sign in sign up
<<
Home , Adiponectin, Adiponectin receptor 1, Adiponectin receptor 2, Resistin

THE PHYSIOLOGICAL ACTIONS OF IN CENTRAL AUTONOMIC NUCLEI: IMPLICATIONS FOR THE INTEGRATIVE CONTROL OF ENERGY

by

Ted Donald Hoyda

A thesis submitted to the Department of Physiology

In conformity with the requirements for

the degree of Doctor of Philosophy

Queen‟s University

Kingston, Ontario, Canada

(September, 2009)

Copyright © Ted Donald Hoyda, 2009 ABSTRACT

Adiponectin regulates feeding behavior, energy expenditure and autonomic function through the activation of two receptors present in nuclei throughout the central nervous system, however much remains unknown about the mechanisms mediating these effects. Here I investigate the actions of adiponectin in autonomic centers of the (the paraventricular nucleus) and brainstem (the nucleus of the solitary tract) through examining molecular, electrical, hormonal and physiological consequences of peptidergic signalling.

RT-PCR and in situ hybridization experiments demonstrate the presence of AdipoR1 and

AdipoR2 mRNA in the paraventricular nucleus. Investigation of the electrical consequences following activation in the paraventricular nucleus indicates that magnocellular- cells are homogeneously inhibited while magnocellular- neurons display mixed responses. Single cell RT-PCR analysis shows oxytocin neurons express both receptors while vasopressin neurons express either both receptors or one receptor. Co-expressing oxytocin and vasopressin neurons express neither receptor and are not affected by adiponectin. Median eminence projecting corticotropin releasing neurons, brainstem projecting oxytocin neurons, and thyrotropin releasing hormone neurons are all depolarized by adiponectin. Plasma adrenocorticotropin hormone concentration is increased following intracerebroventricular injections of adiponectin.

I demonstrate that the nucleus of the solitary tract, the primary cardiovascular regulation site of the medulla, expresses mRNA for AdipoR1 and AdipoR2 and mediates adiponectin induced hypotension. Adiponectin has electrical effects on a majority of medial solitary tract neurons and depolarizes those expressing mRNA for the hypotensive Y, revealing a central mechanism to modulate pressure.

ii

Finally, I show that adiponectin controls paraventricular nucleus neuron excitability by either inhibiting a tetraethyl ammonium-sensitive potassium current thereby depolarizing neurons or activating a glibenclamide-sensitive voltage independent potassium current hyperpolarizing neurons. Therefore, adiponectin differentially modulates potassium current to confer its central effects.

These results are the first to show the physiological and electrical actions of adiponectin on individual neurons in blood barrier protected central autonomic nuclei. This thesis provides a framework for how adiponectin acts centrally to coordinate whole body and feeding behavior in the .

iii

CO-AUTHORSHIP

Chapter 2: RT-PCR experiments were performed with assistance from Dr. Mark Fry. The in situ hybridization study was performed by Dr. Rexford S. Ahima. In all other instances, including conception, design, collection and analysis were performed by Ted D. Hoyda. The manuscript was written by Ted Hoyda with the assistance of Dr. Alastair V. Ferguson.

Chapter 3: Single Cell RT-PCR experiments were performed with assistance from Ms. Christie

Hopf, while icv injections of adiponectin followed by plasma assays for ACTH and TSH were performed by Dr. Willis K. Samson. Design, collection and analysis of all other experiments were performed by Ted D. Hoyda. The manuscript was composed by Ted D. Hoyda with assistance from Dr. Alastair V. Ferguson.

Chapter 4: Experiments were designed by Ted D. Hoyda. Microinjection of adiponectin in the nucleus of the solitary tract and analysis of blood pressure and rate was performed by Ms.

Pauline M. Smith. Single cell RT-PCR experiments were performed with Ms. Christie Hopf‟s assistance. All other data was collected and analyzed by Ted D. Hoyda. The manuscript was written by Ted D. Hoyda with the assistance of Dr. Alastair V. Ferguson.

Chapter 5: All experiments were designed, collected and analyzed by Ted D. Hoyda. The manuscript was written by Ted Hoyda with assistance from Dr. Alastair V. Ferguson.

iv

ACKNOWLEDGEMENTS

First and foremost I acknowledge and thank the „Boss‟, Al Ferguson for the important contributions you made to my development as a student and human-being. On my first „serious‟ day as a graduate student I remember you sharing those important words; “the data is what the data is”, a truth which has served me well throughout my studies and will for the rest of my life.

Thanks for taking a chance and allowing me to pursue an important step in my education (and for that I will never count a „foot wedge‟ on your scorecard, even if we are playing with Dean).

For all of you who I have had the pleasure of sharing a lab with past and present, thank you for making this seem rarely like work and more a time just spent with friends. Tequila

Barbara, Jungle Cat Jenny, „Dish‟, Christie (phone: “h., o., p., f..for..”), Nancy („Drew‟ ie solving the mystery of the coed-softball double play), Matty (July 23, 2009, „Fran‟ time: 9:15, just for posterity‟s sake), Fry Guy and Spud (Friday afternoon/night? I know where to find you), PMS (is that Peter Gabriel singing?) and Dr. Price (Jebus…where to begin?) thanks for the great times and interesting conversations.

Thanks to Coach G and Coach B who reveal me an important outlet with which to take out my (occasional) PhD frustrations, usually to the point where I just can‟t remember them anymore (hypoglycemia). A deep thanks to my parents, who show me the importance of being a

„finisher‟, as they are themselves. Thanks to ES, FD and LT for helping me through the last couple of months.

To the one who knows me best, I send to you my deepest thanks, appreciation and respect. I know I‟m complicated and difficult; often teetering on the brink in the whimsical fantasyland that I live in, but when I get lost Laura, I know I only have to look for your beacon, start walking and I‟ll be home.

v

TABLE OF CONTENTS

ABSTRACT ------ii

CO-AUTHORSHIP ------iv

ACKNOWLEDGEMENTS ------v

TABLE OF CONTENTS ------vi

LIST OF FIGURES ------ix

LIST OF TABLES ------xi

LIST OF ABBREVIATIONS ------xii

Chapter 1 : GENERAL INTRODUCTION ------1

ADIPOSE TISSUE AS AN ENDOCRINE ORGAN ------3

ADIPONECTIN: STRUCTURE AND FUNCTION ------3 The generation of adiponectin knockout mice 5 The sensitizing effects of adiponectin 7

ADIPONECTIN RECEPTORS: DISCOVERY ------8 The generation of knockout mice 12

SIGNAL TRANSDUCTION PATHWAYS OF THE ADIPONECTIN RECEPTORS ---- 13

ADIPONECTIN ACTIONS IN THE CENTRAL NERVOUS SYSTEM ------14 Effects on sympathetic nervous system activity and autonomic function 16

THE PARAVENTRICULAR NUCLEUS OF THE HYPOTHALAMUS ------17 Neural inputs to the PVN 18 Projections of the PVN 19 Neuronal Subtypes in the PVN: Morphological and Electrical Characteristics 20 Type 1: Magnocellular Neurons 20 Type II: Pre-autonomic Parvocellular neurons 21 Type III: Parvocellular Neuroendocrine 22 Neurochemical cell phenotypes in the PVN 23 Oxytocin 23 Vasopressin 24 vi

Thyrotropin releasing hormone 26 Corticotropin releasing hormone 27

THE NUCLEUS OF THE SOLITARY TRACT ------28 Afferent projections to the NTS 28 Efferent projections of the NTS 29 Electrophysiology of the NTS 29 Neurochemical expression in the NTS 30 31

STATEMENT OF PROBLEM ------33

Chapter 2 : ADIPONECTIN SELECTIVELY INHIBITS OXYTOCIN NEURONS OF THE PARAVENCTRICULAR NUCLEUS OF THE HYPOTHALAMUS ------34 ABSTRACT 35 INTRODUCTION 37 MATERIALS AND METHODS 38 RESULTS 44 DISCUSSION 51 ACKNOWLEDGEMENTS 58

Chapter 3 : ADIPONECTIN DEPOLARIZES PARVOCELLULAR PARAVENTRICULAR NUCLEUS NEURONS CONTROLLING NEUROENDOCRINE AND AUTONOMIC FUNCTION ------59 ABSTRACT 60 INTRODUCTION 61 MATERIALS AND METHODS 62 RESULTS 68 DISCUSSION 77 ACKNOWLEDGEMENTS 81

Chapter 4 : ADIPONECTIN ACTS IN THE NUCLEUS OF THE SOLITARY TRACT TO DECREASE BLOOD PRESSURE BY MODULATING THE EXCITABILITY OF NEUROPEPTIDE Y NEURONS ------82 ABSTRACT 83 INTRODUCTION 85 MATERIALS AND METHODS 86 RESULTS 93 DISCUSSION 99 ACKNOWLEDGEMENTS 103

Chapter 5 : IONIC MECHANISMS RESPONSIBLE FOR ADIPONECTIN INDUCED CHANGES TO THE EXCITABILITY OF PARVOCELLULAR PVN NEURONS ------104 vii

ABSTRACT 105 INTRODUCTION 106 MATERIALS AND METHODS 108 RESULTS 111 DISCUSSION 121 ACKNOWLEDGEMENTS 125

Chapter 6 : GENERAL DISCUSSION ------126 REFERENCES ------146

viii

LIST OF FIGURES

Figure 2-1 Adiponectin receptor mRNA is expressed in the PVN

Figure 2-2 Adiponectin influences the membrane potential and spike frequency of MNC PVN neurons

Figure 2-3 The effects of adiponectin are maintained when action potential induced release of neurotransmitter is inhibited

Figure 2-4 Adiponectin exhibits different effects on identified MNC neurons in the PVN

Figure 2-5 Subgroups of MNC neurons in the PVN express different profiles of adiponectin receptors

Figure 3-1 Adiponectin effects on the membrane potential of NE and PA neurons in the PVN

Figure 3-2 The effects of adiponectin are direct and proportionally similar across parvocellular PVN neurons

Figure 3-3 Adiponectin activates the hypothalamic-pituitary stress axis through excitatory actions on NE CRH neurons in the PVN

Figure 3-4 Adiponectin selectively depolarizes type 2 PA TRH neurons

Figure 3-5 The majority of type 2 PA OT neurons are depolarized by adiponectin

Figure 4-1 Microinjections of adiponectin into the mNTS decreases BP

Figure 4-2 Adiponectin's effects on the membrane potential of mNTS neurons

Figure 4-3 Adiponectin depolarizes neurons expressing NPY and GAD67

Figure 5-1 Adiponectin inhibits a voltage gated potassium conductance

Figure 5-2 Action potential broadening accompanies the depolarizing effects of adiponectin

Figure 5-3 Adiponectin induced depolarization is abolished in TEA

ix

Figure 5-4 Adiponectin activates a potassium conductance in hyperpolarized neurons

Figure 5-5 Glibenclamide inhibits adiponectin induced hyperpolarization

Figure 6-1 mRNA for the adiponectin is expressed in regions of the rat CNS

x

LIST OF TABLES

Table 2-1 Primer sets used for RT-PCR analysis of receptor expression and single cell identification

Table 3-1 Primers sets used in the detection of mRNA from single cells

Table 4-1 Primers used in RT-PCR experiments

xi

LIST OF ABBREVIATIONS

α-MSH α-melanocyte stimulating hormone

Acrp30 complement-related of 30 kDa

ACTH adrenocorticotropin hormone

AdipoR1

AdipoR2

AgRP agouti-related

AMPK adenosine monophosphate activated protein kinase

ANGII angiotensin II

ANOVA analysis of variance

ANP atrial

AP area postrema apm1 adipose specific like factor

APPL1 adaptor protein containing pleckstrin domain phosphotyrosine

binding domain and zipper motif

ARC arcuate nucleus

AUC area under curve

BAT brown

BBB blood brain barrier

BKCa large conductance calcium activated potassium channel

BMI body mass index

BNST bed nucleus of the stria terminalis

BP blood pressure

C1q complement 1 q

xii

CART cocaine and amphetamine related transcript

CCK cDNA complementary deoxyribonucleic acid

CNS central nervous system

CRH corticotropin releasing hormone

CSF cerebrospinal fluid

CVLM caudal ventrolateral medulla

CVO circumventricular organ

DA

DE delayed excitation

DEPC diethylpyrocarbonate

DIPI3α deleted in colorectal cancer (DCC)-interacting protein

DMNX dorsal motor nucleus of the vagus

ENK enkephalin

FFA free gAd globular adiponectin

GAPDH glyceraldehyde 3-phosphate dehydrogenase

GBP28 gelatin binding protein 28

HCN hyperpolarization-gated cation non-selective

HMW high molecular weight

HPA hypothalamic-pituitary-adrenal

HR heart rate icv intracerebroventricular ip intraperitoneal

xiii

IRS substrate iv intravenous

Kd dissociation constant

KO knockout

LMW low molecular weight

LTS low threshold stimulus

LXR X receptor

MAPK mitogen activated protein kinase

MMW medium molecular weight

MNC magnocellular mRNA messenger ribonucleic acid

NA noradrenalin

NE

NO nitric oxide

NPY neuropeptide Y

NTS nucleus of the solitary tract

OT oxytocin

OVLT organum vasculosum of the lateral terminalis

PA pre-autonomic

PBN parabrachial nucleus

PBS phosphate buffered saline

PI3-kinase phosphatidylinositol-3-kinase

PIR post-inhibitory rebound

POMC pro-opiomelanocortin

xiv

PPAR peroxisome proliferator-activated receptor

PVN paraventricular nucleus

RT-PCR reverse transcription-polymerase chain reaction

RVLM rostral ventrolateral medulla

SCN suprachiasmatic nucleus

SFO subfornical organ siRNA small interfering ribonucleic acid

SNP single nucleotide polymorphism

SNS sympathetic nervous system

SON supraoptic nucleus

SUR sulfonylurea receptor

TEA tetraethyl ammonium

TNF-α α

TRH thyrotropin releasing hormone

TSH stimulating hormone

TTX tetrodotoxin

UCP-2 -2

VP vasopressin

xv

Chapter 1: GENERAL INTRODUCTION

1

Obesity has become one of the most serious health concerns of the 21st century. As medically defined, is a state of increased body weight derived from augmented adipose tissue content, to levels high enough to produce adverse health consequences (Spiegelman &

Flier, 2001). Excessive weight gain characterizing the obese phenotype is associated with , intolerance, hypertension, dyslipidemia and cardiovascular , collectively termed the „‟ (Moller & Kaufman, 2005). The irony, however, is reflected in the idea that the efficient storage of excessive nutrients is actually an evolutionary adaptation to deal with periods of starvation and famine. In fact, it is estimated that an obese individual weighing up to 250 lbs can survive without food for approximately 150 days!

(Spiegelman & Flier, 2001).

The onset of obesity proceeds when the interaction between genetic, environmental and psychosocial factors disrupts the homeostatic balance between energy intake and energy expenditure. A body can only decrease its weight if total energy expenditure exceeds intake. The mechanisms responsible for such a simple concept are, however, extremely complex. Pioneering studies examining the effects of gross electrolytic lesions of the hypothalamus (see (Elmquist et al, 1999)) and subsequent detailed lesions of hypothalamic nuclei (Gold, 1973) have implicated this region in playing a vital role in controlling energy homeostasis through three different mechanisms: 1) the control of ingestive behavior and activity levels, 2) regulation of autonomic nervous system activity and 3) coordination of neuroendocrine function and thus metabolic homeostasis (Spiegelman & Flier, 2001). These mechanisms rely on complex signaling networks within the central nervous system (CNS) in addition to electrical and hormonal information from the periphery; importantly those signals derived from adipose tissue, required to mediate the control of whole body energy homeostasis and metabolic function.

2

ADIPOSE TISSUE AS AN ENDOCRINE ORGAN

Over the last 20 years, our view of adipose tissue has changed dramatically. The discovery of „‟, derived from , and the investigation of their function in the periphery and CNS, has changed our view of adipose tissue from a one dimensional limitless repository for to a dynamic active player in the regulation of energy homeostasis. Adipose tissue participates in energy homeostasis through the „context specific‟ synthesis and release of a variety of different adipokines such as tumor necrosis factor-α

(TNF-α), IL-6, , angiotensinogen, and adiponectin, among others (Ahima, 2006).

Leptin, for example, is an secreted into the blood in direct proportion to body adiposity (Ahima & Flier, 2000). This targets the hypothalamus, brainstem and other areas in the

CNS, to inhibit feeding and promote positive energy expenditure. Leptin does this through the direct suppression of neurons that synthesize and secrete the orexigenic (appetite stimulating) neuropeptide Y (NPY) and agouti-related peptide (AgRP) while simultaneously activating anorexigenic (appetite inhibiting) neurons that secrete α-melanocyte stimulating hormone (α-MSH) and cocaine-amphetamine related transcript (CART) (Gao & Horvath, 2007).

The disruption of leptin signaling in the CNS results in a morbidly obese phenotype complete with all pathologies associated with the metabolic syndrome (Ahima, 2005), demonstrating the vital importance of circulating adipokines in the control of autonomic function, feeding behavior and energy homeostasis.

ADIPONECTIN: STRUCTURE AND FUNCTION

Adiponectin was originally identified as an adipokine by four groups using different techniques. A cloning strategy using enhanced differential display of mRNA from pre- and post-

3

adipocyte differentiated 3T3-L1 and 3T3-F442A mouse fibroblasts produced the sequence encoding adipocyte complement-related protein of 30 kDa (Acrp30) (Scherer et al, 1995) and

AdipoQ (Hu et al, 1996). The sequence of human adiponectin was similarly cloned by screening adipocyte cDNA libraries, revealing adipose specific collagen like factor (apM1) (Maeda et al,

1996), while human adiponectin was isolated from plasma and named gelatin binding protein (GBP28) (Nakano et al, 1996).

The structure of adiponectin resembles that of a family of involved in the complement system (C1q) and TNF-α (Wong et al, 2004). Similar to C1q, adiponectin is comprised of four distinct regions: an N-terminal , a variable region containing a structurally important at position 39, a collagen-like domain and a C-terminal globular domain (Shapiro & Scherer, 1998). Adiponectin most often exists in the plasma in its full-length form, but can be proteolytically cleaved by leukocyte elastase, an enzyme secreted by activated monocytes, to separate the globular head (gAd) from the tail fragment (Waki et al, 2005). Full- length adiponectin can trimerize and circulate as a low-molecular weight (LMW) form or join with other trimers through the Cys39-dependent disulphide bridge composing medium molecular weight (MMW) hexamers and high molecular weight (HMW) multimers (Waki et al, 2003).

Adiponectin is unique among biologically active molecules secreted from adipocytes in that cellular mRNA expression of the ADIPOQ gene and circulating plasma concentrations of the peptide vary inversely with adiposity (Hu et al, 1996; Arita et al, 1999) and insulin resistance

(Yamamoto et al, 2004). In patients with normal body mass index (BMI), adiponectin is abundant in the plasma, estimated by quantitative western blotting to make up 0.05% of all plasma proteins

(Scherer et al, 1995) and circulates at a concentration in the range 1.9 to 17.0 µg/ml (Arita et al,

1999). Circulating concentrations of adiponectin are sexually dimorphic as females show higher

4

plasma concentrations than males (Combs et al, 2003), a difference which has been suggested to be the result of inhibitory actions of male androgens such as , on the release of adiponectin from adipocytes (Xu et al, 2005).

The ADIPOQ gene encoding human adiponectin is composed of three exons spanning 16 kb and is located on 3q27 (Saito et al, 1999). Several independent studies, examining genetic links to metabolic , have revealed a correlation between single nucleotide polymorphisms (SNPs) within this chromosome and susceptibility to type 2 and the metabolic syndrome (Vionnet et al, 2000; Kissebah et al, 2000; Hara et al, 2002). Indeed,

SNP mutations in the adiponectin gene have been positively correlated with increases in the risk of associated with the development of obesity and insulin resistance (Vasseur et al, 2002; Hara et al, 2002). An SNP in the adiponectin gene of T → G, at nucleotide 276, doubles the risk of developing these pathologies (Stumvoll et al, 2002; Menzaghi et al, 2002). This discovery led to the hypothesis that adiponectin may act as an adipocyte signaling molecule regulating insulin sensitivity and energy homeostasis, with the loss of these actions increasing risk for pathologies associated with either.

The generation of adiponectin knockout mice

Genetic manipulation of the adiponectin gene and the generation of adiponectin

„knockout‟ (KO) mice have lead to an improved understanding of the peptide‟s role in the metabolic regulation of peripheral organ systems. In 2002, three separate groups generated homozygous adiponectin KO mice (adipo - /-) by crossing mice produced from adiponectin gene deleted embryonic stem cells with C57BL/6J mice (Kubota et al, 2002; Ma et al, 2002; Maeda et al, 2002). KO mice maintain normal growth curves with no deviation from normal weight gain when exposed to a high diet. Phenotypic discrepancies concerning insulin receptor signaling, 5

glucose handling and triglycerides/free fatty acids (FFAs) however exist between the different KO mice. For example, Chan‟s group show their mice have normal glucose/insulin levels under fasted conditions, however, in opposition to published reports showing adiponectin induced β-oxidation of fatty acids (Fruebis et al, 2001; Yamauchi et al, 2001), KO mice have still increased oxidation in the muscle and liver compared to controls (Ma et al, 2002). Noda‟s and

Matsuzawa‟s groups, however, show KO mice have moderate to severe insulin resistance and intolerance to glucose (Maeda et al, 2002; Kubota et al, 2002). A comprehensive examination of the adiponectin KO phenotype is provided by Matsuzawa‟s group who show that KO mice have delayed serum FFA clearance from a decreased expression of fatty-acid transport protein-1 mRNA and compromised insulin-receptor substrate 1 (IRS-1)-associated phosphatidylinositol 3 kinase (PI3-kinase) activity in muscle (Maeda et al, 2002). KO mice show elevated serum concentration of TNF-α as well as an increase in TNF-α mRNA in their adipose tissue (Maeda et al, 2002). Re-introduction of adiponectin through adenoviral mediated supplementation decreases serum TNF-α concentration and ameliorates all of the above abnormalities (Maeda et al, 2002).

These results indicate that serum adiponectin concentration varies inversely with TNF-α, a finding that was later confirmed in humans (Kern et al, 2003) and suggest that adiponectin may negatively regulate TNF-α production or release from adipose tissue as it does from activated macrophages (Ouchi et al, 2001). KO mice also have an increased propensity to develop neointimal formations or thickening of the arterial walls, following vascular cuff injury, suggesting adiponectin may play a role in mediating susceptibility to cardiovascular disease (Kubota et al, 2002).

6

The insulin sensitizing effects of adiponectin

Adiponectin acts in peripheral tissues to sensitize cells to insulin and improve glucose tolerance (Kadowaki & Yamauchi, 2005). Intravenous (iv) injections of gAd result in a dramatic decrease in circulating FFA primed by either a high fat meal or iv injections of lipids (Fruebis et al, 2001). FFA are increasingly taken up into cells although cellular content is markedly reduced suggesting gAd is enhancing the oxidation of fatty acids (Fruebis et al, 2001).

Mice receiving chronic injections of gAd show sustainable reductions in body weight compared to wild-type litter mates when fed a high fat/high sucrose diet (Fruebis et al, 2001) suggesting the globular domain of adiponectin acts to mediate weight gain and storage of triglycerides.

A major target for peripheral adiponectin is the liver, where it acts to decrease circulating concentrations of glucose by sensitizing cells to insulin (Berg et al, 2001). The activity rate of hepatic is dynamic, responding to the body‟s need for serum concentrations of glucose. Regardless of energy state, the pathway maintains tonic activity and therefore hepatic glucose makes up some portion of circulating glucose at any given time. Intraperitoneal (ip) injections of full length adiponectin into non-satiated mice result in a marked reduction in circulating glucose levels, independent of an effect on serum insulin concentration (Berg et al,

2001). A similar response is seen in non-obese diabetic mice (mimicking type 1 diabetes) and mice treated with a peroxisome proliferator activated receptor (PPAR)-γ streptozotocin, each model representing different strategies of reducing basal insulin levels and confirming the inhibitory effect on hepatic production of glucose (Berg et al, 2001). Experiments performed on wild type mice receiving adiponectin injections while having their insulin levels regulated by a pancreatic euglycemic clamp, show a decrease in endogenous glucose production which results from inhibition of the flux of glucose through glucose-6-phosphatase, an enzyme critical for the

7

gluconeogenic pathway. This decrease in endogenous glucose production likely occurs through an adiponectin mediated reduction of mRNA for gluconeogenic enzymes, phosphenolpyruvate carboxykinase and glucose-6-phosphatase (Combs et al, 2001).

Yamauchi et al (2001) also investigated the correlation between reduced adiponectin expression and insulin resistance by examining both and liver cell composition in diabetic and lipoatrophic mice (Yamauchi et al, 2001). Re-introducing adiponectin into murine models of obesity, type 2 diabetes and lipoatrophy ameliorates symptoms of insulin resistance through the clearance of triglycerides from tissue and the increase in β-oxidation of FAs

(Yamauchi et al, 2001). This effect is concomitant with an increase in the expression of proteins involved in fatty-acid transport and combustion, as well as energy dissipation, such as acyl-CoA oxidase (Kersten et al, 1999) and uncoupling protein-2 (UCP2) (Lowell & Spiegelman, 2000).

Clearance of cellular triglycerides provides a more favorable environment with which the insulin receptor can signal, as shown by an increase tyrosine kinase phosphorylation of the receptor and insulin receptor substrate-1 as well as an increase in Akt phosphorylation following insulin stimulation compared to controls (Yamauchi et al, 2001). These findings suggest adiponectin acts through multiple mechanisms to regulate blood glucose, insulin sensitivity and cellular triglyceride content thereby preventing the development of diabetes.

ADIPONECTIN RECEPTORS: DISCOVERY

Following the discovery of the peptide, expression cloning studies were performed in an attempt to elucidate the receptor at which adiponectin conferred its biological activity (Yamauchi et al, 2003). Based on adiponectin‟s hydrophobic structure, the receptor at which the peptide binds is most likely membrane bound and present at sites of adiponectin action such as skeletal muscle and the liver. Screening of human skeletal muscle mRNA retrovirally infected into Ba/F3 8

cells for binding with the globular head of adiponectin reveals a 7-transmembrane domain integral which researchers named adiponectin receptor 1 (AdipoR1)

(Yamauchi et al, 2003). Analysis of the mouse genome for membrane proteins with highly significant to AdipoR1 revealed only one other protein; a second receptor

(with 67% homologous identity) also shown to bind both full-length and globular adiponectin termed AdipoR2. The amino acid sequences of AdipoR1 and AdipoR2 are highly conserved from Saccharomyces cerevisiae through Caenorhabditis elegans and Drosophila to

Homo sapiens suggesting these receptors developed early in the evolutionary timeline (Yamauchi et al, 2003). Intriguingly, the yeast homologue YOL002c is known to be a vital signaling component in cellular pathways that regulate phosphate and the oxidation of lipids

(Karpichev et al, 2002).

AdipoR1 and AdipoR2 are composed of 7-transmembrane spanning α-helical domains similar to G-protein coupled receptors; however their topology is rotated with respect to the membrane. In both receptors, the N-terminus is on the intracellular side of the membrane while the C-terminus exists in the extracellular space (Yamauchi et al, 2003). This feature makes them unique in the animal kingdom (Kadowaki et al, 2006). The C-terminus contains the active site(s) for both receptors and Scatchard plot analysis, an equation used to calculate the affinity constant of a ligand binding with a receptor, suggests that there are two binding sites for both globular and full-length adiponectin on either receptor (Yamauchi et al, 2003). gAd preferentially binds a high affinity site with a dissociation constant (Kd) of approximately 0.06 µg/ml, or an intermediate affinity site with a Kd of 0.80 µg/ml, equivalent to 1.14 nM and 14.4 nM respectively. Full-length adiponectin binds either an intermediate (Kd of 6.7 µg/ml) or low affinity site (Kd of 329.3 µg/ml)

9

equivalent to 49.1 nM and 2415 nM of full length adiponectin hexamer respectively (Yamauchi et al, 2003).

Adiponectin receptors are expressed ubiquitously throughout the body and can exist independently or in clusters as either homo- or hetero-multimers (Yamauchi et al, 2003;

Kadowaki & Yamauchi, 2005). Quantitative tissue expression studies show that AdipoR1 is predominantly expressed in skeletal muscle while AdipoR2 is found in the liver and both receptors are expressed in the CNS (Yamauchi et al, 2003). AdipoR1 preferentially binds gAd while AdipoR2 prefers interaction with full length peptide (Yamauchi et al, 2003).

A third receptor, the T-Cadherin molecule, has been proposed to also bind adiponectin

(Hug et al, 2004). Similar expression studies using retrovirally injected cDNA libraries into

Ba/F3 cells demonstrate that T-Cadherin can actively bind hexameric and HMW species of adiponectin but not gAd or trimeric forms. Both quaternary structure and post-translational modification seem to be important for binding as HMW adiponectin produced from non- eukaryotic species and tail-fragments lacking an N-terminal cysteine residue fail to bind the receptor (Hug et al, 2004). The cadherin family of proteins are important for calcium mediated cell-to-cell interactions and extracellular signaling (Hug et al, 2004). T-Cadherin however is unique in the cadherin family because while it is a glycosylphosphatidylinositol-anchored extracellular protein similar to the others, it contains no intracellular signaling domain. This suggests that while the T-cadherin molecule binds HMW adiponectin, it may only be one part of a more complex structure composed of as yet unidentified signaling partners and adaptor proteins.

T-Cadherin is not expressed in hepatocytes or the liver (Hug et al, 2004) and therefore plays no role in adiponectin signaling in the liver. A full understanding of the function of T-Cadherin in the adiponectin system remains to be elucidated.

10

The expression of adiponectin receptors is actively regulated by factors associated with the control of metabolism. For example, the expression of AdipoR1 and AdipoR2 mRNA is increased in skeletal muscle and liver tissue during starvation and subsequently reduced following re-feeding (Tsuchida et al, 2004). AdipoR1 and AdipoR2 receptor mRNA expression is also increased following treatment with streptozotocin to render mice insulin deficient, however, mRNA levels are reduced further then controls following re-exposure to insulin, an inhibition thought to occur through the activation of PI3-kinase and reduced Foxo1 activity (Tsuchida et al,

2004).

Adiponectin receptor mRNA expression is reduced in the ob/ob, hyperinsulinemic mouse; extrapolation of which leads to a “vicious cycle” spiraling insulin resistance and development of the metabolic syndrome. For example, reduces the expression of AdipoR1 and AdipoR2 provoking “adiponectin resistance” followed by a reduction in the anti- diabetic effects of adiponectin thus decreasing insulin signaling efficacy and aggravating hyperinsulinemia from increasing pancreatic output (Kadowaki & Yamauchi, 2005).

Further evidence suggests that of the PPAR family and the

(LXR) can change expression levels of AdipoR1 and AdipoR2. PPARs (α, β and γ) and LXRs are nuclear receptors involved in the control of and glucose homeostasis and play a role in mediating improvements in insulin sensitivity. Treatment of isolated human macrophages with synthetic LXR, PPAR-α and -γ agonists, T0901317, GW647 and GW929 respectively, increase the expression of AdipoR1 and AdipoR2 (Chinetti et al, 2004). Furthermore, injecting PPAR-α and -γ agonists, Wy14643 and rosiglitazone, into KKAy mice (obese, insulin resistant phenotype) restores AdipoR1 and AdipoR2 mRNA expression to near control levels and improves insulin

11

sensitivity (Tsuchida et al, 2005). These results suggest that adiponectin receptors present a viable therapeutic target for the development of anti-diabetic, anti-obesity drugs.

The generation of adiponectin receptor knockout mice

Studies involving the selected disruption of adiponectin receptors and the subsequent effect on phenotype have yielded some interesting and controversial results. Two independent labs have generated adiponectin receptor KO animals (AdipoR1-/-, AdipoR2-/- and in one case the deletion of both receptors (AdipoR1/2-/-)) (Yamauchi et al, 2007; Bjursell et al, 2007). KOs from each lab are viable and fertile, however their phenotypes differ following growth and development. Kadowaki‟s group reports that AdipoR1-/- mice develop normally and gain weight similarly to wild type litter-mates; however, they have a reduced activation of adenosine monophosphate activated-protein kinase (AMPK), a major intracellular signaling target for adiponectin in the liver (Yamauchi et al, 2002) (see below), leading to higher endogenous glucose production and insulin resistance (Yamauchi et al, 2007). Similarly AdipoR2-/- mice show normal weight gain and development irrespective of reduced activation of liver PPAR-α, increased liver triglyceride content and insulin resistance due to high plasma insulin levels (Yamauchi et al,

2007). Double KO animals (AdipoR1/R2-/-) show marked insulin resistance and incredibly poor glucose tolerance in response to oral administration, in addition to a substantial decrease in insulin stimulated IRS-2 tyrosine and Akt phosphorylation compared to wild type litter mates

(Yamauchi et al, 2007).

Bjursell et al (2007) show a different phenotype following the removal of either receptor in mice. AdipoR1-/- mice show an increase in adiposity and weight gain along with a decrease in spontaneous locomotor activity and energy expenditure (Bjursell et al, 2007). Conversely

AdipoR2-/- mice show a lean phenotype, being resistant to the development of obesity following 12

induction of a high fat diet (Bjursell et al, 2007). AdipoR2-/- mice have high spontaneous locomotor activity, increased energy expenditure and reduced plasma cholesterol levels (Bjursell et al, 2007). Similarly, liver AMPK is not activated in the AdipoR1-/- mice while AdipoR2-/- mice show a reduced activation of PPAR-α. Curiously, in each of the KO strains, the remaining mRNA levels of the other receptor are upregulated compared to wild-type litter-mates, suggesting a potential compensatory mechanism for adiponectin signaling and that activation of either receptor may regulate the expression of the other (Bjursell et al, 2007). In another study, the deletion of

AdipoR2 alone produced the same phenotypic outcomes as the above study, decreased lipid levels and improved insulin sensitivity in mice rendered obese due to a high fat diet, however, they eventually became diabetic owing to pancreatic β-cell failure, suggesting AdipoR2 activation is required for pancreatic islet cell maintenance (Liu et al, 2007).

SIGNAL TRANSDUCTION PATHWAYS OF THE ADIPONECTIN RECEPTORS

Adiponectin receptors are not coupled to G-proteins (Yamauchi et al, 2003). Instead, recent evidence obtained from a yeast two-hybrid screen using the intracellular N-terminal domain of AdipoR1 as bait identified APPL1 (adaptor protein containing pleckstrin homology domain, phosphotyrosine-binding domain and leucine zipper motif) otherwise known as DIP13α

(Deleted in colorectal cancer (DCC)-inter-acting protein) as the intracellular molecule that interacts with each receptor (Mao et al, 2006). APPL1 was originally identified as an adaptor protein that interacts with oncoprotein-serine/threonine kinase Akt2. It is located on Chromosome

3p14.3 – 3p21.1 and its deletion or disruption is linked to tumor development (Mitsuuchi et al,

1999). Inhibition of APPL1 expression with small interfering RNA (siRNA) decreases the activation of all major cellular signaling targets of adiponectin including AMPK, p38 mitogen activated protein kinase (MAPK) and PPARα ligand activity, reducing the physiological 13

outcomes of adiponectin treatment such as lipid oxidation and glucose uptake, whereas over- expression of the peptide has the opposite effect (Mao et al, 2006). APPL1 interacts with Rab5, a small intracellular GTPase, to increase the translocation of GLUT4 to the membrane following adiponectin receptor activation (Mao et al, 2006). Furthermore, APPL1 interacts with the PI3K signaling pathway involved in insulin sensitization (Deepa & Dong, 2009). The signaling sequence of adiponectin to increasing PPAR-α ligand activity proceeds sequentially through the activation of APPL1 upon receptor binding, followed by activation of AMPK leading to an increase in p38 MAPK, which finally increases PPAR-α activity to increase the oxidation of fatty acids (Yoon et al, 2006).

Adiponectin confers its physiological effects on glucose utilization and fatty acid oxidation by activating AMPK (Yamauchi et al, 2002). When AMPK is activated through the signaling cascade initiated by adiponectin binding to its receptors, AMPK will shut down energy consuming processes such as cell growth, proliferation and biosynthesis while increasing catabolic processes that generate cellular ATP (Kahn et al, 2005). AMPK is a cellular energy gauge because it is activated by a high AMP/ATP ratio. Additionally AMPK has multiple phosphorylation sites in which upstream molecules such as LKB1 or calmodulin-dependent protein kinase act to regulate its activity (Birnbaum, 2005; Kahn et al, 2005). When expressed together the two receptors may interact to confer signaling of all three downstream pathways known to be activated by adiponectin; AMPK phosphorylation, PPAR-α activation and p38-

MAPK activation (Kadowaki & Yamauchi, 2005).

ADIPONECTIN ACTIONS IN THE CENTRAL NERVOUS SYSTEM

In 2004 the Ahima laboratory provided the first evidence that the brain was a major target for adiponectin (Qi et al, 2004). A single injection of gAd made into the lateral cerebral ventricles 14

of C57Bl/6J mice is associated with a sustained decrease in body weight (for up to 4 days) without a concomitant decrease in food intake suggesting an increase in total energy expenditure

(Qi et al, 2004). The physiological effects of central adiponectin on weight loss are dose dependent and effective at a concentration approximately 1/1000 the effective concentration of adiponectin in the periphery (Qi et al, 2004). Levels of O2 consumption,

(BAT) UCP-1 mRNA and hypothalamic corticotropin releasing hormone (CRH) mRNA are all increased following intracerebroventricular (icv) injections of the peptide (Qi et al, 2004).

Examination of Fos-positive cells (a marker of neuronal activation) in the hypothalamus following icv or iv injections of the gAd reveals that the paraventricular nucleus of the hypothalamus (PVN) is the primary site where activation of neurons occurs. Interestingly, the arcuate nucleus (ARC), which houses feeding related peptides NPY, AgRP, POMC and CART was not activated by either route of peptide administration (icv or iv) illustrated by the lack of Fos activation (Qi et al, 2004).

While the large, hydrophilic adiponectin molecule is unlikely to cross the blood brain barrier (BBB) (Pan et al, 2006; Spranger et al, 2006), Qi et al detects adiponectin in the cerebral spinal fluid (CSF) at 1-4% the concentration of circulating levels (Qi et al, 2004). Additionally, iv injection of adiponectin increases cerebroventricular concentration of the peptide, while the same vascular increase is seen following icv injections of adiponectin consistent with a central transport mechanism for the peptide (Kubota et al, 2007; Qi et al, 2004). Importantly, in leptin deficient obese mice, adiponectin acts in the brain to reduce plasma glucose by 71%, insulin by

52%, triglycerides by 17% and total cholesterol by 29%, revealing a crucial role for the CNS in mediating adiponectin dependent control of glucose homeostasis and metabolic function (Qi et al,

2004).

15

The profile of adiponectin complexes in the cerebrospinal fluid (CSF) is different to serum profiles in humans (Kusminski et al, 2007). Kusminski et al (2007) show that CSF concentrations of full-length adiponectin are approximately 0.1% of serum concentrations and that in contrast to the peripheral circulation where HMW forms dominate, the trimeric and LMW hexameric forms of adiponectin account for the majority of the peptide in the CSF with the HMW form being undetectable (Kusminski et al, 2007).

Kadowaki‟s group has confirmed the presence of adiponectin in the CSF of mice and demonstrated that full length adiponectin is detectable in the CSF of adipo-/- mice 3 hours after iv injection of the peptide supporting a theorized mechanism of transport into the brain (Kubota et al, 2007). In situ hybridization studies using anti-sense probes for the two receptors demonstrate the expression of AdipoR1 and AdipoR2 mRNA in the PVN and the ARC and immunohistochemical analysis suggests that AdipoR1 co-localizes with in the

ARC (Kubota et al, 2007). Kubota et al (2007) show iv injections of adiponectin stimulating food intake through AMPK and ACC activation in the ARC, a phenomenon dependent upon AdipoR1 activation. This activation of AMPK is directly opposed to, and can override, the ability of leptin to inhibit AMPK in the ARC thereby inhibiting feeding (Ahima, 2005).

Effects on sympathetic nervous system activity and autonomic function

An emerging literature suggests that adiponectin controls blood pressure (BP) by modulating the sympathetic nervous system (SNS) through actions at hypothalamic and brainstem centers controlling autonomic function (Wang & Scherer, 2008). As an autonomous regulatory system which responds to physiological stressors, the dysregulation of the SNS has been shown to be associated with obesity, as well as comorbidities including hypertension (Wang

& Scherer, 2008). A reciprocal connection between SNS activation and adiponectin is illustrated 16

both by the reduced serum adiponectin levels following SNS activation using cold stimulus (Imai et al, 2006), and the observations that iv injections of adiponectin dose dependently decreases BP and sympathetic nerve activity (Tanida et al, 2007). The CNS plays a vital role in this sympathetic regulation as electrolytic lesions of the dorsal hypothalamus prevented the decrease in renal sympathetic nerve inhibition (Tanida et al, 2007).

Recent evidence from our laboratory suggests that adiponectin controls the excitability of area postrema (AP) neurons through actions at both receptors inducing a hypertensive effect in urethane anesthetized (Fry et al, 2006). The AP, a circumventricular organ (CVO) situated in the medulla, is a well known homeostatic integration center that is involved in cardiovascular regulation (Ferguson, 1991; Joy & Lowe, 1970). It is part of a diverse regulatory network interacting with other homeostatic nuclei through its neuronal projections to the nucleus of the solitary tract (NTS), parabrachial nucleus (PBN) and the PVN (Price et al, 2008a). These studies clearly indicate important CNS actions for adiponectin, and point to the need for a clearer understanding of the mechanisms by which this adipokine controls specific networks and cellular subtypes in the brain to exert control over integrated autonomic function.

THE PARAVENTRICULAR NUCLEUS OF THE HYPOTHALAMUS

The PVN is a critical nuclei involved in the control of cardiovascular regulation, the immune system, feeding and metabolic function, fluid/electrolyte homeostasis and stress adaptation (Ferguson & Washburn, 1998; Swanson & Sawchenko, 1980). It is a bilateral nucleus positioned around the dorsal section of the 3rd ventricle and is a major integration site for hypothalamic output as it receives a dense array of afferents from a variety of nuclei within the brain. Structurally, the PVN is complex, and although it occupies only a third of a mm3 in the rat, the PVN can be divided into 8 different subdivisions and contains approximately 30 or more 17

different neurotransmitters and in cell bodies or pre-synaptic terminals (Swanson

& Sawchenko, 1983). The PVN is composed of magnocellular (MNC) and parvocellular (pre- autonomic (PA) and neuroendocrine (NE)) neurons, which when grouped together form the 8 subdivisions (3 MNC and 5 parvocellular regions) (Swanson & Sawchenko, 1983). Recent evidence suggests there are glutamatergic and GABAergic interneurons dispersed throughout the nucleus (Daftary et al, 1998; Csaki et al, 2000; Follwell & Ferguson, 2002a; Meister et al, 1988;

Roland & Sawchenko, 1993), while the nucleus is also surrounded by a predominantly

GABAergic „halo-zone‟ feeding synaptic input to the nucleus (Swanson & Sawchenko, 1983).

Cell counting studies performed on either Nissl- or Gomori-stained preparations of PVN estimate there to be between 10000 and 12000 cells within the boundaries of the nucleus (Swanson &

Sawchenko, 1983).

Neural inputs to the PVN

PVN is one of the major integration nuclei in the hypothalamus and therefore afferent projections to it are predictably numerous and diverse. Autoradiographic anterograde and retrograde tracer studies have detailed a structural map of information transfer to the PVN that reveals that the nucleus receives substantial neuronal input from three major areas of the brain: the hypothalamus, extra-hypothalamic sites and the brainstem (Sawchenko & Swanson, 1983;

Saper et al, 1976). Within the hypothalamus, nuclei in close proximity such as the ventromedial medial nucleus (Saper et al, 1976), the preoptic, anterior and lateral hypothalamic areas (Saper et al, 1978; Saper et al, 1979), the suprachiasmatic nucleus (SCN) (Swanson & Cowan, 1975) and the ARC (Sawchenko et al, 1982), all send dense projections to synapse with neurons in the PVN.

Additionally, two of the three sensory CVO‟s send projections to the PVN including the organum vasculosum of the lateral terminalis (OVLT) (Camacho & Phillips, 1981), residing in the 18

hypothalamus and the SFO (Ferguson et al, 1984b; Ferguson et al, 1984a), an extra-hypothalamic midline CVO positioned below the hippocampal commissure and descending into the lateral ventricle (Fry et al, 2007).

Morphological evidence demonstrates an extensive plexus of catecholaminergic fibers and their synaptic vesicles containing dopamine (DA) (Buijs et al, 1984; Decavel et al, 1987), noradrenaline (NA) (Swanson et al, 1981) or (Palkovits et al, 1980a) are present in the

PVN. These neurons originate from brainstem nuclei housing the A1 (and C1), A2, and A6 catecholaminergic cell groups, i.e. the ventrolateral medulla (Palkovits et al, 1980b), the NTS and the coeruleus (Swanson & Hartman, 1980; Swanson & Sawchenko, 1983). Horseradish peroxidase studies suggest that the lateral PBN also sends afferent fibers to innervate the PVN

(Tribollet & Dreifuss, 1981). Neuronal groups in higher forebrain structures also send projections to the PVN, including parts of the limbic system (Swanson & Sawchenko, 1983; Ulrich-Lai &

Herman, 2009), and may play an important role in modulating autonomic stress response in reaction to emotional triggers (Swanson & Sawchenko, 1983; Saper et al, 1976).

Projections of the PVN

Anatomical studies performed on hypophysectomized animals examining retrograde cell loss and retrograde tracer studies have provided an extensive literature describing the efferent projection patterns of PVN neurons, outputs which can be divided into four main areas. The is the primary target for MNC (and a small number of NE) neurons which in conjunction with the supraoptic nucleus (SON) makes up the neurohypophysial system. The intermediolateral cell column of the spinal cord receives axons from PA neurons, which also form reciprocal connections with medullary nuclei (NTS and LPN). Finally, NE neurons send axonal

19

projections to the median eminence to release neurotransmitters into the portal circulation

(Swanson & Sawchenko, 1983).

Neuronal Subtypes in the PVN: Morphological and Electrical Characteristics

The PVN houses three specific types of neurons; type 1 MNC, type 2 PA and type 3 NE which all play specific, extremely important roles in mediating the nucleus‟ function in pituitary hormone secretion, modulation of autonomic function and controlling energy homeostasis. For this reason, each cell type has been the subject of intense investigation as researchers have attempted to delineate unique characteristics that assist in identifying and studying them individually. The following section highlights this breadth of knowledge and details the criteria by which investigators use to classify these cells.

Type 1: Magnocellular Neurons

Of the eight subdivisions of the PVN, three house MNC (or type 1) neurons.

Morphologically, MNC neurons are relatively large (soma diameter 20 – 30 µM) bipolar or multipolar cells with usually one or two large dendrites oriented medially or ventromedially contained within the boundaries of the nucleus. They send axonal projections laterally above and below the fornix to the neurohypophysis where their synaptic terminals release peptides contained in large dense-core vesicles (Swanson & Sawchenko, 1983). MNC neurons are also electrotonically coupled to one another through gap junctions (Andrew et al, 1981) which assist in coordinating action potential pattern generation, an organizational feature which has been suggested to results in efficient and high concentrations peptidergic release into the posterior pituitary (Swanson & Sawchenko, 1983).

20

MNC neurons of the PVN have a distinct electrical phenotype that allows one to distinguish them from surrounding cells when performing electrophysiological experiments. They display a linear current – voltage relationship along with a prominent transient outward rectification carried by the fast-inactivating A-type K+ current (Tasker & Dudek, 1991; Luther &

Tasker, 2000). The manifestation of this current, in current clamp recordings is a prominent delay to first spike following a hyperpolarization. These cells generate spike trains and display little frequency adaptation (Tasker & Dudek, 1991). MNC neurons are the largest cells in the PVN, indicated by cell sizing studies following perikarya staining (Swanson & Sawchenko, 1983) and by whole cell capacitance measurements (Tasker & Dudek, 1991), an direct measurement of membrane surface area based on the ability of the membrane to separate electrical charge.

MNC neurons of the PVN predominantly synthesize and secrete vasopressin (VP) and oxytocin (OT), in addition to much smaller concentrations of CRH and enkephalin (ENK)

(Vandesande & Dierickx, 1975; Swaab et al, 1975b; Sawchenko et al, 1982). OT and VP are synthesized from a common precursor molecule and are packaged individually along with their respective neurophysins, peptide fragments from enzymatic cleavage of the precursor proteins, into large dense core vesicles. These two peptides are highly homologous in structure and sequence, differentiating by only 2 amino acids, and play a number of important roles in autonomic function and energy homeostasis as outline below.

Type II: Pre-autonomic Parvocellular neurons

Apart from being smaller in size, morphologically PA cells are very similar to MNC neurons. Histological studies indicate that PA neurons typically have 2-3 major primary dendrites and 6-10 secondary branches (Swanson & Sawchenko, 1983; Stern, 2001). Retrograde tracing studies using 1, 1‟-dioctadecyl-3,3,3‟,3‟-tetramethylindocarbocyanine and HRP injected into the 21

NTS, rostrol ventrolateral medulla (RVLM), dorsal motor nucleus of the vagus (DMX) and intermediolateral column of the spinal cord demonstrate axonal projection patterns of PA neurons

(Armstrong et al, 1980; Stern, 2001).

Detailed electrophysiological analysis indicates that these cells can be distinguished from others in the PVN by the presence of a low-threshold T-type Ca2+ current which activates at ~

-60 mV and is sensitive to nickel block (Luther & Tasker, 2000; Stern, 2001). A majority of PA neurons display strong inward rectification upon hyperpolarizing current pulses which manifests as a delayed „sag‟ indicative of hyperpolarization-gated cation non-selective (HCN) channel activation (Stern, 2001). PA neurons also display three different types of spontaneous action potential firing; tonic regular, tonic irregular and bursting in which spike broadness becomes progressively larger in action potentials towards the end of the burst (Bains & Ferguson, 1997;

Stern, 2001) suggesting an adaptive component to increase peptide release in target sites. PA neurons also synthesize and secrete CRH, TRH, OT and VP that participate as neuromodulators at their respective target nuclei in the hindbrain and spinal column (Swanson & Sawchenko,

1980).

Type III: Parvocellular Neuroendocrine

Parvocellular NE cells can be distinguished from other PVN neurons by their topographical organization and „lack‟ of distinct electrical characteristics possessed by the other cell subtypes. Studies employing fluoro-gold, a retrograde tracer, intravenously injected into rats suggests that neurosecretory NE cells are positioned medially within the PVN, closest to the boarder of the 3rd ventricle. Electrophysiological recordings made from these fluoro-gold positive cells compared to putative non-neurosecretory PA neurons indicates that NE cells fail to generate a low-threshold spike, seen in PA cells, and display a much smaller T-type Ca2+ current (Luther et 22

al, 2002). In addition, NE cells do not show a delay to first spike, the strong outward rectification current seen in MNC neurons indicative of a robustly activated IA, following a hyperpolarizing pulse. NE neurons send axonal projections to the external lamina of the median eminence and through the synthesis and secretion of CRH, TRH, VP, OT, , , and dopamine among others, regulate adenohypophyseal hormone release from the

(Swanson & Sawchenko, 1983; Vandesande et al, 1977; Sawchenko et al, 1982; Swanson &

Sawchenko, 1980). A detailed Golgi impregnation study suggests that some parvocellular (both

PA and NE) neuron axons also ramify and synapse locally within the PVN (van den Pol, 1982) suggesting a coordinating function for these neurons.

Neurochemical cell phenotypes in the PVN

Oxytocin

OT is a nine amino acid peptide synthesized in neurons distributed throughout the CNS with the highest densities seen in the PVN and supraoptic nucleus (Gimpl & Fahrenholz, 2001). It is translated from a prohormone containing its associated neurophysin, which is transcribed from mRNA encoded in chromosome 20 (Rao et al, 1992). The prohormone is enzymatically cleaved in secretory granules (Gimpl & Fahrenholz, 2001) and released at target sites to bind its G-protein coupled receptor and confer its biological activity (Kimura et al, 1992). OT positive neurons are found in both MNC and parvocellular regions of the PVN where they are uniquely involved in different aspects of endocrine, autonomic and energy homeostasis mechanisms (Swanson &

Sawchenko, 1983).

Pioneering studies examining the physiological effects of OT demonstrated a critical role for the peptide in inducing uterine contractions during birth and the milk-ejection reflex during

23

lactation (Gimpl & Fahrenholz, 2001). OT is also released from MNC neurons in response to changes in blood volume or osmolarity (Gimpl & Fahrenholz, 2001). OT acts to regulate body fluids through the reduction of tubular Na+ reabsorption and increases in glomerular filtration rates in the (Conrad et al, 1993). OT decreases mean arterial pressure through actions to reduce heart rate and atrial contraction force (Gimpl & Fahrenholz, 2001).

Recent evidence suggests that OT may play important roles in energy homeostasis through its actions on consumptive behavior. When injected either ip or icv, OT acts to inhibit both feeding and drinking in rats. In the presence of OT, meal duration times are decreased and latency to first meal following starvation is increased (Arletti et al, 1989; Arletti et al, 1990).

Parvocellular PA neurons expressing OT innervate the NTS, a site of OT receptor expression

(Martins et al, 2005) and area involved in the regulation of feeding. This pathway has been shown to play a critical role in regulating meal size (Blevins et al, 2003; Morton et al, 2005) which further implicates OT as a major regulator of energy homeostasis and a therefore a target of signals involved in mediating this function. Additionally, CSF OT levels are decreased in patients with (Demitrack et al, 1990), a metabolic state which also demonstrates incredibly high concentrations of serum adiponectin (Misra et al, 2007).

Vasopressin

VP (also known as arginine vasopressin or anti-diuretic hormone) is a nine amino acid peptide hormone synthesized in a number of different locations throughout the brain with the highest expression in the SON and the PVN (Swanson & Sawchenko, 1983). The VP gene is located adjacent to the OT gene and is transcribed into mRNA for the precursor protein in the opposite direction of OT (Gimpl & Fahrenholz, 2001). Immunohistochemical staining demonstrates that VP, along with its associated neurophysin, is present in both parvocellular and 24

MNC regions of the PVN (Swanson & Sawchenko, 1983). These studies also provide evidence that VP synthesizing MNC neurons exist as a distinct population of cells from those of OT synthesizing MNC neurons (Vandesande & Dierickx, 1975), a phenomenon which similarly holds true in the SON (Swaab et al, 1975a). Recent analysis on mRNA obtained from single

MNC neurons suggests however that single SON MNC neurons express both VP and OT mRNA and may produce both peptides (Xi et al, 1999). MNC derived VP is packaged into secretory granules and transported down axons to the pars nervosa of the posterior pituitary where it is stored and released upon appropriate stimuli.

MNC VP plays a vital role in regulating body fluid homeostasis through its actions in the kidneys. Hyperosmotic or hypervolemic stimuli induce VP release from the posterior pituitary which acts on V2 receptors expressed in the distal convoluted tubules and apical surface of collecting ducts of the kidney, increasing the expression of aquaporin-2 and thus the permeability to water (miry-Moghaddam & Ottersen, 2003). The VP system acts as a defense against disruptions in body fluid homeostasis and therefore plays an important role in regulating energy homeostasis and metabolic function (Baylis, 1987). VP also acts as a vasoconstrictor (Baylis,

1987) and demonstrates an unusual connection to social behavior in both prairie voles and humans (Walum et al, 2008). A recent examination of genetic variability in the 1a gene reveals that proper functioning of this receptor plays a role in mediating the

„health‟ of monogamous relationships in both voles and human males (Walum et al, 2008) and therefore suggests VP plays a wider role in the context of human physiology and social interaction.

25

Thyrotropin releasing hormone

TRH is a tropic tri-peptide, synthesized from a larger precursor protein within cells of the

CNS (Lechan & Fekete, 2006). TRH is widely distributed throughout the CNS in areas such as the hippocampus, brainstem and in the hypothalamus where it is highly expressed in the parvocellular regions of the PVN (Lechan & Segerson, 1989). NE neurons that express TRH are known as „hypophysiotrophic TRH neurons‟ and release TRH at median eminence into the perivascular space of the pituitary portal system, which carries TRH through the superior hypophyseal to reach its target in the anterior pituitary (Lechan et al, 1983; Lechan et al,

1980). TRH acts by binding its receptor located in thyrotrope and lactotrope membranes in the anterior pituitary to increase the synthesis and release of thyroid-stimulating hormone (TSH) and respectively (Lechan & Fekete, 2006). TRH confers its effects on metabolism primarily through its ability to increase in release of TSH, thus playing a critical role in controlling endocrine function of the thyroid gland.

TRH has been shown to increase thermogenesis, locomotor activity and regulate the activity of autonomic function (Lechan & Fekete, 2006). Central injections of TRH are associated with reduced feeding in normal, fasted and stressed animals (Lechan & Fekete, 2006) indicating this peptide plays important roles as an anorexigenic modulator of food intake and energy homeostasis. Both NE and PA TRH neurons receive monosynaptic connections from α-

MSH/CART and NPY/AgRP neurons of the ARC, cells that have been shown to be involved in mediating the effects of leptin on food intake (Schwartz et al, 2000). Glutamatergic synapses innervate TRH neurons further supporting the integrative nature of this peptide in the involvement of energy homeostatic mechanisms (Wittmann et al, 2005).

26

Corticotropin releasing hormone

CRH is a 41 amino acid neuropeptide produced from a 187 amino acid precursor protein translated from mRNA encoded on chromosome 8 in the human (Vale et al, 1981; Thompson et al, 1987; Arbiser et al, 1988). CRH peptide and mRNA are widely expressed throughout the

CNS, with immunohistochemical and radioimmunoassay studies showing the highest density of positively stained perikarya in the parvocellular region of the PVN (Bloom et al, 1982). CRH released from these neurons into the median eminence acts on G-protein coupled CRH receptors present on anterior pituitary corticotropes to stimulate the calmodulin-kinase dependent release of

POMC derived peptides, namely adrenocorticotropin hormone (ACTH) and β-endorphin (Owens

& Nemeroff, 1991; Wynn et al, 1983; Brown et al, 1985; Hashimoto et al, 1985). Subgroups of

CRH positive parvocellular PVN neurons co-localize with VP, whose expression can be up- regulated following adrenalectomy, while others with OT (Sawchenko et al, 1984), each of which act to potentiate CRH‟s effects on ACTH secretion from anterior pituitary corticotrophs (Owens

& Nemeroff, 1991). ACTH acts in the cortex of the to stimulate the release of in response to the neuropeptide, making CRH the primary hypothalamic factor which orchestrates the CNS derived endocrine response to stress. In addition to endocrine modulation,

CRH also appears to play important roles in the regulation of autonomic and immune function

(SNS activation), as well as having profound effects on behavior (Owens & Nemeroff, 1991).

CRH is an important regulator of autonomic function and metabolic homeostasis in the body. In addition to its mediation of the stress response, CRH also acts to decrease luteinizing and release from the anterior pituitary (Rivier & Vale, 1984b; Rivier & Vale, 1984a) as well as increase SNS activity (Brown et al, 1982). Enhanced sympathetic out-flow results in increased O2 consumption, locomotion, plasma glucose, and concentrations in addition

27

to enhanced thermogenesis via increases in SNS activity innervating brown adipose tissue

(Owens & Nemeroff, 1991; Brown et al, 1982). The physiological manifestation of these phenotypes is an increase in energy metabolism and reduction in feeding, effects which can be mimicked by icv injections of CRH or microinjections of the peptide directly into the PVN

(Britton et al, 1982). These studies clearly demonstrate a vital role for CRH in the integrated CNS control of energy homeostasis.

THE NUCLEUS OF THE SOLITARY TRACT

The nucleus of the solitary tract, abbreviated NTS (nucleus tracti solitarii), is a medullary structure which plays a critical role in the integration of viscerosensory regulation including gustatory, respiratory, hepatic, renal, cardiovascular and metabolic processing among others

(Lawrence & Jarrott, 1996). The NTS is a single nucleus starting at the obex, extending down the rostral-caudal axis where it bifurcates to become a bilateral structure descending adjacent to the

4th ventricle (Lawrence & Jarrott, 1996; Saper, 2002). At the level of the AP, an adjacent sensory

CVO, and before the bifurcation, the nucleus can be divided into three subsections which contain neuronal cell bodies; the commissural (dorsal of the central canal), lateral and medial (ventral to

AP and the 4th ventricle) subnuclei (Lawrence & Jarrott, 1996).

Afferent projections to the NTS

The NTS is unique in the brain as it is the primary relay site for parasympathetic visceral sensory afferents. Sensory information relayed to the CNS through the four cranial nerves; trigeminal, facial, glossopharyngeal and the vagus, all terminate in a highly organized topographical pattern with the NTS (Altschuler et al, 1989). These nerves provide direct input from peripheral chemoreceptors, volume receptors, carotid and arterial baroreceptors, allowing it

28

to play a critical role in the integration of the baroreflex (Guyenet, 2006). The AP and the caudal ventrolateral medulla also send dense projections to the NTS, (Price et al, 2008a; Saper, 2002).

As mentioned above, the PVN is a major source of hypothalamic input to the NTS suggesting these two nuclei directly communicate and perhaps exhibit a reciprocal information processing ability between them (Swanson & Sawchenko, 1983).

Efferent projections of the NTS

Based on the diversity of its role in mediating autonomic function, it is no surprise that the NTS exhibits a dense and diverse array of efferent projections to other sites in the brain. As part of the baroreflex pathway, the NTS contains a defined group of glutamatergic neuronal projections to the CVLM to provide excitatory input to this nucleus (Guyenet, 2006). The PBN is another major target for efferent projections from the NTS (Lawrence & Jarrott, 1996). This connection comprises the beginning of the NTS-PBN-thalamic-cortical relay, a network crucial for the conscious recognition of autonomic sensory input through the brainstem to the forebrain

(Saper, 2002). The NTS also sends information to the forebrain through its projections to the bed nucleus of the stria terminalis (BNST) and the central nucleus of the amygdala. Additionally, the

NTS makes reciprocal connections with the AP and a host of hypothalamic nuclei including the

LH, the ARC and the PVN (Lawrence & Jarrott, 1996; Saper, 2002).

Electrophysiology of the NTS

Though the cytoarchitecture of the NTS is quite complex, studies examining differences in electrical currents have revealed three basic electrophysiological fingerprints of neuronal subgroups in the NTS that bear a striking similarity to the three subgroups in the PVN. These subgroups are assigned the nomenclature; delayed excitation (DE), post-inhibitory rebound (PIR)

29

and non (NON) cells (Vincent & Tell, 1997). Based on morphology and electrical current profiles, DE cells are similar to the MNC neurons of the PVN in that they are the largest cells in the NTS, and express a robust TEA insensitive transient outward current. This current is abolished in the presence of 4-aminopyridine suggesting that it is mediated by the A-type K+ channel (Vincent & Tell, 1997). This current is larger in older animals suggesting post-natal changes in membrane ion channel composition. In contrast, PIR cells display a robust rebound inward current following membrane hyperpolarization which usually elicits a burst (2-8) of action potentials (Vincent & Tell, 1997). This bursting activity is not age dependent but is sensitive to cobalt and cadmium (but not external cesium) suggesting it is mediated by the activation of a T- type Ca2+ channel. These cells are morphologically smaller than DE neurons and resemble PA-

LTS cells in the PVN. The NON cells are similar in shape and size to the PIR cells however they express neither of the noticeable current manifestations seen in the DE and PIR cells. NON cells are also in the minority of NTS neurons and are similar to the NE cells of the PVN (Vincent &

Tell, 1997). A detailed analysis of projection pattern linked to electrophysiological fingerprint expression has yet to be done; however, it should provide insight into the specific function of each of these subgroups of NTS neurons and their role in mediating autonomic function.

Neurochemical expression in the NTS

In addition to its diverse projection profile, the NTS contains a plethora of neurochemicals within its neurons. Along with GABA and glutamate, immunohistochemical analysis of the NTS reveals that individual neurons uniquely express different neuropeptides such as atrial natriuretic peptide (ANP), angiotensin II (ANGII), VP, OT, substance P, cholesystokinin

(CCK) and NPY among others (Lawrence & Jarrott, 1996). In addition, catecholaminergic cells in the NTS have been shown to expression adrenalin, noradrenalin and dopamine, arising from 30

the C2 (adrenalin) or A2 cell group (noradrenalin, dopamine), also shown to produce the enzymes tyrosine hydroxylase and dopa hydroxylase necessary for the biosynthesis of these chemicals

(Armstrong et al, 1982).

Neuropeptide Y

NPY is a 36 amino acid peptide which acts in the brain to potently stimulate feeding and reduce energy expenditure (Clark et al, 1984; Stanley & Leibowitz, 1984). NPY is one of the most widely expressed neuropeptides in the brain however it‟s high expression in the ARC nucleus and its role in feeding induction has dominated interest in the central actions of this peptide (Gao & Horvath, 2007). NPY is expressed in cytoarchitecturally distinct cell groups within the NTS. Immunoreactivity for NPY in neuronal bodies is located in NTS subnuclei adjacent to the AP and co-localized in some cases with adrenaline. NPY immunoreactivity exists in known cardiovascular regulatory locations within the NTS as these are sites of synaptic termination of efferents from the aortic and carotid sinuses (Harfstrand et al, 1987) suggesting a role for NPY in the hindbrain regulation of cardiovascular activity.

Microinjection of NPY into the NTS induces hypotension and facilitates the aortic baroreflex (Kubo & Kihara, 1990). This effect appears to be mediated by the Y2 receptor as only a Y2 selective truncated form of NPY (13-36) could elicit the same hypotensive effects, which could not be blocked by a Y1 selective antagonist (Barraco et al, 1991). Further support for a role for NPY in cardiovascular regulation comes from the observation that in streptozotocin induced diabetic rats (which develop hypertension), intra-NTS microinjections of NPY have a far less significant hypotensive effect then wild type rats (Dunbar et al, 1992). NPY also controls peripheral insulin secretion through actions at NTS neurons (Dunbar et al, 1992) and has been

31

suggested that the pathogenesis of hypertension in diabetic patients may be due to a decrease of intra-NTS efficacy of NPY signaling (Lawrence & Jarrott, 1996).

Adiponectin‟s role in mediating peripheral control of glucose homeostasis, insulin sensitivity and the oxidation of fatty acids is well established, yet its function in the CNS is relatively unknown at the current time. When injected into the lateral ventricles of the brain, this adipokine displays a profound ability to modulate metabolic function and energy expenditure through unknown actions at central autonomic nuclei (Qi et al, 2004). Investigation into the specific control of PVN and NTS neuronal function and the mechanisms by which this occurs is a prudent scientific investment. The plethora of neuropeptides/neurotransmitters in these nuclei including OT, VP, TRH, CRH and NPY, and the role each plays in mediating the stress-axis and central control of energy expenditure, establishes these sites as important targets for central adiponectin actions. This thesis examines results obtained from experiments designed to investigate the central actions of adiponectin and discuss the data further in a global context of the

CNS peptide system.

32

STATEMENT OF PROBLEM

This thesis is comprised of four manuscripts which examine the actions of adiponectin in two autonomic nuclei within the CNS; the PVN and the NTS. Electrophysiological, molecular and in vivo microinjection studies are employed to investigate the adipokines role in mediating energy expenditure, autonomic function including blood pressure regulation and stress-axis hormone concentration regulation. Furthermore, techniques involving post hoc molecular identification of neurons following electrophysiological recordings using both the current and voltage clamp recording configuration are used and are the first to establish central neurons as sites of adiponectin receptor mRNA production and provide data revealing the electrical and physiological consequences of receptor activation.

1. Chapter 2 (manuscript 1) examines the presence of adiponectin receptor mRNA in the

PVN and the effect of activating those receptors expressed in MNC neurons that

synthesize OT, VP or both OT and VP mRNA.

2. Chapter 3 (manuscript 2) focuses on investigating adiponectin actions on parvocellular

PVN neurons involved in the stress axis (CRH) or control of autonomic function and

feeding behavior (TRH, OT). An inquiry into the effect of icv injections of adiponectin

on plasma ACTH concentrations is also made.

3. Chapter 4 (manuscript 3) examines the effect of adiponectin on blood pressure

regulation through actions on identified medial NTS neurons that express NPY.

4. Chapter 5 (manuscript 4) investigates the cellular ionic mechanisms which adiponectin

modulates to confer its effects on excitability in parvocellular PVN neurons.

33

Chapter 2: ADIPONECTIN SELECTIVELY INHIBITS OXYTOCIN NEURONS OF THE

PARAVENCTRICULAR NUCLEUS OF THE HYPOTHALAMUS

Ted D. Hoyda1, Mark Fry2, Rexford S. Ahima3 and Alastair V. Ferguson1

1Department of Physiology, Queen‟s University, ON, Canada, K7L 3N6. 2Department of

Biological Sciences, University of Manitoba, Winnipeg, MB, Canada, R3T 2N2. 3University of

Pennsylvania School of , Division of Endocrinology, Diabetes and Metabolism,

Philadelphia, PA, USA, 19104.

Journal of Physiology. 2007, Vol. 585(3):805-816

34

ABSTRACT

Adiponectin is an adipocyte derived hormone that acts in the brain to modulate energy homeostasis and autonomic function. The paraventricular nucleus of the hypothalamus (PVN), which plays a key role in controlling pituitary hormone secretion, has been suggested to be a central target for adiponectin actions. A number of hormones produced by PVN neurons have been implicated in the regulation of energy homeostasis including oxytocin, corticotropin releasing hormone and thyrotropin releasing hormone. In the present study we investigated the role of adiponectin in controlling the excitability of magnocellular (MNC – oxytocin or vasopressin secreting) neurons within the PVN. Using RT-PCR techniques we have shown expression of both adiponectin receptors in the PVN. Patch clamp recordings from MNC neurons in hypothalamic slices have also identified mixed (27% hyperpolarization, 42% depolarization) effects of adiponectin in modulating the excitability of the majority of MNC neurons tested.

These effects are maintained when cells are placed in synaptic isolation using tetrodotoxin.

Additionally we combined electrophysiological recordings with single cell RT-PCR to examine the actions of adiponectin on MNC neurons, which expressed oxytocin only, vasopressin only, or both oxytocin and vasopressin mRNA and assess the profile of receptor expression in these subgroups. Adiponectin was found to hyperpolarize 100% of oxytocin neurons tested (n=6).

Vasopressin cells, while all affected by adiponectin (n=6), showed mixed responses. Further analysis indicates that oxytocin neurons express both receptors (6/7), while vasopressin neurons express both receptors (3/8), or one receptor (5/8). In contrast 6/6 oxytocin/vasopressin neurons were unaffected by adiponectin. Co-expressing oxytocin and vasopressin neurons express neither receptor (4/6). The results presented in this study suggest that adiponectin plays specific roles in

35

controlling the excitability of oxytocin secreting neurons, actions which correlate with the current literature showing increased oxytocin secretion in the obese population.

36

INTRODUCTION

Adipose tissue plays an active role in modulating energy homeostasis through secreting factors which act at both peripheral and central targets (Ahima, 2005). Adiponectin is an adipose tissue-derived hormone with anti-diabetic, insulin-sensitizing and energy regulatory properties

(Fruebis et al, 2001; Yamauchi et al, 2001; Berg et al, 2001). Intriguingly, and in contrast to leptin, the cellular expression of adiponectin mRNA and serum concentrations are inversely correlated with adiposity (Hu et al, 1996; Yamauchi et al, 2001; Arita et al, 1999). Two receptors

(AdipoR1 and AdipoR2) have recently been characterized and localized to tissues throughout the body including the central nervous system (Yamauchi et al, 2003). When administered centrally, adiponectin stimulates thermogenesis, promotes oxygen consumption and significantly decreases body weight (Qi et al, 2004). Furthermore, intracerebroventricular (ICV) administration of adiponectin increases the numbers of c-fos positive neurons in the paraventricular nucleus of the hypothalamus (PVN) (Qi et al, 2004) suggesting that adiponectin may act in this nucleus to activate subpopulations of PVN neurons.

The PVN is a bilateral hypothalamic nucleus located on either side of the third ventricle and is recognized as a pivotal autonomic control center, which plays essential integrative roles in the control of fluid and electrolyte balance, cardiovascular regulation, the immune response, energy metabolism and stress responses (Ferguson & Washburn, 1998; Swanson & Sawchenko,

1980). Within this nucleus a subpopulation of neuroendocrine cells exists that control the release of vasopressin and oxytocin through axonal projections to the posterior pituitary. These magnocellular (MNC) neurons are located mainly in the lateral region of the nucleus and are responsible for modulating cardiovascular function, fluid balance and the initiation of the milk ejection reflex (Brody, 1988; Gimpl & Fahrenholz, 2001). We have previously reported that

37

magnocellular neurons depolarize in response to leptin administration supporting the hypothesis that these neurons are sensitive to adiposity hormones (Powis et al, 1997) and therefore, the circulating concentrations of neurohypophysial peptides may be regulated by adipokines.

Intriguingly, oxytocin neurons in the PVN have been suggested to play a vital role in coordinating feeding cessation (Olson et al, 1991b; Olson et al, 1991a; Verbalis et al, 1993; Blevins et al,

2003) by acting as targets for factors which induce anorexigenic behavior such as CCK (Olson et al, 1992) and peptide histidine isoleucine (PHI) (Olszewski et al, 2003). The concentrations of oxytocin in brain and periphery have also been shown to be compromised during pathological states of energy homeostasis such as obesity (elevated) (Stock et al, 1989) and restrictive anorexia (reduced) (Demitrack et al, 1990).

The present study was therefore designed to examine the role of adiponectin in controlling the excitability of MNC neurons of the PVN.

MATERIALS AND METHODS

Preparation of hypothalamic slices

All experiments performed in this study used 50-100g male Sprague Dawley rats

(Charles River, PQ), maintained on a 12/12 light/dark cycle and provided with food and water ad libitum. Animals were decapitated and removed from the skull and placed in ice-cold slicing solution containing (in mM) 87 NaCl, 2.5 KCl, 25 NaHCO3, 0.5 CaCl2, 7 MgCl2, 1.25

NaH2PO4, 25 glucose, 75 sucrose, and bubbled with 95% O2/5% CO2 for 1-2 minutes. Brains were then trimmed, mounted on a stage and 300µm coronal sections of the hypothalamus were cut using a vibratome (Leica, Nussloch, Germany) while bathed in ice-cold slicing solution. All

38

procedures were approved by the Queen‟s University Animal Care Committee, and were in accordance with the guidelines of the Canadian Council for Animal Care.

PVN RNA extraction and reverse transcription polymerase chain reaction (RT-PCR)

RNA extraction and cDNA synthesis: The PVN was microdissected from 300µm coronal hypothalamic slices. Total RNA was extracted with TRIzol as per manufacture‟s instructions

(Invitrogen, Burlington, Ontario, Canada) which was then subjected to DNase treatment by adding RNA to a reaction mixture containing 1µl 10X buffer with MgCl2, 7µl diethylpyrocarbonate (DEPC) treated-H20 and 1µl deoxyribonuclease I (Sigma, St. Louis, MO).

This solution was then incubated at 37ºC for 30 minutes, 1µl 25mM EDTA was added and incubated at 65ºC for 10 minutes. Oligo-dT primed cDNA was then synthesized using a

RETROscript cDNA synthesis kit (Ambion, Austin, TX) as per manufacture‟s instructions in a final reaction volume of 20µl.

PCR reactions: 1µl of PVN cDNA was added to a 50µl reaction mixture containing 1X

® PCR buffer, 0.2 mM of each dNTP, 1.5 mM MgCl2, 1 U of Platinum Taq DNA Polymerase, 0.2

µM of each primer and H2O to a final volume of 50µl. Each reaction was denatured at 95ºC for 2-

4 minutes then cycled 35 times through a temperature protocol consisting of 94ºC for 60 seconds,

55ºC for 60 seconds, 72ºC for 60 seconds, and a final extension at 72ºC for 5 minutes. Two separate sets of primers were used to detect AdipoR1 and AdipoR2 mRNA along with a primer set for synaptotagmin (see Table 2.1) which acted as a positive control. PCR products were separated and visualized through gel electrophoresis on a 2% agarose gel containing ethidium bromide and sequenced to confirm identity (Robarts Institute, London, Ontario, Canada).

39

Gene Primer Name Position Sequence Product Size (bp) Reference

Receptor Expression

Adiponectin Receptor 1 AdipoR1 (first set) F ctggtcccatctttctagg 210 This study R tgatagcaggtttctttttaaagc

AdipoR1 (second set) F tcttcctcatggctgtgatg 270 Kharroubi et al., 2003 R agccacttgggaagttcctcc

Adiponectin Receptor 2 AdipoR2 (first set) F tctggttcaacatagcacaaac 367 This study R aaccccaaccttctatgattc

AdipoR2 (second set) F ggagccattctctgcctttc 467 Kharroubi et al., 2003 R accagatgtcacatttgcca

Synaptotagmin Synaptotagmin F aggggctttcctatctaaggg 202 Richardson et al., 2000 R gttggcagtgttgcaagaga

Single Cell RT-PCR

Glyceraldehyde-3- (GAPDH) F (outside) gatggtgaaggtcggtgtg 469 This study phosphate R gggctaagcagttggtggt dehydrogenase F (nested) taccaggctgccttctct 360 R ctcgtggttcacacccatc

γ-amino butyric acid Glutamic acid F (outside) cacaaactcagcggcataga 550 This study decarboxylase-67 R gagatgaccatccggaagaa (GAD67) F (nested) cacaaactcagcggcataga 149 R ctggaagaggtagcctgcac

Tyrosine Hydroxylase TH F (outside) cacctggagtattttgtgcg 1138 This study R cctgtgggtggtaccctatg F (nested) tcgacccagtatatccgcca 376 R tcggacacaaagtacacagg

Vasopressin VP F (outside) cctcacctctgcctgctactt 237 This study R agccagctgtaccagcctaa F (nested) acctctgcctgctacttcca 216 R agccagctgtaccagcctaa

Oxytocin OT F (outside) ctgccccagtctcgcttg 281 This study R cctccgcttccgcaaggcttctggc F (nested) ctgccccagtctcgcttg 244 R gcgagggcaggtagttctcc

Adiponectin receptor 1 AdipoR1 F (outside) gtcccctggctctattactcct 509 This study R agcacttggctgtgatgt F (nested) tcttcctcatggctgtgatgt 223 R ggctcagagaagggagtcatc

Adiponectin receptor 2 AdipoR2 F (outside) ggagccattctctgcctttc 464 This study R ccagatgtcacatttgcca F (nested) actgtaacccacaaccttgcttc 191 R tcaggaacccttctgagatgac

Table 2-1. Primer sets used for RT-PCR analysis of receptor expression and single cell identification.

AdipoR1/R2 (second set) primer set adapted from Kharroubi et al. (Kharroubi et al, 2003), Synaptotagmin primer set adapted from Richardson et al. 2000 (Richardson et al, 2000).

40

Localization of adiponectin receptors by in situ hybridization

Male Sprague Dawley rats were anesthetized with sodium pentobarbital and perfused transcardially with phosphate buffered saline (PBS) prepared with DEPC-treated water, followed by 10% neutral buffered formalin. Brains were removed, immersed in the same fixative overnight, followed by cryoprotection in 20% sucrose/PBS/DEPC at 4°C. Five series of 20 µm coronal sections were cut on a Reichert cryotome, mounted on Superfrost Plus glass slides (Fisher

Scientific, Houston, Texas), air dried, and stored in desiccated boxes at –20°C. Before hybridization, sections were fixed in 4% formaldehyde in DEPC-PBS, pH 7.0, for 20 min at 4°C, dehydrated in increasing concentrations of ethanol, cleared in xylene for 15 min, rehydrated in decreasing concentrations of ethanol, placed in prewarmed sodium citrate buffer (95–100°C, pH

6.0), and then dehydrated in ethanol and air dried. The slides were hybridized with 35S-cRNA probes against AdipoR1 and R2, washed, dehydrated, air dried and exposed to film as previously described (Fry et al, 2006). After dipping in photographic emulsion, the slides were developed, counterstained with thionin, dehydrated in graded ethanols, cleared in xylene, and coverslipped with Permount (Fisher Scientific, Houston, Texas). Brain sections were examined with a Nikon

E600 microscope, and dark-field photomicrographs of the PVN were taken with a SPOT RT digital camera (Phase 3 Imaging Systems, Glen Mills, PA).

Electrophysiology

Hypothalamic slices including the PVN were prepared as described above and were then incubated at 31.5ºC for at minimum 60 minutes prior to recording in artificial cerebral spinal fluid

(aCSF) containing (in mM) 126 NaCl, 2.5 KCl, 26 NaHCO3, 2 CaCl2, 2 MgCl2, 1.25 NaH2PO4,

10 glucose, saturated with 95% O2/ 5% CO2. PVN neurons were visualized through a water immersion 40X objective mounted to an IR-DIC Nikon E600FN microscope (Tokyo, Japan).

41

Borosilicate glass pipettes were pulled on a flaming micropipette puller (Sutter instruments,

USA) to a resistance of 3-6 MΩ when filled with internal recording solution consisting of (in

+ 2+ mM): 140 K -gluconate, 2 MgCl2, 0.1 CaCl2 (calculated free intracellular Ca : 15nM), 1.1

EGTA, 10 HEPES, 2 NaATP adjusted to a pH of 7.3 with KOH). Electrophysiological recordings were made in the whole cell current-clamp configuration from type 1 MNC neurons which were identified morphologically and according to distinct electrophysiological characteristics (Tasker

& Dudek, 1991). Current clamp recordings were made with an EPC-7 amplifier (Heka

Instruments, Germany) sampling at 10 kHz, filtered at 3 kHz, digitalized through a CED micro1401 (CED, Cambridge, England) and displayed using Spike2 (CED, Cambridge, England) data acquisition software to be stored for off-line analysis. Adiponectin (10nM) was perfused onto slices at a rate of 1-2 ml/min, only after stable membrane potential was obtained for more then 100 s. Recordings were performed at 32.5°C and a junction-potential correction of 17.2mV, calculated using p-clamp (Molecular Devices, Union City, CA) was applied to the membrane potential of all recorded neurons by subtraction.

Single cell RT-PCR

Following recording, negative pressure was applied to the pipette in order to collect the cytoplasm and care was taken as the pipette was withdrawn from the recorded cell to pull an outside-out patch in order to seal the cytoplasm contents within the pipette. Following withdrawal from the bath the pipette tip was broken off and the cytoplasmic contents expelled into a siliconized centrifuge tube and stored at -80ºC. This withdrawn cytoplasm was then subjected to

DNase treatment (TURBO DNA-free - Ambion, Austin, TX) before undergoing cDNA synthesis using a random hexamer based Superscript™ III first strand synthesis kit (Invitrogen, Burlington,

Ontario, Canada).

42

Due to the small amount of total RNA harvested from a single cell, PCR was carried out in two separate steps: the first step was a multiplex reaction containing primers (outside) for all the of interest along with cDNA from the single cell. The second reaction was a nested

PCR reaction using a single set of primers (nested) for each gene of interest. The multiplex reaction contained 10µl 10X PCR buffer, 3µl MgCl2, 2µl dNTPs, 0.5µl Platinum® Taq, 1µl cDNA, 9µl H2O, 2µl of each outside primer set (2.5mM) of interest (see table 2.1) and ddH2O up to a final volume of 100µl. Each reaction was denatured at 94ºC for 4 min then cycled 35 times through a temperature protocol consisting of 1 min at 94ºC, 1.5 min at 55ºC, 3 min at 72ºC and finally for 5 min at 72º and held at 4ºC. The final product was diluted 1:1000 and used as a template for a second round PCR.

Second round PCR consists of a series of reactions for each of the genes of interest. Each reaction mixture contains 5µl 10X PCR Buffer, 1.5µl MgCl2, 1µl dNTPs, 0.4µl Platinum® Taq,

2µl cDNA template, 1µl of each nested primer (see table 2.1) and sdH20 up to a final volume of

50µl. The reaction mixture was cycled through a temperature protocol as indicated above following which the products were run out on a 2% agarose gel containing ethidium bromide and sequenced to confirm their identity (Robarts Institute, London, Ontario, Canada).

Statistical Analysis

Cells were classified as responders if within 1000 seconds after the application of adiponectin the membrane potential (mean measured over 100 second periods) changed by more than 2 standard deviations of the mean assessed in the 100 seconds prior to peptide application

(baseline membrane potential). Cells were only included in the responder group if membrane potential showed a return towards baseline membrane potential following return to aCSF perfusion of the slice. Kruskal-Wallis one-way ANOVA tests were used to compare the changes

43

in membrane potential in response to adiponectin between groups of cells. Wilcoxon signed rank tests were used to evaluate changes in spike frequency (paired) before and after adiponectin.

Peptides and Chemicals

Adiponectin (human recombinant globular) was obtained from Phoenix Pharmaceuticals,

Inc. (Belmont, CA) and prepared daily from aliquots stored at -80ºC. Adiponectin was made to a working solution of 10nM in aCSF. Tetrodotoxin (TTX) was obtained from Alomone laboratories

(Jerusalem, Israel) and prepared daily in aCSF to a working concentration of 1µM. All other chemicals used were obtained from Sigma (Scarborough, Ontario, Canada).

RESULTS

Adiponectin Receptor 1 and 2 are expressed in the PVN

Based on evidence of receptor expression in the CNS (Yamauchi et al, 2003) and the induction of c-fos immunoreactivity in the PVN upon ICV injections of adiponectin (Qi et al,

2004) we first examined if AdipoR1 and R2 were expressed in the PVN through the use of RT-

PCR and in situ hybridization. RT-PCR reactions were performed on cDNA derived from mRNA isolated from acutely microdissected PVN tissue with two primer sets designed for different regions on the AdipoR1 and R2 cDNA (Figure 2.1A). Synaptotagmin was used as a positive control along with an RT (-) in which the reverse transcriptase enzyme was omitted from the RT reaction. We amplified product for both receptors indicating the presence of mRNA for AdipoR1 and R2 in PVN. Following this we performed in situ hybridization with probes specific for both receptor mRNA again confirming the presence of AdipoR1 and R2 in both magnocellular and parvocellular regions of the PVN as illustrated in Figure 2.1B.

44

Figure 2-1. Adiponectin receptor mRNA is expressed in the PVN.

Ai) Agarose gel showing PCR analysis of PVN cDNA using two primer sets specific for different locations of each adiponectin receptor mRNA and a primer set specific for synaptotagmin. PCR products for synaptotagmin are not observed in the RT (-) lane in which reverse transcriptase has been omitted from the cDNA synthesis reaction. Aii) A Coronal section (300µm) illustrating the tissue sample taken for mRNA extraction and the RT-PCR reaction. SON = Supraoptic nucleus,

SCN = Suprachiasmatic nucleus, Fx = Fornix, PVN = Paraventricular nucleus. Bi) Dark-field photomicrographs of in situ hybridization for AdipoR1 and AdipoR2, demonstrating mRNA expression in the paraventricular nucleus of the hypothalamus. 3v, third ventricle. Scale bar, 50

µm. Bii) A nissl stained section of PVN with a schematic superimposed on it indicating relative anatomical positions of magnocellular and parvocellular neurons within the nucleus.

45

Adiponectin influences the excitability of MNC neurons in the PVN

Current clamp recordings were obtained from a total of 55 morphologically (based on size) and electrophysiologically (based on an outward rectification caused by activation of the dominant IA current) characterized MNC neurons (mean action potential amplitude of 96.7 ± 1.9 mV, mean resting membrane potential of -60.3 ± 0.9 mV and mean input resistance of 924 ± 56

MΩ). Following a minimum 100 second stable baseline recording period we examined the effects of bath application of 10 nM adiponectin for 200 seconds on both membrane potential and spike frequency. The membrane potential of MNC neurons responded to administration of adiponectin in one of three ways showing either depolarization (6.6 ± 0.5 mV (mean ± SEM), n=23), hyperpolarization (-6.6 ± 0.9 mV, n=15) or no effect (0.2 ± 0.5 mV, n=17) as illustrated in Figure

2.2 A, B. These observations suggest the existence of three distinct populations of MNC neurons with respect to adiponectin sensitivity. In addition the depolarizing effects of adiponectin on membrane potential suggest that 1 nM is the minimum effective concentration. Neurons examined in response to 1 nM adiponectin showed a mean change in membrane potential of 3.7 ±

0.9 mV (n=4) while those subjected to 100 pM showed a change in membrane potential of 1.9 ±

1.0 mV, (n=4) (figure 2.2 C). An EC50 value of 1.8 nM was estimated using: Y=Bottom + (Top-

Bottom)/(1+10^((LogEC50-X))), where X=Log[adiponectin]. Furthermore, the duration of the effect and onset to peak response differed between the groups of adiponectin responsive neurons.

The duration of the response in some cells lasted as long as 1500 seconds before seeing a return towards baseline in the membrane potential.

We also observed adiponectin induced changes in spike frequency with depolarization being associated with an increase in spike frequency (control: 0.18 ± 0.15 Hz, adiponectin: 1.69 ±

0.51 Hz). Conversely cells which hyperpolarized showed a decrease in firing rate (control: 0.62 ±

46

Figure 2-2. Adiponectin influences the membrane potential and spike frequency of MNC PVN neurons. A) Representative current clamp recordings of MNC neurons showing responses to 10nM adiponectin (black bars), i) depolarization, ii) hyperpolarization and iii) no response. Scale bars: 10 mV, 100 seconds. Downward deflections in each trace show the voltage response to hyperpolarizing current pulses for the measurement of input resistance. B, C, D) Graphs representing mean changes in membrane potential (B), concentration dependence of the depolarizing response (C) and spike frequency (D) for all cells examined in response to 10nM adiponectin. Mean changes in membrane potential were significantly different between responsive and non-responsive groups (Kruskal-Wallis test, ***p < 0.001). Concentration- response curve was developed for neurons which depolarized to ADP, numbers above data points represent the cells tested with each dose. EC50 was calculated at 1.8 nM. Mean changes in spike frequency were assessed to be significantly different from control in each of the responding groups (Wilcoxon signed rank test, *p < 0.05, **p < 0.005) (ADP = adiponectin). Cells which depolarized but did not reach threshold to elicit action potentials and cells which were quiescent before hyperpolarization were not included in the analysis of spike frequency.

47

0.31 Hz, adiponectin: 0.04 ± 0.03 Hz) while spike frequency was not changed significantly in cells which showed no change in membrane potential in response to adiponectin (Figure 2.2 D).

Effects of adiponectin on MNC neurons are maintained in TTX

We next examined if the effects of adiponectin on membrane potential were maintained when action potential induced release of neurotransmitter was inhibited by perfusing slices with

1µM TTX. After the efficacy of TTX was established (i.e. the abolition of action potentials in response to depolarizing pulses), adiponectin was administered as above and the resulting membrane potential changes were measured. Both depolarizing (5.50 ± 1.26 mV, n=3) and hyperpolarizing (-4.65 ± 1.36 mV, n=3) effects were observed in response to adiponectin indicating the site of action of adiponectin is most likely directly at the cell membrane of these magnocellular neurons (Figure 2.3).

Adiponectin hyperpolarizes identified OT neurons

In view of the different responses of subsets of magnocellular neurons we next hypothesized that divergent effects of adiponectin may be due to different regulatory actions on oxytocin vs vasopressin neurons. We therefore undertook additional current clamp experiments in which we first defined adiponectin responsiveness of MNC neurons as described above

(depolarization, hyperpolarization or no response) and then used a single cell RT-PCR technique to identify the chemical phenotype of each recorded MNC (see Materials and Methods). Single cell RT-PCR analysis suggests that there are three functional groups of MNC neurons in the

PVN; neurons expressing vasopressin, oxytocin, or both vasopressin and oxytocin (Figure 2.4 C).

Every neuron examined in this study which definitively expressed only oxytocin hyperpolarized in response to 10nM adiponectin (n=6). None of the neurons expressing both vasopressin and oxytocin responded to 10nM adiponectin (n=6). Although all identified vasopressin neurons were

48

Figure 2-3. The effects of adiponectin are maintained when action potential induced release of neurotransmitter is inhibited.

A) Representative current clamp traces of MNC neurons showing the response to 10nM adiponectin (black bars) in the presence of 1µM TTX (diagonal lined bar). Scale bars: 10 mV,

100 seconds. B) Bar graph showing mean membrane potential change of all cells treated with adiponectin (10nM) in the presence of TTX (1µM). Significant changes in mean membrane potential of the responding groups compared to the non-responding group was established through a Kruskal-Wallis test (*p > 0.05).

49

Figure 2-4. Adiponectin exhibits different effects on identified MNC neurons in the PVN. A) Example of a MNC neuron in a slice preparation before (left) and after (right) cytoplasm has been aspirated into the recording pipette. Dashed lines above indicate the outline of the cell before (black) and after cytoplasm removal (red). Cytoplasm is then removed from the bath and subjected to RT-PCR analysis. B) Agarose gels of control protocols run with each cytoplasm processed in this study. Top: Agarose gel illustrating PCR products of each gene of interest derived from whole PVN cDNA which serves as a (+) control. Bottom: Agarose gel showing an RT-PCR analysis in which PVN cDNA has been omitted from the reaction. RT-PCR analysis that did not show a full complement of PCR products in the (+) control and a clean (-) control were omitted from the study. C) Correlation of the molecular phenotype of MNC neurons to the response seen after treatment with 10nM adiponectin. Agarose gels on the left illustrate the single cell RT-PCR products of three distinct functional groups of MNC neurons. Each cell included in the study showed the presence of the housekeeping gene GAPDH in addition to the expression profile of the cell. (GAPDH = Glyceraldehyde 3-phosphate dehydrogenase, GAD67 = Glutamate decarboxylase 67, TH = Tyrosine hydroxylase, VP = Vasopressin, OT = Oxytocin)

50

also responsive to adiponectin, these cells showed heterogeneous effects with equal proportions being either depolarized (3/6) or hyperpolarized (3/6) as shown in Figure 2.4 C.

MNC PVN neurons show diverse adiponectin receptor expression profiles

In light of the differential effects of adiponectin on the different chemical phenotypes of

MNC neurons described above we hypothesized that each individual group expressed a different profile of adiponectin receptors. We tested this hypothesis using the single cell RT-PCR technique in which aspirated cytoplasm from electrophysiologically identified MNC neurons was subjected to RT-PCR analysis as described above (see Materials and Methods) to examine the expression of adiponectin receptors in conjunction with peptidergic expression; vasopressin or oxytocin (Figure 2.5). In neurons expressing exclusively oxytocin, both adiponectin receptors were detected in all but one cell (6/7) (Figure 2.5 A). The expression of adiponectin receptors in vasopressin neurons exhibited more complexity. Both adiponectin receptors were present in 3/8 neurons expressing vasopressin (Figure 2.5 B) whereas in 4/8 neurons, only AdipoR2 was present

(Figure 2.5 C) and 1/8 neurons expressed only AdipoR1. A majority of MNC neurons (4/6) that co-expressed oxytocin and vasopressin did not express either receptor (figure 2.5 D), however 2/6 neurons expressing both vasopressin and oxytocin were found to express AdipoR2.

DISCUSSION

We have shown that PVN tissue expresses both adiponectin receptor mRNAs and that activation of these receptors has direct consequences on the membrane properties of MNC neurons in the PVN. Our data indicate that adiponectin has unique effects on distinct populations of MNC neurons as measured by single-cell RT-PCR; selectively hyperpolarizing oxytocin neurons, exhibiting no effect on those expressing oxytocin and vasopressin and either depolarizing or hyperpolarizing vasopressin neurons. Furthermore our results indicate that 51

Figure 2-5. Subgroups of MNC neurons in the PVN express different profiles of adiponectin receptors.

A) Agarose gels indicating PCR products from an RT-PCR reaction of the cytoplasmic contents of four different MNC neurons. Each gel is an example of the predominant expression pattern in each group: i) An oxytocin neuron expressing both adiponectin receptor 1 and 2, ii) A vasopressin neuron expressing both receptors or AdipoR2 (iii) and a co-expressing MNC neurons expressing neither receptor (iv). B) Ratio of neurons in each group which expressed the profile of receptors indicated in each respective agarose gel. (GAPDH = Glyceraldehyde 3-phosphate dehydrogenase, GAD67 = Glutamate decarboxylase 67, TH = Tyrosine hydroxylase, VP = Vasopressin, OT = Oxytocin, AdipoR1 = adiponectin receptor 1, AdipoR2 = adiponectin receptor 2)

52

oxytocin and vasopressin neurons exhibit distinct adiponectin receptor expression profiles which may underlie the heterogeneous electrophysiological effects seen on these groups of neurons.

These results indicate that this peptide may play important roles in controlling the release of neurohypophysial peptides into the circulation.

In order to examine the responsiveness of individual neurons we used patch-clamp electrophysiology to monitor the membrane potential and action potential frequency of PVN neurons in a slice preparation of the hypothalamus. Previous reports from our lab and others have shown that MNC neurons respond to a variety of signaling factors and display diverse and repeatable responses to treatment (Ferri et al, 2005; Latchford & Ferguson, 2003; Follwell &

Ferguson, 2002b; Hermes et al, 1999). In addition, maintaining cells in a slice preparation allows us to examine a synaptic component; in that performing the experiments in the presence of TTX effectively isolates the recorded neuron from action potential induced release of neurotransmitter and allows us to measure sensitivity to treatment directly at the membrane.

Considerable effort has been directed toward the identification of an electrophysiological fingerprint unique to either vasopressin or oxytocin expressing neurons (Armstrong, 1995;

Armstrong & Stern, 1998a; Armstrong & Stern, 1998b; Hirasawa et al, 2003) although the use of such fingerprints is recognized to leave some doubt regarding the chemical phenotype of recorded neurons (Armstrong, 1995). In our studies, we identified cell type using single-cell RT-PCR analysis after whole cell recording (O'Dowd & Smith, 1996; Lambolez et al, 1992; Qiu et al,

2005) to identify the specific RNAs (OT/VP) expressed in recorded neurons. The single cell RT-

PCR system is remarkably sensitive; being able to detect individual mRNA transcripts from less then 1/100th the total mRNA contents of a single neuron (based on our whole PVN mRNA extraction and serial dilution sensitivity tests). However an important consideration when

53

utilizing single cell RT-PCR technology is to avoid the incorrect interpretation of results due to false-positives or false-negatives. To eliminate the possibility that the detected signal was due to genomic DNA contamination, cytoplasmic RNA was DNased (see Materials and Methods). A sample of bath solution was also taken and an RT (-) in which the reverse transcriptase enzyme was omitted from the reaction were included with each cell tested. A separate reaction which included no PCR template served as an additional negative control. In addition to negative controls, cytoplasm was tested for the presence of the housekeeping gene GAPDH and primer efficacy confirmed by a reaction with cDNA template derived from whole PVN tissue. Every cell included in this study showed a positive reaction for GAPDH and all target genes from whole tissue PVN. The success of each of these controls indicates to us that our single cell RT-PCR analysis is definitive and with it we have identified three groups of PVN MNC neurons based on peptidergic .

Our lab recently showed mRNA for both adiponectin receptors expressed in the area postrema; a sensory circumventricular organ important in autonomic control (Cottrell &

Ferguson, 2004) and showed area postrema neurons to be responsive to adiponectin (Fry et al,

2006). The results presented in this study, however, are the first to show receptor expression in an adult nucleus which exists behind the blood brain barrier (BBB). These observations raise interesting questions as to the source of peptide especially in view of permeability studies indicating that adiponectin likely does not passively cross the BBB (Pan et al, 2006; Spranger et al, 2006), presumably due to its large multimeric circulating size (Waki et al, 2003). However, iv injections of globular adiponectin have been reported to increase cerebral spinal fluid (CSF) concentrations (Qi et al, 2004) and adiponectin complexes have recently been detected in human cerebrospinal fluid (Kusminski et al, 2007) suggesting that circulating peptide could gain access

54

to the CNS through the ventricular space. A recent report suggests CNS endothelial cells express adiponectin receptors activation of which decreases the release of interleukin-6 behind the BBB

(Spranger et al, 2006). This mechanism could explain the effects seen on MNC neurons as vasculature is maintained in our slice, though it seems unlikely that decreases in release could be responsible for the diverse electrophysiological effects. A final possibility suggested by a recent report that adiponectin is also produced in the chicken diencephalon (Maddineni et al,

2005) at least raises the possibility that the CNS may also produce this adipokine. Specific neuronal localization, relevant interconnectivity with PVN MNC neurons and whether central adiponectin exhibits the same inverse adiposity correlation remains to be explored but may provide insight into the source and role of adiponectin in the CNS.

Intriguingly, the data presented in this study suggests that adiponectin differentially modulates neuronal excitability of distinct chemical phenotypes of MNC neurons in the PVN.

Our data indicate that adiponectin specifically hyperpolarizes oxytocin neurons thus inhibiting action potential generation and ultimately oxytocin release. Although many central signals involved in the regulation of energy balance including leptin (Hakansson et al, 1998; Ur et al,

2002), cocaine-amphetamine-related transcript (Vrang et al, 2000), CCK (Olson et al, 1992; Ueta et al, 1993) and PHI (Olszewski et al, 2003) play critical roles in controlling oxytocin secretion, the precise role of this neurohypophysial peptide in the regulation of energy balance has yet to be established. Modified serum concentrations of oxytocin have been reported in pathological states of energy imbalance. Serum oxytocin is approximately four fold higher in clinically obese than control individuals (Stock et al, 1989), and is significantly elevated in a rodent model of diet induced obesity (Northway et al, 1989). Conversely, decreased central (no data on circulating levels) concentrations of oxytocin have been reported in the CSF of underweight restrictive

55

anorexic patients (Demitrack et al, 1990), an abnormality which tends to normalize following restoration of normal weight (Kaye, 1996). There is, however, no direct information to link these changes in oxytocin to the predicted pathological changes in adiponectin function (decrease and increase respectively) associated with such conditions.

The different responses of neurons expressing exclusively vasopressin suggest the possibility that a subgroup of MNC vasopressin neurons exist as a separate functional group.

These may be caudally projecting neurons innervating autonomic centers in the brainstem but with similar electrophysiological characteristics to those of the neurohypophysial system

(dominant IA currents). The function of this group of neurons and what effect they have on autonomic system in response to adiponectin remains to be determined.

The different expression profiles (as assessed by single cell RT-PCR) of adiponectin receptors in subgroups of MNC neurons suggest that hyperpolarizing responses are associated with the expression of both AdipoR1 and R2 (oxytocin and some vasopressin neurons), while depolarizing effects may be associated with expression of AdipoR2 alone (proportion of vasopressin neurons only). As expected, a majority of the co-expressing vasopressin and oxytocin neurons showed no receptor expression correlating with the lack of response seen to adiponectin. These conclusions are slightly different to those from our previous study of area postrema neurons (Fry et al, 2006), in which such correlative data suggested that all cells showing responses to adiponectin expressed both AdipoR1 and R2, and suggest that in addition to receptor expression the broader phenotype of individual neurons (channel expression etc) may all play a role in determining responsiveness of individual neurons to adiponectin. The suggestion that hyperpolarization occurs through the activation of two receptors raises speculation about the mechanism by which this may proceed. A recent study (Kubota et al, 2007) examining the effects

56

of adiponectin in the arcuate nucleus of the hypothalamus suggests that activation of AdipoR1 stimulates the phosphorylation of AMP-activated protein kinase (AMPK) thus increasing the production of ATP. It is well established that altering ATP concentrations within the cell has functional consequences on membrane potential and neuronal firing rate of the neuron through the alteration of KATP channel conductances (Spanswick et al, 2000). Microarray studies show the presence of KATP channel transcripts in the PVN (Hindmarch et al, 2006), however, further studies will be required to examine their response to adiponectin. Additionally further studies will be needed to delineate the signaling pathways of AdipoR2 and examine to what extent those contribute to the membrane potential of magnocellular neurons.

Finally, our recordings from chemically phenotyped vasopressin and oxytocin neurons confirms previous reports of their existence within the magnocellular PVN/SON (Mezey & Kiss,

1991; Xi et al, 1999). The consistently observed lack of response of these neurons to adiponectin suggests they may represent a separate functional group of magnocellular neurons whose properties have yet to be fully elucidated.

In conclusion we have shown that adiponectin modulates the membrane properties of

MNC neurons in the PVN, potentially through the activation of AdipoR1 and/or AdipoR2.

Furthermore though single cell RT-PCR analysis, we have shown that adiponectin specifically regulates the oxytocinergic system by hyperpolarizing the membrane of neurons that express this peptide and both AdipoR1 and R2. Both these and previous results of adiponectin actions in the

CNS highlight the growing complexity with which this peptide exerts its central effects and may provide insight into the pathology of neuronal systems that regulate energy homeostasis.

57

ACKNOWLEDGEMENTS

This work was supported by grants from Heart and Stroke Foundation (AVF), and a

Target Obesity Canadian Heart and Stroke, CIHR Doctoral Research Award (TDH) and Canadian

Diabetes Association postdoctoral fellowship (MF).

58

Chapter 3: ADIPONECTIN DEPOLARIZES PARVOCELLULAR

PARAVENTRICULAR NUCLEUS NEURONS CONTROLLING NEUROENDOCRINE

AND AUTONOMIC FUNCTION

Ted D. Hoyda1, Willis K. Samson2, Alastair V. Ferguson1

1Department of Physiology, Queen's University, Kingston, ON, Canada K7L 3N6.

2Pharmacological and Physiological Science, St Louis School of Medicine, St Louis, MO, USA

63104.

Endocrinology. 2009, Vol. 150(2):832-840

59

ABSTRACT

Adiponectin plays important roles in the control of energy homeostasis and autonomic function through peripheral and central nervous system (CNS) actions. The paraventricular nucleus of the hypothalamus (PVN) is a primary site of neuroendocrine and autonomic integration and thus a potential target for adiponectin actions. Here we investigate actions of adiponectin on parvocellular PVN neurons. Adiponectin influenced the majority (65%) of parvocellular PVN neurons, depolarizing 47% while hyperpolarizing 18% of neurons tested.

Post-hoc identification (single cell RT-PCR) following recordings revealed that adiponectin depolarizes neuroendocrine-CRH neurons, while icv injections of adiponectin in vivo caused increased plasma ACTH concentrations. Adiponectin also depolarized the majority of TRH neurons however neuroendocrine-TRH neurons were unaffected, in accordance with in vivo experiments showing that icv adiponectin was without effect on plasma TSH. In addition bath administration of adiponectin also depolarized both preautonomic TRH and OT neurons. These results suggest adiponectin acts in the CNS to coordinate neuroendocrine and autonomic function through actions on specific functional groups of PVN neurons.

60

INTRODUCTION

Energy homeostasis is coordinated in large part through peripherally produced hormones which interact with the central nervous system (CNS) (Murphy & Bloom, 2006). Adipose tissues produce a number of these signaling molecules, which function in the CNS to dictate feeding patterns and responses to stress, and maintain dynamic equilibrium. Adiponectin is one such protein that, like leptin and resistin, is produced by adipocytes, circulates in the plasma and controls network excitability in important autonomic nuclei (Ahima & Lazar, 2008; Singhal et al,

2007; Spanswick et al, 1997; Powis et al, 1997). Unlike leptin and resistin, however, adiponectin circulates at extremely high concentrations that vary inversely with adiposity (Hu et al, 1996;

Scherer et al, 1995). Adiponectin knockout mice show impaired responses to glucose challenges and have compromised insulin and metabolic function among other pathologies (Kubota et al,

2002). When injected into the brain, adiponectin has a marked effect on autonomic function.

Mice show a substantial decrease in body weight, increased thermogenesis, increased oxygen consumption and increased corticotropin releasing hormone (CRH) mRNA production in the paraventricular nucleus (PVN) (Qi et al, 2004). This evidence suggests that adiponectin plays a critical role in controlling energy homeostasis.

Adiponectin signals through two known receptors: AdipoR1 and AdipoR2 (Yamauchi et al, 2003). Adiponectin receptors are located throughout the CNS, notably in regions of the hypothalamus and brainstem important in controlling autonomic function and feeding behavior, including the area postrema (AP) (Fry et al, 2006), arcuate nucleus (ARC) (Kubota et al, 2007) and the PVN (Hoyda et al, 2007).

The PVN contains neurons which play important roles in controlling autonomic state

(Swanson & Sawchenko, 1980). Electrophysiologically the PVN can be divided into three

61

populations of neurons magnocellular (MNC), pre-autonomic parvocellular (PA) and neuroendocrine parvocellular (NE) neurons (Luther et al, 2002; Tasker & Dudek, 1991). Distinct groups of PA and NE neurons contain CRH, thyrotropin releasing hormone (TRH) and oxytocin

(OT), which participate as neuromodulators (PA) or hormones (NE) translating output information from the PVN to its targets (Swanson & Sawchenko, 1980).

We recently reported that adiponectin controls the excitability of PVN MNC neurons: hyperpolarizing OT neurons, having varying effects on vasopressin neurons and having no effect on neurons which expressed both peptides (Hoyda et al, 2007). Here we examine the effects of adiponectin on the parvocellular neurons in the PVN (PA and NE), specifically investigating three chemically distinct types of neurons within this group: CRH-, TRH- and OT-expressing cells, while also correlating electrophysiological effects with integrated neuroendocrine function in conscious, freely moving animals.

MATERIALS AND METHODS

All procedures involving the use of animals conformed to the standards of the Canadian

Counsel on Animal Care and were approved by the Queen‟s University Animal Care Committee.

All chemicals used in making solutions were obtained from Sigma Pharmaceuticals (Oakville,

ON, Canada). TTX citrate was obtained from Alomone Laboratories (Jerusalem, Israel) made into working aliquots and stored at -80 °C. Globular adiponectin was obtained from Phoenix

Pharmaceuticals (Belmont, CA, USA), made into working aliquots and stored at -20°C until experimentation.

Electrophysiology

Coronal hypothalamic slices (300 µM) containing the PVN were taken daily from Male

Sprague Dawley rats (Charles River, Quebec, Canada) between P19-P26 who were maintained on 62

a 12hr light/dark cycle and provided with food and water ad libitum. Rats were decapitated, brains removed and placed in ice cold slicing solution consisting of (in mM) 87 NaCl, 2.5 KCl,

25 NaHCO3, 0.5 CaCl2, 7 MgCl2, 1.25 NaH2PO4, 25 glucose, 75 sucrose bubbled with 95%

O2/5% CO2 for 3-5 minutes. Brains were then trimmed to size, mounted, immersed with slicing solutions and 300 µM hypothalamic slices containing PVN were cut using a vibratome (Leica,

Nussloch, Germany). Slices were then incubated at 32.5°C in artificial cerebral spinal fluid

(aCSF) containing (in mM) 126 NaCl, 2.5 KCl, 26 NaHCO3, 2 CaCl2, 2 MgCl2, 1.25 NaH2PO4,

10 glucose saturated with 95% O2/5% CO2 for at least 1 hr prior to recording. aCSF was used as external recording solution for all electrophysiological experiments unless otherwise indicated.

Parvocellular PVN neurons were visualized using a 40X water-immersion objective mounted to an upright Nikon E600FN microscope fitted with DIC-IR optics (Tokyo, Japan).

Borosilicate glass (World Precision Instruments, Sarasota FL, USA) was pulled by a flaming micropipette puller (Sutter Instruments Co., Novato CA, USA) and when filled with internal

+ recording solution containing (in mM) 130 K -gluconate, 10 KCl, 2 MgCl2, 0.1 CaCl2 (calculated intracellular free Ca2+: 2.77 nM), 5.5 EGTA, 10 HEPES, 2 NaATP, adjusted to pH 7.2 with KOH had resistances of between 2-5 MΩ. Whole cell configuration was obtained following the formation of a GΩ seal by applying negative pressure to the pipette and monitoring resistance changes on an oscilloscope.

Current clamp recordings from parvocellular PVN neurons were obtained with a

Multiclamp 700B (Molecular Devices, Palo Alto, CA, USA) patch clamp amplifier. Parvocellular

PVN neurons were classified as either preautonomic (type 2) or neuroendocrine (type 3) neurons using a current pulse protocol designed to identify the presence of a low-threshold spike (LTS), or non-accomodating spike firing, identifying electrophysiological features unique to pre-autonomic

63

(PA) and neuroendocrine (NE) neurons in the PVN (Luther et al, 2002; Stern, 2001). The pulse protocol consisted of an initial 250 ms step of -60 pA from baseline membrane potential followed by sweeps of 10 pA (from -60 pA to 40 pA) steps for 250 ms until action potential threshold was reached. Neurons that displayed an LTS were classified as PA (example Figure 3.4 A inset) while parvocellular neurons which showed no discernable LTS and no accommodation of spike frequency during depolarization were classified as NE (example Figure 3.3 A inset). Neurons which had an ambiguous electrophysiological fingerprint were removed from the study.

Experiments were recorded in real-time using acquisition software Spike 2 (Cambridge

Electronic Devices, Cambridge, UK). Capacitance transients and series resistance error was minimized at the start of all recordings.

After establishing a stable control membrane potential for at least 100 seconds, adiponectin, dissolved in aCSF to a concentration of 10 nM (previous studies indicate 1nM is the minimum effective concentration (Hoyda et al, 2007)) was bath applied to the slice at a rate of 1-

2 ml/min. Responsiveness of neurons to adiponectin was assessed by averaging all membrane potential points in 100 second units starting from the time of application of peptide measuring 10 units. The peak mean voltage change in a 100 second unit was compared to the mean membrane potential of points in the control unit. Neurons were categorized as responsive if the peak mean within the ten units exceeded two times the standard deviation of the control membrane potential.

Membrane potential was recorded until values returned to those similar to control. Neurons were eliminated from the data set if they did not show at least partial recovery to baseline.

Single cell RT-PCR

Following experimentation, a post-hoc single cell reverse transcription polymerization chain reaction (single cell RT-PCR) was performed to detect mRNA for prominent neuropeptides

64

expressed in parvocellular neurons (see Table 3.1). Gentle suction was applied to the recording pipette to aspirate cytoplasm. Membrane resistance was continuously monitored as an outside-out patch was pulled sealing the cytoplasmic contents within the pipette. Cells which failed to form outside-out patches were eliminated from single cell RT-PCR analysis but were included in the analysis of changes in membrane potential. Following pipette withdrawal from the bath, the electrode tip was broken and its contents were expelled into a 0.5 mL PCR tube. The following were added to the tube to set up the DNase reaction: (approximate concentrations) 10 U DNase I and 10 mM 10X reaction buffer with MgCl2 (Fermentas Life Sciences, Burlington, ON) and incubated at 37 °C for 30 minutes. Following incubation, 2.5 mM EDTA was added and incubated at 65 °C for 10 minutes to stop the DNase reaction. Reverse transcription reaction proceeded immediately following by adding dithiothreitol (26 mM), dNTPs (3mM), random hexamers (3 µM), MgCl2 (4 mM), RNase inhibitor (20 U) and superscript II (100 U) (Burlington,

ON, and allowed to incubate overnight at 37 °C to allow the reaction to proceed. A negative control reaction in which superscript II was omitted from the reaction mixture was also performed. Samples were stored at -80 °C until the multiplex reaction.

A multiplex PCR strategy was employed to amplify the cDNA produced in the RT reaction. This protocol consisted of two different amplification steps: outside and nested. First, primer sets (0.2µM) specific for all genes of interest (outside) were added to the cDNA from the single cell which then underwent an amplification protocol. Reactions were performed in 100 µl volumes with reagents provided in the Qiagen Multiplex kit (Qiagen, Mississauga, ON, Canada).

The reaction mixture was denatured at 95 °C for 15 minutes then cycled 20 times through a temperature protocol of 94 °C for 30s, 60 °C for 90s, and 72 °C for 90s. The second step consisted of individual reactions for each gene of interest using „nested‟ primer sets. Reactions

65

Product Size Gene Primer Name Position Sequence (bp) Glyceraldehyde-3- GAPDH F (outside) gatggtgaaggtcggtgtg 469 phosphate R gggctaagcagttggtggt dehydrogenase F (nested) taccaggctgccttctct 360 R ctcgtggttcacacccatc γ-amino butyric acid GAD67 F (outside) cacaaactcagcggcataga 550 R gagatgaccatccggaagaa F (nested) cacaaactcagcggcataga 149 R ctggaagaggtagcctgcac Tyrosine Hydroxylase TH F (outside) cacctggagtattttgtgcg 1138 R cctgtgggtggtaccctatg F (nested) tcgacccagtatatccgcca 376 R tcggacacaaagtacacagg Vasopressin VP F (outside) cctcacctctgcctgctactt 237 R agccagctgtaccagcctaa F (nested) acctctgcctgctacttcca 216 R agccagctgtaccagcctaa Oxytocin OT F (outside) ctgccccagtctcgcttg 281 R cctccgcttccgcaaggcttctggc F (nested) ctgccccagtctcgcttg 244 R gcgagggcaggtagttctcc Corticotropin CRH F (outside) ggggaaaggcaaagaaaagg 184 releasing R gacagagccaccagcagcat hormone F (nested) ggagaagagaaaggagaagaggaa 140 R ggacaccagcagccgcag Thyrotropin TRH F (outside) agaggggagacttgggagaa 447 releasing R ctttgcttcaccagggtctc hormone F (nested) attcatgggcagatgaggag 245 R ggcgtttctcaggcattaag Vesicular VGLUT2 F (outside) aggttggctaccacctcctt 583 glutamate R tgagagtagccaacaaccagaagca transporter-2 F (nested) cccgcaaagcatccaacca 164 R cctgcagaagtttgcaacaa Adiponectin AdipoR1 F (outside) gtcccctggctctattactcct 509 receptor 1 R agcacttggctgtgatgt F (nested) tcttcctcatggctgtgatgt 223 R ggctcagagaagggagtcatc Adiponectin AdipoR2 F (outside) ggagccattctctgcctttc 464 receptor 2 R ccagatgtcacatttgcca F (nested) actgtaacccacaaccttgcttc 191 R tcaggaacccttctgagatgac

Table 3-1. Primers sets used in the detection of mRNA from single cells.

66

were performed in 50µl volumes consisting of Qiagen Multiplex reaction reagents, 0.2µM of primers and 2µl of solution from the multiplex reaction to act as template. The reaction mixture was subject to 35 cycles of the same amplification protocol. Two control reactions were performed during the multiplex protocol: 1) A reaction was run with PCR H2O instead of cDNA to detect for genomic contamination and 2) a second reaction was run with cDNA from whole

PVN to assess primer integrity and eliminate false negatives. PCR products from the nested reaction were run on a 2% (w/v) agarose gel containing ethidium bromide and sequenced to confirm their identity (Robarts Institute, London, ON, Canada). PCR reactions that failed to pass any of the control tests were eliminated from the study.

In Vivo Experiments

Under ketamine (Ketaset, Fort Dodge Animal Health, Fort Dodge, IA)/xylazine

(TranquiVed, Vedco Inc. St. Joseph, MO) anesthesia (60mg/8mg mixture/ ml, 0.1 ml/100 gram body weight, intraperitoneal injection) rats were placed in a stereotaxic device and a 23 gauge, stainless steel cannula (17 mm) implanted into the right lateral cerebroventricle as previously described (Antunes-Rodrigues et al, 1985). Minimally five days later, after the animals had returned to preimplantation body weights, an indwelling jugular vein cannula was implanted as previously described (Harms & Ojeda, 1974) under isoflurane-induced anesthesia (3% in O2 for induction, 2% in O2 for maintenance of anesthesia; IsoSol, Vedco Inc.). The jugular cannula was exteriorized at the back of the neck and sealed with heparinized saline (200 units/ml 0.9% NaCl).

On the following day, an extension tubing (PE-50) was attached to the jugular cannula to facilitate blood sampling and rats left undisturbed for minimally 120 minutes. Then, an initial blood sample was withdrawn from the jugular vein without disturbing the animal. All blood samples (0.3 ml) were removed from conscious, unrestrained rats into heparinized syringes and

67

replaced with an equal volume of 0.9% NaCl (37 °C). Sampling was conducted between the hours of 0900 and 1100 (lights on 0600). Blood samples were stored on ice before plasma was separated (10,000 X g, 5 min) and stored at –20 °C until hormone assays conducted. Immediately following the removal of the initial (0 time) blood sample, a 2 microliter injection of isotonic saline vehicle alone or vehicle containing 0.1, 0.3, or 1.0 nanomole adiponectin was conducted via the indwelling cerebroventricular cannula. Subsequent blood samples were removed 5, 15, 30, and 60 minutes after icv injections.

Plasma TSH levels were measured (20 microliter aliquots) using the materials provided in the NIH kit (rTSH-RP-2 standard, minimum detectable level: 0.5 ng/ml; inter-assay and intra- assay coefficients of variability were less than 8%). ACTH concentrations in plasma (33 microliter aliquots) were determined by RIA using the unextracted plasma protocol described by the supplier (Peninsula Laboratories, Inc., San Carlos, CA). Minimum detectable ACTH levels were 1 pg/ml and inter- and intra-assay coefficients of variability were less than 8%.

Data from the RIAs were analyzed by one way ANOVA both within treatment groups across time and between treatment groups at any sampling time point, followed by Scheffé‟s multiple comparison testing. Homogeneity of variance was established using the S test.

Significance was assigned to results that occurred with less than 5% probability.

RESULTS

Adiponectin influences the excitability of parvocellular neurons in the PVN

Current clamp recordings were obtained from 122 identified (58 PA and 64 NE) parvocellular PVN neurons, and the effects of bath administration of 10 nM adiponectin on membrane potential were examined. Adiponectin influenced the membrane potential of the majority of both NE and PA neurons with depolarizing and hyperpolarizing effects observed in 68

Figure 3-1. Adiponectin effects on the membrane potential of NE and PA neurons in the

PVN.

A) Bath administration of 10 nM adiponectin (black bar) induces both depolarizing (i) and hyperpolarizing (ii) effects on membrane potential in different populations of type 3 NE neurons.

Membrane potential routinely recovered to near control membrane potential (red line) following removal of adiponectin from the bath. B) Similar effects of adiponectin on membrane potential were seen in type 2 PA neurons, depolarizing (i) a population of cells while hyperpolarizing (ii) a separate group of neurons.

69

Figure 3-2. The effects of adiponectin are direct and proportionally similar across parvocellular PVN neurons.

A) Histogram showing percentage of each cell type; NE, PA, total combined and when action potential induced neurotransmitter release was inhibited through the application of 1µM TTX of neurons depolarizing (blue) or hyperpolarizing (red) in the presence of 10 nM adiponectin. B)

Summary box and whisker plot indicating for both depolarization and hyperpolarization the smallest observation, the lowest quartile, median, upper quartile and the largest observation in each data set. The median depolarization was 5.5 mV while the median hyperpolarization was -

5.7 mV. 70

different cells in each group (Figure 3.1 A-B). The majority of cells influenced by adiponectin in each group depolarized (7.0 ± 0.5 mV, n = 59) (PA – 50%, NE – 40%, total – 48%), while a smaller proportion of cells hyperpolarized (-6.7 ± 0.7 mV, n = 23) (PA – 17%, NE – 23%, total –

20%) (Figure 3.2 A, B). A third population of parvocellular neurons did not respond to 10 nM adiponectin. Both depolarizing (8.4 ± 1.3 mV, n=15) and hyperpolarizing (-5.9 ± 0.9 mV, n=11) effects of adiponectin were maintained in the presence of 1µM TTX (n=39), and were observed in similar proportions of neurons (Figure 3.2 A) suggesting that mechanistically, adiponectin acts directly at the cell membrane of parvocellular neurons to modulate their electrical excitability.

Adiponectin depolarizes neuroendocrine CRH neurons

In order to assess adiponectin‟s effects on specific groups of neurons in the PVN following membrane potential recordings we performed post hoc single cell RT-PCR analysis for markers of interest (Table 3.1). Thirteen of the 122 neurons recorded from were positively identified as CRH mRNA expressing neurons. Of 13 CRH neurons, 10 were NE neurons and

70% of those neurons showed a depolarizing shift in membrane potential in response to 10 nM adiponectin (5.2 ± 1.1 mV) (Figure 3.3 A, B). In addition 3 PA CRH neurons responded to 10 nM adiponectin (2 depolarized, 1 hyperpolarized). Adiponectin receptor profile in CRH NE neurons was also assessed by single cell RT-PCR and all 7 of the responding CRH neurons expressed mRNA for at least one adiponectin receptor (R1 - n=3, R2 - n=1, R1/R2 - n=3). Neuroendocrine

CRH neurons that did not express either receptor did not respond to adiponectin (0.7 ± 0.9 mV, n=3).

We next examined the effect of adiponectin on hypothalamic-pituitary-adrenal axis by measuring the effects of adiponectin injected into the lateral cerebroventricle (icv) on plasma

ACTH concentrations. Injections (icv) of adiponectin caused an increase in plasma ACTH

71

Figure 3-3. Adiponectin activates the hypothalamic-pituitary stress axis through excitatory actions on NE CRH neurons in the PVN. A) The electrophysiological trace on the bottom is a current clamp recording from an electrophysiologically identified NE neuron (inset) whose post hoc single cell RT-PCR profile (gel inset) identifies it as expressing CRH, depolarizing following bath application of 10 nM adiponectin (black bar). This neuron had a robust increase in action potential frequency as illustrated in a rate meter recording above the trace showing a change in spike frequency before and after adiponectin administration. Scale bars: 30 mV and 100 ms (inset), 10 mV and 50 sec. B) A series of box and whisker plots showing the response of cells that expressed CRH as detected by single cell RT-PCR to adiponectin administration. All but one NE CRH cell depolarized whereas PA CRH neurons were mixed in their response to adiponectin. C) Plasma ACTH levels following icv injection of vehicle or adiponectin at three different doses, measured at different time points up to one hour following injection (*p<0.05, between groups ANOVA, Scheffe Multiple Comparisons).

72

concentrations 15 and 30 minutes after injections when compared to saline injected controls

(Figure 3.3 C). The injection of 1.0 nmole adiponectin showed the largest increase in plasma

ACTH concentration compared to vehicle at 15 minutes (22.3 ± 3.7 pg/ml (1.0 nmole adiponectin) vs. 8.6 ± 1.0 pg/ml (vehicle) p=0.014), (ANOVA, Scheffe Multiple Comparisons)).

Plasma ACTH concentrations had returned to baseline at the 60 minute time point suggesting the actions of adiponectin are transient and reversible. These results indicate that adiponectin acts on

CRH neurons in the PVN to control CRH release from the median eminence, and subsequent

ACTH release from the anterior pituitary gland.

Adiponectin depolarizes preautonomic TRH neurons

We also obtained current clamp recordings from a total of 10 neurons that were identified post hoc as TRH mRNA expressing neurons in which we were able to assess the effects of bath administration of 10 nM adiponectin. The majority of TRH neurons (6/10) depolarized in response to 10 nM adiponectin (7.6 ± 1.9 mV) (Figure 3.4 A). More interestingly, classification of TRH neurons as NE or PA revealed that a majority of the TRH neurons depolarized by adiponectin were PA cells (6/7), while 0/3 NE TRH neurons were affected by adiponectin (Figure

3.4 B). Analysis of the expression patterns of adiponectin receptors in TRH neurons showed that

1/6 responsive PA TRH neurons expressed receptors (R1 – n=0, R2 – n=0, R1/R2 – n=1), while a significant number that responded to adiponectin (4.0 ± 2.9 mV, n=5) did not. We hypothesized that this may be due to the effect of adiponectin on a pre-synaptic interneuron and therefore performed additional current clamp recordings from a total of 4 identified PA TRH neurons that did not express either adiponectin receptor in the presence of 1µM TTX, and observed clear effects of adiponectin in all of these neurons (depolarization in 2 and hyperpolarization in 2), as shown in Figure 3.4 B suggesting that adiponectin may influence PA TRH neurons, at least in

73

Figure 3-4. Adiponectin selectively depolarizes type 2 PA TRH neurons.

A) Current clamp recording of an identified type 2 PA neuron (inset) that expresses mRNA for

TRH (gel inset) and depolarizes to bath application of 10 nM adiponectin (black bar). Scale bars:

20 mV and 100 msec (inset), 10 mV and 100 sec. B) Box and whisker plots showing different subgroups of PVN cells that express TRH as measured by single cell RT-PCR and their response to adiponectin. C) Plasma TSH concentrations following icv injections of vehicle versus 3 increasing doses of adiponectin measured at increasing time points up to 60 minutes following injection.

74

part, through an indirect sodium channel independent mechanism. The mechanisms underlying such indirect effects are not known at the current time.

In addition to measuring plasma ACTH concentrations following lateral ventricle icv injections of adiponectin, plasma TSH concentrations were measured. TSH concentrations showed no significant change compared to vehicle control at any concentration or time point (15 min, vehicle: 8.8 ± 1.2 ng/ml vs. 1.0 nmole: 7.8 ± 0.9 ng/ml, ANOVA, Scheffe Multiple

Comparisons) following adiponectin administration (Figure 3.4). These data suggest that adiponectin selectively regulates ACTH secretion in the hypothalamic-pituitary axis but has no effect on TRH/TSH secretion.

Adiponectin depolarizes preautonomic OT neurons

A significant proportion of PA neurons which express OT send axonal projections to the nucleus of the solitary tract (NTS), and have been suggested to modify feeding by sensitizing

NTS neurons to satiety signals such as CCK (Blevins et al, 2004). We recorded from 15 PA neurons that were positively identified as OT expressing neurons. The majority of these PA OT neurons (11/15) depolarized in response to 10 nM adiponectin (4.1 ± 0.4 mV) (Figure 3.5) while one of the remaining 4 cells tested was hyperpolarized by peptide administration. The majority of adiponectin responsive PA OT neurons expressed adiponectin receptors (R1 - n=2, R2 - n=2,

R1/R2 - n=3, 7/11), while in accordance with our recordings from PA TRH neurons a proportion of these adiponectin responsive OT PA neurons (4/11) did not express mRNA for either of these receptors. These data suggest that adiponectin may interact with OT PA neurons, and importantly highlight opposite actions of this adipokine of functionally distinct subpopulations of OT expressing neurons in the PVN.

75

Figure 3-5. The majority of type 2 PA OT neurons are depolarized by adiponectin.

A) Current clamp recording of a type 2 PA OT neurons in response to bath administration of 10 nM adiponectin. A majority of PA OT neurons showed a depolarizing shift in membrane potential followed by a return to baseline upon removal of adiponectin from the bath. Scale bars:

10 mV and 50 sec. B) Box and whisker summary plot of type 2 PA OT neurons and their response to adiponectin.

76

DISCUSSION

These data show both endocrine and autonomic effects of adiponectin acting upon specific groups of neurons in the PVN, confirming and identifying new and potentially critical roles played by this peptide in regulating energy and autonomic homeostasis. We have shown that adiponectin is primarily an excitatory peptide in the parvocellular regions of the PVN, depolarizing a majority of identified parvocellular neurons while hyperpolarizing only a minority of neurons in the same groups. We have also shown that adiponectin directly excites NE CRH neurons which express adiponectin receptors leading potentially to an increase in CRH release into pituitary portal circulation. Indeed, we have measured an increase in plasma ACTH concentrations following lateral cerebroventricular injections of adiponectin. Additionally, we have identified two distinct groups of PA neurons; TRH expressing and OT expressing cells whose electrical excitability is controlled by adiponectin suggesting the peptide may play alternate regulatory roles in addition to hypothalamic-pituitary axis induction through modulation of caudally projecting, PA cells.

CRH is a potent anorexigenic peptide that when released from parvocellular neurons into the median eminence regulates the secretion of ACTH from anterior pituitary cells ultimately controlling plasma glucocorticoid levels (Charmandari et al, 2005). Our observations that adiponectin depolarizes the majority of NE CRH neurons, while also increasing plasma concentrations of ACTH suggests mechanisms through which this peptide may control activation of the stress axis. Previous reports indicate that when injected icv, adiponectin increases respiration, energy metabolism and body temperature in mice (Qi et al, 2004), all indicators of elevated sympathetic tone. Our results clarify the mechanism responsible for this and suggest that

77

adiponectin‟s actions at CRH neurons involved in HPA axis partially dictate the metabolic and autonomic state of the animal.

PVN TRH neurons also control feeding behavior, participate in the stress response and control metabolic function (Lechan & Fekete, 2006). Increases in TRH secretion at the median eminence are associated with elevated thyroid function ultimately increasing the drive to promote negative energy expenditure. Given that icv injections of adiponectin increase energy expenditure

(Qi et al, 2004), we were surprised to find that plasma concentrations of TSH following central adiponectin injections were unchanged, an observation which ultimately correlates well with our electrophysiological data showing the TRH NE cells tested were also unaffected by adiponectin.

In contrast, PA TRH neurons were selectively depolarized by adiponectin. These neurons have been suggested to project to autonomic preganglionic neurons in the dorsal vagal complex and spinal cord and send information about core body temperature to peripheral targets

(Arancibia et al, 1996). Adipocyte derived adiponectin acting through PA TRH neurons may thus represent a vital connection coupling adiposity stores to central thermoregulation. Though an attractive hypothesis, further experiments will be needed to investigate this possibility.

Activation of PA OT neurons which project to the nucleus of the solitary tract has been shown to exert profound effects on feeding patterns (Blevins et al, 2004). The excitation of these

PVN neurons leading to the release of OT into the hindbrain promotes the sensitization of NTS neurons to satiety cues such as CCK, advancing meal termination (Blevins et al, 2003). The depolarizing effect of adiponectin on PA OT neurons would suggest that adiponectin may act centrally to control meal size. These effects would have profound consequences on feeding behavior and may contribute to the effects seen on weight loss when adiponectin is injected icv

(Qi et al, 2004). The effect of adiponectin on PA OT neurons may not be limited to alterations in

78

feeding behavior but may be heterogeneous in nature. Caudally projecting PA neurons have functionally diverse roles in the autonomic nervous system (Rinaman, 1998; Toth et al, 1999) and further experiments will be needed to examine other systems.

Intriguingly the depolarizing effects of adiponectin on type 2 PA OT neurons are in contrast to adiponectin‟s hyperpolarizing effects on PVN OT magnocellular neurons (Hoyda et al, 2007). While these are two functionally and anatomically independent systems, they share the same signaling molecule OT and apparently are differentially modulated by adiponectin. Taken together these data would indicate that the function of adiponectin in the PVN may not be to control the release of a specific peptide, but to coordinate diverse networks of neurons which may use the same signaling molecule but have unique targets and thus different effects in vivo.

Our data correlating adiponectin receptor expression with cell responsiveness confirm previous reports from our lab (Hoyda et al, 2007; Fry et al, 2006) showing that neurons expressing AdipoR1/R2, AdipoR1 alone, or AdipoR2 alone all respond to adiponectin. Similarly, all responding CRH neurons expressed at least one adiponectin receptor. These observations suggest that the response of the neuron to adiponectin may not be dictated by receptor expression but rather by the ion channel(s) expressed by the neuron and modulated by activation of receptor(s). Elucidation of these target ion channels will provide more information on the intracellular signaling mechanisms responsible for adiponectin induced neuronal excitability.

Several PA neurons that expressed TRH or OT and responded to adiponectin did not express mRNA for either adiponectin receptor. The response to adiponectin in these neurons however was maintained in TTX, suggesting an effect, independent of adiponectin receptor activation at the cell surface of these recorded neurons. Several possibilities could explain these observations. Leptin and have been shown to confer their central effects on feeding

79

through activation of the PVN endocannabinoid system, leptin specifically reducing excitatory synaptic drive to PVN neurons through an endocannabinoid dependent fast inhibitory feedback mechanism (Malcher-Lopes et al, 2006; Tucci et al, 2004). Whether adiponectin also signals through the endocannabinoid system is unknown at the current time but could represent a receptor independent mechanism of signaling in these neurons. A second possibility is that adiponectin could influence these neurons through the modulation of TTX insensitive dendritic release of neuromodulatory peptides. This system has been shown to act as an autocrine or paracrine messaging system and can modulate neuronal output within the PVN (Ludwig & Leng, 2006). A third explanation for the unique effect of adiponectin is the interaction with another receptor. The T-cadherin has been proposed to be a third receptor for circulating adiponectin; however, it is unlikely that activation of this receptor is responsible for the effects seen here as T-cadherin preferentially binds only high-molecular or hexameric forms of adiponectin (Hug et al, 2004). Finally, although our serial dilution controls have shown that both receptors can be detected in dilutions of whole PVN cDNA reflecting far less volume than a single cell (R1 - 1/10 000 000, R2 - 1/10 000 000), we cannot rule out the possibility that adiponectin receptors are expressed in these cells that our single cell RT-PCR technology cannot detect.

In conclusion we have shown that adiponectin, through actions at its two receptors

AdipoR1 and AdipoR2, in addition to potentially a third unique mechanism, controls the excitability of specific groups of neurons in the parvocellular regions of the PVN, depolarizing

NE CRH, PA TRH and PA OT neurons. Furthermore, through in vivo studies we have shown that when injected centrally adiponectin selectively controls the release of ACTH into the plasma while having no effect on TSH release. This study highlights the diverse central roles played by

80

adiponectin in controlling energy homeostasis and autonomic function and may provide insight into the pathophysiology of disrupted adiponectin signaling in diabetes and obesity.

ACKNOWLEDGEMENTS

This work was supported by grants from the Heart and Stroke Foundation (AVF),

National Institutes of Health (WKS) and a Canadian Institute of Health Research Doctoral

Research Award (TDH). We would like to thank Ms. Christie Hopf for her technical assistance with the single cell RT-PCR experiments.

81

Chapter 4: ADIPONECTIN ACTS IN THE NUCLEUS OF THE SOLITARY TRACT TO

DECREASE BLOOD PRESSURE BY MODULATING THE EXCITABILITY OF

NEUROPEPTIDE Y NEURONS

Ted D.Hoyda, Pauline M. Smith and Alastair V. Ferguson

Department of Physiology, Queen's University, Kingston, Ontario, Canada K7L 3N6

Brain Research. 2009, Vol. 1256:76-84

82

ABSTRACT

Adiponectin is an adipocyte derived hormone which acts in the CNS to control autonomic function, energy and cardiovascular homeostasis. Two 7-transmembrane domain receptors: AdipoR1 and AdipoR2, expressed in the hypothalamus and brainstem, mediate the actions of adiponectin. The medulla‟s nucleus of the solitary tract (NTS) is the primary viscerosensory integration site and an important nucleus in the regulation of cardiovascular function. Here we show the localization of both AdipoR1 and AdipoR2 mRNA in the NTS. We have investigated the consequences of receptor activation in response to exogenous application of adiponectin on cardiovascular (blood pressure and heart rate monitoring in vivo), and single neuron (whole cell current-clamp recordings in vitro) function. Microinjection of adiponectin in the medial NTS (mNTS) at the level of the area postrema resulted in a decrease in BP (mean

AUC= -2055 ± 648.1, n=5, mean maximum effect: -11.7 ± 3.6 mmHg) while similar commissural

NTS (cNTS) microinjections were without effect. Patch clamp recordings from NTS neurons in a medullary slice preparation showed rapid (within 200 seconds of application) reversible (usually within 1000 seconds following washout) effects of adiponectin on the membrane potential of

62% of mNTS neurons tested (38/61). In 34% (n=21) of mNTS neurons adiponectin induced a depolarization of membrane potential (6.8 ± 0.9 mV), while the remainder of mNTS cells influenced by adiponectin (n=17) hyperpolarized in response to this adipokine (-5.4 ± 0.7 mV).

Post-hoc single cell RT-PCR analysis of neurons showed that the majority of NPY mRNA positive mNTS neurons were depolarized by adiponectin (7/11), while 4 of these depolarized cells were also GAD67 positive. The results presented in this study suggest adiponectin acts in the NTS to control BP and suggest that such effects may occur as a direct result of the ability of this adipokine to modulate the excitability of discrete groups of neurons in the NTS. These

83

studies identify the mNTS as a new CNS site which adiponectin may act to influence central autonomic processing.

84

INTRODUCTION

The adipocyte hormone adiponectin is positively correlated with insulin sensitization, energy homeostasis and endocrine control through actions at two receptors: AdipoR1 and

AdipoR2, distributed throughout the CNS and peripheral tissue (Ahima & Lazar, 2008; Hoyda et al, 2007; Yamauchi et al, 2003; Kadowaki & Yamauchi, 2005; Fry et al, 2006). Additionally, much recent attention has focused on the cardiovascular effects of adiponectin treatment (Wang

& Scherer, 2008). Hypoadiponectinemia has been positively correlated with high risk for hypertension (Iwashima et al, 2004), whereas reconstitution of adiponectin through adenoviral in the genetically obese KKAy mouse ameliorates elevated systolic BP (Ohashi et al,

2006). Evidence suggests that adiponectin may influence cardiovascular homeostasis through actions on endothelial cells to promote local nitric oxide (NO) production (Tan et al, 2004), through interaction with the -angiotensin system (Furuhashi et al, 2003), or potentially as a consequence of its modulatory effects on the sympathetic nervous system through actions at sites in the central nervous system (Fry et al, 2006; Tanida et al, 2007).

Located in the medulla, the NTS is the principle integration site for cardiovascular and autonomic sensory afferents (Saper, 2002). Through its GABAergic interneurons, the NTS is a critical component of the baroreceptor reflex circuit that regulates sympathetic tone (Guyenet,

2006) and thus represents a potential site at which adiponectin could influence cardiovascular homeostasis. Therefore, using microinjection, current clamp electrophysiology and molecular techniques we have investigated the actions of adiponectin in the NTS and provide evidence to suggest this site represents a potentially crucial autonomic site for actions of this adipokine.

85

MATERIALS AND METHODS

Preparation of brainstem slices

All procedures conformed to the ethical guidelines established by the Canadian Council on Animal Care and were approved by the Queen‟s University Animal Care Committee. Brains from decapitated male Sprague Dawley rats (p21-28) were quickly removed and immersed in a high sucrose based ice cold slicing solution (2-4 °C) containing (in mM): 87 NaCl, 2.5 KCl, 25

NaHCO3, 0.5 CaCl2, 7 MgCl2, 1.25 NaH2PO4, 25 glucose, 75 sucrose bubbled with 95% O2/5%

CO2 for 3-5 minutes. Brains were then blocked, mounted onto a vibratome (Leica Microsystems,

Nussloch, Germany) and immersed in slicing solution bubbled with 95% O2 while 300µm slices containing NTS were cut and placed in 32° C artificial cerebral spinal fluid (aCSF) containing (in mM): 126 NaCl, 2.5 KCl, 26 NaHCO3, 2 CaCl2, 2 MgCl2, 1.25 NaH2PO4, 10 glucose saturated with 95% O2/5% CO2.

Preparation of DNA free mRNA for adiponectin receptor mRNA detection

All procedures involving molecular biology adhered to strict DNA-RNA contamination free protocol. Tissue sections containing exclusively mNTS tissue were microdissected out following slicing procedures (outlined above). NTS RNA was isolated from tissue and diluted a final volume of approximately 100µl solution according to instructions in the ethanol free

Ambion RNA isolation kit (Applied Biosystems, Toronto, ON, Canada). Isolated NTS RNA was then added to a reaction tube set up to run a DNase treatment with the following reagents provided by Fermentas (Burlington, ON, Canada): 10 µl 10X Reaction Buffer with MgCl2 and 10

U deoxyribonuclease I. The DNase reaction proceeded at 37°C for 30 min, then added to it 10

µM 25 mM EDTA and held at 65°C for 10 min to inactivate the enzyme and placed on ice. 10µl of DNA free RNA was used as a template for a reverse transcription reaction according to

86

Superscript III instructions (Invitrogen, Burlington, ON, Canada). A 50 µl reaction volume for the reverse transcription reaction included the following reagents: 50 ng random hexamer, 10 mM dNTP mix, 2µl 10 X RT buffer, 25 mM MgCl2, 0.1M DTT, 40 U RNaseOUT, 200 U Superscript

III RT heated to 50°C for 50 min then to 85 °C for 5 min. Two controls were run in addition to the reaction: one omitting the reverse transcriptase enzyme and one replacing NTS template with

DNA/RNA-free water. Following completion of the RT reaction, aliquots were stored at -80°C or prepared for a polymerization chain reaction (PCR).

PCR detection for adiponectin receptors in NTS

A multiplex PCR strategy (Qiagen, Mississauga, ON, Canada) was employed to detect

PCR products with primers specific for adiponectin receptor 1 and 2, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) which acted as a positive control and genomic DNA, which acted as a negative control. PCR detection for adiponectin receptors in NTS was run in triplicate according to the following protocol. Multiplex reactions were performed in 50µl volumes and contained

0.2µM of each primer (table 4.1), and reagents provided with the Qiagen kit. The reaction volume was denatured at 95°C for 15 minutes and then cycled 30 times according to the following temperature protocol: 30s at 94°C, 90s at 60°C and 90s at 72°C followed by one final extension of 72°C for 10 minutes. PCR products were run on a 2% agarose gel containing ethidium bromide, gel purified and sequenced to confirm their identity (Robarts Institute, London, ON,

Canada).

Microinjection

Urethane-anesthetized (1.4 g/kg) male Sprague Dawley rats (150–350 g) were fitted with a tracheal cannula (PE-205; Intramedic) to facilitate breathing and a femoral arterial catheter for the measurement of blood pressure (BP) and heart rate (HR). Animals were placed on a

87

Table 4-1. Primers used in RT-PCR experiments.

Sequences are always indicated in the 5′–3′ direction. Nested primer sets for AdipoR1 and AdipoR2 were used for tissue RT-PCR experiments (indicated by ).

88

feedback-controlled heating blanket for the duration of the experiment to maintain body temperature at 37°C. The animal was then placed in a stereotaxic frame with its head positioned vertically (nose down). A midline incision was made at the level of the obex to expose the dorsal surface of the medulla and a microinjection cannula (150 µm tip diameter; Rhodes Medical

Instruments) was positioned into commissural nucleus tractus solitarius (cNTS) or mNTS. After a minimum 2 min stable baseline recording was obtained, 0.5 μl of either 10-7 M adiponectin

(Phoenix Pharmaceuticals, Burlingame, CA, USA) or artificial cerebral spinal fluid (aCSF, vehicle control) was unilaterally microinjected into the region by a pressure driven 10 µl

Hamilton micro-syringe over 10 seconds and the effects on BP and HR assessed. Only one injection was made per animal. The concentration of adiponectin for microinjection was chosen to reflect circulating plasma concentrations which, in the male Sprague Dawley rat, is approximately 3µg/ml or 10-7 M (Yang et al, 2004).

At the conclusion of the experiment, animals were overdosed with anesthetic and perfused with 0.9% saline, followed by 10% formalin, through the left ventricle of the heart. The brain was removed and placed in formalin for at least 24 hours. Using a vibratome, 50 µm coronal sections were cut through the region of NTS, mounted, and cresyl violet stained. The anatomical location of the microinjection site was verified at the light microscope level by an observer unaware of the experimental protocol or the data obtained.

Analysis of blood pressure and heart rate data

Animals were assigned to one of two anatomical groups (cNTS or mNTS) according to the location of the microinjection sites. Animals with injection sites that were not wholly confined within either of these regions were excluded from further analysis. Animals with confirmed microinjection sites in mNTS were further divided according to whether adiponectin or

89

aCSF (vehicle control) was microinjected into the region. Normalized BP and HR data (mean baseline BP and HR data were calculated for 60 s before injection and subtracted from all data points before and after injection) were obtained for each animal 60 s before the time of microinjection (control period) until 180 s after microinjection. Area under the curve (AUC) (area between baseline and each BP and HR response) was calculated for each animal for the 180 s time period immediately after the injection, and the mean AUC for BP and HR responses for each group were then calculated. A one way analysis of variance (one way ANOVA) was used to determine whether BP and HR responses observed in response to adiponectin were different according to anatomical location of the microinjection site (cNTS vs mNTS adiponectin) or to substance administered (adiponectin vs aCSF).

Electrophysiology

Slices were placed in a recording chamber and continuously perfused with 30 – 32 °C oxygenated aCSF at a rate of approximately 2 ml/min. Neurons were visualized on an upright differential interface contrast microscopy system (Nikon, Tokyo, Japan). Whole cell current- clamp recordings were made using a Multiclamp 700B amplifier (Molecular Devices, Sunnyvale,

CA, USA) sampled at 5 kHz, filtered at 2.4 kHz from a Micro 1401 interface equipped with Spike acquisition and analysis software (Cambridge Electronic Devices, Cambridge, UK) and then stored for offline analysis. Electrodes were filled with an intracellular solution consisting of (in

2+ mM): 125 potassium gluconate, 2 MgCl2, 0.1 CaCl2 (calculated intracellular free Ca : 2.77 nM),

5.5 EGTA, 10 KCl, 2 NaATP and 10 HEPES (pH-7.2 with KOH). When filled with solution electrodes had resistances between 3-5 MΩ. Only those cells which had action potentials larger then 60 mV and a stable baseline membrane potential after 10 minutes were recorded from. Prior to experimentation a pulse protocol was used to classify neurons as either post-inhibitory rebound

90

(PIR), delayed excitation (DE) or neither (NON) (Vincent & Tell, 1997). After establishing 50 seconds of control baseline, adiponectin (10nM) was applied to the cell through bath perfusion. A cell was considered to be responsive to adiponectin if the membrane potential change measured in

50 second blocks for 500 seconds following the start of application changed more then 2 standard deviations of the baseline membrane potential. Neurons were excluded from analysis if they did not show a recovery towards baseline following removal of adiponectin from the bath.

Single cell RT-PCR

Following the completion of current clamp recording, a post-hoc single cell RT-PCR protocol was performed to examine the expression of molecular markers (see Table 4.1) as described previously (Price et al, 2008b). Briefly, gentle suction was applied to aspirate cytoplasm into the recording pipette. An outside-out patch was pulled sealing the cytoplasmic contents within the pipette. Cells which failed to form outside-out patches were eliminated from single cell RT-PCR analysis but included in membrane potential analysis. The contents of the electrode were expelled into a 0.5 mL PCR tube. The following components were added and a

DNase reaction was performed: (approximate concentrations) 10 U DNase I and 10 mM 10X reaction buffer with MgCl2 (Fermentas, Burlington, ON, Canada) and incubated at 37°C for 30 minutes. 2.5 mM EDTA was added and incubated at 65°C for 10 minutes to stop the DNase reaction. mRNA was reverse transcribed immediately after by adding dithiothreitol (26 mM), dNTPs (3mM), random hexamers (3 µM), MgCl2 (4 mM), RNase inhibitor (20 U) and superscript

II (100 U) (Invitrogen, Burlington, ON, Canada) and allowed to incubate overnight at 37°C. An additional reaction mixture was made up without (RT) and used as a negative control for genomic contamination. Samples were subsequently stored at -80°C until used in PCR detection.

91

A multiplex PCR protocol was used to amplify the cDNA produced in the RT reaction.

This protocol consisted of two different amplification steps: outside and nested. First, „outside‟ primer sets (0.2µM) were added to the cDNA along with reagents provided in the Qiagen multiplex kit (Qiagen, Mississauga, ON, Canada) to a 100 µl volume. The reaction mixture underwent a temperature protocol: 95°C for 15 minutes and then 20 cycles of 94°C for 30s, 60°C for 90s, and 72°C for 90s. The second step consisted of individual reactions for each gene of interest using „nested‟ primer sets. Reactions were performed in 50µl volumes consisting of

Qiagen Multiplex reaction reagents, 0.2µM of primers and 2µl of solution from the multiplex reaction to act as template. The reaction mixture was subject to 35 cycles of the same cycling protocol. PCR products from the nested reaction were run on a 2% agarose gel containing ethidium bromide and sequenced to confirm their identity (Robarts Institute, London, ON,

Canada). With each reaction a positive and negative control was performed in which primers were run against cDNA made from whole NTS (+ control) and in a reaction which PCR grade water replaced the cDNA template (- control). Experiments which did not pass either control were eliminated from the study.

Chemicals and Peptides

All chemicals used to make solutions were obtained from Sigma Pharmaceuticals (Sigma,

Burlington, Ontario, Canada). Salts used to make intracellular solution used in single cell RT-

PCR analysis were from molecular biology grad stock and were kept separate from salts used for other solutions. Adiponectin (recombinant human globular) was obtained from Phoenix

Pharmaceuticals (Belmont, CA, USA) reconstituted in distilled water and made up into working aliquots for experimentation.

92

RESULTS

The NTS contains mRNA for both AdipoR1 and AdipoR2

The NTS is a primary integration site for autonomic control and a site at which adiponectin could potentially act. We first examined if adiponectin receptors were expressed in

NTS tissue through the use of RT-PCR. Acutely dissociated slices containing medial NTS neurons were cut out and used to make cDNA (Figure 4.1 B). Primers designed to detect

AdipoR1 and AdipoR2 mRNA were used to probe mNTS cDNA and the resulting products of the

PCR reaction were visualized on an agarose gel. cDNA of both AdipoR1 and AdipoR2 was detected indicating adiponectin receptor mRNA is localized in the mNTS (Figure 4.1 A). cDNA of GAPDH was amplified along with the adiponectin receptor primers as a positive control and an RT (-) reaction was performed to verify the reactions contained no genomic DNA.

Adiponectin microinjected into the NTS decreases BP

A total of 15 animals were used in this study. Microinjection of 0.5 µl of 10-7M adiponectin into mNTS resulted in a clear decrease in BP (mean AUC = -2055.0 ± 648.1 mmHg*sec, n=5, mean maximum effect: -11.7 ± 3.6 mmHg), without a significant change in HR

(mean AUC = -40.9 ± 19.3 beats, n=5). These depressor effects showed a rapid onset (occurring within 15 seconds of adiponectin administration) with a peak decrease in BP occurring approximately 60 seconds after microinjection and lasting several minutes (at least 5 minutes) before returning toward baseline (see Figure 4.1 B, C). These effects were shown to be the result of peptide administration as microinjection of aCSF (vehicle control) was without effect on BP

(mean AUC = -7.3 ± 183.9 mmHg*sec, n=6, p < .01) or HR (mean AUC = -5.3 ± 15.9 beats, n=6). The effect of adiponectin microinjection into mNTS was also shown to be site specific as adiponectin administration into the immediately adjacent cNTS was without effect on BP (AUC =

93

Figure 4-1. Microinjections of adiponectin into the mNTS decreases BP. A) This agarose gel shows RT-PCR analysis of mNTS cDNA for AdipoR1 and AdipoR2. PCR products for GADPH are observed (positive control) while no products were observed for genomic cDNA (Genomic) or in the RT- lane, in which template has been omitted from the cDNA synthesis reaction. The photomicrograph to the right is a cresyl violet stained coronal section of the medulla outlining NTS (black line) highlighting the portion of mNTS removed for RT-PCR analysis (white line) (AP=Area Postrema, CC=central canal, 4thV=fourth ventricle). B) Raw BP (upper) and HR (lower) recordings showing the response to a bolus microinjection of adiponectin into mNTS in a single animal (time of injection is indicated by the arrow). The photomicrograph to the left shows the anatomical location of the microinjection site in mNTS (black circle) in this animal. C) Normalized mean BP (upper) and HR (lower) traces showing the response to adiponectin microinjection into mNTS (●, n=5) or cNTS (▼, n=4). Microinjection of vehicle (aCSF) into mNTS (, n=6) is also shown. D) These summary bar graphs show mean area under the curve for BP (top) and HR (bottom) in response to adiponectin microinjection in mNTS (blue bar) or cNTS (black bar) or to vehicle control (aCSF) microinjection into mNTS (red bar) (** indicates p<0.001 between mNTS adiponectin vs. mNTS aCSF or mNTS adiponectin vs. cNTS adiponectin, determined by repeated ANOVA).

94

-380.9 ± 96.0 mmHg*sec, n=4, p < .05) or HR (AUC = -13.6 ± 17.8 beats, n=4) (Figure 4.1 C,

D).

Adiponectin influences the membrane potential of NTS neurons

Current clamp recordings were made from 61 mNTS neurons (identified anatomically based on their position in the slice) and the effects of bath administration of 10 nM adiponectin was examined. Following the development of a stable control baseline membrane potential for a minimum of 50 sec, adiponectin was perfused on to these medullary slices at a rate of 1-2 ml/min.

Adiponectin influenced the majority of cells (38/61 affected), inducing both clear depolarizing

(mean 6.8 ± 0.9 mV, n=21) or hyperpolarizing (mean -5.4 ± 0.7 mV, n=17) effects on NTS neurons (Figure 4.2 A, B). Both depolarizing and hyperpolarizing effects were reversible upon removal of adiponectin from the bath and most cells recovered to control membrane potential within 1000 seconds. In some cells the depolarizing effects of adiponectin were accompanied by an initiation of action potentials (AP) or an increase in AP frequency (in previously active cells) which returned to baseline upon washout. Similarly cells that hyperpolarized in the presence of adiponectin decreased or ceased firing APs, an effect that was also reversible. Intriguingly, cells which depolarized reached peak effect faster than those that hyperpolarized in the presence of adiponectin (depolarization: normally within 200 seconds, hyperpolarization: normally 500 seconds) suggesting two independent signaling mechanisms with different time courses of activation. These observations suggest that there are at least two distinct adiponectin responsive cell populations within the mNTS, that adiponectin controls the excitability of a majority of these neurons and that these effects are potentially mediated through different intracellular signaling pathways.

95

Figure 4-2. Adiponectin's effects on the membrane potential of mNTS neurons.

Current clamp recordings showing two different medial NTS neurons A) depolarizing and B) hyperpolarizing to bath administration of 10 nM adiponectin (Black bar, scale bars: 10 mV and

50 sec) and recovery of membrane potential to control following the removal of adiponectin from the bath. Adiponectin influenced the excitability of a majority of neurons tested (38/61). C) A box-and-whisker plot indicating both depolarizing and hyperpolarizing responses seen in medial

NTS neurons: mean depolarization: 6.8 ± 0.8 mV, n=21, mean hyperpolarization: -5.4 ± 0.7 mV, n=17 (mean ± SEM).

96

Adiponectin effects are diverse among electrophysiologically defined NTS neurons

Considerable effort has been made to define the unique electrophysiological properties of cell groups in the NTS, studies, which have led to the classification of three different subpopulations of neurons defined according to the presence (in response to a large hyperpolarizing current pulse) of a delayed return to baseline and first spike (DE cells), a post inhibitory rebound (PIR cells), or neither of these characteristics (NON cells) (Vincent & Tell,

1997). Analysis of the effects of adiponectin on the membrane potential of DE (27% depolarize,

27% hyperpolarize, 46% unaffected, n=26), PIR (36% depolarize, 18% hyperpolarize, 46% unaffected, n=11) and NON (33% depolarize, 25% hyperpolarize, 42% unaffected, n=24) cells demonstrated heterogeneity in the response characteristics in each group (Figure 4.3 A). These data suggest that adiponectin actions on mNTS neurons are not correlated to the electrophysiological properties unique to the three cell groups.

NPY/GAD67 neurons in the NTS are depolarized by adiponectin

A number of different neurotransmitters and neuromodulators are expressed within the

NTS (Lawrence & Jarrott, 1996), each of which has a unique ability to shape the output characteristics of the nucleus. To assess whether adiponectin controls specific groups of chemically defined NTS neurons, following current clamp recordings we used post-hoc single cell RT-PCR to identify the presence of GAPDH, AdipoR1, AdipoR2, cocaine amphetamine related transcript (CART), 4 receptor (MC4R), glutamate (VGLUT), pro- opiomelanocortin (POMC), neuropeptide Y (NPY) and GAD67 (a GABAergic cell marker) mRNA in recorded neurons. Cytoplasm was obtained from a total of 41 mNTS neurons of which

37 cells showed expression of the housekeeping gene GAPDH indicating successful collection and amplification of cytoplasmic mRNA. AdipoR1 mRNA was found to be expressed in 17 of

97

Figure 4-3. Adiponectin depolarizes neurons expressing NPY and GAD67.

A) Scatter plot separating recorded neurons according to their electrophysiological fingerprint; DE, PIR and NON. The plot indicates that each classification group contains neurons that depolarize, hyperpolarize or do not respond. B) A scatter plot of ssRT-PCR results for the effect of 10 nM adiponectin on the membrane potential of cells that expressed GAPDH, only AdipoR1, NPY and GAD67. Green data points indicate a depolarizing response, red a hyperpolarizing response while black indicate no response to adiponectin. The four GABA cells which depolarized also expressed NPY and thus have been included twice in this graph. B) Current clamp recording of the membrane potential response to bath administration of 10 nM adiponectin (black bar) onto a cell whose chemical phenotype was assessed post hoc for GAD67, AdipoR1 and NPY by ssRT-PCR (gel inset). Scale bars: 10 mV, 50 sec.

98

these neurons (depolarization n=8, no response n=7, hyperpolarization n=2) while AdipoR2 was co-expressed with AdipoR1 in only one cell and never expressed independently. Of the 37

GAPDH positive cells a total of 11 of these identified neurons expressed NPY, and the majority

(7/11) of these NPY cells depolarized in response to adiponectin (7.1 ± 1.6 mV), while the remaining 4 NPY cells were unaffected (mean membrane potential change 0.0 ± 1.3 mV)(Figure

4.3 B, C). Five of the depolarized NPY cells also expressed mRNA for AdipoR1 suggesting depolarization of NPY neurons is mediated through activation of this receptor. Four NPY neurons also expressed GAD67 and all of these cells depolarized in response to adiponectin (7.1 ± 2.0 mV). In addition 3/4 NPY/GAD67 neurons expressed AdipoR1. One cell expressing GAD67 but not NPY did not respond. While our experiments also identified small populations of CART

(n=2), MC4R (n=3), VGLUT (n=1), and POMC (n=1), AdipoR1/AdipoR2 (n=1) neurons, these small subpopulations did not show any homogenous patterns of responsiveness to adiponectin.

These data suggest that adiponectin depolarizes populations of both NPY and GAD67/NPY cells in the NTS, a subpopulation which we speculate may be responsible for changes in BP caused by adiponectin microinjected into this nucleus.

DISCUSSION

Here we demonstrate mRNA for both adiponectin receptors to be localized in the NTS, and have provided evidence that adiponectin acts in the mNTS to decrease BP in the whole animal. Through current clamp recordings and single cell RT-PCR we have shown that adiponectin influences the membrane potential of mNTS neurons, depolarizing a subpopulation of mNTS neurons which express NPY and in some cases GAD67. A second subpopulation(s) of unidentified mNTS neurons which do not appear to express AdipoR2, VGLUT, MC4R, POMC

99

or CART are hyperpolarized by adiponectin. These results highlight the mNTS as a CNS site at which adiponectin potentially acts to control autonomic function and energy homeostasis.

These results are the first to identify the presence of adiponectin receptors in the NTS.

Our group and others have localized adiponectin receptors to central sites involved in autonomic control and feeding behavior including the paraventricular nucleus of the hypothalamus (Hoyda et al, 2007), the arcuate nucleus (Kubota et al, 2007) and the area postrema (Fry et al, 2006), a circumventricular organ adjacent to the NTS. Adiponectin has been shown to attenuate hypertension through interaction with the RAS-system (Furuhashi et al, 2003), at the endothelial level (Tan et al, 2004) and through actions in the CNS at the SCN (Tanida et al, 2007)(for review see (Wang & Scherer, 2008)). Additionally, our lab has reported hypertensive effects when microinjected into the area postrema (Fry et al, 2006). Intriguingly while microinjections of adiponectin into the mNTS caused a decrease in BP, injections of the peptide into the commissural NTS exhibited no effect on BP. While only one dose was used for microinjection studies, the clear hypotensive effects following mNTS and not cNTS injections suggest a site specific action for adiponectin in controlling BP. These results suggest that adiponectin plays diverse roles in regulating BP depending on the predominant site of action. Opposite effects of adiponectin on BP seen in two adjacent brainstem nuclei may be due to availability and concentration in vivo. Unlike the NTS, the area postrema is a circumventricular organ with fenestrated capillaries capable of readily sampling membrane impermeable substances from the blood (Price et al, 2008a). Blood brain barrier studies have suggested that adiponectin is unable to access the CNS (Spranger et al, 2006), although Qi et al have suggested that adiponectin can access sites in the hypothalamus to confer its effects on energy homeostasis (Qi et al, 2004).

Recent reports have indicated the presence of adiponectin in human cerebrospinal fluid albeit at

100

1000 fold lower concentrations (Kos et al, 2007; Kusminski et al, 2007) indicating that it is either derived in, or has access to, the CNS. Whether the difference in concentration seen by each nucleus is responsible for their opposite effects on BP is unknown at this time.

Adiponectin was also shown to have mixed effects on the membrane potential of NTS neurons, actions which we originally hypothesized might be associated with separate subpopulations of NTS neurons previously characterized by their electrophysiological fingerprint

(Vincent & Tell, 1997). This was not the case as both depolarizing and hyperpolarizing effects of adiponectin were observed in DE, PIR and NON cells of the mNTS. We next examined if differential responses may correlate with genetic phenotype of recorded mNTS neurons using single cell RT-PCR techniques. We were able to show that mNTS neurons which express NPY and AdipoR1 (and in some cases GAD67) represented a homogenous group the majority of which were depolarized by bath administration of adiponectin. Intriguingly NPY, while abundantly expressed in the NTS (Lawrence et al, 1998) is a well established hypotensive agent when injected into the brainstem nuclei (Tseng et al, 1989; Kubo & Kihara, 1990). The GABAergic interneuron also plays an important role in modulating BP regulation and baroreflex sensitivity in the NTS (McLean et al, 1996). Evidence suggests that activation of GABAA receptors mediate hypertensive responses following bilateral injections of GABA into the NTS (Catelli et al, 1987).

We propose that this neuron which co-expresses NPY and GAD67 could potentially act as an autonomous state-dependent modulator of BP based on firing rate and release patterns of small- synaptic vesicles (containing GABA) and large dense-core vesicles (containing NPY (Paquet et al, 1996)) from nerve terminals. Intriguingly the inefficacy of NPY in NTS seen in diabetes

(Dunbar et al, 1992) has been suggested to play a role in the development of hypertension

(Ergene et al, 1993; Lawrence & Jarrott, 1996). Indeed loss of adiponectin function is a key

101

contributor in the pathogenesis of type 2 diabetes (Hara et al, 2002; Scherer, 2006). Whether adiponectin proves to be an important link between NPY neurons, the NTS and development of diabetic associated hypertension remains to be seen.

The NTS is the primary integration site for information regarding gustatory and gastrointestinal information in the CNS. It projects to major autonomic regulation sites in the brainstem and hypothalamus including the AP, ARC, PVN and lateral hypothalamus and participates in controlling energy homeostasis and feeding behavior (Williams et al, 2006).

Adiponectin has recently been suggested to act in the arcuate nucleus to increase food intake through AdipoR1 and AMPK activation suggesting a role for activation of the NPY neurons in the ARC (Kubota et al, 2007). If a reciprocal feeding pathway exists in the NTS regarding the activation of NPY neurons and the induction of feeding, our results would suggest that adiponectin may also act in the NTS to induce a control over feeding behavior. Further studies will be needed to asses this hypothesis.

Our single cell RT-PCR analysis indicates that cells which hyperpolarize in response to adiponectin do not express AdipoR2, GAD67, NPY, VGLUT, CART, MCR4 or POMC. These data suggest that adiponectin may confer its effects on feeding and metabolism through inhibition of a chemically distinct group of neurons from the ones we have examined in this study. The lack of AdipoR2 mRNA in comparison to AdipoR1 mRNA expression in neurons is also intriguing.

RT-PCR results from tissue experiments indicate that both receptors are present in the nucleus however single cell RT-PCR studies suggest that with the exception of one cell; only AdipoR1 is present in neurons and thus mediates effects on neuronal excitability. Interestingly both depolarizing and hyperpolarizing effects of adiponectin are seen in neurons that express

AdipoR1. Previous studies have shown that activation AMPK, the major intracellular signaling

102

molecule targeted by AdipoR1 (Kubota et al, 2007; Yamauchi et al, 2003) can lead to both excitation or inhibition of membrane potential in a cell type specific manner (Wen et al, 2008;

Wyatt et al, 2007). The unique expression of ion channel complement in mNTS neurons and how they are affected by activated AMPK may be responsible for effects of adiponectin on the membrane potential of these cells. Alternately AdipoR2 may be expressed in supporting cells within the nucleus (endothelial and/or glial cells) and upon activation may contribute to signaling through paracrine actions on recorded neurons.

In conclusion we have shown that through the expression and activation of adiponectin receptors in the NTS, adiponectin decreases BP through controlling the excitability of neurons in the NTS potentially involving a NPY/GABA interneuron. Furthermore we have identified a novel excitatory effect of adiponectin on the orexigenic peptide NPY expressing neurons of the NTS.

These results highlight the growing role of adiponectin in controlling autonomic systems through actions at sites in the CNS and provide important information into the physiology with which this peptide acts.

ACKNOWLEDGEMENTS

This work was supported by grants from the Heart and Stroke Foundation (AVF) (PMS) and a Canadian Institutes of Health Research Doctoral Research Award (TDH).

103

Chapter 5: IONIC MECHANISMS RESPONSIBLE FOR ADIPONECTIN INDUCED

CHANGES TO THE EXCITABILITY OF PARVOCELLULAR PVN NEURONS

Ted D. Hoyda and Alastair V. Ferguson

Department of Physiology, Queen‟s University, Kingston, ON, Canada, K7L 3N6

104

ABSTRACT The adipocyte derived hormone adiponectin acts at two 7-transmembrane domain receptors, AdipoR1 and AdipoR2, present in the paraventricular nucleus of the hypothalamus

(PVN) to regulate energy homeostasis and endocrine function. Adiponectin depolarizes rat parvocellular pre-autonomic neurons that secrete either thyrotropin releasing hormone (TRH) or oxytocin (OT), and parvocellular neuroendocrine corticotropin releasing hormone (CRH) neurons leading to an increase in plasma adrenocorticotropic hormone (ACTH) concentrations while also hyperpolarizing a subgroup of neurons. In the present study we investigate the ionic mechanisms responsible for these changes in excitability in parvocellular PVN neurons. Patch clamp recordings of currents elicited from slow voltage ramps and voltage steps indicate that adiponectin inhibits non-inactivating delayed rectifier potassium current (IK) in a majority of neurons. This inhibition produced a broadening of the action potential (AP) in cells that depolarized in the presence of adiponectin. The depolarizing effects of adiponectin were abolished in cells pre-treated with tetraethyl ammonium (TEA) (0/15 cells depolarize). Slow voltage ramps performed during adiponectin induced hyperpolarization indicate the activation of voltage independent potassium current. These hyperpolarizing responses were abolished in the presence of glibenclamide (a KATP channel blocker) (0/12 cells hyperpolarize). The results presented in this study suggest that adiponectin controls neuronal excitability through the unique modulation of different potassium conductances, effects which contribute to changes in excitability and action potential profiles responsible for peptidergic release into the circulation.

105

INTRODUCTION Central regulation of energy homeostasis and feeding behavior relies on the critical integration of peripheral information with neuronal networks in the hypothalamus. This information describing nutritional intake, storage and use is transmitted from peripheral organs by signaling molecules which function in the central nervous system (CNS) to coordinate the body‟s response to macronutrient intake (Chaudhri et al, 2008). Adiponectin is a 244-amino acid peptide hormone which is synthesized and released from adipocytes at levels varying inversely with body adiposity (Scherer et al, 1995; Hu et al, 1996; Maeda et al, 1996). Adiponectin was originally identified as an insulin-sensitizing hormone and was shown to play important roles in controlling glucose homeostasis and fatty-acid catabolism through its action in metabolic tissues such as skeletal muscle and the liver (Kadowaki & Yamauchi, 2005). Adiponectin deficient animals show insulin resistance, glucose intolerance and hyperlipidemia, whereas reintroducing the peptide ameliorates these dysfunctions (Kadowaki & Yamauchi, 2005; Maeda et al, 2002; Kubota et al,

2002). When introduced into the brains of rodents, adiponectin has been shown to display a wide variety of effects. In rats, the arcuate nucleus (ARC) has been shown to mediate adiponectin‟s induction of feeding through the activation of adenosine mono-phosphate activated protein kinase

(AMPK) and Akt phosphorylation (Kubota et al, 2007) whereas in mice, injections of adiponectin into the lateral ventricle are associated with a reduction in body weight without a preceding decrease in food intake (Qi et al, 2004). This effect of adiponectin is most likely mediated through activation of paraventricular nucleus (PVN) neurons as a robust induction of c-fos activation in this nucleus accompanied these injections (Qi et al, 2004).

Two receptors, AdipoR1 and AdipoR2, through which adiponectin confers its effects have been cloned (Yamauchi et al, 2003) and localized to distinct nuclei in the CNS including the

106

area postrema (Fry et al, 2006), the nucleus of the solitary tract (NTS) (Hoyda et al, 2009b), the

ARC (Kubota et al, 2007) and the PVN (Hoyda et al, 2007). In addition to its effects on energy homeostasis our lab has recently demonstrated an increase in peripheral circulation of ACTH concentrations following icv injections of adiponectin into rats, and an associated increase in the excitability of CRH expressing neurons in the PVN, which in turn are responsible for controlling

ACTH release from anterior pituitary cells (Hoyda et al, 2009a).

The PVN is a bilateral nucleus surrounding the 3rd ventricle and is a critical integrator of information regarding central neuroendocrine and autonomic function (Swanson & Sawchenko,

1983). The PVN contains three distinct groups of neurons, type I magnocellular, type II parvocellular pre-autonomic and type III parvocellular neuroendocrine neurons. In addition to adiponectin‟s depolarizing actions on CRH expressing neurons, the peptide also has been shown to depolarize pre-autonomic TRH neurons, pre-autonomic OT neurons (Hoyda et al, 2009a) and a subset of magnocellular vasopressin neurons while hyperpolarizing magnocellular OT neurons

(Hoyda et al, 2007) and a subgroup of unidentified parvocellular neurons (Hoyda et al, 2009a).

While these previous studies have shown clear effects of adiponectin on the membrane potential of specific groups of PVN neurons, the underlying mechanisms responsible for these changes in excitability are unknown. This study was therefore designed to investigate the ionic mechanisms through which adiponectin elicits these effects on the neuronal excitability of parvocellular PVN neurons through the modulation of specific ion channels. Using patch clamp techniques we provide the first evidence of ionic current modulation by this adipokine in the CNS, supporting the concept that this mediates adiponectin‟s effects on autonomic function.

107

MATERIALS AND METHODS Procedures involving the use of animals in this study were approved by the Queen‟s

University Animal Care Committee and conformed to the standards of the Canadian Counsel on

Animal Care.

Preparation of hypothalamic slices

Hypothalamic coronal slices (300 µm) containing PVN were taken on the day of the recording from male P19-26 Sprague Dawley rats (Charles River, Quebec, Canada) maintained on a 12 hr light/dark cycle and provided water and food ad libitum. Rats were immobilized, decapitated and their brains quickly removed and placed in 4ºC slicing solution consisting of (in mM) 87 NaCl, 2.5 KCl, 25 NaHCO3, 0.5 CaCl2, 7 MgCl2, 1.25 NaH2PO4, 25 glucose and 75 sucrose, bubbled with 95/5% O2/CO2 for 3-5 minutes. Brains were then trimmed to size, mounted on a stage immersed in slicing solution and coronal slices containing the PVN were cut using a vibratome (Leica, Nussloch, Germany). Slices were incubated in a 95/5% O2/CO2 holding chamber containing artificial cerebral spinal fluid (aCSF) consisting of (in mM) 126 NaCl, 2.5

KCl, 26 NaHCO3, 2 CaCl2, 2 MgCl2, 1.25 NaH2PO4, 10 glucose for at a minimum of 1 hr prior to recording.

Electrophysiological recordings

PVN neurons were submerged in aCSF, anchored to the bottom of the recording chamber and visualized using a 40X water-immersion objective mounted to an upright Nikon E600FN microscope outfitted with infrared differential interference contrast optics (Tokyo, Japan).

Electrodes were fabricated from borosilicate glass (World Precision Instruments, Sarasota FL,

USA) using a micropipette puller (P97 Sutter instruments Co., Novato CA, USA) and had resistances of 2-3 megaohms. When filled with internal recording solution consisting of (in mM)

108

+ 2+ 130 K -gluconate, 10 KCl, 2 MgCl2, 0.1 CaCl2 (calculation intracellular free Ca : 2.77 nM), 5.5

EGTA, 10 HEPES, 2 NaATP with pH adjusted to 7.2 using KOH. Whole cell recording configurations were made by applying slight negative pressure on to the end of the patch pipette following the formation of a gigaohm seal, breaking through the cell membrane and obtaining continuity with the internal contents of the cell. Current clamp data were obtained with a

Multiclamp 700B patch clamp amplifier (Molecular Devices, Palo Alto, CA, USA) recorded in real-time using the acquisition software Spike 2 (version 6) and Signal (version 4) (Cambridge

Electronic Devices, Cambridge, UK). Data were acquired at 5 kHz and filtered at 10 kHz using a

Micro1401 mkII interface (Cambridge Electronic Devices, Cambridge, UK). Data were stored for off-line analysis. Capacitance transients and series resistance error was minimized at the start of all recordings. Parvocellular PVN neurons were identified as either pre-autonomic (type II) or neuroendocrine (type III) neurons based on their distinct electrical profiles according to Luther et al (2002) (Luther et al, 2002). After establishing a stable control membrane potential for at least

100 seconds, an experiment specific peptide or drug, dissolved in aCSF was bath applied at a rate of 2 ml/min through gravity fed perfusion. In current clamp recordings, responsiveness of neurons to treatment was assessed by averaging all points in time blocks of 100 seconds for up to

1000 seconds following initial application of treatment. The maximum mean voltage change was compared to baseline control mean voltage and was counted as responsive if the value was greater than two times the standard deviation value of the control membrane potential. Following washout of the treatment, cells were recorded until membrane potential returned to baseline values. A calculated junction potential of 14.4 mV was subtracted from the membrane potential of recorded neurons. Cells which did not show at least a partial recovery towards baseline control membrane potential following drug treatments were omitted from the analysis.

109

Analysis of AP‟s was performed by signal averaging all AP over a given time frame and meaning the data points in time to provide an average profile. Signal averages of APs before and after adiponectin were normalized to the peak of the rising phase and the trough of the repolarization phase. Comparison in the AP repolarization times before and after adiponectin were made by averaging data points taken every 0.6 msec from the peaks of the normalized APs.

Signal version 4 was used for stimulation and data acquisition in all voltage clamp experiments.

Series resistance was routinely compensated up to 50% in these neurons. Peak steady state outward potassium current of the delayed rectifier subtype was measured by averaging all current data points elicited during the final 100 msec of the voltage step (grey bar, Figure 5.1 C). Slow voltage ramps (12 mV/s) were run in most cells from -105 mV to 20 mV. Difference currents were obtained from subtracting control currents from currents elicited following treatment with adiponectin. When used for regression analysis, difference current data points were obtained from averaging 100 data points surrounding the voltage of interest.

Chemicals and peptides

With the exception of those listed below all chemicals used to make solutions in these experiments were purchased from Sigma Pharmaceuticals (Oakville, Ontario, Canada).

Tetrodotoxin (TTX) was purchased from Alomone Labs (Jerusalem, Israel). Lyophilized recombinant human globular adiponectin was purchased from Phoenix Pharmaceuticals

(Burlingame, CA, USA), reconstituted in filtered molecular grade water and made in 20 µM aliquots for daily use. Glibenclamide was purchased from Tocris Biosciences (Cedarlane

Laboratories, Burlington, ON, Canada).

110

Statistical Analysis

Data obtained in these experiments was statistically analyzed using Graph Pad Prism version 5.1 (San Diego, CA, USA).

RESULTS Slow voltage ramps (-100 mV to 20 mV, 12 mV/s) were performed on identified parvocellular PVN neurons in the absence and presence of adiponectin to assess the peptide‟s effect on whole-cell current properties. Recordings were obtained from 15 neurons (6 type II pre- autonomic and 9 type III neuroendocrine) and grouped according to similar „difference current‟

(adiponectin – control current) profiles. The predominant response observed (7/15 neurons) showed a distinct difference current phenotype indicating a large decrease in outward current seen towards the end of the slow voltage ramp (Figure 5.1 A, B).

+ Adiponectin inhibits delayed rectifier K current (IK) in parvocellular PVN neurons

Voltage-gated K+ conductances have been shown to be important for neuronal excitability and modulation of these currents can have significant effects on the membrane potential of neurons (Hille, 1992). Therefore, based on the adiponectin induced decrease in outward current observed in the depolarized range of our voltage ramps, we proceeded to investigate the specific ionic conductance that was responsible for this effect. Whole cell voltage clamp recordings were made from a total of 13 parvocellular neurons within the PVN. A pulse- protocol was used to isolate the non-inactivating delayed-rectifier current (IK) from fast- inactivating outward potassium current (IA) by subjecting cells to a 1 sec pre-pulse to -40 mV followed by 500 msec voltage steps in increments of 10 mV from -80 mV to 20 mV (Figure 5.1,

C, inset). This protocol avoids the activation of IA and therefore allows direct measurement of IK

(Washburn et al, 1999). These recordings were made in the presence of 1 µM TTX to inhibit

111

Figure 5-1. Adiponectin inhibits a voltage gated potassium conductance. A) Representative traces of currents elicited from slow voltage ramps (12 mV/s) starting from a holding potential of -100 mV to 20 mV (inset). Black current traces represent current before the bath application of 10 nM adiponectin whereas green traces represent currents after. Currents shown are an average of 3 ramps (Scale bars: 20 mV, 100 pA). B) Difference current obtained from subtracting currents obtained during control (black) from adiponectin-induced currents (green). C) Voltage clamp currents obtained from performing 500 msec voltage steps (-80 mV to 20 mV) following a 1 sec conditioning pulse to -40 mV from a holding potential of -60 mV (step protocol inset, Scale bars: 250 msec, 20 mV) in the absence (left) and presence (right) of 10 nM adiponectin. Grey boxes represent the data points used to measure average steady-state current (Scale bars: 150 pA, 100 msec). D) I-V Plot of normalized steady-state current, averaged across all cells tested at each voltage step in control conditions (filled black boxes) and following treatment with bath application of 10 nM adiponectin (filled green boxes). Significant differences in steady-state current are indicated by asterisks (*p<0.05, **p<0.005, paired t-test).

112

sodium currents. Following the establishment of control IK amplitudes, the effect of bath administration of 10 nM adiponectin on sustained outward IK was assessed. Of the 13 recorded neurons, 9 cells were electrophysiologically assessed as type 3 neuroendocrine cells and adiponectin decreased steady-state IK in 7 of these cells (Figure 5.1 C). Examination of the normalized I-V relationship in the absence and presence of adiponectin suggested that the delayed rectifier IK is statistically different from control beginning at the -10mV voltage step and the effects of adiponectin become progressively larger up to the 20mV peak voltage step which shows a decrease of 18 ± 4% (n = 7, *p<0.05, paired t-test) (Figure 5.1 D). Steady state IK was unaffected by adiponectin in one cell and increased in the other neuroendocrine cell. The remaining 4 of 13 neurons were electrophysiologically identified as pre-autonomic parvocellular neurons. Adiponectin had mixed effects on the amplitude of steady-state IK with two cells showing an increase in IK at the 20 mV step (22 ± 5%, n = 2) while two neurons showed no change (2 ± 2%, n = 2) (data not shown).

Adiponectin modulates action potential shape in cells that depolarize

We next examined if adiponectin‟s effect on IK induce a change in the shape of the action potential, as would be predicted, in parvocellular neurons which depolarized in response to the peptide. Analysis of action potential shape were made on neurons from a previously developed database of adiponectin responsive cells (Hoyda et al, 2009a) and selected based on the criteria of having spontaneous action potentials during baseline control membrane potential recordings, and a depolarized response concurrent with an increase in action potential frequency following 10 nM adiponectin administration (Figure 5.2 A, i). Six neurons fulfilled these criteria and were examined for changes in action potential shape in response to adiponectin induced depolarization.

Current clamp recordings of action potentials were signal averaged and normalized to peak

113

Figure 5-2. Action potential broadening accompanies the depolarizing effects of adiponectin. Ai) Current clamp recording representing a increase in excitability and depolarizing response to bath application of 10 nM adiponectin (blue bar) along with recovery to baseline following washout (Scale bars: 50 sec, 10 mV). Grey hatched bar indicates control membrane potential before adiponectin treatment. α and β represent regions of current clamp recording used for AP shape analysis. ii) Overlay of normalized action potentials for the current clamp trace on the left, recorded during the control phase (average of all spikes in α (black line)) and following treatment with 10 nM adiponectin (average of all spike in β (green line)) (Scale bars: 2 msec). Note: AP averages were normalized to the peak of the AP and the tough of the repolarization phase. AP profiles following adiponectin treatment reflect initiation from depolarized potentials compared to control. B) Plot of normalized APs averaged over 6 neurons comparing AP profiles before (control: black) and after depolarization to bath application of 10 nM adiponectin (adiponectin: green). Hatched bars represent standard error measurement of the neurons tested in each respective color. C) Plot showing normalized repolarization rate of APs in control phase (black) versus adiponectin treatment (green). Inset: Bar graph showing the calculated AUC for control (black) compared to adiponectin repolarization (green) (**p<0.005).

114

amplitude providing an average action potential shape prior to and following adiponectin treatment (Figure 5.2 A, ii). Adiponectin induced significant spike broadening (Figure 5.2 B), increasing the time course of recovery from the peak of the action potential through the repolarization phase (Figure 5.2 C) (inset: area under the curve (AUC) = adiponectin: 179.8 ±

23.6 mV*msec vs control: 130.5 ± 17.7 mV*msec, n = 6, **p<0.01 paired t-test). A significant difference in normalized voltage of the repolarization phase was first seen at 0.6 msec after the peak of the action potential (control: 67.9 ± 5.0%, n=6 vs. adiponectin: 78.7 ± 4.7%, n = 6, p<0.005, paired t-test) and continued to be significantly different for 4.2 msec (Figure 5.2 C).

These results suggest that during depolarization adiponectin acts to broaden AP width through the inhibition of IK.

Depolarizing effects of adiponectin are blocked in TEA

Based on the observed inhibition of the outward rectifying IK by adiponectin we next tested the hypothesis that depolarizing effects of adiponectin result from inhibition of IK, by assessing membrane potential responses of parvocellular neurons to bath administration of 10 nM adiponectin in the presence of the potassium channel blocker TEA (Figure 5.3 A, i-iii). Current clamp recordings were obtained from 15 parvocellular neurons. Administration of TEA alone induced a depolarizing shift (≥ 2 mV) in the membrane potential of 14/15 neurons (example

Figure 5.3 A, i) (mean depolarization: 6.4 ± 4.1 mV, n = 14, Figure 5.3 B, inset) suggesting baseline membrane potential in parvocellular neurons is in part mediated by a TEA sensitive current. While in the presence of TEA, a stable membrane potential was obtained for a minimum of 50 seconds (using small negative holding current in 5 cells), and 10 nM adiponectin was perfused on to the slice at a rate of 2 ml/min (Figure 5.3 A, i - iii). No depolarizing effects of adiponectin were observed (0 of 15 cells tested - Figure 5.3 B, C) in the presence of either 30 or 3

115

Figure 5-3. Adiponectin induced depolarization is abolished in TEA. A) Current clamp recordings showing the effect of bath application of 10 nM adiponectin (blue bar) in the presence of 30 mM TEA (grey bar). Blue hatched lines in i), ii) and iii) represent the baseline membrane potential used to compare against the effects of adiponectin. Current clamp recordings in i) and ii) show no response to adiponectin treatment while iii) shows a hyperpolarizing response to the peptide (Scale bars: 10 mV, 100 sec). Note in i) the grey hatched bar represents baseline membrane potential before the bath application of 30 mM TEA followed by a depolarization to the new baseline potential (blue bar). B) Scatter-dot plot of the response amplitude to adiponectin observed in the presence of either 30 mM or 3 mM TEA compared to aCSF. Green filled circles indicate a depolarizing response, black filled circles are non-responders while red filled circles are cells that hyperpolarized. Inset: Box and Whisker plot of the effect of TEA alone on the membrane potential of parvocellular neurons. C) Bar graph showing the proportion of neurons responding to adiponectin in the presence and absence of bath administration of 10 nM adiponectin. Red bars indicate a hyperpolarizing response, black bars indicate no response while green bars indicate a depolarizing response.

116

mM TEA, although hyperpolarizing responses to the peptide were still observed in a small number of cells tested (4/15). We performed additional experiments in which we tested the effect of 10 nM adiponectin in the presence of 100 nM iberiotoxin used to block large-conductance calcium activated potassium channels (BKCa). Both depolarizing (9.1 ± 1.7 mV, 2/6) and hyperpolarizing (-3.5 ± 0.3 mV, 2/6) responses were maintained in the presence of the potassium channel blocker, suggesting BKCa is not involved in mediating the effects of adiponectin. These results support the conclusion that the depolarizing effects of adiponectin are mediated through the inhibition of a TEA sensitive current whereas the hyperpolarizing effects are not.

Adiponectin hyperpolarizes PVN neurons through modulation of a voltage independent K+ conductance

We have shown previously that adiponectin hyperpolarizes and inhibits AP firing in a subpopulation of parvocellular PVN neurons (Hoyda et al, 2009a). To investigate the ionic mechanisms responsible for the inhibitory effects of adiponectin we utilized a current clamp and voltage clamp technique which allows us to examine both the underlying currents and membrane potential response to bath application of 10 nM adiponectin. The recording protocol consists of first voltage clamping neurons and performing slow voltage ramps from -105 to -70 mV (12 mV/s) to obtain a profile of the baseline currents. Next the recording is switched to current clamp to record membrane potential and the activity of the cell. Bath application of 10 nM adiponectin is administered in the current clamp configuration and the response is observed. Neurons are once again voltage clamped and the same voltage ramps are performed to measure resultant currents in the presence of adiponectin (Figure 5.4 A, i, ii). This technique was successfully performed on four neurons that hyperpolarized in response to adiponectin. Linear regression analysis of the grouped difference currents (adiponectin – control) for these cells confirms a current reversal of -

117

Figure 5-4. Adiponectin activates a potassium conductance in hyperpolarized neurons. A) Representative current clamp/voltage clamp recording from the same neuron. i) Current-clamp recording of a parvocellular neuron hyperpolarizing in the presence of bath administration of 10 nM adiponectin (blue bar). α indicates control voltage-clamp recording as shown in ii) where as β represents period of time performing the same voltage clamp protocols in the presence of 10 nM adiponectin. Time spent in voltage- clamp configuration during β is indicated by the hatched bars (Scale bars: 10 mV, 25 sec). ii) Representative currents obtained during control (α – black) and following adiponectin induced hyperpolarization (β – red) from the current-clamp recording in i). Currents are an average of three (5-point smoothed) ramps performed during the times indicated. Inset: voltage clamp ramp protocol used to elicit the currents shown. B) Graph showing the mean average difference current obtained from subtracting control currents from adiponectin induced currents. Plotted points are an average of 100 difference current data points around the voltage of interest averaged from all cells tested (mean ± SEM). Linear regression (black hatched bar) intersects when Y=0 at -108.4 mV.

118

108.4 mV for the adiponectin induced current (Figure 5.4 B), a value near the equilibrium potential for potassium (-105 mV) calculated using the Nernst equation, where [K]o = 2.5 mM and [K]i = 140 mM. These data suggest adiponectin hyperpolarizes cells through the activation of a voltage independent potassium current.

Adiponectin induced hyperpolarization is blocked in glibenclamide

Based on these data showing the adiponectin induced difference current reversed at the equilibrium potential for potassium, we hypothesized that hyperpolarizing effects of adiponectin resulted from the modulation of the ATP sensitive potassium conductance KATP. We therefore undertook additional current clamp experiments to determine if pretreatment with glibenclamide

(a KATP channel inhibitor) abolished hyperpolarizing effects of adiponectin on PVN parvocellular neurons. Recordings were obtained from 12 parvocellular neurons and examined first for their response to bath application of 500 nM glibenclamide, which depolarized 6/12 neurons (control: -

51.5 ± 1.4 mV vs. glibenclamide: -46.02 ± 0.4 mV, n = 6, paired t-test, p<0.005) suggesting that in a population of PVN neurons, this conductance contributes to resting membrane potential.

Following pretreatment with glibenclamide, a stable baseline was obtained for a minimum of 100 seconds prior to bath administration of 10 nM adiponectin (Figure 5.5 A, i, ii). No hyperpolarizing responses to adiponectin were observed in this group of 12 neurons, with depolarizing responses observed in 7/12 cells (7.4 ± 1.8 mV, n = 7) and 5 cells being unaffected by bath administration of adiponectin (-0.2 ± 0.9 mV, n = 5) (Figure 5.5 B, C). The fact that no hyperpolarizing responses to adiponectin were observed in the presence of glibenclamide (0/12 neurons) supports the conclusion that a glibenclamide sensitive current (likely KATP) is responsible for adiponectin induced hyperpolarization in parvocellular neurons on the PVN.

119

Figure 5-5. Glibenclamide inhibits adiponectin induced hyperpolarization.

A) Representative current clamp traces showing a parvocellular neuron that depolarizes (i) and parvocellular neuron that does not respond (ii) to bath application of 10 nM adiponectin (blue bar) in the constant presence of 500 nM glibenclamide (black line) (Scale bars: 20 mV, 50 sec). B) Scatter-dot plot showing the response to adiponectin in the presence of 500 nM glibenclamide compared to aCSF. Green data points indicate a depolarizing response, black points are those that did not respond and red indicate a hyperpolarizing response to adiponectin. C) Histogram showing the proportion of neurons that depolarize (green), do not respond (black) or hyperpolarize (red) in response to adiponectin in aCSF (adiponectin) or when the KATP channel is inhibited by bath application of glibenclamide (+ glibenclamide).

120

DISCUSSION In this study we present evidence showing that the effects of adiponectin on the excitability of parvocellular PVN neurons are mediated through the differential modulation of specific potassium currents. Specifically, adiponectin depolarizes and increases excitability in a subpopulation of PVN neurons through the inhibition of a TEA sensitive potassium current, effects that are accompanied by changes in the shape of the AP. Adiponectin‟s inhibitory effects on PVN neurons are mediated through the activation of a glibenclamide sensitive leak potassium conductance. These experiments describe new and potentially critical roles that these ion channels play in mediating the effects of adiponectin on parvocellular neurons in the PVN and identify these channels as critical final pathways through which adiponectin exerts control over energy homeostasis and feeding behavior, through actions in the CNS.

Potassium currents make important contributions to the regulation of neuronal excitability, firing patterns of individual neurons and AP shape (Hille, 1992). The inhibition of sustained outward potassium conductance observed through voltage clamp recordings, both slow voltage ramps and voltage step protocols, suggested that adiponectin‟s effects on these channels would also be reflected by changes in AP shape when examined in current clamp. Analysis of current clamp data obtained from our previous study examining the effects of adiponectin on the excitability of PVN neurons (Hoyda et al, 2009a) suggests that AP shape is significantly different in cells which depolarize in the presence of adiponectin. Normalized action potentials showed a significantly prolonged duration when compared to control confirming the inhibitory effects on potassium current manifested as changes in AP profile. Changes in the kinetics of potassium currents responsible for the repolarization of the AP can have profound consequences at the synaptic terminal, inducing more inward calcium current and increasing the efficacy of synaptic

121

transmission in central synapses (Geiger & Jonas, 2000). Spike broadening in neurohypophysial neurons has been shown to be intimately associated with frequency-dependent facilitation at pre- synaptic terminals increasing release rates of synaptic vesicles containing neurotransmitters

(Jackson et al, 1991). The inhibition of IK by adiponectin may act as a peptide specific mechanism to favor the exocytosis of large dense core vesicles containing hypothalamic peptides to be released into the hypophyseal portal circulation. In addition to the changes in AP profile observed in cells excited by adiponectin, we examined to what effect pretreatment with TEA, a blocker of IK, would have on the responsiveness of cells to adiponectin.

TEA is a quaternary ammonium cation which is commonly used to block delayed rectifier potassium currents. Treatment with TEA is not commonly associated with the modulation of resting membrane potential in a quiescent neuron, however the cation blocker has been known to block other channels and reveal previously inhibited currents in excitable cells

(Hille, 1992). At concentrations used in this study, TEA has been shown to block a number of other currents including NMDA-current (Sobolevsky, 1999) and the transient potassium A- current (Adams et al, 1980) as well as a variety of others (Kavanaugh et al, 1991). Pre-treatment with TEA (30 mM or 3 mM) had a depolarizing effect on the membrane of most neurons examined in this study and in some cases, un-masked a resurgent depolarizing oscillation most likely carried through voltage-gated calcium channels as originally described in the crayfish muscle AP (Fatt & Katz, 1953). This resurgent current was abolished when negative current was injected into the cells to hold membrane potential at sufficiently hyperpolarized values to avoid calcium channel activation. The depolarizing effects of TEA were seen in some neurons which were quiescent upon application, suggesting TEA inhibits a channel that contributes to resting conductance. Our experiments suggest that the effects of adiponectin are also maintained in the

122

presence of the BKCa blocker iberiotoxin suggesting this conductance does not contribute to the adiponectin induced changes in excitability in these cells. While hyperpolarizing responses to adiponectin were still observed in TEA, the depolarizing responses were abolished altogether suggesting this TEA sensitive current mediates the depolarizing effects of adiponectin in this population of neurons.

In contrast our data suggest that adiponectin hyperpolarizes neurons through the activation of a glibenclamide sensitive potassium conductance. Glibenclamide is a membrane permeable sulfonylurea which is used clinically to treat type 2 diabetes because of its specific inhibition of KATP channels; depolarizing pancreatic beta cells and inducing the release of insulin

(Ashcroft, 1988). KATP channels are heteromultimers of inwardly rectifying potassium channel subunits (Kir6x) and sulfonylurea receptors (SUR) (guilar-Bryan et al, 1998), and are expressed in a variety of tissues including the CNS (Ashcroft, 1988). These channels are inhibited by increasing concentrations of intracellular adenosine 5‟-triphosphate independent of membrane potential (Ashcroft, 1988), and have been shown to be critical for the sensing of glucose in neurons of the ventromedial hypothalamus (Miki et al, 2001). Recent microarray analysis indicates that mRNA encoding genes for Kir6.2, sur1 and sur2 are present in the PVN

(Hindmarch et al, 2006), suggesting that these cells have the capacity to form functional KATP channels. Glibenclamide inhibits this channel by binding the sulfonylurea receptor binding pocket on the SUR1 (high affinity block) or SUR2 (low affinity block) moiety (Bryan et al, 2004). That glibenclamide can block the hyperpolarizing effects of adiponectin suggests that the second messenger pathways elicited through either AdipoR1 or R2 receptor activation provides an inhibiting factor which also binds this SUR pocket, albeit with less affinity than glibenclamide.

The specific nature of this second messenger molecule is unknown at the current time.

123

Adiponectin has been shown to stimulate phosphorylation of AMP-activated protein kinase (AMPK) in peripheral and CNS tissue through AdipoR1 activation (Yamauchi et al, 2002;

Yamauchi et al, 2003; Kubota et al, 2007). Activated AMPK transitions a cell from energy consuming anabolic processes to catabolic pathways which generate ATP (Kahn et al, 2005).

Elevation in cellular ATP concentrations would be expected to have an inhibitory effect on KATP channels and therefore potentially contradict the results obtained in these experiments. However

Yamauchi et al (2002) suggest that maximal phosphorylation of AMPK occurs at 15 minutes following exposure to adiponectin in C2C12 myocytes (Yamauchi et al, 2002), whereas our results show the effect on cell membrane excitability usually occurs within 200 seconds of bath application of the peptide. This suggests that the modulation of ion channels occurs prior to the phosphorylation of AMPK. Increasing ATP levels following AMPK activation may even provide an internal tempering mechanism to cell inhibition by reversing the adiponectin induced activation of KATP channels in parvocellular neurons.

In conclusion, we have shown that adiponectin controls the excitability of parvocellular

PVN neurons through the unique modulation of potassium currents. Specifically, we have demonstrated that adiponectin induced depolarization proceeds through the inhibition of a TEA sensitive current and that increases in excitability are associated with a broadening of the AP.

Conversely the adiponectin induced hyperpolarizing responses are due to the activation of a glibenclamide sensitive current. This study highlights, for the first time, conductances responsible for conferring adiponectin‟s effects on CNS neurons controlling energy homeostasis and autonomic functions and may provide specific targets for therapeutic development concerning the pathophysiology of disease states following the disruption of adiponectin signaling.

124

ACKNOWLEDGEMENTS This work was supported by a Doctoral Research Award from the Canadian Institutes of

Health Research (to T. Hoyda), and grants from the Heart and Stroke Foundation of Canada (to

A. Ferguson).

125

Chapter 6: GENERAL DISCUSSION

126

At the present time (circa June 2009), a pubmed search for „adiponectin‟ yields approximately 5400 articles. By comparison „leptin‟, discovered in the same year (Zhang et al,

1994), produces approximately 16200 articles making adiponectin seem like the ugly cousin no one wants to talk about. What could account for this drastic difference in attention? The most obvious explanation might seem to be the glaring obese phenotype displayed by the leptin KO mouse (Halaas et al, 1995) compared to a mild anatomical change in the adiponectin KO mouse

(Kubota et al, 2002; Ma et al, 2002; Maeda et al, 2002). While this is a major factor, there may be a more profound explanation for the difference in interest, related to the relative focus of studies investigating the actions of each adipokine in the brain. Unlike leptin, whose central actions have been the subject of intense focus since 1995 (Campfield et al, 1995; Ahima & Flier,

2000), nearly a decade passed between the discovery of adiponectin and the first investigation of its actions in the CNS (Qi et al, 2004). The work presented in this thesis, along with other recent studies represents the second generation of experiments examining the role of adiponectin in the

CNS, „central adiponectin 2.0‟, which build upon the pioneering studies performed by the Ahima laboratory. In the following pages I will discuss our work in the global context of central adiponectin actions, discuss some of the pitfalls concerning contemporary studies, the difficulties which were responsible for the delay to these studies as well as future directions for the field.

An important question with regards to understanding the central effects of adiponectin is how the peptide accesses nuclei positioned behind the BBB. With the exception of the sensory

CVOs, neurons in the CNS are protected from circulating constituents by a selective barrier formed by „tight junction complexes‟ between endothelial cells which are further wrapped in astrocytic „end feet‟ preventing the simple diffusion of blood into the interstitial fluid of the CNS

(Abbott et al, 2006). Though the BBB allows diffusion of gaseous molecules and some lipophilic

127

substances such as ethanol and testosterone, it prevents the passage of large peptidergic, hydrophilic molecules (Abbott et al, 2006). Alternatively sensory CVOs, which contain „leaky‟ or modified cell-to-cell endothelial barriers are unique CNS regions where such barrier impermeant molecules can directly access CNS neurons. However each CVO is separated from the CNS by an external tanycytes/glial layer (Price et al, 2008a; McKinley et al, 2003) inhibiting the diffusion of substances in the CSF to central neurons. Thus, if adiponectin is to directly access sites in the brain protected by the BBB to exert its central effects, it therefore follows that it must do so through either a specialized transport system, or another mechanism.

Two major studies examining the effects of adiponectin in the brain each provide data indirectly supporting a mechanism for transportation of the peptide into the brain (Qi et al, 2004;

Kubota et al, 2007). Each paper reports the results of injections performed to determine if transport of the peptide across the BBB takes place. Qi et al (2004) and Kubota et al (2007) show that ventricular CSF concentration of adiponectin increases following iv injections and serum adiponectin concentration increases following icv injections of the peptide (Kubota et al, 2007;

Qi et al, 2004). Additionally, adiponectin is found in the CSF following iv injections made into adipo-/- mice, in concentrations which vary linearly with the amount injected into the vasculature

(Kubota et al, 2007). Interpretation of this data suggests that a bi-directional, non-saturating transport system, which moves adiponectin into and out of the ventricular space, exists in the

CNS.

To determine the nature of this transport mechanism, a series of BBB permeability studies examining the unidirectional influx rate (Patlak et al, 1983) of iodinated adiponectin injected intravenously or infused through the left ventricle, were undertaken by two separate laboratories (Pan et al, 2006; Spranger et al, 2006). Both studies suggest that in contrast to leptin

128

(Banks et al, 1996) and ghrelin (Banks et al, 2002), radioactive signal indicative of iodinated adiponectin is not present in the brain following iv injections and therefore presumably does not cross the BBB through leakage or by a transport mechanism (Pan et al, 2006; Spranger et al,

2006). Radioactivity is still present in the serum of these animals indicating adiponectin is stable in the circulation (Pan et al, 2006; Spranger et al, 2006). Spranger et al further report that

AdipoR1 and AdipoR2 mRNA is present in cerebral microvessels and that activation of these receptors with gAd decreases the release of interleukin-6, an inflammatory cytokine (Spranger et al, 2006). These studies suggest that adiponectin does not pass through the endothelial BBB by any transport mechanism. Paradoxically, adiponectin may actually increase the integrity of endothelial cell connections in the CNS, a speculation supported by studies on vascular endothelial cells (Elbatarny et al, 2007).

The demonstration that peripherally administered iodinated adiponectin does not cross the BBB, does however leave the question of how adiponectin gets into the CNS/CSF unexplained. Indeed radioimmuno-assay studies performed by Spranger et al (2006) fail to show the presence of adiponectin oligomers in human CSF, most likely due to the limited sensitivity of their technique (Spranger et al, 2006). Follow up studies clearly demonstrate the presence of the peptide in human CSF, with an emphasis towards the trimeric and LMW hexameric complexes as the predominant form (Kos et al, 2007; Kusminski et al, 2007). Adiponectin exists at substantially lower levels (~1000 fold lower; serum=14.0±5.7 [mean±S.E.] µg/ml vs

CSF=11.9±18.8 ng/ml) in human CSF then in the blood (Kusminski et al, 2007). Taken together, these studies raise the intriguing possibility that the CNS may itself produce the peptide (or a form of it) and contain its own „central adiponectin system‟.

129

The initial studies describing adiponectin indicate that white adipose tissue is the sole source of adiponectin mRNA in mice and rats (Scherer et al, 1995; Hu et al, 1996). However, accumulating evidence derived from experiments performed on a wide range of species suggests that adiponectin production might not be exclusively restricted to white adipocytes. Expression studies demonstrate adiponectin production in a variety of peripheral tissues including skeletal muscle fibers (Punyadeera et al, 2005; Krause et al, 2008), brown adipocytes, (Viengchareun et al, 2002), the heart (Pineiro et al, 2005), primary osteoblasts (Berner et al, 2004), the testis

(Caminos et al, 2008), placental tissue (Caminos et al, 2005) and the CNS (Maddineni et al,

2005; Rodriguez-Pacheco et al, 2007; Psilopanagioti et al, 2009). Of these areas, expression in the brain presents a particularly intriguing area of investigation.

The brain is a prominent site for the biosynthesis of many peptides involved in energy homeostasis and metabolic control. These include, but are not limited to NPY, , MCH, , CRH, TRH, GLP-1, CART, resistin and even leptin (for review see, Karla et al, (Kalra et al, 1999)) (Wilkinson et al, 2007). The first study investigating the expression of adiponectin in the CNS was performed by Ramachandran‟s group, who demonstrate that adiponectin mRNA is expressed in the chicken pituitary gland and diencephalon (Maddineni et al, 2005). Expression of adiponectin has been confirmed in the human pituitary gland where it co-localizes with FSH, LH,

GH and TSH in gonadotropes, somatotropes and thyrotropes respectively (Psilopanagioti et al,

2009). Evidence suggests that adiponectin acts locally to inhibit GH and LH release prompting the hypothesis that an autocrine/paracrine hypophyseal regulatory system controlled by adiponectin exists to modulate pituitary hormone secretion (Steyn et al, 2009; Rodriguez-Pacheco et al, 2007).

130

Recent RT-PCR experiments from our laboratory have confirmed the expression of adiponectin mRNA in specific areas of the rat brain including the hypothalamus and the medulla

(Figure 6.1) (Hoyda et al, unpublished observations). Microarray analysis also indicates that the

SFO and AP are central nuclei which produce adiponectin mRNA (Hindmarch et al, 2008)

(Hindmarch et al, unpublished observations). While each of these experimental procedures are powerful molecular analytical tools, they lack the specificity to indicate which cell type expresses the peptide as the entire nucleus is removed for analysis. Both nuclei however make direct, primary neuronal connections with the specific hypothalamic and medullary nuclei in which our recordings have been focused (SFO to PVN, and AP to NTS) (McKinley et al, 2003) raising the intriguing speculation that neuronal production and release of adiponectin may be a mechanism for in nuclei that exist behind the BBB. A strategy to investigate this hypothesis may include a study using single cell RT-PCR to examine the molecular phenotypes of neurons that show retrograde transport of injected fluorescent microspheres into PVN and NTS. One could probe for adiponectin mRNA thereby definitively assessing the capability to produce the peptide in SFO and AP neurons. Alternatively one may elect to sort these fluorescent neurons and perform western blotting or immunohistochemical studies for the peptide itself.

Central production and release of the peptide may be responsible for the presence of adiponectin in the CSF. The choroid plexuses are the main source of CSF secretion into the ventricles and are endowed with not only peptide transporters, but also synthesize and secrete a variety of peptides into the CSF (Smith et al, 2004). Other CNS peptides, such as VP exhibit a higher concentration in the CSF than in the plasma and are thought to diffuse through the brain and be secreted into the CSF (Ludwig & Leng, 2006). It is unknown whether or not adiponectin is

131

Figure 6-1. mRNA for the adiponectin gene is expressed in regions of the rat CNS.

Left: Agarose gels indicate ethidium bromide stained PCR amplification products for GAPDH (+ control) and adiponectin mRNA (two different primer sets) in addition to CRH mRNA, for each of the tissues examined (right); adipose tissue, hypothalamus, brainstem and subfornical organ.

Adipose tissue was used as a positive control to ensure adiponectin mRNA primer set fidelity.

132

synthesized and secreted by epithelial cells of the choroid plexuses but this may be a likely site of adiponectin production and release to the CSF.

The discovery of CNS derived adiponectin, especially in those areas which can sample constituents of the blood; the sensory CVOs, suggest that the ability of peripheral adiponectin to confer its central effects may not require the peptide to be transported across the BBB. Evidence from both peripheral and central studies suggests that adiponectin regulates its own production from adipocytes (Bauche et al, 2006) and pituitary cells (Rodriguez-Pacheco et al, 2007), a phenomenon that is dependent upon adiponectin receptor activation (Bauche et al, 2006). Our lab has shown the expression of both receptors in the AP (Fry et al, 2006) and the SFO (Hindmarch et al, 2008). Whether peripheral adiponectin can regulate central adiponectin production and change its concentration in the brain is unknown at the present time. Sensory CVOs represent potential sites at which this phenomenon could occur as they have the ability to „sense‟ peripheral adiponectin, express adiponectin receptors and produce the peptide itself. One intriguing question is if cellular expression patterns for central adiponectin mirror the inverse relationship demonstrated for the peptide in the periphery in lean and obese phenotypes.

Mapping the central expression of adiponectin receptor mRNA has been advanced through the use of RT-PCR (both from tissue and single neurons) and in situ hybridization techniques. Our lab and others have demonstrated mRNA expression for AdipoR1 and AdipoR2 to be localized in the AP (Fry et al, 2006), NTS (Hoyda et al, 2009b), SFO (Hindmarch et al,

2008)(Alim et al, unpublished observations), PVN (Kubota et al, 2007; Hoyda et al, 2007; Coope et al, 2008; Guillod-Maximin et al, 2009) and ARC (Guillod-Maximin et al, 2009; Kubota et al,

2007). These experiments demonstrate that cells in these nuclei have the capacity to produce functional protein capable of conferring adiponectin‟s central actions on energy homeostasis and

133

feeding behavior. Studies regarding the expression of adiponectin receptor peptide however must be met with cautious optimism as doubts have been raised about the fidelity of some polyclonal antibodies for the receptors. One must ensure that when performing localization or quantitative protein expression studies using either immunohistochemical or western blotting techniques, the appropriate „blocking peptide‟ controls are performed to make certain the data is properly interpreted.

Within the PVN, NTS and AP, our data obtained from combining single cell RT-PCR with electrophysiological recordings clearly demonstrate that most neurons electrically respond to adiponectin and express mRNA for either AdipoR1, AdipoR2 or both of the receptors in the same cell (Fry et al, 2006; Hoyda et al, 2009b; Hoyda et al, 2009a; Hoyda et al, 2007). An intriguing finding only demonstrated in recordings made from PVN or NTS contained in slice preparations is that adiponectin affected the electrical excitability of some cells that did not express mRNA for either receptor as measured by single cell RT-PCR (Hoyda et al, 2009a; Hoyda et al, 2009b).

This phenomenon may be explained in one of three ways, the first of which explores the technical limitations of our recording configuration and the single cell RT-PCR protocol.

The differential interference contrast microscopic technique, optimized to visualize neurons in a slice preparation, allows whole cell recordings to be made at a maximum depth of approximately 50 – 100 µm. Recordings made at these depths are highly stable and usually accompanied by robust responses to experimental treatments. Ultra-structural analysis from retrograde labeling studies indicates that PVN neuronal dendrites can extend beyond the boarders of the nucleus up to in some cases ~2000 µm, in both the lateral and horizontal planes (Stern,

2001). This suggests that an entire PVN neuron may not be contained within a 300 µm coronal section of the hypothalamus. Aspiration of the cytoplasmic contents into the patch pipette during

134

whole cell recording from a cell contained within the slice provides information as to the molecular phenotype of the cell but is limited by only what comes into the pipette. Complete removal of the cytoplasm in this configuration is impossible based on the morphology of the cells. The proper interpretation of „false negatives‟ in analysis of the molecular phenotype becomes a concern if mRNA for the receptors is shuttled out to distant dendrites and thus not easily obtained. This concern is supported by data obtained from single cell RT-PCR analysis of dispersed AP neurons maintained in culture medium (Fry et al, 2006). These neurons have been enzymatically digested and stripped of their complex dendritic trees, allowed to settle and re- grow (including transcription and translation of mRNA) and therefore provide for a more comprehensive molecular phenotype following cytoplasmic content removal and analysis. While this is beneficial for molecular analysis, a level of „realism‟ is eliminated from the recording preparation as cells are devoid of microenvironment signaling and cell-to-cell communication.

The trade-off therefore, must be justified.

A benefit of the single cell RT-PCR technique, as opposed to in situ hybridization, is the detection of small quantities of mRNA copy number due to the fidelity of the multiplex PCR protocol. Because some proteins can be translated into multiple copies from very few mRNA molecules we performed serial dilutions of PVN tissue mRNA to ensure we could detect limited amounts of mRNA (Hoyda et al, 2007). The multiplex single cell RT-PCR strategy used in these experiments has been shown to reliably detect fewer then 20 copies of a single mRNA within neurons, making the technique robust and sensitive (Cauli et al, 1997). Additionally using two different primer sets for the multiplex and nested reactions provides another level of specificity for the PCR reaction decreasing the chances of false positives. These results would suggest that

135

lack of mRNA expression detected in adiponectin responsive cells may not be due to the limitations of detection but rather may have physiological significance.

Recent immuno-histochemical and -fluorescent studies using the appropriate controls demonstrate ubiquitous AdipoR1 and specific AdipoR2 expression in the brain (Guillod-Maximin et al, 2009; Coope et al, 2008). Importantly these studies show expression of the receptors in the

PVN and ARC nuclei (Guillod-Maximin et al, 2009; Coope et al, 2008) confirming our initial reports of mRNA expression in the PVN (Hoyda et al, 2007). Examination of individual cellular expression patterns for the two receptors presents interesting observations. The two main cell groups in the ARC; NPY/AgRP and POMC/CART neurons, both express the two adiponectin receptors, as measured by fluorescent immunohistochemical detection in either GFP-NPY or

GFP-POMC transgenic mice (Guillod-Maximin et al, 2009). Because the experiments were performed separately, it is unknown if these cell types co-localize both receptors in every cell however based on our molecular phenotyping of receptor mRNA in the PVN (Hoyda et al, 2007;

Hoyda et al, 2009a), co-localization may be a feature of ARC neurons as well. It is important to note also that not all of these neurons express AdipoR1 or R2 and that receptors are expressed in

ARC astrocytes as well (Guillod-Maximin et al, 2009). Precedent therefore exists for the possibility of astrocytes in the PVN to express the receptors. This may provide a mechanism for the effects seen on receptor non-expressing adiponectin sensitive PVN neurons as astrocytes release of host of substances affecting neuronal excitability (Abbott et al, 2006; Gordon et al,

2005; Agulhon et al, 2008).

The expression of adiponectin receptors in the ARC and PVN and the roles they play in mediating the peptides effect on food intake and energy homeostasis remains controversial. Two major studies suggest opposite effects of adiponectin on feeding behavior and energy expenditure

136

in different species of rodent (Kubota et al, 2007; Qi et al, 2004). Differences in injection protocols, sites of peptide action and the feeding status of species used in these studies may explain some of these disparities. The first studies examining adiponectin‟s effects in the CNS showed that in mice, icv injections of gAd decrease body weight through increasing energy expenditure and enhancing autonomic function without a concomitant decrease in food intake (Qi et al, 2004). These effects are dependent upon an activated PVN and are accompanied by an increase in CRH mRNA expression in that nucleus (Qi et al, 2004). Coope et al, confirm that icv injections of adiponectin result in an anorexigenic response in rats and illustrate the importance of activated AdipoR1 signaling in the hypothalamus as oligonucleotide anti-sense primers for the receptor mRNA abolish the anorexic effect of the peptide (Coope et al, 2008). Adiponectin also induces activation of the insulin/leptin like signaling pathways IRS1/2, ERK, Akt, FOXO1, JAK2 and STAT3 in the ARC (Coope et al, 2008). Conversely a study performed by Kadowaki‟s group demonstrates that iv injections of adiponectin increase food intake in mice, an effect dependent upon AdipoR1 induced activation of AMPK in the ARC (Kubota et al, 2007). Close examination of the protocols used in these studies reveals two major differences that may account for the discrepancies in their data. First, as opposed to most studies examining the effects of neuropeptides on feeding behavior (Coope et al, 2008), Kubota et al (2007) made injections into fed, rather then fasted mice (Kubota et al, 2007). The importance of pre-injection feeding status to the central effects of adiponectin (and any other feeding peptide) is underscored by the differences in data and this can become problematic if post-prandial surges in other feeding and satiety peptides such as leptin and insulin are not controlled. Secondly, the route of peptide administration could play a role in the CNS response to adiponectin. Kubota et al (2007) examined the feeding behavior of mice receiving iv injections of adiponectin rather then icv

137

injections as performed by the other two studies (Kubota et al, 2007; Qi et al, 2004; Coope et al,

2008). This difference may have profound effects on animal behavior as sites of action and response time could be markedly different between the two protocols. Kubota et al implicate the

ARC as the primary site of action for iv injected adiponectin, whereas Ahima‟s group suggests the PVN is critical for icv adiponectin. Combined, these studies reveal much about adiponectin‟s role in regulating feeding behavior and energy homeostasis. Kadowaki‟s group proposes that adiponectin acts in a sophisticated feedback loop through peripheral and central function as a

„starvation‟ signal (Kadowaki et al, 2008). In times of limited energy storage, adiposity is low while serum adiponectin is elevated engaging the ARC to promote feeding. This intake behavior over time increases adiposity stores and limits adiponectin in the serum, effectively removing the orexigenic drive (Kadowaki et al, 2008).

Both the ARC and the PVN are positioned behind the BBB, although some evidence suggests that neurons in the ARC, being in close proximity to the median eminence (ME), have distal dendrites that extend through the protective tanycyte layer and can sample constituents in the blood (Garcia-Segura et al, 1996; Cheunsuang & Morris, 2005; Cheunsuang et al, 2006;

Norsted et al, 2008). The expression of adiponectin receptors in ARC astrocytes (Guillod-

Maximin et al, 2009) is an interesting observation in that if these receptors are positioned on dendrites that descend into the serum, they may be part of a mechanism, in conjunction with

CVOs, in which parts of the CNS detect peripheral adiponectin. Alternately, CSF adiponectin or perhaps peripherally induced CNS adiponectin may diffuse or be transported to areas such as the

PVN and NTS, conferring a different effect on feeding through actions at these nuclei. Definitive electrical recordings examining the effect of adiponectin on ARC NPY/AgRP and POMC/CART

138

neurons in either starved or fed animals will assist in determining the action of this peptide in the

ARC and its role in promoting or inhibiting feeding behavior in these animals.

The effect of adiponectin on the membrane potential of PVN neurons occurs in the absence of action potential induced neurotransmitter release (Hoyda et al, 2007; Hoyda et al,

2009a). These data suggest that the effects of adiponectin on PVN neurons are not dependent on intermediary network signaling within the hypothalamus or elsewhere. The presence of mRNA detected for each of the receptors in responding cells supports this conclusion. However the absence of receptor mRNA in a few adiponectin responsive neurons provides data for interesting speculation as to the mechanisms responsible for such indirect actions. The effect of adiponectin on receptor-less cells was a phenomenon seen is parvocellular PA neurons in the PVN (Hoyda et al, 2009a). It is unknown whether electrotonic coupling occurs in PA neurons as it does in the

MNC neurons (Andrew et al, 1981) however evidence suggests that dendritic release of neuropeptides may play an important role in coordinating excitability in important nuclei in the hypothalamus (Ludwig & Leng, 2006; Ludwig & Pittman, 2003).

Dendritic trees are highly organized, highly dynamic structures which receive most of the synaptic input from afferent projections, integrating and transmitting network information on its way to the soma (Ludwig & Leng, 2006). In the hypothalamus, they contain large-dense core vesicles (LDCV) filled with neuropeptides primed for activity dependent release into the surrounding microenvironment (Ludwig & Pittman, 2003). VP and OT for example, are released from nerve terminals into the posterior pituitary but are also present in LDCV and released from dendritic spines into subnuclei. This „non-classical‟ release of neuropeptide is critical for information processing within these nuclei as demonstrated by their ability to augment classical neurotransmitter activity at synaptic spines (Hirasawa et al, 2004), exert autoregulatory effects on

139

their own cell membranes (Ludwig & Leng, 2006; Ludwig et al, 2002) and apply long term influence over neighboring cells within the microenvironment (Ludwig & Leng, 2006). Indeed, both OT and VP are more stable in the interstitium of the CNS than in the blood, demonstrating half-lives of ~20 min and ~2 min respectively (Mens et al, 1983). The limitation of their effect in the extracellular fluid is only their diffusion capability, the anatomy of the microenvironment and the expression and local activity of aminopeptidases (Ludwig & Leng, 2006). These long-term and diverse effects of dendritic peptide release suggest the brain is modulated by its own central hormonal system which can profoundly impact neuronal function. With respect to the effects of adiponectin on non-receptor expressing neurons, it will be interesting to see if these cells express high affinity peptide receptors such as the OT receptor, known to be expressed in the PVN

(Freund-Mercier et al, 1994; Adan et al, 1995). Perhaps a paracrine effect from released neuropeptides of neighboring cells is a mechanism adiponectin uses to coordinate the electrical activity of PVN neurons.

In agreement with previous studies (Qi et al, 2004), our experiments demonstrate an important role for central actions of adiponectin in controlling the stress-axis (Hoyda et al,

2009a). Qi et al report that icv injections of adiponectin are accompanied by increases in PVN c- fos immunoreactivity, an activated autonomic nervous system demonstrated by increases in VO2 usage and thermogenesis, as well as an increase in CRH mRNA within the PVN, culminating in positive energy expenditure and decreased body weight (Qi et al, 2004). Our experiments demonstrate a depolarizing effect on NE CRH producing PVN neurons and NPY producing NTS neurons, and increases in plasma ACTH concentrations following icv injections of adiponectin

(Hoyda et al, 2009a; Hoyda et al, 2009b) each having a profound effect on the stress-axis.

140

Proper coordination of the stress-axis is vital to producing appropriate responses to external physiological stressors. Control of this response is important to emotional well-being, performance, executive function and proper social interaction culminating in an increased chance of survival (Charmandari et al, 2005). Disruptions in the stress-axis from mounting an inadequate or excessive response can lead to pathological consequences including inhibition of growth and development, thyroid function, reproduction, metabolic and gastrointestinal function, immune function and the development of some psychiatric disorders (Charmandari et al, 2005; Ulrich-Lai

& Herman, 2009). This „allostasis‟ underscores the parallel between the stress-axis and metabolic regulation as many of these same pathologies are present in disruptions of either. CRH produced in NE cells of the PVN is the principle hypothalamic regulator of the pituitary-adrenal axis as it increases the secretion of ACTH from anterior pituitary corticotropes (Bale & Vale, 2004). CRH is also a major anorexigenic peptide whose secretion is paradoxically stimulated by NPY, the major orexigenic peptide in the CNS (Charmandari et al, 2005). In addition to our results demonstrating direct depolarizing actions of adiponectin on PVN CRH neurons, recordings from

NTS NPY neurons indicate that this medullary nucleus may be a corollary site where an adiponectin sensitive mechanism for modulating food intake and energy homeostasis exists

(Hoyda et al, 2009b). Single cell RT-PCR analysis combined with electrophysiological recordings reveal a population of NPY only expressing neurons (separate from the co-expressing

NPY/GAD67 neurons), which are depolarized by adiponectin. Immunohistochemical evidence suggests that these NPY neurons project to the medial parvocellular region of the PVN, the anatomical location of CRH expressing neurons involved in the HPA axis, where they act to regulate hypothalamic peptide release (Charmandari et al, 2005; Sawchenko et al, 1985).

Combined, our experiments suggest two mechanisms for increasing CRH secretion; directly at

141

PVN CRH neurons and through influencing the excitability of putative PVN projecting NPY

NTS neurons.

Since the electrical effects of adiponectin and its transport into the CSF are dose dependent (Hoyda et al, 2007; Kubota et al, 2007), we can speculate on the activity level of the stress axis in pathological extremes of body adiposity composition. In the obese state, serum adiponectin is low (Hu et al, 1996) suggesting limited transport to CNS sites behind the BBB and therefore depressed HPA axis activation and further orexigenic feeding behavior due diminished concentrations of serum CRH and ACTH. In contrast, in the anorexic state, adiponectin levels are drastically elevated (Delporte et al, 2003) and presumably increased in the CSF. This would lead to an over active HPA axis potentiating anorexigenic behavior. This begs the question whether extreme concentrations of adiponectin are the cause of such diseases or simply co-incident with them, further aggravating the pathologies associated with the extremes of energy homeostasis and feeding behavior. Clearly there may be important consequences to unregulated adiponectin signaling in the CNS and present interesting opportunities for therapeutic intervention in the peptide system.

In view of adiponectin‟s ability to improve or ameliorate pathologies associated with the metabolic syndrome, its therapeutic potential is under active development. Adiponectin can be used in the treatment of diabetes and obesity because of its glucose lowering, insulin sensitizing and FFA oxidation effects, although a clear understanding of the mechanisms involved is still obscure. Additionally, the differences in multimer composition and the role played by each still require rigorous examination. Interestingly HMW adiponectin is elevated in some forms of type 1 diabetes (Leth et al, 2008) which raises the question whether this unique relationship between diabetes and serum adiponectin (HMW) is causal or in compensatory response to the inability of

142

the molecule to perform its peripheral or central actions. Clearly a full detailing of each form‟s action in each major organ system is crucial to developing the peptide as a therapeutic.

The anti-inflammatory, anti-diabetic action of adiponectin is based largely on its inverse concentration relationship with TNF-α (Wang & Scherer, 2008; Shetty et al, 2009). Though similar in structure, they exhibit opposite effects on and interestingly, down regulate expression and release of each other (Ouchi et al, 2001; Kern et al, 2003; Maeda et al,

2002; Komai et al, 2007). Adiponectin may also act in-conjunction with leptin to organize ectopic fat distribution (Kim et al, 2007), an important variable in the pathology of metabolic syndrome

(Moller & Kaufman, 2005). Adiponectin over-expression in a leptin deficient mouse ameliorates most of the pathologies associated with leptin-deficiency including type 2 diabetes, yet mice increase fat mass above normal leptin deficient weight gain (Kim et al, 2007). The difference is that the fat is distributed away from visceral organs to subcutaneous areas (Kim et al, 2007) suggesting these pathologies may be due to regional adipose tissue distribution highlighting another avenue for potential therapeutic intervention (Shetty et al, 2009; Kim et al, 2007).

Potential interventional therapeutic strategies involving adiponectin are numerous but because serum concentrations of adiponectin change in virtually all major metabolic syndrome pathologies, the peptide is also an informative biomarker and may assist in diagnosis and treatment options. Using agents to enhance the expression and secretion of endogenous adiponectin is also showing promise in pre-clinical work. A variety of mechanisms, including

PPARγ agonists (TZD) and ACE inhibitors, increase plasma concentrations of adiponectin and endocannabinoid receptor (CB1) antagonists which induce weight loss, do so through enhancing the secretion of adiponectin from adipocytes (Shetty et al, 2009).

143

Elucidating the signaling mechanisms responsible for mediating adiponectin effects in the periphery and the CNS provide important therapeutic targets for drug development.

Activation of AMPK, implicated in adiponectin signaling (Yamauchi et al, 2002), with AICAR promotes positive energy expenditure and improves the obesity phenotype through central

(Minokoshi et al, 2004) and peripheral (Ruderman et al, 2003) mechanisms. Our work investigating the ionic conductances responsible for central adiponectin signaling indicate potassium currents, delayed rectifier and inward rectifier ATP-sensitive currents, highlight the importance of these currents in mediating whole body energy homeostasis (Hoyda & Ferguson,

2009 submitted). The hyperpolarizing effects of adiponectin are blocked by glibenclamide, a

sulfonylurea molecule used in the treatment of diabetes based on its ability to inhibit the KATP channel in the thus depolarizing B-cells and enhancing insulin secretion (Ashcroft,

2006). While still a few years away from the development of safe interventional therapeutics which act centrally to „adjust‟ aberrant electrical activity in neuronal networks responsible for mediating food intake, our experiments illuminate potentially interesting targets in the PVN and

NTS which would assist in obtaining energy homeostasis.

Summary and Conclusions

Clearly the last decade has seen an explosion in the adiponectin field, particularly since the cloning of the adiponectin receptors (Yamauchi et al, 2003) and the development of reliable detection assays for the different forms of the peptide. Investigation into the peptides central effects however exists in an infantile state, ready to emerge as data is collected, interpreted and expanded on. Building off the pioneering studies examining adiponectin actions in the CNS (Qi et al, 2004), the research presented in this thesis expands the central role of adiponectin to include specific electrical effects on PVN and NTS neurons; hyperpolarizing PVN MNC OT and some

144

VP neurons while depolarizing PVN NE CRH, some MNC VP, PA OT and NTS NPY neurons through an activated adiponectin receptor dependent modulation of different potassium currents.

Some neurons in the PVN do not require the expression of adiponectin receptors for effects to occur, underscoring the complexity of cell-to-cell communication in this nucleus. These effects have functional consequences for the animal including hypotension through actions in the brainstem and an increase in plasma ACTH concentration through actions in the hypothalamus

(Hoyda et al, 2007; Hoyda et al, 2009b; Hoyda et al, 2009a)(Hoyda & Ferguson, 2009 submitted). These experiments continue in the stream of „foundational‟ studies, beginning to place a few pieces of the puzzle together. But the whole picture is undoubtedly large and these studies in no way represent end points to the depth at which adiponectin acts in the CNS to control autonomic function and energy expenditure. If these studies retain their relevance, they will become starting blocks for researchers to expand on and just as in any new area of investigation; the optimistic and inquisitive mind will seek out new mechanisms and demonstrate roles for the peptide unknown to the field before. Upcoming studies will undeniably build on to and eventually complete the model of adiponectin actions in the brain and once this is achieved is when we can start to realize the full therapeutic potential of this adipokine.

145

REFERENCES

Abbott NJ, Ronnback L & Hansson E (2006). Astrocyte-endothelial interactions at the blood- brain barrier. Nat Rev Neurosci 7, 41-53.

Adams WB, Parnas I & Levitan IB (1980). Mechanism of long-lasting synaptic inhibition in Aplysia neuron R15. J Neurophysiol 44, 1148-1160.

Adan RA, van Leeuwen FW, Sonnemans MA, Brouns M, Hoffman G, Verbalis JG & Burbach JP (1995). Rat in brain, pituitary, mammary gland, and uterus: partial sequence and immunocytochemical localization. Endocrinology 136, 4022-4028.

Agulhon C, Petravicz J, McMullen AB, Sweger EJ, Minton SK, Taves SR, Casper KB, Fiacco TA & McCarthy KD (2008). What is the role of astrocyte calcium in neurophysiology? Neuron 59, 932-946.

Ahima RS (2005). Central actions of adipocyte hormones. Trends Endocrinol Metab 16, 307-313.

Ahima RS (2006). Adipose tissue as an endocrine organ. Obesity (Silver Spring) 14 Suppl 5, 242S-249S.

Ahima RS & Flier JS (2000). Leptin. Annu Rev Physiol 62, 413-437.

Ahima RS & Lazar MA (2008). Adipokines and the peripheral and neural control of energy balance. Mol Endocrinol 22, 1023-1031.

Altschuler SM, Bao XM, Bieger D, Hopkins DA & Miselis RR (1989). Viscerotopic representation of the upper alimentary tract in the rat: sensory ganglia and nuclei of the solitary and spinal trigeminal tracts. J Comp Neurol 283, 248-268.

Andrew RD, MacVicar BA, Dudek FE & Hatton GI (1981). Dye transfer through gap junctions between neuroendocrine cells of rat hypothalamus. Science 211, 1187-189.

Antunes-Rodrigues J, McCann SM, Rogers LC & Samson WK (1985). Atrial natriuretic factor inhibits dehydration and angiotensin II - induced water intake in the conscious unrestrained rat. Proc Natl Acad Sci USA 82, 8720-8723.

146

Arancibia S, Rage F, Astier H & Tapia-Arancibia L (1996). Neuroendocrine and autonomous mechanisms underlying thermoregulation in cold environment. Neuroendocrinology 64, 257-267.

Arbiser JL, Morton CC, Bruns GA & Majzoub JA (1988). Human corticotropin releasing hormone gene is located on the long arm of chromosome 8. Cytogenet Cell Genet 47, 113-116.

Arita Y, Kihara S, Ouchi N, Takahashi M, Maeda K, Miyagawa J, Hotta K, Shimomura I, Nakamura T, Miyaoka K, Kuriyama H, Nishida M, Yamashita S, Okubo K, Matsubara K, Muraguchi M, Ohmoto Y, Funahashi T & Matsuzawa Y (1999). Paradoxical decrease of an adipose-specific protein, adiponectin, in obesity. Biochem Biophys Res Commun 257, 79-83.

Arletti R, Benelli A & Bertolini A (1989). Influence of oxytocin on feeding behavior in the rat. Peptides 10, 89-93.

Arletti R, Benelli A & Bertolini A (1990). Oxytocin inhibits food and fluid intake in rats. Physiol Behav 48, 825-830.

Armstrong DM, Ross CA, Pickel VM, Joh TH & Reis DJ (1982). Distribution of dopamine-, noradrenaline- and adrenaline-containing cell bodies in the rat medulla oblongata: demonstrated by the immunocytochemical localization of catecholamine biosynthetic enzymes. J Comp Neurol 212, 173-187.

Armstrong WE (1995). Morphological and electrophysiological classification of hypothalamic supraoptic neurons. Prog Neurobiol 47, 291-339.

Armstrong WE & Stern JE (1998a). Electrophysiological distinctions between oxytocin and vasopressin neurons in the supraoptic nucleus. Adv Exp Med Biol 449, 67-77.

Armstrong WE & Stern JE (1998b). Phenotypic and state-dependent expression of the electrical and morphological properties of oxytocin and vasopressin neurones. Prog Brain Res 119, 101- 113.

Armstrong WE, Warach S, Hatton GI & McNeill TH (1980). Subnuclei in the rat hypothalamic paraventricular nucleus: a cytoarchitectural, horseradish peroxidase and immunocytochemical analysis. Neuroscience 5, 1931-1958.

147

Ashcroft FM (1988). Adenosine 5'-triphosphate-sensitive potassium channels. Annu Rev Neurosci 11, 97-118.

Ashcroft FM (2006). K(ATP) channels and insulin secretion: a key role in health and disease. Biochem Soc Trans 34, 243-246.

Bains JS & Ferguson AV (1997). Nitric oxide depolarizes type II paraventricular nucleus neurons in vitro. Neuroscience 79, 149-159.

Bale TL & Vale WW (2004). CRF and CRF receptors: role in stress responsivity and other behaviors. Annu Rev Pharmacol Toxicol 44, 525-557.

Banks WA, Kastin AJ, Huang W, Jaspan JB & Maness LM (1996). Leptin enters the brain by a saturable system independent of insulin. Peptides 17, 305-311.

Banks WA, Tschop M, Robinson SM & Heiman ML (2002). Extent and direction of ghrelin transport across the blood-brain barrier is determined by its unique primary structure. J Pharmacol Exp Ther 302, 822-827.

Barraco RA, Ergene E, Dunbar JC, Ganduri YL & Anderson GF (1991). Y2 receptors for neuropeptide Y in the nucleus of the solitary tract mediate depressor responses. Peptides 12, 691- 698.

Bauche IB, it El MS, Rezsohazy R, Funahashi T, Maeda N, Miranda LM & Brichard SM (2006). Adiponectin downregulates its own production and the expression of its AdipoR2 receptor in transgenic mice. Biochem Biophys Res Commun 345, 1414-1424.

Baylis PH (1987). Osmoregulation and control of vasopressin secretion in healthy humans. Am J Physiol 253, R671-R678.

Berg AH, Combs TP, Du X, Brownlee M & Scherer PE (2001). The adipocyte-secreted protein Acrp30 enhances hepatic insulin action. Nat Med 7, 947-953.

Berner HS, Lyngstadaas SP, Spahr A, Monjo M, Thommesen L, Drevon CA, Syversen U & Reseland JE (2004). Adiponectin and its receptors are expressed in bone-forming cells. Bone 35, 842-849.

148

Birnbaum MJ (2005). Activating AMP-activated protein kinase without AMP. Mol Cell 19, 289- 290.

Bjursell M, Ahnmark A, Bohlooly Y, William-Olsson L, Rhedin M, Peng XR, Ploj K, Gerdin AK, Arnerup G, Elmgren A, Berg AL, Oscarsson J & Linden D (2007). Opposing effects of adiponectin receptors 1 and 2 on energy metabolism. Diabetes 56, 583-593.

Blevins JE, Eakin TJ, Murphy JA, Schwartz MW & Baskin DG (2003). Oxytocin innervation of caudal brainstem nuclei activated by cholecystokinin. Brain Res 993, 30-41.

Blevins JE, Schwartz MW & Baskin DG (2004). Evidence that paraventricular nucleus oxytocin neurons link hypothalamic leptin action to caudal brain stem nuclei controlling meal size. Am J Physiol Regul Integr Comp Physiol 287, R87-R96.

Bloom FE, Battenberg EL, Rivier J & Vale W (1982). Corticotropin releasing factor (CRF): immunoreactive neurones and fibers in rat hypothalamus. Regul Pept 4, 43-48.

Britton DR, Koob GF, Rivier J & Vale W (1982). Intraventricular corticotropin-releasing factor enhances behavioral effects of novelty. Life Sci 31, 363-367.

Brody MJ (1988). Central nervous system and mechanisms of hypertension. Clin Physiol Biochem 6, 230-239.

Brown MR, Fisher LA, Rivier J, Spiess J, Rivier C & Vale W (1982). Corticotropin-releasing factor: effects on the sympathetic nervous system and oxygen consumption. Life Sci 30, 207-210.

Brown MR, Fisher LA, Webb V, Vale WW & Rivier JE (1985). Corticotropin-releasing factor: a physiologic regulator of adrenal epinephrine secretion. Brain Res 328, 355-357.

Bryan J, Vila-Carriles WH, Zhao G, Babenko AP & guilar-Bryan L (2004). Toward linking structure with function in ATP-sensitive K+ channels. Diabetes 53 Suppl 3, S104-S112.

Buijs RM, Geffard M, Pool CW & Hoorneman EM (1984). The dopaminergic innervation of the supraoptic and paraventricular nucleus. A light and electron microscopical study. Brain Res 323, 65-72.

149

Camacho A & Phillips MI (1981). Horseradish peroxidase study in rat of the neural connections of the organum vasculosum of the lamina terminalis. Neurosci Lett 25, 201-204.

Caminos JE, Nogueiras R, Gallego R, Bravo S, Tovar S, Garcia-Caballero T, Casanueva FF & Dieguez C (2005). Expression and regulation of adiponectin and receptor in human and rat . J Clin Endocrinol Metab 90, 4276-4286.

Caminos JE, Nogueiras R, Gaytan F, Pineda R, Gonzalez CR, Barreiro ML, Castano JP, Malagon MM, Pinilla L, Toppari J, Dieguez C & Tena-Sempere M (2008). Novel expression and direct effects of adiponectin in the rat testis. Endocrinology 149, 3390-3402.

Campfield LA, Smith FJ, Guisez Y, Devos R & Burn P (1995). Recombinant mouse OB protein: Evidence for a peripheral signal linking adiposity and central neural networks. Science 269, 546- 549.

Catelli JM, Giakas WJ & Sved AF (1987). GABAergic mechanisms in nucleus tractus solitarius alter blood pressure and vasopressin release. Brain Res 403, 279-289.

Cauli B, Audinat E, Lambolez B, Angulo MC, Ropert N, Tsuzuki K, Hestrin S & Rossier J (1997). Molecular and physiological diversity of cortical nonpyramidal cells. J Neurosci 17, 3894-3906.

Charmandari E, Tsigos C & Chrousos G (2005). Endocrinology of the stress response. Annu Rev Physiol 67, 259-284.

Chaudhri OB, Salem V, Murphy KG & Bloom SR (2008). Gastrointestinal satiety signals. Annu Rev Physiol 70, 239-255.

Cheunsuang O & Morris R (2005). Astrocytes in the arcuate nucleus and median eminence that take up a fluorescent dye from the circulation express leptin receptors and neuropeptide Y Y1 receptors. Glia 52, 228-233.

Cheunsuang O, Stewart AL & Morris R (2006). Differential uptake of molecules from the circulation and CSF reveals regional and cellular specialisation in CNS detection of homeostatic signals. Cell Tissue Res 325, 397-402.

150

Chinetti G, Zawadski C, Fruchart JC & Staels B (2004). Expression of adiponectin receptors in human macrophages and regulation by agonists of the nuclear receptors PPARalpha, PPARgamma, and LXR. Biochem Biophys Res Commun 314, 151-158.

Clark JT, Kalra PS, Crowley WR & Kalra SP (1984). Neuropeptide Y and human stimulate feeding behavior in rats. Endocrinology 115, 427-429.

Combs TP, Berg AH, Obici S, Scherer PE & Rossetti L (2001). Endogenous glucose production is inhibited by the adipose-derived protein Acrp30. J Clin Invest 108, 1875-1881.

Combs TP, Berg AH, Rajala MW, Klebanov S, Iyengar P, Jimenez-Chillaron JC, Patti ME, Klein SL, Weinstein RS & Scherer PE (2003). Sexual differentiation, pregnancy, calorie restriction, and aging affect the adipocyte-specific secretory protein adiponectin. Diabetes 52, 268-276.

Conrad KP, Gellai M, North WG & Valtin H (1993). Influence of oxytocin on renal hemodynamics and sodium excretion. Ann N Y Acad Sci 689, 346-362.

Coope A, Milanski M, Araujo EP, Tambascia M, Saad MJ, Geloneze B & Velloso LA (2008). AdipoR1 mediates the anorexigenic and insulin/leptin-like actions of adiponectin in the hypothalamus. FEBS Lett 582, 1471-1476.

Cottrell GT & Ferguson AV (2004). Sensory circumventricular organs: central roles in integrated autonomic regulation. Regul Pept 117, 11-23.

Csaki A, Kocsis K, Halasz B & Kiss J (2000). Localization of glutamatergic/aspartatergic neurons projecting to the hypothalamic paraventricular nucleus studied by retrograde transport of [3H]D-aspartate autoradiography. Neuroscience 101, 637-655.

Daftary SS, Boudaba C, Szabo K & Tasker JG (1998). Noradrenergic excitation of magnocellular neurons in the rat hypothalamic paraventricular nucleus via intranuclear glutamatergic circuits. J Neurosci 18, 10619-10628.

Decavel C, Geffard M & Calas A (1987). Comparative study of dopamine- and noradrenaline- immunoreactive terminals in the paraventricular and supraoptic nuclei of the rat. Neurosci Lett 77, 149-154.

151

Deepa SS & Dong LQ (2009). APPL1: role in adiponectin signaling and beyond. Am J Physiol Endocrinol Metab 296, E22-E36.

Delporte ML, Brichard SM, Hermans MP, Beguin C & Lambert M (2003). Hyperadiponectinaemia in anorexia nervosa. Clin Endocrinol (Oxf) 58, 22-29.

Demitrack MA, Lesem MD, Listwak SJ, Brandt HA, Jimerson DC & Gold PW (1990). CSF oxytocin in anorexia nervosa and bulimia nervosa: clinical and pathophysiologic considerations. Am J Psychiatry 147, 882-86.

Dunbar JC, Ergene E, Anderson GF & Barraco RA (1992). Decreased cardiorespiratory effects of neuropeptide Y in the nucleus tractus solitarius in diabetes. Am J Physiol 262, R865-R871.

Elbatarny HS, Netherton SJ, Ovens JD, Ferguson AV & Maurice DH (2007). Adiponectin, ghrelin, and leptin differentially influence human platelet and human vascular endothelial cell functions: implication in obesity-associated cardiovascular diseases. Eur J Pharmacol 558, 7-13.

Elmquist JK, Elias CF & Saper CB (1999). From lesions to leptin: hypothalamic control of food intake and body weight. Neuron 22, 221-232.

Ergene E, Dunbar JC & Barraco RA (1993). Visceroendocrine responses elicited by neuropeptide Y in the nucleus tractus solitarius. Brain Res Bull 32, 461-465.

Fatt P & Katz B (1953). The electrical properties of crustacean muscle fibres. J Physiol 120, 171- 204.

Ferguson AV (1991). The area postrema: a cardiovascular control centre at the blood-brain interface? Can J Physiol Pharmacol 69, 1026-1034.

Ferguson AV, Day T.A. & Renaud LP (1984a). Influence of subfornical organ stimulation on the excitability of hypothalamic paraventricular neurons projecting to the dorsal medulla. Am J Physiol 247, R1088-R1092.

Ferguson AV, Day TA & Renaud LP (1984b). Subfornical organ efferents influence the excitability of neurohypophysial and tuberoinfundibular paraventricular nucleus neurons in the rat. Neuroendocrinology 39, 423-428.

152

Ferguson AV & Washburn DLS (1998). Angiotensin II: a peptidergic neurotransmitter in central autonomic pathways. Prog Neurobiol 54, 169-192.

Ferri CC, Yuill EA & Ferguson AV (2005). Interleukin-1beta depolarizes magnocellular neurons in the paraventricular nucleus of the hypothalamus through -mediated activation of a non selective cationic conductance. Regul Pept 129, 63-71.

Follwell MJ & Ferguson AV (2002a). influences magnocellular and parvocellular neurons of paraventricular nucleus via separate mechanisms. Am J Physiol Regul Integr Comp Physiol 283, R1293-R1302.

Follwell MJ & Ferguson AV (2002b). Cellular mechanisms of actions on paraventricular nucleus neurones in rat hypothalamus. J Physiol 545, 855-867.

Freund-Mercier MJ, Stoeckel ME & Klein MJ (1994). Oxytocin receptors on oxytocin neurones: histoautoradiographic detection in the lactating rat. J Physiol 480 ( Pt 1), 155-161.

Fruebis J, Tsao TS, Javorschi S, Ebbets-Reed D, Erickson MR, Yen FT, Bihain BE & Lodish HF (2001). Proteolytic cleavage product of 30-kDa adipocyte complement-related protein increases fatty acid oxidation in muscle and causes weight loss in mice. Proc Natl Acad Sci U S A 98, 2005-2010.

Fry M, Hoyda TD & Ferguson AV (2007). Making sense of it: roles of the sensory circumventricular organs in feeding and regulation of energy homeostasis. Exp Biol Med (Maywood ) 232, 14-26.

Fry M, Smith PM, Hoyda TD, Duncan M, Ahima RS, Sharkey KA & Ferguson AV (2006). Area postrema neurons are modulated by the adipocyte hormone adiponectin. J Neurosci 26, 9695- 9702.

Furuhashi M, Ura N, Higashiura K, Murakami H, Tanaka M, Moniwa N, Yoshida D & Shimamoto K (2003). Blockade of the renin-angiotensin system increases adiponectin concentrations in patients with essential hypertension. Hypertension 42, 76-81.

Gao Q & Horvath TL (2007). Neurobiology of feeding and energy expenditure. Annu Rev Neurosci 30, 367-398.

153

Garcia-Segura LM, Chowen JA & Naftolin F (1996). Endocrine glia: roles of glial cells in the brain actions of steroid and and in the regulation of hormone secretion. Front Neuroendocrinol 17, 180-211.

Geiger JR & Jonas P (2000). Dynamic control of presynaptic Ca(2+) inflow by fast-inactivating K(+) channels in hippocampal mossy fiber boutons. Neuron 28, 927-939.

Gimpl G & Fahrenholz F (2001). The oxytocin receptor system: structure, function, and regulation. Physiol Rev 81, 629-683.

Gold RM (1973). Hypothalamic obesity: the myth of the ventromedial nucleus. Science 182, 488- 490.

Gordon GR, Baimoukhametova DV, Hewitt SA, Rajapaksha WR, Fisher TE & Bains JS (2005). Norepinephrine triggers release of glial ATP to increase postsynaptic efficacy. Nat Neurosci 8, 1078-1086. guilar-Bryan L, Clement JP, Gonzalez G, Kunjilwar K, Babenko A & Bryan J (1998). Toward understanding the assembly and structure of KATP channels. Physiol Rev 78, 227-245.

Guillod-Maximin E, Roy AF, Vacher CM, Aubourg A, Bailleux V, Lorsignol A, Penicaud L, Parquet M & Taouis M (2009). Adiponectin receptors are expressed in hypothalamus and colocalized with and neuropeptide Y in rodent arcuate neurons. J Endocrinol 200, 93-105.

Guyenet PG (2006). The sympathetic control of blood pressure. Nat Rev Neurosci 7, 335-346.

Hakansson M-L, Brown H, Ghilardi N, Skoda RC & Meister B (1998). Leptin receptor immunorectivity in chemically defined target neurons of the hypothalamus. The Journal of Neuroscience 18, 559-572.

Halaas JL, Gajiwala KS, Maffei M, Cohen SL, Chait BT, Rabinowitz D, Lallone RL, Burley SK & Friedman JM (1995). Weight-reducing effects of the plasma protein encoded by the obese gene. Science 269, 543-546.

Hara K, Boutin P, Mori Y, Tobe K, Dina C, Yasuda K, Yamauchi T, Otabe S, Okada T, Eto K, Kadowaki H, Hagura R, Akanuma Y, Yazaki Y, Nagai R, Taniyama M, Matsubara K, Yoda M,

154

Nakano Y, Tomita M, Kimura S, Ito C, Froguel P & Kadowaki T (2002). Genetic variation in the gene encoding adiponectin is associated with an increased risk of type 2 diabetes in the Japanese population. Diabetes 51, 536-540.

Harfstrand A, Fuxe K, Terenius L & Kalia M (1987). Neuropeptide Y-immunoreactive perikarya and nerve terminals in the rat medulla oblongata: relationship to cytoarchitecture and catecholaminergic cell groups. J Comp Neurol 260, 20-35.

Harms PG & Ojeda SR (1974). A rapid and simple procedure for chronic cannulation of the rat jugular vein. J Appl Physiol 36, 391-392.

Hashimoto K, Hattori T, Murakami K, Suemaru S, Kawada Y, Kageyama J & Ota Z (1985). Reduction in brain immunoreactive corticotropin-releasing factor (CRF) in spontaneously hypertensive rats. Life Sci 36, 643-647.

Hermes ML, Coderre EM, Buijs RM & Renaud LP (1999). GABA and glutamate mediate rapid neurotransmission from suprachiasmatic nucleus to hypothalamic paraventricular nucleus in rat. J Physiol (Lond) 496, 749-757.

Hille B (1992). Ionic channels of excitable membranes, Second Edition ed. Sinauer Associates Inc., Massachusetts.

Hindmarch C, Fry M, Yao S, Smith PM, Murphy D & Ferguson AV (2008). Microarray Analysis of the Transcriptome of the Subfornical Organ in the Rat: Regulation by Fluid and Food Deprivation. Am J Physiol Regul Integr Comp Physiol.

Hindmarch C, Yao S, Beighton G, Paton J & Murphy D (2006). A comprehensive description of the transcriptome of the hypothalamoneurohypophyseal system in euhydrated and dehydrated rats. Proc Natl Acad Sci U S A 103, 1609-1614.

Hirasawa M, Mouginot D, Kozoriz MG, Kombian SB & Pittman QJ (2003). Vasopressin differentially modulates non-NMDA receptors in vasopressin and oxytocin neurons in the supraoptic nucleus. J Neurosci 23, 4270-4277.

Hirasawa M, Schwab Y, Natah S, Hillard CJ, Mackie K, Sharkey KA& Pittman QJ (2004). Dendritically released transmitters cooperate via autocrine and retrograde actions to inhibit afferent excitation in rat brain. J Physiol 559, 611-624.

155

Hoyda TD, Fry M, Ahima RS & Ferguson AV (2007). Adiponectin selectively inhibits oxytocin neurons of the paraventricular nucleus of the hypothalamus. J Physiol 585, 805-816.

Hoyda TD, Samson WK & Ferguson AV (2009a). Adiponectin depolarizes parvocellular paraventricular nucleus neurons controlling neuroendocrine and autonomic function. Endocrinology 150, 832-840.

Hoyda TD, Smith PM & Ferguson AV (2009b). Adiponectin acts in the nucleus of the solitary tract to decrease blood pressure by modulating the excitability of neuropeptide Y neurons. Brain Res 1256, 76-84.

Hu E, Liang P & Spiegelman BM (1996). AdipoQ is a novel adipose-specific gene dysregulated in obesity. J Biol Chem 271, 10697-10703.

Hug C, Wang J, Ahmad NS, Bogan JS, Tsao TS & Lodish HF (2004). T-cadherin is a receptor for hexameric and high-molecular-weight forms of Acrp30/adiponectin. Proc Natl Acad Sci U S A 101, 10308-10313.

Imai J, Katagiri H, Yamada T, Ishigaki Y, Ogihara T, Uno K, Hasegawa Y, Gao J, Ishihara H, Sasano H & Oka Y (2006). Cold exposure suppresses serum adiponectin levels through sympathetic nerve activation in mice. Obesity (Silver Spring) 14, 1132-1141.

Iwashima Y, Katsuya T, Ishikawa K, Ouchi N, Ohishi M, Sugimoto K, Fu Y, Motone M, Yamamoto K, Matsuo A, Ohashi K, Kihara S, Funahashi T, Rakugi H, Matsuzawa Y & Ogihara T (2004). Hypoadiponectinemia is an independent risk factor for hypertension. Hypertension 43, 1318-1323.

Jackson MB, Konnerth A & Augustine GJ (1991). Action potential broadening and frequency- dependent facilitation of calcium signals in pituitary nerve terminals. Proc Natl Acad Sci U S A 88, 380-384.

Joy MD & Lowe RD (1970). Evidence that the area postrema mediates the central cardiovascular response to angiotensin II. Nature 228, 1303-1304.

Kadowaki T & Yamauchi T (2005). Adiponectin and adiponectin receptors. Endocr Rev 26, 439- 451.

156

Kadowaki T, Yamauchi T & Kubota N (2008). The physiological and pathophysiological role of adiponectin and adiponectin receptors in the peripheral tissues and CNS. FEBS Lett 582, 74-80.

Kadowaki T, Yamauchi T, Kubota N, Hara K, Ueki K & Tobe K (2006). Adiponectin and adiponectin receptors in insulin resistance, diabetes, and the metabolic syndrome. J Clin Invest 116, 1784-1792.

Kahn BB, Alquier T, Carling D & Hardie DG (2005). AMP-activated protein kinase: ancient energy gauge provides clues to modern understanding of metabolism. Cell Metab 1, 15-25.

Kalra SP, Dube MG, Pu S, Xu B, Horvath TL & Kalra PS (1999). Interacting appetite-regulating pathways in the hypothalamic regulation of body weight. Endocr Rev 20, 68-100.

Karpichev IV, Cornivelli L & Small GM (2002). Multiple regulatory roles of a novel Saccharomyces cerevisiae protein, encoded by YOL002c, in lipid and phosphate metabolism. J Biol Chem 277, 19609-19617.

Kavanaugh MP, Varnum MD, Osborne PB, Christie MJ, Busch AE, Adelman JP & North RA (1991). Interaction between tetraethylammonium and amino acid residues in the pore of cloned voltage-dependent potassium channels. J Biol Chem 266, 7583-7587.

Kaye WH (1996). Neuropeptide abnormalities in anorexia nervosa. Psychiatry Res 62, 65-74.

Kern PA, Di Gregorio GB, Lu T, Rassouli N & Ranganathan G (2003). Adiponectin expression from human adipose tissue: relation to obesity, insulin resistance, and tumor necrosis factor-alpha expression. Diabetes 52, 1779-1785.

Kersten S, Seydoux J, Peters JM, Gonzalez FJ, Desvergne B & Wahli W (1999). Peroxisome proliferator-activated receptor alpha mediates the adaptive response to fasting. J Clin Invest 103, 1489-1498.

Kharroubi I, Rasschaert J, Eizirik DL & Cnop M (2003). Expression of adiponectin receptors in pancreatic beta cells. Biochem Biophys Res Commun 312, 1118-1122.

Kim JY, van de WE, Laplante M, Azzara A, Trujillo ME, Hofmann SM, Schraw T, Durand JL, Li H, Li G, Jelicks LA, Mehler MF, Hui DY, Deshaies Y, Shulman GI, Schwartz GJ & Scherer

157

PE (2007). Obesity-associated improvements in metabolic profile through expansion of adipose tissue. J Clin Invest 117, 2621-2637.

Kimura T, Tanizawa O, Mori K, Brownstein MJ & Okayama H (1992). Structure and expression of a human oxytocin receptor. Nature 356, 526-529.

Kissebah AH, Sonnenberg GE, Myklebust J, Goldstein M, Broman K, James RG, Marks JA, Krakower GR, Jacob HJ, Weber J, Martin L, Blangero J & Comuzzie AG (2000). Quantitative trait loci on 3 and 17 influence phenotypes of the metabolic syndrome. Proc Natl Acad Sci U S A 97, 14478-14483.

Komai N, Morita Y, Sakuta T, Kuwabara A & Kashihara N (2007). Anti-tumor necrosis factor therapy increases serum adiponectin levels with the improvement of endothelial dysfunction in patients with . Mod Rheumatol 17, 385-390.

Kos K, Harte AL, da Silva NF, Tonchev A, Chaldakov G, James S, Snead DR, Hoggart B, O'Hare JP, McTernan PG & Kumar S (2007). Adiponectin and resistin in human cerebrospinal fluid and expression of adiponectin receptors in the human hypothalamus. J Clin Endocrinol Metab 92, 1129-1136.

Krause MP, Liu Y, Vu V, Chan L, Xu A, Riddell MC, Sweeney G & Hawke TJ (2008). Adiponectin is expressed by skeletal muscle fibers and influences muscle phenotype and function. Am J Physiol Cell Physiol 295, C203-C212.

Kubo T & Kihara M (1990). Modulation of the aortic baroreceptor reflex by neuropeptide Y, neurotensin and vasopressin microinjected into the nucleus tractus solitarii of the rat. Naunyn Schmiedebergs Arch Pharmacol 342, 182-188.

Kubota N, Terauchi Y, Yamauchi T, Kubota T, Moroi M, Matsui J, Eto K, Yamashita T, Kamon J, Satoh H, Yano W, Froguel P, Nagai R, Kimura S, Kadowaki T & Noda T (2002). Disruption of adiponectin causes insulin resistance and neointimal formation. J Biol Chem 277, 25863-25866.

Kubota N, Yano W, Kubota T, Yamauchi T, Itoh S, Kumagai H, Kozono H, Takamoto I, Okamoto S, Shiuchi T, Suzuki R, Satoh H, Tsuchida A, Moroi M, Sugi K, Noda T, Ebinuma H, Ueta Y, Kondo T, Araki E, Ezaki O, Nagai R, Tobe K, Terauchi Y, Ueki K, Minokoshi Y & Kadowaki T (2007). Adiponectin stimulates AMP-activated protein kinase in the hypothalamus and increases food intake. Cell Metab 6, 55-68.

158

Kusminski CM, McTernan PG, Schraw T, Kos K, O'Hare JP, Ahima R, Kumar S & Scherer PE (2007). Adiponectin complexes in human cerebrospinal fluid: distinct complex distribution from serum. Diabetologia 50, 634-642.

Lambolez B, Audinat E, Bochet P, Crepel F & Rossier J (1992). AMPA receptor subunits expressed by single Purkinje cells. Neuron 9, 247-258.

Latchford KJ & Ferguson AV (2003). Angiotensin II activates a nitric-oxide-driven inhibitory feedback in the rat paraventricular nucleus. J Neurophysiol 89, 1238-1244.

Lawrence AJ, Castillo-Melendez M, McLean KJ & Jarrott B (1998). The distribution of nitric oxide synthase-, adenosine deaminase- and neuropeptide Y-immunoreactivity through the entire rat nucleus tractus solitarius: Effect of unilateral nodose ganglionectomy. J Chem Neuroanat 15, 27-40.

Lawrence AJ & Jarrott B (1996). Neurochemical modulation of cardiovascular control in the nucleus tractus solitarius. Prog Neurobiol 48, 21-53.

Lechan RM & Fekete C (2006). The TRH neuron: a hypothalamic integrator of energy metabolism. Prog Brain Res 153, 209-235.

Lechan RM, Nestler JL, Jacobson S & Reichlin S (1980). The hypothalamic tuberoinfundibular system of the rat as demonstrated by horseradish peroxidase (HRP) microiontophoresis. Brain Res 195, 13-27.

Lechan RM & Segerson TP (1989). Pro-TRH gene expression and precursor peptides in rat brain. Observations by hybridization analysis and immunocytochemistry. Ann N Y Acad Sci 553, 29-59.

Lechan RM, Snapper SB & Jackson IMD (1983). Evidence that spinal cord thyrotropin-releasing hormone is independent of the paraventricular nucleus. Neurosci Lett 43, 61-65.

Leth H, Andersen KK, Frystyk J, Tarnow L, Rossing P, Parving HH & Flyvbjerg A (2008). Elevated levels of high-molecular-weight adiponectin in type 1 diabetes. J Clin Endocrinol Metab 93, 3186-3191.

159

Liu Y, Michael MD, Kash S, Bensch WR, Monia BP, Murray SF, Otto KA, Syed SK, Bhanot S, Sloop KW, Sullivan JM & Reifel-Miller A (2007). Deficiency of adiponectin receptor 2 reduces diet-induced insulin resistance but promotes type 2 diabetes. Endocrinology 148, 683-692.

Lowell BB & Spiegelman BM (2000). Towards a molecular understanding of adaptive thermogenesis. Nature 404, 652-660.

Ludwig M & Leng G (2006). Dendritic peptide release and peptide-dependent behaviours. Nat Rev Neurosci 7, 126-136.

Ludwig M & Pittman QJ (2003). Talking back: dendritic neurotransmitter release. Trends Neurosci 26, 255-261.

Ludwig M, Sabatier N, Bull PM, Landgraf R, Dayanithi G & Leng G (2002). Intracellular calcium stores regulate activity-dependent neuropeptide release from dendrites. Nature 418, 85- 89.

Luther JA, Daftary SS, Boudaba C, Gould GC, Halmos KC & Tasker JG (2002). Neurosecretory and non-neurosecretory parvocellular neurones of the hypothalamic paraventricular nucleus express distinct electrophysiological properties. J Neuroendocrinol 14, 929-932.

Luther JA & Tasker JG (2000). Voltage-gated currents distinguish parvocellular from magnocellular neurones in the rat hypothalamic paraventricular nucleus. J Physiol (Lond) 523 Pt 1, 193-209.

Ma K, Cabrero A, Saha PK, Kojima H, Li L, Chang BH, Paul A & Chan L (2002). Increased beta -oxidation but no insulin resistance or glucose intolerance in mice lacking adiponectin. J Biol Chem 277, 34658-34661.

Maddineni S, Metzger S, Ocon O, Hendricks G, III & Ramachandran R (2005). Adiponectin Gene Is Expressed in Multiple Tissues in the Chicken: Food Deprivation Influences Adiponectin Messenger Ribonucleic Acid Expression. Endocrinology 146, 4250-4256.

Maeda K, Okubo K, Shimomura I, Funahashi T, Matsuzawa Y & Matsubara K (1996). cDNA cloning and expression of a novel adipose specific collagen-like factor, apM1 (AdiPose Most abundant Gene transcript 1). Biochem Biophys Res Commun 221, 286-289.

160

Maeda N, Shimomura I, Kishida K, Nishizawa H, Matsuda M, Nagaretani H, Furuyama N, Kondo H, Takahashi M, Arita Y, Komuro R, Ouchi N, Kihara S, Tochino Y, Okutomi K, Horie M, Takeda S, Aoyama T, Funahashi T & Matsuzawa Y (2002). Diet-induced insulin resistance in mice lacking adiponectin/ACRP30. Nat Med 8, 731-737.

Malcher-Lopes R, Di S, Marcheselli VS, Weng FJ, Stuart CT, Bazan NG & Tasker JG (2006). Opposing Crosstalk between Leptin and Glucocorticoids Rapidly Modulates Synaptic Excitation via Endocannabinoid Release. J Neurosci 26, 6643-6650.

Mao X, Kikani CK, Riojas RA, Langlais P, Wang L, Ramos FJ, Fang Q, Christ-Roberts CY, Hong JY, Kim RY, Liu F & Dong LQ (2006). APPL1 binds to adiponectin receptors and mediates adiponectin signalling and function. Nat Cell Biol 8, 516-523.

Martins AS, Crescenzi A, Stern JE, Bordin S & Michelini LC (2005). Hypertension and exercise training differentially affect oxytocin and oxytocin receptor expression in the brain. Hypertension 46, 1004-1009.

McKinley MJ, McAllen RM, Davern P, Giles ME, Penschow J, Sunn N, Uschakov A & Oldfield BJ (2003). The sensory circumventricular organs of the mammalian brain. Adv Anat Embryol Cell Biol 172, III-122, back.

McLean HA, Champagnat J & avit-Saubie M (1996). A cholecystokinin-B potentiates GABAergic and glycinergic inhibition in the nucleus of the solitary tract of the rat. Eur J Neurosci 8, 1078-1084.

Meister B, Hökfelt T, Geffard M & Oertel W (1988). Glutamic acid decarboxylase- and gamma- aminobutyric acid-like immunoreactivities in corticotropin-releasing factor-containing parvocellular neurons of the hypothalamic paraventriculr nucleus. Neuroendocrinology 48, 516- 526.

Mens WB, Witter A & van Wimersma Greidanus TB (1983). Penetration of neurohypophyseal hormones from plasma into cerebrospinal fluid (CSF): half-times of disappearance of these neuropeptides from CSF. Brain Res 262, 143-149.

Menzaghi C, Ercolino T, Di PR, Berg AH, Warram JH, Scherer PE, Trischitta V & Doria A (2002). A haplotype at the adiponectin locus is associated with obesity and other features of the insulin resistance syndrome. Diabetes 51, 2306-2312.

161

Mezey E & Kiss JZ (1991). Coexpression of vasopressin and oxytocin in hypothalamic supraoptic neurons of lactating rats. Endocrinology 129, 1814-1820.

Miki T, Liss B, Minami K, Shiuchi T, Saraya A, Kashima Y, Horiuchi M, Ashcroft F, Minokoshi Y, Roeper J & Seino S (2001). ATP-sensitive K+ channels in the hypothalamus are essential for the maintenance of glucose homeostasis. Nat Neurosci 4, 507-512.

Minokoshi Y, Alquier T, Furukawa N, Kim YB, Lee A, Xue B, Mu J, Foufelle F, Ferre P, Birnbaum MJ, Stuck BJ & Kahn BB (2004). AMP-kinase regulates food intake by responding to hormonal and nutrient signals in the hypothalamus. Nature 428, 569-574. miry-Moghaddam M & Ottersen OP (2003). The molecular basis of water transport in the brain. Nat Rev Neurosci 4, 991-1001.

Misra M, Miller KK, Cord J, Prabhakaran R, Herzog DB, Goldstein M, Katzman DK & Klibanski A (2007). Relationships between serum adipokines, insulin levels, and bone density in girls with anorexia nervosa. J Clin Endocrinol Metab 92, 2046-2052.

Mitsuuchi Y, Johnson SW, Sonoda G, Tanno S, Golemis EA & Testa JR (1999). Identification of a chromosome 3p14.3-21.1 gene, APPL, encoding an adaptor molecule that interacts with the oncoprotein-serine/threonine kinase AKT2. Oncogene 18, 4891-4898.

Moller DE & Kaufman KD (2005). Metabolic syndrome: a clinical and molecular perspective. Annu Rev Med 56, 45-62.

Morton GJ, Blevins JE, Williams DL, Niswender KD, Gelling RW, Rhodes CJ, Baskin DG & Schwartz MW (2005). Leptin action in the forebrain regulates the hindbrain response to satiety signals. J Clin Invest 115, 703-710.

Murphy KG & Bloom SR (2006). Gut hormones and the regulation of energy homeostasis. Nature 444, 854-859.

Nakano Y, Tobe T, Choi-Miura NH, Mazda T & Tomita M (1996). Isolation and characterization of GBP28, a novel gelatin-binding protein purified from human plasma. J Biochem (Tokyo) 120, 803-812.

162

Norsted E, Gomuc B & Meister B (2008). Protein components of the blood-brain barrier (BBB) in the mediobasal hypothalamus. J Chem Neuroanat 36, 107-121.

Northway MG, Morris M, Geisinger KR & MacLean DB (1989). Effects of a gastric implant on body weight and gastrointestinal hormones in cafeteria diet obese rats. Physiol Behav 45, 331- 335.

O'Dowd DK & Smith MA (1996). Single-cell analysis of gene expression in the nervous system. Measurements at the edge of chaos. Mol Neurobiol 13, 199-211.

Ohashi K, Kihara S, Ouchi N, Kumada M, Fujita K, Hiuge A, Hibuse T, Ryo M, Nishizawa H, Maeda N, Maeda K, Shibata R, Walsh K, Funahashi T & Shimomura I (2006). Adiponectin replenishment ameliorates obesity-related hypertension. Hypertension 47, 1108-1116.

Olson BR, Drutarosky MD, Stricker EM & Verbalis JG (1991a). Brain oxytocin receptor antagonism blunts the effects of anorexigenic treatments in rats: evidence for central oxytocin inhibition of food intake. Endocrinology 129, 785-791.

Olson BR, Drutarosky MD, Stricker EM & Verbalis JG (1991b). Brain oxytocin receptors mediate corticotropin-releasing hormone- induced anorexia. Am J Physiol 260, R448-R452.

Olson BR, Hoffman GE, Sved AF, Stricker EM & Verbalis JG (1992). Cholecytokinin induces c- fos expression in hypothalamic oxytocinergic neurons projecting to the dorsal vagal complex. Brain Res 569, 238-248.

Olszewski PK, Wirth MM, Shaw TJ, Grace MK & Levine AS (2003). Peptides that regulate food intake: effect of peptide histidine isoleucine on consummatory behavior in rats. Am J Physiol Regul Integr Comp Physiol 284, R1445-R1453.

Ouchi N, Kihara S, Arita Y, Nishida M, Matsuyama A, Okamoto Y, Ishigami M, Kuriyama H, Kishida K, Nishizawa H, Hotta K, Muraguchi M, Ohmoto Y, Yamashita S, Funahashi T & Matsuzawa Y (2001). Adipocyte-derived plasma protein, adiponectin, suppresses lipid accumulation and class A scavenger receptor expression in human monocyte-derived macrophages. Circulation 103, 1057-1063.

Owens MJ & Nemeroff CB (1991). Physiology and pharmacology of corticotropin-releasing factor. Pharmacol Rev 43, 425-473.

163

Palkovits M, Mezey E, Zaborszky L, Feminger A, Versteeg DHG, Wijnen HJLM, DeJong W, Fekete MIK, Herman JP & Kanyicska B (1980a). Adrenergic innervation of the rat hypothalamus. Neurosci Lett 18, 237-242.

Palkovits M, Zaborszky L, Feminger A, Mezey E, Fekete MIK, Herman JP, Kanyicska B & Szabo D (1980b). Noradrenergic innervation of the rat hypothalamus: experimental biochemical and electron microscopic studies. Brain Res 191, 161-171.

Pan W, Tu H & Kastin AJ (2006). Differential BBB interactions of three ingestive peptides: , ghrelin, and adiponectin. Peptides 27, 911-916.

Paquet L, Massie B & Mains RE (1996). Proneuropeptide Y processing in large dense-core vesicles: manipulation of prohormone convertase expression in sympathetic neurons using adenoviruses. J Neurosci 16, 964-973.

Patlak CS, Blasberg RG & Fenstermacher JD (1983). Graphical evaluation of blood-to-brain transfer constants from multiple-time uptake data. J Cereb Blood Flow Metab 3, 1-7.

Pineiro R, Iglesias MJ, Gallego R, Raghay K, Eiras S, Rubio J, Dieguez C, Gualillo O, Gonzalez- Juanatey JR & Lago F (2005). Adiponectin is synthesized and secreted by human and murine cardiomyocytes. FEBS Lett 579, 5163-5169.

Powis JE, Bains JS & Ferguson AV (1997). Leptin depolarizes rat hypothalamic paraventricular neurons. Am J Physiol 274, R14682-R1472.

Price CJ, Hoyda TD & Ferguson AV (2008a). The area postrema: a brain monitor and integrator of systemic autonomic state. Neuroscientist 14, 182-194.

Price CJ, Hoyda TD, Samson WK & Ferguson AV (2008b). Nesfatin-1 influences the excitability of paraventricular nucleus neurones. J Neuroendocrinol 20, 245-250.

Psilopanagioti A, Papadaki H, Kranioti EF, Alexandrides TK & Varakis JN (2009). Expression of adiponectin and adiponectin receptors in human pituitary gland and brain. Neuroendocrinology 89, 38-47.

Punyadeera C, Zorenc AH, Koopman R, McAinch AJ, Smit E, Manders R, Keizer HA, Cameron- Smith D & van Loon LJ (2005). The effects of exercise and adipose tissue lipolysis on plasma

164

adiponectin concentration and adiponectin receptor expression in human skeletal muscle. Eur J Endocrinol 152, 427-436.

Qi Y, Takahashi N, Hileman SM, Patel HR, Berg AH, Pajvani UB, Scherer PE & Ahima RS (2004). Adiponectin acts in the brain to decrease body weight. Nat Med 10, 524-529.

Qiu DL, Chu CP, Shirasaka T, Tsukino H, Nakao H, Kato K, Kunitake T, Katoh T & Kannan H (2005). Corticotrophin-releasing factor augments the I(H) in rat hypothalamic paraventricular nucleus parvocellular neurons in vitro. J Neurophysiol 94, 226-234.

Rao VV, Loffler C, Battey J & Hansmann I (1992). The human gene for oxytocin-neurophysin I (OXT) is physically mapped to chromosome 20p13 by in situ hybridization. Cytogenet Cell Genet 61, 271-273.

Richardson PJ, Dixon AK, Lee K, Bell MI, Cox PJ, Williams R, Pinnock RD & Freeman TC (2000). Correlating physiology with gene expression in striatal cholinergic neurones. J Neurochem 74, 839-846.

Rinaman L (1998). Oxytocinergic inputs to the nucleus of the solitary tract and dorsal motor nucleus of the vagus in neonatal rats. J Comp Neurol 399, 101-109.

Rivier C & Vale W (1984a). Corticotropin-releasing factor (CRF) acts centrally to inhibit growth hormone secretion in the rat. Endocrinology 114, 2409-2411.

Rivier C & Vale W (1984b). Influence of corticotropin-releasing factor on reproductive functions in the rat. Endocrinology 114, 914-921.

Rodriguez-Pacheco F, Martinez-Fuentes AJ, Tovar S, Pinilla L, Tena-Sempere M, Dieguez C, Castano JP & Malagon MM (2007). Regulation of pituitary cell function by adiponectin. Endocrinology 148, 401-410.

Roland BL & Sawchenko PE (1993). Local origins of some GABAergic projections to the paraventricular and supraoptic nuclei of the hypothalamus in the rat. The Journal of Comparative Neurology 332, 123-143.

Ruderman NB, Saha AK & Kraegen EW (2003). Minireview: malonyl CoA, AMP-activated protein kinase, and adiposity. Endocrinology 144, 5166-5171.

165

Saito K, Tobe T, Minoshima S, Asakawa S, Sumiya J, Yoda M, Nakano Y, Shimizu N & Tomita M (1999). Organization of the gene for gelatin-binding protein (GBP28). Gene 229, 67-73.

Saper CB (2002). The central autonomic nervous system: conscious visceral perception and autonomic pattern generation. Annu Rev Neurosci 25, 433-469.

Saper CB, Loewy AD, Swanson LW & Cowan WM (1976). Direct hypothalamo-autonomic connections. Brain Res 117, 305-312.

Saper CB, Swanson LW & Cowan WM (1978). The efferent connections of the anterior hypothalamic area of the rat, cat and monkey. J Comp Neurol 182, 575-599.

Saper CB, Swanson LW & Cowan WM (1979). An autoradiographic study of the efferent connections of the lateral hypothalamic area in the rat. J Comp Neurol 183, 689-706.

Sawchenko PE & Swanson LW (1983). The organization of forebrain afferents to the paraventricular and supraoptic nuclei in the rat. J Comp Neurol 218, 121-144.

Sawchenko PE, Swanson LW, Grzanna R, Howe PRC, Bloom SR & Polak JM (1985). Colocalization of neuropeptide Y immunoreactivity in brainstem catecholaminergic neurons that project to the paraventricular nucleus of the hypothalamus. J Comp Neurol 241, 138-153.

Sawchenko PE, Swanson LW & Joseph SA (1982). The distribution and cells of origin of ACTH(1-39)-stained varicosities in the paraventricular and supraoptic nuclei. Brain Res 232, 365-374.

Sawchenko PE, Swanson LW & Vale WW (1984). Co-expression of corticotropin-releasing factor and vasopressin immunoreactivity in parvocellular neurosecretory neurons of the adrenalectomized rat. Proc Natl Acad Sci USA 81, 1883-1887.

Scherer PE (2006). Adipose tissue: from lipid storage compartment to endocrine organ. Diabetes 55, 1537-1545.

Scherer PE, Williams S, Fogliano M, Baldini G & Lodish HF (1995). A novel serum protein similar to C1q, produced exclusively in adipocytes. J Biol Chem 270, 26746-26749.

166

Schwartz MW, Woods SC, Porte D, Jr., Seeley RJ & Baskin DG (2000). Central nervous system control of food intake. Nature 404, 661-671.

Shapiro L & Scherer PE (1998). The crystal structure of a complement-1q family protein suggests an evolutionary link to tumor necrosis factor. Curr Biol 8, 335-338.

Shetty S, Kusminski CM & Scherer PE (2009). Adiponectin in health and disease: evaluation of adiponectin-targeted drug development strategies. Trends Pharmacol Sci 30, 234-239.

Singhal NS, Lazar MA & Ahima RS (2007). Central resistin induces hepatic insulin resistance via neuropeptide Y. J Neurosci 27, 12924-12932.

Smith DE, Johanson CE & Keep RF (2004). Peptide and peptide analog transport systems at the blood-CSF barrier. Adv Drug Deliv Rev 56, 1765-1791.

Sobolevsky AI (1999). Two-component blocking kinetics of open NMDA channels by organic cations. Biochim Biophys Acta 1416, 69-91.

Spanswick D, Smith MA, Mirshamsi S, Routh VH & Ashford MLJ (2000). Insulin activates ATP-sensitive K+ channels in hypothalamic neurons of lean, but not obese rats. Nat Neurosci 3, 757-758.

Spanswick R, Smith MA, Groppi VE, Logan SD & Ashford MLJ (1997). Leptin inhibits hypothalamic neurons by activation of ATP-sensitive potassium channels. Nature 390, 521-525.

Spiegelman BM & Flier JS (2001). Obesity and the regulation of energy balance. Cell 104, 531- 543.

Spranger J, Verma S, Gohring I, Bobbert T, Seifert J, Sindler AL, Pfeiffer A, Hileman SM, Tschop M & Banks WA (2006). Adiponectin does not cross the blood-brain barrier but modifies cytokine expression of brain endothelial cells. Diabetes 55, 141-147.

Stanley BG & Leibowitz SF (1984). Neuropeptide Y: stimulation of feeding and drinking by injection into the paraventricular nucleus. Life Sci 35, 2635-2642.

167

Stern JE (2001). Electrophysiological and morphological properties of pre-autonomic neurones in the rat hypothalamic paraventricular nucleus. J Physiol 537, 161-177.

Steyn FJ, Boehme F, Vargas E, Wang K & Chen C (2009). Adiponectin regulate growth hormone secretion via adiponectin receptor mediated Ca signalling in rat somatotrophs in vitro. J Neuroendocrinol.

Stock S, Granström L, Backman L, Matthiesen AS & Uvnäs-Moberg K (1989). Elevated plasma levels of oxytocin in obese subjects before and after gastric banding. Int J Obes 13, 213-222.

Stumvoll M, Tschritter O, Fritsche A, Staiger H, Renn W, Weisser M, Machicao F & Haring H (2002). Association of the T-G polymorphism in adiponectin (exon 2) with obesity and insulin sensitivity: interaction with family history of type 2 diabetes. Diabetes 51, 37-41.

Swaab DF, Nijveldt F & Pool CW (1975a). Distribution of oxytocin and vasopressin in the rat supraoptic and paraventricular nucleus. J Endocrinol 67, 461-462.

Swaab DF, Pool CW & Nijveldt F (1975b). Immunofluorescence of vasopressin and oxytocin in the rat hypothalamo-neurohypophyseal system. J Neural Transm 36, 195-215.

Swanson LW & Cowan WM (1975). The efferent connections of the suprachiasmatic nucleus of the hypothalamus. J Comp Neurol 160, 1-12.

Swanson LW & Hartman BK (1980). Biochemical specificity in central pathways related to peripheral and intracerebral homeostatic functions. Neurosci Lett 16, 55-60.

Swanson LW & Sawchenko PE (1980). Paraventricular nucleus: a site for the integration of neuroendocrine and autonomic mechanisms. Neuroendocrinology 31, 410-417.

Swanson LW & Sawchenko PE (1983). Hypothalamic integration: organization of the paraventricular and supraoptic nuclei. Ann Rev Neurosci 6, 269-324.

Swanson LW, Sawchenko PE, Berod A, Hartman BK, Helle KB & Vanorden DE (1981). An immunohistochemical study of the organization of catecholaminergic cells and terminal fields in the paraventricular and supraoptic nuclei of the hypothalamus. J Comp Neurol 196, 271-285.

168

Tan KC, Xu A, Chow WS, Lam MC, Ai VH, Tam SC & Lam KS (2004). Hypoadiponectinemia is associated with impaired -dependent vasodilation. J Clin Endocrinol Metab 89, 765-769.

Tanida M, Shen J, Horii Y, Matsuda M, Kihara S, Funahashi T, Shimomura I, Sawai H, Fukuda Y, Matsuzawa Y & Nagai K (2007). Effects of adiponectin on the renal sympathetic nerve activity and blood pressure in rats. Exp Biol Med (Maywood ) 232, 390-397.

Tasker JG & Dudek FE (1991). Electrophysiological properties of neurones in the region of the paraventricular nucleus in slices of rat hypothalamus. J Physiol (Lond ) 434, 271-293.

Thompson RC, Seasholtz AF & Herbert E (1987). Rat corticotropin-releasing hormone gene: sequence and tissue-specific expression. Mol Endocrinol 1, 363-370.

Toth ZE, Gallatz K, Fodor M & Palkovits M (1999). Decussations of the descending paraventricular pathways to the brainstem and spinal cord autonomic centers. J Comp Neurol 414, 255-266.

Tribollet E & Dreifuss JJ (1981). Localization of neurones projecting to the hypothalamic paraventricular nucleus area of the rat: a horseradish peroxidase study. Neuroscience 6, 1315- 1328.

Tseng CJ, Mosqueda-Garcia R, Appalsamy M & Robertson D (1989). Cardiovascular effects of neuropeptide Y in rat brainstem nuclei. Circ Res 64, 55-61.

Tsuchida A, Yamauchi T, Ito Y, Hada Y, Maki T, Takekawa S, Kamon J, Kobayashi M, Suzuki R, Hara K, Kubota N, Terauchi Y, Froguel P, Nakae J, Kasuga M, Accili D, Tobe K, Ueki K, Nagai R & Kadowaki T (2004). Insulin/Foxo1 pathway regulates expression levels of adiponectin receptors and adiponectin sensitivity. J Biol Chem 279, 30817-30822.

Tsuchida A, Yamauchi T, Takekawa S, Hada Y, Ito Y, Maki T & Kadowaki T (2005). Peroxisome proliferator-activated receptor (PPAR)alpha activation increases adiponectin receptors and reduces obesity-related inflammation in adipose tissue: comparison of activation of PPARalpha, PPARgamma, and their combination. Diabetes 54, 3358-3370.

Tucci SA, Rogers EK, Korbonits M & Kirkham TC (2004). The cannabinoid CB1 receptor antagonist SR141716 blocks the orexigenic effects of intrahypothalamic ghrelin. Br J Pharmacol 143, 520-523. 169

Ueta Y, Kannan H, Higuchi T, Negoro H & Yamashita H (1993). CCK-8 excites oxytocin- secreting neurons in the paraventricular nucleus in rats--possible involvement of noradrenergic pathway. Brain Res Bull 32, 453-459.

Ulrich-Lai YM & Herman JP (2009). Neural regulation of endocrine and autonomic stress responses. Nat Rev Neurosci 10, 397-409.

Ur E, Wilkinson DA, Morash BA & Wilkinson M (2002). Leptin immunoreactivity is localized to neurons in rat brain. Neuroendocrinology 75, 264-272.

Vale W, Spiess J, Rivier C & Rivier J (1981). Characterization of a 41-residue ovine hypothalamic peptide that stimulates secretion of corticotropin and beta-endorphin. Science 213, 1394-1397. van den Pol AN (1982). The magnocellular and parvocellular paraventricular nucleus of rat: intrinsic organization. J Comp Neurol 206, 317-345.

Vandesande F & Dierickx K (1975). Identification of vasopressin producing and of the oxytocin producing neurons in the hypothalamic magnocellular neurosecretory system of the rat. Cell Tiss Res 164, 153-162.

Vandesande F, Dierickx K & De MJ (1977). The origin of the vasopressinergic and oxytocinergic fibres of the external region of the median eminence of the rat hypophysis. Cell Tissue Res 180, 443-452.

Vasseur F, Helbecque N, Dina C, Lobbens S, Delannoy V, Gaget S, Boutin P, Vaxillaire M, Lepretre F, Dupont S, Hara K, Clement K, Bihain B, Kadowaki T & Froguel P (2002). Single- nucleotide polymorphism haplotypes in the both proximal promoter and exon 3 of the APM1 gene modulate adipocyte-secreted adiponectin hormone levels and contribute to the genetic risk for type 2 diabetes in French Caucasians. Hum Mol Genet 11, 2607-2614.

Verbalis JG, Blackburn RE, Olson BR & Stricker EM (1993). Central oxytocin inhibition of food and salt ingestion: a mechanism for intake regulation of solute homeostasis. Regul Pept 45, 149- 154.

Viengchareun S, Zennaro MC, Pascual-Le TL & Lombes M (2002). Brown adipocytes are novel sites of expression and regulation of adiponectin and resistin. FEBS Lett 532, 345-350.

170

Vincent A & Tell F (1997). Postnatal changes in electrophysiological properties of rat nucleus tractus solitarii neurons. Eur J Neurosci 9, 1612-1624.

Vionnet N, Hani EH, Dupont S, Gallina S, Francke S, Dotte S, De MF, Durand E, Lepretre F, Lecoeur C, Gallina P, Zekiri L, Dina C & Froguel P (2000). Genomewide search for type 2 diabetes-susceptibility genes in French whites: evidence for a novel susceptibility locus for early- onset diabetes on chromosome 3q27-qter and independent replication of a type 2-diabetes locus on chromosome 1q21-q24. Am J Hum Genet 67, 1470-1480.

Vrang N, Larsen PJ, Kristensen P & Tang-Christensen M (2000). Central administration of cocaine-amphetamine-regulated transcript activates hypothalamic neuroendocrine neurons in the rat. Endocrinology 141, 794-801.

Waki H, Yamauchi T, Kamon J, Ito Y, Uchida S, Kita S, Hara K, Hada Y, Vasseur F, Froguel P, Kimura S, Nagai R & Kadowaki T (2003). Impaired multimerization of human adiponectin mutants associated with diabetes. Molecular structure and multimer formation of adiponectin. J Biol Chem 278, 40352-40363.

Waki H, Yamauchi T, Kamon J, Kita S, Ito Y, Hada Y, Uchida S, Tsuchida A, Takekawa S & Kadowaki T (2005). Generation of globular fragment of adiponectin by leukocyte elastase secreted by monocytic cell line THP-1. Endocrinology 146, 790-796.

Walum H, Westberg L, Henningsson S, Neiderhiser JM, Reiss D, Igl W, Ganiban JM, Spotts EL, Pedersen NL, Eriksson E & Lichtenstein P (2008). Genetic variation in the vasopressin receptor 1a gene (AVPR1A) associates with pair-bonding behavior in humans. Proc Natl Acad Sci U S A 105, 14153-14156.

Wang ZV & Scherer PE (2008). Adiponectin, cardiovascular function, and hypertension. Hypertension 51, 8-14.

Washburn DLS, Beedle AM & Ferguson AV (1999). Inhibition of subfornical organ neuronal potassium channels by vasopressin. Neuroscience 93, 349-359.

Wen JP, Lv WS, Yang J, Nie AF, Cheng XB, Yang Y, Ge Y, Li XY & Ning G (2008). Globular adiponectin inhibits GnRH secretion from GT1-7 hypothalamic GnRH neurons by induction of hyperpolarization of membrane potential. Biochem Biophys Res Commun 371, 756-761.

171

Wilkinson M, Brown R, Imran SA & Ur E (2007). Adipokine gene expression in brain and pituitary gland. Neuroendocrinology 86, 191-209.

Williams DL, Baskin DG & Schwartz MW (2006). Leptin regulation of the anorexic response to glucagon-like Peptide-1 receptor stimulation. Diabetes 55, 3387-3393.

Wittmann G, Lechan RM, Liposits Z & Fekete C (2005). Glutamatergic innervation of corticotropin-releasing hormone- and thyrotropin-releasing hormone-synthesizing neurons in the hypothalamic paraventricular nucleus of the rat. Brain Res 1039, 53-62.

Wong GW, Wang J, Hug C, Tsao TS & Lodish HF (2004). A family of Acrp30/adiponectin structural and functional paralogs. Proc Natl Acad Sci U S A 101, 10302-10307.

Wyatt CN, Mustard KJ, Pearson SA, Dallas ML, Atkinson L, Kumar P, Peers C, Hardie DG & Evans AM (2007). AMP-activated protein kinase mediates carotid body excitation by hypoxia. J Biol Chem 282, 8092-8098.

Wynn PC, Aguilera G, Morell J & Catt KJ (1983). Properties and regulation of high-affinity pituitary receptors for corticotropin-releasing factor. Biochem Biophys Res Commun 110, 602- 608.

Xi D, Kusano K & Gainer H (1999). Quantitative analysis of oxytocin and vasopressin messenger ribonucleic acids in single magnocellular neurons isolated from supraoptic nucleus of rat hypothalamus. Endocrinology 140, 4677-4682.

Xu A, Chan KW, Hoo RL, Wang Y, Tan KC, Zhang J, Chen B, Lam MC, Tse C, Cooper GJ & Lam KS (2005). Testosterone selectively reduces the high molecular weight form of adiponectin by inhibiting its secretion from adipocytes. J Biol Chem 280, 18073-18080.

Yamamoto Y, Hirose H, Saito I, Nishikai K & Saruta T (2004). Adiponectin, an adipocyte- derived protein, predicts future insulin resistance: two-year follow-up study in Japanese population. J Clin Endocrinol Metab 89, 87-90.

Yamauchi T, Kamon J, Ito Y, Tsuchida A, Yokomizo T, Kita S, Sugiyama T, Miyagishi M, Hara K, Tsunoda M, Murakami K, Ohteki T, Uchida S, Takekawa S, Waki H, Tsuno NH, Shibata Y, Terauchi Y, Froguel P, Tobe K, Koyasu S, Taira K, Kitamura T, Shimizu T, Nagai R & Kadowaki T (2003). Cloning of adiponectin receptors that mediate antidiabetic metabolic effects. Nature 423, 762-769. 172

Yamauchi T, Kamon J, Minokoshi Y, Ito Y, Waki H, Uchida S, Yamashita S, Noda M, Kita S, Ueki K, Eto K, Akanuma Y, Froguel P, Foufelle F, Ferre P, Carling D, Kimura S, Nagai R, Kahn BB & Kadowaki T (2002). Adiponectin stimulates glucose utilization and fatty-acid oxidation by activating AMP-activated protein kinase. Nat Med 8, 1288-1295.

Yamauchi T, Kamon J, Waki H, Terauchi Y, Kubota N, Hara K, Mori Y, Ide T, Murakami K, Tsuboyama-Kasaoka N, Ezaki O, Akanuma Y, Gavrilova O, Vinson C, Reitman ML, Kagechika H, Shudo K, Yoda M, Nakano Y, Tobe K, Nagai R, Kimura S, Tomita M, Froguel P & Kadowaki T (2001). The fat-derived hormone adiponectin reverses insulin resistance associated with both lipoatrophy and obesity. Nat Med 7, 941-946.

Yamauchi T, Nio Y, Maki T, Kobayashi M, Takazawa T, Iwabu M, Okada-Iwabu M, Kawamoto S, Kubota N, Kubota T, Ito Y, Kamon J, Tsuchida A, Kumagai K, Kozono H, Hada Y, Ogata H, Tokuyama K, Tsunoda M, Ide T, Murakami K, Awazawa M, Takamoto I, Froguel P, Hara K, Tobe K, Nagai R, Ueki K & Kadowaki T (2007). Targeted disruption of AdipoR1 and AdipoR2 causes abrogation of adiponectin binding and metabolic actions. Nat Med 13, 332-339.

Yang B, Brown KK, Chen L, Carrick KM, Clifton LG, McNulty JA, Winegar DA, Strum JC, Stimpson SA & Pahel GL (2004). Serum adiponectin as a biomarker for in vivo PPARgamma activation and PPARgamma agonist-induced efficacy on insulin sensitization/lipid lowering in rats. BMC Pharmacol 4, 23.

Yoon MJ, Lee GY, Chung JJ, Ahn YH, Hong SH & Kim JB (2006). Adiponectin increases fatty acid oxidation in skeletal muscle cells by sequential activation of AMP-activated protein kinase, p38 mitogen-activated protein kinase, and peroxisome proliferator-activated receptor alpha. Diabetes 55, 2562-2570.

Zhang Y, Proenca R, Maffei M, Barone M, Leopold L & Friedman JM (1994). Positional cloning of the mouse obese gene and its human homologue. Nature 372, 425-432.

173