene expression through interaction with CREB- CBP complex and CRTC2 (Lee, Seo et al. 2010). Furthermore metformin induced inhibition of gluconeogenesis is also supported by the role of Krüpple-like factor 15 (KLF15). Krüpple-like factor 15 is highly expressed in the liver and plays an important role in the regulation of gluconeogenesis in the liver. Previously it was shown that KLF15 is up-regulated in diabetic mice with increased expression of PEPCK gene in the liver (Teshigawara, Ogawa et al. 2005; Gray, Wang et al. 2007). Metformin suppresses the KLF15 gene expression and induces proteosomal degradation by promoting its ubiquitination (Takashima, Ogawa et al. 2010). Interestingly it was further showed by recent reports stating that AMPK and LKB1 activities are dispensable for inhibition of gluconeogenesis by metformin (Foretz, Hebrard et al. 2010). Indeed metformin inhibits the process of gluconeogenesis independent of AMPK activation through decrease in hepatic energy status (Miller and Birnbaum 2010). This decrease in energy status by blocking mitochondrial respiratory chain complex 1 leads to acute inhibition of gluconeogenesis through of key enzymes such fructose-1, 6-biphosphatase independent of gluconeogenic gene expression (Miller and Birnbaum 2010). Metformin also play a beneficial role in improving the disorders of lipid metabolism. It was shown previously in leptin deficient ob/ob mice that metformin improves fatty liver disease (Lin, Yang et al. 2000; Cool, Zinker et al. 2006). Metformin has also been proved useful in patients suffering from alcoholic steatohepatitis (Marchesini, Brizi et al. 2001; Nair, Diehl et al. 2004; Raso, Esposito et al. 2009). Metformin- induced activation of AMPK inactivates acetyle CoA carboxylase (ACC) by phosphorylation and activates fatty acid oxidation. ACC is the rate-limiting enzyme for the production of malonyl-CoA, which is important precursor for the synthesis of fatty

43 Introduction acids and an important inhibitor for mitochondrial fatty acid oxidation (Zhou, Myers et al. 2001). In addition metformin-induced activation of AMPK inhibits fatty acid synthesis through regulating sterol regulatory element-binding protein-1c (SREBP1) gene expression and thereby inhibiting lipogenesis (Li, Xu et al. 2011). In summary metformin plays an important role in inhibition of gluconeogenic gene programme via both AMPK-dependent and AMPK-independent mechanism through controlling several critical regulatory points (Figure 11) (Viollet, Guigas et al. 2012).

Figure 13: Mechanism of metformin action in controlling hepatic gluconeogenesis (Viollet, Guigas et al. 2012). Metformin attenuate the process of gluconeogenesis either AMPK-dependent or AMPK-independent signalling pathways through controlling several critical regulatory points.

44 Introduction

1.6 Aims of study

The interplay between different FoxO family members sharing the conserved DNA binding domain is far from being understood and this may account for the paradoxical findings described above (Biddinger and Kahn 2006). Importantly, in addition to FoxO1, FoxO3 is also significanly expressed in the liver, its contribution to metabolic regulation has however not been thoroughly analysed. Interestingly FoxO3 genotypes are associated with insulin sensitivity phenotypes and longevity in humans suggesting that FoxO3 is critical for metabolic control (Willcox, Donlon et al. 2008; Flachsbart, Caliebe et al. 2009; Pawlikowska, Hu et al. 2009). The relative contribution of individual FoxO proteins has not been addressed the reason behind this might be experimental approaches (most of which are loss of function studies), cell type specific functions and the context-and kinetic-specific stimulation, which might affect the biological outcome of the FoxO signalling system. Therefore a refined study using a conditional gain-of-functional model should be capable to unveil the functions of FoxO3 in the liver under physiological conditions. This would also help to estimate the usefulness of FoxO3 inhibitors in diabetes therapy. To achieve this goal, firstly this study was designed to establish a mouse model that allows for conditional regulation, activation of FoxO3 activity specifically in liver-cells. Further, The morphological, functional, and biochemical consequences of transgene expression should be investigated. Specifically the following questions will be addressed in detail:

1. Does liver-specific expression of constitutively active FoxO3 (LAP-tTA x (tetO)7- FOXO3CA) interfere with metabolic liver functions?

2. What are the consequences of expression of FoxO3CA in regulation of hepatic glucose and lipid homeostasis?

45 Material and Methods

2 Materials an Methods

2.1 Mice

We crossed mice expressing the tetracycline-responsive transactivator (tTA) under the control of the rat liver activator protein (LAP) promoter (Sunami, Leithauser et al. 2012) with mice bearing a constitutively active human FOXO3 allele (FOXO3CA) (Schmidt-Strassburger, Schips et al. 2011; Schips, Wietelmann et al. 2011) under the control of a tTA-regulated promoter. Studies were performed on a C57BL/6xNMRI mixed background. FOXO3CAHep mice contain a bidirectional promoter, which mediates simultaneous expression of FOXO3 and luciferase that was used for in vivo imaging analysis and in vitro luciferase assays as described (Schmidt-Strassburger, Schips et al. 2011; Schips, Wietelmann et al. 2011; Sunami, Leithauser et al. 2012). Mice were kept in a specific pathogen-free (SPF) animal facility at University of Ulm. Experiments were approved by the state of Baden-Württemberg in Germany.

2.1.1 Doxycycline administration

To avoid embryonic activation of FOXO3CA, all mice received 0.1g/L doxycycline in 1% sucrose in drinking water during pregnancy untill mice were born. Afterwards, the doxycycline (DOX) (MP Biomedicals Eschwege) was withdrawn to activate the FOXO3CA expression and the mice were analysed at 4-weeks of age. To avoid the developmental defects and lethality at the earlier time point of FOXO3CA activation the animals received Doxycycline 0.1g/L Doxycycline in 1% sucrose in drinking water untill the newborn animals reached 5-weeks of age and the Doxycycline was withdrawn after 5-weeks to activate the expression of FOXO3CA and the animals were analysed at the age of 7-weeks or at 9-weeks.

2.1.2 Metformin administration

Metformin (1,1-Dimetylbiguanide hydrochloride, D150959, Sigma-Aldrich) was orally administered in drinking water (1.65 g/l). Schedule of metformin administration is shown in the (Figure 34). For the analysis at the age of 7-weeks, metformin administration was performed for the last 3 days untill the animals become 7-weeks.

46 Material and Methods

For the analysis at the age of 9-weeks metformin was administered for three weeks untill the animals become 9-weeks of age. For each experiment mice were fasted for 6 hours prior to the analysis. During 6 hours fasting period mice drink the normal drinking water.

2.2 Genotyping

2.2.1 Isolation of genomic DNA

Extraction of genomic DNA was made from the tail tips of mice. Tails were incubated over night at 56°C for digestion with the DNA lysis buffer (733 µl TNES-buffer with 17 µl proteinase K (20 µg/ µl)). Afterwards, 250µl of 6M NaCl was used and centrifugation was done at 13000 rpm for 10 min at room temperature. The supernatent was transferred to a fresh eppendorf tube. Genomic DNA was precipitated by supplying equal volume of isopropanol to the pellet followed by centrifugation at 13000 rpm for 10 min. DNA pellet was washed with 500 µl of 70% ethanol, dried for 10 min. Dried DNA was dissolved in 250 µl sterile ampuwa water.

2.2.2 Polymerase chain reaction (PCR)

The transgenes tTA and FOXO3CA were detected by PCR with specific pair of primers and applying the following recipe, Reaction mix: DNA sample: 2 µl dNTPs 2mM: 3 µl 5x GoTaq buffer: 6 µl primers 10 µM: 1µl water: 16.8 µl Taq 5000 U/ml: 0.2 µl Primers sequences: tTA 5´ gctaggtgtagagcagcctacattg tTA 3´ gtccagatcgaaatcgtctagcgcg FOXO3-P17 catggagtctgcggccgtctcc FOXO3-P48 ggtacccggggatcctctagtcag

47 Material and Methods

PCR programmes

PCR LAP.tTA (tetO)7.FOXO3CA

Start 94°C /3 min 94°C /3 min

Cycles 33 cycles 33 cycles

Annealing 94°C /45 sec 94°C /45 sec Synthesis 62°C /45 sec 62°C/ 45 sec Denaturation 72°C /1 min 20 sec 72°C /1 min 20 sec

End 72°C 2 min 72°C 2 min

2.3 In Vivo Bioluminescence imaging

To determine liver-specific transgene expression, mice were anesthetized by continous inhalation of 2% vol/vol isoflurane.and injected intraperitoneally with 25 mM Luciferin (Synchem OHG) (150µg/g body weight). Bioluminescence was monitored 1 minute after injection by the IVIS imaging system Xenogen IVIS 200 System (Caliper Life Sciences, Hopkinton, MA, USA) as previously described (Schmidt-Strassburger, Schips et al. 2011; Schips, Wietelmann et al. 2011; Sunami, Leithauser et al. 2012).

2.4 Metabolic studies

2.4.1 Glucose Tolerance Test (GTT) and Insulin Tolerance Test (ITT)

Glucose Tolerance Test (GTT) and Insulin Tolerance Test (ITT) were performed in 7- week-old animals. For GTT, animals were fasted for 6 hours during fasting animals were provided with drinking water, After 6 hour fasting the blood glucose was measured from the tail vein and then injected with glucose (0.75 g / kg body weight) intraperitoneally (i.p.). Tail vein blood was collected and blood glucose values were measured with glucometer (One Touch Ultra II, Life Scan) at the indicated time points 48 Material and Methods after every 30 minutes for two hours. For Insulin tolerance test (ITT), the animals were not fasted they were in the random- fed state and then injected intraperitoneally (i.p.) with insulin (0.75 U / kg body weight, Actrapid Penfill Novo Nordisk). Blood glucose levels were measured from tail vein at the indicated time points after every 15 minutes for 1 hour.

2.4.2 Animal studies

Mice were sacrificed via CO2 inhalation, and blood was collected from vena cava inferior. After brief centrifugation, serum was collected and used for measurement of alanine and aspartate aminotranferase levels (ALT and AST) as well as bilirubin, triglycerides, cholesterol, alkaline phosphatase (ALP), were measured by (Reflotron,Roche). Albumin, pseudocholinesterase, urea, and creatinine levels were measured by (Technical University of Munich). Blood glucose measurement was performed immediately after a collection of tail vein blood with glucometer (One Touch Ultra II, Life Scan).

2.4.3 Enzyme-linked immunosorbent assay (ELISA)

All the ELISA were performed in serum from 6 hour fasted animals with following kits

Kits Catalog number/ Company

Insulin (90080) from Crystal Chem glycogen (K648-100) from BioVision

β-hydroxybutyrate / ketone bodies (700190) from Cayman chemicals

All the ELISA were performed according to the manufacturers instruction.

2.5 Protein biochemistry

2.5.1 Protein extraction with Dignam C buffer

49 Material and Methods

Mice were sacrificed by the way of CO2 inhalation; tissues were collected for protein and RNA isolation and rapidly snap frozen in liquid nitrogen. For preparing the protein extract for SDS-PAGE the liver tissues were incubated in 3-5 volumes of Dignam C buffer additionally supplemented with 1mM DTT, Roche complete mini protease inhibitor (1 tablet/10 ml) and finely grinded with the help of tissue lyser (Tissue lyser Retch, Qiagen, Germany). Afterwards the crude extract was centrifuged (13000rpm for 30min at 4 °C) and the pellet was discarded. Finally the protein concentration in the supernatant was determined by a Bradford assay.

2.5.2 Protein concentration measurement by Bradford method

2 µl of protein extracts made with Dignam C buffer was mixed with 100 µl of 150 mM NaCl and 900 µl of Bradford reagent and vortexed briefly. Optical density of formed protein/coomassie-brilliantblue complex was measured in photometer at wavelengths of 595 nm. Protein concentration in extracts was estimated according standard curve with known concentrations of bovine serum albumin. Protein extracts were shock- freezed in liquid nitrogen and stored at -80°C.

2.5.3 In vitro Luciferase reporter gene activity measurement

For the exact determination of transgene expression in protein extracts, luciferase reporter gene activity was measured with the luminometer Lumat LB9507 (Berthold, Bad Wildbad). For in vitro luciferase assay, 5 µl of protein extract were incubated with 50 µl of luciferase buffer (20 mM Tris, 1.07mM magnesium carbonate, 2.7 mM magnesium sulfate, 0.1 mM dimethyl sulfoxide (DMSO), 60 mM dithiothreitol (DTT), 1.06mM adenosine triphosphate (ATP), 0.54 mM coenzymeA (CoA), 1 mM luciferin) and light emission (RLU) was measured over 15 seconds. From mean values of two measurements the background activity (Dignam C) was subtracted and the obtained values were normalized to the amount of protein (RLU/ µg).

2.5.4 Western blotting

For SDS-PAGE 20-50 µg of protein were diluted with the Dignam C buffer to a specific volume and incubated with the appropriate amount of 4x-Laemmli loading

50 Material and Methods buffer + 5 % β-mercaptoethanol, freshly added for 10 min at 95 °C. The protein samples were separated on polyacrylamide gels with a 5 % polyacrylamide stacking gel (with 0.5 M Tris-HCl pH 6.8, 20 % SDS, APS 10%) and a resolving gel with 8,10,12.5,15 % polyacrylamide depending on the size of the proteins to detect (with 1.5 M Tris/HCl pH 8.8, 20% SDS, APS 10%). The gels were run with the electrophoresis system (CBS Scientific Company, California) in a Tris/glycine/SDS running buffer (25 mM Tris, 192 mM glycine, 0.1% SDS). With the Trans-Blot-Cell system (CBS Scientific Company, California) the proteins were transferred to a nitrocellulose membrane (0.45 µm pore size, Whatman), which was activated with methanol and washed in the blotting buffer (25 mM Tris, 192 mM glycine, 0.025 % SDS, 20 % methanol). The membranes are then washed three times with TBS/Tween (0.1 %). Unspecific protein binding was blocked by incubation with 5 % dry milk powder in TBS/Tween (0.1 %) for 1 h at RT followed by incubation with primary (in blocking solution) overnight at 4° C. After three washes with TBS/Tween the membranes were incubated with the HRP-coupled secondary antibody (1:5000 in blocking solution) for 1 h at RT. After additional three washes chemiluminescence signals were developed with ECL (Super Signal West Dura Extended Duration Substrate Thermoscientific) and detected with X-ray films. Similarly β-tubulin was detected on the same membrane as a control for equal loading. The membrane was sometimes reprobed with several antibodies. If old bands might mask the new bands to detect, the membrane was stripped by incubation with 0.2 M NaOH for 5 min, followed by thorough washing and new blocking. The weterns are performed with the following antibodies listed in (Table 5).

51 Material and Methods

Table 5: Antibodies used for immunostaining and immunoblotting

Antigen source Company Cat.no WB dil IF dil IHC dil FOXO3/FKHRL Rb Santacruz sc-11351 1:1000 1:100 1:100 AKT Rb Cellsignalling 9272 1:1000 T-308phospho-AKT Rb Cellsignalling 4056 1:1000 S-473 phospho-AKT Rb Cell signalling 9271 1:1000 AMPK Rb Cell signalling 2603 1:1000 T172 phospho-AMPK Rb Cell signalling 2535 1:1000 β-Tubulin Abcam Ab6046 1:5000 K8/18 (Ku, Toivola 8592 1:100 et al. 2004) Insulin Rb Cellsignalling 4590 1:100 1:100 Cd45 Rb Abcam 10558 1:200

Abbreviations: IB Immunoblot, IF immunofluorescence, dil. dilution, Rb rabbit, IHC

2.6 RNA Analysis

2.6.1 RNA extraction and cDNA synthesis

RNA was extracted from liver samples kept in RNAlater (Qiagen) by RNAeasy Mini Kit (Qiagen), following the manufacturers instruction. Then the RNA was dissolved in 50 µl of DEPC water and the concentration of RNA was measured with the Nanodrop 1000 spectrophotometer (Thermo Scientific). For purity control the ratio of absorption at 260/280 nm and 260/230 nm was measured. Complementary DNA was generated from 2 µg of RNA using MMLV reverse transcriptase (Promega) using the manufacturers instruction. Quantitative real-time PCR was carried out using qPCR master mix and corresponding universal probe library on Roche LC480 light cycler system (Roche). Primers are listed in (Table 6). PCRs were performed in triplicates according to the manufacturer's instructions, with HPRT used as a reference for relative quantification. Data were corrected for primer efficiency, which was determined by serial dilution of cDNA.

52 Material and Methods

Table 6 : List of primers used in real time PCR analysis.

Transcript Probe (UPL) L primer R primer FoxO3 11 cttcaaggataagggcgaca cgactatgcagtgacaggttg

G6PC 19 Tctgtcccggatctaccttg gaaagtttcagccacagcaa

Pdk4 22 Cgcttagtgaacactccttcg cttctgggctcttctcatgg Pepck1 49 ggagtacccattgagggtatcat gctgagggcttcatagacaag

Igfbp1 66 Ttcagctcccagcatgaag gggctctcagaaagctcatc

Igfbp2 62 gcgggtacctgtgaaaagag cctcagagtggtcgtcatca

Lepr 50 gcacaaggactgaatttccaa ctgatttcttctgaaatgggttc

Pygl 12 Ccagagtgctctaccccaat ccaccacaaagtactcctgtttc

Cpt1α 70 Gactccgctcgctcattc tctgccatcttgagtggtga

Echs 79 Cgggtgagtctctgagtgc ccagagtgctctaccccaat

Acaca 50 Gcctccgtcagctcagatac tttactaggtgcaagccagacat

FAS 19 Gctgctgttggaagtcagc agtgttcgttcctcggagtg

SREBP1 74 ttcctcagactgtaggcaaatct agcctcagtttacccactcct

Acot11 10 Ccatcgtgaataacgccttt ggtacgcctccacacaca

Acss 63 aaatgtgttccaggatacaatgg atcagatatgggctcattttcc

Apom 78 cccagacatgaaaacagacct gagggtgtggtgaccgatt

Aqp7 84 Ggctccagccagaagtca gggccattgtggagttgt

Gckr 53 Gcttcaatccggtgagca tctcctgcatcttctgcatc

Pklr 79 acttagcaaagtctgctttaagtgg tggcacgtctcaggtatcc

Pdk3 7 aagcagatcgagcgctactc ttcacatgcattatcccttcc

Cpt-1b 92 gagtgactggtgggaagaatatg gctgcttgcacatttgtgtt

Fabp5 46 acggctttgaggagtacatga ctcggttttgaccgtgatg

53 Material and Methods

2.6.2 Gene expression analysis

Microarray analyses were performed using 200 ng total RNA as starting material and 5.5 µg ssDNA per hybridization (GeneChip Fluidics Station 450; Affymetrix, Santa Clara, CA). The total RNAs were amplified and labeled following the Whole Transcript (WT) Sense Target Labeling Assay (http://www.affymetrix.com). Labeled ssDNA was hybridized to Mouse Gene 1.0 ST Affymetrix GeneChip arrays (Affymetrix, Santa Clara, CA). The chips were scanned with an Affymetrix GeneChip Scanner 3000 and subsequent images were analyzed using Affymetrix® Expression Console™ Software (Affymetrix). A transcriptome analyses was performed using BRB-ArrayTools (http://linus.nci.nih.gov/BRB-ArrayTools.html). Raw feature data were normalized and log2 intensity expression summary values for each probe set were calculated using robust multiarray average (RMA, Irizarry et al., 2003). Filtering: Genes showing minimal variation across the set of arrays were excluded from the analysis. Genes whose expression differed by at least 1.5 fold from the median in at least 20% of the arrays were retained. Class Comparison: We identified genes that were differentially expressed among the two classes using a 2 sample t-test. Genes were considered statistically significant if their P value was less than 0.05 and displayed a fold change between the two groups of at least 1.5 fold. We used the multivariate permutation test to provide 90% confidence that the false discovery rate was less than 10%. To identify the most affected biological processes, as defined by annotation, we used the GoMINER analysis tool (Zeeberg et al., 2003) with minor modification. Complete microarray data are available at Gene Expression Omnibus (GEO accession number: GSE44977

2.7 Histology

2.7.1 Liver tissue preparation

Livers removed from sacrificed animals were weighed and divided into several pieces that were fixed in 10% formaldehyde for histological/immunohistochemical analysis,

54 Material and Methods snap frozen in liquid nitrogen for immunofluorescence staining. Livers were dehydrated in increasing concentration of ethanol and embedded in paraffin according to standard histological procedures. From the paraffin blocks horizontal sections with a thickness of 2 µm were cut with a microtome and mounted on slides (3-1 sections per slide). For cryosections, the fresh liver tissue was immediately frozen on a plastic cassette that was cooled on liquid nitrogen with the help of tissue tek. The frozen tissue was cut with a cryotome (leica) into 4 µm thick sections (1-2 sections per slide) and stored at –20 °C.

2.7.2 Immunofluorescence staining of paraffin embedded sections

Before staining paraffin sections were incubated for at least 20 minutes then after incubation deparaffinized with 3x xylene, and rehydrated with ethanol (2x 100 %, 1x90 %, 1x70 %, water), for at least 5 min per bath. Then the slides were boiled for antigen retrieval for 15-20 min in a pressure cooker in preheated citric acid buffer (10 mM, pH 6.0, with 0.05 % Tween 20). After cooling for at least 30 min at RT the slides were incubated for 30 min with 0.5 % Triton X-100 in PBS for full permeabilization. Then the slides were washed three times for at least 5 min with PBS. For each slide, liver sections were surrounded with a PAP- pen to create two independent staining compartments per slide. To prevent unspecific antibody binding, sections were incubated for 1 h at RT with blocking solution containing 5 % BSA in PBS in a dark humidified chamber. After removal of the blocking solution the sections were incubated with the primary antibodies in blocking solution, usually overnight at 4 °C. After three washes with 1xPBS the slides were incubated with the fluorescently labeled secondary antibodies (dilution 1:500, labeled with Alexa Fluor 488 (green) or 568/594 (red)) in blocking solution for 1 h at RT, with DAPI (250 ng/ml) for nuclear counterstaining. After two additional washes with PBS (in dark), the slide were shortly washed with water and mounted with a coverglass with mowiol as aqueous mounting medium. TUNEL staining was performed with an in situ TUNEL cell death detection kit (Roche) following the manufacturer's instructions. Immunofluorescence staining for Insulin on pancreas sections was done with the same method as described above. Alexa Fluor-coupled secondary antibodies for immunofluorescence (goat anti-rabbit

55 Material and Methods

A488/A568, donkey anti-rabbit A594, goat anti-mouse A488/A568, donkey anti- mouse A594, donkey anti-goat A488, goat anti-rat A488) were obtained from Life Technologies (Molecular Probes, Paisley, UK) and used in a dilution of 1:500.

2.7.3 Immunofluorescence image aquisition and processing

To avoid bleaching of the fluorescent dyes, the slides were stored in the darkness and analysed within a few days. Images were taken with the Zeiss Axiovert 200M, with filters for DAPI, FITC/Alexa Fluor 488, DsRed and Texas Red/Alexa Fluor 568/594. All directly compared images were taken at the same session with the Zeiss Axiovision-software (version 4.8) with the same settings that were separately adjusted for each channel. After export to jpg files the final adjustment of total brightness (equally for all directly compared images) and the cutting of relevant image parts was performed with Adobe Photoshop.

2.7.4 Quantification of TUNEL staining, Cd45 and Insulin staining

For the quantification of Tunel positive cells and Cd45 positive cells, all the images were taken at the same settings. Images were obtained using Leica CTR5500 microscope. Cell count was performed manually using ImageJ64 software. The quantification of insulin staining for pancreatic sections was done with Image J64 software. All the images were taken in the same settings. The settings were made in the software to calculate the total area (including endocrine as well exocrine), size of islets and number of the islets the results were plotted by calculating the percentage of endocrine area.

2.7.5 Immunohistochemical staining of paraffin embedded sections

Before staininng paraffin sections were incubated for atleast 20 minutes then after incubation deparaffinized with 3x xylene, and rehydrated with ethanol (2x 100 %, 1x90 %, 1x80%, 1x70 %, water), for at least 5 min per bath. Then the slides were boiled for antigen retrieval for 15-20 min in a microwave in preheated citric acid buffer (10 mM, pH 6.0, with 0.05 % Tween 20). After cooling for at least 30 min at room temperature the slides were incubated with primary antibody for 1 hour at RT. The

56 Material and Methods primary antibody was made in blocking solution (0.1% BSA in PBS). The slides were then washed 3x in PBS for at least 5 minutes. After washing the slides were incubated with secondary antibody (Dako/Jackson Immunoresearch) for 1hour. The slides were washed 3x in PBS and after washing the slides were incubated and finally developed with two drops of AEC (Dako) on each liver section for 20 minutes. After 20 minutes incubation the slides were again washed in PBS two times and then immediately transferred in hematoxylin for 10 sec. The slides were then washed for 10 miutes in water bath with continous supply of water. After washing the slides were mounted with mowiol and left for air-drying at room temperature. For k8/k18 staining the sections were blocked with 10% goat serum with 1 %BSA prior to incubation with primary antibody. After incubation with secondary antibody slides were developed with permanent red dye systems (Dako). The other procedure is the same as described above. For Immunohistochemistry secondary antibodies were purchased from Dako/Jackson Immunoresearch and they were applied 50ml on everysection of slide.

2.7.6 PAS staining for Glycogen

For PAS (Periodic acid Schiffs) staining for glycogen content the sections were deparaffinized with 3x xylene, and rehydrated with ethanol (2x 100 %, 1x90 %, 1x80%, 1x70 %, water), for at least 5 min per bath. The slides were incubated in 0.5% periodic acid for 5 minutes. The slides were then washed in distilled water two times for 3 minutes. The slides were then incubated with Schiff´s reagent and provided high power in the microwave for 45-60 seconds until the colour was changed to deep magenta. The slides were then washed in the running tap water for 5 minutes. The slides were counterstained in hematoxylin for 3 minutes. Washed in running tap water and rehydrated with graded ethanol. The slides were mounted with entellan and left for air-drying. All the pictures for Immunohistochemical staining and Hematoxyline staining were taken with (Leica CTR5500) microscope.

2.8 Statistical analyses

Data are shown as means ±SD. Statistical significance was determined using a two- tailed student t-test P<0.05 was considered significant.

57 Material and Methods

2.9 Reagents and Materials

2.9.1 General chemicals

General chemicals (inorganic chemicals, solvents, standard organic salts, acids) were purchased from Applichem (Darmstadt), Roth (Karlsruhe), Merck (Darmstadt) and Sigma (Taufkirchen).

Item Company

Acetic acid (glacial) Merck, Germany Acetic acid Fluka, Switzerland Acrylamide solution 40% AppliChem, Germany Agarose Ultrapure, Electrophoresis Grade Gibco BRL , USA APS Sigma-Aldrich, Germany b- Mercaptoethanol AppliChem, Germany Bromphenolblue Carl Bittmann, Switzerland BSA (Albumin Fraction V) AppliChem, Germany Calcium chloride Sigma-Aldrich, Germany Di- sodium-hydrogen phosphate Sigma-Aldrich, Germany DTT AppliChem, Germany EDTA– dihydrate AppliChem, Germany Ethanol (absolute) Sigma-Aldrich, Germany Glycerol AppliChem, Germany Glycine AppliChem, Germany HEPES Gibco BRL , USA Hydrochloric acid 37% Sigma-Aldrich, Germany Hydrogen peroxide AppliChem, Germany Isopropanol Merck, Germany Methanol Merck, Germany Magnesium chloride Sigma-Aldrich, Germany Non-fat dried milk AppliChem, Germany Potassium chloride Carl Roth GmbH , Germany

58 Material and Methods

Potassium dihydrogen phosphate Merck, Germany SDS AppliChem, Germany Sodium chloride AppliChem, Germany Sodium dihydrogen phosphate Sigma-Aldrich, Germany Sodium hydroxide, pellets AppliChem, Germany Sodium orthovanadate Sigma-Aldrich, Germany TEMED AppliChem, Germany Tris (ultrapure) AppliChem, Germany Tween 20 AppliChem, Germany Urea AppliChem, Germany

2.9.2 Buffers and solutions

TNES-buffer 50 mM Tris-HCl (stock 1M pH 7.5) 0.1M EDTA (stock 0.5 M, pH 8.0) 0.1M NaCl 1 % SDS Proteinase K (Applichem, Darmstadt) 0.45 µg/ µl)

PBS (10x) pH 7.3 87.65 g/l NaCl 2 g/l KCl

11.7 g/l Na2HPO4

2.4 g/l NaH2PO4

TBS(10x) pH 7.6 24.2 g/l Tris 80g/l NaCl

TBE (10x) 121.1 g/l Tris 61.8 g/l hydroboric acid 7.4 g/l EDTA

59 Material and Methods

Dignam C buffer 20 mM Hepes (pH 7.9) 25 % glycerol 0.42 M NaCl

1.5 mM MgCl2 0.2 mM EDTA 1 mM PMSF 1 mM DTT, 1 tablet/10 ml complete mini protease inhibitor (Roche, Mannheim)

RIPA buffer 50 mM Tris-HCl (stock 1M pH 7.4) 150 mM NaCl 1 % Triton X-100 0.5 % sodium desoxycholate 0.1 % SDS 1 mM PMSF, 1 tablet/10 ml complete mini protease inhibitor (Roche, Mannheim)

Luciferin buffer (for luciferase assay) 20 mM Tricin 1.07 mM magnesiumcarbonatehydroxyde 2.67 mM magnesiumsulfate 0.1 mM EDTA 33.3 mM DTT 0.27 mM Coenzyme A 0.47 mM luciferin 0.53 mM ATP

4x Laemmli loading buffer 125 mM Tris-HCl (pH 6.8) 20 % SDS 25 % glycerol bromphenol blue for coloring 60 Material and Methods

(+ 5 % β-mercaptoethanol)

2.9.3 Protein biochemistry materials

5x Bradford reagent (Biorad, München) PageRuler prestained protein ladder (Fermentas, St. Leon-Rot) MagicMarkXP Western Protein Standard (Invitrogen, Life Technologies, Paisley, UK) Protease inhibitor , Complete-Mini (Roche Diagnostics, Germany) PVDF membrane Milipore, USA Whatman paper Invitrogen, USA

2.9.4 Histology and microscopy materials

Cresyl violet (Merck, Darmstadt) DAPI (Merck, Darmstadt) Entellan (Merck, Darmstadt) Mowiol (Merck, Darmstadt) Paraformaldehyde/PFA (Sigma, Taufkirchen) PAP-Pen (Sigma, Taufkirchen) Peel-A-Way disposable embedding molds (Polysciences, Warrington, PA, USA) Microscopy slides “Superfrost Ultra Plus” (Menzel, Braunschweig) Coverglasses 24x50 mm (VWR, Ulm)

2.9.5 Molecular biology materials

1kb plus DNA ladder (Invitrogen, Life Technologies, Paisley, UK) Agarose Ultrapure, Electrophoresis Grade (Invitrogen, Life Technologies, Paisley, UK) Ethidium bromide (Applichem, Taufkirchen) Taq Poymerase and 5x Green GoTaq Reaction buffer (Promega, Mannheim) dNTPs (Genaxxon, Ulm) PCR tubes 0.2 ml (Brand, Wertheim)

61 Material and Methods

Lightcycler 480 Multiwell Plate 96 (Roche, Mannheim)

2.9.6 General laboratory materials

Reaction tubes 1.5/2 ml (Eppendorf, Wesseling-Berzdorf) 15/50 ml High-Clarity Polypropylen Conical Tubes (BD Falcon, San Diego, CA, USA) 5 ml Polystyrene Round Bottom tubes GLKL (Greiner Bio-One, Frickenhausen) Pipette tips 1-1000 ml (Hightech lab, Warsaw, Poland and Axygen, Union City, CA, USA) 1/2/5/10/25 ml glass pipettes (Brand, Wertheim) 1ml Sub-Q syringes (BD, Franklin Lakes, NJ, USA)

3 Laboratoryequipment

3.1 General laboratory equipment

Pipettes 0.1-2 µl 2-20 µl 20-200 µl 200-1000 µl (Abimed, Gilson, Brand) Pipette controller Pipetus (Hirschmann, Eberstadt) Surgical forcipes and scissors (FST, Heidelberg) Biofuge Pico 17 (Thermo Scientific, Waltham, MA, USA) Megafuge 1.0R (Thermo Scientific, Waltham, MA, USA) Digital Heat Block (VWR, Ulm) Centrifuge (max.13.000 rpm) / Biofuge A (Heraeus, Germany) Dryer T5050E (Heraeus, Germany) Freezer (Liebherr, Germany) Deep freezer -80°C, Class N / HFU 586 (Heraeus, Germany) Block thermostat / TCR 100 (Roth AG, Germany) pH-meter Calimatic 761 (Knick, Berlin) ice-machine automatic / AF10 (Scotsman, Italy) Light microscope (Leica, Germany) Lightcycler 480 (Roche, Germany)

62 Material and Methods

Nanodrop (Peqlab, Germany) Magnetic stirrer / RCT-BE (Kika Labortechnik, Germany) Scale Scaltec (Scaltec, Heiligenstadt) Scale BP-210-S (Sartorius, Göttingen) Spectrophotometer Genesys 10S UV/VIS (Thermo Scientific, Waltham, MA, USA) Steam sterilizer H+P (Labortechnick, Germany) Thermocycler Primus 96 plus (MWG, Ebersberg) Gel documentation Genosmart (VWR, Ulm) Vortex Genie-2 / G-560E (Scientific Industries, USA) Cell culture incubator Hera-Safe (Thermo Scientific, Waltham, MA, USA) Laminar flow sterile bank (Thermo Scientific, Waltham, MA, USA) Shaker Polymax 1040 (Heidolph, Schwabach) Waterbath (B.Braun, Germany) Power supplies for electrophoresis Desatronic 500/400 (Desaga, Heidelberg) Agarose gel equipment was constructed by the scientific workshop facilities of Ulm University

3.2 Histology and microscopy equipment

Microtome Microm HM355S (Thermo Scientific, Waltham, MA, USA) Cryotome CM1900 (Leica, Wetzlar) Vibratome VT1200S (Leica, Wetzlar) Leica CTR 5500 microscope (5x,10x,20,40x) Microscope Axiovert 200M (Zeiss, Oberkochen): b/w camera AxioCam MRm objectives Plan-Apochromat 2.5x, 5x, 10x, 20x, 40x, 63x UV-lamp X-Cite 120 EXFO (Lumen dynamics, Mississauga, Canada) Excitation/emission filters for DAPI, FITC/Alexa Fluor 488, DsRed/PE and Texasred/Alexa Fluor 568/594 (AHF Analysentechnik) Stereomicroscope SteREO Discovery.V20 (Zeiss, Oberkochen): RGB camera AxioCam MRc Objective Achromat S 0.63x

63 Material and Methods

3.3 Software

Microsoft Office X Adobe Photoshop CS2 V9.0 Adobe Illustrator CS2 V12.0.0 ImageJ64 1.41 Zeiss Axiovision 4.8

64 Results

3 Results

3.1 Generation of transgenic mice with inducible liver-specific expression of constitutively active FOXO3

The role of FoxO1 in regulation of hepatic glucose production has been showed previously in several mouse models (Matsumoto, Han et al. 2006; Zhang, Patil et al. 2006; Matsumoto, Pocai et al. 2007; Lin and Accili 2012). However the interplay between different FoxO family members sharing the conserved DNA binding domain is far from being understood. Importantly in addition to FoxO1, FoxO3 is also significantly expressed in the liver its contribution to metabolic regulation has however not been thoroughly analysed. Interestingly previous studies have shown the role of FoxO3 in insulin sensitivity phenotypes and longevity in humans suggesting that FoxO3 is critical for metabolic control (Willcox, Donlon et al. 2008; Flachsbart, Caliebe et al. 2009; Pawlikowska, Hu et al. 2009). Therefore the present study was designed to investigate consequences of FOXO3 function in vivo. For this purpose, a transgenic mouse model was generated that allows expression of a mutant (constitutively active) FOXO3 allele specifically in the liver with an inducible transgene expression system to conditionally modulate FOXO3 activity in hepatocytes.

3.1.1 Tetracycline regulated expression of a constitutively active FOXO3 allele driven by the LAP promotor

To activate FoxO3 signaling in the liver a constitutively active allele of FoxO3 (FOXO3-CA) was used in which the three AKT phosphorylation sites (T32, S253, S315) are mutated to alanine residues keeping the FOXO3 in an activated confirmation (Schmidt-Strassburger, Schips et al. 2011; Schips, Wietelmann et al. 2011) (Figure 14). We crossed transgenic mice expessing tetracycline-regulated transactivator (tTA) under the control of LAP promotor which is highly specific for the liver (Sunami, Leithauser et al. 2012) with mice carrying the FoxO3CA allelle and a luciferase reporter gene controlled by a bidirectional promotor (luciferase-(tetO)7- FOXO3-CA) (Figure 15). Tetracyclin transactivator (tTA) binds to tet-operator elements (tetO)7 and drives the transcription of both the luciferase reporter gene and the FOXO3-CA transgene. The tTA binding can be blocked by doxycycline (Dox), 65 Results thereby shutting off transgene expression (Lavon, Goldberg et al. 2000; Sunami, Leithauser et al. 2012)

Figure 14. Mutated FOXO3 structure in which three AKT phosporylation sites T32, S253, S315 are mutated to alanine residues.

Figure 15: Tetracycline regulated (“tet-off”) expression system for the liver specific expression of FOXO3-CA and a luciferase reporter gene.

Animal obtained from breedings (LAP.tTA x (tetO) 7.FOXO3-CA) which contain both the transgenes are called double transgenic (FoxO3CAHep mice) should express FOXO3CA in a doxycycline-regulated manner, whereas single transgenic or wild type littermates should not express FOXO3CA and could be used as controls (Control).

3.2 Perinatal induction of FOXO3CA results in lethal phenotype

3.2.1 liver-specific expression of FOXO3CA

In order to avoid consequences of FoxO3CA expression during embryogenesis, pregnant mothers were kept under doxycycline in the drinking water to repress FoxO3 expression. With this setting, FoxO3CAHep were born at the expected Mendelian ratio. In a first set of experiments doxycycline was removed immediately after birth and the mice were anlaysed at the age of 4-weeks (Figure16A).

66 Results

Macroscopically the FoxO3CAHep mice at this age appeared very sick, they were much smaller then their wildtype littermates, and some mice had already died before the age of 4-weeks (52.94%alive and 47.04 dead) (Figure16B). In vivo luciferase imaging was performed to determine the specificity of transgene expression by injection of luciferin (150µg/g body weight) as an indirect measurement of transgene expression. In vivo imaging confirmed the strongest activity clearly in the liver of double transgenic animals (Figure 16C). Transgene expression was further analysed in different organs from FoxO3CAHep mice by ex vivo luciferase measurement. High luciferase activity was detected only in the livers of 4-week-old FOXO3CA mice and other peripheral tissues had nondetectable luciferase activities (Figure 16D). Hepatic expression of FoxO3CA transgene was further confirmed by immunoblot analysis using antibodies against FOXO3 and β-Tubulin. The results showed strong expression of FOXO3 band only in FoxO3CAHep animals and was undetected in liver from control animals (Figure 16E) Furthermore FoxO3CAHep mice showed a pronounced decrease in liver size and liver weight even more than the body weight (Figure 16F) and the reduced liver weight to body weight ratio (LW/BW) was due to reduction in liver size (Figure 16G).

67 Results

Figure 16: FoxO3CAHep mice exhibit a robust, liver-specific transgene expression A. Schedule of Doxycycline treatment. Doxycyline was administered untill birth. After birth DOX administration is ceased and the animals were analysed at 4-weeks of age. B. Macroscopic appearence of FoxO3CAHep mice as compared to their littermate controls.

68 Results

C. In vivo assessment of transgene expression. Control and double transgenic (FoxO3CAHep) mice were analysed for luciferase expression using an IVIS system. The FoxO3CAHep mice were kept with doxycycline (DOX) until birth then DOX treatment was ceased. Luciferase activity was imaged at the age of 4- weeks. D.Transgene expression was analysed in different organs from FoxO3CAHep mice by ex vivo luciferase measurement. High luciferase activity was detected in livers of 4- week-old mice. Extrahepatic tissues displayed only negligible luciferase activities (RLU: relative light unit). E. Western blot analysis of whole liver extracts from 4-week-old control and FoxO3CAHep mice using antibodies against FOXO3 and β-Tubulin. The latter was used as a loading control. F. Liver from FoxO3CAHep mice and control animals. The livers from transgenic animals were much smaller then their littermate controls. G. Liver weight, bodyweight and liver weight/body weight ratio from 4-week-old mice are shown (n=7, data are represented as mean +/-SD).

3.2.2 FOXO3CA activation affects size and structure of hepatocytes and induces liverdysfunction and metabolic abnormalities

In order to obtain a more detailed insight into the liver structure and morphology, histological sections stained with hematoxylin and eosin were analysed in detail. FoxO3CAHep mice showed much smaller hepatocytes as compared to control animals. Immunohistochemical staining performed with FOXO3 antibody showed the strong nuclear expression of FOXO3CA in these smaller hepatocytes (Figure 17 A,B).

69 Results

Figure 17: Expression of liver-specific FoxO3CA leads to morphological alterations. A: Paraffin sections stained with hematoxylin and eosin; scale bar: 100 µm. B: Paraffin sections stained with an antibody specific for FOXO3CA (Brown). Scale bar: 100µm.

We then analysed the FoxO3CAHep mice with respect to markers of liver damage serum levels of the liver enzymes alanine aminotranferase (ALT) and aspartate amino transferase (AST) were elevated in FoxO3CAHep mice (Figure 18A, B). At the same time, bilirubin levels were increased in FoxO3CAHep mice as compared to control 4-week-old animals (Figure 18C). These changes in ALT, AST, and bilirubin reflect a significant extent of liver damage. Furthermore FoxO3CAHep mice have a moderate reduction in triglyceride (TG) and cholesterol levels (Figure 18D-E) as reported before due to defective insulin action in the liver (Michael, Kulkarni et al. 2000; Zhang, Patil et al. 2006). Interestingly, FoxO3CAHep mice showed reduction of blood glucose levels as compared to control animals and were severely

70 Results hypoglycemic (63.75 ± 4.9 mg/dl vs. 139 ± 11.6 mg/dl, FoxO3CAHep vs. Control, respectively) (Figure 18F). This could be due to dual role of FoxO transcription factors in controlling hepatic insulin sensitivity (Matsumoto, Han et al. 2006) or might be due to some other reasons discussed later.

Figure 18: Expressio of FoxO3CA induces hepatocyte damage and altered metabolic parameters A. Alanine aminotransferase (ALT) and B. aspartate aminotransferase (AST) C. bilirubin, D. Serum triglycerides E. Serum Cholesterol F. Blood glucose levels in the 4-week-old control and FoxO3CAHep mice (n=7, data are represented as mean +/- SD).

71 Results

3.2.3 Expression of FOXO3CA in younger animals induces autophagy

As in a previous study it has already been shown (Schips, Wietelmann et al. 2011) that the consequences of constitutive activation of FOXO3 in cardiomyocytes induces reversible cardiac atrophy and autophagy, we reasoned whether in this mouse model constitutive activation of FoxO3CA in the liver might induce the upregulation of genes involved in autophagy because the animals at 4-weeks of age were extremely atrophic. We therefore performed RTqPCR for autophagy-associated genes (Atg1, Atg12, Atg5). The results showed a moderate upregulation for cell autophagy associated genes investigated namely Atg1, Atg5, Atg12 (Figure 19). In summary FoxO3 activation just after birth results in smaller animals with hypoglycemic phenotype, additionally hepatocyte damage and altered metabolic parameters was observed. Furthermore FoxO3CAHep mice at the age of 4-weeks showed moderate upregulation of autophagy associated genes and this may explain smaller size of hepatocytes and smaller size of liver. Autophagy is the catabolic process of lysosomal degradation and self-destruction. In FoxO3CAHep mice at the age of 4-weeks catabolic pathways such as autophagy was activated that might result in hypoglycemic phenotype or due to some other unknown reasons and also several other parameters were not listed due to severity of phenotype. The results from 4-week-old animals prompted us to analyse FoxO3CA activation in adult animals.

72 Results

Figure 19: FoxO3CAHep shows elevation of various autophagy-associated genes. Expression of several autophagy-associated genes determined by RTqPCR. Bars represent the ratio of gene expression relative to Hprt (n=7).

3.3 FOXO3CA activation in adult animals

3.3.1 Expression pattern of FOXO3CA transgene

In order to analyze FoxO3 function in mature livers, transgene expression was suppressed until 5-weeks of age by administration of doxycycline (DOX) in the drinking water. After removal of doxycyline/ induction of transgene expression mice were analysed at two different time points at the age of 7-weeks (after two weeks of FOXO3CA activation) and at the age of 9-weeks (after four weeks of FOXO3CA activation) (Figure 20A). In order to check for the specificity of transgene expression in vivo imaging was performed both for Control and FoxO3CAHep mice at the age of 7-weeks and at the age of 9-weeks. In vivo imaging again showed strong expression of the transgene only in the liver specifically in double transgenic FoxO3CAHep mice (Figure 20B). Importantly, these mice appeared normal. In vitro luciferase assays were also performed in both 7-week-old and 9-week-old animals. The results revealed that expression of the transgene reporter luciferase was restricted to the liver of FoxO3CAHep mice. Extrahepatic tissues displayed only negligible luciferase activities (Figure 20C).

73 Results

Figure 20: FoxO3CAHep mice reveal a robust, liver-specific transgene expression at the age of 7-weeks and 9-weeks. A. Schedule of doxycycline treatment. DOX administration was started from birth untill 5-weeks of age then DOX administration is ceased and the animals were analysed at 7-weeks and at 9-weeks after 2-weeks and 4-weeks of FOXO3CA activation. B. In vivo assessment of transgene expression. Control and double transgenic (FoxO3CAHep) mice were analysed for luciferase expression using an IVIS system. Luciferase activity was imaged at the age of 7-weeks and at the age of 9-weeks. C. Transgene expression was analysed in different organs from FoxO3CAHep mice by ex vivo luciferase measurement. High luciferase activity was detected in livers of both 7-week-old and 9-week-old mice. Extrahepatic tissues displayed only negligible luciferase activities (RLU: relative light unit).

74 Results

In addition, FoxO3 transgene expression was confirmed at the protein level by performing western blot of whole liver extracts from 7-week-old and 9-week-old mice with antibodies specifically recognizing FoxO3 protein and β-Tubulin as a loading control. The results from western blotting showed strong expression of FoxO3 protein only in FoxO3CAHep mice both at 7-week and 9-week but not in control animals (Figure 21A). We also checked the hepatic expression of FOXO3CA gene at the transcriptional level via real time RT-PCR. The results showed FOXO3CA gene was upregulated in FoxO3CAHep mice both at the age of 7-weeks and 9-weeks. Results are normalized to expression of hypoxanthine-guanine phosphoribosyl transferase gene (Hprt). (Figure 21B).

Figure 21: Detection of FoxO3CA transgene at the protein and transcriptional level in 7-week and 9-week-old mice. A. Western blot analysis of whole liver extracts from 7-week-old and 9-week control and FoxO3CAHep mice using antibodies against FOXO3 and β-Tubulin. The latter was used as a loading control. B. Hepatic expression of FOXO3CA was determined via real time RT-PCR. Results are normalized to expression of hypoxanthine-guanine phosphoribosyl transferase gene (Hprt).

3.3.2 Expression of FOXO3CA in adult animals is not associated with growth defect

75 Results

As already mentioned, mice with hepatic activation of FOXO3CA at birth resulted in growth defect. Accordingly we considered enlisting weight parameters in control and FoxO3CAHep animals at the age of 7-weeks and 9-weeks. We observed that activation of FoxO3 for 2-weeks and 4-weeks did not result in any obvious growth defect as the body weight, liver weight and liver weight/ body weight ratios were normal and the animal appeared active and normal (Figure 22 A,B).

Figure 22: Normal body weight and liver weight characteristics of FoxO3CAHep mice at the age of 7-week and 9-weeks. A. Body weight, liver weight and liver weight/body weight ratio from 7-week-old mice are shown (n=7, data are represented as mean +/-SD). B. Body weight, liver weight and liver weight/body weight ratio from 9-week-old mice are shown (n=7, data are represented as mean +/-SD).

3.3.3 FOXO3CA activation results in altered hepatic morphology

Active FOXO3 should accumulate in the nucleus and activate FOXO3-dependent gene expression. We determined FOXO3 expression and localization by immunohistochemical analysis both in 7-week-old and 9-week-old animals and we found that FOXO3CA indeed localized in the nucleus. Interestingly, FOXO3CA 76 Results expression affected cell morphology, as FoxO3CAHep hepatocytes were much smaller than wildtype controls already seen at 4-weeks (Figure 17). To confirm that all these smaller cells are hepatocytes, Keratin 8/18 staining was performed as hepatocyte marker. The staining displayed all the smaller cells as hepatocytes (Figure 23 A, B).

Figure 23: Expression of FoxO3CA induces hepatocyte atrophy in FoxO3CAHep mice at the age of 7-weeks and 9-weeks. A. Representative H&E-staining from control and FoxO3CAHep mouse livers (A and B, bar: 100 µm). B. Expression of FOXO3 was determined by immunohistochemical staining with FOXO3 antibody. Strong nuclear staining of FOXO3 is seen in livers from FoxO3CAHep mice, Keratin 8/18 staining was performed for hepatocyte staining.

3.3.4 FOXO3CA activation induces moderate hepatocyte damage

77 Results

To examine liver functionality after two weeks of FOXO3CA activation the markers for liver damage and liver dysfunction were assessed from the serum of 7-week-old control and FoxO3CAHep mice after 6 hour of fasting. The transgenic animals showed mildly elevated serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels indicating moderate hepatocyte damage (Figure 24A). Elevated serum bilirubin levels could indicate hepatocyte dysfunction (Figure 24B). However, serum albumin and pseudocholinesterase levels, widely accepted hepatocyte function markers, were not changed (Figure 24C). Alkaline phosphatase levels were elevated in FoxO3CAHep mice suggesting that hepatocyte function is still intact, but biliary tract function is impaired (Figure 24D).

Figure 24: FoxO3CAHep show elevated markers of hepatocyte damage and hepatocyte dysfunction. A. Alanine aminotransferase (ALT) and aspartate aminotransferase (AST) B. Bilirubin C. albumin, pseudocholinesterase, and alkaline phosphatase (ALP) D. levels in the 7- week-old control and FoxO3CAHep mice (n=7, data are represented as mean +/-SD).

Increased liver enzyme levels would be a consequence of hepatocyte death. Indeed FOXO3CA expression in murine forebrain had resulted in massive apoptosis of neuronal progenitors (Schmidt-Strassburger et al., 2012). In order to check for apoptosis in FoxO3CAHep mice TUNEL assays were performed the staining and 78 Results

TUNEL quantification revealed not significant increased levels of apoptosis in FoxO3CAHep mice (Figure 25A). We used TAK1-LPCKO liver as positive control as it was previously reported that TAK1 deficiency mediates hepatocyte cell death (Bettermann, Vucur et al. 2010). We also performed CD45 staining for inflamation the results from staining and quantification of CD45 positive cells showed no signs of inflammation in FoxO3CAHep mice (Figure 25B). We used IKK2CA liver as positive control for inflammation because it has been shown that over-expression of IKK2 cause liver fibrosis through macrophage-mediated chronic inflamation (Sunami, Leithauser et al. 2012). Taken together, hepatic FOXO3 expression led to mild hepatocyte damage independent of inflammation or apoptosis, but hepatocyte function is largely intact.

Figure 25: Expression of constitutively active FoxO3 does not induce strong inflammation and apoptosis. A. Immunofluorescence TUNEL staining for 7-week-old control and FOXO3CAHep mice. Liver-specific deletion of TAK1 (Map3k7) mouse liver (Bettermann, Vucur et al. 2010) served as a positive control (bar: 100 µm). B. Immunohistochemical stainings against CD45 for 7-week-old control and FOXO3CAHep mice were shown. Liver from over-expression of IKK2 mutant (Sunami, Leithauser et al. 2012) served as a positive control (bar: 100 µm). 79 Results

3.3.5 Hepatocellular FOXO3 activation induces metabolic abnormalities

After two weeks of transgene activation, FoxO3CAHep mice displayed markedly elevated fasting blood glucose levels compared to control animals (282.4 ± 22.8 mg/dl vs. 128 ± 4.8 mg/dl, FOXO3CAHep vs. Control, respectively) (Figure 26A). Elevated blood glucose levels could be a consequence of increased gluconeogenesis. Consistently, increased hepatic expression of several genes involved in gluconeogenesis, namely glucose 6 phosphatase (G6pc), phosphoenolpyruvate carboxykinase (Pepck), and pyruvate dehydrogenase kinase isoenzyme 4 (Pdk4), was observed (Figure 26B). Furthermore in order to examine whether hyperglycemic FoxO3CAHep animals were able to tolerate a glucose load through glucosetolerance test. Glucose tolerance test (GTT) was performed as mentioned in materials and methods. The results from intraperitoneal glucose tolerance test performed on FoxO3CAHep mice demonstrated impaired glucose tolerance (Figure 26C). In order to analyze whether glucose intolerance might be a consequence of insufficient insulin production, we measured serum insulin levels. Interestingly, FOXO3CAHep mice showed significantly higher insulin levels (10.01 ± 1.56 ng/ml vs. 1.50 ± 0.82 ng/ml, FoxO3CAHep vs. Control, respectively) (Figure 26D). In order to further confirm that FoxO3CAHep mice are able to lower glucose in response to injected insulin insulin tolerance test (ITT) was performed as mentioned in materials and methods. The results of insulin tolerance test demonstrated that exogenously administered insulin failed to lower blood glucose level in FOXO3CAHep mice (Figure 26E) and further showed that FOXO3CAHep mice was extremely resistant to injected insulin. These data suggest that hepatic FOXO3 affects glucose homeostasis by regulating gluconeogenesis-associated enzymes and it impairs hepatic insulin signalling.

80 Results

Figure 26: Overexpression of FoxO3CA induces hyperglycemia, hyperinsulinemia and impaired glucose tolerance as well as insulin sensitivity. A. Serum glucose levels from 7-week-old control and FoxO3CAHep mice after 6 hours fasting (A- E: n=7, data are represented as mean +/-SD). B. Up-regulation of gluconeogenesis-associated genes in FoxO3CAHep livers. Results 81 Results from quantitative real-time PCR experiments are shown. Bars represent the ratio of the expression of the highlighted genes relative to the hypoxanthine-guanine phosphoribosyltransferase gene (Hprt). C. Glucose tolerance test. Blood glucose levels from control and FoxO3CAHep were measured at the indicated time points after glucose injection. Mice were fasted for 6 hours before performing experiments (*** P<0.001). D. Serum insulin level from control and FoxO3CAHep mice after 6 hours fasting. E. Insulin tolerance test. Blood glucose levels control and FoxO3CAHep were measured after insulin injection (*** P<0.001). Mice were fed ad libitum before performing the experiments.

3.3.6 Activation of catabolic pathways by expression of constitutively active FOXO3

FOXO3 transcription factors are involved in regulation of metabolic homeostasis therefore the observed hyperglycemic metabolic phenotype could be explained by up-regulation of FOXO3 metabolic target genes. To unveil FOXO3-induced transcriptional network, we performed microarray analyses using livers from 7-week- old control (n=4) and FoxO3CAHep (n=4) 6 hours, fasted mice. This analysis revealed 560 up-regulated genes and 421 down-regulated genes in FoxO3CAHep mice compared to control animals (Appendix Table 1). The data confirmed an increase in the mRNA levels for several gluconeogenesis-associated enzymes such as pyruvate dehyrogenase kinase (Pdk4, Pdk3) were upregulated. Pdk4 inhibits pyruvate dehydrogenase by phosphorylation providing pyruvate as gluconeogenic substrates (Wei, Tao et al. 2011). A known regulator of glucose and lipid metabolism peroxisome proliferator activated receptor-coactivator-1γ α (Pparγc1α) was upregulated. Subsets of genes involved in regulation of glucose homeostasis for providing energy during the process of starvation were upregulated (Table 7). Consistently the glycolytic genes such as glucokinase and pyruvate kinase (Gck, Pklr), were downregulated hepatic glucokinase is an important enzyme linking glucose and lipid metabolism by catalysing the conversion of glucose to glucose-6- phosphate, which is an important precursor in support of lipogenesis. In addition, we observed significant alterations in expression levels of genes involved in lipid metabolism and other metabolic pathways (Table 8). Consistently due to enhanced

82 Results activation of lipid catabolic pathways in FoxO3CAHep mice we have observed up- regulation of genes involved in lipid transport and trafficking several fatty acid binding proteins (fabp5, fabp7), apolipoprotein M (Apom), phospholipid transporting ATPase (Atp11) was upregulated. Genes involved in fatty acid synthesis were down-regulated ATP citrate (Acly), ATP citrate lyase provide a sole fuel for cytoplasmic acetyl- CoA, fatty acid synthase Fasn, acetylCoA carboxylase (Acc) acetyl-CoA carboxylase- α which convert acetyl-CoA to malonylCoA which serve as a substrate for lipogenesis, acetyl-CoA synthase (Acsm1). In line with this genes involved in fatty acid oxidation were up-regulated carnitine palmitoyltransferase (cpt1b), enoyl-CoA hydratase/3-hydroxy acyl Coenzyme A dehydrogenase (ehhadh), acetyl-CoA acetyltransferase (Acot2), acety-CoA dehydrogenase (Acad12). This suggests that FOXO3 activates catabolic pathways to provide substrates and energy for gluconeogenesis.

Table 7: Genes involved in glucose metabolism Classification of the main genes that were differentially expressed between control mice and FoxO3CAHep mice. Relevant up-regulated genes were categorized into several ontologies. Data are expressed as the mean-fold difference in gene expression in FoxO3CAHep mouse livers relative to control mouse liver samples.

Fold-change Transgenic (n=4) vs Function Gene Symbol Gene name Control(n=4) Lepr leptin receptor 29.41 Pdk4 pyruvate dehydrogenase kinase, isoenzyme 4 15.38 Igfbp1 insulin-l ke growth factor binding protein 1 10.99 Akr1b7 aldo-keto reductase family 1, member B7 7.14 Igfbp2 insulin-l ke growth factor binding protein 2 4.35 Glucose Irs2 insulin receptor substrate 2 3.23 Metabolism Me2 malic enzyme 2, NAD(+)-dependent, mitochondrial 2.38 Hmox1 heme oxygenase (decycling) 1 2.38 Leprot leptin receptor overlapping transcript 2.38 Aqp7 aquaporin 7 2.08 Pgm3 phosphoglucomutase3 2.22 Akr1b8 aldo-keto reductase family 1, member B10 gucose metabolism 1.96 Pdp1 pyruvate dehydrogenase phosphatase catalytic subunit GM 1.92 Pparγc1α peroxisome proliferative activated receptor, gamma, coactivator 1 alpha 1.92 Igf1r insulin-l ke growth factor-1 receptor 1.92 Gys2 glycogen synthase 2 1.85 Pdk3 pyruvate dehydrogenase kinase isoenzyme 3 GM 1.63 Pgd phosphogluconate dehydrogenase 1.61 Aldh3b1 aldehyde dehydrogenase 3 family, member B1 1.59 Pklr Glycolysis (Pyruvate kinase liver) 0.51 Igf1 insulin-l ke growth factor 1 0.49 Gckr Glucokinase regulatory protein 0.45 Gck glucokinase 0.23

83 Results

Table 8: Genes involved in lipid metabolism

Fold- change Transgeni c (n=4) vs Control(n= Function Gene Symbol Gene name 4) Lipid Metabolism Acmsd amino carboxymuconate semialdehyde decarboxylase 4.17 Acot2 acyl-CoA thioesterase 2 4.17 Acad12 acyl-Coenzyme A dehydrogenase family, member 12 2.7 Cpt1b Carnitine palmitoyltransferase 1b,muscle 2.56 Smpd3 Sphingomyelinphosphodieasterase 3 2.43 Ehhadh enoyl-Coenzyme A, hydratase/3-hydroxyacyl Coenzyme A dehydrogenase 2.33 Ugcg UDP-glucose ceramide glucosyltransferase 2.04 Apom apolipoprotein M 2 Lipo1 lipase, member O1 2 Cerk ceramide kinase 2 Pld4 phospholipase D family, member 4 1.89 B3galnt1 UDP-GalNAc:betaGlcNAc beta 1,3-galactosaminyltransferase, polypeptide 1 1.82 UDP-N-acetyl-alpha-D-galactosamine: polypeptide N- Galnt7 acetylgalactosaminyltransferase 7 1.82 Acsl4 acyl-CoA synthetase long-chain family member 4 1.82 Atp11a Phospholipid transporting ATPase 1.78 Fabp7 fatty acid binding protein 7 1.78 Gltp Glycolipid transfer protein 1.72 Pla2g12a phospholipase A2, group XIIA 1.54 Fabp5 fatty acid binding protein 5, epidermal 1.53 Acsm1 acyl-CoA synthetase medium-chain family member 1 0.66 Ldlr low density lipoprotein receptor 0.66 Acot11 Acyle-CoA thioesterase11 0.58 Acss3 AcetyleCoA synthetase, short chain family member 3 0.36 Fasn Fatty Acid synthase 0.42 Acacα AcetyleCoA carboxylase alpha 0.55 Acly ATP citrate lyase 0.58

The normal action of insulin in the liver is to inhibit glycogenolysis and gluconeogenesis (Jitrapakdee 2012). It is reported peviously in type 2 diabetic conditions that 90% of impaired hepatic glucose production occurs by gluconeogenesis and the rest is supported by glycogenolysis (DeFronzo, Bonadonna et al. 1992). Therefore in order to further confirm the activation of catabolic pathways including glycogenolysis liver glycogen content was examined histochemically by PAS (Periodic Acid Schiffs) staining for liver glycogen. Indeed, FoxO3CAHep mice have significantly reduced glycogen as compared to control animals. Furthermore the results from biochemical analysis revealed reduced glycogen in FoxO3CAHep mice as compared to control animals. Accordingly, glycogen phosphorylase (an enzyme involved in conversion of glycogen to glucose during glycogenolysis) mRNA expression levels are up-regulated in FoxO3CAHep mice as compared to control animals (Figure 27).

84 Results

Figure 27: Increased glycogenolysis in FoxO3CAHep mice. A.Representative PAS (Periodic Acid Schiffs) stainings from control and FoxO3CAHep mice (bar: 100 µm). B. Quantification of hepatic glycogen content. C. Up-regulation of a glycogen phosphorylase gene (Pygl) in the livers from FoxO3CAHep mice. Results from RTqPCR experiments are shown. Bars represent the ratio of the expression relative to Hprt.

85 Results

In addition, serum triglyceride and cholesterol levels were measured from serum of 6 hour fasted FoxO3CAHep and control mice. The results showed fewer triglycerides and cholesterol in FoxO3CAHep mice compared to control animals (Figure 28A), and consistently again mRNA expression of lipid anabolic enzymes namely acetyl-CoA carboxylase (Acc), sterol regulatory element-binding protein 1c (Srebp1), and fatty acid synthase (Fasn) were reduced (Figure 28B). This might be due to reduced effect of insulin in the liver to stimulate lipogenesis together with enhanced insulin activity in extrahepatic tissues to inhibit lipolysis as reported before in liver-specific insulin receptor knockout mice (Michael, Kulkarni et al. 2000).

Figure 28: FoxO3CAHep mice display decreased lipogenesis. A. Serum triglyceride and cholesterol levels from control and FoxO3CAHep mice. B. Expression of lipid synthesis-associated genes determined by RTqPCR. Bars represent the ratio of gene expression relative to Hprt.

In contrast, expression of mRNA levels for lipid catabolic enzymes like carnitine palmitoyltransferase 1a (Cpt1a) and enoyl-CoA hydratase (Echs1), were upregulated (Figure 29A). Consistently, serum ketone bodies (β-Hydroxybutyrate), markers for excessive lipid β-oxidation, were elevated in FoxO3CAHep mice (Figure 29B).

86 Results

Figure 29: Enhanced lipid catabolic pathways in FoxO3CAHep. A. Hepatic mRNA levels for genes involved in lipid catabolic pathway (β-oxidation) were measured in control and FoxO3CAHep mice. Bars represent the ratio relative to Hprt. B. Serum ketone bodies from control and FoxO3CAHep mice.

We also measured serum urea as end product of protein catabolism but observed no significant difference compared to controls. In addition, serum creatinine levels as marker of kidney function were not elevated, indicating that the process of protein catabolism was not significantly activated in FoxO3CAHep mice (Figure 30A).

Figure 30: FoxO3CAHep shows no symptoms of protein catabolism. A. Levels of serum urea as end product during protein catabolism and creatinine as marker of kidney function from 7-week-old mice are shown (n=7, data are represented as mean +/-SD).

87 Results

Taken together, activation of hepatic FOXO3 induces catabolic pathways including glycogenolysis, and β-oxidation of fatty acids, to presumably provide metabolites and energy required for excessive glucose production.

3.3.7 Continuous expression of constitutively active FOXO3 in the liver induces loss of gluconeogenic potential

For assessing the disease progression, FOXO3CAHep mice were kept for another 2 weeks. By 9 weeks of age, to our surprise, FOXO3CAHep mice showed fasting hypoglycemia (70.42 ± 8.3 mg/dl vs. 127.8 ± 4.8 mg/dl, FoxO3CAHep vs. Control, respectively) rather than hyperglycemia (Figure 31A). This could be due to high insulin production by pancreatic langerhans β-cells. Indeed, serum insulin level in 9- week-old FoxO3CAHep mice showed significantly higher insulin levels (8.37 ± 1.63 mg/dl vs. 0.71 ± 0.64 mg/dl, FoxO3CAHep vs. Control, respectively) (Figure 31B).

Figure 31: Expression of FoxO3CA at 9-week results in hypoglycemia and hyperinsulinemia. A. Serum glucose levels from 9-week-old control and FoxO3CAHep mice after 6 hours fasting (A-B: n=7, data are represented as mean +/-SD). B. Serum insulin level from control and FoxO3CAHep mice after 6 hours fasting.

However, this is in contrast with our previous finding that high insulin levels in 7- week-old FoxO3CAHep mice showed fasting hyperglycemia and whereas at 9-weeks of age the transgenic mice have fasting hypoglycemia. One reason could be that in FoxO3CAHep mice gluconeogenesis in the liver for glucose production is less active. To test our hypothesis, as depicted above expression levels of several genes

88 Results involved in promoting gluconeogenesis were analysed. This revealed that the mRNA expression of gluconeogenesis-associated genes was reduced/ normalized in 9- week-old FoxO3CAHep mice compared to 7-week-old mice Figure 26b (Figure 32). Hepatocytes are major place for insulin clearance and another possibility for fasting hypoglycemia is that, hepatocytes in liver failure probably are getting not able to clean up insulin properly as reported previously in liver-specific insulin receptor knockout mice (Michael, Kulkarni et al. 2000).

Figure 32: FoxO3CAHep mice at the age of 9-weeks shows normal gluconeogenesis. A. Up-regulation of gluconeogenesis-associated genes in FoxO3CAHep livers in 7- week-old mice. Results from quantitative real-time PCR experiments are shown. Bars represent the ratio of the expression of the highlighted genes relative to the hypoxanthine-guanine phosphoribosyltransferase gene (Hprt). B. Gluconeogenesis-associated genes in FoxO3CAHep livers in 9-week-old mice. Results from quantitative real-time PCR experiments are shown. Bars represent the ratio of the expression of the highlighted genes relative to the hypoxanthine-guanine phosphoribosyltransferase gene (Hprt).

89 Results

3.3.8 FOXO3CA activation in the liver induces subordinate pancreatic hyperplasia

We characterized further the hyperglycemic phenotype at the age of 7-week and hypoglycemic phenotype at the age of 9-week concerning pancreatic morphology since the pancreas is the major site of insulin secretion and several mouse models points to pancreatic hyperplasia in insulin-resistant states (Michael, Kulkarni et al. 2000; El Ouaamari, Kawamori et al. 2013). Interestingly we detected enlarged pancreatic langerhans islets in 9-week-old FoxO3CAHep mice as compared to control animals however pancreatic islets were normal structured in 7-week-old animals (Figure 33A). Insulin immunostaining demonstrated the prescence of enlarged β-cell- rich islets, which stained strongly for insulin (Figure 33B). Furthermore we quantified the endocrine area that was significantly more in 9-week-old animals then in 7-week animals (Figure 33C). Taken together, continuous activation of FoxO3 signalling leads to initial fasting hyperglycemia and subordinate fasting hypoglycemia, might be due to less insulin clearance capacity in the liver and more insulin production capacity in pancreas. However, several other unknown reasons might explain pancreatic hyperplasia as reported recently the role of liver-derived systemic factors and critical role of betatrophin hormone inducing pancreatic hyperplasia and improving glucose tolerance (El Ouaamari, Kawamori et al. 2013; Yi, Park et al. 2013).

90 Results

Figure 33: Pancreatic hyperplasia and increased insulin content in FoxO3CAHep mice at the age of 9-weeks. A. Representative H&E-staining from control and FoxO3CAHep mouse Pancreas, 9- week-old mouse has enlarged pancreatic Islets (bar: 100 µm). B. Immunoflourescence staining for Insulin 9-week-old mouse has larger area for producing insulin (bar: 100 µm). C. Quantification of endocrine area from both 9-week-old and 7-week-old mouse (n=50, data are represented as mean +/-SD).

3.4 Metformin normalizes FOXO3-induced distorted glucose and insulin metabolism

Metformin (1,1-dimethylbiguanide hydrochloride) is a hypoglycemic agent and one of the most widely used therapeutics for type 2 diabetes (Stumvoll et al., 1995). The principal function of metformin is in part to reduce hepatic glucose production and to

91 Results improve peripheral insulin sensitivity (Stumvoll et al., 1995). Although multiple hepatocyte pathways affected by metformin have been identified, the precise mechanism of metformin action still remains elusive. We wanted to determine, whether metformin administration attenuates FOXO3-induced distorted glucose homeostasis. We therefore administered metformin for the last 3 days until the animals become 7 weeks old (Figure 34A), or to see long-term effect of Metformin, we administered Metformin for 3 weeks (from 6 weeks to 9 weeks old) (Figure 34 B). Metformin was administered with a concentration of 1.65g/L in normal drinking water bottles and the quantity of metformin water used was checked and refilled everyday.

Figure 34: Schematic representation for the administration of Metformin

A. Animals were treated with metformin for 3 days before 7-weeks and the animals were analysed at the age of 7-weeks. B. Animals were treated with metformin for 3 weeks before 9-weeks and the animals were analysed at the age of 9-weeks.

We analysed whether 3-days and 3-weeks of metformin treatment affects FOXO3- induced dysregulation of glucose homeostasis. We first analysed potential effects of metformin treatment on transgene activity, In vitro Luciferase assays were performed both after 3-days of metformin treatment and 3-weeks of metformin treatment. The results showed that luciferase activity was unaffeted in the livers of FoxO3CAHep mice and extrahepatic tissues displayed negligible luciferase activities (Figure 35A-B). Furthermore at the molecular level FOXO3 protein expression was also intact in livers from FoxO3CAHep mice. This showed that metformin treatment has no effect on transgene activity or FoxO3 protein expression both after 3-days and 3-weeks of metformin administration (Figure 35C-D).

92 Results

Figure 35: FoxO3CA transgene expression is intact after administration of metformin. Metformin treatment has no influence on transgene expression in FoxO3CAHep. A. Transgene expression was analysed in different organs from FoxO3CAHep mice by ex vivo luciferase measurement after administration of metformin. Luciferase activity was intact in livers of 7-week-old as well as in 9-week-old mice (RLU: relative light unit). B. Western blot analysis of whole liver extracts from control 3days metformin-treated left and 3-weeks metformin-treated right and FoxO3CAHep mice using antibodies against FOXO3 and β-Tubulin. The latter was used as a loading control.

Several studies have shown a critical role of metformin in attenuating gluconeogenesis by lowering hepatic glucose production (Hundal, Krssak et al. 2000; Natali and Ferrannini 2006). Interestingly, already after 3 days of metformin treatment FoxO3CAHep mice showed normalization in fasting blood glucose level (Figure 36A). We asked whether this was due to normalization in gluconeogenesis and indeed found that several glucogeogenesis-associated genes which were up-regulated in 7- week-old FoxO3CAHep mice were normalized after short-term Metformin treatment (Figure 36B).

93 Results

Figure 36: Metformin administration rapidly normalizes FoxO3-induced distorted glucose level and sustained gluconeogenesis. A. Serum glucose levels from control and FoxO3CAHep mice, either without metformin treatment (-Met) or treated for 3 days (+Met3D) (A-B: n=7 data are represented as mean +/- SD). B. Hepatic mRNA levels for genes involved in gluconeogenesis were measured in control and FoxO3CAHep mice by RTqPCR. Bars represent the ratio of gene to Hprt expression.

Furthermore, at the age of 9 weeks, we observed no hypoglycemic phenotype in FoxO3CAHep mice after 3-weeks of metformin treatment (Figure 37A). In addition, elevated insulin levels in the both 7-week- and 9-week-old FoxO3CAHep mice, as well as enlarged endocrine area at the age of 9 weeks, were also all normalized (Figure 37 B-C). Consistently Threonine-308 and Serine-493 positions of AKT were no more phosphorylated after metformin treatment in FoxO3CAHep mice (Figure 38). Taken together, Metformin administration normalizes FOXO3-induced defective glucose and insulin metabolism.

94 Results

Figure 37: Metformin treatment normalizes glucose, insulin and pancreatic hyperplasia. A. Normalization of blood glucose after 3-weeks metformin treatment (+)Met3W (n=7 data are represented as mean +/- SD) all the animals were fasted for 6.hours before analysis. B. Normalization of serum Insulin after 3 weeks metformin treatment (+)Met3W (n=7 data are represented as mean +/- SD) all the animals were fasted for 6.hours before analysis. C. Normal histology of pancreatic islets form FoxO3CAHep mice after 3-weeks meformin treatment (+) Metformin3W lower panel as compared to pancreatic hyperplasia without metformin treatment at 9-weeks (-) Metformin upper panel.

In order to find out mechanism for metformin action it has been shown previously in

95 Results in vivo mouse model with adenosine mono-phosphate (AMP)-activated protein kinase (AMPK)-defeciency shows no influence on metformin-dependent reduction in blood glucose levels (Foretz, Hebrard et al. 2010), although Metformin has been suggested to perform anti-diabetic effect via activation of adenosine mono-phosphate (AMP)- activated protein kinase (AMPK) (Zhou, Myers et al. 2001). However, it has previously shown that AMPK phosphorylation of FOXO proteins results in activation and nuclear accumulation (Greer, Oskoui et al. 2007). In that respect, metformin would rather worsen than improve the pathology. We nevertheless analyzed state of AMPK activation as measured by Western blot analysis with antibody recognizing AMPKα harboring phosphorylated Threonine-172. Activation of FOXO3 resulted in significant reduction of AMPK phosphorylation (Figure 38). Metformin in our system did not measurably affect AMPK phosphorylation, suggesting that persistant FOXO3 activation may somehow control AMPK activity and might erace potential metformin- driven beneficial activation. Taken together, metformin administration normalizes FOXO3-induced defective glucose and insulin metabolism potentially independent of AMPK.

Figure 38: Metformin normalizes FoxO3CA driven distorted glucose, lipid, insulin metabolism independent of AMPK activation. Hepatic expression of AKT and phopsho-AKT (threonine-308, T308 and serine 473,

96 Results

S473), AMPKα as well as phospho-AMPKα (threonine-172, T172) was analyzed by Western blotting. β-Tubulin was used as a loading control.

3.5. FOXO3-induced activation of global catabolism, hepatocyte atrophy, or hepatocyte damage were reversed after Metformin administration

As described before, expression of FoxO3CA led to hepatic damage as well as hepatocyte atrophy. If glucose metabolism is normalized in FoxO3CAHep mice after Metformin administration, hepatocytes do not have to undergo autophagy. This suggested the hypothesis that metformin treatment in FoxO3CAHep mice could induce histological normalization and normal ALT and AST levels. As expected, there was no FOXO3-induced elevation of ALT and AST levels in serum after Metformin administration (Figure 39 A-B).

Figure 39: Metformin treatment results in normalization of hepatocytic damage. A. ALT B. AST levels in mice with 3-days metformin treatment and 3-weeks metformin treatment the values are compared with mice without any metformin treatment (n=7 data are represented as mean +/- SD) all the animals were fasted for 6 hours before analysis.

Furthermore, we examined liver histology to ascertain that metformin treatment has no influence on hepatocyte structure as well as on nuclear localization of FOXO3 and as we assumed the results displayed that there was no FOXO3 activation-driven hepatocyte atrophy as previously observed without any metformin treatment in both 7-week- and 9-week-old animals. After Metformin treatment there were no smaller 97 Results cells, and livers have the homogenous structure and morphology with strong FoxO3 expression (Figure 40).

Figure 40: Metformin treatment has no influence on FoxO3CA nuclear expression and normalizes hepatocyte atrophy. A. Representative H&E-staining from control and FoxO3CAHep mice livers (A, B the upper panel, bar: 100 µm). Normalization of liver histology after 3-days and 3-weeks of metformin treatment. B. Expression of FOXO3 was determined by immunohistochemical staining (A, B the lower panel, bar: 100 µm) with FOXO3 antibody. Strong nuclear staining of FOXO3 is seen in livers from FoxO3CAHep mice, Normalization of liver histology after 3days and

98 Results

3-weeks of metformin treatment.

Normalization in glucose homeostasis after Metformin administration could normalize general metabolism. To check whether Metformin administration also normalized mRNA expression not only for gluconeogenesis-associated genes, but also for global metabolic system, microarray analyses were performed using mRNA of livers from 6 hours fasted, 7-week-old mice without and with metformin treatment for 72 hours. The results of microarray analysis from 7-week-old animals with 3-days metformin treatment were compared with the 7-week-old microarray data without any metformin treatment. We found that 560 genes up-regulated and 421 genes down-regulated in FoxO3CAHep mice were normalized after metformin treatment (Appendix Table 2). The data confirmed the normalization of expression levels for genes both in glucose metabolism, as well as lipid metabolism by metformin as shown in heatmap (Figure 41A) all those genes involved in gluconeogenesis such as Pdk4 which were upregulated without any metformin treatment were downregulated with metformin treatment. In contrast the glycolytic genes Gck and pklr, which were downregulated without any metformin treatment were normalized with metformin treatment. Similarly genes involved in lipid transport (fabp7, fabp5, Apom) and lipid β-oxidation (Cpt1b, Acot12), which were upregulated without any metformin treatment, were normalized with metformin treatment and as expected the genes involved in lipid synthesis pathways (Acot, Acss), which were downregulated without metformin treatment were now normalized after metformin treatment (Figure 41A). All those genes were further confirmed by RTqPCR by using specific primers and the results were shown in log2 bardiagram (Figure 41B). We therefore conclude that metformin immediately targets FOXO3 activity. However the mechanism how metformin targets FoxO3 still needs further investigation.

99 Results

Figure 41A: Metformin administration reverses global catabolism induced by FoxO3CA activation. A. Heat-map summary of relative changes in gene expression. The left lane indicates fold change in FoxO3CAHep mouse liver compared with those of controls. The right 100 Results lane indicates fold change after administration of metformin in FoxO3CAHep mice compared with FoxO3CAHep mice without metformin.

Figure 41B: Relative up-regulated and down-regulated gene with and without metformin adminstration. B. From the Affymetrix data, several genes involved in glucose and lipid metabolisms were analysed by RTqPCR. Bars represent the ratio of gene to Hprt and data are shown in log2 bar diagram.

101 Discussion

4 Discussion

Type 2 diabetes mellitus and insulin resistance occur as a consequence of uncontrolled gluconeogenesis that is failed to be suppressed by insulin (Rizza 2010). It is a major worldwide health care problem affecting the quality of life. The physiological processes contributing to insulin resistance in type 2 diabetes remain enigmatic (Pajvani, Shawber et al. 2011). Therefore, there is a continuous need to identify molecular events responsible for type 2 diabetes and insulin resistance and this could help to discover new targets for the improvement of antidiabetic drugs.

One of the main signalling pathways involved in the insulin-dependent regulation of hepatic glucose production is Irs/PI3-K/Akt/FoxO pathway. It has been shown that the subcellular localization of FoxO transcription factors is controlled by insulin through Akt-dependent phosphorylation (Nakae, Kitamura et al. 2001; Dong, Copps et al. 2008). The molecular mechanisms that integrate impaired insulin effects to unconstrained gluconeogenesis in type 2 diabetic subjects remains ill defined. Although it had been suggested that FoxO1 is the major executer for regulation of insulin signaling-mediated glucose homeostasis (Gross, van den Heuvel et al. 2008; van der Vos and Coffer 2011) as FoxO1 has been repeatedly described in regulation of hepatic glucose production in the liver (Matsumoto, Han et al. 2006; Zhang, Patil et al. 2006; Matsumoto, Pocai et al. 2007; Zhang, Li et al. 2012). It has been reported in insulin-resistant mice that reducing or antagonizing hepatic FoxO1 activity markedly improves insulin sensitivity and glucose tolerance (Altomonte, Richter et al. 2003; Puigserver, Rhee et al. 2003; Matsumoto, Han et al. 2006; Samuel, Choi et al. 2006; Matsumoto, Pocai et al. 2007; Dong, Copps et al. 2008). In addition to FoxO1, FoxO3 is also significantly expressed in the liver, its contribution to metabolic regulation has however not been thoroughly analysed.

In this study we established and analysed a novel gain-of-function model of hepatic FOXO3 where the constitutively active FOXO3CA transgene is expressed under the control of the LAP promoter system that is specifically active in hepatocytes (LAP-tTA x luciferase-(tetO) 7-FOXO3CA). This mouse model offers a defined genetic system to gain a better understanding for the role of hepatic FOXO3 in regulating glucose homeostasis and lipid metabolism. Previously it was found that FOXO3 genotypes

102 Discussion are associated with insulin sensitivity phenotypes and longevity in human, mice and invertebrates demonstrating that FOXO3 is critical for metabolic control (Willcox, Donlon et al. 2008; Flachsbart, Caliebe et al. 2009; Pawlikowska, Hu et al. 2009). Moreover recently various loss-of-function approaches of FoxO1, FoxO3, FoxO4 in liver have addressed the role of FoxO3 in lipid metabolism (Estall 2012) but the role of liver-specific gain-of-function of FOXO3 in the context of modulating glucose and lipid metabolism in a cell type and context dependent manner was not addressed untill now.

In this study we could demonstrate that constitutively active FoxO3 in the liver critically affects glucose homeostasis and lipid metabolism. Expression of FoxO3CA after birth resulted in perinatal lethality of 47% of FoxO3CAHep mice just within 4- weeks of FoxO3 activation, whereas the rest of 52% FoxO3CAHep mice survived but showed a smaller and severe hypoglycemic phenotype associated with moderate upregulation of autophagy associated genes. Interestingly expression of FoxO3CA in adult animals was sufficient to induce hyperglycemia in FoxO3CAHep mice at the age of 7-weeks through regulation of gluconeogenesis-associated genes and alteration of gene expression in glucose and lipid metabolism. Furthermore, we observed elevated insulin levels as well as impaired glucose tolerance followed by insulin sensitivity and activation of catabolic pathways including glycogenolysis and lipid catabolic pathways in the liver. However the disease progression at the age of 9- weeks in FoxO3CAHep mice was associated with hypoglycemia, hyperinsulinemia and pancreatic hyperplasia. Strikingly, the anti-diabetic drug metformin completely normalized glucose, insulin as well as lipid metabolism in FoxO3CAHep mice.

An important role of insulin signalling in regulating glucose homeostasis and systemic growth has been shown previously in various experimental paradigms (Accili, Drago et al. 1996; Dong, Park et al. 2006). FoxO transcription factors are the potential mediators of hepatic gucose production downstream of insulin signalling and deregulated during insulin resistance conditions (Eijkelenboom and Burgering 2013). In our mouse model FoxO3CA expression just after birth result in perinatal lethality of FoxO3CAHep mice with in 4-weeks of FoxO3 activation. The rest of FoxO3CAHep animal analysed were severely smaller, sick and showed hypoglycemic phenotype. It was previously reported that liver-specific deletion of Irs1/Irs2 results in 103 Discussion hyperglycemia, hyperinsulinemia and smaller mice. In this mouse model it was further evidenced that hepatic deletion of FoxO1 in triple knockout mice (TKO) reverses the phenotype. This study pointed to the role of FoxO1 in directly regulating the genes involved in glucose metabolism and indirectly inhibiting growth hormone regulated genes (Dong, Copps et al. 2008). Additionally the smaller size of liver was reported in liver-specific insulin receptor knockout (LIRKO)(Michael, Kulkarni et al. 2000) mice and in liver-specific deletion of catalytic subunit of PI3K (p110α) L- p110α KO mice (Sopasakis, Liu et al. 2010) and was associated with defects in growth signalling pathways. It is still not known whether FoxO1 and FoxO3 share similar target genes or upregulate specific genes. In our mouse model it might be possible that growth-signalling pathways were suppressed due to overactivation of FoxO3 but this needs further investigation.

However it has been recently reported that expression of FoxO3CA in cardiomyocytes resulted in a small heart phenotype and this was associated with cardiomyocyte atrophy and activation of autophagy (Schips et al., 2011). In line with this finding we also observed moderate upregulation of autophagy-associated genes (e.g Atg5, Atg12, Atg1) in FoxO3CAHep mice. Additionaly hepatocytes were much smaller and appeared atrophic in FoxO3CAHep mice. As autophagy is a lysosomal degradation and self-destruction process, we could assume that the hypoglycemic phenotype in FoxO3CAHep mice at the age of 4-weeks might be due to activation of catabolic pathways like autophagy. Due to the severity of the phenotype at the age of 4-weeks in FoxO3CAHep mice most of the parameters could not be investigated and this prompted us to analyse the animals at adult age.

When FoxO3CAHep mice were analysed at the age of 7-weeks, we observed a hyperglycemic phenotype due to enhanced gluconeogenesis. This was followed by hyperinsulinemia, impaired glucose and insulin tolerance, consistent with studies showing constitutive activation of FoxO1 in the liver (Nakae, Biggs et al. 2002; Qu, Altomonte et al. 2006; Zhang, Patil et al. 2006). However, our data suggest that the interplay between different FoxO family members could be critical for the fine-tuning of glucose metabolism. Previous work demonstrated that expression of a gain-of- function allele of FoxO1 (FoxO1ADA) in the mouse liver paradoxically induced hypoglycemia and increased insulin sensitivity through its increased ability to 104 Discussion phosphorylate Akt in Trb3 (tribbles homologue 3)-dependent or independent manner (Matsumoto, Han et al. 2006). In addition, deletion of FoxO1 in hepatocytes did not alter the expression of gluconeogenic enzymes during fasting or after feeding (Lu, Wan et al. 2012). Finally, additional members of the FoxO family might be involved in regulating liver glucose homeostasis as hepatic activation of FoxO6 also induces hyperglycemia in mice (Kim, Perdomo et al. 2011; Kim, Zhang et al. 2013). The discrepencies observed in these studies with respect to insulin sensitivity might be due to variation in experimental animal models, level and duration of overexpression, as well as time of evaluation with respect to feeding versus fasting. Alternatively, these data might be difficult to understand because different mutated forms of FoxO may behave differently in unexpected ways. In contrast to these studies synergistic contributions of FOXO transcription factors was also evidenced (Haeusler, Kaestner et al. 2010) (Estall 2012; Zhang, Li et al. 2012) (Xiong, Tao et al. 2013) although peviously no glucose metabolic defects have been outlined in either FoxO3-/- or FoxO4-/- mice (Hosaka, Biggs et al. 2004). It had been shown that liver-specific deletion of FoxO1, FoxO3, FoxO4 causes more profound fasting hypoglycemia, improved insulin sensitivity and enhanced glucose tolerance with reduced plasma insulin levels and this study presented an important role of FoxO1 for regulating glucose metabolism and FoxO3 regulating lipid metabolism. However, the authors claim whether it is achieved in a synergistic or coordinated pattern is still far from being understood and remains largely open and the relative contributions of individual FoxO transcription factors was not described. Accordingly, the FoxO3CAHep mouse model is advantageous and a unique model for studying the metabolic disease pathogenesis with respect to tissue specific gain-of function of FOXO3 in the liver.

Our mice show substantial hyperinsulinemia. The normal action of insulin in the liver involves an inactivation of FoxO transcription factors. However, this pathway is blunted in our mouse model. Therefore, the hyperglycemic phenotype predominates. Our data suggest that FoxO3 in hepatocytes could be the central effector downstream of the insulin receptor/PI3K/Akt axis. This conclusion is supported by the fact that the FoxO3CAHep mice among the FoxO gain-of-function studies show the phenotype closest related to the mouse model of hepatic insulin receptor deficiency. Both the studies are consistent in terms of FoxO3CAHep and liver-specific insulin 105 Discussion receptor knock out mice (LIRKO) producing hyperglycemia due to increased hepatic glucose production and increased expression of gluconeogenic target genes G6pc, Pepck followed by marked hyperinsulinemia, impaired glucose and insulin tolerance. These effects in LIRKO mice could arise due to deficiency of hepatic insulin signalling resulting in insulin resistance and diabetes. In our mouse model our phenotype of FoxO3CAHep mice also mimics several insulin resistance phenotypes such as liver- specific deletion of Irs1 and Irs2 (DKO-mice) (Dong, Copps et al. 2008), liver-specific deletion of Akt1 and Akt2 (DLKO) mice (Lu, Wan et al. 2012), and liver-specific deletion of catalytic subunit of PI3K (p110α) L-p110α KO mice (Sopasakis, Liu et al. 2010) develop hyperglycemia due to unbridled gluconeogenesis, hyperinsulinemia followed by impaired glucose and impaired insulin tolerance. All these studies points to defective insulin signalling culminating in insulin resistance and type 2 diabetes.

The metabolic system reacts to environmental conditions by storing or utilizing nutrients. For example lack of metabolites in the glucose metabolic pathway will stimulate catabolic pathways like gluconeogenesis, glycogenolysis or lipid β-oxidation (Sun, Miller et al. 2012). Consistently, in our study we observed that continuous glucose production by FoxO3CA expression in FoxO3CAHep mice led to inhibition of lipid synthesis and to activation of glycogenolysis and lipid catabolic pathways. We saw an inhibition of the fatty acid synthesis pathway and activation of lipid β-oxidation most likely in order to provide energy and metabolites for gluconeogenesis.

A previous study with over-expression of FoxO1 (FoxO1ADA) in the liver observed improved insulin sensitivity, lipid accumulation and steatosis in the liver, with reduced fatty acid oxidation which is quite distinct from our model. The lipid catabolic pathway was actually inhibited in mice with FoxO1 over-expression (Matsumoto, Han et al. 2006). In this study the increase in insulin sensitivity and lipid accumulation (Aoyama, Peters et al. 1998) was associated with increase FoxO1 activity to suppress pseudokinase tribble 3 (Trb3). Trb3 is a modulator of Akt activity and it has been shown that Trb3 when it binds to Akt and prevents insulin-induced phosphorylation of Akt (Du, Herzig et al. 2003). The suppression of Trb3 was not dependent on FoxO1 but FoxO1 may interact with the PPARγ coactivator 1α(Pgc1 α) a well-known modulator of Trb3. However, the importance of (Pgc1α) has been previously reported in disorders of lipid metabolism in the liver in several mouse models (Leone, 106 Discussion

Weinheimer et al. 1999; Hashimoto, Cook et al. 2000). In our mouse model, we obeserved an increase in lipid and glucose catabolic pathway a possible explanation might be that critical regulators of glucose and lipid metabolism such as Pgc1α (Li, Monks et al. 2007) are differently affected in FoxO1 and FoxO3 gain-of-function models. In line with this using a similar gain-of function model Qu et al. (Qu, Altomonte et al. 2006) have observed enhanced lipogenesis and liver steatosis due to increased FoxO1 activity resulted in upregulation of Pgc-1α, fatty acid synthase and acetyl CoA carboxylase expression. The inconsistency oberved with respect to improved insulin sensitivity and increased lipogenesis can be due to variation in experimental animal models and experimental settings. However there are only few gene expression studies comparing target genes of different FoxO members (Eijkelenboom and Burgering 2013), and further investigation of specific gene regulation by individual FoxO members also in the context of lipid metabolism is required.

We have seen a moderate reduction in triglycerides and cholesterol levels in FoxO3CAHep mice. In line with this (Zhang, Patil et al. 2006), liver-specific FoxO1 activation has resulted in decreased triglycerides in transgenic mice which were due to decreased expression of genes involved in glycolysis and lipid synthesis resulting in reduced triglyceride concentrations in transgenic mice (Zhang, Patil et al. 2006). In our mouse model in FoxO3CAHep mice we also have observed downregulation of genes involved in glycolysis, such as glucokinase and pyruvate kinase are downregulated as well as lipid synthesis genes ascertained by microarray analysis and quantitative PCR studies. Consistently, the reduced triglycerides in liver-specific Irs1/ Irs2 knockout mice are associated with decreased SREBP-1c expression (Dong, Copps et al. 2008). Furthermore the decreased triglycerides in liver-specific deletion of catalytic subunit of PI3K (p110α) L-p110α KO mice (Sopasakis, Liu et al. 2010) are also linked to decreased lipogenic gene expression. Another possibility for reduced triglycerides in FoxO3CAHep mice might be due to a decreased effect of insulin in the liver to promote triglyceride synthesis coupled with suppression of extrahepatic lipolysis due to high insulin levels in FoxO3CAHep mice as reported previously in LIRKO mice (Michael, Kulkarni et al. 2000).

107 Discussion

Additionally, expression of FoxO3CA in the liver led to loss of glycogen storage in FoxO3CAHep mice, a phenotype also observed in insulin resistant LIRKO mice (Michael, Kulkarni et al. 2000), liver-specific Irs1/ Irs2 knockout mice (Dong, Copps et al. 2008) and in liver-specific FoxO1 gain-of-function mice (Qu, Altomonte et al. 2006). The normal action of insulin in the liver is to suppress glycogenolysis and gluconeogenesis. Loss of glycogen in these studies is associated with impaired insulin signalling during insulin resistant states. However, it has also been reported in subjects of type 2 diabetes that the increase in hepatic glucose production occurs by enhanced gluconeogenesis and enhanced glycogenolysis (DeFronzo, Bonadonna et al. 1992). Consistently, in our mouse model we have observed an increase in mRNA- level of glycogen phosphorylase an important enzyme for glycogenolysis showing that both gluconeogenesis and glycogenolysis are activated to produce more glucose production in FoxO3CAHep mice as reported recently in diabetic db/db mouse liver (Zhang, Xu et al. 2013).

Our gene expression profile ascertained by microarray analysis in FoxO3CAHep mice showed a global metabolic profile with the regulation of several genes involved in glucose and lipid metabolism. The most upregulated gene was leptin receptor (Lepr). It has been shown in the liver that proper insulin signalling plays a critical role in expression and clearence of leptin receptor as well as fine-tuning of leptin action and leptin homeostasis. The upregulation of Lepr during insulin resistance states was reported by (Cohen, Kokkotou et al. 2007) in liver-specific insulin receptor knockout mice LIRKO mice (Michael, Kulkarni et al. 2000). Additionally the increase in leptin receptor expression was also reported in liver-specific deletion of the catalytic subunit of PI3K (p110α) L-p110α KO mice (Sopasakis, Liu et al. 2010) and was due to insulin resistance phenotype. Consistently, in our mouse model we can postulate that increased expression of Lepr is correlated with insulin resistance in FoxO3CAHep mice.

In diabetes, the increased rate of hepatic gluconeogenesis has been shown previously (DeFronzo, Bonadonna et al. 1992; Pilkis and Granner 1992; Samuel and Shulman 2012). Consistent with the previous reports, genes involved in promoting hepatic gluconeogenesis such as pyruvate dehydrogenase kinase-4 (Pdk4) were upregulated and glycolytic genes such as Glucokinase (Gck), and pyruvate kinase 108 Discussion liver (Pklr) were downregulated as reported before in liver-specific gain-of-function of FoxO1 and in type 2 diabetic stated in human liver (Pilkis and Granner 1992; Zhang, Patil et al. 2006). Interestingly genes providing substrates for gluconeogenesis are increased such as aquaporin7 as shown previously (Zhang, Patil et al. 2006). Consistently due to enhanced activation of lipid catabolic pathways in FoxO3CAHep mice we have observed upregulation of genes involved in lipid transport and trafficking (fabp5, fabp7, Apom, Atp11) and genes involved in fatty acid oxidation (cpt1b, ehhadh, Acot2, Acad12). Conversly genes involved in fatty acid synthesis were downregulated (Acly, Fasn, Acc, Acsm1). Moreover our microarray analyses reveal that the livers from FoxO3CAHep mice are accelerated towards uncontrolled gluconeogenesis by inhibiting the lipid synthesis pathway and activation of lipid β- oxidation. Interestingly, our gene expression profiles are very similar to microarray analysis of recently reported 9-week-old type 2 diabetic db/db mouse liver (Zhang, Xu et al. 2013).

Interestingly, when we evaluated the disease progression by analysing the FoxO3CAHep mice at the age of 9-weeks, surprisingly FoxO3CAHep mice showed fasting hypoglycemia and higher insulin levels. These findings were in contrast to our previous findings where FoxO3CAHep mice showed fasting hyperglycemia at the age of 7-weeks. What could be the reason for this unanticipated hypoglycemic response? It could be due to high insulin secretion from pancreatic β-cells as the insulin levels were elevated in FoxO3CAHep mice at the age of 9-weeks. Similarly, it was previously reported in liver-specific LIRKO mouse model that LIRKO mice showed progressive decline in blood glucose at 6 months of age, LIRKO mice showed fasting hypoglycemia that resembles the hypoglycemic phenotype of FoxO3CAHep mice at the age of 9 weeks. The hypoglycemic effect could be due to the development of acquired liver failure or plausibly imitating the reduced sufficiency of livers for gluconeogenesis. When FoxO3CAHep mice were analysed for gluconeogenesis they showed reduced effect of gluconeogenesis. Additionally, another possibility for fasting hypoglycemia is that it might arise due to less insulin clearence capacity in the liver as observed in LIRKO mice. Hepatocytes are the major place for insulin clearence and hepatocytes in liver failure probably are not able to clean up insulin properly. Furthermore it was reported in human patients that any excess of endogenous or exogenous insulin could cause hypoglycemia in diabetes (Cryer,

109 Discussion

Davis et al. 2003; De Leon and Stanley 2013).

Strikingly in our mouse model we have demonstrated pancreatic hyperplasia at the age of 9-weeks in FoxO3CAHep mice and this finding is similar to pancreatic hyperplasia of insulin-resistant LIRKO mouse model (Michael, Kulkarni et al. 2000). It could arise as a compensatory mechanism in response to increased insulin levels and impaired insulin clearence. However, the mechanisms controlling proliferation, expansion of β-cells and its adaptation to insulin resistance are incompletely defined although it was evidenced that massive bulk of new β-cells in both rodent and humans develop by simple self-duplication of preexisting β-cells (Dor, Brown et al. 2004; Teta, Rankin et al. 2007; Meier, Butler et al. 2008). It is known that pancreatic β-cells preserve the potential to raise their replication rate in reply to physiological challenges, including hyperglycemia (Alonso, Yokoe et al. 2007), pancreatic injury (Nir, Melton et al. 2007; Cano, Rulifson et al. 2008) insulin resistance (Bruning, Winnay et al. 1997; Pick, Clark et al. 1998; Michael, Kulkarni et al. 2000; Kulkarni, Jhala et al. 2004) and gestation (Parsons, Brelje et al. 1992; Rieck, White et al. 2009). It has been reported that circulating and systemic factors can influence β-cell mass and replication. Importantly glucose infusion in mice induces β-cell replication showing glucose itself is a β-cell mitogen (Bonner-Weir, Deery et al. 1989; Bernard, Thibault et al. 1998; Alonso, Yokoe et al. 2007). In addition to the factors described above several hormones including prolactin, placental lactogen, insulin, glucagons- like peptide 1, and glucose-dependent insulinotropic polypeptide also play a critical role in regulation of β-cell mass (Parsons, Brelje et al. 1992; Bernard, Thibault et al. 1998; Paris, Bernard-Kargar et al. 2003; Drucker 2006; Sachdeva and Stoffers 2009). The role of circulating islet cell growth factor during insulin resistance condition independent of obesity and glucose has also been shown in transplanted mice (Flier, Kulkarni et al. 2001).

Specifically how the liver signals pancreatic β-cells to proliferate is unknown, but recently using LIRKO mouse model it was demonstrated that non-cell-autonomous, nonneural, humoral factors causes pancreatic hyperplasia (El Ouaamari, Kawamori et al. 2013). This study showed that the liver-derived systemic factors trigger human and mouse β-cell proliferation and further showed that the liver serves as a source for

110 Discussion

β-cell growth factors in response to insulin resistance. In line with this study recently betatrophin a hormone that controls pancreatic β-cell proliferation and β-cell mass expansion in insulin resistant mouse model was identified (Yi, Park et al. 2013). This study critically showed the role of betatrophin produced in mouse liver inducing β-cell proliferation, β-cell mass expansion and improving glucose tolerance and further pointing to the therapeutic importance of betatrophin during insulin resistance conditions. Similarly constitutive activation of mitogen- activated protein kinase / ERK kinase (MEK-1) in the mouse liver induces pancreatic β-cell replication and enhances glucose tolerance showing that the increase in hyperinsulinemia and compensatory pancreatic β-cell mass is multifactorial and requires the central nervous system (CNS) presenting an inter-organ communication system controlling pancreatic β-cell proliferation (Imai, Katagiri et al. 2008). Recently the role of hepatocyte growth factor (HGF) in determining increase in islet cell mass along with hyperinsulinemia in mouse models of insulin-resistance was reported (Araujo, Oliveira et al. 2012).

However in our mouse model it might be that glucose and insulin itself are potential candidates for the observed pancreatic hyperplasia as reported before (Bonner-Weir, Deery et al. 1989; Assmann, Hinault et al. 2009; Assmann, Ueki et al. 2009). However, we cannot exclude the possibility of liver-derived unknown factors inducing β-cell hyperplasia or betatrophin like hormone contribution augmenting β-cell hyperplasia in FoxO3CAHep mice at the age of 9 weeks. Importantly the expression of betatrophin was not affected in our gene expression analyses. Therefore further investigation is needed to identify specific β-cell growth factor. Indeed our mouse model may serve as a tool for identifying yet unknown liver-derived factors/ hormones augmenting β-cell hyperplasia in insulin resistance states.

Since decades, several types of anti-diabetic drugs have been generated. Among those, biguanides including phenformin and metformin function by inhibition of glucose production (Stumvoll, 1995). Especially metformin is currently the most prescribed drug for type 2 diabetes. In our mouse model metformin administration completely normalizes glucose level with marked decrease in expression of gluconeogenic target genes in FoxO3CAHep mice. It has been reported that metformin primarily functions to decrease hepatic glucose production by attenuating hepatic gluconeogenesis both in diabetic mouse models (Mithieux, Guignot et al. 2002; 111 Discussion

Shaw, Lamia et al. 2005; Kim, Park et al. 2008; He, Sabet et al. 2009; Hou, Venier et al. 2010; Tahara, Matsuyama-Yokono et al. 2011) and in humans with type 2 diabetes (Davidson and Peters 1997; Hundal, Krssak et al. 2000; Natali and Ferrannini 2006). Additionally metformin administration in our mouse model normalizes insulin levels in FoxO3CAHep mice both at the age of 7-week as well as at 9-weeks. Metformin is the best insulin sensitizer during insulin resistance conditions and it improves insulin sensitivity due to its positive effects on insulin receptor tyrosine kinase activity in streptozotosin-induced diabetic rats (Rossetti, DeFronzo et al. 1990) and it increases the activation of insulin receptor specifically insulin receptor substrate 2 (Irs2) activation in human liver (Gunton, Delhanty et al. 2003). In our mouse model we have seen downregulation of Irs2 in gene expression analysis after metformin treatment and actually the normalization of insulin levels in our mouse model is associated with normalization of glucose levels after metformin treatment as evidenced in several studies (Hou, Venier et al. 2010; Martin-Montalvo, Mercken et al. 2013). We have also seen the normalization of pancreatic islet morphology in FoxO3CAHep mice at the age of 9-weeks, as it has been already known that metformin has no immediate effect on the function of pancreatic beta cells. Fasting and feeding insulin levels are frequently reduced after metformin treatment in patients and in mice with diabetes showing the normal compensatory response of the pancreas to enhanced insulin sensitivity (Klip and Leiter 1990; Rossetti, DeFronzo et al. 1990; DeFronzo, Barzilai et al. 1991; Johnson, Webster et al. 1993; Stumvoll, Nurjhan et al. 1995; Cusi, Consoli et al. 1996). From these studies it can be inferred that in our mouse model the normalization of pancreatic islets after metfromin treatment in FoxO3CAHep mice at the age of 9-weeks is associated with normal insulin levels and improved insulin sensitivity.

Despite the wide acceptance of metformin as first-line therapy, the molecular mechanisms of action remain incompletely understood. It was suggested that metformin reduces glucose levels via activating adenosine mono-phosphate (AMP)- activated protein kinase (AMPK) and its upstream kinase serine/threonine kinase 11 (STK11) also known as liver kinase B1 (LKB1) both in vivo and in vitro (Zhou, Myers et al. 2001; Shaw, Lamia et al. 2005; Kim, Park et al. 2008; Lee, Seo et al. 2010). Interestingly, in vivo mouse models with liver-specific AMPK-deficient mice (AMPKα1α2-/-) and liver-specific LKB1-deficient mice (Lkb1-KO) show no 112 Discussion influence on metformin-dependent reduction in blood glucose levels (Foretz, Hebrard et al. 2010). We also assume that metformin effects in our model were AMPK- independent. We did not see significant changes of amount and phosphorylation of AMPK in FoxO3CAHep mice. Furthermore, as AMPK is known to activate FoxO family members metformin-induced activation of AMPK should actually result in phosphorylation and activation of FoxO3 (Greer, Oskoui et al. 2007). Therefore it could be reasonable that AMPK activity is suppressed by active FoxO3. This could explain why independent of metformin treatment FoxO3CAHep mice displayed lower AMPK activity. A recent study suggested that metformin suppresses hepatic glucagon signaling by decreasing production of cyclic AMP and PKA activity in an AMPK-dependent manner (Miller, Chu et al. 2013). PKA signaling activates the transcription factor CREB that regulates target gene expression for glucose and lipid metabolism together with CREB-binding protein (CBP). In addition, metformin was shown to inactivate CBP by phosphorylation (He, Sabet et al. 2009). Given that CBP is a general transcription co-regulator, this axis of metformin action might affect FoxO3-dependent transcription. However, direct effects of metformin on FoxO3 activity are more likely, as modulation of CBP should result in a much more generalized alteration of gene expression.

Additionally, it has also been reported that metformin-induced activation of AMPK lowers hepatic gluconeogenesis in livers from B6-Lepob/ob mice through modulation of orphan nuclear receptor small heterodimer partner (SHP) (Kim, Park et al. 2008). The role of SHP in repressing the transcriptional activities of hepatic nuclear factor 4- α (HNF4α), FoxO1 and a variety of nuclear receptors that regulate the gluconeogenic programme has been reported (Lee and Moore 2002; Kim, Kim et al. 2004; Yamagata, Daitoku et al. 2004). SHP also interacts with the CREB-CBP complex and CRTC2 to attenuate gluconeogenic gene expresssion (Lee, Seo et al. 2010). In our mouse model it is possible that SHP might interact with FoxO3 or other transcriptional regulators to lower gluconeogenesis. Nevertheless we performed various experiments to assess the level of SHP activation but we have not found any striking results one plausibility could be an unknown exciting mechanism to attenuate distorted glucose and lipid metabolism in FoxO3CAHep mice after metformin treatment and this mechanism is on the way of further exploration.

113 Discussion

We observed complete normalization in lipid metabolism, as well as hepatocyte morphology after metformin administration in FoxO3CAHep mice. Metformin-mediated re-normalization of lipid metabolism is probably a consequence of normalization in glucose metabolism. Previously it was shown that metformin reverses fatty liver disease in leptin-deficient ob/ob mice (Lin et al., 2000) and in rat fatty liver induced by high fat feeding (Gao, Lu et al. 2005; Zhang, Zhang et al. 2006), suggesting that metformin is beneficial in improving lipid metabolism. In line with this it has also been recently reported that metformin reduces the fat content of the liver in obese overweight women (Sanchez-Munoz, Salas-Romero et al. 2013). The gene expression analysis after metformin treatment in FoxO3CAHep mice showed complete normalization of all the genes involved in lipid and glucose metabolism. The possible explanations for normalization of genes after metformin treatment in FoxO3CAHep mice are correlated with normalization of glucose metabolism. Given that our model depends on the expression of a transcription factor that was rendered non- responsive to insulin signalling suggests that this terminal step is crucially targeted by metformin. In summary, our study provides new insights into the role of FoxO3 in glucose and lipid metabolism. Furthermore, our data suggest that FoxO3 could be an immediate target of metformin action.

Conclusions

In the present study we demonstrate that constitutive activation of FoxO3 results in impaired glucose and lipid metabolism in transgenic mice. FoxO3 activation resulted in impaired glucose tolerance due to upregulation of gluconeogenic target genes followed by hyperinsulinemia and impaired insulin sensitivity. Moreover expression of constitutively active FoxO3 results in activation of catabolic pathways including glycogenolysis and lipid catabolic pathways in the liver. On the other hand FoxO3 inhibits lipid synthesis pathways. Consistently our gene expression analysis has identified several genes important for regulation of glucose and lipid catabolic pathways as reported in type 2 diabetes conditions. However the disease progression in transgenic mice resulted in hypoglycemia might be as a consequence of beginning of liver failure, hyperinsulinemia and pancreatic hyperplasia. Strikingly the administration of metformin rapidly and completely normalizes glucose, insulin as

114 Discussion well as lipid metabolism in mice expressing constitutively active FoxO3 (Figure 42). Our findings identify FoxO3 as a critical metabolic regulator and a potential hepatic target of metformin. Taken together the results obtained from constitutive activation of FoxO3 in the liver imply that dysbalanced FoxO3 activity in the liver augments insulin resistance and type 2 diabetes. Therefore tight regulation of FoxO3 activity would be very crucial and agents controlling hepatic FoxO3 system may represent therapeutic tools to prevent type 2 diabetes.

Figure 42: Effects of liver-specific FoxO3 activation in FoxO3CAHep mice. Expression of constitutively active FoxO3 leads to hyperglycemia due to up-regulation of gluconeogenic target genes followed by impaired glucose and insulin tolerance and furthermore resulted in activation of catabolic pathways including enhanced glycogenolysis and decresed lipid synthesis. The disease progression in transgenic mice resulted in hypoglycemia, hyperinsulinemia and pancreatic hyperplasia. The administration of metformin for 3 days and 3 weeks completely normalizes the distorted glucose and lipid metabolism in FoxO3CAHep mice.

115 References

5 References

Accili, D. and K. C. Arden (2004). "FoxOs at the crossroads of cellular metabolism, differentiation, and transformation." Cell 117(4): 421-6. Accili, D., J. Drago, et al. (1996). "Early neonatal death in mice homozygous for a null allele of the insulin receptor gene." Nat Genet 12(1): 106-9. Adler, A. I., E. J. Shaw, et al. (2009). "Newer agents for blood glucose control in type 2 diabetes: summary of NICE guidance." BMJ 338: b1668. Agius, L. (2008). "Glucokinase and molecular aspects of liver glycogen metabolism." Biochem J 414(1): 1-18. Alonso, L. C., T. Yokoe, et al. (2007). "Glucose infusion in mice: a new model to induce beta-cell replication." Diabetes 56(7): 1792-801. Altomonte, J., L. Cong, et al. (2004). "Foxo1 mediates insulin action on apoC-III and triglyceride metabolism." J Clin Invest 114(10): 1493-503. Altomonte, J., A. Richter, et al. (2003). "Inhibition of Foxo1 function is associated with improved fasting glycemia in diabetic mice." Am J Physiol Endocrinol Metab 285(4): E718-28. Aoyama, T., J. M. Peters, et al. (1998). "Altered constitutive expression of fatty acid- metabolizing enzymes in mice lacking the peroxisome proliferator-activated receptor alpha (PPARalpha)." J Biol Chem 273(10): 5678-84. Araujo, T. G., A. G. Oliveira, et al. (2012). "Hepatocyte growth factor plays a key role in insulin resistance-associated compensatory mechanisms." Endocrinology 153(12): 5760-9. Argaud, D., H. Roth, et al. (1993). "Metformin decreases gluconeogenesis by enhancing the pyruvate kinase flux in isolated rat hepatocytes." Eur J Biochem 213(3): 1341-8. Ashcroft, F. M. and P. Rorsman (2012). "Diabetes mellitus and the beta cell: the last ten years." Cell 148(6): 1160-71. Assmann, A., C. Hinault, et al. (2009). "Growth factor control of pancreatic islet regeneration and function." Pediatr Diabetes 10(1): 14-32. Assmann, A., K. Ueki, et al. (2009). "Glucose effects on beta-cell growth and survival require activation of insulin receptors and insulin receptor substrate 2." Mol Cell Biol 29(11): 3219-28.

116 References

Authier, F. and B. Desbuquois (2008). "Glucagon receptors." Cell Mol Life Sci 65(12): 1880-99. Balani, J., S. L. Hyer, et al. (2009). "Pregnancy outcomes in women with gestational diabetes treated with metformin or insulin: a case-control study." Diabet Med 26(8): 798-802. Bar, R. S., L. C. Harrison, et al. (1979). "Regulation of insulin receptors in normal and abnormal physiology in humans." Adv Intern Med 24: 23-52. Barthel, A., D. Schmoll, et al. (2005). "FoxO proteins in insulin action and metabolism." Trends Endocrinol Metab 16(4): 183-9. Bechmann, L. P., R. A. Hannivoort, et al. (2012). "The interaction of hepatic lipid and glucose metabolism in liver diseases." J Hepatol 56(4): 952-64. Ben Sahra, I., Y. Le Marchand-Brustel, et al. (2010). "Metformin in cancer therapy: a new perspective for an old antidiabetic drug?" Mol Cancer Ther 9(5): 1092-9. Bergman, R. N. (2005). "Minimal model: perspective from 2005." Horm Res 64 Suppl 3: 8-15. Bernard, C., C. Thibault, et al. (1998). "Pancreatic beta-cell regeneration after 48-h glucose infusion in mildly diabetic rats is not correlated with functional improvement." Diabetes 47(7): 1058-65. Bettermann, K., M. Vucur, et al. (2010). "TAK1 suppresses a NEMO-dependent but NF-kappaB-independent pathway to liver cancer." Cancer Cell 17(5): 481-96. Biddinger, S. B. and C. R. Kahn (2006). "From mice to men: insights into the insulin resistance syndromes." Annu Rev Physiol 68: 123-58. Bonnefond, A., P. Froguel, et al. (2010). "The emerging genetics of type 2 diabetes." Trends Mol Med 16(9): 407-16. Bonner-Weir, S., D. Deery, et al. (1989). "Compensatory growth of pancreatic beta- cells in adult rats after short-term glucose infusion." Diabetes 38(1): 49-53. Borkhardt, A., R. Repp, et al. (1997). "Cloning and characterization of AFX, the gene that fuses to MLL in acute leukemias with a t(X;11)(q13;q23)." Oncogene 14(2): 195-202. Bouche, C., S. Serdy, et al. (2004). "The cellular fate of glucose and its relevance in type 2 diabetes." Endocr Rev 25(5): 807-30. Brunet, A. (2004). "[The multiple roles of FOXO transcription factors]." Med Sci (Paris) 20(10): 856-9.

117 References

Brunet, A., A. Bonni, et al. (1999). "Akt promotes cell survival by phosphorylating and inhibiting a Forkhead transcription factor." Cell 96(6): 857-68. Brunet, A., F. Kanai, et al. (2002). "14-3-3 transits to the nucleus and participates in dynamic nucleocytoplasmic transport." J Cell Biol 156(5): 817-28. Brunet, A., L. B. Sweeney, et al. (2004). "Stress-dependent regulation of FOXO transcription factors by the SIRT1 deacetylase." Science 303(5666): 2011-5. Bruning, J. C., J. Winnay, et al. (1997). "Development of a novel polygenic model of NIDDM in mice heterozygous for IR and IRS-1 null alleles." Cell 88(4): 561-72. Buteau, J. and D. Accili (2007). "Regulation of pancreatic beta-cell function by the forkhead protein FoxO1." Diabetes Obes Metab 9 Suppl 2: 140-6. Butler, A. E., J. Janson, et al. (2003). "Beta-cell deficit and increased beta-cell apoptosis in humans with type 2 diabetes." Diabetes 52(1): 102-10. Calnan, D. R. and A. Brunet (2008). "The FoxO code." Oncogene 27(16): 2276-88. Cano, D. A., I. C. Rulifson, et al. (2008). "Regulated beta-cell regeneration in the adult mouse pancreas." Diabetes 57(4): 958-66. Carling, D. (2004). "The AMP-activated protein kinase cascade--a unifying system for energy control." Trends Biochem Sci 29(1): 18-24. Chatzigeorgiou, A., A. Halapas, et al. (2009). "The use of animal models in the study of diabetes mellitus." In Vivo 23(2): 245-58. Cheng, Z., S. Guo, et al. (2009). "Foxo1 integrates insulin signaling with mitochondrial function in the liver." Nat Med 15(11): 1307-11. Cho, H., J. Mu, et al. (2001). "Insulin resistance and a diabetes mellitus-like syndrome in mice lacking the protein kinase Akt2 (PKB beta)." Science 292(5522): 1728-31. Clement, S., U. Krause, et al. (2001). "The lipid phosphatase SHIP2 controls insulin sensitivity." Nature 409(6816): 92-7. Cohen, S. E., E. Kokkotou, et al. (2007). "High circulating leptin receptors with normal leptin sensitivity in liver-specific insulin receptor knock-out (LIRKO) mice." J Biol Chem 282(32): 23672-8. Cool, B., B. Zinker, et al. (2006). "Identification and characterization of a small molecule AMPK activator that treats key components of type 2 diabetes and the metabolic syndrome." Cell Metab 3(6): 403-16. Corton, J. M., J. G. Gillespie, et al. (1994). "Role of the AMP-activated protein kinase in the cellular stress response." Curr Biol 4(4): 315-24. 118 References

Craparo, A., R. Freund, et al. (1997). "14-3-3 (epsilon) interacts with the insulin-like growth factor I receptor and insulin receptor substrate I in a phosphoserine- dependent manner." J Biol Chem 272(17): 11663-9. Cryer, P. E., S. N. Davis, et al. (2003). "Hypoglycemia in diabetes." Diabetes Care 26(6): 1902-12. Cuesta-Munoz, A. L., T. Tuomi, et al. (2010). "Clinical heterogeneity in monogenic diabetes caused by mutations in the glucokinase gene (GCK-MODY)." Diabetes Care 33(2): 290-2. Cusi, K., A. Consoli, et al. (1996). "Metabolic effects of metformin on glucose and lactate metabolism in noninsulin-dependent diabetes mellitus." J Clin Endocrinol Metab 81(11): 4059-67. Daitoku, H., K. Yamagata, et al. (2003). "Regulation of PGC-1 promoter activity by and the forkhead transcription factor FKHR." Diabetes 52(3): 642-9. Davidson, M. B. and A. L. Peters (1997). "An overview of metformin in the treatment of type 2 diabetes mellitus." Am J Med 102(1): 99-110. Davis, R. J., C. M. D'Cruz, et al. (1994). "Fusion of PAX7 to FKHR by the variant t(1;13)(p36;q14) translocation in alveolar rhabdomyosarcoma." Cancer Res 54(11): 2869-72. De Leon, D. D. and C. A. Stanley (2013). "Determination of insulin for the diagnosis of hyperinsulinemic hypoglycemia." Best Pract Res Clin Endocrinol Metab 27(6): 763-9. Defronzo, R. A. (2009). "Banting Lecture. From the triumvirate to the ominous octet: a new paradigm for the treatment of type 2 diabetes mellitus." Diabetes 58(4): 773-95. DeFronzo, R. A., N. Barzilai, et al. (1991). "Mechanism of metformin action in obese and lean noninsulin-dependent diabetic subjects." J Clin Endocrinol Metab 73(6): 1294-301. DeFronzo, R. A., R. C. Bonadonna, et al. (1992). "Pathogenesis of NIDDM. A balanced overview." Diabetes Care 15(3): 318-68. Detaille, D., B. Guigas, et al. (2002). "Obligatory role of membrane events in the regulatory effect of metformin on the respiratory chain function." Biochem Pharmacol 63(7): 1259-72.

119 References

Donath, M. Y., E. Dalmas, et al. (2013). "Inflammation in obesity and diabetes: islet dysfunction and therapeutic opportunity." Cell Metab 17(6): 860-72. Donath, M. Y. and S. E. Shoelson (2011). "Type 2 diabetes as an inflammatory disease." Nat Rev Immunol 11(2): 98-107. Dong, X., S. Park, et al. (2006). "Irs1 and Irs2 signaling is essential for hepatic glucose homeostasis and systemic growth." J Clin Invest 116(1): 101-14. Dong, X. C., K. D. Copps, et al. (2008). "Inactivation of hepatic Foxo1 by insulin signaling is required for adaptive nutrient homeostasis and endocrine growth regulation." Cell Metab 8(1): 65-76. Dor, Y., J. Brown, et al. (2004). "Adult pancreatic beta-cells are formed by self- duplication rather than stem-cell differentiation." Nature 429(6987): 41-6. Drucker, D. J. (2006). "The biology of incretin hormones." Cell Metab 3(3): 153-65. Du, K., S. Herzig, et al. (2003). "TRB3: a tribbles homolog that inhibits Akt/PKB activation by insulin in liver." Science 300(5625): 1574-7. Durham, H. A. and G. E. Truett (2006). "Development of insulin resistance and hyperphagia in Zucker fatty rats." Am J Physiol Regul Integr Comp Physiol 290(3): R652-8. Eijkelenboom, A. and B. M. Burgering (2013 ). "FOXOs: signalling integrators for homeostasis maintenance." Nat Rev Mol Cell Biol 14(2): 83-97. El Ouaamari, A., D. Kawamori, et al. (2013). "Liver-derived systemic factors drive beta cell hyperplasia in insulin-resistant states." Cell Rep 3(2): 401-10. El-Mir, M. Y., V. Nogueira, et al. (2000). "Dimethylbiguanide inhibits cell respiration via an indirect effect targeted on the respiratory chain complex I." J Biol Chem 275(1): 223-8. Emanuelli, B., P. Peraldi, et al. (2000). "SOCS-3 is an insulin-induced negative regulator of insulin signaling." J Biol Chem 275(21): 15985-91. Estall, J. L. (2012). "The Foxo family: partners in crime or silent heroes." Endocrinology 153(2): 549-51. Evans, J. M., L. A. Donnelly, et al. (2005). "Metformin and reduced risk of cancer in diabetic patients." BMJ 330(7503): 1304-5. Flachsbart, F., A. Caliebe, et al. (2009). "Association of FOXO3A variation with human longevity confirmed in German centenarians." Proc Natl Acad Sci U S A 106(8): 2700-5. 120 References

Flier, S. N., R. N. Kulkarni, et al. (2001). "Evidence for a circulating islet cell growth factor in insulin-resistant states." Proc Natl Acad Sci U S A 98(13): 7475-80. Foretz, M., S. Hebrard, et al. (2010). "Metformin inhibits hepatic gluconeogenesis in mice independently of the LKB1/AMPK pathway via a decrease in hepatic energy state." J Clin Invest 120(7): 2355-69. Frescas, D., L. Valenti, et al. (2005). "Nuclear trapping of the forkhead transcription factor FoxO1 via Sirt-dependent deacetylation promotes expression of glucogenetic genes." J Biol Chem 280(21): 20589-95. Furuyama, T., K. Kitayama, et al. (2003). "Forkhead transcription factor FOXO1 (FKHR)-dependent induction of PDK4 gene expression in skeletal muscle during energy deprivation." Biochem J 375(Pt 2): 365-71. Galili, N., R. J. Davis, et al. (1993). "Fusion of a fork head domain gene to PAX3 in the solid tumour alveolar rhabdomyosarcoma." Nat Genet 5(3): 230-5. Gao, Z., D. Hwang, et al. (2002). "Serine phosphorylation of insulin receptor substrate 1 by inhibitor kappa B kinase complex." J Biol Chem 277(50): 48115-21. Gao, Z. Q., F. E. Lu, et al. (2005). "[Study on therapeutic effects of metformin on rat fatty livers induced by high fat feeding]." Zhonghua Gan Zang Bing Za Zhi 13(2): 101-4. Graham, G. G., J. Punt, et al. "Clinical pharmacokinetics of metformin." Clin Pharmacokinet 50(2): 81-98. Gray, S., B. Wang, et al. (2007). "Regulation of gluconeogenesis by Kruppel-like factor 15." Cell Metab 5(4): 305-12. Greer, E. L., M. R. Banko, et al. (2009). "AMP-activated protein kinase and FoxO transcription factors in dietary restriction-induced longevity." Ann N Y Acad Sci 1170: 688-92. Greer, E. L. and A. Brunet (2005). "FOXO transcription factors at the interface between longevity and tumor suppression." Oncogene 24(50): 7410-25. Greer, E. L., P. R. Oskoui, et al. (2007). "The energy sensor AMP-activated protein kinase directly regulates the mammalian FOXO3 transcription factor." J Biol Chem 282(41): 30107-19. Gross, D. N., A. P. van den Heuvel, et al. (2008). "The role of FoxO in the regulation of metabolism." Oncogene 27(16): 2320-36.

121 References

Gunton, J. E., P. J. Delhanty, et al. (2003). "Metformin rapidly increases insulin receptor activation in human liver and signals preferentially through insulin- receptor substrate-2." J Clin Endocrinol Metab 88(3): 1323-32. Guo, S., G. Rena, et al. (1999). "Phosphorylation of serine 256 by protein kinase B disrupts transactivation by FKHR and mediates effects of insulin on insulin-like growth factor-binding protein-1 promoter activity through a conserved insulin response sequence." J Biol Chem 274(24): 17184-92. Haeusler, R. A., K. H. Kaestner, et al. (2010). "FoxOs function synergistically to promote glucose production." J Biol Chem 285(46): 35245-8. Hall, R. K., T. Yamasaki, et al. (2000). "Regulation of phosphoenolpyruvate carboxykinase and insulin-like growth factor-binding protein-1 gene expression by insulin. The role of winged helix/forkhead proteins." J Biol Chem 275(39): 30169-75. Hanson, R. W. and L. Reshef (1997). "Regulation of phosphoenolpyruvate carboxykinase (GTP) gene expression." Annu Rev Biochem 66: 581-611. Hardie, D. G. (2004). "The AMP-activated protein kinase pathway--new players upstream and downstream." J Cell Sci 117(Pt 23): 5479-87. Hardie, D. G., J. W. Scott, et al. (2003). "Management of cellular energy by the AMP- activated protein kinase system." FEBS Lett 546(1): 113-20. Hashimoto, T., W. S. Cook, et al. (2000). "Defect in peroxisome proliferator-activated receptor alpha-inducible fatty acid oxidation determines the severity of hepatic steatosis in response to fasting." J Biol Chem 275(37): 28918-28. He, L., A. Sabet, et al. (2009). "Metformin and insulin suppress hepatic gluconeogenesis through phosphorylation of CREB binding protein." Cell 137(4): 635-46. Herzig, S., F. Long, et al. (2001). "CREB regulates hepatic gluconeogenesis through the coactivator PGC-1." Nature 413(6852): 179-83. Hirosumi, J., G. Tuncman, et al. (2002). "A central role for JNK in obesity and insulin resistance." Nature 420(6913): 333-6. Hosaka, T., W. H. Biggs, 3rd, et al. (2004). "Disruption of forkhead transcription factor (FOXO) family members in mice reveals their functional diversification." Proc Natl Acad Sci U S A 101(9): 2975-80. Hou, M., N. Venier, et al. (2010). "Protective effect of metformin in CD1 mice placed

122 References

on a high carbohydrate-high fat diet." Biochem Biophys Res Commun 397(3): 537-42. Housley, M. P., N. D. Udeshi, et al. (2009). "A PGC-1alpha-O-GlcNAc transferase complex regulates FoxO transcription factor activity in response to glucose." J Biol Chem 284(8): 5148-57. Huang, H. and D. J. Tindall (2007). "Dynamic FoxO transcription factors." J Cell Sci 120(Pt 15): 2479-87. Hundal, R. S., M. Krssak, et al. (2000). "Mechanism by which metformin reduces glucose production in type 2 diabetes." Diabetes 49(12): 2063-9. Hurst, R. T. and R. W. Lee (2003). "Increased incidence of coronary atherosclerosis in type 2 diabetes mellitus: mechanisms and management." Ann Intern Med 139(10): 824-34. Hwangbo, D. S., B. Gershman, et al. (2004). "Drosophila dFOXO controls lifespan and regulates insulin signalling in brain and fat body." Nature 429(6991): 562- 6. Ide, T., H. Shimano, et al. (2004). "SREBPs suppress IRS-2-mediated insulin signalling in the liver." Nat Cell Biol 6(4): 351-7. Imai, J., H. Katagiri, et al. (2008). "Regulation of pancreatic beta cell mass by neuronal signals from the liver." Science 322(5905): 1250-4. Ito, Y., N. Azrolan, et al. (1990). "Hypertriglyceridemia as a result of human apo CIII gene expression in transgenic mice." Science 249(4970): 790-3. Jiang, G. and B. B. Zhang (2003). "Glucagon and regulation of glucose metabolism." Am J Physiol Endocrinol Metab 284(4): E671-8. Jitrapakdee, S. (2012). "Transcription factors and coactivators controlling nutrient and hormonal regulation of hepatic gluconeogenesis." Int J Biochem Cell Biol 44(1): 33-45. Johnson, A. B., J. M. Webster, et al. (1993). "The impact of metformin therapy on hepatic glucose production and skeletal muscle glycogen synthase activity in overweight type II diabetic patients." Metabolism 42(9): 1217-22. Kahn, B. B., T. Alquier, et al. (2005). "AMP-activated protein kinase: ancient energy gauge provides clues to modern understanding of metabolism." Cell Metab 1(1): 15-25. Kahn, S. E. (2003). "The relative contributions of insulin resistance and beta-cell

123 References

dysfunction to the pathophysiology of Type 2 diabetes." Diabetologia 46(1): 3- 19. Kallwellis-Opara, A., X. Zaho, et al. (2003). "Characterization of cis-elements mediating the stimulation of glucose-6-phosphate transporter promoter activity by glucocorticoids." Gene 320: 59-66. Kamagate, A., S. Qu, et al. (2008). "FoxO1 mediates insulin-dependent regulation of hepatic VLDL production in mice." J Clin Invest 118(6): 2347-64. Kashyap, S., R. Belfort, et al. (2003). "A sustained increase in plasma free fatty acids impairs insulin secretion in nondiabetic subjects genetically predisposed to develop type 2 diabetes." Diabetes 52(10): 2461-74. Kenyon, C. J. (2010). "The genetics of ageing." Nature 464(7288): 504-12. Kim, D. H., G. Perdomo, et al. (2011). "FoxO6 integrates insulin signaling with gluconeogenesis in the liver." Diabetes 60(11): 2763-74. Kim, D. H., T. Zhang, et al. (2013). "FoxO6 in glucose metabolism (FoxO6)." J Diabetes 5(3): 233-40. Kim, D. H., T. Zhang, et al. (2011). "Targeting FoxO1 for hypertriglyceridemia." Curr Drug Targets 12(9): 1245-55. Kim, J. Y., H. J. Kim, et al. (2004). "Orphan nuclear receptor small heterodimer partner represses hepatocyte nuclear factor 3/Foxa transactivation via inhibition of its DNA binding." Mol Endocrinol 18(12): 2880-94. Kim, Y. D., K. G. Park, et al. (2008). "Metformin inhibits hepatic gluconeogenesis through AMP-activated protein kinase-dependent regulation of the orphan nuclear receptor SHP." Diabetes 57(2): 306-14. Kitamura, T. (2013). "The role of FOXO1 in beta-cell failure and type 2 diabetes mellitus." Nat Rev Endocrinol 9(10): 615-23. Kitamura, T., J. Nakae, et al. (2002). "The forkhead transcription factor Foxo1 links insulin signaling to Pdx1 regulation of pancreatic beta cell growth." J Clin Invest 110(12): 1839-47. Klip, A. and L. A. Leiter (1990). "Cellular mechanism of action of metformin." Diabetes Care 13(6): 696-704. Kluth, O., F. Mirhashemi, et al. (2011). "Dissociation of lipotoxicity and glucotoxicity in a mouse model of obesity associated diabetes: role of forkhead box O1 (FOXO1) in glucose-induced beta cell failure." Diabetologia 54(3): 605-16.

124 References

Koo, S. H., L. Flechner, et al. (2005). "The CREB coactivator TORC2 is a key regulator of fasting glucose metabolism." Nature 437(7062): 1109-11. Ku, N. O., D. M. Toivola, et al. (2004). "Studying simple epithelial keratins in cells and tissues." Methods Cell Biol 78: 489-517. Kulkarni, R. N., U. S. Jhala, et al. (2004). "PDX-1 haploinsufficiency limits the compensatory islet hyperplasia that occurs in response to insulin resistance." J Clin Invest 114(6): 828-36. Kwon, H. and J. E. Pessin (2013). "Adipokines mediate inflammation and insulin resistance." Front Endocrinol (Lausanne) 4: 71. Kwon, H. S., B. Huang, et al. (2004). "Protein kinase B-alpha inhibits human pyruvate dehydrogenase kinase-4 gene induction by dexamethasone through inactivation of FOXO transcription factors." Diabetes 53(4): 899-910. Lavon, I., I. Goldberg, et al. (2000). "High susceptibility to bacterial infection, but no liver dysfunction, in mice compromised for hepatocyte NF-kappaB activation." Nat Med 6(5): 573-7. Lee, C. S., N. J. Sund, et al. (2002). "Foxa2 controls Pdx1 gene expression in pancreatic beta-cells in vivo." Diabetes 51(8): 2546-51. Lee, J. M., W. Y. Seo, et al. (2010). "AMPK-dependent repression of hepatic gluconeogenesis via disruption of CREB.CRTC2 complex by orphan nuclear receptor small heterodimer partner." J Biol Chem 285(42): 32182-91. Lee, Y. K. and D. D. Moore (2002). "Dual mechanisms for repression of the monomeric orphan receptor liver receptor homologous protein-1 by the orphan small heterodimer partner." J Biol Chem 277(4): 2463-7. Leone, T. C., C. J. Weinheimer, et al. (1999). "A critical role for the peroxisome proliferator-activated receptor alpha (PPARalpha) in the cellular fasting response: the PPARalpha-null mouse as a model of fatty acid oxidation disorders." Proc Natl Acad Sci U S A 96(13): 7473-8. Li, X., B. Monks, et al. (2007). "Akt/PKB regulates hepatic metabolism by directly inhibiting PGC-1alpha transcription coactivator." Nature 447(7147): 1012-6. Li, Y., S. Xu, et al. (2011). "AMPK phosphorylates and inhibits SREBP activity to attenuate hepatic steatosis and atherosclerosis in diet-induced insulin- resistant mice." Cell Metab 13(4): 376-88. Lin, H. V. and D. Accili (2012). "Hormonal regulation of hepatic glucose production in health and disease." Cell Metab 14(1): 9-19. 125 References

Lin, H. Z., S. Q. Yang, et al. (2000). "Metformin reverses fatty liver disease in obese, leptin-deficient mice." Nat Med 6(9): 998-1003. Liu, Y. F., K. Paz, et al. (2001). "Insulin stimulates PKCzeta -mediated phosphorylation of insulin receptor substrate-1 (IRS-1). A self-attenuated mechanism to negatively regulate the function of IRS proteins." J Biol Chem 276(17): 14459-65. Lu, M., M. Wan, et al. (2012). "Insulin regulates liver metabolism in vivo in the absence of hepatic Akt and Foxo1." Nat Med 18(3): 388-95. Lu, T. T., M. Makishima, et al. (2000). "Molecular basis for feedback regulation of bile acid synthesis by nuclear receptors." Mol Cell 6(3): 507-15. Lyssenko, V., R. Lupi, et al. (2007). "Mechanisms by which common variants in the TCF7L2 gene increase risk of type 2 diabetes." J Clin Invest 117(8): 2155-63. MacDonald, P. E., W. El-Kholy, et al. (2002). "The multiple actions of GLP-1 on the process of glucose-stimulated insulin secretion." Diabetes 51 Suppl 3: S434- 42. Marchesini, G., M. Brizi, et al. (2001). "Metformin in non-alcoholic steatohepatitis." Lancet 358(9285): 893-4. Martin-Montalvo, A., E. M. Mercken, et al. (2013). "Metformin improves healthspan and lifespan in mice." Nat Commun 4: 2192. Matsumoto, M., S. Han, et al. (2006). "Dual role of transcription factor FoxO1 in controlling hepatic insulin sensitivity and lipid metabolism." J Clin Invest 116(9): 2464-72. Matsumoto, M., A. Pocai, et al. (2007). "Impaired regulation of hepatic glucose production in mice lacking the forkhead transcription factor Foxo1 in liver." Cell Metab 6(3): 208-16. Mayr, B. and M. Montminy (2001). "Transcriptional regulation by the phosphorylation- dependent factor CREB." Nat Rev Mol Cell Biol 2(8): 599-609. McCarthy, M. I. (2010). "Genomics, type 2 diabetes, and obesity." N Engl J Med 363(24): 2339-50. McCarthy, M. I., P. Rorsman, et al. (2013). "TCF7L2 and diabetes: a tale of two tissues, and of two species." Cell Metab 17(2): 157-9. Meier, J. J., A. E. Butler, et al. (2008). "Beta-cell replication is the primary mechanism subserving the postnatal expansion of beta-cell mass in humans." Diabetes 57(6): 1584-94. 126 References

Menge, B. A., A. Tannapfel, et al. (2008). "Partial pancreatectomy in adult humans does not provoke beta-cell regeneration." Diabetes 57(1): 142-9. Michael, M. D., R. N. Kulkarni, et al. (2000). "Loss of insulin signaling in hepatocytes leads to severe insulin resistance and progressive hepatic dysfunction." Mol Cell 6(1): 87-97. Miller, R. A. and M. J. Birnbaum (2010). "An energetic tale of AMPK-independent effects of metformin." J Clin Invest 120(7): 2267-70. Miller, R. A., Q. Chu, et al. (2013). "Biguanides suppress hepatic glucagon signalling by decreasing production of cyclic AMP." Nature 494(7436): 256-60. Miller, S. P., G. R. Anand, et al. (1999). "Characterization of glucokinase mutations associated with maturity-onset diabetes of the young type 2 (MODY-2): different glucokinase defects lead to a common phenotype." Diabetes 48(8): 1645-51. Mithieux, G., L. Guignot, et al. (2002). "Intrahepatic mechanisms underlying the effect of metformin in decreasing basal glucose production in rats fed a high-fat diet." Diabetes 51(1): 139-43. Morris, A. P., B. F. Voight, et al. (2012). "Large-scale association analysis provides insights into the genetic architecture and pathophysiology of type 2 diabetes." Nat Genet 44(9): 981-90. Myatt, S. S. and E. W. Lam (2007). "The emerging roles of forkhead box (Fox) proteins in cancer." Nat Rev Cancer 7(11): 847-59. Nadal, A., P. F. Marrero, et al. (2002). "Down-regulation of the mitochondrial 3- hydroxy-3-methylglutaryl-CoA synthase gene by insulin: the role of the forkhead transcription factor FKHRL1." Biochem J 366(Pt 1): 289-97. Naimi, M., N. Gautier, et al. (2007). "Nuclear forkhead box O1 controls and integrates key signaling pathways in hepatocytes." Endocrinology 148(5): 2424-34. Nair, S., A. M. Diehl, et al. (2004). "Metformin in the treatment of non-alcoholic steatohepatitis: a pilot open label trial." Aliment Pharmacol Ther 20(1): 23-8. Nakae, J., W. H. Biggs, 3rd, et al. (2002). "Regulation of insulin action and pancreatic beta-cell function by mutated alleles of the gene encoding forkhead transcription factor Foxo1." Nat Genet 32(2): 245-53. Nakae, J., T. Kitamura, et al. (2003). "The forkhead transcription factor Foxo1 regulates adipocyte differentiation." Dev Cell 4(1): 119-29.

127 References

Nakae, J., T. Kitamura, et al. (2001). "The forkhead transcription factor Foxo1 (Fkhr) confers insulin sensitivity onto glucose-6-phosphatase expression." J Clin Invest 108(9): 1359-67. Nakae, J., M. Oki, et al. (2008). "The FoxO transcription factors and metabolic regulation." FEBS Lett 582(1): 54-67. Nakashima, N., P. M. Sharma, et al. (2000). "The tumor suppressor PTEN negatively regulates insulin signaling in 3T3-L1 adipocytes." J Biol Chem 275(17): 12889- 95. Natali, A. and E. Ferrannini (2006). "Effects of metformin and thiazolidinediones on suppression of hepatic glucose production and stimulation of glucose uptake in type 2 diabetes: a systematic review." Diabetologia 49(3): 434-41. Nathan, D. M., J. B. Buse, et al. (2009). "Medical management of hyperglycemia in type 2 diabetes: a consensus algorithm for the initiation and adjustment of therapy: a consensus statement of the American Diabetes Association and the European Association for the Study of Diabetes." Diabetes Care 32(1): 193- 203. Nedumaran, B., S. Hong, et al. (2009). "DAX-1 acts as a novel corepressor of orphan nuclear receptor HNF4alpha and negatively regulates gluconeogenic enzyme gene expression." J Biol Chem 284(40): 27511-23. Nemoto, S., M. M. Fergusson, et al. (2005). "SIRT1 functionally interacts with the metabolic regulator and transcriptional coactivator PGC-1{alpha}." J Biol Chem 280(16): 16456-60. Nielsen, J. H., E. D. Galsgaard, et al. (2001). "Regulation of beta-cell mass by hormones and growth factors." Diabetes 50 Suppl 1: S25-9. Nir, T., D. A. Melton, et al. (2007). "Recovery from diabetes in mice by beta cell regeneration." J Clin Invest 117(9): 2553-61. Nolan, C. J., P. Damm, et al. (2011). "Type 2 diabetes across generations: from pathophysiology to prevention and management." Lancet 378(9786): 169-81. O'Brien, R. M. and D. K. Granner (1996). "Regulation of gene expression by insulin." Physiol Rev 76(4): 1109-61. O'Brien, R. M., R. S. Streeper, et al. (2001). "Insulin-regulated gene expression." Biochem Soc Trans 29(Pt 4): 552-8. Owen, M. R., E. Doran, et al. (2000). "Evidence that metformin exerts its anti-diabetic

128 References

effects through inhibition of complex 1 of the mitochondrial respiratory chain." Biochem J 348 Pt 3: 607-14. Paik, J. H., R. Kollipara, et al. (2007). "FoxOs are lineage-restricted redundant tumor suppressors and regulate endothelial cell homeostasis." Cell 128(2): 309-23. Pajvani, U. B., C. J. Shawber, et al. (2011). "Inhibition of Notch signaling ameliorates insulin resistance in a FoxO1-dependent manner." Nat Med 17(8): 961-7. Paris, M., C. Bernard-Kargar, et al. (2003). "Specific and combined effects of insulin and glucose on functional pancreatic beta-cell mass in vivo in adult rats." Endocrinology 144(6): 2717-27. Parry, P., Y. Wei, et al. (1994). "Cloning and characterization of the t(X;11) breakpoint from a leukemic cell line identify a new member of the forkhead gene family." Genes Cancer 11(2): 79-84. Parsons, J. A., T. C. Brelje, et al. (1992). "Adaptation of islets of Langerhans to pregnancy: increased islet cell proliferation and insulin secretion correlates with the onset of placental lactogen secretion." Endocrinology 130(3): 1459- 66. Pawlikowska, L., D. Hu, et al. (2009). "Association of common genetic variation in the insulin/IGF1 signaling pathway with human longevity." Aging Cell 8(4): 460-72. Perseghin, G., K. Petersen, et al. (2003). "Cellular mechanism of insulin resistance: potential links with inflammation." Int J Obes Relat Metab Disord 27 Suppl 3: S6-11. Pick, A., J. Clark, et al. (1998). "Role of apoptosis in failure of beta-cell mass compensation for insulin resistance and beta-cell defects in the male Zucker diabetic fatty rat." Diabetes 47(3): 358-64. Pilkis, S. J. and D. K. Granner (1992). "Molecular physiology of the regulation of hepatic gluconeogenesis and glycolysis." Annu Rev Physiol 54: 885-909. Puigserver, P., J. Rhee, et al. (2003). "Insulin-regulated hepatic gluconeogenesis through FOXO1-PGC-1alpha interaction." Nature 423(6939): 550-5. Qu, S., J. Altomonte, et al. (2006). "Aberrant Forkhead box O1 function is associated with impaired hepatic metabolism." Endocrinology 147(12): 5641-52. Radziuk, J., Z. Zhang, et al. (1997). "Effects of metformin on lactate uptake and gluconeogenesis in the perfused rat liver." Diabetes 46(9): 1406-13. Raso, G. M., E. Esposito, et al. (2009). "Comparative therapeutic effects of metformin

129 References

and vitamin E in a model of non-alcoholic steatohepatitis in the young rat." Eur J Pharmacol 604(1-3): 125-31. Rieck, S., P. White, et al. (2009). "The transcriptional response of the islet to pregnancy in mice." Mol Endocrinol 23(10): 1702-12. Rizza, R. A. (2010). "Pathogenesis of fasting and postprandial hyperglycemia in type 2 diabetes: implications for therapy." Diabetes 59(11): 2697-707. Rodgers, J. T. and P. Puigserver (2007). "Fasting-dependent glucose and lipid metabolic response through hepatic sirtuin 1." Proc Natl Acad Sci U S A 104(31): 12861-6. Rossetti, L., R. A. DeFronzo, et al. (1990). "Effect of metformin treatment on insulin action in diabetic rats: in vivo and in vitro correlations." Metabolism 39(4): 425- 35. Roth, U., K. Curth, et al. (2004). "The transcription factors HIF-1 and HNF-4 and the coactivator p300 are involved in insulin-regulated glucokinase gene expression via the phosphatidylinositol 3-kinase/protein kinase B pathway." J Biol Chem 279(4): 2623-31. Rowan, J. A., W. M. Hague, et al. (2008). "Metformin versus insulin for the treatment of gestational diabetes." N Engl J Med 358(19): 2003-15. Rui, L., M. Yuan, et al. (2002). "SOCS-1 and SOCS-3 block insulin signaling by -mediated degradation of IRS1 and IRS2." J Biol Chem 277(44): 42394-8. Sachdeva, M. M. and D. A. Stoffers (2009). "Minireview: Meeting the demand for insulin: molecular mechanisms of adaptive postnatal beta-cell mass expansion." Mol Endocrinol 23(6): 747-58. Saltiel, A. R. and C. R. Kahn (2001). "Insulin signalling and the regulation of glucose and lipid metabolism." Nature 414(6865): 799-806. Samuel, V. T., C. S. Choi, et al. (2006). "Targeting foxo1 in mice using antisense oligonucleotide improves hepatic and peripheral insulin action." Diabetes 55(7): 2042-50. Samuel, V. T. and G. I. Shulman (2012). "Mechanisms for insulin resistance: common threads and missing links." Cell 148(5): 852-71. Sanchez-Munoz, V., R. Salas-Romero, et al. (2013). "[Decrease of liver fat content by aerobic exercise or metformin therapy in overweight or obese women]." Rev Invest Clin 65(4): 307-17. 130 References

Schilling, M. M., J. K. Oeser, et al. (2006). "Gluconeogenesis: re-evaluating the FOXO1-PGC-1alpha connection." Nature 443(7111): E10-1. Schips, T. G., A. Wietelmann, et al. (2011 ). "FoxO3 induces reversible cardiac atrophy and autophagy in a transgenic mouse model." Cardiovasc Res 91(4): 587-97. Schmidt-Strassburger, U., T. G. Schips, et al. (2011). "Expression of constitutively active FoxO3 in murine forebrain leads to a loss of neural progenitors." FASEB J 26(12): 4990-5001. Schmoll, D., K. S. Walker, et al. (2000). "Regulation of glucose-6-phosphatase gene expression by protein kinase Balpha and the forkhead transcription factor FKHR. Evidence for insulin response unit-dependent and -independent effects of insulin on promoter activity." J Biol Chem 275(46): 36324-33. Schoneveld, O. J., I. C. Gaemers, et al. (2004). "Mechanisms of glucocorticoid signalling." Biochim Biophys Acta 1680(2): 114-28. Sedding, D. G. (2008). "FoxO transcription factors in oxidative stress response and ageing--a new fork on the way to longevity?" Biol Chem 389(3): 279-83. Shachter, N. S. (2001). "Apolipoproteins C-I and C-III as important modulators of lipoprotein metabolism." Curr Opin Lipidol 12(3): 297-304. Shaw, R. J., K. A. Lamia, et al. (2005). "The kinase LKB1 mediates glucose homeostasis in liver and therapeutic effects of metformin." Science 310(5754): 1642-6. Shibata, M., K. Yoshimura, et al. (2009). "The MAP1-LC3 conjugation system is involved in lipid droplet formation." Biochem Biophys Res Commun 382(2): 419-23. Shimomura, I., M. Matsuda, et al. (2000). "Decreased IRS-2 and increased SREBP- 1c lead to mixed insulin resistance and sensitivity in livers of lipodystrophic and ob/ob mice." Mol Cell 6(1): 77-86. Shu, Y., S. A. Sheardown, et al. (2007). "Effect of genetic variation in the organic cation transporter 1 (OCT1) on metformin action." J Clin Invest 117(5): 1422- 31. Singh, R., S. Kaushik, et al. (2009). "Autophagy regulates lipid metabolism." Nature 458(7242): 1131-5. Sopasakis, V. R., P. Liu, et al. (2010). "Specific roles of the p110alpha isoform of

131 References

phosphatidylinsositol 3-kinase in hepatic insulin signaling and metabolic regulation." Cell Metab 11(3): 220-30. Srinivasan, K. and P. Ramarao (2007). "Animal models in type 2 diabetes research: an overview." Indian J Med Res 125(3): 451-72. Stefan, N., K. Kantartzis, et al. (2008). "Identification and characterization of metabolically benign obesity in humans." Arch Intern Med 168(15): 1609-16. Stephenne, X., M. Foretz, et al. (2011). "Metformin activates AMP-activated protein kinase in primary human hepatocytes by decreasing cellular energy status." Diabetologia 54(12): 3101-10. Stumvoll, M., N. Nurjhan, et al. (1995). "Metabolic effects of metformin in non-insulin- dependent diabetes mellitus." N Engl J Med 333(9): 550-4. Sun, Z., R. A. Miller, et al. (2012). "Hepatic Hdac3 promotes gluconeogenesis by repressing lipid synthesis and sequestration." Nat Med 18(6): 934-42. Sunami, Y., F. Leithauser, et al. (2012). "Hepatic activation of IKK/NFkappaB signaling induces liver fibrosis via macrophage-mediated chronic inflammation." Hepatology 56(3): 1117-28. Tahara, A., A. Matsuyama-Yokono, et al. (2011). "Effects of antidiabetic drugs in high-fat diet and streptozotocin-nicotinamide-induced type 2 diabetic mice." Eur J Pharmacol 655(1-3): 108-16. Tahrani, A. A., C. J. Bailey, et al. (2011). "Management of type 2 diabetes: new and future developments in treatment." Lancet 378(9786): 182-97. Takashima, M., W. Ogawa, et al. (2010). "Role of KLF15 in regulation of hepatic gluconeogenesis and metformin action." Diabetes 59(7): 1608-15. Taneera, J., S. Lang, et al. (2012). "A systems genetics approach identifies genes and pathways for type 2 diabetes in human islets." Cell Metab 16(1): 122-34. Teshigawara, K., W. Ogawa, et al. (2005). "Role of Kruppel-like factor 15 in PEPCK gene expression in the liver." Biochem Biophys Res Commun 327(3): 920-6. Teta, M., M. M. Rankin, et al. (2007). "Growth and regeneration of adult beta cells does not involve specialized progenitors." Dev Cell 12(5): 817-26. Tikhanovich, I., J. Cox, et al. (2013). "Forkhead box class O transcription factors in liver function and disease." J Gastroenterol Hepatol 28 Suppl 1: 125-31. Tsuchida, A., T. Yamauchi, et al. (2004). "Insulin/Foxo1 pathway regulates expression levels of adiponectin receptors and adiponectin sensitivity." J Biol Chem 279(29): 30817-22. 132 References

Utter, M. F. and D. B. Keech (1960). "Formation of oxaloacetate from pyruvate and carbon dioxide." J Biol Chem 235: PC17-8. Vague, J. (1996). "The degree of masculine differentiation of obesities: a factor determining predisposition to diabetes, atherosclerosis, gout, and uric calculous disease. 1956." Obes Res 4(2): 204-12. van der Horst, A. and B. M. Burgering (2007). "Stressing the role of FoxO proteins in lifespan and disease." Nat Rev Mol Cell Biol 8(6): 440-50. van der Vos, K. E. and P. J. Coffer (2011). "The extending network of FOXO transcriptional target genes." Antioxid Redox Signal 14(4): 579-92. Viollet, B., B. Guigas, et al. (2009). "AMP-activated protein kinase in the regulation of hepatic energy metabolism: from physiology to therapeutic perspectives." Acta Physiol (Oxf) 196(1): 81-98. Viollet, B., B. Guigas, et al. (2012). "Cellular and molecular mechanisms of metformin: an overview." Clin Sci (Lond) 122(6): 253-70. Wajchenberg, B. L. (2000). "Subcutaneous and visceral adipose tissue: their relation to the metabolic syndrome." Endocr Rev 21(6): 697-738. Wei, D., R. Tao, et al. (2011). "Feedback regulation of hepatic gluconeogenesis through modulation of SHP/Nr0b2 gene expression by Sirt1 and FoxO1." Am J Physiol Endocrinol Metab 300(2): E312-20. Wilcock, C. and C. J. Bailey (1994). "Accumulation of metformin by tissues of the normal and diabetic mouse." Xenobiotica 24(1): 49-57. Wilcox, G. (2005). "Insulin and insulin resistance." Clin Biochem Rev 26(2): 19-39. Willcox, B. J., T. A. Donlon, et al. (2008). "FOXO3A genotype is strongly associated with human longevity." Proc Natl Acad Sci U S A 105(37): 13987-92. Xiong, X., R. Tao, et al. (2013). "Deletion of hepatic FoxO1/3/4 genes in mice significantly impacts on glucose metabolism through downregulation of gluconeogenesis and upregulation of glycolysis." PLoS One 8(8): e74340. Yamagata, K., H. Daitoku, et al. (2004). "Bile acids regulate gluconeogenic gene expression via small heterodimer partner-mediated repression of hepatocyte nuclear factor 4 and Foxo1." J Biol Chem 279(22): 23158-65. Yeagley, D., S. Guo, et al. (2001). "Gene- and activation-specific mechanisms for insulin inhibition of basal and glucocorticoid-induced insulin-like growth factor binding protein-1 and phosphoenolpyruvate carboxykinase transcription. Roles

133 References

of forkhead and insulin response sequences." J Biol Chem 276(36): 33705-10. Yi, P., J. S. Park, et al. (2013). "Betatrophin: a hormone that controls pancreatic beta cell proliferation." Cell 153(4): 747-58. Yoon, J. C., P. Puigserver, et al. (2001). "Control of hepatic gluconeogenesis through the transcriptional coactivator PGC-1." Nature 413(6852): 131-8. Zhande, R., J. J. Mitchell, et al. (2002). "Molecular mechanism of insulin-induced degradation of insulin receptor substrate 1." Mol Cell Biol 22(4): 1016-26. Zhang, D. M., G. Y. Zhang, et al. (2006). "[Therapeutic effects of insulin-sensitizing drugs on nonalcoholic fatty liver disease: experiment with rats]." Zhonghua Yi Xue Za Zhi 86(18): 1279-83. Zhang, F., X. Xu, et al. (2013). "Gene expression profile analysis of type 2 diabetic mouse liver." PLoS One 8(3): e57766. Zhang, K., L. Li, et al. (2012). "Hepatic suppression of Foxo1 and Foxo3 causes hypoglycemia and hyperlipidemia in mice." Endocrinology 153(2): 631-46. Zhang, W., S. Patil, et al. (2006). "FoxO1 regulates multiple metabolic pathways in the liver: effects on gluconeogenic, glycolytic, and lipogenic gene expression." J Biol Chem 281(15): 10105-17. Zhou, G., R. Myers, et al. (2001). "Role of AMP-activated protein kinase in mechanism of metformin action." J Clin Invest 108(8): 1167-74. Zhou, Y., J. Lee, et al. (2011). "Regulation of glucose homeostasis through a XBP-1- FoxO1 interaction." Nat Med 17(3): 356-65.

134

Declaration

Declaration I hereby confirm that I have made the work independently and without using any sources or aids other than those stated. In each individual case, I have clearly identified the source of the passages that are taken word for word or paraphrased from other works.

Ulm,

Sarah Gul

154 Acknowledgement

Acknowledgement I would like to thank Prof. Dr. Thomas Wirth for giving me the opportunity to work on this project under his supervision, for his keen interest, and fruitful discussions on this project and his great support even in personal matters. I would like to express my gratitude to Dr. Yoshiaki Sunami for his supervision, skilful guidance, practical help, for his informative discussions and speacial thanks for preparing the manuscript and arranging further cooperations on this project. I am also thankful to Dr. Pavel Strnad for his informative discussions and suggestions on this project. I would like to express my gratitude to Prof. Dr. Bernhard Böhm for his interest, nice suggestions and informative discussions on this project. I am thankful to Dr. Karl Lenhard Rudolph for his keen interest and useful discussions on this project. It’s a pleasure to thank Dr. Frank Leithaüser and other staff members of Pathology department for analysing and processing the histology from the animals related to this project. I am also thankful to Dr. Karl-Heinz Holzmann for the help with gene expression microarray experiments. I am also thankful to Melanie Gerstlauer and Nina Ushmorov for the excellent technical assistance. I do appreciate Dr. Katja Fiedler for her kind help in various issues in the lab. I am also very thankful to the former and present members of the Institute of Physiological Chemistry for their help and special thanks to Hansgeorg Glöckler for his help in computer issues. I am thankful to my friends Ayesha Maqbool, Heba Hamed Salem, Gandhari Maity for being cooperative and helpful in various issues. My heartfelt gratitude is reserved for my beloved parents and my in-laws for allowing me to complete this educational carrier and for their continuous support, great care and encouragement through out my studies. Lastly, but perhaps most importantly I am thankful to my caring and loving husband Dr. Muazzam Ali Khan Khattak for his great support, his patience and his encouragement during this time. A very special thanks to my little son Mohammad Zaween a nice companion during the last year of

155 Acknowledgement my P. hD studies. I do appreciate all those who remembered me in their prayers.

156 Curriculum Vitae

Curriculum Vitae

Personal Information

Name: Sarah Gul Date of Birth: 28-08-1982 Place of Birth: Peshawar-Pakistan Nationality: Pakistani Gender: Female Address: Institute of Physiological Chemistry Albert-Einstein-Allee 11 D-89081 Ulm, Germany Telephone: +49 (0) 731 (500) 33834 +49 (0) 731 (500) 23263 Email: [email protected]

Qualification: 08/2009 – to date PhD Student at Institute of Physiological Chemistry, University of Ulm. Member of the International Graduate School in Molecular Medicine Ulm, Germany. Supervisor: Prof. Dr. Thomas Wirth Thesis: “Functional analysis of hepatic FOXO3 signalling in glucose and lipid metabolism ” 2004 – 2006 Master of Philosophy (M. Phil) (Molecular biology/Biochemistry) Quaid-e-Azam University (QAU) Islamabad, Pakistan Thesis: "Iron-Regulated Outer Membrane Protein Profiles of Pasteurella multocida”. Research done under Vaccine Quality control section in National Veterinary Laboratories

157 Curriculum Vitae

(NVL) National agricultural research centre (NARC) Islamabad, Pakistan 2002 – 2004 Master of Science (M. Sc) (Molecular biology/Biochemistry) Quaid-e-Azam University (QAU) Islamabad, Pakistan 2000 – 2002 Bachelor of Science (B. Sc) Botany/Zoology /Chemistry Punjab University, Pakistan 1998 – 2000 Higher Secondary School (HSSC) (F. Sc) Biology/Physics/Chemistry PAF Intermediate college, Rawalpindi, Pakistan 1996 – 1998 Matriculation (Secondary School) Biology/Physics/Chemistry PAF Intermediate college, Rawalpindi, Pakistan

Publications:

S Gul, KH Holzmann, F Leithäuser, H Maier, M Vogel, T Luedde, B Boehm, P Strnad, Y Sunami, T Wirth. Metformin reverses FOXO3-induced hyperactivation of hepatic gluconeogenesis and catabolic pathways (Submitted).

Y Sunami, F Leithäuser, S Gul, K Fiedler, N Güldiken, S Espenlaub, KH Holzmann, N Hipp, A Sindrilaru, T Luedde, B Baumann, S Wissel, F Kreppel, M Schneider, KS Kochanek, S Kochanek, P Strnad, and T Wirth. Hepatic activation of IKK/NFκB signaling induces liver fibrosis via macrophage-mediated chronic inflammation. Hepatology. 2012 Sep;56(3):1117-28. doi: 10.1002/hep.25711. Epub 2012 Jul 10.

Posters:

S Gul , KH Holzmann , F Leithäuser, H Maier , B Böhm, P Strnad, Y Sunami, T Wirth. FOXO3 regulates hepatic glucose production. Poster presented by Dr. Yoshiaki Sunami in Zeitschrift für Gastroenterologie meeting, 158 Curriculum Vitae

Hannover, Germany, January 2013

S Gul , KH Holzmann , F Leithäuser, H Maier , B Böhm, P Strnad, Y Sunami, T Wirth. FOXO3 regulates hepatic glucose production. Poster presented by Dr. Yoshiaki Sunami in Europeon association for the study of liver diseases EASL, Amsterdam. The Netherlands, April 2013

S Gul, KH Holzmann, F Leithäuser, H Maier, M Vogel, T Luedde, B Böhm, P Strnad, Y Sunami, T Wirth. Metformin reverses FOXO3-induced hyperactivation of hepatic gluconeogenesis and catabolic pathways. Poster presented by Dr. Yoshiaki Sunami in European association for the study of diabetes in 49th EASD virtual meeting. Barcelona, Spain September 2013

S Gul , KH Holzmann , F Leithäuser, H Maier , B Böhm, P Strnad, Y Sunami, T Wirth. Hepatic activation of FOXO3 induces initial hyperglycemia and subordinate hypoglycemia and pancreatic β-cell proliferation. Poster presented in Europeon association for the study of liver diseases EASL Barcelona, Spain April 2012

S Gul , KH Holzmann , F Leithäuser, H Maier , B Böhm, P Strnad, Y Sunami, T Wirth. Activation of FoxO3a in the liver induces hypoglycemia and metabolic dysfunction. Poster presented in American Association for the study of Liver diseases AASLD San Francisco, California, USA November 2011

S Gul , F Leithäuser, H Maier , P Strnad, Y Sunami, T Wirth. Activation of FoxO3a in the liver induces hypoglycemia and metabolic dysfunction. Poster presented in DGVS spring conference Complications of liver diseases, Berlin, Germany May 2011

S Gul , F Leithäuser, H Maier , P Strnad, Y Sunami, T Wirth. Characterization of FOXO3a function in the liver. Poster presented in Retreat, Signalling pathways in development and regeneration, Günzberg, Germany June 2010 159 Curriculum Vitae

Meeting/Conferences:

04/2012: Europeon association for the study of liver diseases (EASL) Barcelona, Spain 11/ 2011: American Association for the study of Liver diseases (AASLD) Sanfrancisco, California, USA 05/2011: DGVS spring conference 2011 Complications of liver diseases Berlin, Germany

160