The Role of Crebh in Hepatic Energy Regulation Under Metabolic Stress Roberto Mendez Wayne State University

Wayne State University

Wayne State University Dissertations

1-1-2015 The Role Of Crebh In Hepatic Energy Regulation Under Metabolic Stress Roberto Mendez Wayne State University,

Follow this and additional works at: http://digitalcommons.wayne.edu/oa_dissertations

Recommended Citation Mendez, Roberto, "The Role Of Crebh In Hepatic Energy Regulation Under Metabolic Stress" (2015). Wayne State University Dissertations. Paper 1154.

This Open Access Dissertation is brought to you for free and open access by DigitalCommons@WayneState. It has been accepted for inclusion in Wayne State University Dissertations by an authorized administrator of DigitalCommons@WayneState. THE ROLE OF CREBH IN HEPATIC ENERGY REGULATION UNDER METABOLIC STRESS

ROBERTO MENDEZ

DISSERTATION

Submitted to the Graduate School

of Wayne State University,

Detroit, Michigan

in partial fulfillment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

2015

MAJOR: MOLECULAR BIOLOGY AND GENETICS

Approved By:

______Advisor Date

______

DEDICATION This work is dedicated to my loving family. My parents, Roberto and Helen, my brother, Shane, my sister, Julia, and my wife, Jenny, have supported me during my academic endeavors. Without them by my side, this journey would have been neither possible nor worthwhile.

ACKNOWLEDGMENTS

I would like to thank my advisor, Dr. Kezhong Zhang for being a source of guidance and motivation during my time in the program. I would also like to thank my committee members, Dr. James Granneman, Dr. Maik Huttemann, and Dr. Li Li, for their insightful critiques and suggestions for this project.

My work was made possible with the generous aid and advice of my fellow lab members, past and present, particularly Aditya Dandekar, Hyunbae Kim, Jin-sook Lee,

Chunbin Zhang, Xuebao Zhang, and Ze Zheng.

Finally, I would like to thank the DLAR staff for kindly and patiently providing a clean and friendly work environment.

iii

TABLE OF CONTENTS

Dedication…………………………………………………………………………...... ………...... ii

Acknowledgements……………………………………………………………………...... …. iii

List of Figures………………………………………………………………………...... vi

Chapter I: Introduction……………………………………………………….…………...... 1

Hepatic energy metabolism...... 2

Hepatic energy maintenance...... 5

Transcriptional regulators of hepatic energy metabolism...... 7

Non-alcoholic fatty liver disease...... 10

CREBH...... 11

FGF21...... 14

Chapter II: Defining the phenotype of CrebH -null mice under metabolic stress……...... 17

Summary…………………………………………………………………………...... 17

Materials and methods…………………………………………………………...... 18

Results………………...………………………………………………………...... 22

Discussion...... 33

Chapter III: CREBH regulates FGF21 expression...... 38

Summary…………………………………………………………………………...... 38

Materials and Methods...... 39

Results...... 44

Discussion...... 58

Chapter IV: Exogenous FGF21 administration rescues CrebH -null phenotype...... 61

Summary………………………………………………………………………………...... 61

Materials and Methods...... 62

Results...... 65

Discussion...... 72

Chapter V: Conclusions and significance…………………………………………………...... 74

Appendix A...... 78

Appendix B………………………………………………………………………………………… 79

References...... 80

Abstract…………………………………………………………………………………………...... 96

Autobiographical statement………………………………………………………………...... 98

LIST OF FIGURES

Figure 1: Simplified model of energy metabolism...... 2

Figure 2: Diagram of insulin signaling in the hepatocyte...... 6

Figure 3: Illustration of the two-hit model of NASH pathology...... 11

Figure 4: Diagram of CREBH activation...... 12

Figure 5: Metabolite levels in serum from wild-type and CrebH-null mice under metabolic stress...... 22

Figure 6: Hepatic lipid content analysis in CrebH -null and wild-type control mice after AHF diet...... 23

Figure 7: Physiological comparison of wild-type and CrebH -null mice after AHF diet feeding...... 24

Figure 8: Quantitative real-time PCR analysis of fasting-induced gene transcriptional regulation in CrebH -null and wild-type control mice under fasting and fed conditions. 26

Figure 9: Quantitative real-time PCR analysis of variable-length fasting-induced gene expression in CrebH -null and wild-type control mice...... 27

Figure 10: Quantitative real-time PCR results for high-sucrose and normal diet-fed mice...... 28

Figure 11: GTT and ITT analyses with CrebH -null and wild-type control mice after normal or AHF diet...... 30

Figure 12: Glucose tolerance test and AKT phosphorylation in high-sucrose diet-fed mice...... 31

Figure 13: CREBH activity increases FGF21 serum concentration...... 32

Figure 14: Inspection of conserved Fgf21 gene promoter region elements across mammalian species...... 44

Figure 15: Site-directed mutagenesis of the human FGF21 gene promoter...... 46

Figure 16: Description of luciferase reporter assay...... 47

Figure 17: Reporter assay for the mutated Fgf21 promoter region...... 48

Figure 18: Western blot analysis to validate adenoviral overexpression of CrebH and PPARα in CrebH - or Pparα -null mice...... 49

Figure 19: Verification of CrebH-expressing adenovirus...... 49

Figure 20: Quantitative real-time PCR analysis of CrebH -null, Pparα-null, and wild-type control mice under fed and fasting conditions...... 50

Figure 21: Western blot analysis of null mice under fed and fasting conditions...... 51

Figure 22: Quantitative real-time PCR analysis of hepatic Fgf21 mRNA levels in CrebH - null, Pparα-null, and wild-type control mice under fed and fasting conditions with and without adenoviral rescue procedure...... 53

Figure 23: Western Blot analysis of serum FGF21 protein levels in CrebH- and Pparα- null mice under fed or fasting conditions with and without adenoviral rescue procedure...... 54

Figure 24: ChIP assay for CREBH, PPARα, and CREB binding activities at the human FGF21 gene promoter region...... 56

Figure 25: Immunoprecipation of liver protein lysates from wild-type mice under metabolic stress...... 57

Figure 26: Effects of administration of recombinant FGF21 protein on serum FGF21 levels and body weights...... 65

Figure 27: Infusion of recombinant FGF21 protein reverses hypertriglyceridemia, and hypoketonemia in CrebH-null mice after the AHF diet...... 66

Figure 28: Infusion of recombinant FGF21 protein partially rescued NASH phenotype of the CrebH-null mice after the AHF diet...... 68

Figure 29: Administration of recombinant FGF21 protein reverses hepatic steatosis phenotype developed in the CrebH-null mice under the AHF diet...... 70

Figure 30: Gross anatomical inspection of liver morphology after FGF21 protein administration...... 70

Figure 31: GTT analysis with CrebH-null (KO) and wild-type (WT) control mice after administration with vehicle PBS or FGF21 protein...... 71

Figure 32: Proposed model of CREBH-mediated Fgf21 regulation...... 77

vii

CHAPTER I: Introduction

In the liver, lipid content is regulated by dietary FA (FA) or carbohydrate uptake, as well as hepatic biosynthesis, esterification, oxidation, and export of FA (1,

2). After consumption, dietary carbohydrates are both used for normal functions and, if present in excess, stored for later use. Enzymes involved in glycolytic and lipogenic pathways are dynamically regulated at both transcriptional and post-translational levels by various factors such as substrate concentrations and hormones (3).

Metabolic signals, such as an excess of FA, glucose, or insulin, can regulate the activity or abundance of key transcription factors to modulate hepatic lipid metabolism

(1). Many liver-specific transcription factors have been identified as prospective targets for de novo lipogenesis and FA oxidation, including sterol regulatory element binding protein-1c (SREBP-1c) , peroxisome proliferator-activated receptors

(PPARs), liver X receptor (LXRα), and carbohydrate-responsive element-binding protein (ChREBP) (4-7). These factors integrate signals from multiple pathways and organize the activity of the metabolic enzymes necessary for hepatic lipid metabolism with the supply of energy and FA. Cyclic-AMP response element binding protein H

(CREBH) is an endoplasmic reticulum (ER)-membrane-bound transcription factor that is preferentially expressed in liver (8-10 ). CREBH is a major player in regulating lipid and glucose metabolism in the liver. Metabolism of these two classes of molecules is important in maintaining energy production, usage and storage in individual cells as well as maintaining energy homeostasis at the systemic level(11, 12 ). Disturbances to this delicate balance may lead to energy imbalances and serious health

2 consequences, particularly non-alcoholic fatty liver disease (NAFLD) or non-alcoholic steatohepatitis (11-13 ).

1.1 Hepatic Energy Metabolism

Hepatocytes, the parenchymal cell of the liver, exhibit a unique strategies for processing metabolites (13-18 ). Hepatocytes carry out multiple metabolic processes including glycolysis, gluconeogenesis, lipolysis, lipogenesis, FA oxidation, and ketogenesis (Figure 1). Glycolysis is the breakdown of glucose to pyruvate, which is carried out in ten steps (19 ). During glycolysis, the conversion of glucose to glucose-

6-phosphate, the conversion of fructose-6-phosphate to fructose-1,6-bisphosphate and the conversion of phosphoenolpyruvate (PEP) to pyruvate, are regulated by the enzymes glucokinase, phosphofructokinase, and pyruvate kinase, respectively. The end product of glycolysis, pyruvate, can be irreversibly converted to acetyl coenzyme

A (acetyl CoA), a metabolite involved in multiple metabolic pathways (20 ).

Figure 1. Simplified model of energy metabolism. Glucose taken from blood enters the glycolytic pathway to generate pyruvate and glycogen. Glucose may be generated by the gluconeogenesis or glycogenolysis. Acetyl CoA may convert between ketone bodies or triglycerides. Acetyl CoA also may generate ATP via the TCA cycle and oxidative phosphorylation (not shown).

Gluconeogenesis is the production of glucose from non-sugar sources (19,

21 ). The source used may be lactate, pyruvate, or amino acids. Hepatocytes utilize lactate produced in peripheral tissues via anaerobic respiration. The multistep process of gluconeogenesis is regulated by four enzymes: glucose-6-phosphatase, fructose-1,6-bisphosphatase, pyruvate carboxylase, and PEP carboxykinase

(PEPCK). Interestingly, the rate of activity of both glucose-6-phosphatase and

PEPCK are primarily regulated at the transcriptional level. Glucose can also be stored as or generated from glycogen, a polysaccharide that consists of branched chains of glucose molecules (22 ). These chains can be quickly broken down to produce glucose that can enter the bloodstream or the glycolytic pathway. The breakdown of glycogen, known as glycogenolysis, is regulated by the phosphorylation of its rate-limiting enzyme, glycogen phosphorylase (22 ). Excess levels of glucose are stored as glycogen via the glycogen synthesis pathway. This pathway is regulated by phosphorylation of the glycogen synthase enzyme.

In addition to its involvement in the glucose metabolism, pyruvate also can be converted irreversibly to acetyl-CoA by the pyruvate dehydrogenase complex (23 ).

Acetyl-CoA is a predecessor metabolite of both lipid and ketone body production.

Increased levels of acetyl-CoA increase lipogenesis. Activation of the enzyme, acetyl-

CoA carboxylase, induces the conversion of acetyl-CoA to malonyl-CoA which is then converted to malonyl-ACP and shuttled into the mitochondria. Through multiple conversion processes, the malonyl-ACP molecules are condensed to produce a FA chain (24 ). FA chains are joined to a glycerol-3-phosphate back bone to form a triglyceride (TG). After a TG is formed, it may be stored in cytoplasmic lipid droplets

4 or incorporated into VLDL particles and secreted into the blood. Additionally, in a state of energetic need, TG can also be hydrolyzed to release the esterified FA.

These FA are destined to undergo oxidation. Conversion of carbohydrates to TG involves two steps: glycolysis which generates acetyl-CoA from glucose, and lipogenesis which converts acetyl-CoA to FA.

Retrieving energy from TG requires hydrolysis of the TG molecule to yield the stored FA. Lipolysis occurs in three steps, each of which releases one FA chain (25 ).

First, the TG is cleaved to produce one FA chain and a diacylglyceride. Next, the diacylglyceride is cleaved to produce another FA and a monoacylglyceride. The final step is the cleavage of the last FA chain and a glycerol molecule. Glycerol can be converted to dihydroxyacetone phosphate to feed into the glycolytic and gluconeogenic pathways. TG carried by lipoproteins must also undergo lipolysis to release FA which can be taken up by the target cells. After free FA have been released from the TG, they undergo oxidation (26 ). The primary method of FA oxidation is mitochondrial β-oxidation. FA oxidation is a primary source of energy during fasting. Oxidation of FA yields acetyl-CoA. FA that cross the plasma membrane are rapidly converted to fatty acyl-CoA. This process is known as FA activation.

Ketogenesis is the process by which acetyl-CoA is used to generate ketone bodies (27, 28 ). Ketone bodies are metabolites that can be converted to energy by converting them back to acetyl-CoA. Ketogenesis occurs mainly in the liver during prolonged states of fasting, when available blood glucose levels are low. During fasting, ketones generated in the liver are used as energy in peripheral tissues,

5 especially the brain which cannot produce energy from FA oxidation. Insulin signaling represses ketogenesis. There are three types of ketone bodies: acetoacetone, acetone, and 3-hydroxybutyric acid (BOH). The rate of conversion from acetyl-CoA to these ketone bodies is limited by the enzyme HMG-CoA synthase, which converts acetoacetyl-CoA to HMG-CoA. The first ketone body that can be formed from acetyl-

CoA is acetoacetate. Acetoacetate can generate acetone by non-enzymatic decarboxylation. Acetoacetate can also generate BOH in a reaction catalyzed by the enzyme D-β-hydroxybutryate dehydrogenase (BDH1). Typically, during long-term fasting, BOH is the most abundant ketone body in serum (29, 30 ).

1.2 Hepatic Energy Maintenance

The liver is an organ that is heavily involved in energy homeostasis. The liver monitors and maintains metabolite levels through complex regulatory mechanisms.

Hormones, such as insulin and glucagon, induce signaling pathways in hepatocytes that lead to altered metabolic profiles. Insulin signaling decreases blood sugar, while up-regulating glycolysis, glycogenesis, and lipid storage (31 ). Binding of insulin to the insulin receptor induces autophosphorylation and activation of the receptor. Activation of this transmembrane receptor induces recruitment, phosphorylation and activation of insulin receptor substrate (IRS) proteins. Active IRS1 and IRS2 initiate a signaling cascade that induces glucose receptor transport from the Golgi apparatus to the cell membrane to increase glucose uptake. Additionally, the IRS substrates activate PI3 kinase and, consequently, protein kinase B (AKT). AKT induces both glycogen synthesis and lipid storage (Figure 2).

Glucagon is another metabolic hormone that is produced in the pancreas (32 ).

Production of glucagon is induced by fasting conditions leading to decreased levels of blood glucose. Glucagon is released into the bloodstream where it interacts with hepatic receptors. Hepatocytes under the regulation of glucagon exhibit upregulated gluconeogenesis and glycogenolysis resulting in increased blood glucose.

Additionally, both lipolysis and ketogenesis are upregulated in hepatocytes under glucagon signaling (33, 34 ).

Figure 2. Diagram of insulin signaling in the hepatocyte. Insulin activates insulin receptors (IR), initiating a signaling cascade resulting in increased glucose uptake via the Glucose transporter (GLUT), lipogenesis and glycogen storage, as well as decreased gluconeogenesis.

Glucagon signaling functions through activation of the glucagon receptor, a G- protein coupled receptor. Activation of this receptor results in recruitment of GTP to the associated G-protein complex. This allows the alpha subunit of the G-protein complex to dissociate from the beta and gamma subunits and activate adenylate cyclase. Adenylate cyclase is an enzyme that converts ATP to cyclic AMP (cAMP).

Increased intracellular levels of cAMP result in activation of protein kinase A (PKA).

Active PKA initiates multiple signaling cascades, including activation of the CREB transcription factor and phosphorylase kinase. Targets of CREB transcriptional regulation include Pepck and G6pc , two major players in the gluconeogenic pathway.

Phosphorylase kinase induces signaling cascades that both induce glycogenolysis and inhibit glycogen synthesis.

In contrast to incoming hormone signals, the liver is also responsible for producing hormones that deliver signals on the systemic level, as well as, to specific peripheral organs. These liver-derived hormones include hepcidin, betatrophin, and fibroblast growth factor 21 (FGF21) (35-37 ). Hepcidin is a powerful regulator of iron concentration. FGF21 is a systemic regulator of energy metabolism. Hepcidin and

FGF21 are systemic regulators, while betatrophin acts as a blood lipid regulator by regulating serum TG levels, possibly through promoting ANGPTL3 cleavage (38 ).

1.3 Transcriptional Regulators of Hepatic Energy Metabolism

Hepatic energy metabolism is often regulated at the transcriptional level by modulating transcription of mRNAs of key rate-limiting enzymes. This regulation can be observed in various pathways, including gluconeogenesis, lipolysis, lipogenesis,

FA metabolism, and ketogenesis. Metabolic stress, in the form of hormone signaling or altered metabolite levels, can trigger these changes.

The rate of gluconeogenesis is regulated by activity of two enzymes, PEPCK and G6PC, the catalytic subunit of the glucose-6-phosphatase enzyme complex (21 ).

Both Pepck and G6pc contain segments of DNA known as insulin response elements in their promoter regions. This element allows the transcription factor, FOXO1, to bind and increase the transcriptional rate of these genes. AKT phosphorylation, induced by the insulin signaling cascade, results in increased AKT activity. AKT irreversibly phosphorylates FOXO1 on the promoter region of the gluconeogenic genes, causing

FOXO1 to dissociate from the DNA and exit the nucleus. This causes a decrease in transcription of both Pepck and G6pc, and subsequently, inhibits gluconeogenesis

(Figure 2).

FA metabolism is the avenue by which cells maintain homeostasis of FA levels by way of oxidation and synthesis of FA. The balance of these opposing processes is maintained by transcriptional regulation of key enzymes and activation of pertinent transcription factors. Dietary intake of lipid metabolites is a major factor in regulating the expression and activation of these proteins. The peroxisome proliferator- activating receptor (PPAR) family, liver x receptor alpha (LXRα), carbohydrate response element binding factor (ChREBP), and sterol response element-binding protein 1c (SREBP1c) are integral transcription factors in the maintaining FA homeostasis (39 ).

PPAR family members PPARα is a nuclear receptor that is important in the maintenance of lipid homeostasis. PPARα activity in the liver can be induced by

9 fasting or influx of dietary lipids (40, 41 ). This allows PPARα to act as a FA sensor and induce FA oxidation (42 ). Transcriptional targets of PPARα include, apolipoproteins, which are involved in lipolysis, and carnitine palmitoyl-transferases, which are involved in FA oxidation (43-46 ). Additionally, PPARα-null mice display a defective fasting response characterized by low ketone and glucose levels in blood

(45 ).

LXRα is an isoform of LXR that is highly expressed in liver tissue. LXRα is a key regulator of hepatic lipogenesis and is induced by FA (47, 48 ). Lxrα-null mice demonstrate defective in FA synthesis and lipogenesis, coinciding with down- regulation of the Srebp-1, stearoyl CoA desaturase-1 (Scd1 ), and FA synthase (Fasn) genes (49 ). SREPB1c is a transcription factor that targets genes involved in de novo lipogenesis. Unlike SREBP1a, another SREBP1 isoform, SREBP1c is not regulated by intracellular sterol concentration. Instead SREBP1c regulation is controlled by insulin signaling (50 ). SREBP1c targets genes required for glycolysis and lipogenesis

(51 ). Upon activation, the ER-membrane bound SREBP1c protein migrates to the

Golgi and is cleaved, releasing the active cytosolic portion of the protein which acts as a transcription factor (52 ). ChREBP is a transcription factor that is activated in response to glucose (53 ). Upon activation, ChREBP induces lipogenesis by up- regulating Scd1 , Fasn , and acetyl-CoA carboxylase 1 (54, 55 ). Knockdown of

ChREBP in ob/ob mice resulted in reduced hepatic steatosis and increased insulin sensitivity (56 ).

1.4 Non-alcoholic fatty liver disease (NAFLD)

The above-mentioned methods of regulation describe a delicate balancing act that relies on proper function in all metabolic pathways to maintain homeostasis.

Imbalances in hepatic energy metabolism result in disruption of this homeostasis. If the disruption results in excess lipid accumulation in the liver, NAFLD can occur.

NAFLD is defined by hepatic steatosis in an individual who does not excessively consume alcohol (57 ). Inhibition or deletion of some genes involved in lipid metabolism can protect against development of NAFLD (58 ). For example, the Scd1 - null mouse model is resistant to diet-induced hepatic steatosis (59 ). Prolonged or harmful disruptions may result in NAFLD, a disease state in the liver defined by an excessive accumulation of fat. NAFLD places an individual at risk for NASH, a more advanced form of liver disease that involves both steatosis and inflammation of the liver.

NASH develops as a progression of NAFLD. The progression of a healthy liver to NAFLD to NASH is described by the two-hit model (60 ) (Figure 3). NAFLD is the result of TG buildup in the liver. This buildup is caused by an inability of the liver to balance import and synthesis of TG with the export and degradation or TG. The imbalance is the first hit and is often a result of increased lipid delivery to the liver, increased hepatic de novo lipogenesis, or insulin resistance. After the first hit, a liver that has NAFLD is susceptible to a second hit. The second hit is a stress-inducing factor that results in inflammation. These factors include oxidative stress, endotoxins, saturated FA, or inflammatory cytokines (61, 62 ). Healthy livers are resistant to second hits, but livers with NAFLD are at higher risk for liver damage caused by second hit exposure.

Increased FA delivery; Oxidative stress; hepatic lipogenesis; Inflammatory cytokines; Insulin resistance Saturated FA; Endotoxin

First hits Second hits

Fatty liver Normal liver Steatohepatitis (Steatosis)

Figure 3. Illustration of the two-hit model of NASH pathology.

NAFLD does not always present with symptoms, but NAFLD patients do commonly suffer from fatigue (63 ). Additionally, NAFLD often presents with insulin resistance (64 ). Patients with NAFLD present altered liver enzyme levels, particularly those of aspartate transaminase (AST) and alanine transaminase (ALT). In healthy individuals, the ratio of AST to ALT is less than one. Upon the onset of NAFLD, that ratio increases (57, 65 ). Patients with NASH often present with an AST to ALT ratio that is higher than one, but a proper diagnosis of NASH must be made by liver biopsy

(66, 67 ).

1.5 CREBH

The ER is an organelle that is heavily involved in protein folding and plays an important role in stress response (68 ). The ER membrane contains bound signaling molecules that respond to incoming metabolic stress to regulate lipid and glucose metabolism. CREBH is a bZip transcription factor that is primarily expressed in the liver (10). It is localized in the ER of hepatocytes as an inactive, full-length, membrane-bound protein. In order to become an active transcription factor, CREBH must be cleaved, releasing the cytosolic portion of the protein (9). Activation can be

12 caused by dietary, ER, or inflammatory stress conditions, such as proinflammatory cytokines TNFα, IL6, and IL1β, bacterial endotoxin lipopolysaccharide, insulin signaling, saturated FA, nutrient starvation, or AHF diet feeding (9, 12 ). CREBH migrates from the ER to Golgi apparatus where it is cleaved by Site-1 and Site-2 proteases in a manner similar to that observed in the activation of SREBPs (9).

Afterwards, the active, cytosolic region of CREBH may enter the nucleus to act as a transcription factor and induce expression of downstream target genes involved in lipid and glucose metabolism (Figure 4).

Figure 4. Diagram of CREBH activation. Upon stress, CREBH migrates from the ER to the Golgi Apparatus to undergo cleavage. The cleaved portion of CREBH induces transcription of downstream targets.

CREBH regulates many lipid and glucose metabolic pathways (13 ). In fact, previous studies performed in our laboratory have shown that this multi-faceted behavior maintains homeostasis by inducing opposing metabolic processes, such as lipolysis and lipogenesis or FA elongation and oxidation (12 ). Upon longer-term feeding of AHF, more advanced manifestations of metabolic syndrome become apparent. Upon six-month feeding of AHF, CrebH -null mice develop symptoms of

NASH not found in wild-type mice under the same feeding conditions (12 ).

Hepatocytes from these diet-treated CrebH -null mice exhibit increased size and lipid droplet content. Livers from these mice demonstrate inflammation and morphological changes accompanying the increase in lipid content. Additionally, serum TG levels are increased in CrebH -null mice compared to wild-type mice after diet treatment.

This elevated level of serum TG content is known as hypertriglyceridemia. CREBH was later shown to regulate to expression of ApoA4, a gene involved in lipolysis (69 ).

The role of CrebH in metabolite regulation is not limited to mice. A 2012 study showed that rare variants in the human CREBH gene contribute to the development of hypertriglyceridemia in human patients (70 ).

CREBH has been shown to play a vital role lipogenesis upon induction of ER stress. Primary hepatocytes from wild-type and CrebH -null mice were treated with the

ER stress inducing substances, tunicamycin and palmitic acid (12 ). Wild-type primary hepatocytes exhibited the development of lipid droplets indicative of functional ER stress-induced lipogenesis. Primary hepatocytes from CrebH -null mice were unable to produce lipid droplets at levels comparable to cells from the wild-type mice.

CREBH was also shown to transcriptionally regulate the Fsp27 gene, which is

14 important in lipid droplet formation (71 ). These results indicate a complex and, at times, paradoxical role for CREBH in lipid metabolism.

CREBH has also been demonstrated as major player in gluconeogenesis in the liver. Lee et al. showed that CrebH mRNA, along with that of gluconeogenic enzymes, was upregulated during fasting (13 ). Additionally, CrebH overexpression upregulated the mRNA levels of Pepck and G6pc. In rat primary hepatocytes, CrebH overexpression increased glucose output. CrebH -knockdown reduced fasting blood glucose levels in both wild-type and hyperglycemic db/db mice. These results show that CREBH activation during fasting induces gluconeogenesis.

1.6 FGF21

FGF21 is a metabolic hormone that is expressed preferentially in the liver and regulates lipid and glucose metabolism at a systemic level (72 ). It is a member of the

FGF family of proteins that is commonly associated with cell proliferation. FGF21 differs from this convention by not playing a role in cell proliferation and instead regulating energy metabolism pathways. Activity of FGF21 is associated with antiobesity and antidiabetic effects (73, 74 ). These characteristics have caused

FGF21 to be a focus of clinical trials.

In mice, FGF21 is secreted by multiple organs, though primarily liver, in response to metabolic stress. These stresses include fasting, high-fat diet, and high- sucrose diet. Under high-fat diet and fasting conditions, Fgf21 expression is upregulated by PPARα (75, 76 ). Upon secretion, the FGF21 protein enters circulation where it signals to target tissues, which include liver, adipose, pancreas and brain.

Signaling is carried out by interaction with β-Klotho and FGF receptor (77 ). After

FGF21 has entered circulation, the protein half-life is approximately 30 minutes, but its metabolic effects last for 24 hours in peripheral tissues (78, 79 ). In human hypertriglyceridemia patients, fenofibrate treatment resulted in increased serum

FGF21 levels (80 ).

Under fasting conditions, FGF21 activity stimulates FA oxidation and ketogenesis (37 ). During the fed state, FGF21 produced in the adipose tissue regulates Ppar γ expression. The metabolic consequences of FGF21 signaling vary from tissue to tissue. In liver, FGF21 signaling induces gluconeogenesis, FA oxidation, and ketogenesis. Adipose tissue undergoes decreased lipolysis and increased thermogenesis and glucose uptake in response to FGF21 signaling.

FGF21 signaling in brain alters behavior, increasing food intake and energy expenditure.

Due to its effects on multiple tissues and control over metabolic function,

FGF21 has been the focus many studies aimed at rescuing disease phenotypes. In obese mice, prolonged administration of FGF21 protein resulted in pronounced weight loss (81 ). Furthermore, overexpression of Fgf21 in mouse liver protected these mice from diet-induced weight gain (79 ). In mice that have developed diet- induced NAFLD, exogenous FGF21 protein administration was proven effective in reducing liver lipid content (82 ). In diabetic cynomolgus monkeys, FGF21 mimetic administration improved the overall serum lipid profile (83 ). These animals showed decreased serum TG, cholesterol and LDL content. This improvement coincided with decreased fasting serum glucose and insulin levels and an increase in serum HDL levels. Obese mice, while responsive to exogenous FGF21 administration, have

16 elevated FGF21 serum levels and are FGF21-resistant (84 ). Elevated serum levels of

FGF21 in humans have been associated with hepatic steatosis, insulin resistance, hypertriglyceridemia, and type 2 diabetes (37 ).

The promising results of rescue experiments in animal disease models, coupled with the mildness of side effects, FGF21 has been a focal point for therapeutic research. Research in this area has aimed toward improving stability of the protein in circulation, sensitizing tissues to FGF21 signaling, and increasing bioavailability of administered protein (79, 85, 86 ). Despite evidence of FGF21 resistance in patients who are most likely to benefit from the treatment, therapeutic

FGF21 administration remains a viable avenue of research, due to effectiveness of

FGF21 treatment in FGF21 resistant animal models. Currently, an FGF21 analogue, called LY2405319, has shown promise in clinical testing (87, 88 ).

CHAPTER II: Defining the Phenotype of CrebH -null Mice Under Metabolic Stress

2.1 Summary

In this chapter, we define the phenotype of CrebH -null mice under three sources metabolic stress: fasting, AHF diet, and high-sucrose diet. CrebH-null mice were shown to be susceptible to hyperlipidemia during fasting and AHF diet-feeding.

These mice were also defective in ketogenesis. Quantitative real-time PCR analysis reveals CREBH-mediated regulation of genes involved in lipid clearance and ketogenesis. CrebH -null mice have been shown to be more susceptible to AHF diet- induced non-alcoholic fatty liver disease than wild-type counterparts. Enzymatic and histological analysis of liver tissue from AHF diet-fed CrebH-null and wild-type mice indicated elevated levels of lipid content in CrebH -null mice. High-sucrose diet- feeding is proposed as a novel inducer of CREBH activity. After 5 weeks of high- sucrose feeding, evidence suggests that CrebH -null mice were protected from diet- induced insulin resistance. Additionally, CREBH regulates expression of the metabolic hormone, Fgf21 , under fasting, high-sucrose diet, and AHF diet stresses.

CREBH-mediated regulation of Fgf21 was confirmed by gain-of-function analysis in mice expressing exogenous CREBH.

2.2 Materials and methods

Materials

Chemicals were purchased from Sigma unless indicated otherwise. Synthetic oligonucleotides were purchased from Integrated DNA Technologies, Inc. (Coralville,

IA). Human insulin was purchased from Eli Lilly (Indianapolis, IN). Antibody against murine CREBH was made in our laboratory as previously described (9). Mouse monoclonal anti-FGF21 antibody and FGF21 ELISA kit were from R & D Systems,

Inc. (Minneapolis, MN). Kits for measuring TG, FA, and the ketone body, 3- hydroxybutyric acid were purchased from BioAssay Systems, Inc. (Hayward, CA).

Mouse experiments

The CrebH-null mice with deletion of exons 4–7 of the CrebH gene have been previously described (8). Wild-type control mice on a C57Bl/6J background of approximately 3-months of age were used for the metabolic diet study. Unless otherwise stated, all animals were fed ad libitum with Laboratory Rodent Diet 5001 from Lab Diet. For the atherogenic high-fat (AHF) diet study, the AHF diet (16% fat,

41% carbohydrate, 20% protein, 1.25% cholesterol , supplemented with 0.5% sodium cholate by weight) was purchased from Harlan Laboratories, Inc. (TD.88051). For the high-sucrose diet study, the high-sucrose diet (5.2% fat, 18.3% protein, and 61.7% carbohydrate by weight, with 53% of carbohydrate calories coming from sucrose) was provided by Harlan Laboratories Inc. (TD.88122). All the animal experiments were approved by the Wayne State University IACUC committee and carried out under the institutional guidelines for ethical animal use. For all fasting procedures, mice were housed in cages containing Pure-o’Cel bedding. Mice were provided water, but no

19 food for 6, 12, 14, or 24 hours. Tissue collection or glucose tolerance test were performed immediately after the fasting period.

Food intake and fat mass measurements

The CrebH -null and wild-type mice were subjected to food intake measurement under the normal chow diet. Food consumption was determined by measuring diet weights every 12 hours. Daily food intake was determined based on food consumption in 24 hours, measured daily for one week. The whole body fat mass of the CrebH -null and wild-type mice was determined by using the EchoMRI system for living animals according to manufacturer’s instruction (Echo Systems,

Houston, TX).

Measurement of mouse lipid contents

Liver tissue and blood plasma samples were isolated from the mice under the normal chow diet, after fasting, or after the AHF diet. To determine hepatic TG levels, approximately 100 mg liver tissue was homogenized in PBS followed by centrifugation. The supernatant was mixed with 10% Triton-100 in PBS for TG measurement using a commercial kit (BioAssay Systems, Hayward, CA). Mouse blood plasma samples were subjected to quantitative analyses of plasma TG, total cholesterol, HDL, and LDL as previously described (12 ). Levels of the ketone body metabolite, 3-hydroxybutyric acid in the plasma were determined using commercial kits (BioAssay Systems).

Oil-red O staining

Frozen liver tissue sections were stained with Oil-red O for lipid contents according to standard protocol. Briefly, frozen liver tissue sections of 8 microns were

20 air-dried, and then fixed in formalin. The fixed sections were rinsed with 60% isopropanol before they were stained with freshly prepared Oil-red O solution for 15 minutes. After Oil-red O staining, liver sections were rinsed again with 60% isopropanol followed by washing with water before subjected to microscope analysis.

Quantitative real-time PCR

For real-time PCR analysis, the reaction mixture containing cDNA template, primers, and SYBR Green PCR Master Mix (Applied Biosystems) was run in a 7500

Fast Real-time PCR System (Applied Biosystems, Carlsbad, CA). Fold changes of mRNA levels were determined after normalization to internal control β-actin mRNA levels. The sequences of real-time PCR primers used in this study are shown in appendix A.

Adenovirus injection

Adenovirus-expressing GFP (Ade-GFP) or the activated CREBH (Ade-

CREBH) were injected to wild-type mice through tail-vein injection. Approximately 1 ×

10 10 PFU of adenovirus in 0.3 ml PBS were injected into a mouse of approximately

20 gram body weight. Three days after the injection, the liver tissues and blood samples were collected from the CrebH -null and wild-type control mice under feeding or fasting condition for further analyses.

FGF21 Enzyme-linked Immunosorbent Assay (ELISA)

Mouse blood serum were collected for FGF21 ELISA analysis. The analysis was performed using mouse monoclonal anti-FGF21 antibody and FGF21 ELISA kit from R & D Systems, Inc. (Minneapolis, MN) according to the manufacture’s instruction.

Glucose tolerance test (GTT) and insulin tolerance test (ITT)

Glucose tolerance was measured after intraperitoneal (i.p.) injection of 2mg glucose/gram body weight into overnight-fasted CrebH -null and wild-type control animals. For the insulin tolerance test, mice were fasted for 4 hours before i.p. injection of human insulin (0.75 mU/g body weight). Blood glucose levels were measured from tail blood using an OneTouch Ultra glucometer (Johnson & Johnson).

Statistics

Experimental results are shown as mean ± SEM (for variation between animals or experiments). The mean values for biochemical data from the experimental groups were compared by a paired or unpaired, 2-tailed Student’s t test.

Statistical tests with P < 0.05 were considered significant.

2.3 Results

Previous studies have shown that CREBH cleavage and activation can be effected by ER stress, inflammatory stimuli, and metabolic stress (8, 9, 12 ). In addition to mediating hepatic acute-phase response, CREBH has been shown to have a drastic effect on the ability of mice to cope with metabolic stress caused by high-fat diet feeding. To further explore the role of CREBH on physiological adaptation to metabolic stress, CrebH -null and wild-type control mice were exposed to fasting conditions or fed with the AHF or high-sucrose diet.

Figure 5. Metabolite levels in serum from wild-type and CrebH-null mice under metabolic stress. Levels of (A) triglyceride, (B) fatty acid, and (C) BOH were analyzed under feeding, fasting, and AHF diet conditions. Each bar denotes the mean SEM (n = 8 WT or KO mice under feeding or fasting condition; n = 6 WT or KO mice under the AHF diet). *, P ≤ .05; **, P ≤ .01

Fasting has been previously shown to induce CREBH activity (13, 89 ). Serum

TG were elevated in CrebH -null mice compared to those of the wild-type control under feeding conditions, but serum FA and BOH levels remained similar between the groups (Figure 5). However, after overnight fasting, more discrepancies between

23 the CrebH -null and wild-type mice serum metabolites begin to manifest. These include increased levels of serum TG and FA that coincide with decreased levels of

BOH in the CrebH -null mice. After feeding wild-type and CrebH -null mice with an atherogenic high fat diet for 6 months, CrebH -null mice possess higher levels of hepatic TG compared to wild-type mice, as indicated by both Oil-red staining of liver tissue sections and quantitative enzymatic analysis of hepatic TG levels (Figure 6).

Additionally, serum TG and FA levels were significantly increased in CrebH -null mice, while serum BOH levels were lowered in these mice (Figure 5). Furthermore, CrebH - null mice exhibited decreased body fat mass compared to wild-type mice under a normal chow diet (Figure 7B). This reduction in fat mass coincided with a slightly decreased weight compared to wild-type mice (Figure 7A). Food consumption was measured under these conditions, and CrebH -null and wild-type mice were found to consume similar amounts of food (Figure 7C). This finding excludes food intake as a factor in the differences in body and fat mass observed in our testing groups.

Figure 6. Hepatic lipid content analysis after AHF diet. (A) Oil Red O staining of liver tissue sections from mice after 6 months of AHF diet feeding. CrebH KO mice exhibit drastically increased lipid droplet content as well as hepatocyte size compared to wild-type mice. (B) Quantitative, colorimetric assay of hepatic triglyceride content. ** p ≤ 0.01.

Figure 7. Physiological comparison of wild-type and CrebH-null mice after AHF diet- feeding. (A) Body weight comparison (B) Body fat mass determined by MRI. (C) Food consumption. * p ≤ 0.05.

Previously, we performed gene microarray analysis with CrebH -null and wild- type control mice fed either normal chow or AHF diet (12 ). The microarray analysis implicated that most CREBH-regulated genes are involved in lipid metabolism, specifically TG metabolism, cholesterol and sterol biosynthesis, FA oxidation, and ketogenesis. In order to identify genes that undergo CREBH-mediated, fasting- induced upregulation, wild-type and CrebH -null mice were exposed to feeding or 16- hour fasting conditions. Liver tissues were collected from these mice and quantitative real-time PCR was performed to identify CREBH-regulated genes (Figure 8). My analysis identified that the genes Fgf21 , apolipoprotein (Apo) C-II (ApoC2), ApoA5, and lactate dehydrogenase A (Ldha) are regulated by CREBH in mouse liver under fasting conditions. Expression levels of these genes did not differ in wild-type and

CrebH -null mice under feeding conditions. Upon fasting, wild-type mice showed a

25 modest but significantly higher level of expression for these targets than CrebH -null mice.

Just as the targets of CREBH regulation varied under differing metabolic stresses, the fasting-induced effect of CREBH on target gene expression may vary under differing fasting conditions. In our experiment, groups of three male wild-type mice were paired with groups of three male CrebH -null mice and fasted for 6, 12 or

24 hours beginning at 6 pm. Immediately after fasting, mice were sacrificed and liver tissue samples were collected. Fed wild-type and CrebH -null mice liver tissue samples were collected as controls. Quantitative, real-time PCR was performed to determine hepatic gene expression. In this experimental setting, I identified that

CREBH regulates hepatic expression of the Fgf21 , Bdh1 , and Ldha genes in mice upon different periods of fasting (Figure 9). Fed wild-type and CrebH -null mice showed relatively low baseline expression levels of Fgf21 mRNA. Fgf21 upregulation was strongest in the 6- and 12-hour fasted wild-type mice, but showed a more modest increase in the 24-hour fasted wild-type group. In contrast, CrebH -null mice displayed a similar pattern in Fgf21 expression pattern, but mRNA levels were roughly 10-fold lower than in their wild-type counterparts (Figure 9).

* *

Figure 8. Quantitative real-time PCR analysis of fasting-induced gene transcriptional regulation in CrebH-null (KO) and wild-type (WT) control mice under fasting and fed conditions. Fold changes of mRNA expression levels were calculated after normalization to that of β-actin. Bars denote mean values ± SEM (n=3). * p≤ 0.05.

Expression levels of the Bdh1 and Ldha mRNAs were similar in both wild-type and CrebH -null mice under feeding conditions. However, under 12-hour fasting, expression levels of the Bdh1 and Ldha mRNA in CrebH -null mice were significantly lower than those in the control mice (Figure 9). This difference persisted in Bdh1 expression levels after 24-hour fasting, as well.

Figure 9. Quantitative real-time PCR analysis of variable-length fasting-induced gene expression in CrebH-null (KO) and wild-type (WT) control mice. Liver tissue mRNA samples from wild-type and CrebH-null mice under fed and 6-, 12-, and 24-hour fasting conditions were compared. Fold changes of mRNA expression levels were calculated after normalization to that of β-actin. Bars denote mean values ± SEM (n=3). * p≤ 0.05. ** p≤0.01.

In addition to fasting, I tested the regulation of FGF21 expression by CREBH under metabolic stress-inducing diets. Previous studies have shown that Fgf21 expression is augmented by diets that are high in sucrose (90 ). To test whether

CREBH plays a role in this upregulation, wild-type and CrebH -null mice were fed a normal or high-sucrose diet for 5 weeks. After the feeding period, mice were euthanized and mouse livers were collected. Quantitative real-time PCR analysis revealed that Fgf21 expression shows a marked increase in wild-type mice under the high-sucrose feeding regimen that is absent in CrebH -null mice under the same treatment (Figure 10). Moreover, CrebH -null and wild-type control mice were fed with

28 the AHF diet for 6 months. The AHF diet (also called the Paigen diet) is known to induce atherosclerosis and NAFLD in rodent models (91, 92 ). Similar to the fasting and high-sucrose-diet-fed conditions, the liver of the CrebH -null mice expressed significantly lower levels of the Fgf21 mRNA, compared to that of the wild-type control mice, after AHF diet-feeding.

Figure 10. Quantitative real-time PCR results for high-sucrose and normal diet-fed mice. Wild-type mice showed a statistically significant increase in Fgf21 expression under high-sucrose diet that is not present in CrebH-null mice. High-sucrose diet- treated wild-type mice showed no increase in CrebH expression. Fold changes of mRNA expression levels were calculated after normalization to that of β-actin. Bars denote mean values ± SEM (n=3). * p≤ 0.05.

High-fat diets have been shown to cause NAFLD and high sucrose diets are known to induce insulin resistance (92, 93 ). In order to outline the physiological effects of CREBH under dietary stress, I performed a glucose tolerance test (GTT) and insulin tolerance test (ITT) on wild-type and CrebH -null mice under normal, high- sucrose, and AHF diet-feeding conditions. Under the normal chow diet, CrebH -null

29 and wild-type mice displayed similar patterns of blood glucose clearance in response to exogenous glucose or insulin challenge (Figures 11 A and B). After feeding the mice with AHF diet for six months, CrebH -null and wild-type mice still had similar response patterns to infusion of exogenous glucose (Figure 11C). However, the

CrebH -null mice showed no response to insulin challenge during ITT (Figure 11D). It is notable that CrebH -null mice displayed lower levels of blood glucose compared to the wild-type control mice, which can be explained by the role of CREBH in gluconeogenesis (Figure 11) (13 ). Surprisingly, when GTT was performed on mice that were fed the high-sucrose diet, the CrebH -null mice were better able to clear infused glucose than were wild-type mice (Figure 12A). This effect was quantified using the incremental area under the curve method and found to be statistically significant. To determine if this was due to insulin resistance that was present in wild- type mice and absent in CrebH -null mice, these mice were sacrificed and livers were collected. Because AKT phosphorylation is a downstream effect of insulin signaling in liver, the phosphorylation state of AKT was assessed by Western blot analysis

(Figure 12C). Corroborating the results of the GTT, high-sucrose fed CrebH -null mice possessed higher levels of phosphorylated AKT, despite having somewhat lower levels of total AKT.

Figure 11. GTT and ITT analyses with CrebH-null and wild-type control mice after normal or AHF diet. Fasted mice were infused with glucose or insulin via I.P. injection. Blood glucose was measured before injection and at intervals during the 2- hour recovery period. (A,B) GTT and ITT of mice after normal chow diet feeding. (C,D) GTT and ITT of mice after AHF diet feeding. Bars denote mean values ± SEM (n=3).

Figure 12. Glucose tolerance test and AKT phosphorylation in high-sucrose diet-fed mice. (A-B) After high-sucrose feeding, CrebH KO mice showed increased ability in clearing blood glucose in response to exogenous glucose challenge compared to WT mice. (C) KO mice showed higher levels of AKT phosphorylation than WT mice after high-sucrose feeding.

Bars denote mean values ± SEM (n=8). * p≤ 0.05.

Taken together, my studies suggested that CREBH regulates expression of

FGF21, a hepatic hormone that regulates whole-body energy metabolism, under the metabolic stress conditions induced by fasting, the AHF diet, or the high-sucrose diet.

To confirm this regulation at the protein level, I examined serum levels of FGF21 in the CrebH -null and wild-type control mice upon fasting or in the mice over-expressing the activated form of CREBH or GFP control in livers. Upon 14 hours of fasting, serum levels of FGF21 in the CrebH -null mice were significantly reduced, compared to those of the control mice, as indicated by ELISA analysis (Figure 13A).

Furthermore, an activated form of CREBH or GFP control was over-expressed in livers of wild-type control mice by using an adenoviral-based over-expression system.

Over-expression of the activated CREBH in mouse liver dramatically increased serum levels of FGF21 in the mice after the fasting (Figure 13B), thus confirming the role of CREBH in regulating FGF21 production from the liver.

Figure 13. CREBH activity increases FGF21 serum concentration. (A) CrebH KO mice show decreased Fgf21 expression upon fasting. (B) Overexpression of CrebH upregulates Fgf21 expression. Serum levels of FGF21 were determined by ELISA. The bars denote mean values ± SEM (n=3). * p≤ 0.05; ** p≤ 0.01.

2.4 Discussion

It has been documented that both CrebH and Fgf21 are upregulated by fasting

(76, 89 ). Although CREBH is known to play important roles in transcriptional regulation of hepatic lipid metabolism, the regulatory targets of CREBH under metabolic stress have only begun to be investigated. After identifying the key role that

CREBH plays in lipid and ketone regulation under AHF diet-induced metabolic stress, a logical next step would be to characterize the role of CREBH in managing these metabolites under fasting-induced metabolic stress. Our quantitative real-time PCR analysis revealed fasting-induced targets of CREBH (Figure 8). One of these targets,

HMG-CoA synthase, is an enzyme responsible for production of HMG-CoA, an upstream metabolite of both ketone body and cholesterol synthesis (94 ). This corroborates the notable decrease in serum ketone body concentration (Figure 5).

Additionally, this provides an explanation for the previous finding that CrebH -null mice present lower levels of cholesterol (total cholesterol, HDL, and LDL) (12 ).

The Ldha gene codes the M subunit of the LDH complex (95, 96 ). This subunit entirely composes the LDH-5 complex, the isozyme present in liver tissue (95 ). This enzyme converts lactic acid to pyruvate in the liver as part of the Cori cycle (97 ). The resulting pyruvate then feeds in to gluconeogenesis pathway. The lower level of Ldha in liver tissues of CrebH -null mice under fasting conditions could be a contributing factor to the low blood glucose level seen in these animal models (Figure 8) (12, 13 ).

The Ldha fasting expression pattern shows a steady increase as fasting length increases in wild-type mice (Figure 11). This increase is not present in the CrebH -null mice, which maintains a similar expression level at all time points. This pattern

34 indicates a third regulation pattern that consistently increases over the course of the fast and is entirely dependent on CREBH activity. The fact that the CrebH -null mice maintain a steady level of Ldha expression that is similar to that of wild-type mice under the feeding condition shows that CREBH is not necessary to maintain the base line level of Ldha expression.

Bdh1 is an enzyme involved in ketogenesis that was previously shown to be regulated by CREBH (11, 98 ). The fasting-induced expression profile of Bdh1 shows that wild-type mice have peak expression during the 12- and 24-hour fasting periods

(Figure 11). These longer fasting periods represent time points when ketogenesis would become active. The CrebH -null mice do not show increased expression during these periods. This shows the necessity of CREBH in Bdh1 induction upon fasting and indicates the importance of CREBH in ketogenesis.

Apolipoproteins are proteins involved in lipid transport and lipolysis (99 ). Two apolipoproteins, ApoA5 and ApoC2, were down-regulated in fasting CrebH -null mice compared to fasting wild-type mice (Figure 10). These two genes have previously been shown to be regulated by CREBH under a high-fat diet (12 ). ApoA5 is involved in the formation of lipid-carrying lipoprotein particles, such as HDL and VLDL (100 ).

ApoC2 facilitates the interaction between lipoprotein particles and lipoprotein lipase

(101 ). This interaction is important for delivery of the TG to peripheral tissues.

Because apolipoproteins are core components of lipid-carrying particles, they are essential to lipid clearance in blood, and human polymorphisms in these genes are linked to hyperlipidemia (102 ). Our previous studies have shown that CrebH -null mice demonstrate hyperlipidemia when fed an AHF diet (12 ). I have demonstrated that

35 hyperlipidemia occurs in CrebH -null mice under fasting conditions, as well (Figure 5).

The hyperlipidemia phenotype is likely due to down-regulation of ApoC2 and ApoA5 in CrebH -null mice. Additionally, owing to the importance of these proteins in delivery of lipids to peripheral tissues, the absence of ApoC2 and ApoA5 in CrebH -null mice may be responsible for the small fat pads present in these mice under the AHF diet compared to wild-type mice under the same diet (12 ).

A critical gene regulated by CREBH under fasting is Fgf21 (Figure 8).

Previously, through microarray gene expression analysis, we identified several groups of CREBH-regulated genes that are involved in lipolysis, FA oxidation, and lipogenesis (12 ). However, how CREBH modulates perplex genetic programs involved in multiple lipid metabolic pathways remains to be elucidated. FGF21 is a fasting-induced hormone responsible for activating multiple metabolic stress response pathways (76 ). FGF21 exerts pleiotropic effects on hepatic FA oxidation, ketogenesis, and lipolysis by reprogramming gene transcription in lipid metabolism in the liver upon fasting or under a methionine- and choline-deficient (MCD) diet (76,

103, 104 ). Identifying Fgf21 as a target of CREBH may explain the role of CREBH in lipid metabolism during states of metabolic stress. Fgf21 displays an interesting expression pattern upon fasting (Figure 9). In wild-type mice, Fgf21 expression peaks during the 6- and 12-hour fast, but begins to lower during the 24-hour fast. This pattern is the same in CrebH -null mice, but the relative mRNA levels are approximately 10-fold lower. In this case, another factor seems to be regulating the pattern of expression based on the length of the fast, but CREBH appears to be upregulated FGF21 during all of the fasting periods studied. Fgf21 has also been

36 shown to be up-regulated by dietary sucrose ingestion (90 ). Our own high-sucrose diet experiments corroborate this observation (Figure 10). However, Fgf21 mRNA levels are not increased in CrebH -null mice under high-sucrose diet. This implies that sucrose-dependent Fgf21 mRNA level increases are mediated by CREBH.

Interestingly, mRNA levels of CrebH are not affected by the high-sucrose diet. This is similar to the CrebH mRNA level observed in mice under the AHF diet, which also showed an increase in Fgf21 mRNA levels. A distinction between the high-sucrose and the AHF diets, though, is the lack of increased expression of Ppar α mRNA in high-sucrose diet-fed mice (105 ). The link established between FGF21 and CREBH in these experiments led us to investigate the relationship between these two proteins and possible role of FGF21 as a mediator of CREBH.

Previous studies have shown that CrebH mediates different metabolic processes depending on the type of stress endured by the hepatocyte (9, 12, 13, 89 ).

Some of the processes that CrebH regulates are even in opposition to each other

(e.g. lipolysis under AHF-feeding and lipogenesis under ER stress). We hypothesized that CREBH may initiate different fasting-induced pathways based on the length of the fasting stress. We designed an experiment to test the effect of various fasting conditions on CREBH target gene expression. If CREBH does differ in its regulatory targets based on different fasting conditions, we should expect to see CREBH targets displaying different expression profiles for individual genes. Indeed, multiple profiles were identified, confirming this hypothesis (Figure 9). However, CrebH itself shows only a modest increase in expression due to fasting. This may be misleading, though, as CREBH activity cannot be accurately expressed by its mRNA level. CREBH

37 activity is also regulated at the protein level by RIP-mediated cleavage (Figure 4).

Additionally, as we will demonstrate in the following chapter, a binding partner can modify the transcriptional activity of CREBH.

CHAPTER III: CREBH Works in Concert with PPARα to Regulate Fgf21 Expression

3.1 Summary

In the previous chapter, the regulatory relationship between CREBH and Fgf21 was established. In this chapter, we define the mechanism by which CREBH regulates Fgf21 expression. Analysis of the Fgf21 gene promoter identified highly- conserved CRE and PPRE1/2 response elements in close juxtaposition, indicating protein-DNA binding sites for CREBH and PPARα. DNA binding of CREBH and

PPARα was confirmed by ChIP analysis. Site-mutation was performed to generate mutant FGF21 promoter regions, representing mutations in each of newly discovered response elements, for luciferase reporter assay. Reporter assay indicated that the

CRE and PPRE2 are vital to CREBH- and PPARα-mediated Fgf21 expression.

CREBH and PPARα act in synergy to induce expression of the Fgf21 gene promoter.

Moreover, the proximity of the binding sites suggested a possible protein-protein interaction between CREBH and PPARα. This discovery of the possible interaction was followed by verification of the interaction and investigation of its importance. IP-

Western blotting analysis showed that CREBH and PPARα interact under fasting- and AHF diet-induced stress. CREBH and PPARα were shown to be necessary for proper expression of Fgf21 during fasting-induced metabolic stress. Our study found that CREBH and PPARα regulate each other transcriptionally and interact with each other to induce Fgf21 transcription.

3.2 Materials and Methods

Materials

Chemicals were purchased from Sigma unless indicated otherwise. Synthetic oligonucleotides were purchased from Integrated DNA Technologies, Inc. (Coralville,

IA). Antibodies against flag and β-actin were purchased from Sigma. Mouse monoclonal anti-PPARα antibody was purchased from Millipore (Billerica, MA),

Mouse monoclonal anti-FGF21 antibody and FGF21 ELISA kit were purchased from

R & D Systems, Inc. (Minneapolis, MN). Polyclonal anti-CREBH antibody used for

Western blot analysis was raised in our laboratory as previously described (106 ). Kits for measuring TG and FA were from BioAssay Systems (Hayward, CA).

Mouse experiments

The CrebH-null mice were previously generated as described (8). The Pparα- null mice were purchased from the Jackson Laboratories. The CrebH -null, Pparα-null, and wild-type control mice on a C57Bl/6J background at approximately 3-month-old were used for the experiments. All the animal experiments were approved by the

Wayne State University IACUC committee and carried out under the institutional guidelines for ethical animal use.

Cell culture experiments

The hepatocarcinoma cell line Huh-7 was kindly provided by Dr. Charles Rice at the Rockefeller University [1]. Huh-7 cells were plated on 10-cm plates and maintained in DMEM/High Glucose media containing 100 units/ml penicillin, 100

μg/ml streptomycin, and 10% fetal bovine serum. The cells at about 60% confluence were infected with adenovirus-expressing GFP (Ade-GFP), adenovirus-expressing

CREBH (Ade-CREBH), adenovirus-expressing PPARα (Ade-PPARα), or combination of Ade-CREBH plus Ade-PPARα at an MOI of 100 for 48 hours before they were collected for ChIP analyses.

ChIP assay

The Huh-7 cells were infected with equal amounts of Ade-GFP, Ade-CREBH,

Ade-PPARα, or combination of Ade-CREBH and Ade-PPARα. To determine the binding of CREBH and/or PPARα to the FGF21 promoter region, we used a commercial ChIP assay kit (Millipore, MA) according to the manufacturer’s instruction with modifications as previously reported (106 ). A polyclonal anti-CREBH antibody or a monoclonal anti-PPARα antibody (Millipore, MA) was used for IP. The genomic

DNA controls isolated from sonicated cell lysates and immunoprecipitated DNAs were both extracted with the chelex-100 kit (Bio-Rad, CA). For the cross-link procedure, formaldehyde was added directly to the medium to a final concentration of

1%. The cultures were incubated for 10 min at 37 ˚C. Cells were washed with ice-cold

PBS containing protease inhibitors and collected by centrifugation. Cell pellets were re-suspended in 200 μl SDS lysis buffer, and were sonicated to obtain DNA fragments of 200-1000 bp. For the PCR analysis, 1 μl DNA sample was mixed with 5

μM human FGF21 forward primer (5’-ATTCAGGCTGCCCTTGCCACG-3’), 5 μM human FGF21 reverse primer (5’-GCCTGGCTCCTCTTTCGGC-3’), and 18 μl PCR supermix. As a control, PCR reaction was 3 carried out with the control primers

(forward; 5’ CACCAGGGTCTCCGATCTCCCA-3 and reverse; 5’-

TGCTGGAGGTGGGCCTTGGAC-3’).

Quantitative Real-time RT-PCR

For real-time PCR analysis, the reaction mixture containing cDNA template, primers, and SYBR Green PCR Master Mix (Applied Biosystems) was run in a 7500

Fast Real-time PCR System (Applied Biosystems, Carlsbad, CA). Fold changes of mRNA levels were determined after normalization to internal control β-actin mRNA levels. The primer sequences are described in Appendix A.

Western Blot analysis

Total cell lysates were prepared from either liver tissue or cultured cells using

NP-40 lysis buffer (1% NP-40, 50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.05% SDS,

0.5 mM Na vanadate, 100 mM NaF, 50 mM β-glycerophosphate, and 1 mM phenylmethylsulfonyl fluoride) supplemented with protease inhibitors (EDTA-free

Complete Mini, Roche) as previously described (107 ). Denatured proteins were separated by SDS-PAGE on 10% Tris-glycine polyacrylamide gels and transferred to a 0.45-mm PVDF membrane (GE Healthcare) followed by Western blot analysis with anti-CREBH, anti-PPARα, anti-FGF21, or anti-β actin antibody. Membrane-bound antibodies were detected by chemiluminescence detection reagents (GE Healthcare).

Immunoprecipitation (IP) – Western blot analysis

The interaction between CREBH and PPARα was determined by IP-Western blot assay. For preparation of the whole lysates, liver tissues were homogenized and lysed with Nonidet P-40 lysis buffer containing protease inhibitor cocktail as previously described (107 ). For IP, 200 μg of the whole lysate were incubated with 3

μg of specific protein antibody for overnight at 4°C. Immunoprecipitated proteins were

42 resolved by SDS-PAGE followed by immunoblotting using an anti-CREBH or anti-

PPARα antibody.

Site-directed mutagenesis

Site mutation was introduced into the CREB-PPRE binding motif in the FGF21 gene promoter using the QuickChange II Site-directed Mutagenesis Kit (Stratagene) according to manufacturer’s instruction. All the constructed plasmid vectors were confirmed by DNA sequencing. All the reporter assays were performed in triplicate.

The primers used for site-directed mutagenesis are: 5’-

CAAATATCACGCCACAGGAGTGGG-3’; and 5’-

CCCACTCCTGTGGCGTGATATTTG-3’.

Luciferase reporter analysis

For luciferase reporter analysis, we used Dual-Luciferase Reporter Assay

System (Promega). Hepa1-6 cells were co-transfected with the reporter vectors and control reporter vector pGL4.7 and infected with adenovirus Ade-GFP, Ade-CREBH,

Ade-PPARα, or Ade-CREBH plus Ade-PPARα. Luciferase assay was performed at

48 hours after transfection/infection. Data graphs were presented after normalization of Firefly luciferase reporter activities to the control Renilla luciferase activities.

Statistics

Experimental results are shown as mean ± SEM (for variation between animals or experiments). All in vitro experiments were repeated with biological triplicates at least three times independently. Mean values for biochemical data from the experimental groups were compared by paired or unpaired, 2-tailed Student’s t tests. Multiple comparisons were compared with ANOVA and proceeded by ad hoc

43 statistical test when necessary. Statistical tests with P < 0.05 were considered significant.

3.3 Results

After establishing the regulatory effect of CREBH on Fgf21 expression, our next step was to delineate the mechanism of this effect. Analysis of the Fgf21 gene revealed a highly-conserved cAMP response element (CRE) within the promoter region, 63 nucleotides upstream from the transcription start site (Figure 14). This site serves as a binding region for CREB protein family members, including CREBH, to initiate transcription (10, 108 ). Upstream from this element, in close proximity, is an also highly conserved PPAR response element (PPRE), which serves a binding region for PPAR proteins (109 ). The PPRE is composed of two individual binding elements, PPRE1 and PPRE2. The proximity of the binding sites suggests a possible interaction between CREBH and a PPAR protein.

Figure 14. Inspection of conserved Fgf21 gene promoter region elements across mammalian species. cAMP response element (CRE) and PPAR response elements 1 and 2 (PPRE-1 and PPRE-2) are highly conserved and in close proximity within the Fgf21 promoter region.

In order to determine the contribution of the CRE, PPRE1, and PPRE2 binding elements in CREBH- and/or PPARα-mediated Fgf21 expression, we performed site directed mutagenesis for each of these regions (Figure 15). This process was used to produce four mutant forms of the Fgf21 promoter region. The FM13 mutant contained

45 an AC to TG conversion in the CRE element. The PE-M1 mutant contained an AT to

CG conversion in the PPRE1 element. The PE-M2 mutant contained an AC to CG conversion in the PPRE2 element. The fourth mutant promoter region, PE-M12, contained both mutations present in the PE-M1 and PE-M2 mutants. The mutant promoter regions were fused to luciferase reporter genes (Figure 15A). The resulting plasmids were used to transfect Hepa1-6 cells. Reporter assays were performed after infection of the Hepa1-6 cells with adenovirus expressing GFP, CREB, CREBH, or

PPARα (Figures 16 and 17). CrebH and PPARα overexpression were verified by

Western blot analyses (Figure 18). CrebH overexpression was also verified by measurement of serum and hepatic TG (Figure 19). Exogenous expression of CREB showed no increase in reporter activity from the wild-type, PE-M1 or PE-M2 promoters (Figure 16B). The wild-type, PE-M1, and PE-M2 promoters showed reporter activity under exogenous CrebH expression, however. Reporter activity from the PE-M2 promoter was slightly lower than from the wild-type and PE-M1 promoters under this condition. Exogenous expression of CrebH showed no reporter activity with the FM13 promoter (Figure 17A). The wild-type, PE-M1, and PE-M2 promoters showed reporter activity under co-overexpression of CrebH and GFP (Figure 17C).

Reporter activities from the PE-M2 and PE-M12 promoters were slightly lower than from the wild-type and PE-M1 promoters under this condition, as well. Co- overexpression of CrebH and GFP showed no reporter activity with the FM13 promoter (Figure 17B). Co-overexpression of PPAR α and GFP induced reporter activity of the PE-M1, PE-M2, PE-M12, and wild-type promoters; PE-M2 and PE-M12 showed lower activity than PE-M1 and wild-type (Figure 17C). This condition showed

46 no reporter activity from the FM13 promoter (Figure 17B). Co-overexpression of

CrebH and PPAR α-induced the highest reporter activity in the wild-type, PE-M1, PE-

M2, and PE-M12 promoters (Figure 17C), suggesting a synergy of CREBH and

PPARα in driving Fgf21 expression. This condition did not induce reporter activity with the FM13 promoter (Figure 17B).

Figure 15. Site-directed mutagenesis of the human FGF21 gene promoter. WT, CRE- mutated (FM13), PPRE1-mutated (PE-M1), and PPRE2-mutated (PE-M2) forms of the FGF21 promoter region. A fourth mutant that contained the mutations present in PE-M1 and PE-M2, called PE-M12, was also created.

Figure 16. Description of luciferase reporter assay. (A) Illustration of luciferase fused to WT and mutant promoter regions. (B) Reporter assay demonstrating the specificity of CREB binding in the Fgf21 promoter region. Hepa 1-6 cells were transduced with the WT or mutant form of the Fgf21 promoter fused to the luciferase promoter gene. Cells were transfected with a control plasmid, plasmid expressing CREB, or plasmid expressing CREBH. The bars denote mean values ± SEM (n=3 biological repeats). ** p≤ 0.01.

Figure 17. Reporter assay for the mutated Fgf21 promoter region. (A-C) Hepa 1-6 cells were transiently transduced with expression vectors containing the WT or mutated Fgf21 promoter region linked to a luciferase reporter. (A) Cells expressing WT or FM1 linked reporter gene were infected with adenovirus expressing active CrebH. (B) Cells with WT or FM1 linked reporter gene were infected with adenovirus expressing active combinations of CrebH, Pparα, and GFP. (C) Cells with WT, PE- M1, PE-M2, or PE-M12 linked reporter gene were infected with adenovirus expressing active CrebH, Pparα, and GFP. The bars denote mean values ± SEM (n=3 biological repeats). * p≤ 0.05; ** p≤ 0.01.

Figure 18. Western blot analysis to validate adenoviral overexpression of CrebH and PPARα in CrebH- or Pparα-null mice. (A) Hepatic CREBH content in Pparα-null mice overexpressing CrebH or GFP. (B) Hepatic PPARα content in CrebH-null mice overexpressing PPARα or GFP.

Figure 19. Verification of CrebH- expressing adenovirus using measurements of (A) serum TG, (B) hepatic TG, and (C) body weight in male wild-type mice after four months of AHF feeding. The bars denote mean values ± SEM (n=3 for GFP expressing mice and 4 for CREBH- expressing mice). ** p≤ 0.01.

Real-time PCR analysis of CrebH -null and wild-type mice after a 14-hour overnight fast was performed and this analysis revealed targets of CREBH transcriptional regulation. Both CrebH and Fgf21 mRNA levels were upregulated in response to fasting in wild-type mice. This pattern is not observed in CrebH -null mice.

Similar pattern is observed in PPARα mRNA levels (Figure 20). Real-time PCR for these genes was performed on Ppar α-null mice and wild-type counterparts under fasting conditions. The expression profile of CrebH in Ppar α-null mice resembled that of Ppar α in CrebH -null mice. The mRNA levels of these genes are reflected in the levels of associated proteins in the null mouse models (Figure 21). Together, these data suggest a reciprocal regulation between CREBH and PPARα (i.e. CREBH and

PPARα regulate each other for their induction upon fasting).

Figure 20. Quantitative real-time PCR analysis of CrebH-null, Pparα-null, and wild- type control mice under fed and fasting conditions. (A) CrebH, Fgf21, and Pparα mRNA levels in CrebH-null and wild-type mice. (B) CrebH, Fgf21, and Pparα mRNA levels in Pparα-null and wild-type mice. Fold changes of mRNA expression levels were calculated after normalization to that of β-actin. Bars denote mean values ± SEM (n=3). * p≤ 0.05; ** p≤0.01.

Figure 21. Western blot analysis of null mice under fed and fasting conditions. (A) CrebH, Fgf21, and Pparα protein levels in CrebH-null mice. (B) CrebH, Fgf21, and Pparα protein levels in Pparα-null mice. Results compared to β-actin expression.

In order to determine the individual contribution of PPARα and CREBH to the hepatic expression of Fgf21 , an adenoviral rescue experiment was performed. Ppar α- null and CrebH -null mice were fasted and injected with CrebH - or Ppar α-expressing adenovirus, respectively. These rescue-treated mice were compared to wild-type and untreated null mice under fed and fasting conditions. Liver and blood serum were collected. Real-time PCR was performed on liver tissue (Figure 22). Wild-type, fed mice showed a low, baseline level of Fgf21 mRNA expression, which was several- fold lower than the expression level of wild-type mice under the fasting condition. This baseline level was matched in Ppar α-null and CrebH -null mice under both fasting and feeding conditions. Fasted CrebH -null and Ppar α-null mice that were injected with

GFP -expressing adenovirus displayed baseline or lower than baseline levels of Fgf21 expression. Ppar α-null mice that overexpressed CrebH showed a significant increase in Fgf21 expression levels, compared to the fed, Pparα -null mice. However, these mice showed lower levels of Fgf21 expression than the wild-type, fasted mice.

CrebH -null mice that overexpressed Ppar α displayed only a baseline expression level of Fgf21 mRNA. Western blot analysis showed that serum FGF21 protein levels varied slightly from hepatic Fgf21 mRNA levels (Figure 23). Serum FGF21 protein levels were low or undetectable in wild-type mice under the fed condition. Fasted, wild-type mice had observably higher FGF21 serum levels than fed, wild-type mice.

Serum FGF21 levels of CrebH -overexpressing, fasted Ppar α-null mice were similar to those from the fasted wild-type mice. Serum FGF21 levels of fasted CrebH -null mice over-expressing Pparα were detectable, but notably lower than wild-type, fasted levels.

Figure 22. Quantitative real-time PCR analysis of hepatic Fgf21 mRNA levels in CrebH-null, Pparα-null, and wild-type control mice under fed and fasting conditions with and without adenoviral rescue procedure. (A) Fgf21 levels in WT and Pparα-null mice under fed and fasting conditions. Additional Pparα-null mice were treated with adenovirus expressing GFP or CrebH, and Fgf21 mRNA levels were assessed. (B) Fgf21 levels in WT and CrebH-null mice under fed and fasting conditions. Additional CrebH-null mice were treated with adenovirus expressing GFP or Pparα, and Fgf21 mRNA levels were assessed. Fold changes of mRNA expression levels were calculated after normalization to that of β-actin. Bars denote mean values ± SEM (n=3). * p≤ 0.05; p ≤ 0.01.

Figure 23. Western Blot analysis of serum FGF21 protein levels in CrebH- and Pparα-null mice under fed or fasting conditions with and without adenoviral rescue procedure. (A) FGF21 protein levels in WT and Pparα-null mice under fed and fasting conditions. Additional Pparα-null mice were treated with adenovirus expressing GFP or CrebH, and FGF21 protein levels were assessed by Western blot analysis. (B) FGF21 protein levels in WT and CrebH-null mice under fed and fasting conditions. Additional CrebH-null mice were treated with adenovirus expressing GFP or Pparα, and FGF21 protein levels were assessed. Serum albumin protein levels were determined as loading controls.

The capability of PPARα and CREBH to bind to the promoter region of the

Fgf21 gene was determined in order to evaluate the proteins as possible initiators of

Fgf21 gene transcription. DNA-protein binding was activity assessed via ChIP assay

(Figure 24A). Huh7 cells were infected with adenovirus-expressing GFP, active

CrebH , Ppar α, or a combination of active CrebH and Ppar α. PCR was performed on the precipitate product using primers targeting the Fgf21 gene promoter region

(Appendix B). Protein-DNA binding complexes in Huh7 cells expressing CrebH ,

Ppar α or a combination of the two were pulled-down using anti-PPARα or anti-

CREBH antibodies and subjected to ChIP-PCR amplification of the Fgf21 gene promoter region containing the CRE-PPRE binding elements (Appendix B). The

ChIP-PCR results showed that CREBH and PPARα can bind to the Fgf21 gene promoter (Figure 24A). Interestingly, co-expression of CREBH and PPARα yielded stronger binding activity to the Fgf21 promoter than expression of either CREBH or

PPARα, thus confirming the synergistic effect of CREBH and PPARα on driving

Fgf21 gene transcription. As a control, expression of GFP yielded negative results with PPARα and CREBH pull-down.

Because the cAMP response element is a binding site for multiple members of the CREB protein family, we tested the specificity of the element for CREBH binding via ChIP assay (Figure 24B). Huh7 cells were infected with plasmid expressing Creb1 or CrebH or an empty vector. Pull-down was performed with cells over-expressing

Creb1 or vector control using anti-CREB and anti-CREBH antibodies. PCR was used to amplify the same Fgf21 promoter sequence as the previous experiment. Pull-down using anti-CREB antibodies failed to produce positive Fgf21 promoter-binding signals

(Figure 24B), suggesting that it is CREBH, not CREB, that binds to the proximal

Fgf21 gene promoter.

Figure 24. ChIP assay for CREBH, PPARα, and CREB binding activities at the human FGF21 gene promoter region. (A) ChIP with Huh7 cells infected with adenovirus expressing GFP CrebH, Pparα, or an equal-titer combination of CrebH and Pparα. PCR was used to amplify the target sequence on the Fgf21 promoter region containing the CRE-PPRE consensus sequence (Appendix B). Input used as positive control. Negative controls include the Huh7 cells with GFP overexpression, no antibody pull-down, IgG antibody pull-down, and PCR using primers targeting the FUT1 gene. (B) ChIP using Huh7 cells transfected with plasmid expressing CREB or CREBH or empty vector. PCR was used to amplify the target sequence on the Fgf21 promoter region. Input used as positive control. Negative controls include empty vector transfection, no antibody pull-down, IgG antibody pull-down, and PCR using primers targeting the FUT1 gene. An additional positive control was included for CREB-transfected cells. In this control, PCR was performed using primers targeting the promoter region of the known CREB target, Pck1.

To identify interactions between CREBH and PPARα immunoprecipitation (IP)

Western blotting analyses were performed with liver tissue samples from mice under fasting or after AHF diet feeding (Figure 25). Liver tissues from wild-type mice under normal chow feeding, 14-hour overnight fasting, or after 6-month AHF diet feeding were homogenized and protein was isolated. Pull-down was performed with the anti-

CREBH or anti-PPARα antibody. Pull-down fraction and whole lysate were analyzed by Western blot and probed with the anti-CREBH or anti-PPARα antibody. Western blot showed the interaction between PPARα and CREBH that is particularly favored under the fasting or AHF diet-fed condition.

Figure 25. Immunoprecipation of liver protein lysates from wild-type mice under metabolic stress. Liver lysate from wild-type mice under normal chow, fasting, or AHF diet feeding. Precipitation was induced by pull-down with the anti-PPARα or anti- CREBH antibody. Pull-down fractions and whole lysate were probed with the PPARα or CREBH antibody. Levels of β-actin in the total lysates were determined as loading controls.

3.4 Discussion

Owing to the potential of Fgf21 as a key component of CREBH metabolic control, we sought to define the relationship underlying this regulation and its mechanism. This resulted in the discovery of the conserved and adjacent CRE and

PPRE-1/2 elements on the promoter region of the Fgf21 gene (figure 14). The conserved nature of the elements marks them as potentially integral for proper expression of Fgf21 . The proximity of these elements indicates that a PPAR protein family member may be functioning as a binding partner for CREBH during Fgf21 induction. A previous study had revealed that, under an AHF-diet, CREBH upregulates Ppar α expression. Figures 20 and 21 showed the shared regulatory pattern of CrebH - and Ppar α-null mice under feeding and fasting conditions. These analyses also identify PPARα as a mediator of fasting-induced Fgf21 up-regulation.

In order to determine the importance of the CRE and PPRE1/2 elements in

Fgf21 expression, we carried out a reporter assay using the wild-type Fgf21 gene promoter and its mutant isoforms (FM13, PE-M1, PE-M2, and PE-M12). The FM promoter with the site mutation in the CRE binding element was unable to produce reporter activity under CREBH overexpression conditions (Figure 17). This shows the importance of the CRE element. Indirectly, this indicates that Fgf21 expression is highly dependent on CREBH. The reporter activity linked to the PE-M1 promoter with the site mutation in one of the PPRE binding elements in the Fgf21 gene promoter was similar to that of the wild-type promoter in all testing conditions (Figures 16 and

17). Additionally, reporter activities driven by the PE-M2, which harbors a site mutation at the second PPRE binding element in the Fgf21 gene promoter, and PE-

M12, in which both PPRE elements were mutated, were similar in all overexpression conditions and both were lower than that of the wild-type promoter. This leads us to the conclusion that PPRE1 does not play a role in CREBH- or PPARα-dependent

Fgf21 upregulation, but PPRE2 does play a role in this regulation. Due to the higher level of reporter activity in cells that co-expressed Ppar α and CrebH , compared to expression of Pparα or CrebH alone, it is clear that both proteins are required to reach proper levels of Fgf21 expression.

The next step of our study was to determine the individual contributions of

CREBH and PPARα as inducers of Fgf21 . We attempted to rescue Fgf21 expression in both CrebH - and Ppar α-null mice with Pparα and CrebH overexpression, respectively (Figures 22 & 23). The results of this experiment showed that, while both

PPARα and CREBH play a role in the regulation of Fgf21 , exogenous expression of

CREBH alone is sufficient to induce Fgf21 transcription, but PPARα requires interaction with CrebH to accomplish the same result.

In Figure 25, we showed that inducing metabolic stress by fasting or AHF diet results in interaction between CREBH and PPARα. This indicates the existence of a

CREBH-PPARα complex under metabolic stress. By employing ChIP analyses, we were able to show direct binding of the CREBH-PPARα complex to the Fgf21 promoter (Figure 24). We were also able to show that the CRE binding site on the promoter is specific to CREBH. This leads us to conclude that PPARα interacts and forms a complex with CREBH under metabolic stress, such as fasting or AHF diet- feeding. This complex binds to the CRE and PPRE1/2 elements of the Fgf21 promoter region and initiates Fgf21 transcription. The transcriptional upregulation

60 results in increased FGF21 protein content in mouse liver and serum under fasting conditions (Figures 13 and 20).

CHAPTER IV: Exogenous FGF21 Administration Rescues CrebH - null Phenotype

4.1 Summary

CrebH -null mice are susceptible to AHF diet-induced NAFLD and NASH.

FGF21, a transcriptional target of CREBH, has proven to be a potent treatment option for these diseases. In this chapter, we describe the rescue of the NASH phenotype of

CrebH -null mice developed under the AHF diet by injections of recombinant human

FGF21 protein. Treatment with recombinant protein was an effective method of increasing serum FGF21 protein levels and decreasing body weight in CrebH-null mice when compared to vehicle control. Injection of FGF21 improved hyperlipidemia, hypoketonemia, hepatic steatosis, hepatocyte ballooning, and hepatomegaly.

Additionally, glucose tolerance was improved in AHF diet-fed CrebH -null mice upon

FGF21 protein treatment. The success of this treatment in rescuing the AHF diet-fed

CrebH -null mouse disease phenotype indicates the importance of FGF21 in the

CREBH-regulated metabolic stress response. Furthermore, effectiveness of the treatment puts FGF21 forward as a potential exciting therapeutic option for hypertriglyceridemia and NASH.

4.2 Materials and Methods

Materials

Synthetic oligonucleotides were purchased from Integrated DNA

Technologies, Inc. (Coralville, IA). The crude blood serum antibody was purified through affinity purification. Kits for measuring TG, FA, and ketone body 3- hydroxybutyric acid were from BioAssay Systems (Hayward, CA). Purified, bioactive recombinant human FGF21 protein was purchased from ProSpec-Tany TechnoGene

Ltd (Ness Ziona, Israel). Mouse monoclonal anti-FGF21 antibody and FGF21 ELISA kit were from R & D Systems, Inc. (Minneapolis, MN). OneTouch Ultra glucometer and blood glucose measurement strips were purchase form Johnson & Johnson, Inc.

Animal experiments

For the FGF21 rescue animal experiment, daily injections of purified, bioactive rhFGF21 protein and control injections of vehicle (sterile water) were carried out for 8 days. CrebH-null and wild-type control male mice after the AHF diet for 6 months were injected i.p. with 0.1 g purified recombinant human FGF21 protein per gram body weight or the vehicle PBS. On the fifth day of injection, mice were fasted for 12 hours. Following fasting, on the sixth day of injection, glucose tolerance tests were performed. Mice were allowed to recover for 2 days with ongoing injections before they were euthanized and tissues were harvested. All the animal experiments were approved by the Wayne State University IACUC committee and carried out under the institutional guidelines for ethical animal use.

Measurement of mouse lipid metabolites

Triton-100 in PBS for TG measurement using a commercial kit (BioAssay Systems,

CA). Mouse hepatic TG levels were determined by normalization to liver tissue mass used for TG measurement. Mouse blood plasma samples were subjected to quantitative analyses of TG, free FA, and the ketone body metabolite 3- hydroxybutyric acid using the commercial kits (BioAssay Systems).

Oil-red O staining

Frozen liver tissue sections were stained with Oil-red O for lipid contents according to standard protocol (107 ). Briefly, frozen liver tissue sections of 8 microns were air-dried, and then fixed in formalin. The fixed sections were rinsed with 60% isopropanol before they were stained with freshly prepared Oil-red O solution for 15 minutes. After Oil-red O staining, liver sections were rinsed again with 60% isopropanol followed by washing with water before subjected to microscope analysis.

Sirius-red staining of hepatic collagens

Sirius-red staining of collagen deposition in liver tissue section for fibrosis evaluation was as previously described (107 ). Briefly, paraffin-embedded mouse liver tissue sections (5μm) were pretreated with xylene, ethanol, and water according to the standard protocol. The liver tissue sections were stained for 1 hr with Sirius red solution (Sigma) and then washed with acidified water (0.5% acetic acid). The stained

64 tissue sections were dehydrated with 100% ethanol and cleared in xylene before being mounted with mounting solution.

GTT assay

Glucose tolerance tests of the CrebH-null and wild-type control mice after 6 days of administration of the FGF21 protein or vehicle. Animals were fasted for 12 hours before infused with 1 mg D-glucose per gram body weight. Blood glucose levels were measured from tail blood using an OneTouch Ultra glucometer (Johnson

& Johnson).

Statistics

4.3 Results

In order to determine whether Fgf21 is the major target of CREBH in regulating lipid metabolism and to evaluate the potential effect of FGF21 on reversing the NASH phenotype resulting from CrebH deletion, we fed the AHF diet to CrebH -null mice to establish the NASH disease model.

Figure 26. Effects of administration of recombinant FGF21 protein on serum FGF21 levels and body weights. (A) CrebH-null mice showed increased circulating levels of FGF21 upon FGF21 protein administration. (B) FGF21 protein administration induced a modest but significant decrease in body weight compared to the vehicle control. Each bar denotes the mean ± SEM (n=5 mice treated with vehicle and 4 mice treated with FGF21). * P≤0.05.

We administered a series of recombinant human FGF21 protein injections to

CrebH -null mice after the AHF diet. FGF21 protein was administered by intraperitoneal injection daily for 8 days. Effectiveness of FGF21 protein administration was assessed by FGF21 serum concentration after the series of injections in CrebH -null mice (Figure 26). FGF21 protein administration successfully increased circulating FGF21 protein levels. Additionally, since long term FGF21 protein administration has been shown to induce weight loss in mice, mouse body

66 weight was measured upon the 8-day administration of FGF21 protein (82, 84 ).

Protein injection was accompanied by small but significant decrease in body weight for the FGF21-injected mice (Figure 26).

Figure 27. Infusion of recombinant FGF21 protein reverses hypertriglyceridemia, and hypoketonemia in CrebH-null mice after the AHF diet. CrebH-null (KO) and wild-type (WT) control mice were fed with AHF diet for 6 months before they were injected i.p. with purified recombinant FGF21protein or vehicle once a day for 8 days. Levels of serum TG (A), ketone body BOH (B), and FA (C) in the WT and CrebH-null mice administrated with recombinant FGF21 protein or vehicle (Veh). The blood samples were collected from the mice after 12-hour fasting on the sixth day of administration of FGF21 protein or vehicle. Each bar denotes the mean ± SEM (n = 5 WT or KO mice treated with vehicle and 4 KO mice treated with FGF21). *, p ≤ .05; **, P ≤ 0.01.

To determine the ability of FGF21 to rescue metabolic syndrome in AHF diet- fed CrebH -null mice, wild-type and CrebH -null mice were fed an AHF-diet for 6 months. After the feeding period, CrebH -null mice were administered either FGF21 protein or vehicle control (PBS). Wild-type mice were given vehicle control only.

CrebH -null mice that received FGF21 injection showed improved lipid metabolite

67 profiles and liver pathological states compared to vehicle-injected CrebH -null mice controls (Figure 27). Vehicle-injected control CrebH -null mice exhibit increased TG and free FA levels in serum that indicate onset of NAFLD and metabolic syndrome.

Additionally, serum ketone levels were reduced compared to wild-type controls. In contrast, FGF21-injected mice showed comparable levels of serum TG and free FA to wild-type controls. These mice also showed increased serum ketone levels compared to both wild-type and CrebH -null mice with vehicle treatment. These results indicate that FGF21 is the major downstream factor targeted by CREBH in regulating TG and FA metabolism. This also shows that the capacity of FGF21 to treat hyperlipidemia extends to diseases that are caused by CREBH deficiency.

Figure 28. Infusion of recombinant FGF21 protein partially rescued NASH phenotype of the CrebH-null mice after the AHF diet. CrebH-null (KO) and wild-type (WT) control mice were fed with AHF diet for 6 months before they were injected i.p. with purified recombinant FGF21protein or vehicle once a day for 8 days. Histologic staining analysis of liver tissue sections from the wild-type and CrebH-null mice after 8 days of FGF21 protein or vehicle treatment. (A) Oil-red O staining of hepatic lipid droplets. (B) Hematoxylin and eosin (HE) staining of liver tissue sections. (C) Sirius red staining of hepatic collagens. Magnification: 200x.

FGF21 protein injection was also able to rescue hepatic symptoms of NASH.

Oil Red O staining of lipid droplets in liver tissue showed that the AHF-diet regimen was successful in inducing NAFLD in vehicle-treated CrebH -null mice (Figure 28A).

This effect was ameliorated CrebH -null mice that were injected with FGF21 protein.

Enzymatic analysis of liver TG levels confirmed reduced TG content in FGF21- injected CrebH -null mice (Figure 29A). Hepatocyte ballooning was reduced in CrebH - null mice after FGF21 treatment, as well (Figure 28B). This was corroborated by a significant decrease in hepatocyte size (Figure 29B). Portal and lobular inflammation were unaffected by FGF21 treatment CrebH -null mice (Figure 28B). Fibrosis

(assessed by Sirius Red staining of collagen) was not decreased in these mice, either

(Figure 28C). The long-term stress caused by AHF diet-feeding (6 months) may have caused morphological changes that were unable to be overcome by the relatively short treatment period (8 days). Effects of FGF21 injection were detectable from a gross anatomical observation (Figure 30). CrebH -null mice that received FGF21 treatment displayed improved liver size and coloration compared to vehicle-injected

CrebH -null mice. Additionally, FGF21, as a potent metabolic hormone, modulates lipid metabolism in adipose tissues (110-112 ).

Figure 29. Administration of recombinant FGF21 protein reverses hepatic steatosis phenotype developed in the CrebH-null mice under the AHF diet. (A) Levels of hepatic TG in the livers of the wild-type (WT) and CrebH-null mice treated with recombinant FGF21 protein (FGF21) or vehicle. (B) Fold change of hepatocyte sizes in the livers of the wild-type and CrebH-null mice administrated with recombinant FGF21 protein or vehicle. For the tissue section from each animal, sizes of hepatocytes from 3 representative tissue areas per mouse were calculated to obtain the average hepatocyte size for each treatment group. Each bar denotes the mean ± SEM (n=5 WT or KO mice treated with vehicle and 4 KO mice treated with FGF21). * P ≤ 0.05.

Figure 30. Gross anatomical inspection of liver morphology after FGF21 protein administration.

Because NASH typically presents with glucose intolerance and insulin resistance, we assessed the effect of FGF21 protein injection on the improvement of glucose tolerance in CrebH -null mice (64 ). GTT was performed and indicated improved glucose tolerance in CrebH -null mice injected with FGF21 compared to

CrebH -null mice injected with vehicle control (Figure 31). These results not only confirm that FGF21 is the major CREBH-target in regulating lipid metabolism, but also indicate that FGF21 protein administration may be an effective treatment to rescue NASH and its related metabolic symptoms.

Figure 31. GTT analysis with CrebH-null (KO) and wild-type (WT) control mice after administration with vehicle PBS or FGF21 protein. KO and WT control mice were fed with AHF diet for 6 months before they were injected i.p. with purified recombinant FGF21protein or vehicle once a day for 6 days. Mice were subjected to fasting blood sample collection and glucose tolerance test after 12-hour fasting Glucose tolerance tests of the CrebH-null and wild-type control mice after administration of the FGF21 protein or vehicle. Animals were fasted for 12 hours before being infused with 1 mg D-glucose per gram body weight. Each bar denotes the mean ± SEM (n = 5 WT or KO mice treated with vehicle and 4 KO mice treated with FGF21).

4.4 Discussion

Having characterized the molecular mechanism by which CREBH regulates

Fgf21 expression, we sought to delineate the functional impact of the CREBH-FGF21 regulatory axis in lipid homeostasis and metabolic disorders. CrebH -null mice develop profound hyperlipidemia and NASH under the AHF diet (Figure 5) (12 ).

Therefore, the AHF-fed CrebH -null animal model is a perfect model system for evaluating therapeutic potential of FGF21 protein treatment in metabolic symptoms associated with impaired lipid metabolism. Through administration of recombinant

FGF21 protein to the CrebH -null mice, we demonstrated the effects of FGF21 treatment in decreasing blood TG and FA of the CrebH -null mice (Figure 27). FGF21 treatment can partially rescue hepatic inflammation and fibrosis in the CrebH-null mice after the AHF diet (Figures 28 and 29). These results confirm that FGF21 is a major target of CREBH in regulating lipid homeostasis under metabolic stress. Our findings also identify FGF21 as a potential therapeutic agent in the treatment of

NASH.

The significance of these findings is also implicated by the fact that the transcriptional regulation of Fgf21 by CREBH that we have outlined is reflected at the physiological level. The CREBH-dependent increase in serum FGF21 levels coupled with the ability of circulating FGF21 to induce signaling at a systemic level opens up the possibility that CREBH may have a more direct role in modulating metabolism outside of the liver.

CrebH -null mice are susceptible to diet-induced NASH. FGF21 has been shown to rescue metabolic disorders in diet-induced disease models. We performed

73 similar experiments using our CrebH -null model of NASH. Previous reports of FGF21 protein injection experiments indicate that this procedure induced significant weight loss in mice that have been fed a high-fat diet. In our study, administration of FGF21 protein to the CrebH -null mice for 8 days led to a modest loss in body weight (Figure

26B). This is likely due to the relatively short time frame of the treatment over which the experiment was carried out. It is likely that a more sustained regimen would produce more drastic weight loss in our mouse model. To further support the effectiveness of the treatment, serum FGF21 levels radically increased in mice that received the rescue treatment (Figure 26A). The ability of the FGF21 treatment to rescue the hyperlipidemia and hepatocyte size increase present in the CrebH -null disease model mice implicates that CREBH can significantly mediate hepatic metabolic state by modulating Fgf21 expression.

The effectiveness of the treatment in reversing the hyperlipidemia associated with our model shows that these effects can reach the circulating serum and have systemic effects. Functional mutations in CREBH have already been identified as a contributing factor in hypertriglyceridemia in human patients (70 ). FGF21 is currently a topic of interest as a therapeutic option. The findings from my rescue experiment confirm the pleiotropic effects of exogenous FGF21 administration on lipid metabolism and its promising therapeutic potential in metabolic disorders. Our study offers important evidence for FGF21 as a potential treatment option for patients suffering from hypertriglyceridemia linked to CREBH mutation.

CHAPTER V: Conclusions and Significance

CREBH is emerging as a key player in hepatic response to metabolic stress.

This work identifies the phenotypic effect of CrebH deficiency under various metabolic stress conditions. Previous studies suggested a major role for CREBH in lipid metabolism and gluconeogenesis under high-fat diet-induced stress. However, the method by which CREBH regulates complex metabolic programs involved in multiple pathways under metabolic stress remains elusive. My studies established that CREBH plays this role under fasting-induced stress and that the length of the fast affects the regulation. Additionally, it was demonstrated that CREBH regulates ketogenesis by targeting Bdh1 and Hmgcs2 . Importantly, Fgf21 was identified as a target of CREBH. Because of the importance of FGF21 as a metabolic hormone, we focused our attention on defining the mechanism of this regulation. At the molecular lever, CREBH interacts with PPARα to activate FGF21 expression through a multi- layered regulatory mechanism. CREBH and PPARα regulate each other at the transcriptional level in liver upon metabolic stress, such fasting or AHF diet (Figure 20 and 21). Metabolic stress-induced expression of Fgf21 requires both CREBH and

PPARα. The coordination between CREBH and PPARα in regulating Fgf21 expression involves protein-protein interaction during fasting or AHF diet-feeding

(Figure 25). Inspection of the Fgf21 promoter led to the recognition of the juxtaposed

CRE and PPRE binding sites where CREBH and PPARα are enriched under metabolic stress (Figure 15). Through protein-protein and protein-DNA interactions,

CREBH and PPARα function in synergy to activate expression of the Fgf21 gene.

CREBH and PPARα appear to act as binary trans-activators that regulate each other

75 for their expression and interact with each other for their function. In the future, understanding how CREBH interacts with other proteins to modify its transcriptional effects will be key to understanding the regulatory mechanism of CREBH in stress response.

An important part of my work is the demonstration of the therapeutic potential of FGF21 in the treatment of NASH and its common comorbidities. By administering the AHF diet to CrebH -null mice, previous studies in our lab have generated a disease model of NASH that is dependent on CREBH deficiency (12 ). CrebH -null mice develop profound hypertriglyceridemia, hypoketonemia, and NASH under the

AHF diet. Therefore, the AHF-fed CrebH -null mouse is a perfect model system for evaluating the therapeutic potential of FGF21 protein as a treatment option in NASH and its associated symptoms. Owing to the severity of NASH and associated metabolic dysfunction, as well as the increasing interest in FGF21 as a therapeutic option, we chose to treat our disease model mice using recombinant human FGF21 protein. CrebH -null mice showed a partial recovery from the NASH state in response to this treatment. Previous studies have used exogenous FGF21 protein to treat metabolic disease model animals (82, 86 ). However, the effectiveness of FGF21 protein in the treatment of NASH and CREBH-deficiency were novel findings.

Because this treatment lowered serum TG levels and human CREBH deficient patients suffer from hypertriglyceridemia, this discovery is pertinent to the management of human disease.

This body of work was created with the goal of understanding of the mechanism of CREBH regulatory action. Based on my findings, I can make three

76 conclusions that make strong headway in achieving this goal (Figure 32). First,

CREBH is essential for hepatic adaptation to metabolic stress, particularly for the response involved in lipid metabolism. Second, FGF21 plays a role in mediating these effects and CREBH works in tandem with PPARα to promote Fgf21 expression.

Third, the metabolic disease phenotype observed in CrebH -deficient mice under an

AHF diet is relieved by exogenous FGF21 administration. Therefore, these findings extend the relationship between the nuclear receptors and ER-associated trans- activators in regulating energy metabolism.

However, my research does not fully describe the role of CREBH in metabolic regulation. Much work remains to be done in this field in order to meet that goal. Of particular importance is the role of CREBH in the regulation of metabolism under high-sucrose diet-induced stress. Fasting and high-fat diet-feeding activate some common response pathways, such as lipolysis and ketogenesis (113-116 ). High- sucrose diet-feeding activates a distinct set of responses, including glycolysis and lipogenesis(117 ). The potential activation of CREBH by a stress factor that differs so fundamentally from previously described inducers greatly broadens the prospective range of CREBH-mediated metabolic control. Addressing these questions in the future will provide novel insights into the key transcriptional programs that control energy homeostasis and has important implications in the understanding and treatment of metabolic disease.

Figure 32. Proposed model of CREBH-mediated Fgf21 regulation. Under fasting or high-fat diet feeding conditions, CREBH associates with PPARα at the Fgf21 promoter region under metabolic stress. Fgf21 is up-regulated at the RNA and protein levels. Lipid and ketone metabolism pathways are activated. The original metabolic stress is alleviated by the action of the CREBH-PPARα-FGF21 regulatory axis.

APPENDIX A

Gene name Forward primer Reverse primer CrebH TCGTGCCAGTGCGAGTGT AGCCACGCGGGATGC PPARα GAAGGGCACACGCGTGCGAGTTTTCAG CTGTGATGACAACGTCTTGTTCCCGAACT Apoc2 CTCTGCTGGGCACGGTGCA GCCGCCGAGCTTTTGCTGTAC Apoa4 GCATCTAGCCCAGGAAACTG ATGTATGGGGTCAGCTGGAG Fgf21 GTGTCAAAGCCTCTAGGTTTCTT GGTACACATTGTAACCGTCCTC ApoA5 TCCTCGCAGTGTTCGCAAG GAAGCTGCCTTTCAGGTTCTC Bdh1 TGACACCCGTCGGACCTAC TTCTCAGTCGGTCACTCTTCA Hmgcs2 GAAGAGAGCGATGCAGGAAAC GTCCACATATTGGGCTGGAAA Ldha TGTCTCCAGCAAAGACTACTGT GACTGTACTTGACAATGTTGGGA

Sequence information for primers used in quantitative real-time PCR studies.

APPENDIX B

Human FGF21 Promoter

1 CGAGTGGAGC GATCGTCCGC CCTCTGCCGC GTCCTCCTGG GGGATGGAGA ACCTGTTTGC 61 CAGGCTGCCG CCGGAGCGCG CGAGGGCCTC CAGCCTGCAG GTGCTCCCGC GCC CACCAGG 121 GTCTCCGATC TCCCA CACCC CCCGCCTCAC CGTCCACACT CAAATTCCAG AATCCCAGCC 181 CCTAGTCACT CCTTTTGGGG AATTCTAGGG TCCCGTGCCA AACTGCTACT TTCCAGACCT 241 GCACTGCACG AAGGCAACCA TTCCAAGGCT GGAGACCGGA GTGGAACAGG GTCTGTTTCC 301 CACGCCGTCA GCGGGGGCCG TGGCTGCCGG CGCCAGAGAC AAAGGCCTCT GCCCGCTTGT 361 CACCTAATTA AAGCATTTTC TCTGGTCCAA GGCCCACCTC CAGCATCGCC AACCCCGAGC CAGGTT CCGGGTGGAG GTCGT 421 AGGGCAGGCT AACGTAGGGT CCAGCCCTAC GGGGCTTTGC GGGTGGGTGT TCTTCCCCTG 481 TGAGGAGAGG AGAGATTGTA AGAAATAAAG ACACAAGACA AAGAGATAAA GAGAAAACAG 541 CTGGGCCCGG GGGACCACTA CCATCAAGAC GCAGAGACCG GTAGTGGCCC CGAATGGCTG ------2641 AGATGGTCCA AGAGGTGGTT TTTCCAGGAG CCCAGCACCC CTCCTCCCTC CGACTCAGGT 2701 GCTTGAGACC CCAGATCCTT CTCTCTGAGA CTCAGGAATG TGGGCCCCCA GCCCCTTTCA 2761 CCTGGGTCCC AGCTAACCCG ATCCTCCCCT CCCTCATCCC CTAGACCCAG GAGTCTGGCC 2821 CTCCATTGAA AGGACCCCAG GTTACATCAT CC ATTCAGGC TGCCCTTGCC ACG ATGGAAT 2881 TCTGTAGCTC CTGCCAAAT G GGTCA AATAT CA TG GTTCA G GCGCAGGGAG GGTGATTGGG PPRE2 PPRE1 CREB

2941 CGGGCCTGTC TGGGTATAAA TTCTGGAGCT TCTGCATCTA TCCCAAAAAA CAAGGGTGTT 3001 CTGTCAGCTG AGGATCCAGC CGAAAGAGGA GCCAGGCACT CAGGCCACCT GAGTCTACTC CG GCTTTCTCCT CGGTCCG 3061 ACCTGGACAA CTGGAATCTG GCACCAATTC TAAACCACTC AGCTTCTCCG AGCTCACACC CDS 3121 CCGGAGATCA CCTGAGGACC CGAGCCATTG ATG GACTCGG ACGAGACCGG GTTCGAGCAC 3181 TCAGGACTGT GGGTTTCTGT GCTGGCTGGT CTTCTGCTGG GAGCCTGCCA GGCACACCCC 3241 ATCCCTGACT CCAGTCCTCT CCTGCAATTC GGGGGCCAAG TCCGGCAGCG GTACCTCTAC 3301 ACAGATGATG CCCAGCAGAC AGAAGCCCAC CTGGAGATCA GGGAGGATGG GACGGTGGGG 3361 GGCGCTGCTG ACCAGAGCCC CGAAAGTGAG TGTGGGCCAG AGCCTGGGTC TGAGGGAGGA 3421 GGGGCTGTGG GTCTGGATTC CTGGGTCTGA GGGAGGAGGG GCTGGGGGCC TTGGCCCCTG 3481 GGTCTGAGGG AGGAGGGGCT GGGGATCTGG ACTCCTGGGT CTGAGGGAGG AGGGGCTGGG 3541 GATCTGGGCC CCTGGGTCTG AGGGAGGAGG GGCTGGGTCT GGACCCCTGG GTCTGAGGGA 3601 GGAGGGGCTG GGGGTCTGGA CTCTTGGGTT TGAAGAAGGA AGGGCTGGGG TCCTGGACTC 3661 TTGGGTCTGA GTTGGGAGGG GGCTTTGGCT TGGGCTTCTC CTGGGTCTGA GGGAGGAGGT 3721 AGGCTGTGGG CTTGGACTCC CAGGGCTGGG ACAGAGCCGG ATGGTGGGAC AGAGTCGGGT 3781 GGTGGGACAG TCCCGGGTGG GAGAGGTCCT CGAACCACCT TATCGCTTTC ACCCCTTAGG 3841 TCTCCTGCAG CTGAAAGCCT TGAAGCCGGG AGTTATTCAA ATCTTGGGAG TCAAGACATC 3901 CAGGTTCCTG TGCCAGCGGC CAGATGGGGC CCTGTATGGA TCGGTGAGTT TCCAGGACCC 3961 TCCTCACCAC CCACCATGCT CCTCCTATAT GTCGCCCTCA C

hFGF21 ChIP (CREBH)-B

hFGF21 ChIP (NS)-B: ChIP negative control primers

Sequence of the human FGF21 gene promoter highlighting potential CREBH and PPARα binding sites, FGF21 ChIP primers, and ChIP negative control primers.

REFERENCES

(1) Jump, D.B., Botolin, D., Wang, Y., Xu, J., Christian, B., and Demeure, O. (2005) Fatty acid

regulation of hepatic gene transcription. The Journal of nutrition . 135 , 2503-2506

(2) Nguyen, P., Leray, V., Diez, M., Serisier, S., Le Bloc'h, J., Siliart, B., and Dumon, H. (2008) Liver

lipid metabolism. Journal of animal physiology and animal nutrition . 92 , 272-283

(3) Saltiel, A.R., and Kahn, C.R. (2001) Insulin signalling and the regulation of glucose and lipid

metabolism. Nature . 414 , 799-806

(4) Xu, J., Nakamura, M.T., Cho, H.P., and Clarke, S.D. (1999) Sterol regulatory element binding

protein-1 expression is suppressed by dietary polyunsaturated fatty acids. A mechanism for

the coordinate suppression of lipogenic genes by polyunsaturated fats. The Journal of

biological chemistry . 274 , 23577-23583

(5) Zelcer, N., and Tontonoz, P. (2006) Liver X receptors as integrators of metabolic and

inflammatory signaling. The Journal of clinical investigation . 116 , 607-614

(6) Schoonjans, K., Peinado-Onsurbe, J., Lefebvre, A.M., Heyman, R.A., Briggs, M., Deeb, S.,

Staels, B., and Auwerx, J. (1996) PPARalpha and PPARgamma activators direct a distinct

tissue-specific transcriptional response via a PPRE in the lipoprotein lipase gene. The EMBO

journal . 15 , 5336-5348

(7) Schoonjans, K., Staels, B., and Auwerx, J. (1996) The peroxisome proliferator activated

receptors (PPARS) and their effects on lipid metabolism and adipocyte differentiation.

Biochimica et biophysica acta . 1302 , 93-109

(8) Luebke-Wheeler, J., Zhang, K., Battle, M., Si-Tayeb, K., Garrison, W., Chhinder, S., Li, J.,

Kaufman, R.J., and Duncan, S.A. (2008) Hepatocyte nuclear factor 4alpha is implicated in

endoplasmic reticulum stress-induced acute phase response by regulating expression of

cyclic adenosine monophosphate responsive element binding protein H. Hepatology . 48 ,

1242-1250

(9) Zhang, K., Shen, X., Wu, J., Sakaki, K., Saunders, T., Rutkowski, D.T., Back, S.H., and Kaufman,

R.J. (2006) Endoplasmic reticulum stress activates cleavage of CREBH to induce a systemic

inflammatory response. Cell . 124 , 587-599

(10) Omori, Y., Imai, J., Watanabe, M., Komatsu, T., Suzuki, Y., Kataoka, K., Watanabe, S.,

Tanigami, A., and Sugano, S. (2001) CREB-H: a novel mammalian transcription factor

belonging to the CREB/ATF family and functioning via the box-B element with a liver-specific

expression. Nucleic acids research . 29 , 2154-2162

(11) Lee, J.H., Giannikopoulos, P., Duncan, S.A., Wang, J., Johansen, C.T., Brown, J.D., Plutzky, J.,

Hegele, R.A., Glimcher, L.H., and Lee, A.H. (2011) The transcription factor cyclic AMP-

responsive element-binding protein H regulates triglyceride metabolism. Nature medicine .

17 , 812-815

(12) Zhang, C., Wang, G., Zheng, Z., Maddipati, K.R., Zhang, X., Dyson, G., Williams, P., Duncan,

S.A., Kaufman, R.J., and Zhang, K. (2012) Endoplasmic reticulum-tethered transcription factor

cAMP responsive element-binding protein, hepatocyte specific, regulates hepatic lipogenesis,

fatty acid oxidation, and lipolysis upon metabolic stress in mice. Hepatology . 55 , 1070-1082

(13) Lee, M.W., Chanda, D., Yang, J., Oh, H., Kim, S.S., Yoon, Y.S., Hong, S., Park, K.G., Lee, I.K.,

Choi, C.S., Hanson, R.W., Choi, H.S., and Koo, S.H. (2010) Regulation of hepatic

gluconeogenesis by an ER-bound transcription factor, CREBH. Cell metabolism . 11 , 331-339

(14) Postic, C., and Girard, J. (2008) Contribution of de novo fatty acid synthesis to hepatic

steatosis and insulin resistance: lessons from genetically engineered mice. The Journal of

clinical investigation . 118 , 829-838

(15) Wu, C., Khan, S.A., Peng, L.J., Li, H., Carmella, S.G., and Lange, A.J. (2006) Perturbation of

glucose flux in the liver by decreasing F26P2 levels causes hepatic insulin resistance and

hyperglycemia. American journal of physiology. Endocrinology and metabolism . 291 , E536-

543

(16) Terrettaz, J., Assimacopoulos-Jeannet, F., and Jeanrenaud, B. (1986) Inhibition of hepatic

glucose production by insulin in vivo in rats: contribution of glycolysis. The American journal

of physiology . 250 , E346-351

(17) Rhee, J., Inoue, Y., Yoon, J.C., Puigserver, P., Fan, M., Gonzalez, F.J., and Spiegelman, B.M.

(2003) Regulation of hepatic fasting response by PPARgamma coactivator-1alpha (PGC-1):

requirement for hepatocyte nuclear factor 4alpha in gluconeogenesis. Proceedings of the

National Academy of Sciences of the United States of America . 100 , 4012-4017

(18) Pegorier, J.P., Garcia-Garcia, M.V., Prip-Buus, C., Duee, P.H., Kohl, C., and Girard, J. (1989)

Induction of ketogenesis and fatty acid oxidation by glucagon and cyclic AMP in cultured

hepatocytes from rabbit fetuses. Evidence for a decreased sensitivity of carnitine

palmitoyltransferase I to malonyl-CoA inhibition after glucagon or cyclic AMP treatment. The

Biochemical journal . 264 , 93-100

(19) Pilkis, S.J., el-Maghrabi, M.R., and Claus, T.H. (1988) Hormonal regulation of hepatic

gluconeogenesis and glycolysis. Annual review of biochemistry . 57 , 755-783

(20) Chen, Y., Siewers, V., and Nielsen, J. (2012) Profiling of cytosolic and peroxisomal acetyl-CoA

metabolism in Saccharomyces cerevisiae. PloS one . 7, e42475

(21) Nuttall, F.Q., Ngo, A., and Gannon, M.C. (2008) Regulation of hepatic glucose production and

the role of gluconeogenesis in humans: is the rate of gluconeogenesis constant?

Diabetes/metabolism research and reviews . 24 , 438-458

(22) Roach, P.J., Depaoli-Roach, A.A., Hurley, T.D., and Tagliabracci, V.S. (2012) Glycogen and its

metabolism: some new developments and old themes. The Biochemical journal . 441 , 763-787

(23) Kersten, S. (2001) Mechanisms of nutritional and hormonal regulation of lipogenesis. EMBO

reports . 2, 282-286

(24) Wakil, S.J., Stoops, J.K., and Joshi, V.C. (1983) Fatty acid synthesis and its regulation. Annual

review of biochemistry . 52 , 537-579

(25) Nagle, C.A., Klett, E.L., and Coleman, R.A. (2009) Hepatic triacylglycerol accumulation and

insulin resistance. Journal of lipid research . 50 Suppl , S74-79

(26) McGarry, J.D., and Foster, D.W. (1980) Regulation of hepatic fatty acid oxidation and ketone

body production. Annual review of biochemistry . 49 , 395-420

(27) Cotter, D.G., Ercal, B., Huang, X., Leid, J.M., d'Avignon, D.A., Graham, M.J., Dietzen, D.J.,

Brunt, E.M., Patti, G.J., and Crawford, P.A. (2014) Ketogenesis prevents diet-induced fatty

liver injury and hyperglycemia. The Journal of clinical investigation . 124 , 5175-5190

(28) Garber, A.J., Menzel, P.H., Boden, G., and Owen, O.E. (1974) Hepatic ketogenesis and

gluconeogenesis in humans. The Journal of clinical investigation . 54 , 981-989

(29) Laffel, L. (1999) Ketone bodies: a review of physiology, pathophysiology and application of

monitoring to diabetes. Diabetes/metabolism research and reviews . 15 , 412-426

(30) Cahill, G.F., Jr., Herrera, M.G., Morgan, A.P., Soeldner, J.S., Steinke, J., Levy, P.L., Reichard,

G.A., Jr., and Kipnis, D.M. (1966) Hormone-fuel interrelationships during fasting. The Journal

of clinical investigation . 45 , 1751-1769

(31) Perry, R.J., Samuel, V.T., Petersen, K.F., and Shulman, G.I. (2014) The role of hepatic lipids in

hepatic insulin resistance and type 2 diabetes. Nature . 510 , 84-91

(32) Ramnanan, C.J., Edgerton, D.S., Kraft, G., and Cherrington, A.D. (2011) Physiologic action of

glucagon on liver glucose metabolism. Diabetes, obesity & metabolism . 13 Suppl 1 , 118-125

(33) Jiang, G., and Zhang, B.B. (2003) Glucagon and regulation of glucose metabolism. American

journal of physiology. Endocrinology and metabolism. 284 , E671-678

(34) McGarry, J., Wright, P.H., and Foster, D.W. (1975) Hormonal control of ketogenesis. Rapid

activation of hepatic ketogenic capacity in fed rats by anti-insulin serum and glucagon. The

Journal of clinical investigation . 55 , 1202-1209

(35) Nemeth, E., and Ganz, T. (2006) Regulation of iron metabolism by hepcidin. Annual review of

nutrition . 26 , 323-342

(36) Crunkhorn, S. (2013) Metabolic disorders: Betatrophin boosts beta-cells. Nature reviews.

Drug discovery . 12 , 504

(37) Woo, Y.C., Xu, A., Wang, Y., and Lam, K.S. (2013) Fibroblast growth factor 21 as an emerging

metabolic regulator: clinical perspectives. Clinical endocrinology . 78 , 489-496

(38) Zhang, R., and Abou-Samra, A.B. (2014) A dual role of lipasin (betatrophin) in lipid

metabolism and glucose homeostasis: consensus and controversy. Cardiovascular

diabetology . 13 , 133

(39) Pegorier, J.P., Le May, C., and Girard, J. (2004) Control of gene expression by fatty acids. The

Journal of nutrition . 134 , 2444S-2449S

(40) Kersten, S., Seydoux, J., Peters, J.M., Gonzalez, F.J., Desvergne, B., and Wahli, W. (1999)

Peroxisome proliferator-activated receptor alpha mediates the adaptive response to fasting.

The Journal of clinical investigation . 103 , 1489-1498

(41) Mandard, S., Muller, M., and Kersten, S. (2004) Peroxisome proliferator-activated receptor

alpha target genes. Cellular and molecular life sciences : CMLS . 61 , 393-416

(42) Sanderson, L.M., Boekschoten, M.V., Desvergne, B., Muller, M., and Kersten, S. (2010)

Transcriptional profiling reveals divergent roles of PPARalpha and PPARbeta/delta in

regulation of gene expression in mouse liver. Physiological genomics . 41 , 42-52

(43) Vu-Dac, N., Schoonjans, K., Kosykh, V., Dallongeville, J., Fruchart, J.C., Staels, B., and Auwerx,

J. (1995) Fibrates increase human apolipoprotein A-II expression through activation of the

peroxisome proliferator-activated receptor. The Journal of clinical investigation . 96 , 741-750

(44) Prieur, X., Coste, H., and Rodriguez, J.C. (2003) The human apolipoprotein AV gene is

regulated by peroxisome proliferator-activated receptor-alpha and contains a novel farnesoid

X-activated receptor response element. The Journal of biological chemistry . 278 , 25468-

25480

(45) Leone, T.C., Weinheimer, C.J., and Kelly, D.P. (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. Proceedings of the

National Academy of Sciences of the United States of America . 96 , 7473-7478

(46) Barrero, M.J., Camarero, N., Marrero, P.F., and Haro, D. (2003) Control of human carnitine

palmitoyltransferase II gene transcription by peroxisome proliferator-activated receptor

through a partially conserved peroxisome proliferator-responsive element. The Biochemical

journal . 369 , 721-729

(47) Tobin, K.A., Steineger, H.H., Alberti, S., Spydevold, O., Auwerx, J., Gustafsson, J.A., and Nebb,

H.I. (2000) Cross-talk between fatty acid and cholesterol metabolism mediated by liver X

receptor-alpha. Molecular endocrinology . 14 , 741-752

(48) Ulven, S.M., Dalen, K.T., Gustafsson, J.A., and Nebb, H.I. (2005) LXR is crucial in lipid

metabolism. Prostaglandins, leukotrienes, and essential fatty acids . 73 , 59-63

(49) Peet, D.J., Turley, S.D., Ma, W., Janowski, B.A., Lobaccaro, J.M., Hammer, R.E., and

Mangelsdorf, D.J. (1998) Cholesterol and bile acid metabolism are impaired in mice lacking

the nuclear oxysterol receptor LXR alpha. Cell . 93 , 693-704

(50) Foretz, M., Pacot, C., Dugail, I., Lemarchand, P., Guichard, C., Le Liepvre, X., Berthelier-

Lubrano, C., Spiegelman, B., Kim, J.B., Ferre, P., and Foufelle, F. (1999) ADD1/SREBP-1c is

required in the activation of hepatic lipogenic gene expression by glucose. Molecular and

cellular biology . 19 , 3760-3768

(51) Rome, S., Lecomte, V., Meugnier, E., Rieusset, J., Debard, C., Euthine, V., Vidal, H., and Lefai,

E. (2008) Microarray analyses of SREBP-1a and SREBP-1c target genes identify new regulatory

pathways in muscle. Physiological genomics . 34 , 327-337

(52) Owen, J.L., Zhang, Y., Bae, S.H., Farooqi, M.S., Liang, G., Hammer, R.E., Goldstein, J.L., and

Brown, M.S. (2012) Insulin stimulation of SREBP-1c processing in transgenic rat hepatocytes

requires p70 S6-kinase. Proceedings of the National Academy of Sciences of the United States

of America . 109 , 16184-16189

(53) Iizuka, K. (2013) Recent progress on the role of ChREBP in glucose and lipid metabolism.

Endocrine journal . 60 , 543-555

(54) Iizuka, K., Bruick, R.K., Liang, G., Horton, J.D., and Uyeda, K. (2004) Deficiency of carbohydrate

response element-binding protein (ChREBP) reduces lipogenesis as well as glycolysis.

Proceedings of the National Academy of Sciences of the United States of America . 101 , 7281-

7286

(55) Ishii, S., Iizuka, K., Miller, B.C., and Uyeda, K. (2004) Carbohydrate response element binding

protein directly promotes lipogenic enzyme gene transcription. Proceedings of the National

Academy of Sciences of the United States of America. 101 , 15597-15602

(56) Dentin, R., Benhamed, F., Hainault, I., Fauveau, V., Foufelle, F., Dyck, J.R., Girard, J., and

Postic, C. (2006) Liver-specific inhibition of ChREBP improves hepatic steatosis and insulin

resistance in ob/ob mice. Diabetes . 55 , 2159-2170

(57) Alba, L.M., and Lindor, K. (2003) Review article: Non-alcoholic fatty liver disease. Alimentary

pharmacology & therapeutics . 17 , 977-986

(58) Gentile, C.L., and Pagliassotti, M.J. (2008) The role of fatty acids in the development and

progression of nonalcoholic fatty liver disease. The Journal of nutritional biochemistry . 19 ,

567-576

(59) Miyazaki, M., Flowers, M.T., Sampath, H., Chu, K., Otzelberger, C., Liu, X., and Ntambi, J.M.

(2007) Hepatic stearoyl-CoA desaturase-1 deficiency protects mice from carbohydrate-

induced adiposity and hepatic steatosis. Cell metabolism . 6, 484-496

(60) Day, C.P., and James, O.F. (1998) Steatohepatitis: a tale of two "hits"? Gastroenterology . 114 ,

842-845

(61) Csak, T., Ganz, M., Pespisa, J., Kodys, K., Dolganiuc, A., and Szabo, G. (2011) Fatty acid and

endotoxin activate inflammasomes in mouse hepatocytes that release danger signals to

stimulate immune cells. Hepatology . 54 , 133-144

(62) Takaki, A., Kawai, D., and Yamamoto, K. (2013) Multiple hits, including oxidative stress, as

pathogenesis and treatment target in non-alcoholic steatohepatitis (NASH). International

journal of molecular sciences . 14 , 20704-20728

(63) Newton, J.L. (2010) Systemic symptoms in non-alcoholic fatty liver disease. Digestive

diseases . 28 , 214-219

(64) Fabbrini, E., Magkos, F., Mohammed, B.S., Pietka, T., Abumrad, N.A., Patterson, B.W.,

Okunade, A., and Klein, S. (2009) Intrahepatic fat, not visceral fat, is linked with metabolic

complications of obesity. Proceedings of the National Academy of Sciences of the United

States of America . 106 , 15430-15435

(65) Sorbi, D., Boynton, J., and Lindor, K.D. (1999) The ratio of aspartate aminotransferase to

alanine aminotransferase: potential value in differentiating nonalcoholic steatohepatitis from

alcoholic liver disease. The American journal of gastroenterology . 94 , 1018-1022

(66) Obika, M., and Noguchi, H. (2012) Diagnosis and evaluation of nonalcoholic fatty liver

disease. Experimental diabetes research . 2012 , 145754

(67) Paschos, P., and Paletas, K. (2009) Non alcoholic fatty liver disease and metabolic syndrome.

Hippokratia . 13 , 9-19

(68) Walter, P., and Ron, D. (2011) The unfolded protein response: from stress pathway to

homeostatic regulation. Science . 334 , 1081-1086

(69) Xu, X., Park, J.G., So, J.S., Hur, K.Y., and Lee, A.H. (2014) Transcriptional regulation of

apolipoprotein A-IV by the transcription factor CREBH. Journal of lipid research . 55 , 850-859

(70) Johansen, C.T., Wang, J., McIntyre, A.D., Martins, R.A., Ban, M.R., Lanktree, M.B., Huff, M.W.,

Peterfy, M., Mehrabian, M., Lusis, A.J., Kathiresan, S., Anand, S.S., Yusuf, S., Lee, A.H.,

Glimcher, L.H., Cao, H., and Hegele, R.A. (2012) Excess of rare variants in non-genome-wide

association study candidate genes in patients with hypertriglyceridemia. Circulation.

Cardiovascular genetics . 5, 66-72

(71) Xu, X., Park, J.G., So, J.S., and Lee, A.H. (2014) Transcriptional activation of Fsp27 by the liver-

enriched transcription factor CREBH promotes lipid droplet growth and hepatic steatosis.

Hepatology .

(72) Harms, M., and Seale, P. (2013) Brown and beige fat: development, function and therapeutic

potential. Nature medicine . 19 , 1252-1263

(73) Kharitonenkov, A., and Adams, A.C. (2014) Inventing new medicines: The FGF21 story.

Molecular metabolism . 3, 221-229

(74) Mu, J., Pinkstaff, J., Li, Z., Skidmore, L., Li, N., Myler, H., Dallas-Yang, Q., Putnam, A.M., Yao, J.,

Bussell, S., Wu, M., Norman, T.C., Rodriguez, C.G., Kimmel, B., Metzger, J.M., Manibusan, A.,

Lee, D., Zaller, D.M., Zhang, B.B., DiMarchi, R.D., Berger, J.P., and Axelrod, D.W. (2012) FGF21

analogs of sustained action enabled by orthogonal biosynthesis demonstrate enhanced

antidiabetic pharmacology in rodents. Diabetes . 61 , 505-512

(75) Badman, M.K., Pissios, P., Kennedy, A.R., Koukos, G., Flier, J.S., and Maratos-Flier, E. (2007)

Hepatic fibroblast growth factor 21 is regulated by PPARalpha and is a key mediator of

hepatic lipid metabolism in ketotic states. Cell metabolism . 5, 426-437

(76) Inagaki, T., Dutchak, P., Zhao, G., Ding, X., Gautron, L., Parameswara, V., Li, Y., Goetz, R.,

Mohammadi, M., Esser, V., Elmquist, J.K., Gerard, R.D., Burgess, S.C., Hammer, R.E.,

Mangelsdorf, D.J., and Kliewer, S.A. (2007) Endocrine regulation of the fasting response by

PPARalpha-mediated induction of fibroblast growth factor 21. Cell metabolism . 5, 415-425

(77) Yie, J., Wang, W., Deng, L., Tam, L.T., Stevens, J., Chen, M.M., Li, Y., Xu, J., Lindberg, R., Hecht,

R., Veniant, M., Chen, C., and Wang, M. (2012) Understanding the physical interactions in the

FGF21/FGFR/beta-Klotho complex: structural requirements and implications in FGF21

signaling. Chemical biology & drug design . 79 , 398-410

(78) Huang, J., Ishino, T., Chen, G., Rolzin, P., Osothprarop, T.F., Retting, K., Li, L., Jin, P., Matin,

M.J., Huyghe, B., Talukdar, S., Bradshaw, C.W., Palanki, M., Violand, B.N., Woodnutt, G.,

Lappe, R.W., Ogilvie, K., and Levin, N. (2013) Development of a novel long-acting antidiabetic

FGF21 mimetic by targeted conjugation to a scaffold antibody. The Journal of pharmacology

and experimental therapeutics . 346 , 270-280

(79) Kharitonenkov, A., Shiyanova, T.L., Koester, A., Ford, A.M., Micanovic, R., Galbreath, E.J.,

Sandusky, G.E., Hammond, L.J., Moyers, J.S., Owens, R.A., Gromada, J., Brozinick, J.T.,

Hawkins, E.D., Wroblewski, V.J., Li, D.S., Mehrbod, F., Jaskunas, S.R., and Shanafelt, A.B.

(2005) FGF-21 as a novel metabolic regulator. The Journal of clinical investigation . 115 , 1627-

1635

(80) Galman, C., Lundasen, T., Kharitonenkov, A., Bina, H.A., Eriksson, M., Hafstrom, I., Dahlin, M.,

Amark, P., Angelin, B., and Rudling, M. (2008) The circulating metabolic regulator FGF21 is

induced by prolonged fasting and PPARalpha activation in man. Cell metabolism . 8, 169-174

(81) Coskun, T., Bina, H.A., Schneider, M.A., Dunbar, J.D., Hu, C.C., Chen, Y., Moller, D.E., and

Kharitonenkov, A. (2008) Fibroblast growth factor 21 corrects obesity in mice. Endocrinology .

149 , 6018-6027

(82) Xu, J., Lloyd, D.J., Hale, C., Stanislaus, S., Chen, M., Sivits, G., Vonderfecht, S., Hecht, R., Li,

Y.S., Lindberg, R.A., Chen, J.L., Jung, D.Y., Zhang, Z., Ko, H.J., Kim, J.K., and Veniant, M.M.

(2009) Fibroblast growth factor 21 reverses hepatic steatosis, increases energy expenditure,

and improves insulin sensitivity in diet-induced obese mice. Diabetes . 58 , 250-259

(83) Foltz, I.N., Hu, S., King, C., Wu, X., Yang, C., Wang, W., Weiszmann, J., Stevens, J., Chen, J.S.,

Nuanmanee, N., Gupte, J., Komorowski, R., Sekirov, L., Hager, T., Arora, T., Ge, H., Baribault,

H., Wang, F., Sheng, J., Karow, M., Wang, M., Luo, Y., McKeehan, W., Wang, Z., Veniant,

M.M., and Li, Y. (2012) Treating diabetes and obesity with an FGF21-mimetic antibody

activating the betaKlotho/FGFR1c receptor complex. Science translational medicine . 4,

162ra153

(84) Fisher, F.M., Chui, P.C., Antonellis, P.J., Bina, H.A., Kharitonenkov, A., Flier, J.S., and Maratos-

Flier, E. (2010) Obesity is a fibroblast growth factor 21 (FGF21)-resistant state. Diabetes . 59 ,

2781-2789

(85) Moyers, J.S., Shiyanova, T.L., Mehrbod, F., Dunbar, J.D., Noblitt, T.W., Otto, K.A., Reifel-Miller,

A., and Kharitonenkov, A. (2007) Molecular determinants of FGF-21 activity-synergy and

cross-talk with PPARgamma signaling. Journal of cellular physiology . 210 , 1-6

(86) Huang, Z., Wang, H., Lu, M., Sun, C., Wu, X., Tan, Y., Ye, C., Zhu, G., Wang, X., Cai, L., and Li, X.

(2011) A better anti-diabetic recombinant human fibroblast growth factor 21 (rhFGF21)

modified with polyethylene glycol. PloS one . 6, e20669

(87) Gaich, G., Chien, J.Y., Fu, H., Glass, L.C., Deeg, M.A., Holland, W.L., Kharitonenkov, A., Bumol,

T., Schilske, H.K., and Moller, D.E. (2013) The effects of LY2405319, an FGF21 analog, in obese

human subjects with type 2 diabetes. Cell metabolism . 18 , 333-340

(88) Kharitonenkov, A., Beals, J.M., Micanovic, R., Strifler, B.A., Rathnachalam, R., Wroblewski,

V.J., Li, S., Koester, A., Ford, A.M., Coskun, T., Dunbar, J.D., Cheng, C.C., Frye, C.C., Bumol, T.F.,

and Moller, D.E. (2013) Rational design of a fibroblast growth factor 21-based clinical

candidate, LY2405319. PloS one . 8, e58575

(89) Danno, H., Ishii, K.A., Nakagawa, Y., Mikami, M., Yamamoto, T., Yabe, S., Furusawa, M.,

Kumadaki, S., Watanabe, K., Shimizu, H., Matsuzaka, T., Kobayashi, K., Takahashi, A., Yatoh,

S., Suzuki, H., Yamada, N., and Shimano, H. (2010) The liver-enriched transcription factor

CREBH is nutritionally regulated and activated by fatty acids and PPARalpha. Biochemical and

biophysical research communications . 391 , 1222-1227

(90) Lundåsen, T. (2007) Studies on the hormonal regulation of bile acid synthesis. in Department

of Medicine at Huddinge University Hospital ed.)^eds.), Karolinska Universitetssjukhuset,

Huddinge, Sweden

(91) Gauthier, M.S., Favier, R., and Lavoie, J.M. (2006) Time course of the development of non-

alcoholic hepatic steatosis in response to high-fat diet-induced obesity in rats. The British

journal of nutrition . 95 , 273-281

(92) Kohli, R., Kirby, M., Xanthakos, S.A., Softic, S., Feldstein, A.E., Saxena, V., Tang, P.H., Miles, L.,

Miles, M.V., Balistreri, W.F., Woods, S.C., and Seeley, R.J. (2010) High-fructose, medium chain

trans fat diet induces liver fibrosis and elevates plasma coenzyme Q9 in a novel murine

model of obesity and nonalcoholic steatohepatitis. Hepatology . 52 , 934-944

(93) Huang, W., Metlakunta, A., Dedousis, N., Zhang, P., Sipula, I., Dube, J.J., Scott, D.K., and

O'Doherty, R.M. (2010) Depletion of liver Kupffer cells prevents the development of diet-

induced hepatic steatosis and insulin resistance. Diabetes . 59 , 347-357

(94) Hegardt, F.G. (1999) Mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase: a control

enzyme in ketogenesis. The Biochemical journal . 338 ( Pt 3) , 569-582

(95) Drent, M., Cobben, N.A., Henderson, R.F., Wouters, E.F., and van Dieijen-Visser, M. (1996)

Usefulness of lactate dehydrogenase and its isoenzymes as indicators of lung damage or

inflammation. The European respiratory journal . 9, 1736-1742

(96) Markert, C.L., Shaklee, J.B., and Whitt, G.S. (1975) Evolution of a gene. Multiple genes for LDH

isozymes provide a model of the evolution of gene structure, function and regulation.

Science . 189 , 102-114

(97) Roef, M.J., de Meer, K., Kalhan, S.C., Straver, H., Berger, R., and Reijngoud, D.J. (2003)

Gluconeogenesis in humans with induced hyperlactatemia during low-intensity exercise.

American journal of physiology. Endocrinology and metabolism . 284 , E1162-1171

(98) Williamson, D.H., Bates, M.W., Page, M.A., and Krebs, H.A. (1971) Activities of enzymes

involved in acetoacetate utilization in adult mammalian tissues. The Biochemical journal . 121 ,

41-47

(99) Li, W.H., Tanimura, M., Luo, C.C., Datta, S., and Chan, L. (1988) The apolipoprotein multigene

family: biosynthesis, structure, structure-function relationships, and evolution. Journal of

lipid research . 29 , 245-271

(100) O'Brien, P.J., Alborn, W.E., Sloan, J.H., Ulmer, M., Boodhoo, A., Knierman, M.D., Schultze,

A.E., and Konrad, R.J. (2005) The novel apolipoprotein A5 is present in human serum, is

associated with VLDL, HDL, and chylomicrons, and circulates at very low concentrations

compared with other apolipoproteins. Clinical chemistry . 51 , 351-359

(101) Jong, M.C., Hofker, M.H., and Havekes, L.M. (1999) Role of ApoCs in lipoprotein metabolism:

functional differences between ApoC1, ApoC2, and ApoC3. Arteriosclerosis, thrombosis, and

vascular biology . 19 , 472-484

(102) McNeely, M.J., Edwards, K.L., Marcovina, S.M., Brunzell, J.D., Motulsky, A.G., and Austin,

M.A. (2001) Lipoprotein and apolipoprotein abnormalities in familial combined

hyperlipidemia: a 20-year prospective study. Atherosclerosis . 159 , 471-481

(103) Fisher, F.M., Chui, P.C., Nasser, I.A., Popov, Y., Cunniff, J.C., Lundasen, T., Kharitonenkov, A.,

Schuppan, D., Flier, J.S., and Maratos-Flier, E. (2014) Fibroblast growth factor 21 limits

lipotoxicity by promoting hepatic fatty acid activation in mice on methionine and choline-

deficient diets. Gastroenterology . 147 , 1073-1083 e1076

(104) Zhang, Y.K., Yeager, R.L., Tanaka, Y., and Klaassen, C.D. (2010) Enhanced expression of Nrf2 in

mice attenuates the fatty liver produced by a methionine- and choline-deficient diet.

Toxicology and applied pharmacology . 245 , 326-334

(105) Kang, J.H., Yun, S.I., Park, M.H., Park, J.H., Jeong, S.Y., and Park, H.O. (2013) Anti-obesity

effect of Lactobacillus gasseri BNR17 in high-sucrose diet-induced obese mice. PloS one . 8,

e54617

(106) Kim, H., Mendez, R., Zheng, Z., Chang, L., Cai, J., Zhang, R., and Zhang, K. (2014) Liver-

enriched transcription factor CREBH interacts with peroxisome proliferator-activated

receptor alpha to regulate metabolic hormone FGF21. Endocrinology . 155 , 769-782

(107) Zheng, Z., Zhang, C., and Zhang, K. (2011) Measurement of ER stress response and

inflammation in the mouse model of nonalcoholic fatty liver disease. Methods in enzymology .

489, 329-348

(108) Montminy, M.R., Sevarino, K.A., Wagner, J.A., Mandel, G., and Goodman, R.H. (1986)

Identification of a cyclic-AMP-responsive element within the rat somatostatin gene.

Proceedings of the National Academy of Sciences of the United States of America . 83 , 6682-

6686

(109) Juge-Aubry, C., Pernin, A., Favez, T., Burger, A.G., Wahli, W., Meier, C.A., and Desvergne, B.

(1997) DNA binding properties of peroxisome proliferator-activated receptor subtypes on

various natural peroxisome proliferator response elements. Importance of the 5'-flanking

region. The Journal of biological chemistry . 272 , 25252-25259

(110) Veniant, M.M., Hale, C., Helmering, J., Chen, M.M., Stanislaus, S., Busby, J., Vonderfecht, S.,

Xu, J., and Lloyd, D.J. (2012) FGF21 promotes metabolic homeostasis via white adipose and

leptin in mice. PloS one . 7, e40164

(111) Hotta, Y., Nakamura, H., Konishi, M., Murata, Y., Takagi, H., Matsumura, S., Inoue, K., Fushiki,

T., and Itoh, N. (2009) Fibroblast growth factor 21 regulates lipolysis in white adipose tissue

but is not required for ketogenesis and triglyceride clearance in liver. Endocrinology . 150 ,

4625-4633

(112) Markan, K.R., Naber, M.C., Ameka, M.K., Anderegg, M.D., Mangelsdorf, D.J., Kliewer, S.A.,

Mohammadi, M., and Potthoff, M.J. (2014) Circulating FGF21 is liver derived and enhances

glucose uptake during refeeding and overfeeding. Diabetes . 63 , 4057-4063

(113) Finn, P.F., and Dice, J.F. (2006) Proteolytic and lipolytic responses to starvation. Nutrition . 22 ,

830-844

(114) Estrany, M.E., Proenza, A.M., Llado, I., and Gianotti, M. (2011) Isocaloric intake of a high-fat

diet modifies adiposity and lipid handling in a sex dependent manner in rats. Lipids in health

and disease . 10 , 52

(115) Sunny, N.E., Satapati, S., Fu, X., He, T., Mehdibeigi, R., Spring-Robinson, C., Duarte, J.,

Potthoff, M.J., Browning, J.D., and Burgess, S.C. (2010) Progressive adaptation of hepatic

ketogenesis in mice fed a high-fat diet. American journal of physiology. Endocrinology and

metabolism . 298 , E1226-1235

(116) Sengupta, S., Peterson, T.R., Laplante, M., Oh, S., and Sabatini, D.M. (2010) mTORC1 controls

fasting-induced ketogenesis and its modulation by ageing. Nature . 468 , 1100-1104

(117) Lin, W.J., and Anderson, J.W. (1977) Effects of high sucrose or starch-bran diets on glucose

and lipid metabolism of normal and diabetic rats. The Journal of nutrition . 107 , 584-595

ABSTRACT

THE ROLE OF CREBH IN HEPATIC ENERGY REGULATION UNDER METABOLIC STRESS by

ROBERTO MENDEZ

May 2015

Advisor: Dr. Kezhong Zhang

Major: Molecular Biology and Genetics

Degree: Doctor of Philosophy

Lipid metabolism is tightly regulated by nuclear receptors, transcription factors, and cellular enzymes in response to nutritional, hormonal, and stress signals .

Hepatocyte specific, cyclic AMP responsive element-binding protein (CREBH) is a transcription factor that is preferentially expressed in the liver and localized in the endoplasmic reticulum (ER) membrane. CREBH is known to be activated by ER stress, inflammatory stimuli, and metabolic signals to regulate hepatic acute-phase response, lipid metabolism, and glucose metabolism. In my thesis research, I have characterized the roles and mechanisms of CREBH in these functions, as well as the overall phenotype of CrebH -null mice. I demonstrated that CREBH interacts with peroxisome proliferator-activated receptor α (PPARα) to regulate expression of the genes involved in lipolysis, FA oxidation, and ketogenesis by activating fibroblast growth factor 21 (FGF21), a hepatic hormone that regulates whole-body energy homeostasis. Gain- and loss-of-function analyses indicated that CREBH and PPARα

97 regulate each other for their expression. Activated CREBH protein interacts with

PPARα under fasting or metabolic diet, and both factors are required for induction of the FGF21 gene. CREBH and PPARα bind to integrated CREH-PPARα binding motifs in the FGF21 gene promoter across mammalian species. CREBH and PPARα function in synergy to activate the FGF21 gene promoter. Administration of recombinant human FGF21 protein can rescue hypertriglyceridemia and hypoketonemia and, partially, nonalcoholic steatohepatitis (NASH) in the CrebH-null mice developed under an atherogenic high-fat diet, corroborating the role of CREBH in regulating FGF21. In summary, my research demonstrated the transcriptional mechanism for CREBH in regulating lipid metabolism by activating FGF21. In particular, the functional interaction between CREBH and PPARα in regulating

FGF21 may represent an important transcriptional coactivation mechanism in maintaining lipid homeostasis under metabolic stress.

AUTOBIOGRAPHICAL STATEMENT Roberto Mendez Education BS Ball State University Muncie, Indiana Major: Biology Option: Genetics Minor: Chemistry Graduation: Spring 2010

Ph.D. Wayne State University Detroit, Michigan Major: Molecular Biology and Genetics Graduation: Winter 2015

Awards College Board National Hispanic Scholar 2006 Wayne State University Graduate School Dean’s Diversity Fellowship: 2010 R25GM058905 Initiative for Maximizing Student Development (IMSD) Research Education grant to WSU: 2011 Publications Lee J-S, Zheng Z., Mendez R. , Ha S-W., Xie Y., Zhang K. 2012. Pharmacologic ER stress induces non-alcoholic steatohepatitis in an animal model. Toxicology Letters. 211 (1), 29-38. Lee J-S., Mendez R ,. Heng H.H., Yang Z-Q., Zhang K. 2012. Pharmacological ER stress promotes hepatic lipogenesis and lipid droplet formation. Am. J. Transl. Res. 4 (1), 102-113 Mendez R. , Zheng Z., Fan Z., Rajagopalan S., Sun Q., Zhang K. 2013. Exposure to fine airborne particulate matter induces macrophage infiltration, unfolded protein response, and lipid deposition in white adipose tissue. Am. J. Transl. Res. 5 (2), 224-234 Kim, H., Mendez, R. , Zheng, Z., Chang, L., Cai, J., Zhang, R., and Zhang, K. 2014. Liver-enriched Transcription Factor CREBH Interacts with Peroxisome Proliferator-activated Receptor α to Regulate Metabolic Hormone FGF21. Endocrinology 155: 769-782 .