UC San Diego UC San Diego Electronic Theses and Dissertations

Title Exploring the role of Urocortin 2 and Urocortin 3 in energy metabolism

Permalink https://escholarship.org/uc/item/4cj0p89r

Author Falkenhagen, Katherine M.

Publication Date 2012

Peer reviewed|Thesis/dissertation

eScholarship.org Powered by the California Digital Library University of California

UNIVERSITY OF CALIFORNIA, SAN DIEGO

Exploring the Role of Urocortin 2 and Urocortin 3 in Energy Metabolism

A dissertation submitted in partial satisfaction of the requirements for the degree of Doctor of Philosophy

Biomedical Sciences

Katherine M. Falkenhagen

Committee in charge:

Professor Richard Lieber, Chair Professor Ronald Evans Professor Mark Geyer Professor Jerrold Olefsky Professor Simon Schenk Professor Nicolas Webster

2012

Katherine M. Falkenhagen, 2012

The dissertation of Katherine M. Falkenhagen is approved, and it is acceptable in quality and form for publication on microfilm and electronically:

Chair

University of California, San Diego

2012

iii

Dedication

This dissertation is dedicated to the memory of Dr. Wylie Walker Vale.

Table of Contents

Signature Page…………….………………………………………………………………....iii

Dedication.……………………………………………………………………………………iv

Table of Contents..……………………………………………………………………………v

List of Abbreviations………………………………………………………………………..viii

List of Figures…………………………………………………………………………………x

List of Tables…………………………………………………………………………………xii

Acknowledgements…………………………………………………………………...…….xiii

Vita……………………………………………………………………………………………xv

Abstract of the Dissertation………………………………………………………………..xvi

Chapter 1: General Introduction…………………………………………………………….1

Dissertation Overview……………………………………………………………….2

Discovery of the CRF family of peptides and receptors…………………………3

Parameters of energy homeostasis………………………………………………..7

Skeletal muscle and energy homeostasis……………………………………….10

Objective…………………………………………………………………………….13

Significance…………………………………………………………………………15

References………………………………………………………………………….16

Chapter 2: Metabolic characterization of Ucn 2/Ucn 3 double knockout mice…...…..21

Abstract……………………………………………………………………………...22

Introduction………………………………………………………………………….23

Methods……………………………………………………………………………..25

Results………………………………………………………………………………28

Discussion…………………………………………………………………………..31

Figures………………………………………………………………………………35

References………………………………………………………………………….39

Chapter 3: Effect of Ucn 2 deficiency on skeletal muscle fatigability and mitochondrial biogenesis in mice…………………………………………………………………………..42

Abstract……………………………………………………………………………...43

Introduction………………………………………………………………………….44

Methods……………………………………………………………………………..46

Results………………………………………………………………………………52

Discussion…………………………………………………………………………..55

Figures………………………………………………………………………………57

Supplemental Figure…………………………………………………………….....63

References………………………………………………………………………….64

Chapter 4: Ucn 2 mRNA expression in mouse skeletal muscle.………………………66

Abstract……………………………………………………………………………...67

Introduction………………………………………………………………………….68

Methods……………………………………………………………………………..70

Results………………………………………………………………………………75

Discussion…………………………………………………………………………..77

Figures………………………………………………………………………………80

Supplemental Figures……………………………………………………………...84 .

References………………………………………………………………………….86

Chapter 5: General Discussion...………………………………………………………….88

References………………………………………………………………………….95

vii

List of Abbreviations

ACTH adrenocorticotropic hormone AICAR 5-Aminoimidazole-4-carboxyamide ribonucleoside Akt protein kinase B AMPK AMP-activated protein kinase ATP adenosine-5'-triphosphate cAMP cyclic adenosine monophsophate cDNA complementary deoxyribonucleic acid CPT1 carnitine palmitoyltransferase I CRF corticotropin releasing factor CRF-BP corticotropin releasing factor binding protein CRFR corticotropin releasing factor receptor CVS chronic variable stress dKO double knockout EDL extensor digitorum longus Epac cAMP-regulated guanine nucleotide exchange factor Erk1/2 extracellular-signal-regulated kinase 1/2 GLUT-4 glucose transporter 4 GPCR G-protein coupled receptor GTT glucose tolerance test HFD high fat diet HPA hypothalamic-pituitary-adrenal i.p. intraperitoneal ITT insulin tolerance test LFD low fat diet MAPK mitogen-activated kinase MHC myosin heavy chain mRNA messenger ribonucleic acid mtTFA mitochondrial transcriptional factor A NRF-1 nuclear respiratory factor 1 NRF-2 nuclear respiratory factor 2

viii

OXPHOS oxidative phosphorylation system PDK4 pyruvate dehydrogenase lipoamide kinase isozyme 4 PGC-1α peroxisome proliferator-activated receptor gamma coactivator 1α SDH succinate dehydrogenase TA tibialis anterior Ucn urocortin WT wild type

List of Figures

Figure 2-1. Weight gain and food intake of Ucn 2/Ucn 3 dKO mice and WT controls on LFD or HFD………………………………………………………………………………35

Figure 2-2. Fasting glucose and insulin levels of Ucn 2/Ucn 3 dKO mice compared to WT controls………………………………………………………………………………36

Figure 2-3. Glucose homeostasis phenotypes of Ucn 2/Ucn3 dKO mice……………37

Figure 2-4. Lipid profiles of Ucn 2/Ucn 3 dKO mice after overnight fast…………..….38

Figure 3-1. Fatigue test of isolated EDL muscle from Ucn 2 KO mice and WT controls……………………………………………………………………………………….57

Figure 3-2. Nuclear gene expression of mitochondrial-related genes in Ucn 2 KO skeletal muscle relative to expression in WT controls…………………………………..58

Figure 3-3. Fiber type composition of Ucn 2 KO skeletal muscle and WT controls….59

Figure 3-4. Skeletal muscle mitochondrial density in Ucn 2 KO mice and WT controls……………………………………………………………………………………….60

Figure 3-5. Mitochondrial activity of Ucn 2 KO skeletal muscle and WT controls……61

Figure 3-6. Gene expression of PGC-1α and PDK4 in skeletal muscle of exercised mice relative to control……………………………………………………………………...62

Supplemental Figure 3-1. Body composition after 21-day voluntary exercise……….63

Figure 4-1. Effect of nutritional status on Ucn 2 mRNA expression in mouse skeletal muscle………………………………………………………………………………………..80

Figure 4-2. Effect of hypoxia on Ucn 2 mRNA expression in mouse skeletal muscle………………………………………………………………………………………..81

Figure 4-3. Effect of exercise on Ucn 2 mRNA expression in mouse skeletal muscle………………………………………………………………………………………..82

Figure 4-4. Effect of AICAR on Ucn 2 mRNA expression in C2C12 myotubes and mouse skeletal muscle……………………………………………………………………..84

Supplemental Figure 4-1. Body weight and body composition of mice after HFD...... 85

Supplemental Figure 4-2. Blood glucose levels of fasted mice……..……………..…..84

List of Tables

Table 5-1. Summary of metabolic phenotypes of genetically manipulated CRF family animal models……………………………………………………………………………….93

Table 5-2. Summary of metabolic phenotypes of genetically manipulated CRF family animal models after HFD…………………………………………………………………..94

xii

Acknowledgements

I am extremely grateful for all the people that have contributed to this dissertation in one way or another. I cannot possibly thank everyone, but I would like to acknowledge the people that have been most influential during my time as a graduate student.

Firstly, I thank all past and current members of the Vale Lab. I especially thank

Alissa Blackler, Anna Pilbrow, Cindy Donaldson, Elizabeth Flandreau, Eran Gershon,

Ezra Wiater, Hannah Park, Joan Vaughan, Jonathon Kelber, Kathy Lewis, Louise

Bilezikjan, Lykke “Li” Blaabjerg, Marilyn Perrin, Mark Huising Nick Justice, Peter

Grey, Remy Manuel, Sandra Guerra, and Talitha van der Meulen. Each of you have helped me in some way, whether that be teaching me a technique, assisting me in an experiment, sharing reagents, providing feedback, or simply making me laugh.

Thank you PBL-V.

I have been very fortunate for all the collaborators who made the science in this dissertation possible. In the Montminy lab, I thank Naomi Goebel (metabolic cages), Jose Paz (Seahorse assays), Judith Altarejos (fat pads, lipid levels), Nina

Miller (muscle homogenization), and Sam Van de Velde (islet isolation). I thank

Shannon Bremner (SDH assay, fiber typing) and Caryn Urbanczyk (fatigue tests) of the Lieber lab. In the Evans Lab, I thank Weiwei Fan (mitochondrial copy number),

Vihang Narkar (mitochondrial primers), and Michael Downes (AICAR). I thank the

Powell lab for assistance with the hypoxic muscle studies. Finally, big thanks to the

Schenk lab for numerous techniques including muscle extraction and

xiii

homogenization, glucose uptake assays, use of treadmills, citrate synthase assays and everything else.

Of course I thank my committee members, Dr. Rick Lieber, Dr. Ron Evans, Dr.

Mark Geyer, Dr. Jerrold Olefsky, Dr. Simon Schenk and Dr. Nick Webster for their guidance and wisdom over the past 4 years. It has been a pleasure working with you all. A special thanks goes out to Dr. Lieber for “adopting” me during that difficult time that was January 2012. Also, I am deeply grateful to Dr. Simon Schenk for his huge contributions to my work. Thank you for your endless support, for welcoming me into your lab, for teaching me so many techniques and for being so generous with your time. I cannot express my gratitude enough.

Also, I send a special thanks halfway across the world to Dr. Alon Chen.

Thank you so much for everything you have done to make this dissertation possible.

While at the Salk Institute I received support in part from the National Institute of Diabetes and Digestive and Kidney Diseases Program Project Grant DK026741-

30-32, the National Skeletal Muscle Rehabilitation Research Center R24 HD050837-

04 (UCSD), the Elizabeth Keadle Fund, the Helen McLoraine Chair and the H.A. and

Mary K. Chapman Charitable Trust.

Lastly, and most importantly, I thank Dr. Wylie Vale for affording me with a unique graduate school experience that I will cherish forever. Thank you for your warmth, wisdom, wit, trust, and encouragement. You are a true hero and I will always, always strive to make you proud.

xiv

Vita

2002 Student Researcher, Plant Biology Laboratory, Willamette University

2003 Bachelor of Arts in Biology, Chemistry Minor, Willamette University, Departmental Honors

2003-2006 Research Assistant II, Casey Eye Institute, Oregon Health and Science University

2012 Doctor of Philosophy in Biomedical Sciences, University of California, San Diego

ABSTRACT OF THE DISSERTATION

Exploring the Role of Urocortin 2 and Urocortin 3 in Energy Metabolism

Katherine M. Falkenhagen

Doctor of Philosophy in Biomedical Sciences

University of California, San Diego, 2012

Professor Richard Lieber, Chair

Members of the corticotropin-releasing factor (CRF) family of peptides are powerful modulators of numerous metabolic pathways including mechanisms of insulin secretion and glucose homeostasis. Drugs targeted to modulate the CRF system may be relevant treatments for human metabolic disorders. However, a more integrated understanding of the CRF system is needed before such drugs can be developed. The general aims of this dissertation were to investigate the influence of

xvi

the CRF family of peptides on energy metabolism with a focus on the most newly discovered members of this family, Urocortin 2 (Ucn 2) and Urocortin 3 (Ucn 3). Ucn 2 inhibits insulin signaling in skeletal muscle and Ucn 3 enhances glucose-induced insulin secretion from the pancreas. We generated a mouse deficient in both peptides to investigate how these urocortins physiologically contribute to energy homeostasis parameters such as body weight and glucose homeostasis. Additionally, we further characterized the role of endogenous Ucn 2 in skeletal muscle by analyzing metabolic properties, such as fiber type composition, mitochondrial density and fatigue of skeletal muscle from Ucn 2 KO mice. Lastly, we sought to determine if Ucn

2 is regulated in skeletal muscle by metabolic status. Our three main findings were that (1) the loss of both Ucn 2 and Ucn 3 in the dKO mouse model contributes abnormal metabolic phenotypes such as increased body weight and higher concentrations of blood lipids, (2) endogenous Ucn 2 is not involved in mitochondrial pathways, but may be involved in muscle fatigue, and (3) Ucn 2 mRNA expression levels in skeletal muscle may be influenced by energy availability status with overnight fasting decreasing Ucn 2 mRNA expression and high fat feeding trending towards an increase of Ucn 2 mRNA expression. Overall, our studies suggest a complex role of Ucn 2 and Ucn 3 in energy metabolism and specifically show that endogenous Ucn 2 may contribute to muscle fatigue with mechanisms that do not involve mitochondrial density or activity. These findings support the continued investigation of CRF system as potential targets for the treatment of metabolic disorders.

xvii

Chapter 1:

General Introduction

1 2

Dissertation Overview

Maintaining energy homeostasis involves carefully orchestrated metabolic pathways of both central and peripheral mechanisms that constantly adapt to changes in the environment. Impaired adaptive mechanisms can have severe consequences potentially leading to metabolic disorders such as obesity and diabetes. As the prevalence of metabolic disorders continues to rise worldwide, a thorough investigation of mechanisms leading to such metabolic disturbances is necessary in order to develop novel therapeutics.

Members of the corticotropin-releasing factor (CRF) family of peptides and receptors are powerful modulators of metabolic pathways [1-5]. In mammals, the CRF family is comprised of four structurally related ligands CRF, Urocortin 1 (Ucn 1),

Urocortin 2 (Ucn 2), and Urocortin (Ucn 3). These ligands signal through two receptors: corticotropin-releasing factor receptor type 1 (CRFR1) and corticotropin- releasing factor receptor type 2 (CRFR2). CRF family members are expressed in many major metabolic tissues including skeletal muscle and pancreas [6-8]. Genetic manipulations of the CRF family in mouse models demonstrate distinct metabolic phenotypes, which suggest crucial endogenous roles for the peptides and receptors in metabolic pathways [9-13].

Modulating the CRF system has been suggested as treatment for human metabolic disorders [3, 5], but a more integrative understanding of the CRF system in metabolism is needed. The goal of this dissertation is to further dissect the roles of the CRF family of peptides with focus on Ucn 2 and Ucn 3, generally on overall whole body energy homeostasis and specifically in skeletal muscle metabolism.

Discovery of CRF family of peptides and receptors

In 1981, Vale and colleagues isolated CRF for the first time from sheep hypothalami [14]. CRF was identified as a 41-amino acid hypothalamic peptide and a major regulator of the median eminence of the hypothalamic-pituitary-adrenal (HPA) axis. CRF is released from the hypothalamus in response to stress to initiate the release of adrenocorticopic hormone (ACTH) from the anterior pituitary, which then initiates the release of glucocorticoids from the adrenal cortex. In addition to the hypothalamus, CRF is also present in other regions of the brain and in numerous peripheral tissues where it can exert various other endocrine, autonomic and behavioral effects.

The first receptor for CRF, CRFR1, was isolated from a human Cushing’s corticotrophic cell tumor in 1993 [15]. CRFR1 has since been cloned out of cDNA libraries from rat brain [16], rat cerebellum [17] and mouse corticotropes [18]. CRFR1 is a seven transmembrane G-protein-coupled receptor (GPCR) that belongs to the class B GPCR family, also known as the secretin receptor family. In many cells, the receptor is coupled to Gs to activate adenylyl cyclase, which stimulates the release of intracellular cAMP. It has also been shown to activate other signaling pathways such as phospholipase C with production of insositol -1.4.5-triphosphate which activates protein kinase C-dependent and calcium-dependent pathways, mitogen-activated protein kinase-dependent pathways, nitric oxide production and interactions with calcium channels [19-21]. Though multiple splice variants have been identified, including eight in human, three in rats, four in mice and nine in hamster, only one splice variant (CRFR1α, but known as CRFR1) has been shown to be biologically active.

A gene encoding a second receptor, CRFR2, was identified in 1995 from a mouse heart cDNA library using a probe for CRFR1 [22]. CRFR2 is also a seven transmembrane GPCR and has 70% amino acid homology to the CRFR1 receptor.

Like CRFR1, CRFR2 is often coupled to Gs leading to an increase intracellular cAMP upon stimulation, but can also activate other pathways such the induction of MAP kinase pathways [23]. There have been four functional splice variants identified in humans, including membrane-bound and soluble isoforms of CRFR2α, membrane- bound CRFR2β and membrane-bound CRFR2γ. CRFR2β and both membrane-bound and soluble isoforms of CRFR2α have also been identified in rodents.

Following the discovery of urotensin-1-like immunoreactivity in rat brain, Ucn

1 was also identified in 1995 by screening a rat midbrain cDNA library with a probe for urotensin, a CRF homologue in suckerfish [24]. Ucn 1 was named for its similar structure and bioactivity to urotensin and CRF; it is a 40-amino acid peptide with 63% homology to urotensin and 45% homology to CRF at the amino acid level. The Ucn 1 gene encodes a 122-amino acid preprotein, with Ucn 1 in the C-terminus.

The most newly discovered family members, Ucn 2 and Ucn 3, were identified in 2001 [16, 7]. Ucn 2 was identified by a sequence homology search of the human genome [25]. It was predicted as a 38-amino acid peptide with 34% homology to

CRF and 42% homology to Ucn 1 at the amino acid level. The Ucn 2 gene codes for a 112-amino acid residue preprotein and the putative 38-amino acid Ucn 2 peptide is contained in the C-terminus. Ucn 2 preproteins have been identified in human, rat, chicken, fish, dog, and rhesus monkey.

Ucn 3 was identified by a sequence homology search of the human genome with a probe containing a sequence related to urocortin in pufferfish [7]. The putative

38-amino acid Ucn 3 peptide is encoded in a 161-amino acid residue preprotein. Ucn

3 preproteins have been identified in human, rat, chicken, fish, dog, frog, and rhesus monkey.

Also included in the CRF family of peptides is a binding protein (CRF-BP), which was originally isolated from human liver and rat brain in 1991 [26]. The CRF-

BP is a 37-kDa secreted glycoprotein that is capable of binding both CRF and Ucn 1 with similar potencies. It is hypothesized that CRF-BP acts to limit CRFR activation by sequestering CRFR agonists.

Ligands of the CRF family vary in binding affinities to the CRF receptors. Only

CRF and Ucn 1 have high binding affinity to CRFR1, with Ucn 1 being about 3 times more potent than CRF. Ucn 2 has very low affinity and Ucn 3 has almost no binding affinity to CRFR1. For membrane-bound CRFR2, the Ucns bind with high affinity, and CRF has very low binding affinity. Ucn 1 and Ucn 2 are 1-2 orders more potent than CRF and Ucn 3 is slightly less potent than Ucn 1 and Ucn 2 at binding of

CRFR2. Soluble CRFR2, however, shows high affinity for Ucn 1 and CRF and low affinity for Ucn 2 and Ucn 3. The soluble CRFR2 isoform may act similar to the CRF-

BP by competitively sequestering ligand from the CRFR1 receptor.

This dissertation focuses on the metabolic roles of Ucn 2 and Ucn 3, which are both selective for CRFR2. Expression of Ucn 2 has been reported in rodent brain including the paraventricular, supraoptic, and arcuate nuclei of the hypothalamus and the locus coeruleus in the brainstem [25]. In the periphery, Ucn 2 is expressed at highest levels in skeletal muscle, skin, adrenal gland, lung, and gastrointestinal tract

[6]. Ucn 3 is also expressed throughout the brain including the hypothalamus and

medial amygdala, and in peripheral tissues such as the pancreas, adrenal gland, and gastrointestinal tract [8, 27, 28].

Parameters of energy homeostasis

Metabolic disorders are associated with disruptions in energy homeostasis.

Characteristics of obese and diabetic patients may include high body weight, hyperglycemia, and hyperinsulemia. In this section we provide a brief overview of such parameters in relation to metabolic disease and also how the CRF family may be involved.

Body Weight and composition

Food intake and energy expenditure are the major factors that contribute to body weight and body composition. Ultimately, the brain governs these mechanisms by activating pathways of behavior, the endocrine system, and the autonomic nervous system [29]. Behavioral mechanisms determine food intake, while endocrine and autonomic mechanisms influence energy expenditure. Central infusion of CRF reduces food intake [30-32], increases thermogenesis [30, 33], and decreases weight gain [34-36]. The urocortins also exert anorectic and thermogenic effects that result in decreased body weight. For example, central administration of Ucn 3 reduces food intake [27, 37-39] and overexpression of Ucn 3 in mouse hypothalamus increases energy expenditure [40].

Related to the maintenance of body weight is the reported effect of Ucn 2 on skeletal muscle mass. Several studies have shown that Ucn 2 increases skeletal muscle mass and protects muscle from atrophy [41, 42]. Rats treated with a daily subcutaneous injection of Ucn 2 have increased skeletal muscle mass [42, 43].

These studies also found Ucn 2 treatment to increase muscle absolute force in

skeletal muscle [42, 43]. In rodents, Ucn 2 treatment also protects skeletal muscle from atrophy caused by casting, dexamethasone, or denervation [42, 43].

Glucose homeostasis

Insulin is the primary regulator of blood glucose concentrations. Insulin acts on target tissues to promote anabolic mechanisms such as glucose uptake and storage. In skeletal muscle, insulin binds to its receptor to activate a signaling cascade, which results in the translocation of glucose transporter 4 to the cell membrane to transport glucose into the cell. Many metabolic disorders are characterized by insulin resistance, a condition in which tissues fail to respond adequately to insulin.

Ucn 2, through activation of CRFR2, has been shown to inhibit insulin signaling in skeletal muscle [10]. In C2C12 myotubes, Ucn 2 dose-dependently inhibits insulin-induced phosphorylation of Akt and ERK1/2 and inhibits insulin- induced glucose uptake [10]. In line with observations in vitro, Ucn 2 KO mice are more insulin sensitive as shown by insulin and glucose tolerance tests and hyperinsulinemic euglycemic clamp experiments [10]. Furthermore, WT mice treated with the CRFR2 antagonist, astressin 2B, are more glucose tolerant than untreated animals [10].

Beta cells in the pancreas are responsible for the storage and release of insulin. Ucn 3 and both receptors, CRFR1 and CRFR2, are found in pancreatic beta cells where they influence insulin secretion and beta cell mass [8, 11, 13] and in the presence of high glucose, the pancreatic beta cell line MIN6 have been shown to secrete Ucn 3 [8]. Interestingly, the obese mouse model ob/ob and high fat diet-fed

rats express Ucn 3 mRNA to a greater extent than controls [13]. Activation of either

CRFR1 or CRFR2 increases glucose-stimulated insulin release from beta cells [8,

11]. Additionally, It was shown that activation of CRFR1 increases B cell proliferation through a MAP kinase mechanism [11]. Collectively, these data suggest that Ucn 3 is released by beta cells and acts locally on CRFRs to increase glucose-induced insulin secretion and perhaps contribute to the beta cell growth.

Skeletal muscle and energy homeostasis

Skeletal muscle is a plastic tissue that is constantly adapting to changes in energy requirements or work demand in order to maintain whole body energy homeostasis. Ucn 2 and CRFR2 are both highly expressed in skeletal muscle where they influence a variety of mechanisms including muscle mass [41], respiratory rate

[44], and insulin signaling [10]. In this dissertation, we were interested in exploring potential roles for Ucn 2 in other metabolic mechanisms in skeletal muscle such as fiber type switching, mitochondrial biogenesis and muscle fatigue.

Fiber Type Switching

Skeletal muscle is composed of muscle fiber types with distinct metabolic properties. These fibers can be classified according to the myosin heavy chain

(MHC) gene that is expressed. In rodent skeletal muscle, four different isoforms of

MHC are present: type 1, type IIA, type IIx, and type IIB [45]. Because these fiber types differ in contraction speed, type I fibers are referred to as slow-twitch fibers and type II are fast-twitch. Type I slow twitch fibers are highly oxidative, with high levels of oxidative enzymes that provide a slow, but steady supply of ATP to the muscle, which makes them resistant to fatigue. Type II fast twitch fibers, on the other hand, are highly glycolytic with glycolytic enzymes providing fast supply of ATP and fatigue quickly due to depletion of glycogen stores after brief usage [46]. Different types of muscle will have different fiber type composition depending on function. For example, the soleus muscle is used constantly (standing, walking, running) and is comprised of more type I fibers than type II. Conversely, extensor digitalis longus

(EDL) muscle, responsible for the extension of the toes and ankles, is representative

of a muscle primarily composed of type II fibers. Fiber type transformation occurs as an adaptation to meet changes in energy demands. Repeated electrical stimulation of muscle [47, 48] and endurance exercise [49, 50] have been shown to induce the transformation of type II fibers to type I fibers. It has been suggested that fiber type plasticity is correlated with the pathogenesis of metabolic disorders [51]. For example, diabetic patients tend to have more glycolytic skeletal muscle fibers than healthy individuals [51].

Mitochondrial biogenesis

Skeletal muscle can adapt to energy deprivation by increasing mitochondrial number and activity, a process known as mitochondrial biogenesis. Mitochondrial biogenesis requires the coordinated expression of both mitochondrial and nuclear genes. Mitochondrial DNA encodes 13 subunits of the oxidative phosphorylation system (OXPHOS) and nuclear DNA encodes the remaining OXPHOS subunits along with other mitochondrial proteins. Peroxisome proliferator-activated receptor gamma co-activator 1α (PGC-1α) is a transcriptional co-activator that stimulates mitochondrial biogenesis by activating the expression of nuclear respiratory factors

(NRF-1 and NRF-2) and mitochondrial transcriptional factor A (mtTFA) which induce transcription of mitochondrial and nuclear genes that encode mitochondrial proteins

[31-33, 52]. Mitochondrial biogenesis increases the capacity of muscle to generate

ATP by oxidative phosphorylation [50, 53, 54]. Skeletal muscle mitochondrial biogenesis occurs in response to exercise [50, 55] or glucocorticoid treatment [56] and can occur in vitro in the rat muscle myotube cell line L6 by increasing cystolic calcium to mimic exercise [57]. Mitochondrial dysfunction is correlated with metabolic

disorders and a suggested therapy for metabolic disorders is to stimulate mitochondrial biogenesis [58, 59].

Muscle Fatigue

Muscle fatigue is the decline in muscle force generation (muscle performance) in response to repeated, intense use [46]. Early muscle fatigue may indicate disruptions in skeletal muscle function such as impairment of contractile proteins, failure of sarcoplasmic reticulum Ca2+ release, ionic changes on action potential, or effects of reactive oxygen species. “Fatigability” varies in skeletal muscle and may be correlated with fiber type composition; fast fibers fatigue quickly, while slow fibers are resistant to fatigue. Increased muscle fatigue has been shown in diabetic patients

[60], and this may be due in part to reduced mitochondrial function, vascular factors, or inability to mobilize glycogen stores [60].

Objective

The overall aim of this dissertation is to investigate the influence of the CRF family of peptides on energy homeostasis with focus on the most newly discovered family members, Ucn 2 and Ucn 3. We chose these particular peptides because Ucn

2 has been shown to inhibit insulin signaling in skeletal muscle and Ucn 3 enhances glucose-induced insulin secretion from the pancreas. Thus, Ucn 2 and Ucn 3 are positioned to influence energy metabolism and worthy targets for the treatment of metabolic disease.

Our first objective was to investigate how these two pathways interact. To accomplish this goal, we generated a mouse deficient in both peptides. In these double knock out mice, we tested energy homeostasis parameters such as body weight and glucose homeostasis to understand how Ucn 2 and Ucn 3 physiologically contribute to energy homeostasis

Our second objective was to further characterized the role of endogenous Ucn

2 in skeletal muscle by analyzing metabolic properties, such as fiber type composition, mitochondrial biogenesis and fatigue of skeletal muscle from Ucn 2 deficient mice. We chose to focus on Ucn 2 due to the reported metabolic phenotypes of Ucn 2 KO mice that include increased insulin sensitivity and decreased body fat after HFD compared to WT controls. Additionally, as Ucn 2 is present in skeletal muscle at relatively higher levels than other peripheral tissues it is possible that many of the metabolic phenotypes observed in the Ucn 2 KO mice are due to the loss of Ucn 2 from skeletal muscle. We were curious to know if the loss of Ucn 2 in skeletal muscle affects other metabolic pathways in addition to insulin signaling pathways.

As a third objective, we sought to investigate if Ucn 2 is regulated in skeletal muscle by metabolic status by measuring levels of Ucn 2 mRNA in mouse skeletal muscle after metabolic perturbations such as fasting, high fat diet and exercise. Our general hypothesis was that skeletal muscle Ucn 2 is increased in times of nutrient deprivation where it can play a “glucose-sparing” role by inhibiting glucose uptake into muscle and sparing glucose for tissues that rely exclusively on glucose as a fuel source, such as the brain. These experiments are necessary in order to validate the use of Ucn 2 and Ucn 3 as therapeutic targets for metabolic disorders.

Significance

Due to the increasing trend of metabolic diseases rates worldwide, novel therapies are needed. GPCRs are common drug targets due to their location on cell membranes and frequent relevance to disease. As the CRF family receptors are

GPCRs, it is of particular interest to understand how the actions of these molecules are relevant to metabolic disorders. The purpose of this dissertation is to further dissect the complex and integrative roles of the CRF family, particularly Ucn 2 and

Ucn 3, in energy homeostasis. We seek to provide new insights into potential therapeutic interventions for the treatment of metabolic disorders.

References

1. Cullen, M.J., et al., Urocortin, corticotropin releasing factor-2 receptors and energy balance. Endocrinology, 2001. 142(3): p. 992-9.

2. Hashimoto, K., et al., Physiological roles of corticotropin-releasing hormone receptor type 2. Endocr J, 2001. 48(1): p. 1-9.

3. Kuperman, Y. and A. Chen, Urocortins: emerging metabolic and energy homeostasis perspectives. Trends Endocrinol Metab, 2008. 19(4): p. 122-9.

4. Richard, D., Q. Huang, and E. Timofeeva, The corticotropin-releasing hormone system in the regulation of energy balance in obesity. Int J Obes Relat Metab Disord, 2000. 24 Suppl 2: p. S36-9.

5. Richard, D., Q. Lin, and E. Timofeeva, The corticotropin-releasing factor family of peptides and CRF receptors: their roles in the regulation of energy balance. Eur J Pharmacol, 2002. 440(2-3): p. 189-97.

6. Chen, A., et al., Urocortin II gene is highly expressed in mouse skin and skeletal muscle tissues: localization, basal expression in corticotropin- releasing factor receptor (CRFR) 1- and CRFR2-null mice, and regulation by glucocorticoids. Endocrinology, 2004. 145(5): p. 2445-57.

7. Lewis, K., et al., Identification of urocortin III, an additional member of the corticotropin-releasing factor (CRF) family with high affinity for the CRF2 receptor. Proc Natl Acad Sci U S A, 2001. 98(13): p. 7570-5.

8. Li, C., et al., Urocortin III is expressed in pancreatic beta-cells and stimulates insulin and glucagon secretion. Endocrinology, 2003. 144(7): p. 3216-24.

9. Bale, T.L., et al., Corticotropin-releasing factor receptor-2-deficient mice display abnormal homeostatic responses to challenges of increased dietary fat and cold. Endocrinology, 2003. 144(6): p. 2580-7.

10. Chen, A., et al., Urocortin 2 modulates glucose utilization and insulin sensitivity in skeletal muscle. Proc Natl Acad Sci U S A, 2006. 103(44): p. 16580-5.

11. Huising, M.O., et al., CRFR1 is expressed on pancreatic beta cells, promotes beta cell proliferation, and potentiates insulin secretion in a glucose- dependent manner. Proc Natl Acad Sci U S A, 2010. 107(2): p. 912-7.

12. Jamieson, P.M., et al., Urocortin 3 transgenic mice exhibit a metabolically favourable phenotype resisting obesity and hyperglycaemia on a high-fat diet. Diabetologia, 2011. 54(9): p. 2392-403.

13. Li, C., et al., Urocortin 3 regulates glucose-stimulated insulin secretion and energy homeostasis. Proc Natl Acad Sci U S A, 2007. 104(10): p. 4206-11.

14. Vale, W., et al., Characterization of a 41-residue ovine hypothalamic peptide that stimulates secretion of corticotropin and beta-endorphin. Science, 1981. 213(4514): p. 1394-7.

15. Chen, R., et al., Expression cloning of a human corticotropin-releasing-factor receptor. Proc Natl Acad Sci U S A, 1993. 90(19): p. 8967-71.

16. Perrin, M.H., et al., Cloning and functional expression of a rat brain corticotropin releasing factor (CRF) receptor. Endocrinology, 1993. 133(6): p. 3058-61.

17. Chang, C.P., et al., Identification of a seven transmembrane helix receptor for corticotropin-releasing factor and sauvagine in mammalian brain. Neuron, 1993. 11(6): p. 1187-95.

18. Vita, N., et al., Cloning and expression of a complementary DNA encoding a high affinity human neurotensin receptor. FEBS Lett, 1993. 317(1-2): p. 139- 42.

19. Brar, B.K., et al., Urocortin-II and urocortin-III are cardioprotective against ischemia reperfusion injury: an essential endogenous cardioprotective role for corticotropin releasing factor receptor type 2 in the murine heart. Endocrinology, 2004. 145(1): p. 24-35; discussion 21-3.

20. Hillhouse, E.W. and D.K. Grammatopoulos, The molecular mechanisms underlying the regulation of the biological activity of corticotropin-releasing hormone receptors: implications for physiology and pathophysiology. Endocr Rev, 2006. 27(3): p. 260-86.

21. Wiesner, B., et al., Intracellular calcium measurements of single human skin cells after stimulation with corticotropin-releasing factor and urocortin using confocal laser scanning microscopy. J Cell Sci, 2003. 116(Pt 7): p. 1261-8.

22. Perrin, M., et al., Identification of a second corticotropin-releasing factor receptor gene and characterization of a cDNA expressed in heart. Proc Natl Acad Sci U S A, 1995. 92(7): p. 2969-73.

23. Rossant, C.J., et al., Corticotropin-releasing factor type 1 and type 2alpha receptors regulate phosphorylation of calcium/cyclic adenosine 3',5'- monophosphate response element-binding protein and activation of p42/p44 mitogen-activated protein kinase. Endocrinology, 1999. 140(4): p. 1525-36.

24. Vaughan, J., et al., Urocortin, a mammalian neuropeptide related to fish urotensin I and to corticotropin-releasing factor. Nature, 1995. 378(6554): p. 287-92.

25. Reyes, T.M., et al., Urocortin II: a member of the corticotropin-releasing factor (CRF) neuropeptide family that is selectively bound by type 2 CRF receptors. Proc Natl Acad Sci U S A, 2001. 98(5): p. 2843-8.

26. Potter, E., et al., Cloning and characterization of the cDNAs for human and rat corticotropin releasing factor-binding proteins. Nature, 1991. 349(6308): p. 423-6.

27. Jamieson, P.M., et al., Urocortin 3 modulates the neuroendocrine stress response and is regulated in rat amygdala and hypothalamus by stress and glucocorticoids. Endocrinology, 2006. 147(10): p. 4578-88.

28. Li, C., et al., Urocortin III-immunoreactive projections in rat brain: partial overlap with sites of type 2 corticotrophin-releasing factor receptor expression. J Neurosci, 2002. 22(3): p. 991-1001.

29. Swanson, L.W. and G.J. Mogenson, Neural mechanisms for the functional coupling of autonomic, endocrine and somatomotor responses in adaptive behavior. Brain Res, 1981. 228(1): p. 1-34.

30. Arase, K., et al., Effects of corticotropin-releasing factor on food intake and brown adipose tissue thermogenesis in rats. Am J Physiol, 1988. 255(3 Pt 1): p. E255-9.

31. Krahn, D.D., B.A. Gosnell, and M.J. Majchrzak, The anorectic effects of CRH and restraint stress decrease with repeated exposures. Biol Psychiatry, 1990. 27(10): p. 1094-102.

32. Morley, J.E. and A.S. Levine, Corticotrophin releasing factor, grooming and ingestive behavior. Life Sci, 1982. 31(14): p. 1459-64.

33. LeFeuvre, R.A., N.J. Rothwell, and M.J. Stock, Activation of brown fat thermogenesis in response to central injection of corticotropin releasing hormone in the rat. Neuropharmacology, 1987. 26(8): p. 1217-21.

34. Bornstein, S.R., et al., Chronic effects of a nonpeptide corticotropin-releasing hormone type I receptor antagonist on pituitary-adrenal function, body weight, and metabolic regulation. Endocrinology, 1998. 139(4): p. 1546-55.

35. Brown, M.R., et al., Corticotropin-releasing factor: actions on the sympathetic nervous system and metabolism. Endocrinology, 1982. 111(3): p. 928-31.

36. Rohner-Jeanrenaud, F., et al., Central corticotropin-releasing factor administration prevents the excessive body weight gain of genetically obese (fa/fa) rats. Endocrinology, 1989. 124(2): p. 733-9.

37. Fekete, E.M., et al., Delayed satiety-like actions and altered feeding microstructure by a selective type 2 corticotropin-releasing factor agonist in rats: intra-hypothalamic urocortin 3 administration reduces food intake by

prolonging the post-meal interval. Neuropsychopharmacology, 2007. 32(5): p. 1052-68.

38. Ohata, H. and T. Shibasaki, Effects of urocortin 2 and 3 on motor activity and food intake in rats. Peptides, 2004. 25(10): p. 1703-9.

39. Pelleymounter, M.A., et al., Behavioral and neuroendocrine effects of the selective CRF2 receptor agonists urocortin II and urocortin III. Peptides, 2004. 25(4): p. 659-66.

40. Kuperman, Y., et al., Perifornical Urocortin-3 mediates the link between stress-induced anxiety and energy homeostasis. Proc Natl Acad Sci U S A, 2010. 107(18): p. 8393-8.

41. Hinkle, R.T., et al., Activation of the CRF 2 receptor modulates skeletal muscle mass under physiological and pathological conditions. Am J Physiol Endocrinol Metab, 2003. 285(4): p. E889-98.

42. Hinkle, R.T., et al., Corticotropin releasing factor 2 receptor agonists reduce the denervation-induced loss of rat skeletal muscle mass and force and increase non-atrophying skeletal muscle mass and force. J Muscle Res Cell Motil, 2004. 25(7): p. 539-47.

43. Hinkle, R.T., et al., Urocortin II treatment reduces skeletal muscle mass and function loss during atrophy and increases nonatrophying skeletal muscle mass and function. Endocrinology, 2003. 144(11): p. 4939-46.

44. Solinas, G., et al., Corticotropin-releasing hormone directly stimulates thermogenesis in skeletal muscle possibly through substrate cycling between de novo lipogenesis and lipid oxidation. Endocrinology, 2006. 147(1): p. 31-8.

45. Schiaffino, S., et al., Three myosin heavy chain isoforms in type 2 skeletal muscle fibres. J Muscle Res Cell Motil, 1989. 10(3): p. 197-205.

46. Allen, D.G., G.D. Lamb, and H. Westerblad, Skeletal muscle fatigue: cellular mechanisms. Physiol Rev, 2008. 88(1): p. 287-332.

47. Mayne, C.N., et al., Stimulation-induced expression of slow muscle myosin in a fast muscle of the rat. Evidence of an unrestricted adaptive capacity. FEBS Lett, 1993. 327(3): p. 297-300.

48. Windisch, A., et al., Fast to slow transformation of denervated and electrically stimulated rat muscle. J Physiol, 1998. 510 ( Pt 2): p. 623-32.

49. Gollnick, P.D., et al., Effect of training on enzyme activity and fiber composition of human skeletal muscle. J Appl Physiol, 1973. 34(1): p. 107-11.

50. Holloszy, J.O., Biochemical adaptations in muscle. Effects of exercise on mitochondrial oxygen uptake and respiratory enzyme activity in skeletal muscle. J Biol Chem, 1967. 242(9): p. 2278-82.

51. Oberbach, A., et al., Altered fiber distribution and fiber-specific glycolytic and oxidative enzyme activity in skeletal muscle of patients with type 2 diabetes. Diabetes Care, 2006. 29(4): p. 895-900.

52. Wu, Z., et al., Mechanisms controlling mitochondrial biogenesis and respiration through the thermogenic coactivator PGC-1. Cell, 1999. 98(1): p. 115-24.

53. Constable, S.H., et al., Energy metabolism in contracting rat skeletal muscle: adaptation to exercise training. Am J Physiol, 1987. 253(2 Pt 1): p. C316-22.

54. Dudley, G.A., P.C. Tullson, and R.L. Terjung, Influence of mitochondrial content on the sensitivity of respiratory control. J Biol Chem, 1987. 262(19): p. 9109-14.

55. Davies, K.J., L. Packer, and G.A. Brooks, Biochemical adaptation of mitochondria, muscle, and whole-animal respiration to endurance training. Arch Biochem Biophys, 1981. 209(2): p. 539-54.

56. Weber, K., et al., Glucocorticoid hormone stimulates mitochondrial biogenesis specifically in skeletal muscle. Endocrinology, 2002. 143(1): p. 177-84.

57. Ojuka, E.O., et al., Raising Ca2+ in L6 myotubes mimics effects of exercise on mitochondrial biogenesis in muscle. FASEB J, 2003. 17(6): p. 675-81.

58. Patti, M.E. and S. Corvera, The role of mitochondria in the pathogenesis of type 2 diabetes. Endocr Rev, 2010. 31(3): p. 364-95.

59. Joseph, A.M., et al., Mitochondrial dysregulation in the pathogenesis of diabetes: potential for mitochondrial biogenesis-mediated interventions. Exp Diabetes Res, 2012. 2012: p. 642038.

60. Halvatsiotis, P., et al., Synthesis rate of muscle proteins, muscle functions, and amino acid kinetics in type 2 diabetes. Diabetes, 2002. 51(8): p. 2395- 404.

Chapter 2:

Metabolic characterization of Ucn 2/Ucn 3 double knockout mice

21 22

Abstract

Urocortin 2 (Ucn 2) and Urocortin 3 (Ucn 3), members of the corticotropin- releasing factor (CRF) family of peptides, are emerging as important central and peripheral regulators of energy homeostasis and metabolic functions. Both peptides, which are differentially expressed in metabolism-related peripheral tissues and brain nuclei, function through specific activation of the CRF receptor type 2 (CRFR2). To further characterize the role of these peptides in regulating energy balance under basal and metabolically challenged (high fat diet) conditions we generated a novel mouse deficient for both Ucn 2 and Ucn 3. The Ucn 2/Ucn 3 double knockout mice

(dKO) had increased body weight, elevated basal fasting glucose and insulin concentrations, and higher concentrations of circulating cholesterol and triglycerides.

Data from the Ucn 2/Ucn 3 dKO mice support a critical role for the Ucns/CRFR2 system in the regulation of energy balance through effects on glucose homeostasis and metabolic functions. These complex findings further emphasize the need for specific mouse models that will allow the dissection of central and peripheral contributions of these CRF family members to the regulation of energy homeostasis.

Introduction

Urocortin 2 (Ucn 2) and Urocortin 3 (Ucn 3) are members of the corticotropin- releasing factor (CRF) family of peptides that play numerous roles in regulating central and peripheral responses to homeostatic challenge including the regulation of energy balance and metabolism [1-10]. Both peptides function through specific activation of the G-protein coupled receptor CRF receptor type 2 (CRFR2) [6, 9, 11].

Ucn 2, Ucn 3 and CRFR2 are differentially expressed throughout the central nervous system and the periphery [7, 12-23]. Previous studies have characterized CRF family knockout models to investigate the endogenous roles of the Ucns/CRFR2 system in regulating energy homeostasis. Mice deficient for Ucn 2 or Ucn 3 exhibited unique metabolic characteristics and reveal a potential role for these peptides in mediating energy homeostasis and metabolic functions under basal and metabolically challenged conditions [24-28].

Ucn 2 has been shown to inhibit insulin signaling and glucose uptake in skeletal muscle [26]. Mice null for Ucn 2 displayed a robust increase in insulin sensitivity, attributed to the absence of Ucn 2-induced inhibition of insulin. Clamp studies reveal that Ucn 2 KO animals remain sensitive to insulin after high-fat diet

(HFD), while WT mice became characteristically insulin resistant [26]. Also, on a

HFD Ucn 2 KO mice consumed similar amounts of food and gained similar weights, but had more lean tissue mass compared to WT littermates [26]. Additionally, Ucn 2

KO animals were protected against HFD-induced hyperglycemia and hyperinsulimia

[26].

Ucn 3 is highly expressed by the pancreatic beta cells and has been shown to enhance glucose-induced insulin secretion [28]. When placed on a HFD, Ucn 3 KO

mice were protected against HFD-induced hyperinsulimia, hyperglycemia, glucose intolerance, and insulin resistance [28]. However, on a regular chow diet the Ucn 3

KO animals, unlike the Ucn 2 KO mice, show similar responses to glucose and insulin challenges as WT controls. Thus, the metabolic phenotypes reported in the Ucn 3 KO mice have been suggested to be a result of lower insulin concentration secondary to the absence of Ucn 3-induced increase of insulin secretion after HFD [28].

While both Ucn 2 and Ucn 3 activate CRFR2 with high affinity, they are differentially expressed in metabolically related peripheral tissues and brain nuclei [6,

9, 16, 29], which may suggest a distinct role for these peptides in regulating energy homeostasis and metabolic functions. To explore the combined contribution of Ucn 2 and Ucn 3 in regulating glucose homeostasis and metabolic functions at the whole animal level, we generated a mouse model deficient in both Ucn 2 and Ucn 3. We hypothesized the double deficient mice (Ucn 2/ Ucn 3 dKO) would have the combined advantages of the lack of Ucn 2-induced insulin inhibition in skeletal muscle and the lack of Ucn 3-induced insulin secretion from the pancreas to result in improved insulin sensitivity and protection against HFD-induced abnormalities such as hyperinsulemia and hyperglycemia. To test this hypothesis, the energy homeostasis parameters of

Ucn 2/Ucn 3 dKO mice were assessed under basal and HFD conditions.

Methods

Animals

Ucn 2/Ucn 3 dKO mice were generated by crossing the previously characterized Ucn 2 [26] and Ucn 3 [28] single KO mice. WT mice were obtained from the same colony. Mutant mice are fertile and mutant alleles were passed on in a

Mendelian fashion. Animals were housed in an approved animal facility with 12 hr light (0600 hr to 1800 hr), dark (1800 hr to 0600 hr) cycle with free access to food and water. All animal protocols were approved by the Salk Institute Institutional Animal

Care and Use Committee.

Food intake and high fat diet

At four weeks of age, male mice (N=13) were individually housed and fed either low fat (LFD, 10% kcal fat) or high fat (HFD, 45% kcal fat) diets (Research

Diets, Inc., New Brunswick, NJ, USA). Food intake was monitored by weighing food pellets at the beginning and the end of the light cycle using a digital scale. Animal weights were recorded weekly.

Glucose and Insulin Challenge Tests

For glucose tolerance tests (GTT), animals (N=10) were fasted overnight for

14 hr before receiving an i.p. glucose injection (2 g/kg body weight). Venous tail blood was sampled before glucose injection and at 15, 30, 60 and 90 min after injection using an automatic glucometer (Novamax; Nova Biomedical, Waltham, MA,

USA). For insulin tolerance tests (ITT), animals were fasted for 2 hr and then injected

i.p. with recombinant insulin (0.75 U/kg) and blood glucose was measured as described above at 30, 60 and 90 min after injection.

Epididymal fad pad measurements

Epididymal fat pads were surgically removed and weighed using a digital scale.

Fasting glucose, insulin and lipid measurements

After an overnight fast (14 hr), tail blood was collected to measure plasma glucose concentration using an automated glucometer (Novamax, Nova Biomedical,

Waltham, MA, USA). For insulin concentration, blood was collected via cardiac puncture and centrifuged to separate the plasma. Plasma was stored at -20 C° until assay. Insulin was measured using a commercially available rat/mouse RIA kit

(Millipore, St. Charles, MO, USA) per manufacturer’s instructions. Lipid concentrations were determined using CardioChek Analyzer (Polymer Technology

Systems, Inc. Indianapolis, Indiana, USA).

Statistical Analysis

Statistical analysis was performed using GraphPad Prism 5 Software

(GraphPad Software Inc., La Jolla, CA, USA). Results are expressed as means ±

S.E.M. Comparisons between experimental groups were made using two-way analysis of variance (ANOVA) or an unpaired two-tailed student’s t-tests where appropriate. Post hoc analysis was done using Bonferroni tests. P values < 0.05

were considered to be statistically significant. Statistical measures are specified in figure legends.

Results

Ucn 2/ Ucn 3 dKO mice gain more weight despite similar food consumption compared to WT controls

HFD feeding is a well-established model to study the ontogeny of type 2 diabetes. Ucn 2/Ucn 3 dKO mice and WT controls were placed on a HFD or LFD for

17 weeks. A main effect of diet on weight was apparent at week 2 (F(1,44) = 11.20, p=0.002) and lasted throughout the study (week 17: F(1,43) = 10.71, p<0.0001) (Fig.

2-2A). There was a main effect of genotype on weight differences starting at week 6 of the study with the Ucn 2/Ucn 3 dKO mice weighing more than WT controls

(F(1,44)=4.06, p=0.05) (Fig. 2-1A). On LFD, Ucn 2/Ucn 3 dKO mice weighed more than WT controls at weeks 9-11 and 13-17. On HFD, the Ucn 2/Ucn 3 dKO mice weighed more than WT controls at weeks 7, 12, and 13. At the end of the study Ucn

2/Ucn 3 dKO mice on LFD weighed significantly more than WT controls (WT = 34.4 ±

1.5, dKO = 39.8 ± 1.3, p<0.01) and Ucn 2/Ucn 3 dKO mice trended towards weighing more on a HFD (WT = 41.1 ± 0.7, dKO = 44.0 ± 1.1) (Fig. 2-1B). Interestingly, the WT animals on HFD showed similar a weight gain profile as Ucn 2/Ucn 3 dKO mice placed on LFD. These weight differences were not due to differences in food intake

(Fig. 2-1C).

Ucn 2/Ucn 3 dKO mice exhibit higher fasting glucose and insulin concentrations compared to WT controls

Ucn 2/Ucn 3 dKO mice had higher fasting glucose concentration (F(1,41) =

10.68, p = 0.002) (Fig. 2-2A) and fasting insulin concentration (F(1,37) = 6.41, p=0.16) (Fig. 2-2B). In the HFD groups, Ucn 2/Ucn 3 dKO had significantly higher

concentration of blood glucose (WT = 104 ± 7, dKO = 126 ± 5, p < 0.05) and insulin

(WT = 0.98 ± 0.30, dKO = 1.92 ± 0.29, p<0.05) (Fig. 2-2A and 2-2B, right bars). In the

LFD group, Ucn 2/Ucn 3 dKO mice showed a trend towards higher glucose (WT = 93

± 5, dKO = 112 ± 8) and insulin concentration (WT = 0.67 ± 0.22, dKO = 1.04 ± 0.22), however, these differences did not reach statistical significance (Fig. 2-2A and 2-2B, left bars).

Ucn 2/Ucn 3 dKO mice have improved glucose tolerance but abnormal insulin sensitivity compared to WT controls

Glucose tolerance was significantly improved in the Ucn 2/Ucn 3 dKO mice compared to WT controls on LFD (F(1,22) =9.71, p = 0.005)(Fig. 2-3A) and HFD

F(1,19) = 4.70, p = 0.04)(Fig. 2-3B). In the LFD groups, percent change of blood glucose during the GTT is significantly lower in Ucn 2/Ucn 3 dKO mice compared to

WT at 30 min (WT = 274% ± 18, KO = 159% ± 31, p<0.01), 60 min (WT = 271% ± 19,

KO = 172% ± 33, p<0.01) and 90 min (WT = 221% ± 16, KO = 137% ± 31, p<0.05) time points (Fig. 2-3A). In the HFD, post hoc Bonferroni analysis shows significance only at the 15 min time point (WT =267 ± 48, KO = 147 ± 17, p<0.05) (Fig 2-3B).

Interestingly, insulin sensitivity in Ucn 2/Ucn 3 dKO mice, determined using an ITT, was similar to WT control mice in the LFD groups (Fig. 3B), but significantly decreased in the Ucn 2/Ucn 3 dKO mice in the HFD (F(1,16) = 6.50, p = 0.02)(Fig. 2-

3C). In the in HFD group, the Ucn 2/Ucn 3 dKO had a lower percent decrease of blood glucose at the 30 min time point of the ITT compared to WT controls (WT = -80

± 12, KO = -37 ± 3, p < 0.05) (Fig. 2-3D).

Ucn 2/Ucn 3 dKO mice have higher total cholesterol and triglyceride concentrations compared to WT controls

Ucn 2/Ucn 3 dKO mice demonstrated significantly higher concentrations of circulating cholesterol (F(1,40) = 60.3, p<0.0001) (Fig. 2-4A) and triglycerides

(F(1,40) = 11.9, p = 0.001) (Fig. 2-4B). Cholesterol concentration was significantly higher in Ucn 2/Ucn 3 dKO in both LFD (WT = 173 ± 20, dKO = 330 ± 19, p < 0.0001) and HFD (WT = 187 ± 13, dKO = 315 ± 19, p < 0.0001) groups. Triglyceride concentration was significantly higher in Ucn 2/Ucn 3 dKO on a LFD (WT = 60 ± 3, dKO = 80 ± 7, p < 0.05) and showed a similar trend in the HFD group (WT = 64 ± 7, dKO = 82 ± 3)(Fig. 2-4B). Intriguingly, although the Ucn 2/Ucn 3 dKO mice had increased body weight (Fig. 2-1), there was a main effect of genotype with lower percentage of epididymal fat pad weight to body weight in the Ucn 2/Ucn 3 dKO mice compared to WT controls (F(1,40)=33.54, p<0.0001) (Fig. 2-4C). On HFD, Ucn 2/Ucn

3 dKO mice had a significantly lower percentage of epididymal fat pad weight to body weight compared to WT controls (WT = 5.2% ± 0.4, dKO = 2.9% ± 0.2, p<0.0001). In the LFD group, there was a trend for lower percentage of epididymal fat pad weight to body weight in Ucn 2/Ucn 3 dKO mice (WT = 4.5% ± 0.3, dKO = 3.8% ± 0.1), but this did not reach significance (Fig. 2-4C).

Discussion

Maintaining energy homeostasis and body weight is achieved by balancing energy intake and expenditure. Afferent signals indicating the presence of physiological and/or psychological challenges are integrated centrally, and the efferent pathways controlling feeding behavior and energy expenditure are modified accordingly. Our results add to the growing literature centered on the roles of Ucn 2 and Ucn 3 acting through CRFR2 [7, 14, 18, 24-34] and support the hypothesis that these peptides are critical modulators of both centrally controlled metabolic functions and fuel utilization in key metabolic peripheral tissues.

The Ucns/CRFR2 system is highly expressed in hypothalamic regions directly associated with control of feeding and energy balance [2, 6, 9, 14, 16, 17, 23, 35].

While the single Ucn 2 KO or Ucn 3 KO mice models did not differ in their body weight compared to WT controls on either LFD or HFD, the Ucn 2/Ucn 3 dKO mice presented here gained more weight than WT controls in both LFD and HFD conditions, despite consuming similar amounts of food. Intriguingly, the Ucn 2/Ucn 3 dKO mice on HFD, showed lower percentages of epididymal fat pad weight to total body weight, suggesting that dKO mice are heavier due to increased lean tissue.

Comparing this phenotype with the single KOs after HFD, Ucn 2 KO mice had higher percentage of lean tissues, while Ucn 3 KO mice had similar body composition to WT controls. The current data and the published studies support the involvement of both

Ucn 2 and Ucn 3 in body weight regulation. Mechanisms responsible for higher body weight in the Ucn 2/Ucn 3 dKO without increased food intake have not been determined, but potential explanations include differences in mechanisms of energy efficiency or energy expenditure between genotypes.

The Ucns/CRFR2 system-deficient mouse models that have been metabolically characterized to date (Ucn 2 KO, Ucn 3 KO, CRFR2 KO and Ucn 2/Ucn

3 dKO) show improved glucose tolerance after HFD, suggesting a major role of this system in regulating glucose metabolism under challenge [24-26, 28]. Since Ucn 2 has been shown to inhibit insulin signaling in muscle, the increased insulin sensitivity phenotype of the Ucn 2 KO mice has been attributed in part to the absence of the endogenously expressed Ucn 2 in the skeletal muscle. In contrast, the improved glucose tolerance in the Ucn 3 KO mice has been attributed to the absence of pancreatic Ucn 3, which has been demonstrated increase glucose-induced insulin secretion from the pancreatic beta cells [28]. The improved glucose tolerance profile of the Ucn 2/Ucn 3 dKO mice is similar to that reported for the Ucn 2 KO [26].

Whether this phenotype is primarily due to lack of Ucn 2 in skeletal muscle, lack of

Ucn 3 in pancreatic beta cells, a combination of both, or entirely different mechnisms, remains to be determined. The ITTs in this study reveal that on LFD, while Ucn 2 and

CRFR2 deficient mice models showed a significant increase in insulin sensitivity [24,

26], the dKO mice had comparable glucose concentration following insulin administration. On HFD, the Ucn 2/Ucn 3 dKO mice actually had decreased insulin sensitivity, which is the opposite of what was reported in the single KO mice. One possible explanation for these differences is the higher concentration of circulating insulin measured in the Ucn 2/Ucn 3 dKO mice compared to WT controls, a phenotype unique to the dKO mice and not present in the Ucn 2, Ucn 3 or CRFR2 single KOs. For reasons that remain unknown, the Ucn 2/Ucn 3 dKO have higher insulin concentration which confounds the interpretation of the GTTs and ITTs.

Serum cholesterol and triglycerides were significantly higher in the Ucn 2/Ucn

3 dKO mice compared to WT controls on LFD and HFD. While these parameters were not reported for the single Ucn KO models, the CRFR2 KO mice did have lower serum concentrations of triglycerides, cholesterol, and free fatty acids compared to

WT controls when fed a HFD [24, 37]. At present, the role of the Ucns/CRFR2 system in adipose tissue and fatty acids metabolism is poorly investigated. Our current findings provide rationale to investigate a possible role for the Ucns/CRFR2 system in adipose tissue metabolic functions.

Since the overall observed metabolic phenotype of the Ucn 2/Ucn 3 dKO mice described in this study is different from the reported CRFR2 KO phenotype, the possible involvement of Urocortin 1 (Ucn 1) should be investigated further. Ucn 1, an additional member of the CRF family of peptides has high and equal affinity for both

CRFR1 and CRFR2 [37]. Thus, a difference between the Ucn 2/Ucn 3 dKO mice and the CRFR2 KO mice would be Ucn 1 actions on CRFR2. Since Ucn1 is expressed both centrally and peripherally it may contribute to the observed differences, however, the role of Ucn 1 in mediating metabolic functions and glucose homeostasis has not been explored in detail. Determining the metabolic phenotype of the Ucn 1 KO mouse [38] and the recently established (but not yet metabolically characterized)

Ucn1/2/3 triple KO mouse model [32] may provide a better understanding of this complex regulatory system.

It is also important to recognize the possibility of developmental compensation mechanisms contributing to the observed phenotypes. Tissue specific and conditional

KOs are needed to fully understand the roles of these peptides in specific metabolic tissues in regulating energy balance under basal and metabolically challenged

conditions. Additionally, an investigation of expression levels of the CRF family of peptides and receptors could reveal if compensation is playing a role in the manifestation of these phenotypes. Future studies should also include the side-by- side comparison of single KOs and multiple KOs to elucidate the individual roles of the peptides.

In summary, the novel Ucn 2/Ucn 3 dKO mouse model presented in this study supports complex roles for the Ucns/CRFR2 system in the regulation of energy balance by affecting both glucose homeostasis and lipid metabolism. Specifically, we found that the Ucn 2/ Ucn 3 dKO mice had higher body weight, lower body epididymal fat percentage and higher concentrations of insulin, glucose, and lipids compared to

WT controls on the same diet. The present establishment and characterization of the

Ucn 2/Ucn 3 dKO mouse model emphasizes the need for specific mouse models that will separate the central and peripheral contributions of these CRF family members to the regulation of energy balance under basal and metabolically challenged conditions. Availability of metabolic tissue-specific mouse models for the Ucn/CRFR2 system will provide more detailed models for studying the mechanisms of action of

Ucn 2 and Ucn 3.

Figures

A 50 * * * * * 40 * * * * WT - LFD * * WT - HFD 30 Ucn2/Ucn3 dKO - LFD Ucn2/Ucn3 dKO - HFD

0 5 10 15 Week on Diet B C WT WT 50 Ucn 2/Ucn 3 dKO 30 Ucn 2/Ucn 3 dKO

40 * 20 30

20 10 10

0 0 LFD HFD LFD HFD

Figure 2-1. Weight gain and food intake of Ucn 2/ Ucn 3 dKO mice and WT controls on LFD or HFD. (A), Growth curves of Ucn 2/ Ucn 3 dKO mice and WT during a 17- week feeding study. Ucn 2/Ucn 3 dKO mice weighed more than WT controls starting at week 6 of the study (F(1,44) = 4.06). (B), End weights of Ucn 2/Ucn 3 dKO mice at end of the 17-week study. On LFD, Ucn 2/Ucn 3 dKO mice weighed significantly more than WT controls and Ucn 2/Ucn 3 dKO mice trended towards weighing more on a HFD. (C), Average daily food intake measurements show similar food consumption between Ucn 2/Ucn 3 dKO and WT mice. Daily food measurements were collected for 7 days during week 8 of the study. Significance was determined using two-factor ANOVA followed by post hoc Bonferroni tests. All data are displayed as mean ± SEM. * = p < 0.05.

A B WT WT 150 Ucn 2/Ucn 3 dKO 2.5 * Ucn 2/Ucn 3 dKO 2.0 * 100 1.5

1.0 50 0.5

0 0.0 LFD HFD LFD HFD

Figure 2-2. Fasting glucose and insulin concentrations of Ucn 2/Ucn 3 dKO mice compared to WT controls. (A) Concentration of glucose F(1,41) = 10.68, p = 0.002) and (B) concentration of insulin (F(1,37) = 6.41, p = 0.016) were higher in Ucn 2/Ucn 3 dKO mice compared to WT controls. Differences reached significance in the HFD but not the LFD group. Significance was determined using two-factor ANOVA followed by post hoc Bonferroni tests. All data are displayed as mean ± SEM. * = p<0.05.

A B 350 350 300 300 250 250 200 ** ** 200 150 * 150 * 100 100 WT WT 50 Ucn 2/Ucn 3 dKO 50 Ucn 2/Ucn 3 dKO 0 0 0 20 40 60 80 100 0 20 40 60 80 100 Time after injection (min) Time after injection (min)

C D 0 0 * -50 -50

-100 -100

-150 WT -150 WT Ucn 2/Ucn 3 dKO Ucn 2/Ucn 3 dKO -200 -200 0 20 40 60 80 100 0 20 40 60 80 100 Time after injection (min) Time after injection (min)

Figure 2-3. Glucose homeostasis phenotypes of Ucn 2/Ucn 3 dKO mice. (A-B) Glucose and (C-D) insulin tolerance tests in Ucn 2/Ucn 3 dKO and WT controls on a (A-C) LFD or (B-D) HFD graphed as percent change from baseline blood glucose. (A- B) Glucose tolerance was significantly improved in the dKO mice compared to WT controls on LFD (F(1,22) = 9.71, p = 0.005) and HFD (F(1,19) = 4.70, p = 0.043). In the LFD group, percent chance of blood glucose was significantly higher in the Ucn 2/Ucn 3 dKO compared to WT at t =30, 60, and 90 min. In the HFD group, percent chance of blood glucose was significantly higher in the Ucn 2/Ucn 3 dKO compared to WT at t = 30 min. (C-D) On LFD, there was no difference in insulin tolerance between Ucn 2/Ucn 3 dKO mice and WT, but on HFD, Ucn 2/ Ucn 3 dKO animals have a lower percent decrease of blood glucose in response to insulin at t= 30 min. Significance was determined using ANOVA with repeated measures analysis followed by post hoc Bonferroni tests. All data are displayed as mean ± SEM. * = p<0.05, ** = p<0.01.

WT Ucn 2/Ucn 3 dKO A 400 **** **** 300

200

100

0 LFD HFD

B WT Ucn 2/Ucn 3 dKO 100 * 75

0 LFD HFD C WT 6 Ucn 2/Ucn 3 dKO

3 ****

0 LFD HFD

Figure 2-4. Lipid profiles of Ucn 2/Ucn 3 dKO mice after overnight fast. (A) Cholesterol concentration was higher in Ucn 2/Ucn 3 dKO mice compared to WT controls (F(1,40)=60.33, p<0.0001) and was significant in both LFD (p < 0.0001) and HFD (p < 0.0001) groups. (B) Lipid concentration was higher in Ucn 2/Ucn 3 dKO mice compared to WT controls (F(1,40)=11.9, p=.0013) and was significant between genotypes in the in the LFD group (p < 0.05) but not the HFD group. (C) Epididymal fat pad as a percentage of total body weight was lower in Ucn 2/ Ucn 3 dKO compared to WT mice (F(1,40)=33.5, p<0.0001) and was significant in the HFD group (p<0.0001) but not the LFD group. Significance was determined using two-factor ANOVA followed by post hoc Bonferroni tests. All data are displayed as mean ± SEM. * = p<0.05, **** = p<0.0001

References

1. Carlin, K.M., W.W. Vale, and T.L. Bale, Vital functions of corticotropin- releasing factor (CRF) pathways in maintenance and regulation of energy homeostasis. Proc Natl Acad Sci U S A, 2006. 103(9): p. 3462-7.

2. Cullen, M.J., et al., Urocortin, corticotropin releasing factor-2 receptors and energy balance. Endocrinology, 2001. 142(3): p. 992-9.

3. Dautzenberg, F.M. and R.L. Hauger, The CRF peptide family and their receptors: yet more partners discovered. Trends Pharmacol Sci, 2002. 23(2): p. 71-7.

4. Fekete, E.M. and E.P. Zorrilla, Physiology, pharmacology, and therapeutic relevance of urocortins in mammals: ancient CRF paralogs. Front Neuroendocrinol, 2007. 28(1): p. 1-27.

5. Kuperman, Y. and A. Chen, Urocortins: emerging metabolic and energy homeostasis perspectives. Trends Endocrinol Metab, 2008. 19(4): p. 122-9.

6. Lewis, K., et al., Identification of urocortin III, an additional member of the corticotropin-releasing factor (CRF) family with high affinity for the CRF2 receptor. Proc Natl Acad Sci U S A, 2001. 98(13): p. 7570-5.

7. Pelleymounter, M.A., et al., Behavioral and neuroendocrine effects of the selective CRF2 receptor agonists urocortin II and urocortin III. Peptides, 2004. 25(4): p. 659-66.

8. Perrin, M.H. and W.W. Vale, Corticotropin releasing factor receptors and their ligand family. Ann N Y Acad Sci, 1999. 885: p. 312-28.

9. Reyes, T.M., et al., Urocortin II: a member of the corticotropin-releasing factor (CRF) neuropeptide family that is selectively bound by type 2 CRF receptors. Proc Natl Acad Sci U S A, 2001. 98(5): p. 2843-8.

10. Richard, D., Q. Lin, and E. Timofeeva, The corticotropin-releasing factor family of peptides and CRF receptors: their roles in the regulation of energy balance. Eur J Pharmacol, 2002. 440(2-3): p. 189-97.

11. Hsu, S.Y. and A.J. Hsueh, Human stresscopin and stresscopin-related peptide are selective ligands for the type 2 corticotropin-releasing hormone receptor. Nat Med, 2001. 7(5): p. 605-11.

12. Chen, A., et al., Urocortin II gene is highly expressed in mouse skin and skeletal muscle tissues: localization, basal expression in corticotropin- releasing factor receptor (CRFR) 1- and CRFR2-null mice, and regulation by glucocorticoids. Endocrinology, 2004. 145(5): p. 2445-57.

13. Fukuda, T., et al., Urocortin 1, urocortin 3/stresscopin, and corticotropin- releasing factor receptors in human adrenal and its disorders. J Clin Endocrinol Metab, 2005. 90(8): p. 4671-8.

14. Inoue, K., et al., Human urocortin II, a selective agonist for the type 2 corticotropin-releasing factor receptor, decreases feeding and drinking in the rat. J Pharmacol Exp Ther, 2003. 305(1): p. 385-93.

15. Li, C., et al., Urocortin III is expressed in pancreatic beta-cells and stimulates insulin and glucagon secretion. Endocrinology, 2003. 144(7): p. 3216-24.

16. Li, C., et al., Urocortin III-immunoreactive projections in rat brain: partial overlap with sites of type 2 corticotrophin-releasing factor receptor expression. J Neurosci, 2002. 22(3): p. 991-1001.

17. McCrimmon, R.J., et al., Corticotrophin-releasing factor receptors within the ventromedial hypothalamus regulate hypoglycemia-induced hormonal counterregulation. J Clin Invest, 2006. 116(6): p. 1723-30.

18. Ohata, H. and T. Shibasaki, Effects of urocortin 2 and 3 on motor activity and food intake in rats. Peptides, 2004. 25(10): p. 1703-9.

19. Samuelsson, S., et al., Corticotropin-releasing factor 2 receptor localization in skeletal muscle. J Histochem Cytochem, 2004. 52(7): p. 967-77.

20. Saruta, M., et al., Urocortin 3/stresscopin in human colon: possible modulators of gastrointestinal function during stressful conditions. Peptides, 2005. 26(7): p. 1196-206.

21. Takahashi, K., et al., Expression of urocortin III/stresscopin in human heart and kidney. J Clin Endocrinol Metab, 2004. 89(4): p. 1897-903.

22. Takahashi, K., et al., Expression of urocortin 3/stresscopin in human adrenal glands and adrenal tumors. Peptides, 2006. 27(1): p. 178-82.

23. Van Pett, K., et al., Distribution of mRNAs encoding CRF receptors in brain and pituitary of rat and mouse. J Comp Neurol, 2000. 428(2): p. 191-212.

24. Bale, T.L., et al., Corticotropin-releasing factor receptor-2-deficient mice display abnormal homeostatic responses to challenges of increased dietary fat and cold. Endocrinology, 2003. 144(6): p. 2580-7.

25. Bale, T.L., et al., Mice deficient for corticotropin-releasing hormone receptor-2 display anxiety-like behaviour and are hypersensitive to stress. Nat Genet, 2000. 24(4): p. 410-4.

26. Chen, A., et al., Urocortin 2 modulates glucose utilization and insulin sensitivity in skeletal muscle. Proc Natl Acad Sci U S A, 2006. 103(44): p. 16580-5.

27. Chen, A., et al., Urocortin 2-deficient mice exhibit gender-specific alterations in circadian hypothalamus-pituitary-adrenal axis and depressive-like behavior. J Neurosci, 2006. 26(20): p. 5500-10.

28. Li, C., et al., Urocortin 3 regulates glucose-stimulated insulin secretion and energy homeostasis. Proc Natl Acad Sci U S A, 2007. 104(10): p. 4206-11.

29. Jamieson, P.M., et al., Urocortin 3 modulates the neuroendocrine stress response and is regulated in rat amygdala and hypothalamus by stress and glucocorticoids. Endocrinology, 2006. 147(10): p. 4578-88.

30. Chen, P., et al., Injection of Urocortin 3 into the ventromedial hypothalamus modulates feeding, blood glucose levels, and hypothalamic POMC gene expression but not the HPA axis. Am J Physiol Endocrinol Metab, 2010. 298(2): p. E337-45.

31. Fekete, E.M., et al., Delayed satiety-like actions and altered feeding microstructure by a selective type 2 corticotropin-releasing factor agonist in rats: intra-hypothalamic urocortin 3 administration reduces food intake by prolonging the post-meal interval. Neuropsychopharmacology, 2007. 32(5): p. 1052-68.

32. Neufeld-Cohen, A., et al., A triple urocortin knockout mouse model reveals an essential role for urocortins in stress recovery. Proc Natl Acad Sci U S A, 2010. 107(44): p. 19020-5.

33. Tabarin, A., et al., Role of the corticotropin-releasing factor receptor type 2 in the control of food intake in mice: a meal pattern analysis. Eur J Neurosci, 2007. 26(8): p. 2303-14.

34. Zorrilla, E.P., et al., Human urocortin 2, a corticotropin-releasing factor (CRF)2 agonist, and ovine CRF, a CRF1 agonist, differentially alter feeding and motor activity. J Pharmacol Exp Ther, 2004. 310(3): p. 1027-34.

35. Chao, H., et al., Type 2 corticotropin-releasing factor receptor in the ventromedial nucleus of hypothalamus is critical in regulating feeding and lipid metabolism in white adipose tissue. Endocrinology, 2012. 153(1): p. 166-76.

36. Kuperman, Y., et al., Perifornical Urocortin-3 mediates the link between stress-induced anxiety and energy homeostasis. Proc Natl Acad Sci U S A, 2010. 107(18): p. 8393-8.

37. Vaughan, J., et al., Urocortin, a mammalian neuropeptide related to fish urotensin I and to corticotropin-releasing factor. Nature, 1995. 378(6554): p. 287-92.

38. Zalutskaya, A.A., et al., Impaired adaptation to repeated restraint and decreased response to cold in urocortin 1 knockout mice. Am J Physiol Endocrinol Metab, 2007. 293(1): p. E259-63.

Chapter 3:

Effect of Ucn 2 deficiency on skeletal muscle fatigability and mitochondrial biogenesis

in mice

42 43

Abstract

Impairments in mitochondrial function have been correlated with metabolic diseases such as obesity and diabetes. Urocortin 2 (Ucn 2) and its receptor, corticotropin-releasing factor receptor type 2, are highly expressed by skeletal muscle and have been hypothesized to affect metabolic pathways such as insulin signaling and skeletal muscle mass, but their roles in mitochondrial mechanisms and muscle fatigue have not been explored. Using the Ucn 2 KO model, we sought to determine if Ucn 2 plays a role in mitochondrial biogenesis and muscle fatigue. Our studies reveal that muscle from Ucn 2 KO mice fatigued faster that WT controls and had decreased nuclear gene expression of the key mitochondrial proteins CPT1, PGC-1α, and PDK4. However, there were no differences in fiber type composition, mitochondrial copy number, or mitochondrial enzyme activity between skeletal muscle of Ucn 2 KO mice and WT controls. Taken together, these studies suggest that Ucn 2 plays an endogenous role in skeletal muscle fatigability without changes in mitochondrial density or activity.

Introduction

Skeletal muscle mitochondria are key regulators of energy balance and metabolism. Skeletal muscle can adapt to changes in energy need or work demand by altering mitochondrial density or mitochondrial enzyme activity, which in turn alters oxidative capacity and the production of ATP [1-3]. For example, exercise increases mitochondrial density and activity in skeletal muscle [4, 5]. Mitochondrial biogenesis, or the increase of mitochondrial proteins and activity, is a complex process that involves many orchestrated pathways of both nuclear and mitochondrial genomes [6,

7]. Such mitochondrial adaptations in skeletal muscle are fundamental to the health of an organism. Decreased mitochondrial oxidative capacity and abnormal mitochondrial biogenesis have been correlated with age and diseases such as metabolic syndrome and obesity [8, 9].

Urocortin 2 (Ucn 2) and its cognate G-protein coupled receptor, corticotropin releasing factor receptor type 2 (CRFR2) are both expressed by skeletal muscle [10-

12]. Whole-body Ucn 2 knockout (KO) mice display metabolic phenotypes that have been attributed to the loss of Ucn 2 in skeletal muscle. Because Ucn 2 inhibits insulin signaling, Ucn 2 KO mice remain sensitive to insulin after HFD while wild-type (WT) controls become characteristically insulin resistant [13]. Likewise, Ucn 2 KO mice maintain normal concentrations of insulin and glucose while WT controls develop high fat diet-induced hyperglycemia and hyperinsulemia [13]. Body composition analyses reveal that the Ucn 2 KO mice remain leaner than WT controls after HFD [13]. The mechanisms responsible for the “leanness” phenotype have not been determined.

Body weight represents a balance between food intake and energy expenditure. Since Ucn 2 KO mice have similar food intake to WT controls [13], we

were interested in exploring the potential role of Ucn 2 in mechanisms that regulate energy expenditure. Because skeletal muscle is a major determinant of whole-body energy expenditure, we focused on the role of Ucn 2 in regulating skeletal muscle mitochondrial biogenesis. Considering the lean phenotype of the Ucn 2 KO mice, we hypothesized that skeletal muscle in Ucn 2 KO mice contain increased mitochondrial density and activity.

To address this hypothesis, our studies focused on analyzing the skeletal muscle of Ucn 2 KO and littermate control, WT mice. We aimed to identify the role of

Ucn 2 in skeletal muscle fatigability and mitochondrial biogenesis by comparing the skeletal muscle of Ucn 2 KO mice and WT controls in terms of muscle fatigue, gene expression, mitochondrial protein content, fiber type differences, mitochondrial copy number, and mitochondrial enzyme activity. We also considered the possibility that the Ucn 2 KO mice needed to be metabolically perturbed in order for phenotypes to manifest. To test this, groups of mice were exposed to both acute (treadmill) and chronic (running wheel) exercise protocols before analyzing skeletal muscle for differences in mitochondrial phenotypes.

Methods

Animals

Generation of the Ucn 2 KO mouse has been reported [13]. All experiments were done in adult male mice and WT controls were age-matched littermates.

Animals were housed in 12 hr light (0600 hr to 1800 hr), 12 hr dark (1800 hr to 0600 hr) cycle with free access to water and standard rodent chow. All experiments and procedures were approved by the Salk Institutional Animal Care and Use Committee.

Fatigue test

Mice were anesthetized with CO2 and sacrificed with rapid neck disarticulation

(N=12). The fifth toe muscle of the extensor digitorum longus (EDL) was surgically removed. The muscle was placed into a specialized muscle chamber containing cold

Ringer’s solution and muscle heads were attached with sutures to a force transducer and a fixed post. A single supramaximal current pulse (100Hz, 400ms pulse duration) was delivered to the muscle every 3 sec by platinum plate electrodes. LabView software (National Instruments, Austin, TX) was used to collect and automatically record tension measurements.

Fiber type composition

Fiber type composition was determined by separating myosin heavy chain

(MHC) isoforms using SDS-PAGE and silver staining methods. Frozen muscles

(N=5) were homogenized in Buffer A (250mM sucrose, 100mM potassium chloride,

20mM trizma base, 5mM EDTA, pH 6.8). The muscle homogenate was centrifuged at 1000 x g for 10 min (4°C). The supernatant was aspirated and the resulting pellet

was resuspended in Buffer B (175mM potassium chloride, 20mM trizma base, 2mM

EDTA, 0.50% triton x-100, pH 6.8) and centrifuged at 1000 x g for 10 min (4°C). New supernatant was aspirated and the pellet was resuspended in solution C (150mM potassium chloride, 20mM trizma base) and vortexed thoroughly. Protein concentration was determined by the Pierce® BSA Protein Assay Kit (Thermo Fisher

Scientific, Waltham, MA, USA). Equal protein concentrations were loaded on acrylamide gels (4% acrylamide for stacking and 8% acrylamide for resolving) at a constant current of 10 mA at 275 V for 21 h (4°C). Silver staining was performed using Silver Stain Plus (Bio-Rad Laboratories, Hercules, CA, USA) according to manufacturer’s recommendations and MHC bands were quantified using a densitometer.

Gene expression studies

Skeletal muscle (N=8) was surgically removed and then flash frozen in liquid nitrogen. Muscle was made into a powder with a mortar and pestle. Powdered muscle in lysis buffer and B-ME was then homogenized using a tissue homogenizer and a sonicator. Muscle homogenate was treated with proteinase K (Qiagen,

Valencia, CA, USA) for 10 min @ 55°C. To complete the homogenization process, samples were spun through a Qiagen shredder according to manufacturer’s specifications. RNA was extracted from skeletal muscle using RNeasy Mini Kit

(Qiagen, Valencia, CA, USA) according to manufacturer’s instructions. RNA was quantified using a NanoDrop spectrophotometer at 260 and 280 nm. Complementary

DNA was synthesized from 1 ug RNA using the High Capacity cDNA Reverse

Transcription Kit (Applied Biosystems, Carlsbad, CA, USA). To measure relative

gene expression, RT-PCR was performed using the LightCycler 480 System (Roche

Applied Biosciences, Indianapolis, IN). Real time PCR experiments were carried out using 364 well plates with each well containing 5µL LightCycler 480 SYBR Green I

Master mix (Roche Applied Biosciences, Indianapolis, IN, USA), 1µL forward primer,

1µL reverse primer, 2µL H2O and 1µL cDNA. Primers were designed using Universal

Probe Library (Roche Applied Biosciences, Indianapolis, IN, USA). All CT values were normalized to the housekeeping gene, HPRT, and set relative to appropriate control.

Measurement of mitochondrial copy number

Genomic DNA was isolated from skeletal muscle (N=5) using a commercially available DNA isolation kit (Qiagen, Valencia, CA, USA). DNA concentration was measured using a NanoDrop. RT-PCR was performed using the LightCycler 480

System (Roche Applied Biosciences, Indianapolis, IN, USA). Primers were designed using Universal Probe Library (Roche Applied Biosciences, Indianapolis, IN, USA).

Mitochondrial DNA copy number was assessed by determining the ratio of mitochondrial DNA (mtDNA) to nuclear (nDNA).

Western blotting

Skeletal muscle (N=4) was surgically removed and flash frozen in liquid nitrogen. Muscle was homogenized in glass tubes on ice with a motorized glass homogenizer. Protein concentrations were determined using the Pierce® BSA

Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA). An antibody cocktail containing 5 mAbs was used to label different subunits of the OXPHOS

system (MitoSciences ab110412, Abcam, Cambridge, MA, USA). Proteins were visualized using standard chemoilluminescence techniques.

Succinate dehydrogenase assay

Muscle (N=8) was surgically removed, pinned lengthwise to cork and flash froze in liquid nitrogen-cooled isopentane. Cross-sections (10 µm) were prepared on glass slides and incubated with 100 mL substrate solution (0.124 g Nitro blue tetazolium, 0.186 g EDTA, 1.296 g succinic acid, 98 mL 0.1M PO4 buffer [8.7 mL

Dibasic PO4, 1.3 mL Monobasic PO4, 90mL H2O, pH 7.6], 750 µL 0.1M Azide, and 1 mL 0.1M MPMS) for 30 min at RT. Samples were then dehydrated and cover-slipped prior to analysis.

Mitochondrial isolation and citrate synthase assay

Mitochondria were isolated from homogenized muscle by centrifugation (N=8).

Briefly, muscle was homogenized with a motorized homogenizer in buffer A (210 mM mannitol, 70 mM sucrose, 1 mM EGTA, 0.5% BSA, 5 mM HEPES, 10 µg/mL sutilisin, pH 7.2). The homogenate was centrifuged at 600x g for 10 min (4°C) and supernatant was transferred to a new microtube and centrifuged at 10,000 x g for 20 min (4°C). The supernatant was aspirated and the resulting pellet was resuspended in buffer B (210 mM mannitol, 70 mM sucrose, 1 mM EGTA, 5 mM HEPES, pH 7.2).

Protein concentrations were determined using a standard BSA assay. Samples were frozen at -80°C until activity assay. Mitochondrial activity was determined using a standard citrate synthase assay as described previously [14]. Absorbance was measured at 412 nm.

Acute exercise protocol

Ucn 2 KO mice and WT littermates (N=3) were acclimated to a motorized treadmill with 3 x 10 min sessions (1 min at 5 m/min, 4 min at 10 m/min, 5 min at 5 m/min). Acclimation runs were held 48 hr, 36 hr and 24 hr before experimental exercise bout. On the day of experiment, mice were fasted 3 hr and then subjected to a 60 min running exercise bout at repetitions of variable speeds (5-25 m/min).

After exercise, mice were returned to home cages without food, but with free access to water. Muscle was collected 3 hr after exercise.

Voluntary exercise protocol

Ucn 2 KO mice and WT littermates (N=5) were singly housed for 21 days in cages equipped with free running wheels. Sedentary control mice were singly housed in cages without running wheels. After the 21-day period, running wheels were removed from the cages for 24 h. Animals were fasted, but had access to water for 4 h before tissue collection.

Body composition analysis

The EchoMRI – 100HTM Body Composition Analyzer was used to determine body composition (EchoMRI, Houston, TX, USA). Mice were weighed on a digital scale and then immobilized in a glass tube with breathing holes and scanned in the

MRI machine. Values for lean tissue mass, fat mass free water, and total water weight were automatically recorded.

Statistics

Statistical analysis was performed with GraphPad Prism 5 (GraphPad

Software Inc., La Jolla, CA, USA) for two-way analysis of variance (ANOVA) or unpaired two-tailed student’s t-test where appropriate. Statistical measures are specified in figure legends.

Results

Ucn 2 KO muscle fatigues faster than WT control

Fatigue experiments revealed that extensor digitorum longus (EDL) muscle from Ucn 2 KO mice fatigued faster than WT controls (Fig. 3-1). Absolute tension graphs show that skeletal muscle from Ucn 2 KO mice had the same tension as WT control muscle at the beginning of the fatigue test, but fatigued at a faster rate as the test progressed (Fig. 3-1A). There was a main effect of genotype on absolute tension measurements between the Ucn 2 KO mice and WT control muscle (F(1,21) = 6.65, p

= 0.018). Absolute tension measurements of Ucn 2 KO skeletal muscle and WT control muscle reached significance between time points 54 s and 81 s. Relative tension (percent decline from initial tension) graphs show that Ucn 2 KO skeletal muscle reached 50% fatigue at 51 s whereas WT control muscle reached 50% fatigue at 63 s (Fig. 3-1B). There was a main effect of genotype on relative tension between Ucn 2 KO muscle and WT control muscle (F(1,21) = 11.68, p = 0.003).

Differences in relative tension values reached significance between 45 s and 102 s.

Skeletal muscle from Ucn 2 KO mice express lower mRNA levels of mitochondrial- related genes in compared to WT controls

In both gastrocnemius and EDL muscle, we found Ucn 2 KO skeletal muscle to express lower levels of carnitine palmitoyltransferase I (CPT1), peroxisome proliferator-activated receptor gamma coactivator 1-α (PGC-1α), and pyruvate dehydrogenase lipoamide kinase isozyme 4 (PDK4) mRNA compared to WT controls

(Fig. 3-2). There were no significant differences in mRNA expression levels of cytochrome c oxidase subunit 5a (Cox5a) or cytochrome c oxidase subunit 6a

(Cox6a). In gastrocnemius muscle, Ucn 2 KO mice had a 31% reduction of CPT1,

18% reduction of PGC-1α and 66% reduction of PDK4 expression compared to controls (Fig. 3-2A). In EDL, Ucn 2 KO mice had a 41% reduction of CPT1, 32% reduction of PGC-1α and 70% reduction of PDK4 (Fig. 3-2B).

Skeletal muscle from Ucn 2 KO mice have similar fiber type composition as WT controls

As a way to further explore mitochondrial differences and to see if different fiber type composition could explain the fatigability phenotype seen in the Ucn 2 KO muscle, we determined fiber type composition of the Ucn 2 KO mice compared to WT controls. Analysis of MHC isoform composition in gastrocnemius (Fig. 3-3A), soleus

(Fig. 3-3B), EDL (Fig. 3-3C) and tibialis (TA) (Fig. 3-3D) show no differences in isoform percentages between the Ucn 2 KO mice and WT controls.

Skeletal muscle from Ucn 2 KO mice have similar mitochondrial protein levels and copy number as WT controls

The protein abundance of mitochondrial proteins comprising complexes I-V were not different in skeletal muscle of Ucn 2 KO mice and WT controls (Fig. 3-4A).

Likewise, there were no differences in mitochondrial copy number in the skeletal muscle of Ucn 2 KO mice and WT controls as determined by quantitative PCR or mitochondrial DNA and nuclear DNA (Fig. 3-4B).

Skeletal muscle from Ucn 2 KO mice have similar mitochondrial enzyme activity as

WT controls

Succinate dehydrogenase (SDH) and citrate synthase are two key mitochondrial enzymes found in the mitochondrial matrix that catalyze reaction in the citric acid cycle. SDH assays (Fig. 3-4A) and citrate synthase assays (Fig. 3-4B) revealed no differences between Ucn 2 KOs and WT controls, suggesting that the lack of Ucn 2 in skeletal muscle does not affect mitochondrial enzyme activity.

Acute exercise-induced gene expression is comparable in Ucn 2 KO mice and WT controls

We subjected the mice to an acute exercise protocol of 1 hr of treadmill running at varying speeds and then quantified relative mRNA levels of genes expected to increase with exercise. Two-factor ANOVA revealed a main effect of exercise in both PGC-1α mRNA (F(1,4) = 169.6, p = 0.0002)(Fig. 3-6A) and PDK4 mRNA (F(1,4)=15.3, p = 0.0174)(Fig. 3-6B). However, there were no significant effects of genotype.

Discussion

We conclude that Ucn 2 does not play a physiological role in skeletal muscle mitochondrial biogenesis or function as our studies find no differences in fiber type composition, mitochondrial protein content, mitochondrial copy number, or mitochondrial activity between Ucn 2 KO mice and WT controls. However, our studies reveal for the first time, that isolated muscle lacking Ucn 2 fatigues faster than control muscle, but the mechanism remains to be determined. One potential mechanism of early muscle fatigue in the Ucn 2 KO mice could be abnormal Ca2+ handling [15]. In line with our findings of fatigability in Ucn 2 KO muscle, Reutenauer-

Patte et al. recently reported that treatment of dystrophic muscle with Ucn 2 causes the muscle to be more resistant to fatigue in a mechanism attributed to Ca2+ influx pathways [16]. Dystrophic muscle, characterized by muscle wasting, is correlated with excessive Ca2+ influx [17, 18] and Ucn 2 treatment results in a normalization this Ca2+ influx via a mechanism that involves cAMP elevation and activation of protein kinase

A and Epac [16]. It is possible that higher fatigability in the Ucn 2 KO skeletal muscle is due to the lack proper Ca2+ influx, normally maintained by Ucn 2. Further studies are needed to elucidate this.

Our findings that Ucn 2 KO mice have decreased mRNA expression CPT1,

PGC-1α and PDK4 genes in skeletal muscle compared to WT control muscle are also in line with our hypothesis that Ucn 2 KO mice would have decreased expression of mitochondrial-related genes in skeletal muscle. These results are intriguing because

CPT1, PGC-1α and PDK4 are all involved in the regulation of carbohydrate and fat utilization. CPT1 is a mitochondrial enzyme that mediates the transport of fatty acids into the mitochondria [19]; less CPT1 activity would result in less fatty acid transport

and therefore more glucose utilization. PGC-1α is a transcriptional regulator and stimulator of mitochondrial biogenesis; less PGC-1α could result in less mitochondrial biogenesis. PDK4 inhibits pyruvate dehydrogenase complex, which converts glucose to acetyl-CoA [19, 20]; less PDK4 would result in more PDC activity and therefore more glucose (carbohydrate) utilization and less fatty acid transport. However, since we did not find differences in mitochondrial activity, these gene expression differences may not reflect differences in protein level and may not be physiologically significant.

The mice in our studies were in a “basal state,” without metabolic perturbation or challenge. If Ucn 2 did play a role in mitochondrial biogenesis, Ucn 2 KO mice would display abnormal exercise-induced skeletal muscle adaptations such as increase in slow type I fibers, increased mitochondrial density and activity and therefore decreased percent body fat. We have initiated such studies and find no differences in body composition (Supplemental Fig. 3-1 p. 63) between exercised Ucn

2 KO mice and exercised WT controls. Thus, we do not believe that the negative results found in our studies are due to the lack of metabolic challenge.

We report that Ucn 2 does not play a physiological role in skeletal muscle mitochondrial biogenesis. However, our fatigability finding of Ucn 2 KO muscle is novel and warrants further studies to elucidate molecular mechanisms. Furthermore,

Ucn 2 and other agonists of CRFR2 can be considered for therapeutic purposes in the treatment of diseases associated with muscle fatigue and function.

Figures

Figure 3-1. Fatigue test of isolated EDL muscle from Ucn 2 KO mice and WT controls. Supramaximal pulses (100Hz, 400ms pulse duration) were delivered to isolated EDL muscle every 3 sec for a total of 3 min. Measurements of tension were automatically recorded. (A) Graph of absolute isometric tension values show Ucn 2 KO skeletal muscle and WT control muscle began experiment with similar tension, but the rate of decline (fatigue) was significantly greater in the Ucn 2 KO muscle (F(1,21) = 6.65, p = 0.0175). (B) Graph of percent change of isometric tension from initial tension shows that the Ucn 2 KO muscle reached 50% fatigue faster than WT controls. Ucn 2 KO measurements were significantly lower than WT controls (F(1,21) = 11.68, p = 0.0026). Significance determined by ANOVA with repeated measures.

A B

WT WT 1.5 Ucn 2 KO 1.5 Ucn 2 KO

1.0 * 1.0 * * * 0.5 * 0.5 *

0.0 0.0 CPT1 Cox5a PGC-1a PDK4 Cox6a CPT1 Cox5a PGC-1a PDK4 Cox6a

Figure 3-2. Nuclear gene expression of mitochondrial-related genes in Ucn 2 KO skeletal muscle relative to expression in WT controls. Relative expression levels in (A) gastrocnemius and (B) EDL reveal lower levels of CPT, PGC-1α, PDK4 but not Cox 5a or Cox6a in Ucn 2 KO mice compared to WT controls. In gastrocnemius muscle, Ucn 2 KO mice had a 31.2% reduction of CPT1, 17.9% reduction of PGC-1α and 65.7% reduction of PDK4 expression compared to controls. In EDL, Ucn 2 KO mice had a 40.8% reduction of CPT1, 32.2% reduction of PGC-1α and 70.1% reduction of PDK4. Significance determined by unpaired two-tailed Student’s t-test.

A B 100 WT 100 WT Ucn 2 KO Ucn 2 KO 80 80

60 60

40 40

20 20

0 0 I IIa IIx IIb I IIa IIx IIb C Fiber Type D Fiber Type 100 WT 100 WT Ucn 2 KO Ucn 2 KO 80 80

60 60

40 40

20 20

0 0 I IIa IIx IIb I IIa IIx IIb Fiber Type Fiber Type

Figure 3-3. Fiber type composition of Ucn 2 KO skeletal muscle and WT controls. MHC isoforms were separated by MHC and detected by silver stain. Relative quantification was determined using a densitometer. There were no differences between Ucn 2 KO mice and WT controls in fiber type composition of (A) gastrocnemius, (B) soleus, (C) extensor digitorum longus, or (D) tibialis anterior muscle. Significance was determined by unpaired two-tailed Student’s t-test.

WT KO WT KO WT KO WT KO

ATP synthase subunit alpha Complex IV subunit I

Complex III subunit Core 2 Complex II subunit 30kDa

Complex I Subunit NDUFB8

Figure 3-4. Skeletal muscle mitochondrial density in Ucn 2 KO mice and WT controls. (A) Representative Western Blot of skeletal muscle homogenates immunoblotted with an antibody cocktail to different subunits of the OXPOS system shows no differences between Ucn 2 KO skeletal muscle and WT controls. (B) Graph of mtDNA:nDNA ratios as determined by quantitative PCR shows no differences between Ucn 2 KO mice and WT controls. Significance was determined by unpaired two-tailed Student’s t-test.

WT Ucn 2 KO

Figure 3-5. Mitochondrial activity of Ucn 2 KO skeletal muscle and WT controls. (A) Representative photographs of WT and Ucn 2 KO skeletal muscle sections stained for SDH show no differences between phenotypes. (B) Bar graphs of mitochondrial activity as determined by a citrate synthase assay show no difference in skeletal muscle mitochondria activity between Ucn 2 KO mice and WT controls. Significance was determined by unpaired two-tailed Student’s t-test.

Figure 3-6. Gene expression of PGC-1α and PDK4 in skeletal muscle of non- exercised and exercised mice. All values relative to non-exercised WT mice. Two factor ANOVA revealed a main effect of exercise in both PGC-1α mRNA (Fig 3-6A, F(1.4) = 169.57, p = 0.0002) and PDK4 mRNA (Fig 3-6B, F(1,4)=15.29, p = 0.0174). However, there were no significant effects of genotype. Significance was determined by two-factor ANOVA * = p < 0.05

Supplemental Figure 3-1. Body composition after 21-day voluntary exercise. There was no effect of exercise on (A) body weight but a main effect of exercise on (B) body fat percentage. * = P < 0.05.

References

1. Allen, D.L., et al., Cardiac and skeletal muscle adaptations to voluntary wheel running in the mouse. J Appl Physiol, 2001. 90(5): p. 1900-8.

2. Hood, D.A., Mechanisms of exercise-induced mitochondrial biogenesis in skeletal muscle. Appl Physiol Nutr Metab, 2009. 34(3): p. 465-72.

3. Rockl, K.S., et al., Skeletal muscle adaptation to exercise training: AMP- activated protein kinase mediates muscle fiber type shift. Diabetes, 2007. 56(8): p. 2062-9.

4. Henriksson, J., Training induced adaptation of skeletal muscle and metabolism during submaximal exercise. J Physiol, 1977. 270(3): p. 661-75.

5. Holloszy, J.O., Biochemical adaptations in muscle. Effects of exercise on mitochondrial oxygen uptake and respiratory enzyme activity in skeletal muscle. J Biol Chem, 1967. 242(9): p. 2278-82.

6. Cannino, G., C.M. Di Liegro, and A.M. Rinaldi, Nuclear-mitochondrial interaction. Mitochondrion, 2007. 7(6): p. 359-66.

7. Goffart, S. and R.J. Wiesner, Regulation and co-ordination of nuclear gene expression during mitochondrial biogenesis. Exp Physiol, 2003. 88(1): p. 33- 40.

8. Nisoli, E., et al., Defective mitochondrial biogenesis: a hallmark of the high cardiovascular risk in the metabolic syndrome? Circ Res, 2007. 100(6): p. 795-806.

9. Lowell, B.B. and G.I. Shulman, Mitochondrial dysfunction and type 2 diabetes. Science, 2005. 307(5708): p. 384-7.

10. Chen, A., et al., Urocortin II gene is highly expressed in mouse skin and skeletal muscle tissues: localization, basal expression in corticotropin- releasing factor receptor (CRFR) 1- and CRFR2-null mice, and regulation by glucocorticoids. Endocrinology, 2004. 145(5): p. 2445-57.

11. Kishimoto, T., et al., A sauvagine/corticotropin-releasing factor receptor expressed in heart and skeletal muscle. Proc Natl Acad Sci U S A, 1995. 92(4): p. 1108-12.

12. Perrin, M., et al., Identification of a second corticotropin-releasing factor receptor gene and characterization of a cDNA expressed in heart. Proc Natl Acad Sci U S A, 1995. 92(7): p. 2969-73.

13. Chen, A., et al., Urocortin 2 modulates glucose utilization and insulin sensitivity in skeletal muscle. Proc Natl Acad Sci U S A, 2006. 103(44): p. 16580-5.

14. Philp, A., et al., Sirtuin 1 (SIRT1) deacetylase activity is not required for mitochondrial biogenesis or peroxisome proliferator-activated receptor-gamma coactivator-1alpha (PGC-1alpha) deacetylation following endurance exercise. J Biol Chem, 2011. 286(35): p. 30561-70.

15. Allen, D.G., G.D. Lamb, and H. Westerblad, Skeletal muscle fatigue: cellular mechanisms. Physiol Rev, 2008. 88(1): p. 287-332.

16. Reutenauer-Patte, J., et al., Urocortins improve dystrophic skeletal muscle structure and function through both PKA- and Epac-dependent pathways. Am J Pathol, 2012. 180(2): p. 749-62.

17. Hopf, F.W., P.R. Turner, and R.A. Steinhardt, Calcium misregulation and the pathogenesis of muscular dystrophy. Subcell Biochem, 2007. 45: p. 429-64.

18. Millay, D.P., et al., Calcium influx is sufficient to induce muscular dystrophy through a TRPC-dependent mechanism. Proc Natl Acad Sci U S A, 2009. 106(45): p. 19023-8.

19. Jeukendrup, A.E., Regulation of fat metabolism in skeletal muscle. Ann N Y Acad Sci, 2002. 967: p. 217-35.

20. Sugden, M.C. and M.J. Holness, Recent advances in mechanisms regulating glucose oxidation at the level of the pyruvate dehydrogenase complex by PDKs. Am J Physiol Endocrinol Metab, 2003. 284(5): p. E855-62.

Chapter 4:

Ucn 2 mRNA expression in mouse skeletal muscle

66 67

Abstract

Urocortin 2 (Ucn 2) is a member of the corticotropin-releasing factor (CRF) family that affects several pathways of skeletal muscle metabolism and function such as increasing respiratory rate and inhibiting insulin signaling. The aim of this study was to identify metabolic perturbations that can modulate the expression of Ucn 2 in skeletal muscle. We found that overnight fasting of mice caused a 31% reduction of

Ucn 2 transcript in skeletal muscle and that high-fat diet (HFD) feeding caused a trend of increased Ucn 2 expression compared to controls. In a cell culture model of muscle cells, differentiated C2C12 myotubes, we found that simulated exercise

(treatment with the AMPK activator, AICAR) caused a ∼5-fold increase in Ucn 2 compared to controls. However, in vivo, neither hypoxia nor exercise significantly altered skeletal muscle Ucn 2 expression. The results of the present study reveal that energy status and AICAR can modulate the expression of Ucn 2 in skeletal muscle. Future studies are needed to determine whether the current findings have identified a mechanism through which Ucn 2 modulation regulates glucose uptake in skeletal muscle in response to nutrient status.

Introduction

Urocortin 2 (Ucn 2) is a member of the corticotropin-releasing factor (CRF) family of peptides that is endogenously expressed in mouse skeletal muscle and affects numerous pathways involved in skeletal muscle metabolism and function [1-

5]. Exogenous Ucn 2 administration has been shown to be involved in important metabolic pathways in skeletal muscle such as increasing respiratory rate [5], increasing muscle mass [3], and inhibiting insulin signaling [2]. However, the physiological roles of endogenous skeletal muscle Ucn 2 remain unclear.

Ucn 2 is also expressed in cardiac muscle, and upregulation of Ucn 2 transcript has been shown to have cardioprotective effects. Several perturbations can cause Ucn 2 transcript upregulation. For example, Ucn 2 expression is increased by hypoxia in rat cardiomyocytes [6] and by inflammatory stress in mouse cardiomyocytes [7]. Ucn 2 has been shown to inhibit apoptosis, improve cardiomyocyte contractile function, reduce oxidative stress and vasodilation [8-10].

Thus, in conditions of hypoxia and inflammatory stress, Ucn 2 upregulation is proposed to be an adaptive protective mechanism that is critical for cardiac function following perturbations.

The transcriptional regulation of Ucn 2 in cardiac muscle led us to question whether Ucn 2 is also transcriptionally regulated in skeletal muscle. To date, regulation of Ucn 2 in skeletal muscle has not been well-studied. The present study investigates the effects of metabolic perturbations such as overnight fasting, high fat diet (HFD), hypoxia, and exercise on Ucn 2 mRNA expression in skeletal muscle.

Because Ucn 2 inhibits insulin signaling and therefore glucose uptake, we predict that

Ucn 2 transcript could have a “glucose-sparing” role to inhibit glucose uptake into

muscle in order to spare glucose for tissues such as the brain that rely exclusively on glucose as an energy source. Evaluating the regulation of Ucn 2, will contribute to our understanding of the physiological role of this peptide in skeletal muscle.

Methods

Animals

Adult male C57Bl/6J mice (Jackson Laboratory, Bar Harbor, ME, USA) were used in all experiments. Animals were housed on a 12 hr light (0600 hr to 1800 hr),

12 hr dark (1800 hr to 0600 hr) cycle with free access to water and food. All experiments and procedures were approved by the Salk Institutional Animal Care and

Use Committee or the University of California, San Diego Animal Care and Use

Committee.

Fasting studies

Mice were weighed and blood glucose was measured from tail blood using an automated glucometer on the night before fasting (Novamax; Nova Biomedical

Corporation, Waltham, MA, USA). Mice were placed in a clean cage with no food but with free access to water. The next morning, after 15 hr, blood glucose was measured again and animals were sacrificed by decapitation. Muscles were surgically removed and flash frozen in liquid nitrogen for later analysis.

High-fat diet studies

Male mice were individually housed and fed ad libitum either low fat (LFD,

10% kcal fat) or high fat (HFD, 45% kcal fat) diets (Research Diets, Inc., New

Brunswick, NJ, USA) for 15 weeks. After the 15 weeks, animals were euthanized for tissue collection. Body weights and epididymal fat pads were weighed using a digital scale.

Hypoxic chambers

Animals (N=5) were placed in a hypobaric chambers maintained at 0.5 atmospheric pressure (380 mmHg), which is equivalent to approximately 10% of sea level inspired O2 or about 6,000 m above sea level. Animals were initially placed in the chambers at 1.0 atmospheric pressure and over a 5 min period the pressure was gradually lowered to 0.5 atmospheric pressure and held there for 7 days. Control animals were held in the same room as the hypoxic chambers. Tissue collection was done at normal atmospheric pressure.

Acute exercise

Mice (N=4) were acclimated to a motorized treadmill with 3 x 10 min sessions

(1 min at 5m/min, 4 min at 10m/min, 5 min at 5 m/min). Acclimation runs were held

48 hr, 36 hr and 24 hr before experimental exercise bout. On the day of experiment, mice were fasted 3 hr and then subjected to 60 min or running exercise bout at repetitions of variable speeds (5-25 m/min). After exercise, mice were returned to home cages without food, but with free access to water. Muscle was collected 3 hr after exercise.

Voluntary exercise

Adult C57Bl/6J mice (Jackson Laboratory, Bar Harbor, ME, USA) were singly housed in cages equipped with free running wheels for 21 days (N=6). Sedentary control mice were singly housed in the same room in cages without running wheels.

After the 21-day period, running wheels were removed from the cages for 24 hr.

Animals were fasted but had access to water for 4 hr before tissue collection.

Cell culture

C2C12 myoblasts (ATCC, Manassas, VA, USA) were maintained in DMEM containing 20% fetal bovine serum and 1% (vol/vol) penicillin/streptomycin at 37°C in a 5% CO2-humidified atmosphere. For differentiation, cells were grown to 90% confluency and medium was replaced with DMEM containing 2% horse serum and

1% (vol/vol) penicillin/streptomycin for 5-10 d until formation of myotubes as determined by microscope.

AICAR studies

C2C12 cells were differentiated in 12-well plates and then treated with 1mM

AICAR (Toronto Research Chemicals, Ontario, CANADA) for 24 hr (N=4). At 24 hr, media was aspirated and plate was flash froze in liquid nitrogen. On ice, buffer RLT with beta-mercaptoethanol (from Qiagen RNeasy Kit, Qiagen, Valencia, CA, USA) was added to the cells and a cell scrapper was used to detach the cells from the plate. Cell lysates were transferred to a Qiagen shredder column and the RNA was isolated with the Qiagen RNAeasy Kit according to manufacturer’s recommendations

(Qiagen, Valencia, CA, USA).

For in vivo studies, mice (N=4) were injected i.p. with 250mg/kg AICAR

(Toronto Research Chemicals, Ontario, CANADA). Control animals were injected with saline at the same time as the AICAR injections. At 24 hr after injections, animals

were sacrificed by disarticulation and skeletal muscle was surgically removed and flash frozen in liquid nitrogen.

RNA isolation

Skeletal muscle was surgically removed and then flash froze in liquid nitrogen.

Muscle was made into a powder with a mortar and pestle. Powdered muscle was homogenized in lysis buffer and B-ME using a tissue homogenizer and a sonicator.

Muscle homogenate was treated with proteinase K (Qiagen, Valencia, CA, USA) for

10 min @ 55°C. To complete the homogenization process, samples were spun through a Qiagen shredder according to manufacturer’s recommendations. RNA was extracted from skeletal muscle using RNeasy Mini Kit (Qiagen, Valencia, CA, USA) according to manufacturer’s instructions. RNA was quantified using a NanoDrop spectrophotometer at 260 and 280 nm.

cDNA synthesis and quantitative PCR

Complementary DNA was synthesized from 1 ug RNA using the High

Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Carlsbad, CA, USA).

To measure relative gene expression, RT-PCR was performed using the LightCycler

480 System (Roche Applied Biosciences, Indianapolis, IN, USA). Experiments were carried out using 364-well plates with each well containing 5uL LightCycler 480 SYBR

Green I Master mix (Roche Applied Biosciences, Indianapolis, IN, USA), 1uL forward primer, 1uL reverse primer, 2uL H2O and 1uL cDNA. Primers were designed using

Universal Probe Library (Roche Applied Biosciences, Indianapolis, IN, USA). All CT

values were normalized to the housekeeping gene, HPRT and set relative to appropriate control.

Statistical Analysis

Statistical analysis was performed using GraphPad Prism 5 Software

(GraphPad Software Inc., La Jolla, CA). Results are expressed as means ± S.E.M.

Comparisons between experimental groups were made using two-way analysis of variance (ANOVA) or an unpaired two-tailed student’s t-tests where appropriate. P values < 0.05 were considered to be statistically significant. Statistical measures are specified in figure legends.

Results

Skeletal muscle Ucn 2 mRNA expression decreases with fasting

We analyzed mRNA expression levels in skeletal muscle in mice after an overnight fast and after 15 weeks of HFD. Fasted mice displayed a significant drop in blood glucose (Supplemental Fig. 4-1 p. 83) and significantly lower levels of skeletal muscle Ucn 2 mRNA compared to controls (control: 1.0 ± 0.18, fasted: 0.31 ± 0.05, p

= 0.027)(Fig. 4-1). Mice fed a high fat diet showed significantly higher body weights and trended towards higher percent epididymal fat pad to body weight compared to mice in the control group (Supplemental Fig. 4-2 p. 84). In these HFD-fed mice, skeletal muscle Ucn 2 mRNA expression levels showed a trend to be higher than in skeletal muscle of control mice (LFD: 1.0 ± 0.2, HFD: 1.5 ± 0.5)(Fig. 4-2).

Skeletal muscle Ucn 2 mRNA expression does not change with hypoxia

Skeletal muscle Ucn 2 mRNA levels in the mice exposed to 7 days of hypoxia

(0.5 atmospheric pressure, 380 mmHg), were similar to control (control: 1.0 ± 0.14, hypoxia: 1.1 ± 0.28)(Fig. 4-3), suggesting that oxygen levels do not influence skeletal muscle Ucn 2 mRNA expression levels.

Skeletal muscle Ucn 2 mRNA expression does not change with acute or chronic exercise

To explore the potential regulation of Ucn 2 mRNA expression by exercise, we subjected mice to both acute and chronic exercise protocols. We found no significant difference in Ucn 2 mRNA expression after acute exercise (control: 1.0 ± 0.27 N=4,

acute exercised: 0.85 ± 0.41) or chronic exercise (Fig 3-3B; control: 1.0 ± 0.26 N=6, chronic exercised: 0.68 ± 0.08).

Skeletal muscle Ucn 2 expression increases in vitro with exercise mimetic, AICAR

Exercise protocols in animals can often invoke stress responses in mice. To separate exercise effects from stress effects, we used 5-amino-1-β-D-ribofuranosyl- imidazole-4-carboxamide (AICAR), an exercise mimetic. AICAR mimics the effect of exercise by activating cAMP-dependent protein kinase (AMPK) [11]. Interestingly, we found a robust 5-fold increase in Ucn 2 mRNA expression in C2C12 myotubes after 24-hr treatment with AICAR (Fig 3-4A; untreated cells: 1.0 ± 0.044, AICAR- treated: 5.3 ± 0.42, p = 0.002). We also injected mice with AICAR and determined

Ucn 2 mRNA expression levels in skeletal muscle. We found a slight trend in Ucn 2 mRNA increase in skeletal muscle of mice injected with AICAR relative to levels in saline-treated mice, but this did not reach significance (Fig 3-4B; saline-treated: 1.0 ±

0.12, AICAR-treated: 1.29 ± 0.28). Thus, AICAR leads to an increase in expression of Ucn 2 in vitro, in a manner that is not carried out in vivo.

Discussion

The present study reports for the first time that transcription of the Ucn 2 gene in mouse skeletal muscle is influenced by nutritional status. After fasting, Ucn 2 mRNA levels in skeletal muscle were decreased by 31% compared to controls and

HFD feeding for 15 weeks led to a trend in increased Ucn 2 mRNA expression levels compared to control mice. We found that 7 d in hypoxic conditions, 1 hr of acute exercise, or 21 d of voluntary exercise had no effect on Ucn 2 mRNA expression levels in skeletal muscle. Additionally, the AMPK activator AICAR caused a robust 5- fold increase in Ucn 2 mRNA in differentiated C2C12 myotubes. Injection of AICAR in vivo showed a trend in Ucn 2 mRNA increase, but this did not reach significance.

In short, skeletal muscle Ucn 2 decreased with fasting, trended towards an increase with HFD and increased with exercise mimetics in vitro. This regulation of Ucn 2 in skeletal muscle provides important insight into the physiological role of Ucn 2 in skeletal muscle.

Our finding that Ucn 2 message decreases with fasting and increases with

HFD was surprising and does not support a “glucose-sparing” role in skeletal muscle.

However, fasting does induce expression of genes that favor glucose uptake such as

GLUT4 [12], possibly as a way to increase the energy uptake into the cell. A potential adaptive mechanism to increase glucose uptake, is to decrease activity of Ucn 2

(which inhibits insulin signaling). We report a trend of increased skeletal muscle Ucn

2 expression after exercise. In this case, Ucn 2 could be playing a glucose-sparing role for the brain. However, another potential explanation could be that since Ucn 2 increases oxygen consumption [5], upregulation of Ucn 2 could be an adaptive mechanism in response to higher oxygen demand in response to exercise.

We are also unable to fully explain the increase of Ucn 2 expression upon exposure to AICAR in muscle cells but not in vivo or with exercise. AICAR mimics exercise by activation of AMPK, but there are many effects of exercise such as hypoxia that would not occur with AICAR treatment. It is possible that other effects of exercise not produced by AICAR modulate the expression of Ucn 2.

To complete a comprehensive understanding of the role of Ucn 2 in skeletal muscle analysis of the regulation of its receptor CRFR2 is needed. In skeletal muscle cells, CRFR2 has been reported to be upregulated during myogenic differentiation

[13]. Interestingly, expression of CFRF2 mRNA was found to be increased in skeletal muscle of mice exposed to HFD or chronic-variable stress [13]. CRFR2 expression also increases in skeletal muscle after exposure to endotoxins [14]. Regulation of

CRFR2 by hypoxia and exercise is an interesting question that needs to be addressed in future studies.

There are several limitations to the present study. Firstly, though our studies showed significant changes of Ucn 2 mRNA expression, we were unable to demonstrate the differences at a protein level due to an unavailability of specific Ucn

2 antibody. Other mechanisms of regulation must be considered such as post- translational modifications where protein levels may not always correlate with mRNA levels. Development of better antibodies is needed in order to pursue these experiments. Further, we cannot exclude the possibility that we missed critical time points of Ucn 2 expression in our experiments. Future studies with additional time points or adjusting the time of fasting, HFD, or exercise session would address such concerns.

In conclusion, we report that nutritional status modulates Ucn 2 gene expression in skeletal muscle possibly as a protective mechanism to regulate glucose uptake in response to nutrient status or regulate oxygen consumption in response to exercise. The findings of no effect of hypoxia or exercise on Ucn 2 mRNA expression, suggest that Ucn 2 is not involved in physiological mechanisms regulating skeletal muscle during such metabolic perturbations. Our AICAR studies suggest that activation of AMPK is correlated with Ucn 2 expression levels, but more studies are necessary to determine specific mechanisms.

Figures

Figure 4-1. Effect of nutritional status on Ucn 2 mRNA expression in skeletal muscle. (A) Skeletal muscle Ucn 2 mRNA levels after overnight fast. Fasted mice had significantly lower levels of Ucn 2 mRNA compared to fed mice (control: 1.0 ± 0.18, fasted: 0.31 ± 0.05, p = 0.027) (B) Skeletal muscle Ucn 2 mRNA levels after HFD. HFD fed mice had higher Ucn 2 mRNA expression levels compared to mice fed a low fat diet (LFD: 1.00 ± 0.19, HFD: 1.55 ± 0.51). Significance was determined by unpaired two-tailed Student’s t-test, * = p < 0.05.

Figure 4-2. Effect of hypoxia on Ucn 2 mRNA expression in skeletal muscle. Expression levels of Ucn 2 in the skeletal muscle of mice exposed to hypoxia for 7 days are similar to controls (control:1.00 ± 0.14, hypoxia: 1.13 ± 0.28). Significance was determined by unpaired two-tailed Student’s t-test.

Figure 4-3. Effect of exercise on Ucn 2 mRNA expression in skeletal muscle. (A) Acute or (B) Chronic exercise had no effect on Ucn 2 RNA expression in skeletal muscle (control: 1.00 ± 0.27, acute exercised: 0.86 ± 0.41; control: 1.00 ± 0.26, chronic exercised: 0.68 ± 0.08). Significance was determined by unpaired two-tailed Student’s t-test.

Figure 4-4. Effect of AICAR on Ucn 2 mRNA expression in C2C12 myotubes and skeletal muscle. (A) Ucn 2 mRNA expression levels in C2C12 myotubes treated with 1mM AICAR for 24 hours are higher relative to levels in untreated cells (untreated cells: 1.00 ± 0.04, AICAR-treated: 5.28 ± 0.42, p = 0.0020). (B) Ucn 2 mRNA expression levels of skeletal muscle of mice injected i.p. with 250 mg/kg AICAR show a tread towards higher levels compared to saline-injected control mice (saline-treated: 1.00 ± 0.12, AICAR-treated: 1.29 ± 0.28). Significance was determined by unpaired two-tailed Student’s t-test, * = p < 0.05.

Supplemental Figures

Supplemental Figure 4-1. Blood glucose levels of fasted mice. Mice fasted overnight had significantly lower levels of blood glucose compared to control mice (control: 135.6 ± 8.8, fasted: 82.3 ± 8.6, p = 0.007). Significance was determined by unpaired two-tailed Student’s t-test, * = p < 0.05.

Supplemental Figure 4-2. Body weight and composition of mice after HFD. Mice placed on HFD had higher (A) body weight (LFD: 35.5 ± 1.2, HFD: 42.6 ± 0.7, p = 0.0001) and showed a trend towards higher (B) percentage epididymal fat pad mass to total body weight (LFD: 4.49 ± 0.27, HFD: 5.237 ± 0.4737, p = 0.1793) compared to controls. Significance was determined by unpaired two-tailed Student’s t-test, * = p < 0.05.

References

1. Chen, A., et al., Urocortin II gene is highly expressed in mouse skin and skeletal muscle tissues: localization, basal expression in corticotropin- releasing factor receptor (CRFR) 1- and CRFR2-null mice, and regulation by glucocorticoids. Endocrinology, 2004. 145(5): p. 2445-57.

2. Chen, A., et al., Urocortin 2 modulates glucose utilization and insulin sensitivity in skeletal muscle. Proc Natl Acad Sci U S A, 2006. 103(44): p. 16580-5.

3. Hinkle, R.T., et al., Urocortin II treatment reduces skeletal muscle mass and function loss during atrophy and increases nonatrophying skeletal muscle mass and function. Endocrinology, 2003. 144(11): p. 4939-46.

4. Reutenauer-Patte, J., et al., Urocortins improve dystrophic skeletal muscle structure and function through both PKA- and Epac-dependent pathways. Am J Pathol, 2012. 180(2): p. 749-62.

5. Solinas, G., et al., Corticotropin-releasing hormone directly stimulates thermogenesis in skeletal muscle possibly through substrate cycling between de novo lipogenesis and lipid oxidation. Endocrinology, 2006. 147(1): p. 31-8.

6. Buhler, K., et al., The human urocortin 2 gene is regulated by hypoxia: identification of a hypoxia-responsive element in the 3'-flanking region. Biochem J, 2009. 424(1): p. 119-27.

7. Ikeda, K., et al., Regulation of urocortin I and its related peptide urocortin II by inflammatory and oxidative stresses in HL-1 cardiomyocytes. J Mol Endocrinol, 2009. 42(6): p. 479-89.

8. Brar, B.K., et al., Urocortin-II and urocortin-III are cardioprotective against ischemia reperfusion injury: an essential endogenous cardioprotective role for corticotropin releasing factor receptor type 2 in the murine heart. Endocrinology, 2004. 145(1): p. 24-35; discussion 21-3.

9. Yang, L.Z., et al., Urocortin II enhances contractility in rabbit ventricular myocytes via CRF(2) receptor-mediated stimulation of protein kinase A. Cardiovasc Res, 2006. 69(2): p. 402-11.

10. Kageyama, K., et al., Vasodilative effects of urocortin II via protein kinase A and a mitogen-activated protein kinase in rat thoracic aorta. J Cardiovasc Pharmacol, 2003. 42(4): p. 561-5.

11. Narkar, V.A., et al., AMPK and PPARdelta agonists are exercise mimetics. Cell, 2008. 134(3): p. 405-15.

12. Heijboer, A.C., et al., Sixteen hours of fasting differentially affects hepatic and muscle insulin sensitivity in mice. J Lipid Res, 2005. 46(3): p. 582-8.

13. Kuperman, Y., et al., Expression and regulation of corticotropin-releasing factor receptor type 2beta in developing and mature mouse skeletal muscle. Mol Endocrinol, 2011. 25(1): p. 157-69.

14. Heldwein, K.A., et al., Endotoxin regulates corticotropin-releasing hormone receptor 2 in heart and skeletal muscle. Mol Cell Endocrinol, 1997. 131(2): p. 167-72.

Chapter 5:

General Discussion

88 89

The studies in this dissertation provide further evidence that Ucn 2 and Ucn 3 play significant and complex roles in regulating energy balance and metabolism. We report that the loss of both Ucn 2 and Ucn 3 in our dKO mouse model contributes to very different phenotypes compared to the loss of Ucn 2 or Ucn 3 alone. We also explored the involvement of Ucn 2 in skeletal muscle mitochondrial biogenesis and skeletal muscle function. We found that endogenous Ucn 2 is not involved in mitochondrial pathways, but is potentially involved in muscle fatigue. In addition to these studies, we report for the first time that Ucn 2 mRNA expression levels in skeletal muscle is influenced by energy availability status. The latter finding provides the intriguing insight that Ucn 2 may be involved in adaptive mechanisms to energy availability other than mitochondrial pathways.

Compared to WT controls, the Ucn 2/Ucn 3 dKO mice have increased body weight, elevated basal fasting glucose and insulin concentrations and higher concentration of circulating cholesterol and triglycerides. Due to the complexity of the dKO model, specific mechanisms linking the urocortins to these changes are difficult to isolate. The data from the present research allow for comparisons to be made with other CRF family knockout models and offer valuable direction for future studies.

To date, numerous mouse models have been generated to investigate the roles of the CRF family of peptides. Of these knockout mice, the Ucn 2 KO [10], Ucn

3 KO [13], CRFR2 KO [9] and CRFR1 KO [11] have been characterized metabolically. A mouse model overexpressing Ucn 3 has also been recently generated and metabolically characterized [12]. Additionally, there is a mouse model in which CRFR2 message was knocked down in the ventromedial nucleus of the hypothalamus [14]. In order to obtain a more integrative understanding of the roles of

the CRF family of peptides, a comparison of the phenotypes found in these mouse models is advantageous. A summary of the baseline metabolic phenotypes of the major CRF family animal models is shown in Table 1.

As a model of obesity and metabolic syndrome, the Ucn 2 KO, Ucn 3 KO,

CRFR 2 KO and Ucn 3-overexpressing mice have all been characterized after HFD in order to gain insight into the roles of these peptides in the pathogenesis of metabolic disease. Though the protocols were slightly different for each KO model, all studies suggested a possible involvement of urocortins in weight gain regulation and the development of obesity and type 2 diabetes. These phenotypes, summarized in

Table 2, collectively provide further evidence that the CRF system is involved in situations of metabolic challenge and may be a valuable target for the treatment of metabolic disorders.

Our studies provide a strong rationale to investigate the roles of the CRF family in metabolic tissues such as adipose tissue and liver, which have not been fully explored. It is possible that the higher lipid concentrations in the Ucn 2/Ucn 3 dKO mice is due to direct effects of adipose tissue or liver. Additionally, direct effects of the

CRF system in liver could contribute to the high blood glucose phenotypes observed in the Ucn 2/Ucn 3 dKO mice. Both receptor subtypes have been reported in human adipose tissue [15] and liver [16]. However, future studies are needed to reveal functional roles of the receptors in these tissues.

The complexity of our findings in Chapter 2 compelled the investigation of the roles of the CRF family of peptides to be conducted in a more specific manner. Since skeletal muscle mitochondria are major contributors to proper metabolic function, we investigated the role of Ucn 2 in skeletal muscle by characterizing skeletal muscle

from Ucn 2 KO mice. We hypothesized that the Ucn 2 KO mice would have increased mitochondrial biogenesis. Contrary to our hypothesis, we found no difference in fiber type composition, mitochondrial copy number, or mitochondrial enzyme activity between skeletal muscle of Ucn 2 KO mice and WT controls, suggesting that endogenous Ucn 2 is not involved in mitochondrial biogenesis.

However, we reveal a novel and exciting role of Ucn 2 in skeletal muscle fatigue; isolated EDL muscle from Ucn 2 KO mice fatigues faster than controls in a standard ex vivo muscle fatigue test. Though we were unable to describe specific mechanisms, we can rule out the involvement of mitochondrial–related mechanisms contributing to the fatigability phenotype. We propose mechanisms such as Ca2+ handling may be involved and should be investigated further.

For the first time, we reveal evidence that Ucn 2 transcript is regulated by energy status in skeletal muscle. We see a significant reduction in skeletal muscle

Ucn 2 mRNA after an overnight fast and a trend towards increased levels after 15 weeks HFD. Additionally, treatment of differentiated C2C12 myotubes with the exercise mimetic, AICAR, caused a robust increase of Ucn 2 message. If Ucn 2 plays an endogenous role in mechanisms affected by nutrient status, transcriptional regulation of Ucn 2 could be an adaptive mechanism in response to metabolic stress.

This exciting possibility provides further incentive for a continued investigation of the metabolic roles of Ucn 2 in skeletal muscle such as the regulation of oxygen consumption in response to energy demand.

Overall, our studies suggest a complex role for the CRF family in metabolism and specifically show that, though Ucn 2 does not play an endogenous role in skeletal muscle mitochondrial biogenesis, it may be involved in other metabolic pathways.

Based on results from our studies, the next logical step would be to develop more sophisticated animal models, such as tissue specific knockouts to determine the endogenous roles of the CRF family in specific metabolic tissues. Such studies are warranted for the continued investigation of the use of Ucn 2 and Ucn 3 as therapeutic targets for metabolic disorders.

Table 1. Summary of metabolic phenotypes of genetically manipulated CRF family animal models. x denotes that the phenotype was not reported.

R2 U2/U3 PHENOTYPE U2 KO U3 KO R2 KO U3+ VMH R1 KO KO KD Body weight Same Same Same Higher Higher Higher x

Fat Mass Same x Same Same Same Higher Lower

Food Intake Same Higher Same Same Higher Higher x Glucose Higher Same Higher Higher x Higher Higher Tolerance Insulin Higher Same Higher Same Same Higher x Sensitivity Fasting Same Lower Same Higher Lower Same Lower Insulin Fasting x Same x Higher Lower Same Lower Glucose Triglycerides x Same Same Higher x Same x

Table 2. Summary of metabolic phenotypes of genetically manipulated CRF family animal models after HFD. x denotes that the phenotype was not reported

U2/U3 PHENOTYPE U2 KO U3 KO R2 KO U3+ KO Body Weight Same Same Same Same Higher

Fat Mass Lower Same Lower Lower Lower

Food Intake Same Same Higher Same x Glucose x Higher Higher Higher Same Tolerance Insulin x Higher Higher Lower Same Tolerance Fasting insulin Lower Lower Lower Higher Lower

Fasting glucose Lower Lower Lower Higher Lower

Triglycerides x x Lower Same x