<%BANNER%>

Potent Anorexic and Lipopenic Effects of Central Leptin Gene Therapy Are Blocked by Diet-Induced Obesity: Evidence for ...


PAGE 1

POTENT ANOREXIC AND LIPOPENIC EFFECTS OF CENTRAL LEPTIN GENE THERAPY ARE BLOCKED BY DIET-INDUCED OBESITY: EVIDENCE FOR IMPAIRED LEPTIN RECEPTOR EXPR ESSION/SIGNAL TRANSDUCTION IN OBESITY AND REVERSAL BY CALORIC RESTRICTION By JARED TIMOTHY WILSEY A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2003

PAGE 2

Copyright 2003 by Jared Timothy Wilsey

PAGE 3

This dissertation is dedica ted my mother, Christine.

PAGE 4

iv ACKNOWLEDGMENTS The research described in this disserta tion was made possible through the help and support of University of Flor ida faculty, fellow graduate st udents, post-docs, scientists, and other collaborators of the Philip J. Scarp ace laboratory. First, I would like to thank Dr. Scarpace for both allowing me to conduct a rotation project in his lab in the fall of 1999, and then inviting me to join his la boratory group in May 2000. Dr. Scarpace has been a pleasure to work for. He has encouraged independence, which has greatly enhanced my development as a student and sc ientist. Whenever needed, his advice and direction on experimental desi gn and the development of nov el research hypotheses has been invaluable. In addition, his counsel has enhanced my ability to both prepare manuscripts and, more importantly, to a ppropriately respond to reviewers. All members of Dr. Scarpace’s lab, both past and present, have contributed to my education and these projects. I thank Michael Matheny, who provided invaluable technical advice on a daily basis despite his ow n busy research schedule. Mike has also been a lot of fun to work with, and we are all lucky to have him around. Dr. Yi “Edi” Zhang has been enormously helpful with bot h guidance on scientific issues and with helping to keep me caught up on the literature. My fellow graduate student, Dr. Gang Li, has likewise given me some great advice ove r the years and has calmed me down on numerous occasions when I was convinced I ru ined an assay or experiment. All of Dr. Scarpace’s past technicians have been of assistance as well including “Thomas” Kit-Yan

PAGE 5

v Tang (the best rat weigher in the business), Min Li, and Constanza Diana Victoria Frase (yes, that is all one name). Several faculty members have helped me over the years. Dr. Nihal Tumer, who frequently collaborates with Dr. Scarpace, has been a great positive influence in the laboratories and is a truly kind woman. She also makes a deliciously potent cup of Turkish coffee. I thank my other committee members for their great suggestions and comments over the last couple of years, incl uding Dr. Charles Wood, Dr. Daniel Driscoll, and Dr. Steven Borst. Dr. Wood, along with Dr. David Weiner in the VA Hospital, were very helpful in getting me st arted on real-time PCR when I ha d to quickly repeat all of my expression data. Dr. Borst was one of th e first College of Medicine Faculty I met, way back in 1998 he gave me a lesson on rat brain excision and anatomy. Dr. Driscoll has enlightened me as to the clinical signifi cance of obesity researc h, particularly in the treatment of the severe, early-onset obesity caused by specific genetic defects. He explained that in such cases, gene delive ry—even into the central nervous system—may prove to be a reasonable st rategy. Dr. Scarpace’s co llaboration with Dr. Sergei Zolotukhin and Dr. Victor Prima was of criti cal importance in all of the gene therapy experiments. I thank them both for their generous provision of vectors and related materials, as well as their intellectual contributions to these projects. Most importantly, I would lik e to thank my friends and family for giving me moral support and making me laugh over the years. I thank my mom and brother Darren for being sources of first-class advice and positive words. My mother, especially, has been there for me through it all: track meets, football games, 40 yard dash time trials ad nauseum weightlifting and powerlifting competiti ons—she really is a Super-mom and a

PAGE 6

vi great friend. I also thank my dad, my grandpa rents, aunts and uncles, and my little halfsisters. I thank Amie Dirks, my best friend during the graduate school years and a wonderful person. I thank my “industry cont act” and friend, Mike Ferguson. Last but not least, I thank all my friends and traini ng partners from Gainesville Gym, including Anthony (an ever-aspiring Olympian) and Josh.

PAGE 7

vii TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES............................................................................................................xii LIST OF FIGURES.........................................................................................................xiii ABSTRACT.....................................................................................................................xv i CHAPTER 1 BACKGROUND AND SIGNIFICANCE....................................................................1 Leptin......................................................................................................................... ...1 The Leptin Receptor.....................................................................................................3 Isoforms.................................................................................................................3 Leptin Binding to Leptin Receptor........................................................................4 Leptin Receptor Signal Transduction...........................................................................5 JAK/STAT Pathway..............................................................................................5 ERK-c-fos Pathway...............................................................................................6 PIA3 Kinase and Insulin-Like Signaling...............................................................6 Negative Regulators of Leptin Signaling..............................................................8 Down-Stream Leptin Signaling and Neurope ptide Regulation in the Hypothalamus..9 Arcuate Nucleus....................................................................................................9 Projections from the Arcuate...............................................................................12 Extra-Hypothalamic Signaling and Energy Balance..................................................13 Other Molecules Regulat ing Energy Balance.............................................................14 Insulin..................................................................................................................14 Glucose................................................................................................................16 Orexins/Hypocretins............................................................................................16 Ghrelin.................................................................................................................17 Uncoupling Proteins...................................................................................................18 Leptin Resistance........................................................................................................20 Adeno-Associated Virus and Gene Therapy..............................................................21 Wild-Type AAV..................................................................................................21 Recombinant Adeno-Associated Virus and Gene Delivery................................23 Experimental Design and Rational.............................................................................24 Incorporating Regulation into Leptin Gene Therapy..........................................24 Central Leptin Over-Expression in Diet-Induced Obese Animals......................25

PAGE 8

viii Effects of Diet-Induced Obesity and Ca loric Restriction on Leptin Receptor Expression and Signal Transduction Capacity................................................26 Chapter Summary and Conclusions............................................................................26 2 GENERAL METHODS AND MATERIALS............................................................28 Experimental Animals................................................................................................28 Construction of rAAV Vector Plasmid.......................................................................28 Packaging of rAAV Vectors.......................................................................................29 Stereotaxic Injections..................................................................................................29 Third Cerebroventricle........................................................................................29 Hypothalamic......................................................................................................30 Oxygen Consumption.................................................................................................30 Tissue Harvesting.......................................................................................................31 Serum Measurements..................................................................................................31 Leptin...................................................................................................................31 Insulin..................................................................................................................31 Glucose................................................................................................................31 Free Fatty Acids..................................................................................................31 CSF Leptin...........................................................................................................32 Probes......................................................................................................................... 32 RNA Isolation and RNA Dot Blot..............................................................................33 Relative-Quantitative RT-P CR Using 18S Competimers...........................................33 Real-time RT-PCR for Leptin Receptor.....................................................................34 STAT3/Phospo-STAT3 Assay...................................................................................35 UCP-1 Protein in Brown Adipose Tissue...................................................................36 Statistical Analysis......................................................................................................36 3 INCORPORATING REGULATION IN TO LEPTIN GENE THERAPY................38 Introduction.................................................................................................................38 Methods and Materials...............................................................................................39 Animals................................................................................................................39 Construction of rAAV Vector Plasmid...............................................................40 Packaging of rAAV vectors.................................................................................41 Vector Administration.........................................................................................41 Third ventricle injection...............................................................................41 Hypothalamic injection................................................................................46 Experimental Design...........................................................................................46 Experiment 1................................................................................................46 Experiment 2................................................................................................47 Tissue Harvesting................................................................................................47 Serum Leptin, FFA, Insulin, and Glucose...........................................................47 Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR)..........................48 STAT3/Phospo-STAT3 Assay............................................................................48 Probes..................................................................................................................49 mRNA Levels (Dot Blot Analysis)......................................................................49

PAGE 9

ix Statistical Analysis ..............................................................................................50 Results........................................................................................................................ .51 Experiment 1.......................................................................................................51 Food consumption and body mass...............................................................51 Serum leptin and adiposity...........................................................................51 Experiment 2.......................................................................................................52 Food consumption and body mass...............................................................52 Serum leptin and adiposity...........................................................................53 Serum free fatty acids, insulin, and glucose.................................................53 Leptin expression.........................................................................................53 Signal transduction in hypothalamus...........................................................54 Leptin receptor expression in the hypothalamus..........................................54 CSF leptin.....................................................................................................54 Brown adipose tissue....................................................................................55 Discussion...................................................................................................................55 4 CENTRAL LEPTIN GENE THER APY FAILS TO OVERCOME THE LEPTIN RESISTANCE ASSOCIATED WITH DIET-INDUCED OBESITY .......................72 Introduction.................................................................................................................72 Methods and Materials...............................................................................................73 Animals................................................................................................................73 Experimental Design...........................................................................................73 Blood Collection..................................................................................................74 Oxygen Consumption..........................................................................................75 Construction of rAAV Vector Plasmid...............................................................75 Packaging of rAAV Vectors................................................................................75 Vector Administration.........................................................................................76 Tissue Harvesting................................................................................................76 Serum Leptin and FFA........................................................................................77 Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR)..........................77 STAT3/Phospo-STAT3 Assay............................................................................78 Probes..................................................................................................................78 mRNA Levels (Dot Blot Analysis).....................................................................79 Statistical Analysis..............................................................................................79 Results........................................................................................................................ .80 Part I: High-Fat Feeding......................................................................................80 Food consumption and body mass...............................................................80 Oxygen consumption....................................................................................81 Serum leptin and free fatty acids..................................................................81 Part II: Post rAAV-Leptin Delivery....................................................................83 Food consumption and body mass...............................................................83 Oxygen consumption....................................................................................84 Adiposity......................................................................................................85 Serum leptin and free fatty acids..................................................................85 Leptin transgene expression in the hypothalamus........................................86 Leptin receptor expression and signal transduction in the hypothalamus....86

PAGE 10

x Downstream neuropeptide regul ation in the hypothalamus.........................86 UCP-1 in brown adipose tissue....................................................................87 Discussion...................................................................................................................88 5 CALORIC RESTRICTION REVERSES IMPAIRMENTS IN LEPTIN RECEPTOR EXPRESSION AND MAXIMAL SIGNALI NG CAPACITY IN DIET-INDUCED OBESE ANIMALS..................................................................................................101 Introduction...............................................................................................................101 Methods and Materials.............................................................................................102 Animals..............................................................................................................102 Experimental Design.........................................................................................103 Experimental design 1: leptin signaling in diet-induced obese..................103 Experimental design 2: leptin signa ling following caloric restriction.......104 Leptin Administration.......................................................................................104 Tissue Harvesting..............................................................................................105 Real-Time PCR.................................................................................................105 STAT3/Phospo-STAT3 Assay..........................................................................106 Leptin mRNA Levels in White Adipose Tissue................................................107 Serum Leptin.....................................................................................................107 Statistical Analysis............................................................................................107 Results.......................................................................................................................1 08 Food Intake and Body Weight...........................................................................108 Adiposity...........................................................................................................109 Leptin Receptor Expression in the Hypothalamus............................................109 Hypothalamic STAT3 Phosphorylation............................................................110 Leptin Expression in White Adipose Tissue.....................................................111 Serum Leptin.....................................................................................................111 Discussion.................................................................................................................116 6 GENERAL DISCUSSION AND CONCLUSIONS................................................123 Major Findings..........................................................................................................123 Physiological and Biochemical Effect s of Leptin Gene Therapy are Reversible......................................................................................................123 Leptin May Regulate Leptin Receptor Gene Expression..................................124 There is a Central Nervous System Component to Leptin Resistance..............125 Diet-Induced Obese Animals Have Im paired Leptin Signaling Capacity.........126 Impairments in Leptin Receptor Expr ession and Signaling in Diet-Induced Obese are Reversed by Caloric Restriction...................................................126 New Insights on Leptin Resistance...........................................................................127 Evidence for Role of Leptin Signal Transduction.............................................127 Evidence Against Major Role for Blood:Brain Barrier.....................................127 Evidence for Down-Stream Contributors to Leptin Resistance........................128 Conclusion................................................................................................................130

PAGE 11

xi LIST OF REFERENCES.................................................................................................132 BIOGRAPHICAL SKETCH...........................................................................................148

PAGE 12

xii LIST OF TABLES Table page 3-1 Brown adipose tissue parameters, serum free fatty acids, insulin, and glucose, and CSF leptin at sacrifice .......................................................................................65 4-1 Serum leptin and free fatty acids after 75 days of high-fat feeding (DIO and DR)...........................................................................................................83 4-2 Oxygen consumption and 24 hour caloric in take on day 7 following rAAV-leptin or control vector delivery .........................................................................................90 4-3 Serum leptin and FFA at sacrifice............................................................................92 4-4 Hypothalamic POMC NPY, AGRP, a nd SOCS3 expression at sacrifice..............94

PAGE 13

xiii LIST OF FIGURES Figure page 1-1 Model of leptin receptor signal transduction...........................................................7 3-1 pTR-ObW. pTR-ObW encodes rat leptin cDNA..............................................42 3-2 The TET-responsive promotor/leptin plas mid construct pTR-tetR-Ob (top), and the assessory plasmid constr uct pTR-rtTA/tTS (bottom)......................................43 3-3 Schematic representation of leptin transgene regulation by doxycycline in dual vector “TET-Ob” system, “OFF” state..................................................................44 3-4 Schematic representation of leptin transgene regulation by doxycycline in dual vector “TET-Ob” system, “ON” state....................................................................45 3-5 Body mass following intrace rebroventricular administ ration of rAAV-leptin......56 3-6 Food intake following in tracerebroventricu lar administration of rAAV-leptin....56 3-7 Disappearance of visceral adipose tissue in 4 month old F344xBN rats administered rAAV-leptin.....................................................................................57 3-8 Leptin is undetectable in serum of animals following central leptin gene delivery..................................................................................................................57 3-9 Experiment 2 design, TET-Ob gene delivery and regulation................................58 3-10 Body mass following TETOb or control vector delivery.....................................59 3-11 Daily food consumption following TETOb or control vector delivery................59 3-12 Serum leptin following TETOb or control vector delivery..................................60 3-13 Visceral adiposity following TETOb or control vector delivery..........................60 3-14 Hypothalamic leptin expression 66 days after TETOb or control vector delivery..................................................................................................................61 3-15 Hypothalamic P-STAT3 66 days following TETOb or control vector delivery. Values represent means SEM (Bar Graph, Top)................................................62

PAGE 14

xiv 3-16 POMC expression 66 days following TETOb or control vector delivery............63 3-17 SOCS3 expression 66 days following TETOb or control vector delivery...........63 3-18 Long-form leptin receptor (Ob-Rb) expression in the hypothalamus 66 days following TETOb or control vector delivery........................................................64 3-19 UCP-1 protein in BAT 66 days followi ng TET-Ob or control vector delivery.....64 4-1 Body mass during high fat feed ing (pre-vector delivery)......................................82 4-2 Caloric intake during high fat feeding (pre-vector delivery).................................82 4-3 Oxygen consumption on day 30 and on day 70 after commencing HF-feeding (pre-vector delivery)..............................................................................................83 4-4 Changes in body mass during 29 days post-vector delivery..................................89 4-5 Caloric intake during 29 days post-vector delivery...............................................89 4-6 Visceral adiposity 30 days post-vector delivery....................................................90 4-7 Leptin transgene expression in th e hypothalamus 30 days post-vector delivery. ...............................................................................................................91 4-8 Hypothalamic leptin receptor expr ession 30 days post-vector delivery................92 4-9 STAT3 phosphorylation in the hypothalamus at sacrifice.....................................93 4-10 UCP-1 concentration in BAT at sacrifice..............................................................94 4-11 UCP-1 per total interscapular brow n adipose tissue (IBAT) at sacrifice...............95 5-1 Body mass during high fat feeding. ...................................................................112 5-2 Body mass during calorie restriction. ................................................................112 5-3 Visceral adiposity at sacrifice..............................................................................113 5-4 Ob-Rb expression in the hypothalamus...............................................................113 5-5 Maximal leptin-induced STAT3 phosphoryl ation capacity is reduced in DIO and increased by caloric restriction. ....................................................................114 5-6 Leptin expression in retroperiton eal white adipose tissue (RTWAT) at sacrifice................................................................................................................115

PAGE 15

xv 5-7 Serum leptin at sacrifice.......................................................................................115 5-8 Serum leptin corrected for leakage of infused leptin from the CSF....................116

PAGE 16

xvi Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy POTENT ANOREXIC AND LIPOPENIC EFFECTS OF CENTRAL LEPTIN GENE THERAPY ARE BLOCKED BY DIET-INDUCED OBESITY: EVIDENCE FOR IMPAIRED LEPTIN RECEPTOR EXPR ESSION/SIGNAL TRANSDUCTION IN OBESITY AND REVERSAL BY CALORIC RESTRICTION By Jared Timothy Wilsey August 2003 Chair: Philip J. Scarpace Major Department: Pharmacology and Therapeutics Obesity is estimated to result in 300,000 deaths each year in the United States alone. The purpose of this dissertation was to study the molecular basis of obesity. Specifically, we studied how the brains respon se to the hormone leptin is impaired in obese animals. First, we demonstrated that young adult F 344XBN rats are highly responsive to a single intracerebroventricular injection of adeno-associated virus encoding leptin (rAAVleptin). Within 1 week of leptin gene de livery, animals exhibited a significant anorectic response without the concomitant decrease in resting energy expenditure that is normally recorded during anorexia. Th is anorectic response persiste d for more than two months. As a result of this negative shift in ener gy balance, rAAV-leptin treated animals lost significant weight with respect to controls. More impressively, there was a near complete disappearance of visceral white fat following rAAV-leptin treatment.

PAGE 17

xvii We then incorporated regulation into rAAV-leptin by placing leptin under the control of the tetracycline transactivator (rtT A) and operon (tetR). Using this system, we showed that all of the physio logical and biochemical responses to leptin gene therapy are completely reversible. We also demonstrat ed that central leptin may negatively regulate the expression of leptin recep tor in the hypothalamus, and that this is reversed upon transgene inactivation. This was the first ever experiment to use an externally regulated transgene to control energy balance. Next, we showed that diet-induced obes e (DIO) animals are non-responsive to the anorectic, lipopenic, and biochemical effect s of rAAV-leptin whereas high-fat fed animals that did not become obese (diet resistan t) retained leptin responsiveness. It was found that the leptin resist ance in the DIO animals was associated with reduced hypothalamic leptin receptor expression along with an absence of signal transduction response to rAAV-leptin. DIO animals were then tested with an acute supramaximal bolus of i.c.v. leptin, and were found to have reduced leptin receptor signal transduction capacity compared to non-obese controls. Finally, the deficits in leptin receptor expression and signaling capacity in DIO were completely reve rsed by 30 days of caloric restriction. Thus, short-term cal orie restriction may be a viab le strategy to restore leptin sensitivity in previously leptin-resis tant models of acquired obesity.

PAGE 18

1 CHAPTER 1 BACKGROUND AND SIGNIFICANCE One in three Americans is obese (Spiegelman and Flier, 2001). It is estimated that obesity results in 300,000 deaths per year in the Unites St ates (Kopelman, 2000), and the World Health Organization now recognizes obe sity as one of the top 10 global health problems (Kelner and Helmuth, 2003). Ob esity promotes diabetes, hypertension, cardiovascular disease, stroke, and cancer (Khaodhiar et al., 1999). Being overweight can decrease quality of life by triggering oste oarthritis, joint pain and degeneration, gout, and sleep apnea (Scapinelli, 1975; Khaodhiar et al., 1999). In excess of 5% of U.S. national health expenditures ar e directed at costs associated with obesity (Khaodhiar et al., 1999; Thompson et al. 1998). Moreover, th e incidence of obesity in children has escalated dramatically (Troiano and Flegal, 1999), foreshadowing greater harm to come. Leptin Leptin, the product of the Ob gene cloned in 1994 (Zha ng et al., 1994), is a 16 KD peptide hormone secreted prim arily from white adipose tissue (WAT). When leptin was found to rapidly reve rse obesity in the ob/ob mutant mouse, excitement soared in the scientific community and the popular press alike for what c ould be a cure for the ever growing obesity epidemic in Western societie s. Before research began to accumulate suggesting otherwise, it was even referred to as the “anti-obesity hormone” (Elmquist et al., 1998). Leptin acts on satiety centers in the hypothalamus to both decrease food intake and increase energy expenditure (Ahima and Flier, 2000b). These effects can combine to cause dramatic lipopenia, or loss of adipos e tissue, in many animal models. In some

PAGE 19

2 cases, a complete disappearance of visibl e adipose tissue can be achieved following pharmacological leptin infusion or Ob gene delivery (Chen et al., 1996; Halaas et al., 1997; Scarpace et al., 2002b). There are documented cases of leptin de ficiency leading to obesity in humans, analogous to the ob/ob mouse. One such case is described below. . a 200-pound 9-year-old English girl, whos e legs were so larg e she could barely walk, was found to lack the weight-regul ating hormone leptin. Treatment with leptin dramatically reduced her food inta ke, and that of her similarly affected cousin, to the point where they both now have body weights within the normal range for their age and live normal lives. Before leptin treatment, the younger child consumed in excess of 1100 calories at a si ngle meal, which is approximately half the average daily intake of an adult. W ith only a few leptin injections, this was reduced by 84% to 180 calories, the typica l intake of a normal child (Friedman, 2003, page 856). Unfortunately, such a miraculous transf ormation following leptin treatment is atypical. Common human obesity is a more complex diseas e than the monogenic leptin deficiency described above. The prevailing obe sity is believed to involve multiple genes that participate in the intricate, and often redundant, feeding and metabolic pathways— systems that are essential for survival. This complexity has greatly impeded the search for safe and effective treatments for obesity, a search that has been largely unsuccessful to date (Gura 2003). Most regrettably, hum an obesity is normally accompanied by high levels of endogenous leptin and resistance to the weight-reducing effe cts of the hormone. Leptin resistance will be discusse d in detail later in this chapter as well as Chapters 4-6 of this dissertation. Leptin has diverse biological functions in addition to its role in energy balance. These include roles in reproduction, hematopoiesis, angiogenesis, immune responsiveness, blood pressure regulation, and bone formation (Fruhbeck et al., 2001). Leptin stimulates the differentiation of macrophages and the production of cytokines by

PAGE 20

3 macrophages (Gainsford et al., 1996). Leptin accelerates the onset of puberty in wildtype rodents and appears to increase reproductive behavior (A hima and Flier, 2000b). In fact, leptin may be the signal that acts on th e gonadotrophin releasi ng hormone system in the hypothalamus at the time of puberty (Elmqui st et al., 1998). Leptin stimulates the growth of blood vessels and increases the ra te of wound healing (Sierra-Honigmann et al., 1998), and yet inhibits bone formation (Ducy et al., 2000). The effect of pharmacogical leptin on these systems likely de pends on the route of administration. For example, chronic central delivery of the hormone results in low peripheral leptin. Thus, direct administration of leptin into the central nervous system is expected to inhibit any direct actions of leptin in the periphery. The Leptin Receptor Isoforms The leptin receptor was first cloned in 1995 (T artaglia et al., 1995). It is a member of the class I cytokine receptor family that includes the IL-6, IL-11, IL-12, LIF, and CNTF receptors (Baumann et al., 1996). The ab sence of a functional leptin receptor (as in the db/db mouse) results in obesit y and diabetes-like symptoms. Alternative splicing of a single gene yields at least six isoforms of the leptin receptor, named Ob-Ra, Ob-Rb, Ob-Rc, Ob-Rd, Ob-Re, and Ob-Rf. Ob-Rb is the “long form” of the leptin receptor and is found primarily in the brain, although lim ited Ob-Rb expression has been detected elsewhere including the adrenal gland (Takekoshi et al., 1999), the intestine (Morton et al., 1998), and both brown and white adipose tis sue (Siegrist-Kaiser et al., 1997). Ob-Rb is the only form of the leptin receptor believed to be capable of full intracellular signaling capacity (Spiegelman and Flier, 2001). Ob-R e is a soluble form of the leptin receptor and may negatively regulate the bioavailability of circulating leptin by acting as a binding

PAGE 21

4 protein (Brabant et al., 2000; Fruhbeck et al., 2001). The remaining isoforms (Ob-Ra, Ob-Rc, Ob-Rd, and ObRf) have shortened intr acellular domains. Ob-Ra appears to be involved in transport across the blood-br ain barrier (Kastin and Pan, 2000). The intracellular domain of all leptin recep tor isoforms contains an identical 29 amino acid “Box 1” Janus-family kinase (J AK) binding domain in the juxtamembrane region. Only the long-form receptor Ob-Rb has a “Box 2” motif with s ignal t ransducer and a ctivator of t ranscription (STAT) binding sites. The intracellular domain of Ob-Rb contains approximately 306 amino acids whereas the intracellular domain of the shortened forms range from 32-40 am ino acids (Tartaglia et al., 1995). Leptin Binding to Leptin Receptor Both Ob-Rb and short forms of the leptin receptor exist as homodimers in the absence of ligand, and the formation of homodi mers does not seem to be affected by the presence of leptin (Nakashima et al., 1997; Sweeney, 2002; White and Tartaglia, 1999). Each leptin receptor in a homodimer unit bi nds leptin in a 1:1 stoichiometry, hence forming tetrameric complexes with 2 lept in molecules bound by the receptor dimer (Devos et al., 1997). A conformational change in the receptor takes place upon the formation of this tetramer that is a critical first step in leptin si gnaling (Fong et al., 1998). It appears that these leptin-Ob-Rb comple xes are internalized subsequent to ligand binding via clatherin-coated ve sicles (Sweeney, 2002; Lundin et al., 2000). Similar to other members of the cytokine receptor family, it is believed that internalized receptors are targeted for degradation or recycled back to the cell surface (S weeney, 2002). It has been recently estimated that 25% or less of le ptin receptors are locate d at the cell surface, with the majority sequestered in intracellular pools (Barr et al., 1999). While leptin exposure may promote receptor internalization, it is presently unclear what other factors

PAGE 22

5 may influence transport of receptors betw een intracellular compartments and the cell surface. Leptin Receptor Signal Transduction JAK/STAT Pathway Like all members of the class I cytokine receptor family, Ob-Rb operates through a JAK/STAT signal transduction pathway (Vaisse et al., 1996 ). Activated Ob-Rb can phosphorylate STAT-3, 5, and 6 (Baumann et al., 1996; Ghilardi et al., 1996), but most research to date has focused on the STAT-3 pathway (Spiegelman and Flier, 2001). The naked leptin receptor does not have intrinsi c tyrosine kinase activity, so STAT signaling is dependent upon Ob-Rb association with kinase s such as JAK2 (Figure 1-1) (Vaisse et al., 1996). Docking of JAK2 occurs follo wing ligand binding and formation of the leptin:receptor tetramer discu ssed above (Morton et al., 1999). JAK2 then activates the Ob-Rb by phosphorylation of Tyr 985 and Tyr 1138, th e latter of which is critical for the recruitment and activation of STAT3 (see below). Once Tyr 1138 on Ob-Rb is phophorylated, STAT3 is recruited to the recepto r via SH2 domains where it is tyrosine phophorylated by the receptor-associated JAK2 (Sweeney, 2002). Activated STAT3 (PSTAT3) then dissociates from the receptor a nd forms homoor heterodimers. These activated STAT dimers travel to the nucleus where they act as transcription factors (Sweeney, 2002). While other signaling pathways are empl oyed by the leptin receptor (discussed below), the JAK/STAT pathway is required fo r leptin’s effects on energy balance and bodyweight regulation. Target ed disruption of Tyr 1138, the critical residue for STAT3 signaling by the leptin receptor, results in both hyperphagia a nd morbid obesity. This is despite the fact that other leptin receptor-m ediated signaling pathways appear to remain

PAGE 23

6 intact (Bates et al., 2003). STAT3 indepe ndent signaling, on the other hand, seems to play an important role in reproduction, linear growth, and glucose homeostasis (Bates et al., 2003). A simplified diagram of Ob-Rb signal transduction, including JAK-STAT signaling, is presente d in Figure 1-1. ERK-c-fos Pathway Leptin has been shown to activate a signa ling cascade via extr acellular regulated kinase (ERK) (Figure 1-1) (Kim et al., 2000b) ERKs are serine/threonine kinases and belong to the MAPK family. Activation of the ERK pathway is apparently mediated by the phosphorylation of Tyr 985 on the leptin receptor by JAK2 (see above), although there may also be an ERK pathway independe nt of Tyr 985 (Bjorbaek et al., 2001). The physiological significance of the ERC-c-fos pa thway with respect to leptin signaling is still under investigation, but it appears to regulate the expr ession of a completely different set of genes than the STAT3 path way (Figure 1-1) (Sweeney, 2002). PIA3 Kinase and Insulin-Like Signaling Furthermore, leptin can act through so me components of the insulin signal transduction pathway (Figure 1-1) (Kim et al., 2000b). Via JAK2, leptin can activate insulin-receptor substrates IRS-1 and IRS-2 which, in tu rn, can activate PI3-kinase (PI3K) (Figure 1-1) (Szanto and Kahn, 2000). Activation of PI3K occurs when SH2 domains of the regulatory subunit bind to phos phorylated tyrosine residues of IRS molecules. PI3K then catalyzes the phosphorylation of phospha tidylinositol (4,5) bisphosphate (PIP2) to phosphatidylinositol ( 3,4,5) trisphosphate (PIP3). Subsequently, PIP3 activates a variety of downstream si gnaling molecules that contain pleckstrin

PAGE 24

7 Figure 1-1. Model of leptin receptor signal transduction. homology domains, which recognize PIP3 (Niswender and Schwartz, 2003). Some examples of molecules activated by PIP3 are se rine-threonine kinases, tyrosine kinases, and GTPases (Shepherd et al., 1998). Perhap s the most studied downstream target of IRS-induced PIP3 is protein kinase B (P KB), also known as Akt (Niswender and Schwartz, 2003). Based on knowledge of insu lin signal transduction, it is believed that PIP3 activates PKB via a phosphatidyl-inositol dependent kinase (PDK) (Shepherd et al., 1998). PKB then activates phosphodiesterase 3B (PDE3B) which, in turn, decreases intracellular cAMP. It has been demonstrated that this decrease in cAMP levels may enhance leptin-dependent STAT3 phosphorylatio n (Niswender et al., 2001; Zhao et al.,

PAGE 25

8 2002). This PDE3B-mediated decrease in cAMP ma y also be a critical step in the leptindependent suppression of NPY (Akabayashi et al., 1994). Activation of this IRS-PI3K signaling cas cade is required for several physiological effects normally associated with insuli n, including glucose transporter (GLUT) recruitment to the cell membrane and glyc ogenesis (Kanai et al., 1993). Because of overlap between insulin and leptin signaling, leptin can modify certain insulin-induced changes in gene expression in vivo (Kim et al., 2000b; Szanto and Kahn, 2000). Negative Regulators of Leptin Signaling SOCS3. Leptin signaling activates negative-feedback messages within the cell. The most studied inhibitory signal is suppressor of cytokine signaling-3 (SOCS-3), which is so reliably activated by leptin that it serves as a marker for cells responding to the hormone (Bjorbaek et al., 1998; Bjorbaek et al., 1999). It is believed that the transcription factor P-STAT3, activated by leptin, directly upregulates SOCS-3 expression (Niswender and Schwartz, 2003). SOCS -3 inhibits leptin signaling in part by suppressing the activity of JAK kinase (Bjorbae k et al., 1999). SOCS-3 is also a negative regulator of insulin signali ng (Fruhbeck et al., 2001). Thus, SOCS-3 induction in a hyperleptinemic state could be one mechanism linking obesity and insulin resistance (see discussion of leptin resistance below). PIAS. Another negative regulator of leptin signaling is protein inhibitor of activated STATS (PIAS) (Liu et al., 1998). This fa mily of peptides was discovered by a yeast two-hybrid screen for molecules that inte ract with STAT. PIAS proteins can inhibit STAT-induced transcriptional activation, pr esumably by binding to activated STAT dimers and blocking their passage into th e nucleus (Greenhalgh and Hilton, 2001). However, unlike SOCS-3, expression of PIAS family members does not appear to be up-

PAGE 26

9 regulated by cytokine signa ling, including leptin signaling (unpublished observations by Philip Scarpace’s Laboratory). Rather, PIAS family members (specifically PIAS-1 and PIAS-3) are constitutively expressed in ma ny cell types (Greenhalgh and Hilton, 2001). Although not proven at this time, it is possible the PIAS activity increases in response to leptin signaling, thus allowing it to act as a negative feedback effector. Down-Stream Leptin Signaling and Neuropep tide Regulation in the Hypothalamus Leptin signaling in the hypothalamus is the first step in a complex pathway involving multiple neuropeptides and vari ous intra-hypothalamic as well as extrahypothalamic centers involved in the regulati on of energy balance. This section will discuss both the neuroanatomy of this pa thway within the hypothalamus and the key neuropeptides and neurotransmitters that participate. Arcuate Nucleus The arcuate nucleus (ARC) is a critical area for leptin si gnaling with respect to the regulation of energy balance. The ARC sits ju st above the median eminence, at the base of the third ventricle (Williams et al., 2001). Leptin acts on two di stinct populations of neurons in the arcuate nucleus via the Ob -Rb receptor (as described above). One population of neurons co-expresses the orexig enic (appetite-stimulating) peptides NPY and AGRP, and leptin signaling reduces th eir expression (York and Bouchard, 2000). The second population co-expresse s the anorectic peptides -MSH (derived from POMC processing) and CART, and lept in up-regulates their expres sion (Elias et al., 1999; York, 1999). -MSH and AGRP -MSH and AGRP are the endogenous agonist and antagonist, respectively, for a common recepto r, the melanocortin 4 receptor (MC4R),

PAGE 27

10 which is expressed primarily in the br ain (Cone, 1999). Activation of MC4R by -MSH reduces food intake while suppression of MC4R activity by AGRP or a pharmacological antagonist increases food intake and diminishes the response to leptin (Fan et al., 1997). Signaling through the MC4R can also increase en ergy expenditure (Pierroz et al., 2002). Another melanocortin receptor, MC3R, is also implicated in ener gy homeostasis, but the MC4R appears to be the most cr itical for the hypophagic action of -MSH and its analogues whereas both receptors may be involved in regulating energy expenditure (Williams et al., 2001). The mechanism of enhanced metabolic rate following the delivery of an MC3/4 agonist appears to involve both enhanc ed uncoupling protein expression in the brown adipose tissue and in increased capacity of skeletal muscle to oxidize fatty acids (Shek et al., 2002). NPY NPY is a particularly potent stimula tor of feeding in the hypothalamus. Acute NPY injections into the brain can increase food intake several fold and, surprisingly, chronic NPY infusion continues to stimulate feeding (Stanley et al., 1986). Thus, NPY can apparently override short-term and long-term satiety signals. NPY can also decrease energy expenditure. One mechanism behind this is an NPY-induced reduction in thermogenesis in brown adi pose tissue and, presumably, a reduction in uncoupling protein expression and activity (Zar jevski et al., 1993). The activity of NPY/AGRP neurons in the ARC is increased during fasting and in states of negative energy balance in an apparent homeostatic a ttempt to limit or reverse the loss of body fat stores (Williams et al., 2001). CART. Cocaineand amphetamine-regulated tran script (CART) is an anorectic neuropeptide expressed by firstorder leptin responsive neurons in the ARC. Alternative

PAGE 28

11 splicing of the CART mRNA re sults in two peptide products consisting of 116 and 129 amino acids in the rat (Douglass et al., 1995) These pro-peptides are then tissuespecifically processed into smaller pep tides and packaged into vesicles as neurotransmitters (Hill ebrand et al., 2002). Several CART peptide fragments have been identified in the hypothalamus and other pa rts of the brain, and many of these have proven to be biologically activ e (Kuhar and Dall Vechia, 1999). CART was initially identified in the striatum, where its ex pression is potently induced by various psychoactive drugs (Douglass et al., 1995). Ho wever, CART is highly expressed in parts of the hypothalamus, including the ARC, paraventricular nucleus, dorsomedial hypothalamic nucleus, and lateral hypothalamus. As mentioned above, in the ARC it is co-expressed with POMC in the anorex ic class of first order neurons. Central CART delivery causes anorexia and can block the orexigenic effect of NPY (Kristensen et al., 1998; Lambert et al., 1998) Moreover, i.c.v. CART can increase uncoupling protein (discussed later in Chapter 1) expression in brown and white adipose tissue and muscle, suggesting that CART ma y promote increased energy expenditure. CART knockouts have been shown to have in creased susceptibil ity to diet-induced obesity (Asnicar et al., 2001) Recent data also sugge st that CART infusion may promote lipid oxidation, particul arly in high-fat fed animal s (Rohner-Jeanrenaud et al., 2002). However, animals appear to begin de sensitizing to the eff ects of chronic CART infusion, including the anorexic effect, with in 6 days (Larsen et al., 2000). The mechanism behind this rapid desensitiz ation is unknown, but it has undoubtedly contributed to the lack of interest in the CART pathway as a target for the pharmacological treatment of obesity.

PAGE 29

12 Projections from the Arcuate The ARC has extensive afferent and e fferent connections to “second order neurons” in other parts of the hypothalamus, including the lateral hypothalamus (LH), the paraventricular nucleus (PVN), the dorsome dial hypothalamic nucleus (DMH), and the ventromedial hypothalamus (VMH) (Williams et al., 2001). Most of these centers have functional Ob-Rb and, thus, may respond to leptin directly or indirectly via projections from the arcuate. Lateral hypothalamus. The lateral hypothalamus was cl assically described as the “feeding center” since stimulaton of the LH increases food intake and its destruction leads to profound anorexia, sometimes resulti ng in starvation. Neurons in the LH express peptides that potently stimulate f ood intake, including or exins/hypocretins and melanin-concentrating hormone (MCH). The LH is also richly enervated with NPY terminals (originating in the ARC) and is densely populated with NPY-Y5 receptors (Williams et al., 2001). It is believed that the anorectic signal -MSH acts, in part, via projections from the arcuate to the LH wh ere the expression of MCH and orexins are suppressed (Elias et al., 1999). Ventromedial hypothalamus. In opposition to the LH, the VMH was classically viewed as the “satiety center .” It has been known for half a century that stimulation of the VMH leads to anorexia and weight loss whereas destruction of the VMH causes overfeeeding and obesity (Stellar, 1954). The VM H has connections with other key centers affecting energy balance in the hypothalamus including the PVN, LH, and DMH. In addition, the VMH may be a direct target of leptin as it has been shown to be richly populated with Ob-Rb (Williams et al., 2001).

PAGE 30

13 Dorsomedial hypothalamus. The DMH is located immediately above the VMH. It has reciprocal connections with th e PVN and LH, and also projects to extrahypothalamic areas includ ing the brainstem. ARC-NPY/AGRP neurons terminate in the DMH, and the DMH may act in concert wi th the PVN to help initiate and maintain food intake (Christophe, 1998). The DMH is ri ch in both Ob-Rb and insulin receptors. Paraventricular nucleus. The PVN is located in the anterior hypothalamus. If functions as an “integrati on center” (Williams et al., 2001). The PVN receives projections from the ARC and the LH. The PVN has abundant nerve-terminals secreting neurotransmitters that potent ly modify appetite, including -MSH and NPY from the ARC, serotonin, galanin, norepinephrine, and en dorphins. Corticotroph in releasing factor (CRF) is synthesized in the PVN and then re leased via axonal projections in the median eminence, where it may inhibit NPY/AGRP ne urons (Williams et al., 2001; Arase et al., 1998). In addition to contributing to an a norectic response, arcuate melanocortin nerve terminals in the PVN may regulate pituitar y hormone release and autonomic nervous system activity (Spiegelman and Flier, 2001). Extra-Hypothalamic Signaling and Energy Balance Signals originating in the hypothalamus projec t to other parts of the nervous system to affect both food intake and energy expend iture. For example, CART neurons in the arcuate also project to autonom ic centers in the spinal cor d, thus providing a pathway for the CART-induced increase in sympathetic acti vity and energy expenditure (Elias et al., 1999). Melanocortin signaling originating in the hypothalamus may also enhance sympathetic activity and, ergo, energy expe nditure (Spiegelman and Flier, 2001).

PAGE 31

14 Moreover, signals from other parts of th e body may travel afferently to various parts of the brain involved in energy homeostasis. For example, messages from the digestive tract are conveyed via the vagus nerve to the nucleus of the tractus solitarius (NTS) in the medulla (Travers et al., 1987). These messages include information about taste, gastric distension, and portal vein gl ucose levels (Travers et al., 1987). Cholecystokinin (CCK), an intestinal satiet y signal, also signals to the NTS via CCK receptors on the vagus. CCK is believed to be involved in meal termination, and its release is potently stimulated by fatty meals. The NTS po ssesses receptors for leptin and -MSH, and it expresses POMC. Administrati on of an MC4 receptor agonist (i.e., an MSH analogue) adjacent to the NTS results in potent anorexia, suggesting that this may be an important extra-hypothalamic feedi ng/satiety center (Gri ll et al., 1998). Certain signals originating in other pa rts of the brain may project to the feeding/satiety centers in the hypothalamus. For example, the NTS has projections to the LH. Neurons originating in the raphe nuclei of the caudal brain stem project adjacent to the ARC and PVN, where they release sero tonin (Williams et al. 2001). Serotonergic signaling causes feelings of satiety and may also increase energy expenditure (Bray, 2000). As such, it has been the target of numerous weight loss drugs, including the infamous dexfenfluramine (a main ingredient of “Fen-Phen” or Redux ) (Vickers et al., 1999) and Meridia (Gura 2003). Other Molecules Regulating Energy Balance Insulin As discussed previously, leptin and in sulin share a common signal transduction pathway through IRSs and PI3Kinase. Their physiological effects, especially in the central nervous system, are also similar. Li ke leptin, intracerebr oventricular infusion of

PAGE 32

15 insulin causes potent anorexia and weight loss (Woods et al., 1979). Insulin receptors are found in areas of the brain known to be key hunger/satiety centers, including the arcuate nucleus of the hypothalamus (Niswender and Schwartz, 2003). Similar to what occurs in the leptin-deficient ob/ob mouse, absolute insulin deficien cy is associated with sustained hyperphagia known as “diabetic hyperphagia” (S ipols et al., 1995). Twenty-four hour integrated and fasting serum insulin levels corre late with total adiposity (Bagdade et al., 1967), much like what is observed with leptin. Finally it is be lieved that both leptin and insulin can activate POMC neurons and suppr ess the activity of NP Y/AGRP neurons in the arcuate nucleus (Niswende r and Schwartz, 2003). Despite the signaling overlap, there ar e some key differences between the physiological actions of leptin and insulin. On e example is that insulin facilitates fatty acid and triglyceride deposition in adipose ti ssue whereas leptin promotes lipolysis via both central signaling and direct effects on white fat (Wang et al., 1999). Thus, from a whole-organism perspective, insulin is an anabolic hormone while leptin tends to be catabolic in nature. However, it is important to point out that th e anabolic actions of peripheral insulin may be different from the specific effects of centr al insulin signaling, which, as discussed above, cause anorexia and weight loss. Obesity is a powerful risk factor for T ype II diabetes, which is characterized by insulin resistance. As we will discuss below, common obesity is typified by hyperleptinemia and leptin resistance. The relationship between insu lin and leptin, both peripherally and in the centra l nervous system, is likely complex and not fully understood at this time. However, given the common si gnaling pathway and the apparent synergy of

PAGE 33

16 leptin and insulin action in the CNS, it is pl ausible that leptin re sistance may be one factor linking obesity an d insulin resistance. Glucose In 1955, Mayer first published his theory of the “glucostat” hypothesis on the regulation of food intake and body weight (M ayer, 1955). This hypothesis proposed that special glucose sensing neurons helped to stim ulate feeding at the e nd of a fast. Taken alone, this theory does not appear to explain feeding behavior very well. For example, Type II diabetics, who are ex tremely hyperglycemic, are also hyperphagic. Still, glucose does appear to be one participant in the co mplex regulation of feeding. Hypoglycemia or the blockade of neuronal glucose metabo lism stimulates acute feeding behavior (Williams et al., 2001). Several hypothalamic ar eas critical to the regulation of energy homeostasis have been shown to have gl ucose-sensing neurons including the ARC, DMH, PVN, VMH, and LH (Williams et al., 2001). Extrahypothalamic areas also have glucose sensing neurons, including the NTS, s ubstantia nigra, locus coeruleus, neocortex, and hippocampus (Williams et al., 2001). Sele ctive destruction of glucose-sensing neurons in the VMH of mice leads to obesity, demonstrating that th ese neurons are, in fact, involved in energy homeostasis (Ber gen et al., 1996). The mechanism by which these neurons respond to changes in glucos e concentration appears to involve ATPsensitive K+ channels (Ashford et al., 1990). Orexins/Hypocretins Orexin A, orexin B, and hypocretins 1 and 2 are all derived from a common precursor peptide known as “prepro-orexin” or “prepro-hypocretin” (Williams et al., 2001). These four peptides, collectively refe rred to as orexins/hypocretins, are closely related structurally and functionally. As me ntioned previously, orexins/hypocretins are

PAGE 34

17 expressed by neurons in the lateral hypothala mus. They are also expressed in the perifornical nucleus and dorsal area of the hypothalamus (Williams et al., 2001). Orexin/hypocretin neurons project to the P VN, ARC, and the NTS as well as the dorsal motor nucleus of the vagus nerve (de Lecea et al., 1998; Peyron et al., 1998). As their name would suggest, these pe ptides are stimulators of feeding (orexigenic). Prepro-orexin expression is stimulated by fasting and hypoglycemia and suppressed by food in the gut (gastric distension) (Williams et al., 2001). Orexins/ hypocretins appear to be shortterm regulators of feeding, acting on a meal-to-meal basis. Consistent with this, acute i. c.v. administration of orexin-A stimulates acute feeding, but does not alter 24 hour energy intake (de Lecea et al., 1998; Haynes et al., 1999). Moreover, chronic infusion of orexin-A does not cause weight gain (Yamanaka et al., 1999). Ghrelin Ghrelin is an orexigenic peptide first isolated from the stomach, although it is expressed in lesser amounts by the pancr eas, kidney, ARC, and pituitary gland (Hillebrand et al., 2002). Ghrelin release is stimulated in states of negative energy balance and suppressed in conditions positive energy balance or obesity (Otto et al., 2001; Toshinai et al., 2001). Ghreli n can enhance hypothalamic NPY and AgRP expression in the hypothalamus (Kamegai et al., 2001), and chronic central delivery of ghrelin results in increased food intake, bodyweight, and adiposity (Hillebrand et al., 2002; Tschop et al. 2000). Moreover, it a ppears that ghrelin may inhibit CART/POMC neurons in the ARC (Riediger et al., 2003). As such, ghrelin opposes many of the effects of hypothalamic leptin signaling. Central ghrelin can increase ACTH release and decrease thyroid stimulating hormone (TSH) le vels, perhaps contribu ting to the reduction

PAGE 35

18 in energy expenditure following ghrelin administration (Hilleb rand et al., 2002). Ghrelin increases growth hormone levels (Kojima et al., 1999), but the physio logical significance of this is still under investigation. Uncoupling Proteins Aerobic organisms synthesize most of th eir ATP through the process of oxidative phosphorylation. This process starts w ith NADH and FADH2, high energy molecules formed during glycolysis, fat oxidation, and th e citric acid cycle. NADH and FADH2 pass their electrons to oxygen via a series of electron carr iers in the mitochondria, and this process leads to the pu mping of protons out of the mitochondrial matrix. This ultimately creates a proton gradient. Prot ons flow through the protein channel “ATPsynthase,” which uses the energy releas ed when the protons flow down their concentration gradient to phosphorylate ADP to ATP (Stryer, 1995). However, mammalian metabolism is not perfectly efficient. In fact, it has been estimated that proton leak across the inne r mitochondrial membrane accounts for ~25% of basal metabolic rate in mammals (Porter, 2001). Specialized proteins called “uncoupling prot eins” contribute to the inefficiency of oxidative phosphorylation. The first unc oupling protein discove red was UCP-1, which carries electrons from the cytosol into th e mitochondrial matrix. UCP-1 is found in brown adipose tissue (BAT), where it allows the energy liberated from fat oxidation to be dissipated as heat. As such, BAT UCP-1 is important for thermoregulation in mammals and is rapidly activated during exposure to co ld environments (Scarpace et al., 1994). These proteins may also play a key role in determining resting metabolic rate. Indeed, mutations in uncoupling protei ns have been linked to reduced basal metabolism in humans (Porter, 2001).

PAGE 36

19 UCP-1 is one peripheral mediator of th e increase in energy expenditure caused by leptin (Scarpace et al., 1997). In a recent study by Scarpace et al. (2002b), leptin gene therapy induced a dramatic 34-fold increase in UCP-1 protein levels in BAT. According to one study, UCP-1 is an absolute requireme nt for leptin’s lipolytic effects in mice (Commins et al., 2001), but it is unc lear if this is universally tr ue in other mammals. It is clear that central leptin signa ling powerfully induces UCP-1 expression and protein levels in the BAT pads of rats and this is associat ed with thermogenesis and lipolysis (Scarpace et al., 1997). Two other UCP’s have been discovered, since named UCP-2 and UCP-3. UCP-2 is widely distributed, expressed in spleen, l ung, intestine, white adipose tissue (WAT), BAT, uterus, kidneys, testes, brain, and heart w ith low levels detected in muscle and liver (Porter, 2001). UCP-3 is found in BAT and skeletal muscle (Boss et al., 1997). The physiological roles of UCP-2 and -3 are somewh at controversial, but there is evidence that they can uncouple mitoc hondrial respiration and pot entially increase energy expenditure. Both UCP-2 and -3 have been shown to decrease resting membrane potential in transfected yeast, and yeast mitochondria transfected with UCP-3 show increased oxygen consumption and heat product ion (Hagen et al., 2000 ; Porter, 2001). Moreover, mice overexpressing UCP-3 are hyper phagic and leaner than wild-type mice (Clapham et al., 2000). Such a phenotype is co nsistent with increased energy expenditure due to mitochondrial uncoupling. Leptin has been shown to upregulate UCP-2 in WAT (Scarpace et al., 1998; Commins et al., 2001) an d UCP-3 in BAT (Scarpace et al., 1998), but the effect is less potent than what is observed with UCP-1. Nonetheless, UCP-2 and 3 may play an important role in energy homeostasis.

PAGE 37

20 Leptin Resistance Human obesity is associated with hyp erleptinemia and leptin resistance (Heymsfield et al., 1999). As such, research that can describe th e molecular basis of leptin resistance and how to overcome it co uld have a profound impact on the clinical treatment of obesity. Several animal models have been described that mimic the common human pattern of hyperleptinemia in the obese state. Like obese humans in recent clinical trials (Heymsfield et al., 1999), these animals are re sistant to exogenous leptin. Aged-obese rats are partially resistant to both injected leptin peptide (Scarpace et al., 2000a) and constitutively active rAAV-leptin gene therapy (Scarpace et al., 2002b). This includes impaired signal transduction res ponse (Scarpace et al ., 2000b) and impaired anorectic, thermogenic, and lipolytic respons es to leptin (Shek and Scarpace, 2000). Data from this laboratory (Scarpace et al., 2002b) demons trate both a reduced maximal response and a complete attenuation of res ponse to rAAV-leptin in aged-obese but not young male rats observed for 46 days post-transf ection. It is not known if the impaired responses to leptin in thes e aged-obese animals are due to age itself, obesity, or a combination of the two. However, there is evidence that diet-induced obesity can reduce leptin responsiveness in young animals. Friedm an et al. demonstrated that diet-induced obese mice are less responsive to the lipolytic and anorectic effect s of chronic leptin infusion as compared to lean age-matched c ontrols (Halaas et al., 1997). In a similar study, 16 days on a high fat diet resulted in resistance to the an orectic effects of peripheral leptin and this resi stance increased in severity by day 56 of continued high fat feeding (Van Heek et al., 1997). It was al so demonstrated that diet-induced obesity prevents the normal increase in lumbar sy mpathetic nerve activity in response to pharmacologically administered leptin in rats (Lu et al., 1998). Increased sympathetic

PAGE 38

21 drive is a key component to leptin’s ther mogenic and lipolytic effects (Astrup, 2000; Scarpace and Matheny, 1998). These results sugg est that obesity may play a significant role in leptin desensitization in our aged -obese model, but it do es not eliminate the possibility that age also plays a role. As of the commenceme nt of the research described in this dissertation, it was not known if th e desensitization and rapid attenuation of response to rAAV leptin gene therapy obser ved in our aged-obese model could be duplicated in a young animal via diet-indu ced obesity; nor was it known if leptin responsiveness could be restored in an agedobese model by a period of caloric restriction that reduces adiposity. Although there is evidence that obesity plays some role in leptin resistance independent of aging, the mechanism of th is obesity-induced leptin resistance is unknown. One possibility is that the copious visceral and/or subcutaneous fat may be secreting one or more substances that inhibit leptin’s thermogenic and anorectic effects. Another possibility is that when adipose depots expand, th ey stop producing a substance that is required for normal leptin action. Indeed, discoveries in the past decade have changed our view of the adipocyte from a si mple storage depot to a complex endocrine organ that plays a critical role in metabolic regulation. Hormones secreted from white adipose tissue include leptin, TNF, TGF-, IL-6, adipophilin, adipsin, ASP, MIF, IGF1, and the recently identified adiponectin (Fruhbe ck et al., 2001). A deficiency in this latter peptide could be one factor linking obesi ty and both leptin and insulin resistance. Adeno-Associated Virus and Gene Therapy Wild-Type AAV Adeno-associated viruses (AAV) are members of the Parvoviridae family (Meneses, 1999). AAV virions range fr om 20-30 nm with icosahedral symmetry

PAGE 39

22 (Meneses, 1999). The AAV genome is a single stranded DNA (ssDNA) molecule containing ~4680 nucleotides (Meneses, 1999) The genome contains only two open reading frames (ORFs), which are flanked by inverted terminal repeats (ITRs) (Linden and Berns, 2000). These ITRs can serve as origins of replication (Meneses, 1999). The 5’ OPR encodes the “Rep” proteins which, as their name implies, are involved in the replication of AAV and are also believed to play a critical role in the site-specific integration of AAV into human chromo some 19 (specifically locus 19q13.3-qter) (Linden and Berns, 2000). The 3’ OPF contains the capsid ( cap ) gene which encodes the structural proteins of the virion. This cap gene expresses the th ree cap proteins (Vp1, Vp2, and Vp2) via alternativ e splicing (Meneses, 1999). Approximately 80% of adult humans are seropositive for antibodies against AAV, with seroconversion typically occurring by age 8 (Berns and Bohenzky, 1987). Despite this apparent wide spread AAV infection in humans, AAV appears to be benign as it has never been associated with any disease (Ber ns and Bohenzky, 1987). In the absence of co-infection or super-infection with other viruses such as herpes or adenovirus (Ad), AAV enters a latent pathway, lying dormant in chromosome 19. There are 5 known serotypes in humans: AAV-1 and AAV-2 are t hought to be of simian origin, AAV-3 and AAV-4 were first isolated from throat sw abs of humans suffering from adenovirus infection, and AAV-5 was initially isolated from a genital co ndyloma (Blacklow et al., 1967; Blacklow et al., 1968; Georg-Fries et al., 1984; Meneses, 1999). The adenoassociated virus used for gene delivery in th is dissertation is a recombinant form of AAV serotype-2.

PAGE 40

23 Recombinant Adeno-Associated Virus and Gene Delivery The recombinant adeno-associated virus (rAAV) typically used for gene delivery has been modified by removal of viral coding sequences (both the rep and cap encoding OTRs), thereby preventing repl ication (Meneses, 1999; Hermona t et al., 1984). Since the rep proteins are missing, rAAV is not capable of site-specific integra tion (Kearns et al., 1996). Nonetheless, long term expression of transgene has been observed and there is evidence of viral integration, albeit at random locations (Kearns et al., 1996). Most AAV vectors, including those used in this dissertation, consists of the two ITRs flanking the transgene of interest under the control of a non-AAV promotor such as chicken -actin (CBA) (Hauswirth et al., 2000; Zolotukhin et al., 2002). rAAV is a useful vector for stable gene delivery because it is capable of transfecting non-divid ing cells, including neuronal tissue. This makes it an ideal tool for overexpressing peptides in the central nervous system. rAAV has been successfully employed as a vector for gene delivery, or “gene therapy,” at the University of Florida and elsewhere (Muzyczka, 1992). For example, erythropoietin has been produced by the skeletal muscle of mice after transfection with a rAAV v ector encoding the human epo gene (Fisher et al., 1997). rAAV has been used to over-express tyro sine hydroxylase and aromatic amino acid decarboxylase in the CNS of animal mode ls of Parkinson’s Disease, including nonhuman primates (Kaplitt et al ., 1994). Our laboratory has successfully used rAAV to over-express leptin in both the hypothalamus a nd cells lining the cereb roventricles. We have also used rAAV to overexpress proopiomelanocortin (POMC) in the central nervous system. rAAV transfection of nondividing neuronal tissue allows for very stable transgene expression. In a recent st udies by our group, no a ttenuation of leptin

PAGE 41

24 transgene expression has been noted up to 300 days after intracereb roventricular gene delivery (Scarpace et al., 2002a; Scarpace et al., 2002b; Scarpace et al., 2003). Experimental Design and Rational One purpose of the studies described in this doctoral dissertati on was to test the efficacy of leptin gene delivery as a stra tegy to reverse acquired obesity. Acquired obesity refers to obesity that is not du e to a specific mutant gene (as in the ob/ob or db/db mouse). Rather, acquired obesity forms over tim e as a result of environmental conditions interacting with many genes. This form of obesity more closely mimics common, polygenic human obesity. Specifically, we used diet-induced obese animal models, where obesity gradually occurs due to excessive consumption of a calorically dense diet. A second purpose was to use central leptin gene and peptide delivery as tools to study leptin resistance. Diet -induced obese animals were used as a model of leptin resistance, and various physiological and bioc hemical responses to leptin were compared to those in lean cohorts. The general e xperimental design and objectives of each major experiment included in this diss ertation are described below. Incorporating Regulation into Leptin Gene Therapy Constitutively active leptin gene therapy ha s been shown to have potent anorexic, thermogenic, and lipopenic affects in both lep tin deficient and genetically normal animal models (Chen et al., 1996; Murphy et al., 1997; Scarpace et al., 2002b). One disadvantage of all leptin gene delivery systems studied to da te is lack of posttransfectional control. This is undesirabl e clinically as it would be difficult or impossible to reverse the progression of dele terious side effects as they appeared. Perhaps more importantly in the immediate context of this research, incorporating regulation into the rAAV-encodi ng-leptin could create an excellent research tool. For

PAGE 42

25 example, an on/off leptin gene therapy syst em may be useful for studying leptin’s longterm regulation of hypothalamic signal tran sduction, neuropeptide expression, and leptin receptor expression. Moreover, it would allo w us to determine whether the various biochemical and physiological changes attribut ed to leptin transgene overexpression are completely reversible. Thus, the objective of th is experiment was to create a leptin gene therapy system that can be externally regulat ed and to then test and study this system in vivo Central Leptin Over-Expression in Diet-Induced Obese Animals Unlike the ob/ob mouse, diet-induced obese (DIO) animals have an obesity characterized by hyperleptinemia and leptin resi stance. DIO occurs when animals are fed a high-fat, high-energy diet for several weeks to months. Most anim al models, including those with no known metabolic abnormalities, ar e susceptible to varying degrees of DIO. It has been hypothesized that deficient transport of leptin across the blood:brain barrier (BBB) plays a key role in leptin resistance a nd, as such, central over-expression of leptin may overcome this leptin resistance (Dhillon et al., 2001). In a very recent report, central leptin gene therapy was shown to pr event diet-induced obesity when the gene was delivered prior to commencing high-fat feedi ng (Dube et al., 2002). However, since the leptin transgene was delivered prior to high-fa t feeding, the ability of central leptin gene delivery to overcome leptin resistance and obes ity could not be evaluate d. Thus, the first objective of this set of experiments was to de termine if intracerebroventricular delivery of rAAV-encoding-leptin can reverse obesity in ge netically normal DIO rats. A second, yet critical, objective of this inve stigation was to use the resu lts of this study to further understand leptin resistance in obese animals.

PAGE 43

26 Effects of Diet-Induced Obesity and Ca loric Restriction on Leptin Receptor Expression and Signal Transduction Capacity This set of experiments spawned from th e results of the rAAV-leptin delivery in DIO study (previous paragraph). We hypot hesized that reduced leptin receptor expression in DIO animals was the result of chronic hyperleptin emia leading to chronically elevated leptin in the central nervous system, i.e., negative regulation of leptin receptor expression by leptin. We further hypothesized that this downregulation in leptin receptor expression would reduce maximal leptin-induced STAT3 phosphorylation capacity. Finally, we hypothesized that de ficits in leptin receptor expression and signaling capacity could be revers ed if we lowered leptin levels via caloric restriction. Chapter Summary and Conclusions In this chapter, we have discussed the many pathways involving a multitude of hormones, neuropeptides, and neurotransmitters that contribute to the regulation of energy balance. Given the escalating rates of obesity in modern societies, it appears that these homeostatic pathways did not evolve to effectively defend against the overabundance of highly palatable, over-processed fo ods and sedentary lifesty les of our era. While the regulation of energy balance is complex, leptin appears to be a key player in the process. Indeed, many of the signals participating in energy homeostasis including melanocortins, NPY, AGRP, CA RT, and uncoupling proteins occur downstream of leptin signaling. Put another way, leptin regulat es the expression and activity of most of the critical molecules involved in the regulation of food intake and energy expenditure. Moreover, it is clear that th e most common forms of human obesity are typified by leptin resistance. Thus, ther apeutic strategies that can overcome or circumvent this leptin resistance could have an enormous clinical impact in treating not

PAGE 44

27 only obesity, but its many associated disord ers including diabetes and cardiovascular disease. The major objective of this doctoral dissertation was to explore the molecular mechanisms of leptin resistance in obese an imal models, with a particular emphasis on leptin signal transduction in th e hypothalamus. It is hoped that the results of this dissertation will make a meaningful contributio n to our understanding of leptin resistance in the obese state. Ultimately, a better understa nding of leptin resistance will likely lead to more effective treatments for obesity and other metabolic disorders.

PAGE 45

28 CHAPTER 2 GENERAL METHODS AND MATERIALS Experimental Animals Young adult (age 2-4 months) male Fischer 344 x Brown Norway rats were obtained form Harlan Sprague-Dawley (Indianap olis, IN) for all experiments described in this dissertation. Upon arrival, rats were examined and remained quarantined for one week. Animals were individually caged w ith a 12:12 hour light:d ark cycle (07:00 to 19:00 hr). Animals were cared for in accordance with the principles of the NIH Guide to the Care and Use of Experimental Animals. Construction of r AAV Vector Plasmid pTR-ObW encodes rat leptin cDNA (a kind gift from Roger Unger (Chen et al., 1996) and green fluorescent protein (GFP) reporter gene cDNA under the control of chicken -actin promoter linked to CMV enhan cer (CBA). Parenthetically, this vector was referred to as “rAAV-leptin” after it was packaged into recombinant adenoassociated virus. The woodchuck hepatitis virus posttranscriptiona l regulatory element (WPRE) was placed downstream to enhance the expression of the transgenes (Loeb et al., 1999). The control vector (referred to as “r AAV-con” subsequent to packaging) encodes GFP driven by a CBA. Vectors contain AAV terminal repeats at both sides of the cassette to mediate replication and packaging of the vector (Bell et al., 1999). A vector system with an inducible promoter, “TETOb ”, was also prepared. Like pTR-ObW described above, pTR-tetROb encodes rat leptin cDNA and GFP cDNA. However, in this case both genes are under th e control of a tet-inducible pr omoter (tetR). This tetR

PAGE 46

29 promoter is activated by the product of the accessory vector, pTR-rtTA/tTS, expressing mutually exclusive reverse transactivator rt TA (Tet-On) and transcriptional silencer (tTS). In the accessory vector the rtTA and tTS transgenes are linked within dicistronic cassettes through an IRES element for coordi nate expression. C onstruction of rAAVvector plasmids, including the TETOb system (Chapter 3) a nd constitutively active rAAV-leptin (Chapters 3 and 4), was done vi a a collaborative arrangement with Drs. Sergei Zolotukhin and Victor Prima of the University of Florida’s Department of Molecular Genetics. Packaging of rAAV Vectors Vectors were packaged, purified, concentrat ed, and titered as described previously (Conway et al., 1999). The titer of rAAVOb was 2.3E13 physical particles/mL. A mini-adenovirus helper plasmid (pDG) (Gri mm et al., 1998) was used to produce rAAV vectors with no detectable adenovirus or wild type AAV contamination. rAAV vectors were purified using iodixanol gradient/hepar in-affinity chromatography and were more than 99% pure as judged by PAAG/silver-stained gel elect rophoresis (not shown). Packaging of rAAV viral vectors, was done via a collaborative arrangement with Drs. Sergei Zolotukhin and Victor Prima of the University of Florida’s Department of Molecular Genetics. Stereotaxic Injections Third Cerebroventricle Rats were anesthetized with 60 mg/kg pe ntobarbital and heads were prepared for surgery. Animals were placed into a stereota xic frame and a small incision (1.5 cm) was made over the midline of the skull to expos e the landmarks of the cranium (Bregma and Lamda). The following coordinates were used for injection into 3rd cerebroventricle: 1.3

PAGE 47

30 mm posterior to Bregma and 9.4 mm ventra l from the skull surface on the midline (medial fissure), with the nose bar set at 3.3 mm below the ear bars (below zero) and the canula set at 20 posterior fr om vertical. A small hole was drilled through the skull and a 23-gauge stainless steel guide canula was lowered to the 3rd cerebroventricle. This was followed by an injection canula attached to a 10uL syringe. Hypothalamic Rats were anesthetized with 60mg/kg pe ntobarbital and heads were prepared for surgery. Animals were placed into a stereota xic frame and a small incision (1.5 cm) was made over the midline of the skull to expos e the landmarks of the cranium (Bregma and Lamda). The following coordinates were used for direct hypothalamic injection: 1.8 mm posterior to Bregma, 0.8 mm right of midline (Medial Fissure), and 9.0 mm ventral from the skull surface. The nose bar was set at zer o (on same plane with ear bars) and the canula was set vertically. A small hole was drilled through the skull and a 23-gauge stainless steel guide canula was lowered to the hypothalamus. This was followed by an injection canula attached to a 10uL syringe. Oxygen Consumption O 2 consumption was assessed in up to four rats simultaneously with an Oxyscan analyzer (OXS-4; Omnitech Electronics, Columbus, OH) as described previously (Scarpace et al., 1997). Flow rates were 2 L/min with a 30-s sampling time at 5-min intervals. The rats were placed into the chamber for 150 min with the lowest 6 consecutive O 2 consumption values during this peri od used in the calculations (basal resting VO2) Food was not available. Animal rooms were free of human activity and kept as quiet as possible during measuremen ts. All measurements were made between

PAGE 48

31 09:00 and 14:00 hrs. Resu lts were expressed as O2 consumption relative to metabolic body size (ml min -1 kg2/3). Tissue Harvesting Anesthetized rats (85 mg/kg pentobarbital) were sacrificed by cervical dislocation. Blood was collected by cardiac puncture and serum was harvested by a 10 minute centrifugation in serum separator tubes. The circulatory system was perfused with 20 mL of cold saline. Perirenal and retroperitone al white adipose tissue and hypothalami were excised, weighed, and immediately frozen in liquid nitrogen. The hypothalamus was removed by making an incision medial to pirifo rm lobes, caudal to the optic chiasm, and anterior to the cerebral crus to a depth of 2-3 mm. Tissues were stored at –80 C until analysis. Serum Measurements Leptin Serum leptin was measured using a ra t leptin radioimmunoassay kit (Linco Research, St. Charles, MO) or a rodent leptin ELISA kit (Crystal Chem, Chicago IL). Insulin Serum insulin was measured using a ra t insulin radioimm unoassay kit (Linco Research, St. Charles, MO). Glucose Serum glucose was via a colormetric reac tion with Trinder, the Sigma Diagnostics Glucose reagent (Sigma, St. Louis MO). Free Fatty Acids Serum free fatty acids were measured using the NEFA C colorimetric kit from WAKO Chemicals GmbH (Neuss, Germany).

PAGE 49

32 CSF Leptin CSF leptin was measured by a modification to the Crystal Chem leptin ELISA. Standards were prepared ranging from 100 dow n to 3.1 pg/mL. All modifications were applied to “reaction 1” of the ELISA, where leptin from the biological sample (or standards) binds to solid phase antibody on the bottom surface of the wells within the microplate. These modifications are as follows : 50 L of standard or CSF were used in reaction 1 along with 50 L of “guinea pig antimouse leptin serum” (provided in kit). “Sample diluent” was omitted from reaction 1 in an attempt to enhance the detection limit of the kit. In the manufacturer’s protocol, 5 L of serum or plasma is used in reaction 1 along with 45 L of sample diluent.. “R eaction 2,” which consists of the binding of enzyme-linked secondary antibody to the pr imary antibody-bound lept in, was performed as described in the manufacturer’s instruc tions. Using this m odification, we have a detection limit of approximately 6 pg/mL C SF. The final absorbance of our 6.25 pg/uL standard is significantly greater than that of a blank. Probes Leptin mRNA was detected using a 33 -mer antisense oligonucleotide (5’GGTCTGAGGCAGGGAGCAGCTC TTGGAGAAGGC-3’) probe. POMC mRNA was detected using a 24-mer antisen se oligonucleotide probe (5’CYYGCCCACCGGCTTGCCCCAGCG-3’). Oli gonucleotide probes were end labeled by terminal deoxynucleotidyl tran sferase (Promega). The UCP-1 probe is a full length cDNA clone and was obtained from Dr. Le slie Kozak, Pennington Research Center, Baton Rouge, LA. The AgRP cDNA probe was provided by Dr. Michael Schwartz (University of Washington). The rat pre pro NPY cDNA was provided by Janet Allen (University of Glasgow, UK). SOCS3 cDNA was a gift from Christian Bjorbaek

PAGE 50

33 (Harvard University). The cDNA probes were labeled using a random primer kit (Primea-Gene, Promega, Madison, WI). Probes were purified with Nick columns (Pharmacia) and, except for oligonucleotide probes, were heat-denatured for 2 minutes. All probes have been verified to hybridized to th e corresponding specific mRNAs by Northern Analysis prior to use in Do t Blot assay (below). RNA Isolation and RNA Dot Blot Tissue was sonicated in guanidine buffe r, phenol extracted, and isopropanol precipitated using a modificat ion of the method of Chomczynski and Sacchi (1987). Isolated RNA was quantified by spectrophotomet ry and integrity is verified using 1% agarose gels stained with ethidium bromide. For dot blot analysis, multiple concentrations of RNA were immobilized on nylon membranes using a dot blot apparatus (BioRad, Richmond, CA). Membranes were baked in a UV crosslinking apparatus. Membranes were then prehybridized in 10 mL Quickhyb (Stratagene, LaJolla, CA) for 30 minutes followed by hybridization in the pr esence of a labeled probe and 100 ug salmon sperm DNA. After hybridization for 2 hours at 65 C, the membranes were washed and exposed to a phosphor imaging screen for 24-72 hours (depending on anticipated strength of signal). The screen was then scanned using a Phosphor Imager (Molecular Dynamic, Sunnyvale, CA) and analyzed by Image Quant Software (Molecular Dynamics). Expression data was obtained for the following transcripts using dot blot analysis: UCP1, NPY, POMC, SOCS3, and leptin. Relative-Quantitative RT-P CR Using 18S Competimers Relative quantitative RT-PCR was perfor med using QuantumRNA 18s Internal Standards kit (Ambion, Austin, Tx). Tota l RNA (3 g) was treated with RNase-free DNase using a DNA-free kit (Ambion), and fi rst-strand cDNA synthesis generated from

PAGE 51

34 1 g RNA in a 20 l volume using random primers (Gibco BRL) containing 200 units of M-MLV reverse transcriptase (Gibco BRL). Relative PCR was performed by multiplexing target gene primers and 18s primers and coamplifing for a number of cycles found to be in the linear range of the targ et. Linearity for the leptin amplicon, for example, was determined to be 25-30 cycles. The optimum ratio of 18s primer to competimer is 1:9. PCR was performed at 94C denaturation for 120 sec, 59C annealing temperature for 60 sec, and 72 C elongation temperature for 120 sec for 27 cycles. The PCR product was electrophoresed on a 5% acrylamide gel and stained with SYBR green (Molecular Probes, Eugene, OR.) Gels were scanned using a STORM fluorescent scanner and digitized data analyzed using imagequant. (Molecular Dynamics). Real-time RT-PCR for Leptin Receptor Real-time RT-PCR was used to quantify the effects of obesity and caloric restriction on leptin receptor expression. We designed primers and a Taqman probe specific for the long-form leptin receptor (O b-Rb) using Primer Express software, version 1.5 (Perkin-Elmer Applied Biosystems, Inc., Fo ster City, CA). The sequences for the ObRb primers were forward primer: 5’-GGGAACCTGTGAGGATGAGTGT-3’, reverse primer: 5’-TTTCCACTGTTTTCACGTTGCT-3’. The fluorescent probe sequence was: 6FAM-AGAGTCAACCCTCAGTTAAATATGCAACGCTG-TAMRA. Optimization experiments showed that 300 nM of forwar d primer, 900 nM of reverse primer, and 50 nM Taqman probe gave the most reproducib le results and maximally efficient PCR (i.e., lowest threshold cycle (CT) values). Total RNA (6 g) was treated with RNase-free DNase using a DNA-free kit (Ambion). First-strand cDNA was generated from 1.6 g

PAGE 52

35 RNA in a 40 L volume using random primers (Gibco BRL) containing 200 units of MMLV reverse transcriptase (Gibco BRL). Real-time PCR for Ob-Rb was performed on 100 ng cDNA template in a 50 L total vo lume including Taqman RT-PCR Master Mix (Applied Biosystems, Inc., Foster Cit y, CA) using an ABI Prism GeneAmp 5700 Sequence Detection System (Applied Biosys tems). Ob-Rb expression was quantified using an 18S rRNA standard (Applied Biosystems) and the CT method (Bustin, 2000). The mean CT in the control group (CHOW) was chosen as the calibrator for CT calculation. STAT3/Phospo-STAT3 Assay These methods were described in detail pr eviously (Scarpace et al., 2000b). Briefly, hypothalamus was sonicated in 10 mM Tris -HCL, pH 6.8, 2% SDS, and 0.08 ug/mL okadaic acid plus protease inhibitors (PMFS, benzamidine, and leupeptin) [an aliquot of this sonicate was frozen for RNA analysis ]. Sonicate was diluted and quantified for protein using a detergent comp atible Bradford Assay. Samples were boiled and separated on an SDS-PAGE gel and electrotransfe rred to nitrocellulose membrane. Immunoreactivity was assessed with an antibody specific to tyrosine-705phosphorylated-STAT3 (antibody kit from Ne w England Biolabs, Beverly, MA). Immunoreactivity was visualized by chemilu minescents detection (Amersham Life Sciences, Piscataway, NJ) and quantified by vi deo densitometry (BioRad, Hercules, CA). Following P-STAT3 quantification, membra nes were stripped of antibody with Immunopure (Pierce, Rockford, IL) and immuno reactivity was re-assessed using a total STAT3 antibody. STAT3 phosphorylation is expressed as P-STAT3/Total STAT3 in each sample.

PAGE 53

36 UCP-1 Protein in Brown Adipose Tissue Approximately 30 mg of interscapular brown adipose tissue was homogenized in 300 uL 10 mM Tris-HCL, pH 6.8, 2% SDS, a nd 0.08 ug/mL okadaic acid plus protease inhibitors (PMFS, benzamidine, and leupeptin ). Samples were boiled for 5 minutes to and an aliquot of this homogenate was wit hdrawn and diluted for detergent compatible Bradford protein analysis. Samples were boiled and separated on an SDS-PAGE gel (20 g protein/lane) and electrot ransferred to nitrocellulose membrane. Immunoreactivity was assessed with an antibody specific to UC P-1 (Linco Research, St. Charles, MO). Immunoreactivity was visualized by chemilu minescents detection (Amersham Life Sciences, Piscataway, NJ) and quantified by vi deo densitometry (BioRad, Hercules, CA). Statistical Analysis All data are expressed as mean standard error of mean. level was set at 0.05 for all analyses. Data were analyzed by 2-Way ANOVA, 1-Way ANOV A, or student’s ttest, as appropriate. A Tukey’s pos t-hoc was used for 1-Way ANOVAS post-hoc analysis or a bonferoni multiple comparison test correcting for the num ber of contrasts. The post-hoc analysis of 2-Way ANOVAS de pended on the presence or absence of interactions. When only 2-Wa y ANOVA main effects were signi ficant, relevant pairwise comparisons were made using the Bonferroni Multiple Comparison method with the error rate corrected for the number of contrasts. When there was an inte raction, factors were separated and a further 1-Way ANOVA was applied with a Bonferroni Multiple Comparison post-hoc. When separation of f actors resulted in onl y two population means to compare, the 1-WAY ANOVA was replaced with student’s t-test. GraphPad Prism software version 3.0 (San Diego, CA) was used for all statistical analysis and graphing.

PAGE 54

37 GraphPad QuickCalc (graphpad.com) was used for post-hoc analysis of all 2-WAY ANOVAS.

PAGE 55

38 CHAPTER 3 INCORPORATING REGULATION INTO LEPTIN GENE THERAPY Introduction Leptin, the product of the Ob gene, acts on satiety cente rs in the hypothalamus to both decrease food intake and increase energy expenditure. Virus-me diated leptin gene delivery has been shown to cause a rapid a nd complete disappearance of white adipose tissue in genetically normal animals (Chen et al., 1996). Leptin gene delivery is substantially more effective in correcting genetic obesity than daily leptin infusion despite higher peak serum leptin levels with the latter (Morsy et al., 1998). Peripheral administration of virally-encode d leptin requires that the leptin cross the blood-brain barrier (BBB) to reach its pr imary hypothalamic targets, but this transport system is saturated at serum leptin levels seen in obese animals (Banks et al., 1999). Since several obese, leptin resistant models respond better to central versus peripheral administration of recombinant leptin (Halaas et al., 1997; Niim i et al., 1999; Van Heek et al., 1997), it has been reasoned that central delivery of the leptin gene would be superior to peripheral delivery (Lundberg et al., 2001). Indeed, central delivery of virally-encoded leptin causes dramatic lipopenia while avoiding systemic hyperleptinemia and the limitations of BBB transport (Lundberg et al., 2001; Scarpace et al., 2002b). In Experiment 1 of this chapter, I administered rAAV-encoding leptin under the control of a constitutively active CBA promoter (Figure 3-1) in an attempt to duplicate our previous finding of potent weight and fat loss in with this vector.

PAGE 56

39 The main limitation of viral-mediated lept in-gene therapy systems studied to date, including the vector used in Experiment 1 is the lack of post-transf ectional control. This is undesirable clinically as it would be diffi cult or impossible to reverse the progression of deleterious side effects as they appeared. Given leptin ’s diverse biological functions, central overexpression of leptin (coupled with a precipitous fall in serum leptin) could compromise the immune (Matarese, 2000) a nd reproductive systems (Ahima and Flier, 2000a). Central leptin signaling is also known to decrease bo ne mass (Ducy et al., 2000). Peripheral overexpession of leptin, on the other hand, leads to pa ncreatic beta cell disfunction (Koyama et al., 1997) and upregul ates suppressor of cytokine signaling, which may interfere with insulin sign aling (Spiegelman and Flier, 2001). We have attempted to solve these proble ms of uninhibited transgene expression by placing leptin under the control of the tetr acycline transactivator (rtTA) and operon (tetR). In this dual-vector system, one vector encodes the tetR promotor and the leptin gene while an assessory vector endodes rtTA and a transcrip tional silencer (tTS) [Figure 3-2]. This dual vector “TETOb ” system allows for the control leptin transgene expression via doxycycline (doxy) in drinking water. In Experiment 2 we administered the TETOb dual-vector system into the hypothalami of young adult, non-obese rodents, and examined the effects on anorectic signali ng in the brain, peripheral thermogenesis, and adiposity. Moreover, we de termined if these effects can be completely reversed if the transgene is silenced. Methods and Materials Animals Young adult (age ~4 months) male Fischer 344 x Brown Norway rats were obtained form Harlan Sprague-Dawley (Indian apolis, IN). Upon arrival, rats were

PAGE 57

40 examined and remained quarantined for one week. Animals were individually caged with a 12:12 hour light:dark cy cle (07:00 to 19:00 hr). Animals were cared for in accordance with the principles of the NIH Gu ide to the Care and Use of Experimental Animals. Construction of r AAV Vector Plasmid Constitutively active leptin and control vector. pTR-ObW (Fig 3-1) encodes rat leptin cDNA (a gift from Guoxun Chen (Chen et al., 1996) and green fluorescent protein (GFP) reporter gene cDNA under the co ntrol of chicken -actin promoter linked to CMV enhancer (CBA). The woodchuck hepa titis virus posttranscr iptional regulatory element (WPRE) was placed downstream to e nhance the expression of the transgenes (Loeb et al., 1999). Control ve ctor (pTR-control) is identic al to pTR-ObW but without the leptin cDNA and IRES element. The control vector encodes GFP alone. TETOb vector system with inducible promoter. Like pTR-ObW described above, pTR-tetROb (Figure 3-2, top) encodes rat leptin cDNA and GFP cDNA. However, in this case both genes are under the control of a tet-inducible promoter (tetR). The woodchuck hepatitis virus posttranscrip tional regulatory el ement (WPRE) was placed downstream to enhance the expression of the transgenes (Loeb et al., 1999). This tetR promoter is activated by the product of the accessory vector, pTR-rtTA/tTS (Figure 3-2, bottom), expressing mutually exclusive re verse transactivator rtTA (Tet-On) and transcriptional silencer (tTS). In the accessory vector, the rtTA and tTS transgenes are linked within dicistronic casse ttes through an IRES element for coordinate expression. All vectors contain AAV terminal repeats at both sides of the cassette to mediate replication and packaging of the vector (Bell et al., 1999).

PAGE 58

41 Packaging of rAAV Vectors Vectors were packaged, purified, concentrat ed, and titered as described previously (Conway et al., 1999). The titers of rAAV-tetROb and rAAV-AS (accessory virus) used were 6.6E12 and 1.37E12 infectious particle s/mL, respectively. The titer of the constitutively active rAAV-leptin was 6.0E 9 infectious particles/mL. The physical particle titer is approximately 100 fold greater than the infecti ous titer in all cases. A mini-adenovirus helper plasmid (pDG) (Gri mm et al., 1998) was used to produce rAAV vectors with no detectable adenovirus or wild type AAV contamination. rAAV vectors were purified using iodixanol gradient/hepar in-affinity chromatoagraphy and were more than 99% pure as judged by PAAG/silver-stained gel electro phoresis (not shown). Vector Administration Third ventricle injection Vectors were delivered into the 3rd cerebroventricle in Experiment 1 Rats were anesthetized with 60 mg/kg pe ntobarbital and heads were prepared for surgery. Animals were placed into a stereotaxic frame and a small incision (1.5 cm) was made over the midline of the skull to expos e the landmarks of the craniu m (Bregma and Lamda). The following coordinates were us ed for injection into 3rd cerebroventricle: 1.3 mm posterior to Bregma and 9.4 mm ventral from the s kull surface on the midline (medial fissure), with the nose bar set at 3.3 mm below the ear bars (below ze ro) and the canula set at 20 posterior from vertical. A sm all hole was drilled through the skull and a 23-gauge stainless steel guide canula was lowered to the 3rd cerebroventricle. This was followed by an injection canula attached to a 10uL syringe. I injected 3.0 uL containing 1.8E9 infectious

PAGE 59

42 Figure 3-1: pTR-ObW. pTR-ObW encode s rat leptin cDNA (and green fluorescent protein (GFP) reporter gene cDNA under the control of a constitutively active chicken -actin promoter linked to CM V enhancer (CBA). Control vector (pTR-control) is identical to pTR-O bW but without the leptin cDNA and IRES element. The control vector en codes GFP. The cassette flanked by the AAV terminal repeats (TR) in each plasmid were packaged into rAAV to yield the rAAV-leptin vector. pTR-betaObW7139 bp ApR ColE1 ori f1(+) origin TR TR IRES WPRE hGFP F64L/S65T Ob cDNA, rat CMV ie enhancer Chimeric intron bGH poly(A) Chiken b-actin promote r Exon1 Chicken –actin

PAGE 60

43 Figure 3-2: The TET-responsive promotor/leptin plasmid construct pTR-tetR-Ob (top), and the assessory plasmid cons truct pTR-rtTA/tTS (bottom). pTR-tetR-Ob6198 bp ApR TR TR IRES GFP (F64L, S65 T Ob cDNA, r at SV40 poly(A) Tet responsive promot er ColE1 ori f1(+) origin pTR-rtTA/tTS7157 bp TR TR IRES rtTA tTS CMV ie enhancer Chimeric intron SV40 poly(A) bGH poly(A) Chiken b-actin promot e ColE1 ori f1(+) origin Exon1 Chicken -Actin

PAGE 61

44 Figure 3-3: Schematic repres entation of leptin transgene regulation by doxycycline in dual vector “TET-Ob” system, “OFF” state. In this dual vector system, the accessory vector (top) encodes a transc riptional activator (rtTA, green ovals) and a transcriptional silencer (tTS, bl ue diamonds), both of which are under the control of a constitutively active chicken -actin (CBA) promoter. The TET-leptin vector (bottom) encodes leptin under the control of a Tet responsive promoter. In the default state, the transcriptional silencer binds the Tet Responsive Promoter, preventing tran scriptional activat or binding to the promoter and blocking leptin expression. Thus, in the abse nce of tetracycline analogue such as doxycyline, this system is in the “OFF” state.

PAGE 62

45 Figure 3-4: Schematic repres entation of leptin transgene regulation by doxycycline in dual vector “TET-Ob” system, “ON” st ate. In presence of doxycycline (yellow circles), the relative affinities of the transcriptional activator (green ovals) and the transcriptional silencer (blue diamonds) for the Tet Responsive Promoter are reversed. The transcrip tional silencer is displaced from the promoter and the transcriptional activato r binds to the promoter, activating the expression of leptin transgene. Thus, in the presence of doxycyline, this system is in the “ON” state.

PAGE 63

46 viral particles dissolv ed in Ringer’s solution at approxi mately 0.25 uL/minute. Animals received either the constitutively active r AAV-leptin vector (n=7) or rAAV-con encoding GFP (n=3). Hypothalamic injection The dual vector TETOb system was micro-injected into the center of the right hypothalamus to maximize co-infecti on of cells. As with the 3rd cerebroventricle injections, rats were anesthetized with 60 mg/kg pentobarbital and heads were prepared for surgery. Animals were placed into a st ereotaxic frame and a small incision (1.5 cm) was made over the midline of the skull to expose the landmarks of the cranium (Bregma and Lamda). The following coordinates were used for direct hypothalamic injection: 1.8 mm posterior to Bregma, 0.8 mm right of mid line (Medial Fissure), and 9.0 mm ventral from the skull surface. The nose bar was set at zero (on same plane with ear bars) and the canula was set vertically. A small hole was drilled through the skull and a 23-gauge stainless steel guide canula was lowered to the hypothalamus. This was followed by an injection canula attached to a 10uL syringe. I injected 5 uL of viral particles in Ringer’s solution at approximately 0.25 uL/minute. Animals received either the two-vector TETOb system containing equal amounts (6E9 inf ectious particles) of rAAV-TetR-Ob and accessory vector (n=14) or control virus encoding GFP (n=6). Experimental Design Experiment 1 The aim of Experiment 1 was to study the effects of constitutively active rAAVleptin on body weight, food intake, and adiposity. I injected rAAV-leptin (n=7) or control virus encoding GFP (rAAV-con) (n=3) in to the 3rd ventricle of 4 month old male

PAGE 64

47 F344xBN rats using a stereotaxic apparatus and monitored food intake and body weight for ~60 days, at which point animals we re euthenized for tissue analysis. Experiment 2 The primary aim of Experiment 2 was to test our ability to regulate leptin transgene expression using the TETOb system. Experiment 2 is sub-divided into two stages. During STAGE 1, doxycycline hydrochloride (S igma, St. Louis, MO) (400 ug/mL) and 0.1% sacharrine (Sigma) were provided in the drinking water of all animals beginning day 9 post-transfection. All rats received doxy for a total of 34 days. At the end of STAGE 1, doxy was withdrawn from half of the TETOb rats for 32 days (TETOb -OFF, n=7) while half continued to receive Doxy (TETOb -ON, n=7). This second phase of the study is designated as STAGE 2. Controls received doxy throughout the study (i.e., both STAGE 1 and STAGE 2). Tissue Harvesting Rats were anesthetized with 85 mg/kg pentobarbital and sacrificed by cervical dislocation. Blood was collected by cardiac puncture and serum was harvested by a 10 minute centrifugation in serum separator tubes. The circulatory system was perfused with 20 mL of cold saline. Inguinal, perirenal, and retrop erioneal white adipose tissue, brown adipose tissue, and hypotha lami were excised, weighed, and immediately frozen in liquid nitrogen. Tissues were st ored at –80 C until analysis. Serum Leptin, FFA, Insulin, and Glucose Serum leptin and insulin were measured using rat radioimmunoassay kits (Linco Research, St. Charles, MO). Serum free fa tty acids were measured using the NEFA C colorimetric kit from WAKO Chemicals Gm bH (Neuss Germany). Serum glucose was

PAGE 65

48 measured via a colormetric reaction with Tri nder, the Sigma Diagnostics Glucose reagent (Sigma, St. Louis MO). Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR) Hypothalamic leptin transgene expressi on and Ob-Rb expression were evaluated by using relative quantitative RT-PCR through the use of QuantumRNA 18s Internal Standards kit (Ambion, Austin, Tx). Total RNA (3 g) was treated with RNase-free DNase using a DNA-free kit (Ambion), and first-strand cDNA synthesis was generated from 1 g RNA in a 20 l volume using random primers (Gibco BRL) containing 200 units of M-MLV reverse transc riptase (Gibco BRL). Rela tive PCR was performed by multiplexing leptin primer s (sense: 5’GGAGGAATCC-CTG CTCCAGC; antisense: 5’ CCTCTCCTGAGGATACCTGG) and 18s primers and coamplifing for a number of cycles found to be in the linear range for leptin. Line arity for the leptin amplicon was determined to be 25-30 cycles. The optimum ratio of 18s primer to competimer was 1:9. PCR was performed at 94C denaturation for 120 sec, 59C annealing temperature for 60 sec, and 72C elongation temperature for 120 sec; for 27 cycles. Similarly, the number of cycles found to be at the mid point of the linear range was 26 cycles for Ob-Rb. The sequence for Ob-Rb primers were se nse: GGGAACCTGTGAGGATGAGTG; antisense: TAGCCCCTTGCTCTTCATCAG. The PCR pr oducts were electrophoresed on a 5% acrylamide gel and stained with SYBR green (Molecular Probes, Eugene, OR.) Gels were scanned using a STORM fluorescent scan ner and digitized data analyzed using imagequant. (Molecular Dynamics). STAT3/Phospo-STAT3 Assay These methods were described in detail previously (Scarpace et al., 2000b). Briefly, hypothalamus was sonicated in 10 mM Tris-HCL, pH 6.8, 2% SDS, and 0.08

PAGE 66

49 g/mL okadaic acid plus proteas e inhibitors (PMFS, benzam idine, and leupeptin) [an aliquot of this sonicate was frozen for RNA analysis]. Sonicate was diluted and quantified for protein using a detergent compat ible Bradford Assay. Samples were boiled and separated on an SDS-PAGE gel and electrot ransferred to nitrocellulose membrane. Immunoreactivity was assessed with an an tibody specific to phosphorylated-STAT3 (antibody kit from New England Biolabs, Beverly MA). Immunoreactivity was visualized by chemiluminescents detection (Amersham Life Scien ces) and quantified by video densitometry (BioRad). Followi ng P-STAT3 quantification, membranes were stripped of antibody (Pierce) and immunoreactivity was re-assessed using a total STAT-3 antibody Probes The UCP-1 probe is a full length cDNA clone and was obtained from Dr. Leslie Kozak, Pennington Research Ce nter, Baton Rouge, LA. The rat pre pro NPY cDNA was provided by Janet Allen (University of Glasgow, UK). SOCS3 cDNA was a gift from Christian Bjorbaek (Harvard Univers ity). The cDNA probes were labeled using a random primer kit (Prime-a-Gene, Promega, Madison, WI). Probes were purified with Nick columns (Pharmacia) and, except for o ligonucleotide probes, were heat-denatured for 2 minutes. All probes have been verified to hybridized to the corresponding specific mRNAs by Northern Analysis prior to use in Dot Blot assay (below). mRNA Levels (Dot Blot Analysis) Tissue was sonicated in guanidine buffer, phenol extracted, and isopropanol precipitated using a modification of the method of Chomczynski (Chomczynski and Sacchi, 1987). Isolated RNA was quantified by spectrophotometry and integrity was verified using 1% agarose gels stained with ethidium bromide. For dot blot analysis,

PAGE 67

50 multiple concentrations of RNA were immobilized on nylon membranes using a dot blot apparatus (BioRad, Richmond, CA). Memb ranes were baked in a UV crosslinking apparatus. Membranes were then prehybrid ized in 10 mL Quickhyb (Stratagene, LaJolla, CA) for 30 minutes followed by hybridization in the presence of a labeled probe and 100 g salmon sperm DNA. After hybridizati on for 2 hours at 65 C, the membranes were washed and exposed to a phosphor imaging screen for 24-72 hours (depending on anticipated strength of signal). The screen was then scanned using a Phosphor Imager (Molecular Dynamic, Sunnyvale, CA) and analyzed by Image Quant Software (Molecular Dynamics). Statistical Analysis Food intake and bodyweight comparisons were made by a 2-Way ANOVA with rAAV vector and time as factors. Comparis ons of absolute body weights, mean daily food intake, adiposity, and serum lepin in rAAV-leptin versus rAAV-con were made using unpaired t-test. Comparisons between control and TET-Ob during stage 1 were made using unpaired t tests with a one-tailed p value. For st age 2 and endpoint statistics, relevant pairwise comparisons were made using unpaired t tests with the Bonferroni correction for multiple comparisons applied to the level, which was set at 0.05. We chose this test over a 1-WAY ANOVA because we did not have 3 independent groups receiving 3 treatments during this experi ment as assumed in a 1-way ANOVA. TETOb ON and TETOb -OFF were one group (TETOb ) during the first half of the experiment and then were split at the midpoint. Th e controls, on the other hand, were always independent of the other two groups. More over, we are not testing the single null hypothesis that Control = Ob -ON = Ob -OFF for each of our experimental parameters, as in the ANOVA. Rather, we our testing two sepa rate null hypothesis at the conclusion of

PAGE 68

51 the experiment: (1) Ob -ON = Control, and (2) Ob -ON = Ob -OFF. The first test is to determine whether or not the TETOb system had an effect. The second test is to determine if we were able to de-activate the transgene. In the case of leptin expression, we tested a third null hypothesis, Ob -OFF = Control, and adjusted the alpha level accordingly (i.e., =0.05/3). Rejection of the null hypothesis in this latter test would indicate some leaky expression of transgen e after withdrawing doxycycline. Since we corrected for the error rate associated with multiple pairwise comparisons, this method does not increase our chance of a Type I error. GraphPad Prism software (San Diego, CA) was used for all statistical analysis and graphing. Results Experiment 1 Food consumption and body mass On the day of vector delivery, body ma ss did not differ between controls and rAAV-leptin treated animals (252.711.3 vs 260.91.7 g, respectively) [Fig 3-5]. Following gene delivery, control (rAAV-con) rats steadily gained mass while rAAVleptin rats had a slight decr ease in body mass such that rAAV-leptin rats weighed 31.5% less than controls 61 days after ge ne delivery (329.817.2 vs. 250.911.0 g, p<0.0001) [Fig 3-5]. rAAV-leptin trea tment also resulted in a signi ficant anorectic effect versus controls beginning on day 6 post-transfection [Figure 3-6]. Between days 6 and 61, mean daily food intake was 20.1% less in rAAV-leptin treated rats (17.280.19 vs. 13.800.30 g/day, p<0.0001) [Figure 3-6]. Serum leptin and adiposity The suppression in age-associated wei ght gain in the rAAV-leptin group was accompanied by the complete disappearance of white adipose tissue [Figure 3-7]. The

PAGE 69

52 sum of two major visceral white fat depots (retroperitoneal white adipose [RTWAT] and perirenal white adipose [PWAT] depots) was 3.08 0.29 g in the control rats ~61 days after gene delivery, whereas visceral fat pads were absent in rAAV-leptin treated group. We know from other experiments in this la boratory that this complete catabolism of white adipose depots takes place within 9 da ys (Scarpace et al., 2002b). Moreover, this loss of white fat is accompanied by an appa rent preservation of lean body mass (Dhillon et al., 2001). Neither caloric restriction nor any drug known to the author can replicate this dramatic, rapid and selective loss of white fat. Consistent w ith the loss of adipose tissue in rAAV-leptin treated rats, serum lep tin (an index of adipos ity) was undetectable in these animals [Figure 3-8]. Experiment 2 Food consumption and body mass Beginning day 9 post-transfection, doxy was provided in the drinking water of all animals for 34 days (STAGE 1), presumably activating the leptin transgene in all TETOb treated animals [Figure 3-9]. During STAGE 1, TETOb rats gained 51.7 % less mass [Figure 3-10, between arrows] and ate 11.4 % less food [Figure 3-11] than controls. To begin STAGE 2, the TETOb group was divided into two subgroups. Doxy was withdrawn from half of the TETOb treated animals for 32 days (TETOb -OFF) while half continued to receive doxy (TETOb -ON) [Figure 3-9]. During STAGE 2, TETOb ON rats gained 44.8 % less mass than TETOb -OFF rats [Figure 3-10], and ate significantly less food than both TETOb -OFF and controls [Figure 3-11]. Rates of body mass gain were not different between TETOb -ON and controls during STAGE 2. The same degree of anorexia was maintained in TETOb -ON animals throughout both stages of the study. Doxy/saccharin did not affect fluid intake [data not shown].

PAGE 70

53 Serum leptin and adiposity Consistent with leptin-dependent catabolism of adipose tissue in TETOb rats, serum leptin fell to 22.5% of control valu es during STAGE 1 [Figure 3-12]. During STAGE 2, serum leptin increased to 83.4% of control values in the TETOb -OFF animals but remained very low (21.9% of controls) in the TETOb -ON group [Figure 3-12]. At sacrifice, average visceral adiposity (sum RTWAT and PWAT) was 14.6% of controls in TETOb -ON animals [Figure 3-13], and three of the seven TETOb -ON had no visible intra-abdominal adipose tissue [similar to Fi gure 3-7, right]. Visceral adiposity was 76.9% of controls in TETOb -OFF at sacrifice, suggesti ng these animals recovered adipose tissue during STAGE 2 [Figure 3-13]. Serum free fatty acids, insulin, and glucose At sacrifice, average serum free fa tty acids were 15.2% lower in the TETOb -ON group compared to the TETOb -OFF group and were 12.5% lower in TETOb -ON versus controls [Table 3-1]. Neither of these findings was statistically significant after employing the Bonferroni correction for multiple comparisons. Serum insulin was significantly reduced in the TETOb -ON as compared to both controls and TETOb -OFF [Table 3-1]. Serum glucose was unaffected by treatment [Table 3-1]. Leptin expression Leptin expression in the hypothalamus of TETOb -ON animals was increased 8.5 fold versus controls [Figure 3-14]. The level of hypothalamic leptin expression did not differ between TETOb -OFF and control, indicating that the transgene was silenced when doxy was withdrawn [Figure 3-14].

PAGE 71

54 Signal transduction in hypothalamus Hypothalamic phosphorylated STAT3 (P-STA T3), a component of the leptinsignaling cascade, was elevated by 46% in TETOb -ON group but not TETOb -OFF [Figure 3-15]. Total STAT3 was not affected [d ata not shown]. It is believed that an increase in POMC expression (the precursor to -MSH) contributes to the anorectic effect of leptin (Elias et al., 1999). In this study, hypothalami c POMC mRNA was found to be elevated by 41.2% in TETOb -ON animals versus controls and by 16.4 % versus TETOb -OFF [Figure 3-16]. Leptin is also known to up-regulate expression of suppressor of cytokine signaling (SOCS3) (W ang et al., 2000; Scarpace, 2002). SOCS3 mRNA was increased by 67.3% in TETOb -ON versus controls and by 34.9% versus TETOb -OFF [Figure 3-17]. Finally, leptin signaling is known to depress NPY expression in arcuate neurons (Spiegelman a nd Flier, 2001). However, we did not detect any differences in NPY expressi on (100.0 12.4 arbitrary units in control, 94.919.7 in TETOb -ON, 115.810.0 in TETOb -OFF). Leptin receptor expression in the hypothalamus Hypothalamic leptin rece ptor expression in TETOb -ON was 24% lower than that in control and 31% lower than TETOb -OFF [Figure 3-18]. However, only the difference between TETOb -ON and TETOb -OFF reached statistical significance. Nevertheless, this novel data suggests th at central overexpression of leptin may negatively regulate leptin receptor expression. Moreover, this effect appears to be reversed upon transgene silencing. CSF leptin Data from Dr. Scarpace’s laboratory dem onstrates an increase in CSF leptin concentrations when a cons titutively active leptin gene encoded by rAAV is delivered

PAGE 72

55 i.c.v. (Scarpace et al., 2002b). In the present study, I gave the TETOb system directly into the hypothalamus. In this case, I detected a significant decrease in CSF leptin in the TETOb -ON but not TETOb -OFF animals as compared to controls [Table 3-1]. Brown adipose tissue TETOb -ON animals had a 3-fold increase in UCP-1 protein levels per unit of BAT compared to both controls and TETOb -OFF [Figure 3-19]. Furthermore, TETOb -ON animals tended to have elevated UCP-1 mRNA in BAT compared to TETOb -OFF and controls [Table 3-1]. Consiste nt with activated BAT in the TETOb -ON group, their BAT appeared dark red (stimulated) whereas the BAT extracted from both controls and TETOb -OFF animals appeared pale brown (dor mant). Total BAT tissue mass declined in the TETOb -ON animals while the protein concentration (per unit BAT) increased [Table 3-1], suggesting that the decrease in mass was due to lipolysis. Even when the decrease in BAT mass is accounted for, there wa s still a 2-fold increase in UCP-1 protein per total interscapular BAT pad in TETOb -ON compared to both controls and TETOb OFF [Table 3-1]. Discussion A single injection of rAAV encoding leptin under the control of a constitutively active promoter has been shown to cause su stained (>2 months) anorexia and severe reductions in adiposity, and we repeated these findings in Experiment 1 We then hypothesized that a direct hypot halamic injection of rAAV encoding leptin under control of the tetracycline transa ctivator and operon (TETOb ) would allow us to regulate leptin transgene expression and the subsequent anorexic and thermogenic effects ( Experiment 2 ). In animals given the TETOb system, we demonstrated a reversible suppression in

PAGE 73

56 -10 0 10 20 30 40 50 60 70 200 225 250 275 300 325 350 375 rAAV-con rAAV-leptin DayBody Mass (g) Figure 3-5: Body mass following intracerebrove ntricular administrati on of rAAV-leptin. Values represent mean SEM. By 2-Way ANOVA, significance was found for effects of vector (F=83.91, p<0.0001), time (F=5.40, p<0.0001), and the interaction between vector and time (F=6.52, p<0.0001). -10 0 10 20 30 40 50 60 70 10.0 12.5 15.0 17.5 20.0 22.5 rAAV-con rAAV-leptin DayIntake (g)/24 hours Figure 3-6: Food intake follo wing intracerebrovent ricular administrati on of rAAV-leptin or rAAV-con. Values represent mean SEM. By 2-Way ANOVA, significance was found for effects of vector (F=77.50, p<0.0001), time (F=8.21, p<0.0001), and the interacti on of vector and time (F=3.54, p<0.0001).

PAGE 74

57 Figure 3-7: Disappearance of visceral adi pose tissue in 4 month old F344xBN rats administered rAAV-leptin. The an imal on the left received an intracerebroventicular inj ection of control rAAV encoding GFP. The animal on the right received a similar inject ion of rAAV-encoding-leptin. These pictures were taken 46 days after gene delivery, but no recovery in adiposity was observed at 61 days. rAAV-conrAAV-le p tin 0 1 2 3 4 rAAV-con rAAV-leptin <0.05 ng/mLng/mL serum Figure 3-8: Leptin is undete ctable in serum of animals following central leptin gene delivery, consistent with severe reducti on in adiposity. Values represent mean SEM. Measurements taken 61 days after vector delivery rAAV-con rAAV-leptin white adipose

PAGE 75

58 Figure 3-9: Experiment 2 design, TET-Ob gene delivery and regulation. Experiment 2 is sub-divided into two stages. Animal s were given a direct hypothalamic injection containing the dual-vector rAAV-TET-Ob system (top, n=14) or control rAAV encoding GFP (bottom, n=6). During STAGE 1, which began 9 days after gene delivery, doxycycli ne hydrochloride was provided in the drinking water of all animals, presumably activating the leptin transgene in the TET-Ob treated rats. STAGE 1 lasted 34 days. At the end of STAGE 1, doxy was withdrawn from half of the TET-O b rats for 32 days (TET-Ob-OFF, n=7) while half continued to receive Doxy (TET-Ob-ON, n=7). This second phase of the study is designated as STAGE 2. Controls receiv ed doxy throughout the study (i.e., both STAGE 1 and STAGE 2). During STAGE 2, the leptin transgene should be silen ced in the TET-Ob-OFF group, yet remain active in the TET-Ob-ON.

PAGE 76

59 -10 0 10 20 30 40 50 60 70 280 300 320 340 360 380 Control TET-Ob-ON TET-Ob-OFF STAGE 1 STAGE 2DayBody Mass (g) Figure 3-10: Body mass following TET-Ob or control vector delivery. rAAV-TET-Ob or control was administered on day (-)9 Left arrow represents start of STAGE 1. Doxy was withdr awn from half of the TET-Ob treated animals on day 34 (right arrow, start of STAGE 2) Values represent means SEM. p<0.001 for difference in body mass gained during STAGE 1 in TET-Ob and control. p<0.01 for difference in body mass gained during STAGE 2 in ObON and Ob-OFF. STAGE 1STAGE 2 15 16 17 18 19 20 Control TET-Ob(ON) Ob-OFF ** **Food g/day Figure 3-11: Daily food consumption following TET-Ob or control vector delivery. Values represent means SEM.. TET-Ob was divided into TET-Ob-ON and TET-Ob-OFF to begin STAGE 2. p<0.01 during STAGE 1 for difference between TET-Ob and control. p<0.01 during STAGE 2 for difference between TET-Ob-ON and both control and TET-Ob-OFF.

PAGE 77

60 Pre-DOXYSTAGE 1STAGE 2 0 1 2 3 4 5 Control TET-Ob(ON) Ob-OFF *** *** Serum Leptin (ng/mL) Figure 3-12: Serum leptin following TET-Ob or control vector delivery. Values represent means SEM. TET-Ob was divided into TET-Ob-ON and TET-Ob-OFF to begin STAGE 2. p<0.0001 for difference between TET-Ob and control at the end of STAGE 1. By the end of STAGE 2, serum leptin had increased to control levels in TET-Ob-OFF but remained low in TET-Ob-ON (p<0.0001 vs. control and TET-Ob-OFF). ControlOb-ONOb-OFF 0 1 2 3 4***Sum of RTWAT and PWAT (g) Figure 3-13: Visceral adiposity (sum of re troperitoneal [RTWAT] and perirenal white adipose tissue [PWAT] depots) following TET-Ob or control vector delivery. Values represent means SEM. p<0.0001 for difference between TET-ObON and controls and p<0.001 for difference between TET-Ob-ON and TET-Ob-OFF.

PAGE 78

61 ControlOb-ONOb-OFF 0 2 4 6 8**Leptin Expression (relative to 18S rRNA) Leptin 18S rRNA 1 2 3 4 5 6 Figure 3-14: Hypothalamic leptin expression 66 days after TET-Ob or control vector delivery. Values represent means SEM (Bar Graph, Top). p<0.001 for difference between TET-Ob-ON and control, p<0.01 for difference between TET-Ob-ON and TET-Ob-OFF. Image (below) is 5% acryilamide gel of representative relative quantitative PCR product. Top band is leptin amplicon, bottom band is 18S rRNA control amplicon for same sample, produced in same PCR reaction (multiplexing). La nes 1, 4: Control; lanes 2, 5: TET-ObON; lanes 3, 6: TET-Ob-OFF.

PAGE 79

62 ControlOb-ONOb-OFF 0 50 100 150 200***STAT3 Phosphorylation P-STAT3 1 2 3 4 5 6 7 8 9 Figure 3-15: Hypothalamic PSTAT3 66 days following TET-Ob or control vector delivery. Values represent means SE M (Bar Graph, Top). p<0.001 for difference between TET-Ob-ON and both control and TET-Ob-OFF. Image (Below) is of representative Western Blot for P-STAT3. Primary antibody is specific for tyrosine-phosphor ylated STAT3. Lanes 1, 3, 5: Control; lanes 2, 4, 7, 9: TET-Ob-ON; lanes 6, 8: TET-Ob-OFF. No differences across groups were detected for total hypothalamic STAT3 (image not shown).

PAGE 80

63 ControlOb-ONOb-OFF 0 50 100 150 200*POMC mRNA Figure 3-16: POMC expression 66 days following TET-Ob or control vector delivery. Values represent means SEM. p<0.01 for difference between TET-Ob-ON and control, p<0.025 for difference between TET-Ob-ON and TET-Ob-OFF. ControlOb-ONOb-OFF 0 50 100 150 200*SOCS3 mRNA Figure 3-17: SOCS3 expression 66 days following TET-Ob or control vector delivery. Values represent means SEM. p<0.025 difference between TET-Ob-ON and control, p<0.025 for difference between TET-Ob-ON and TET-Ob-OFF.

PAGE 81

64 ControlOb-ONOb-OFF 0.00 0.25 0.50 0.75 1.00 1.25 *Ratio ObRb/18S rRNA Figure 3-18: Long-form leptin receptor (O b-Rb) expression in the hypothalamus 66 days following TET-Ob or control vector delivery. Values represent means SEM. p<0.05 for difference between TET-Ob-ON and TET-Ob-OFF. Difference between TET-Ob-ON and control did not reach statistical significance (p=0.0582). 1 2 3 4 5 6 7 8 UCP-1 ControlOb-ONOb-OFF 0 100 200 300 400**Arbitrary Units/ mg BAT Figure 3-19: UCP-1 protein in BAT 66 days following TET-Ob or control vector delivery. Values represent means SEM (Bar Graph, Top). p<0.001 for difference between TET-Ob-ON and bot h control and TET-Ob-OFF. Image (Below) is of representative Western Bl ot for UCP-1. Lanes 1, 3: Control; lanes 2, 4, 6, 8: TET-Ob-ON; lanes 5, 7: TET-Ob-OFF. Note that lanes 1-4 and lanes 5-8 came from separate gels.

PAGE 82

65 Table 3-1: Brown adipose tissue parameters, serum free fatty acids, insulin, and glucose, and CSF leptin at sacrifice Parameter Control TETOb -ON TETOb -OFF BAT weight (mg) 302 24 211 18* 290 23 BAT protein (mg/g tissue) 102.7 7. 133.1 12.2† 101.4 6.6 UCP-1 mRNA (arbitrary units/g RNA) 100.0 6.0 119.5 8..9 92.6 9.7 UCP-1 protein (arbitrary units/total BAT) 100 12 208 27** 99 15 Serum free fatty acids (nMol/L) 0.442 .0.038 0.393 0.024 0.464 0.019 Serum insulin (U/mL) 27.8 5.5 11.6 3.8‡ 46.6 7.2 Serum glucose (mg/dL) 169.1 12.7 154.9 6.0 173.6 8.7 CSF leptin (pg/mL) 11.6 2.47 5.18 0.55* 6.49 1.61 Food was withdrawn 2 hours prior to collecting serum and tissue samples. Samples were collected between 10:00 and 13:00. Data represent the mean SEM. *p<0.025 vs. controls and TETOb -OFF; **p<0.01 vs. controls and TETOb -OFF, ***p<0.0001 vs. controls and TETOb -OFF, †p<0.025 vs. TETOb -OFF only, ‡ p<0.025 vs. controls and p<0.01 vs. TETOb -OFF. the rate of body mass gain and reversible anorexia. There was a dramatic 85.4% reduction in visceral adiposity in TET-Ob animals continuously administered doxycycline for 66 days (TET-Ob-ON). In a sub-population of TET-Ob animals from which we withdrew Doxy (TET-Ob-OFF), visceral adiposity was indistinguishable from controls. We believe this represen ts a recovery of adiposity in TET-Ob-OFF animals during STAGE 2. Consistent with the adipos ity data, serum leptin was 22.5% of control levels in TET-Ob at the end of STAGE 1. By the end of STAGE 2, serum leptin in the TET-Ob-OFF group was similar to controls and n early quadruple the se rum leptin in the TET-Ob-ON. Hypothalamic leptin expression was increased 8.5 fold in the TET-Ob-ON animals, and this was reversed in TET-Ob-OFF. Leptin signal transduction and changes in the expression of POMC and SOCS3 in the hypothalamus were also consistent with a

PAGE 83

66 reversible increase in central leptin activity. UCP-1 protein levels per unit BAT tripled in TET-Ob-ON animals and this increase was completely reversed in TET-Ob-OFF. Finally, serum insulin was reduced in the TET-Ob-ON but not TET-Ob-OFF group. In the first leptin gene therapy study, Unge r et al. demonstrated that an intravenous infusion of adenovirus (Ad) encoding leptin maintained hyperleptinemia for 28 days and resulted in a complete disappearance of white adipose tissue in Wistar rats (Chen et al., 1996). Pairfed animals experienced a mode st reduction in adiposity despite similar weight loss, thus revealing that leptin gene therapy triggers an unprecedented, selective catabolism of white fat (Chen et al., 1996). Given that uptake of leptin across the blood:brain barrier (BBB) app ears to be a saturable proc ess, it recently has been hypothesized that central delivery of the le ptin gene may be mo re therapeutically efficacious in models resistant to periphera l leptin (Lundberg et al., 2001). In a recent investigation, central delivery of the leptin gene using an Ad vector resulted in both anorexia and weight loss in the normally leptin-resistant obese fa/fa rat (Muzzin et al., 2000). However, leptin transgene expression appeared to decrease dramatically between day 7 and 14 post-transfection, as did physiologi cal responses. This transient transgene expression could be the biproduct of a host-i mmune reaction that appears to be mounted in response to adenoviral antigens (Yang et al., 1994). Recombinant adeno-associated virus appear s to circumvent the limitations of Ad (Conway et al., 1999). The rAAV vector lacks virally encoded genes, thus eliminating the risk of immune response to viral-specifi c antigens. Moreover, rAAV-mediated gene delivery results in unabated transgene expres sion for at least 6 months (Klein et al., 1999), whereas transgene expressi on appears to dec line within days after Ad vector

PAGE 84

67 delivery (Koyama et al., 1998; Muzzin et al ., 2000). Further, rAAV gene delivery is suitable for delivery directly into neuronal ti ssue (Peel et al., 1997). This allows for transgene expression in the CNS, thus bypassi ng potential limitations of BBB transport. For all of these reasons, rAAV app ears to be the superior vector for leptin gene delivery. In the present study, the leptin gene wa s delivered centrally by rAAV, thereby avoiding both the problem of saturable BBB transport and the probl em of unpredictable silencing of transgene with Ad. Moreover, this is the first study to incorporate regulation into leptin gene therapy—presumably a desira ble feature for any gene delivery system to be studied in humans. The importance of post-transfectional control becomes clear when one considers the potential side effects of le ptin transgene overexpres sion. For example, a single intracerebroventricular (i.c.v) injection of constituti vely active rAAV-leptin was recently shown to impair T-lymphocyte-mediated immunity in rats (Zhang, 2002). As discussed previously, leptin overexpression could also have negative effects on the reproductive system (Ahima and Flier, 2000a) and skeletal system (Ducy et al., 2000). Should such side effects appear, transgene expression could be reduced or silenced if the gene delivery system incorporates regulation. This is the principle advantage of the TET-Ob system described here. Without such re gulation, one would have to resort to complicated (and likely less effective) pharmacol ogical strategies or an tisense delivery to reduce the severity of side effects. Recently, it has been suggested that long-term over-expression of leptin can lead to leptin resistance even in the absence of obes ity (Qiu et al., 2001; Scarpace et al., 2002a). In one study, transgenic mice that over-expressed leptin displayed minimal adiposity and low body weights early in life, yet eventually stopped respondi ng to the tran sgene despite

PAGE 85

68 continuous expression (Qiu et al., 2001). Both published (Scarpace et al., 2002a; Scarpace et al., 2003) and unpublished data from our laboratory suggest that the same is true of leptin gene therapy. Namely, chronica lly elevated levels of leptin in or around the hypothalamus following rAAV-leptin gene delivery lead to leptin resistance, even in nonobese animals that are fully leptin responsiv e at the onset of gene therapy. These observations bring to light a nother potential advantage of a system that incorporates regulation. We speculate that by periodically silencing the leptin transgene, this phenomenon of acquired leptin-i nduced-leptin resistance may be circumvented. After a brief period of silencing the leptin transgene, the defective step(s) in the leptin signaling cascade may be allowed to re-sensitize prior to another period of tran sgene activation. One particularly novel finding was that leptin overexpression appears to reduce hypothalamic leptin receptor expression, and th is attenuation is reve rsed within 32 days of silencing the transgene. Expression of the long-form leptin receptor, Ob-Rb, was reduced by 30.5% in TET-Ob-ON versus TET-Ob-OFF [Figure 3-14]. This suggests that leptin is a negative regulator of leptin recep tor expression and that silencing the transgene restores receptor expression by reducing lep tin concentration in the vicinity of the hypothalamic Ob-Rb receptors. Long term peripheral hyperleptinemia following leptin infusion has previously been associated reduced Ob-Rb expression in the hypothalamus (Martin et al., 2000), but our presen t data is the first to demonstr ate that this is a specific effect of elevated leptin in the brain and, moreov er, that this effect can be reversed. It is unknown what role, if any, this leptin-induced decrease in leptin receptor expression plays in leptin resistance. However, these re sults prompted us to further study the role of hyperleptinemia in Ob-Rb expression and f unction. Experiments exploring leptin

PAGE 86

69 receptor expression and signaling capacity in hyperleptimemic and hypoleptinemic states will be discussed in Chapter 5. In the present study, hypothalamic P-ST AT3, a primary Ob-Rb second messenger, remained elevated 66 days after beginning doxy administration in the TET-Ob-ON group. These results are consistent with those of an earlier investiga tion by Dr. Scarpaces’s group using constitutively active leptin gene delivery, where P-ST AT3 was elevated to the same extent 9 and 46 days post-transf ection (Scarpace et al., 2002b). Taken together, these results suggest that there is no leptin-d ependent desensitizati on of leptin signal transduction over the time frame of these stud ies. Moreover, the elevation in STAT3 phosphorylation in TET-Ob-ON in this study persisted desp ite an increase in expression of SOCS3, a putative negative regulator of leptin signaling in the hypothalamus. No elevation in hypothalamic P-STAT 3 was detected in the TET-Ob-OFF group, confirming that the amplified leptin signal transduction in the TET-Ob-ON group was dependent on the externally regulated leptin transgene e xpression. Moreover, PO MC mRNA levels in the hypothalamus and UCP-1 in BAT mirrore d hypothalamic P-STAT3 levels, implying that there was no desensitization anywhere downstream of the leptin receptor. The physiological data confirms that the TET-Ob-ON animals remained responsive to the transgene as anorexia co ntinued through the final days of the study. One surprising finding was that NPY expres sion was not signifi cantly depressed in the TET-Ob-ON group at sacrifice. This is par ticularly perplexing in light of the increased POMC expression in TET-Ob-ON, as leptin signaling has been shown to have opposite and simultaneous effects on POMC a nd NPY expressing neurons (Elias et al., 1999; York, 1999). In previous studies, we have reliably detected reduced hypothalamic

PAGE 87

70 NPY mRNA following central leptin peptid e and leptin gene delivery (Shek and Scarpace, 2000; Scarpace, 2002). One possibility is that the TET-Ob-mediated enhanced leptin signaling did, indeed, suppress NPY expre ssion, but to an extent that was too small for us to detect. A second possibility is that sometime before day 66, compensatory pathways reversed the effects of the tran sgene on NPY expression. Our previous constitutively active leptin gene therapy tr ial was 20 days shorter, which may not have been long enough for this compensatory path way to override leptin’s effect on NPY expression (Scarpace, 2002). Sahu (2002) rece ntly demonstrated that one of the first steps in leptin-induced-leptin resistance is desensitization of NPY neurons. Thus, our NPY data may be a preliminary sign of acquired leptin resistance in the TET-Ob-ON animals. The relationship between the dese nsitization of NPY neurons and physiological leptin resistance will require fu rther investigation. However, the results of the present study suggest that the desensitization of NPY neurons may precede the loss of physiological responses to leptin. The reduction in CSF leptin concentrations in the TET-Ob-ON rats versus both controls and TET-Ob-Off animals was also unexpected. Apparently, the amount of leptin leaking (if any) from the tr ansfected hypothalamic cells was insufficient to compensate for the reduced amount of leptin coming acro ss the blood brain barrier Indeed, there was an approximate 80% reduction in serum leptin levels in the TET-Ob-ON group at sacrifice. Periphera l leptin, not hypothalam ic leptin transgene, appears to be the dominant factor in determin ing CSF leptin levels in th is study. Nevertheless, the hypothalamic leptin expression and leptin si gnal transduction data confirm that local levels of leptin were elevated.

PAGE 88

71 In summary, constitutively active leptin gene delivery by rAAV (“rAAV-leptin”) has been shown to cause dramatic reductions in adiposity in young adult animal models, and we repeat these observations here (Experiment 1). Our major objective was to incorporate regulation into this potent rAAV-leptin system (Experiment 2), and the results of this chapter demonstrate that we achieved this goal. The TET-Ob system allows for post-transfection control of le ptin transgene expression, and ergo leptin signaling and associated bioc hemical and physiological responses. Although this experiment examined only an “on” and “off” st ate, it is known that expression of genes under the control of the TET ope ron occurs in a doxy-dose de pendent manner (Huang et al., 1999). This suggests that, by adjusting the dose of doxy, this system could be used to titer leptin expression to achieve optimal levels of adiposity while minimizing side effects.

PAGE 89

72 CHAPTER 4 CENTRAL LEPTIN GENE THERAPY FA ILS TO OVERCOME THE LEPTIN RESISTANCE ASSOCIATED WITH DIET-INDUCED OBESITY Introduction When a few injections of leptin were found to rapidly reverse obesity in the ob/ob mouse, excitement soared over a potential cure for the growing obesity epidemic. Unfortunately, the ob/ob mouse did not turn out to be a model of common obesity. Typical obese humans and animals are hyperleptin emic and resistant to exogenous leptin. Clinical trials with leptin have been disappointing du e to this pheno menon of leptin resistance in the obese state and interest in pharmacological leptin to treat obesity has waned. Since several obese, leptin resist ant models resp ond better to central versus peripheral administration of recombinant leptin (Halaas et al., 1997; Niimi et al., 1999; Van Heek et al., 1997), it has been reasoned th at leptin is not reaching its hypothalamic targets. Transport of leptin across the blood:b rain barrier appears to be saturated at serum leptin levels observed in obesity (Banks et al., 1999), suggesting that deficient blood:brain barrier transport may play a role in leptin resist ance. Thus, central delivery of leptin may be effective in cases of pe ripheral leptin resistance in obese humans and animal models. A single intracerebroventricula r injection of rAAV-leptin has been shown to cause a rapid and complete disappearance of white adipose tissue in genetically normal young adult F344XBN male rats (Scarpace, 2002). In a recent report, a single injection of rAAV encoding leptin prevented diet induced obesity in young male Sprague-Dawley

PAGE 90

73 rats when administered before commencing hi gh-fat (HF) feeding (Dube et al., 2002). In the clinical treatment of obesity, however, a mo re typical objective is to reverse obesity in an already obese subject. The purpose of the present investigation was to see if a single intracerebroventricular injection of rAAV encoding leptin could reverse the obesity caused by 100 days of high fat feeding. To this end, we administ ered rAAV-leptin or control vector to high fat fed obese and diet resistant animals as we ll as lean, chow fed animals. Physiological responses to th e rAAV-leptin versus control vector were measured in all dietary groups, including a norexia, weight loss, and whole body energy expenditure via indirect calor imetry. After sacrifice, biochemical responses to rAAVleptin were evaluated, includi ng leptin signal transduction an d neuropeptide expression in the hypothalamus, and uncoupling protein co ncentration in brown adipose tissue. Methods and Materials Animals Three-month old male Fischer 344 x Brow n Norway rats were obtained form Harlan Sprague-Dawley (Indianapolis, IN). Upon arrival, rats were examined and remained quarantined for one week. Animal s were individually caged with a 12:12 hour light:dark cycle (07:00 to 19:00 hr). Animal s were cared for in accordance with the principles of the NIH Guide to the Care and Use of Experimental Animals. Experimental Design All animals were maintained on standard rat chow (diet 2018 from Harlan Teklad, Madison, WI) from weaning until two weeks after arriving in our laboratory, at which point animals were approximately 3 months old. This CHOW diet provides 3.3 kcal/g of digestible energy and 15% of ener gy as fat. At this point, 22 animals were switched to a high fat/high sucrose (HF) diet (F3282 from BioServ, Frenchtown, NJ, USA) while 10

PAGE 91

74 continued to receive chow. This HF diet pr ovides 5.3 kcal/g and 59.4% of energy as fat. Animals were maintained on these respect ive diets through the conclusion of the experiment. Over several weeks, the HF fed animals spontaneously divided into two distinct groups: those that were becoming obese on the HF diet (Diet Induced Obese, DIO) and those that were not gaining extra we ight on the HF diet (Diet Resistant, DR). After 4 weeks of HF-feeding, the top 40% of weight gainers were designated as “DIO” and the rest of the animals on the HF-fed diet were designated as “DR”. This is similar to the designation system used previously by Le vin and Keesey (1998), who defined the top 37.5 % of weight gainers on a high energy di et as DIO. DIO and DR groups were subsequently analyzed separately. After approximately 100 days of HF feeding, all animals (including CHOW-fed) were prepared for stereotaxis surgery as described below. Animals were given a single in jection into the third cerebroventricle of rAAV-encodingleptin (rAAV-leptin) or contro l vector encoding green fluores cents protein (rAAV-con). This yielded 6 groups of anim als: an rAAV-leptin and an rAAV-con subgroup in each of the three original dietary gr oups (CHOW, DIO, and DR). These 6 groups are abbreviated CHOW-con, CHOW-leptin, DIO-con, DIO-leptin, DR-con, and DR-leptin. Animals were monitored for another 30 days and then sacrificed for tissue analysis. Blood Collection Blood was harvested from all animals on day 75. Animals were placed under temporary anesthesia by ethrane inhalation. Using a sterile, razor sharp scapula, a small piece (~2 mm) of the tip of the tail was excise d. Then, using a gentle “milking” motion, 0.5 mL of blood was collected from the tail into a sterile microfuge tube. The blood was allowed to clot, and then was immediately spun at 1300 G for 10 minut es. The top serum layer was placed in a fresh tube and stored at -20 C.

PAGE 92

75 Oxygen Consumption O2 consumption was assessed in up to four rats simultaneously with an Oxyscan analyzer (OXS-4; Omnitech Electronics, Columbus, OH) as described previously (Scarpace et al., 1997). O2 consumption is used to estimate energy expenditure in a procedure known as “indirec t calorimetry.” Flow rates were 2 L/min with a 30-s sampling time at 5-min intervals. The ra ts were placed into the chamber for 120 min with the lowest 6 consecutive O2 consumption values during this period used in the calculations (basal resting VO2) Food was withdrawn 2 hours prior to commencing measurements. All measurements were made between 09:00 and 14:00 hrs. Results were expressed as O2 consumption relative to metabolic body size (ml min -1 kg2/3) or L/Kcal food intake. Construction of r AAV Vector Plasmid The vector used in this study is identical to that used in Experiment 1 of Chapter 3. pTR-Ob encodes rat leptin cDNA [a kind gift fr om Roger Unger (Chen et al., 1996)] and green fluorescent protein (GFP) reporter gene cDNA. Th e woodchuck hepatitis virus posttranscriptional regulatory element (WPR E) was placed downstream to enhance the expression of the transgenes (Loeb et al., 1999). The control vector encodes GFP driven by a CBA. Vectors contain AAV terminal repeat s at both sides of the cassette to mediate replication and packaging of the vector (Bell et al., 1999). Packaging of rAAV Vectors Vectors were packaged, purified, concentrat ed, and titered as described previously (Conway et al., 1999). The titer of rAAV-lep tin was 2.3E13 physical particles/mL. A mini-adenovirus helper plasmid (pDG) (Gri mm et al., 1998) was used to produce rAAV

PAGE 93

76 vectors with no detectable adenovirus or w ild type AAV contamination. rAAV vectors were purified using iodixanol gradient/hepar in-affinity chromatography and were more than 99% pure as judged by PAAG/silver-stained gel electro phoresis (not shown). Vector Administration Rats were anesthetized with 60 mg/kg pe ntobarbital and heads were prepared for surgery. Animals were placed into a stereotaxic frame and a small incision (1.5 cm) was made over the midline of the skull to expos e the landmarks of the cranium (Bregma and Lamda). The following coordina tes were used for injection into 3rd cerebroventricle: 1.3 mm posterior to Bregma and 9.4 mm ventral from the s kull surface on the midline (medial fissure), with the nose bar set at 3. 3 mm below the ear bars (below zero) and the canula set at 20 posterior from vertical. A small hole was drilled through the skull and a 23-gauge stainless steel guide canula was lowered to the 3rd cerebroventricle. This was followed by an injection canula attached to a 10uL syringe. We injected 2.5 uL of viral particles dissolved in Ringer’s solution at approximately 0.25 uL/minute. Animals received either the rAAV-leptin (n=5 CHOW n=5 DIO, and n=5 DR) or control virus encoding GFP (n=5 CHOW, n=4 DIO, and n=8 DR). Tissue Harvesting Rats were anesthetized with 85 mg/kg pentobarbital and sacrificed by cervical dislocation. Blood was collected by cardiac puncture and serum was harvested by a 10 minute centrifugation (1300 G) in serum separa tor tubes. The circulatory system was perfused with 20 mL of cold sa line. Inguinal, perirenal, a nd retroperitoneal white adipose tissue, brown adipose tissue, and hypothalami were excised, wei ghed, and immediately frozen in liquid nitrogen. The hypothalamu s was removed by making an incision medial

PAGE 94

77 to piriform lobes, caudal to the optic chiasm, and anterior to the cerebral crus to a depth of 2-3 mm. Tissues were stor ed at –80 C until analysis. Serum Leptin and FFA Serum leptin was measured using a rat radioimmunoassay kit (Linco Research, St. Charles, MO). Serum free fatty acids were measured using the NEFA C colorimetric kit from WAKO Chemicals GmbH (Neuss Germany) Day 75 data was derived from serum harvested from tail bleeding. Endpoint meas urements were made on serum harvested from a cardiac puncture at sacrifice. Reverse Transcriptase-Polymera se Chain Reaction (RT-PCR) Leptin transgene expression and long-form leptin receptor (Ob-Rb) expression were evaluated by using relative quantita tive RT-PCR through the use of QuantumRNA 18s Internal Standards kit (Ambion, Austin, Tx). Total RNA (3 g) was treated with RNase-free DNase using a DNA-free kit (Ambio n), and first-strand cDNA synthesis was generated from 1 g RNA in a 20 L volume using random primers (Gibco BRL) containing 200 units of M-MLV reverse transc riptase (Gibco BRL). Relative quantitative PCR for rAAV-leptin expression was perfor med by multiplexing rAAV-leptin specific primers (sense: 5’GGCTCTGACTGACCGCGTTA; antisense: 5’ CTGCCAGGGTCTGGTCCATC) and 18s primer s and coamplifing for 28 cycles, the midpoint of the linear range fo r signal intensity versus num ber of cycles. The optimum ratio of 18s primer to competimer was 1:9. PCR was performed at 94C denaturation for 60 sec, 59C annealing temperature for 45 s ec, and 72C elongation temperature for 120 sec. Similarly, the number of cycles found to be at the mid point of the linear range was 26 cycles for ObRb. The sequence for Ob-Rb primers were sense: GGGAACCTGTGAGGATGAGTG; antisense : TAGCCCCTTGCTCTTCATCAG. The

PAGE 95

78 PCR product was electrophoresed on a 5% acry lamide TBE gel and stained with SYBR green (Molecular Probes, Eugene, OR). Gels were scanned usi ng a STORM fluorescent scanner and digitized data were analyzed using Imagequant software (Molecular Dynamics, Sunnyvale, CA). STAT3/Phospo-STAT3 Assay These methods were described in detail previously (Scarp ace et al., 2000b). Briefly, hypothalamus was sonicated in 10 mM Tris-HCL, pH 6.8, 2% SDS, and 0.08 ug/mL okadaic acid plus prot ease inhibitors (PMFS, benzam idine, and leupeptin) [an aliquot of this sonicate was frozen for RNA analysis]. Sonicate was diluted and quantified for protein using a detergent compa tible Bradford Assay. Samples were boiled and separated on an SDS-PAGE gel and electr otransferred to nitrocellulose membrane. Immunoreactivity was assessed with an an tibody specific to phosphorylated-STAT3 (antibody kit from New England Biolabs, Beverly MA). Immunoreactivity was visualized by chemiluminescents detection (Amersham Life Sciences, Piscataway, NJ) and quantified by video densitometry (BioRa d, Hercules, CA). Following P-STAT3 quantification, membranes were stripped of antibody using Immunopur e reagent (Pierce, Rockford, IL) and immunoreactivity was re-a ssessed using a total STAT-3 antibody. Probes POMC mRNA is detected using a 24-mer antisense oligonucleotide probe (5’CYYGCCCACCGGCTTGCCCCAGCG-3’). The oligonucleotide probe was end labeled by terminal deoxynucleotidyl transferase (Promega). The AgRP cDNA probe was provided by Dr. Michael Schwartz (University of Washington). The rat pre pro NPY cDNA was provided by Janet All en (University of Glasgow UK). SOCS3 cDNA was a gift from Christian Bjorbaek (Harvard University). The cDNA probes were labeled using

PAGE 96

79 a random primer kit (Prime-a-Gene, Promega, Madison, WI). Probes were purified with Nick columns (Pharmacia) and, except for o ligonucleotide probes, were heat-denatured for 2 minutes. All probes have been verified to hybridize to the corresponding specific mRNAs by Northern Analysis prior to use in Dot Blot assay (below). mRNA Levels (Dot Blot Analysis) Tissue was sonicated in guanidine buffe r, phenol extracted, and isopropanol precipitated using a modifi cation of the method of Chomczynski (Chomczynski and Sacchi, 1987). Isolated RNA was quantif ied by spectrophotometry and integrity was verified using 1% agarose gels stained with ethidium bromide. For dot blot analysis, multiple concentrations of RNA were imm obilized on nylon membranes using a dot blot apparatus (BioRad, Richmond, CA). Me mbranes were baked in a UV crosslinking apparatus. Membranes were then prehybrid ized in 10 mL Quickhyb (Stratagene, LaJolla, CA) for 30 minutes followed by hybridization in the presence of a labeled probe and 100 g salmon sperm DNA. After hybridization for 2 hours at 65C, the membranes were washed and exposed to a phosphor imaging screen for 24-72 hours (depending on anticipated strength of signal). The screen was then scanned using a Phosphor Imager (Molecular Dynamic, Sunnyvale, CA) and analyzed by Image Quant Software (Molecular Dynamics). Statistical Analysis All data are expressed as mean standard error of m ean. Body mass and food intake during the HF-feeding period were compared by 2-WAY ANOVA, with dietary group and time serving as factors. Total we ight gain through day 75 was compared by 1WAY ANOVA with Tukey’s post-hoc. Comp arisons in cumulative caloric intake

PAGE 97

80 (through day 5) and average daily caloric in take (day 6-100) were made by 1-way ANOVA. A 1-way ANOV A was used to compare serum leptin, serum free fatty acids, and oxygen consumption in the three dietary groups (CHOW, DIO, and DR) prior to vector delivery. When 1-way ANOVA wa s significant, Tukey post-hoc was used to evaluate pairwise comparisons. After vector delivery, comparisons were made by 2-Way ANOVA on the now 6 groups (each dietary gr oup was divided into rAAV-leptin and control subgroups) with dietar y group and vector as fact ors. When only 2-Way ANOVA main effects were significant, relevant pa irwise comparisons were made using the Bonferroni Multiple Comparison method with th e error rate corrected for the number of contrasts. When there was an interaction, factors were separated and a further 1-Way ANOVA was applied with a Bonf erroni Multiple Comparison post-hoc. When separation of factors resulted in onl y two population means to co mpare, the 1-WAY ANOVA was replaced with student’s t-test. -level was set at 0.05 for a ll analyses. GraphPad Prism software (San Diego, CA) was used for all st atistical analysis and graphing. Post-hoc analysis of 2-Way ANOVAS wa s done using GraphPad QuickCalc (GraphPad.com). Results Part I: High-Fat Feeding Food consumption and body mass Male F344XBN rats, age 3 months, were ei ther maintained on standard chow diet (CHOW) or a high-fat high sucros e (HF) diet as described in the Methods section. The HF-fed animals spontaneously divided into two distinct groups: those that were becoming obese on the HF diet (Diet Induced Obese, DIO) and those that were not gaining extra weight on the HF diet (Diet Resistant, DR). DIO animals gained mass at a greater rate than DR and CHOW, and this trend became significant by day 25 (Figure 4-

PAGE 98

81 1). By day 75, DIO animals had gained ove r 25% more mass than both CHOW and DR [p<0.001] (Figure 4-1). During the firs t week of high fat feeding, acute hyperphagia was observed in all HF-fed animals (both DI O and DR). Cumulative caloric intake was nearly 30% greater in HFfed animals than CHOW-fed animals during the first 5 days of HF-feeding (p<0.001), but this acute hyperphagi a attenuated by day 7 (Figure 4-2). After this acute phase, caloric intake was similar in DIO and CH OW while DR animals consumed significantly less calories than bot h DIO and CHOW during this phase of the experiment (p<0.01, Figure 4-2). Oxygen consumption Oxygen consumption (VO2) was first measured 30 days after beginning HFfeeding. At this point, oxygen consumption (mL/min/kg2/3) was significantly reduced in DIO compared to both CHOW and DR (p <0.01), suggesting that reduced energy expenditure in DIO contributes to their accelerated rate of weight gain during this phase. By day 70, there was no longer a difference in VO2 in the three groups (Figure 4-3). Serum leptin and free fatty acids At day 75, serum leptin was significantly greater in DIO compared to DR (p<0.01, Table 4-1). CHOW animals had lower serum leptin than both DIO and DR (p<0.001, p<0.01, respectively). Serum FFA did not diffe r across groups at this point in the study (Table 4-1).

PAGE 99

82 0 25 50 75 100 250 300 350 400 450 DIO DR CHOW DayBody Mass (grams) Figure 4-1: Body mass during hi gh fat feeding (pre-vector delivery). Values represent means SEM. By 2-Way ANOVA, significance was found for effects of dietary group (F=46.38, p<0.0001), time (F=263.00, p<0.0001), and the interaction between group and time (F=1.71, p<0.05). 0 25 50 75 100 50 60 70 80 90 100 DIO CHOW DR DayKcal/24hr Figure 4-2: Caloric intake duri ng high fat feeding (pre-vector de livery). Values represent means SEM. By 2-Way ANOVA, significance was found for effects of dietary group (F=31.59, p<0.0001), time (F=12.00, p<0.0001), and the interaction between group and time (F=2 .62, p<0.0001). Note the biphasic pattern of caloric intake in all HF-f ed (DIO and DR) animals demonstrated acute hyperphagia after being sw itched to the high fat di et. After this transient hyperphagic response to the diet subsided (day 7), average caloric intake did not differ between DIO and CHOW for th e rest of the study. Mean daily caloric intake was significantly lower in DR compared to both DIO and CHOW after day 7 (p<0.01).

PAGE 100

83 20 30 40 50 60 70 80 70 80 90 100 110 120 CHOW DIO DR **Day% of CHOW-fed Control Figure 4-3: Oxygen consumption on day 30 and on day 70 after commencing HF-feeding (pre-vector delivery). Values represen t mean percent of control SEM. p<0.01 for difference between DIO and CHOW on day 30 and difference between DIO and DR on day 30. Table 4-1: Serum leptin and free fatty acids after 75 days of high-fat feeding (DIO and DR). CHOW DIO DR Serum Leptin (ng/mL) 4.900.52 13.291.43a 9.050.58b Serum FFA (meq/L) 0.550.059 0.440.045 0.640.055 Data represent meanSEM. p values represent resu lts of post-hoc analysis following 1-WAY ANOVA. a p<0.001 versus CHOW, p<0.01 versus DR b p<0.01 versus CHOW and DIO Part II: Post rAAV-Leptin Delivery Food consumption and body mass After approximately 100 days of HF-feeding in DIO and DR, animals in all dietary groups (i.e., CHOW, DIO, DR) were given a single i.c.v. injection containing 5.75E10 physical particles of rAAV encoding leptin (rA AV-leptin) or control virus (rAAV-con). CHOW-leptin had a robust response to rAAV-leptin, losing 6.6% of their body mass while CHOW-con animals increa sed their body mass by 3.2% dur ing the same 29 days of

PAGE 101

84 observation (p<0.0001) (Figure 4-4). CHOW-leptin also ha d a significant (p<0.001) anorectic response to rAAV-leptin, consumi ng an average of 20.6% less calories than controls from day 7 through 29 post vector deliv ery (Figure 4-5). In contrast, DIO were completely unresponsive to the weight reduc ing (Figure 4-4) and anorectic effects (Figure 4-5) of rAAV-leptin. DR rats responded to rAAV-lep tin, but in a more variable fashion than CHOW. DR-leptin animals lo st an average of 8.4% of their body mass while DR-con increased their BM by 4.7% dur ing 29 days post-vector delivery (p<0.01, Figure 4-4). DR-leptin consumed an averag e of 17% less food than DR-con days 7-29 post gene delievery (p<0.05), but unlike in CHOW, this anorectic response rapidly attenuated and was no longer si gnificant by day 14 post-vect or delivery (Figure 4-5). Oxygen consumption rAAV-leptin did not significantly incr ease oxygen consumption 7 days following vector delivery in any of the three dietary groups (Table 4-2). This was not a surprise as it has been previously demonstrated that leptin prevents the decrease in energy expenditure normally observed following reduced caloric intake, yet does not necessarily increase energy expenditure above that of untreated, ad libitum fed controls (Scarpace et al., 1997; Scarpace, 2002). At this point in th e present study (day 7), significant anorexia was recorded in CHOW-leptin and DR-leptin animals but not in DIO-leptin animals (compared to animals in the respective dietar y group given control v ector) (Table 4-2). For this reason, oxygen consumption was e xpressed relative to caloric intake ( L/kCal). By this method, rAAV-leptin treatment was found to increase energy expenditure per Kcal consumed in CHOW (CHOW-leptin vs CHOW-con, p<0.05), but not DIO (Table 4-2).

PAGE 102

85 Adiposity Animals were sacrificed 30 days after vect or delivery. At this point, visceral adiposity (sum of retroperit oneal and perirenal white adipose tissue) of CHOW-leptin was 19.5% of that in CHOW-con (p<0.0001, Figure 4-6). Although DIO animals had nearly 3-fold greater visceral adiposity co mpared to CHOW-con, there was no effect of rAAV-leptin on adiposity in DIO (Figure 4-6). This confirms that DIO animals were completely unresponsive to a dose of rAAVleptin that causes a near complete disappearance of white fat in age-matched chow-fed animals. DR-con had more than double the visceral adiposity of CHOW-con de spite similar body masse s, suggesting that DR animals have reduced lean body mass coupled with increased fat mass (Figure 4-6). Visceral adiposity was reduced by 57% in DR-leptin vs. DR-con (p<0.01), suggesting these animals were indeed leptin responsive de spite high initial adiposity. Unexpectedly, adiposities were not statisti cally different between DIO and DR control rats despite significant differences in tota l body mass and rates of weight gain and despite different leptin responsiveness. Serum leptin and free fatty acids As expected, serum leptin data correlated with the adiposity data (Table 4-3). rAAV-leptin caused an 84% reduction in seru m leptin in CHOW (p<0.0001) and a 67% reduction in DR (p<0.01), yet had no effect in DIO (Table 4-3). DIO-con animals had 3fold greater serum leptin (p<0.0001) and DR-con had 2.5-fold greater serum leptin (p<0.001) than CHOW-con (Table 4-3). At sacrifice, there was a significant dietary group main effect on serum FFA (2Way ANOVA p<0.05, Table 4-3). Serum FFA tended to be higher in DIO compared to CHOW and DR, but this was not sign ificant by post-hoc analysis.

PAGE 103

86 Leptin transgene expressi on in the hypothalamus Primers were designed to specifically detect leptin expressed by rAAV-leptin transgene via RT-PCR. mRNA for the leptin transgene was detected in the hypothalami of all animals administered rAAV-leptin, but not in the animals administered rAAV-con (Figure 4-7). No differences in transgene expression were found across dietary groups. Leptin receptor expression and signal transduction in the hypothalamus Leptin receptor (Ob-Rb) expression was si gnificantly reduced in both DIO-con and DR-con as compared to CHOW-con (p<0.05) (Figure 4-8). r AAV-leptin treatment tended to reduce Ob-Rb expression in CHOW anim als, but this failed to reach statistical significance (p=0.07). rAAV-leptin did not affect the already suppressed Ob-Rb expression in DIO and DR (Figure 4-8). Basal P-STAT3 was significantly elevated in DIO-con as compared to CHOW-con (p<0.01) and DR-con (p<0.05), and P-STAT 3 was greater in DR-con than CHOW-con (p<0.01). rAAV-leptin treatment cau sed a significant increase in STAT3 phosphorylation in CHOW (p<0.01) and DR an imals (p<0.05), but not in DIO animals (Figure 4-9). Total STAT levels were not affected by diet or r AAV-leptin in any group (data not shown). Downstream neuropeptide regulation in the hypothalamus Leptin normally causes an increase in Pro-opiomelanocortin (POMC, the precursor to -melanocyte stimulating hormone [-MSH]), and this is on e downstream component of the leptin signaling cascade that ultimately results in anorexia and increased energy expenditure (Elias et al., 1999). In the pr esent study, rAAV-leptin treatment caused a significant increase in POMC expression in CHOW animals (p<0.01) and in DR animals (p<0.05), but not in DIO (Table 4-4). There was no effect of diet on POMC expression.

PAGE 104

87 NPY and AGRP are orexigenic signals whose expression may be down-regulated by leptin (Elias et al., 1999; Spiegelman and Flier, 2001). rAAV-leptin treatment significantly reduced NPY expression in CHOW animals (p<0.05), but not in DIO or DR (Table 4-4). AGRP expression tended to be suppressed by rAAVleptin treatment in both CHOW and DR, but this barely failed to reach statistical si gnificance in both cases (p= 0.064 and p=0.061, respectively) (Table 44). No effect of rAAV-leptin on AGRP was observed in DIO. SOCS3 expression tended to be elevated by rAAV-leptin treatment in CHOW and DR, but this failed to reach statistical significance (Table 4-4). Expression of AGRP, NPY, and SOCS3 did not differ across the three dietary groups (Table 4-4). Hypothalamic expression of PIAS-3, a putative negative regulator of PSTAT3 DNA binding, was also measured. We di d not detect an effect of diet or rAAVleptin on PIAS-3 expression (data not shown). UCP-1 in brown adipose tissue The increase in energy expenditure in leptin treated compared to pair-fed animals is mediated in part by uncoupling protei n-1 (UCP-1) (Scarpace and Matheny, 1998; Scarpace, 2002). UCP-1 uncouples mitochondri al respiration from ATP production, thus decreasing the efficiency of metabolism and increasing heat production. In the present study, UCP-1 protein concentration in th e interscapular brow n adipose tissue was significantly increased by HF feeding (Fi gure 4-10). HF feeding increased UCP-1 protein concentration by more than 2-fold in both DIO and DR (F igure 4-10). rAAVleptin treatment caused a significant increase in UCP-1 concentration in CHOW (p<0.05), but had no effect in DIO (Figure 4-10). Although UCP-1 was already elevated in DR-con to the same degree as DIO-con, rAAV-leptin caused a significant 2-fold further increase in UCP-1 concentration in BAT in DR-leptin (p<0.05, Figure 4-10).

PAGE 105

88 High fat feeding causes a significant infilt ration of BAT with white adipocytes. Thus, we also expressed UCP-1 levels per to tal interscapular BAT pa d. When expressed this way, the difference in the UCP-1 re sponse to rAAV-leptin in DIO and DR was eliminated (Figure 4-11). Thus, the apparent robust UCP-1 effect of rAAV-leptin in DR was actually a result of the maintenance of to tal UCP-1 protein coupled with a significant decrease in BAT pad mass, most likely due to enhanced lipolysis and the loss of white fat. rAAV-leptin enhanced UCP-1 per total IBAT pad in CH OW (Figure 4-11). Discussion The major objective of the present study was to compare the effects of leptin gene therapy in CHOW-fed and DIO animals with hopes of gaining new insight into the mechanisms of obesity-associated leptin resistance. To our knowledge, this is the first study to measure responses to leptin gene delivery after a period of chronic high-fat feeding. During the initial 100 days of hi gh fat feeding, 40% of HF-fed animals became obese (DIO), gaining significantly more body mass and adiposity than CHOW-fed controls. The remaining HF-fed animals were “diet resistant,” gain ing body weight at a similar rate to CHOW-fed cont rols. Barry Levin’s group ha s previously described this phenomenon of DIO and DR in detail using Sprague-Dawley rats (Levin et al., 1987; Levin et al., 1997; Levin and Keesey, 1998). In our F344xBN model, we recorded increased adiposity in DR animals with resp ect to CHOW-fed controls despite similar body masses. Thus, DR rats appear to have an unfavorable change in body composition as a result of the HF-feeding. The appa rent loss in lean body mass in DR may be

PAGE 106

89 -15 -10 -5 0 5 10 rAAV-Con rAAV-Leptin *** **CHOWDIO DR% Change in BM Figure 4-4: Changes in body mass during 29 days post-vector de livery. Values represent mean percent change of starting body mass SEM. By 2-Way ANOVA, significance was found for effects of diet ary group (F=3.74, p<0.05), vector (F=19.86, p<0.001), and the interacti on between group and vector (F=4.22, p<0.05). By post hoc analysis: p<0. 0001 for effect of rAAV-leptin on body mass in CHOW; p<0.01 for effect of rAAV-leptin in DR. (a) CHOW 0 4 8 12 16 20 24 28 0 25 50 75 100 rAAV-Con rAAV-Leptin * * * * *DAYKCal/day (b) DIO 0 4 8 12 16 20 24 28 0 25 50 75 100 DAYKCal/day (c) DR 0 4 8 12 16 20 24 28 0 25 50 75 100* *DAYKCal/day Figure 4-5: Caloric intake duri ng 29 days post-vector delivery. Values represent mean SEM. (a) CHOW: By 2-Way ANOVA, si gnificance was found for effects of vector (F=106.51, p<0.0001), time (F=7 .45, p<0.0001), and the interaction between vector and time (F=4.03, p< 0.0001). By post hoc analysis: p<0.05 for effect of rAAV-leptin on food intake on days 7-29 in CHOW. (b) DIO: Only time main effect was significant (F=23.16, p<0.0001). rAAV-leptin had no effect. (c) DR: Both vector (F=1 7.87, p<0.0001) and time (F=8.08, p<0.0001) main effects were significant. By post hoc analysis, p<0.05 for effect of rAAV-leptin on food intake days 7-12 in DR.

PAGE 107

90 Table 4-2: Oxygen consump tion and 24 hour caloric inta ke on day 7 following rAAVleptin or control vector delivery CHOW-Con CHOW-Lep DIO-Con DIO-Lep DR-Con DR-Lep VO2 (ml/kg2/3/m) 11.290.44 12.380.82 11.440.22 11.910.31 12.540.43 12.070.19 Food Intake (Kcal) 63.462.07 50.992.13a 57.643.97 55.701.86 58.381.01 38.9612.38b VO2 per Kcal (L/Kcal consumed)* 99.853.66 135.59.21c 117.68.69 124.95.20 Data represent mean SEM. By 2-Way ANOVA analysis of day 7 food intake, significance was found for vector (rAAV-leptin) ma in effect (F=6.25, p<0.05). By 2-Way ANOVA analysis of VO2/Kcal, significance was found for vector main effect (F=9.54, p<0.01). p values below are results of post-hoc analysis of rAAV-leptin effect in each dietary group following 2-way ANOVA with dietary group and vector as factors. *Due to highly variable acute anorectic response to rAAV-leptin in DR group, VO2/Kcal was not evaluated for these animals. a p<0.01 versus CHOW-Con b p<0.05 versus DR-Con c p<0.05 versus CHOW-Con CHOWDIODR 0 5 10 15 rAAV-Con rAAV-Leptin *** **† †Visceral Adiposity [RTWAT + PWAT] (g) Figure 4-6: Visceral adiposity 30 days post-vector delivery. Values represent mean SEM. By 2-Way ANOVA, significan ce was found for effects of vector (F=20.63, p<0.0001), dietary group (F= 53.44, p<0.0001), and the interaction between group and vector (F=4.03, p< 0.0001). p<0.0001 for effect of rAAVleptin on adiposity in CHOW; p<0.05 for effect of rAAV-leptin on adiposity in DR. (†)p<0.0001 for difference in adi posity between CHOW-con and both DIO-con and DR-con.

PAGE 108

91 A CHOWDIODR 0.00 0.25 0.50 0.75 1.00 rAAV-Leptin rAAV-Con Arbitrary Units Relative to 18S rRNA B Figure 4-7: (A) Leptin transgene expressi on in the hypothalamus 30 days post-vector delivery. Primers were designed to sp ecifically detect leptin expressed by rAAV-leptin transgene via RT-PCR. mRNA for the leptin transgene was detected in the hypothalami of all anim als administered rAAV-leptin, but not in the animals administered rAAV-con. No differences in transgene expression were found across dietary groups (B) Representative images of PCR product electrophoresed on a 5% acrylamide TBE gel, stained with SYBR green, and scanned using a STORM fluorescent scanner. Top gel is CHOW, middle is DIO, and bottom is DR. On all gels, lanes 1-4 were administered rAAV-con and lanes 5-9 we re administered rAAV-leptin. Top band (where visible) repres ents leptin transgene; bottom band represents 18S rRNA control amplicon. 123456789 CHOW DIO DR Leptin Transgene 18S rRNA Leptin Transgene 18S rRNA Leptin Transgene 18S rRNA

PAGE 109

92 Table 4-3: Serum leptin and FFA at sacrifice. CHOW-Con CHOW-Lep DIO-Con DIO-Lep DR-Con DR-Lep Serum Leptin (ng/mL) 2.330.148 0.370.114 a 7.340.497 a 6.560.643 5.760.945b 1.930.830c Serum FFA (meq/L) 0.590.015 0.490.033 0.640.037 0.790.079 0.680.066 0.590.051 Data represent meanSEM. By 2-Way ANOVA analysis of serum leptin, significance was found for vector (rAAV-leptin) main effect (F=12.64, p<0.01) and dietary group main effect (F=25.36, p<0.0001). By 2Way ANOVA analysis of FFA, significance was found only for dietary group main effect (F=4.40, p<0.05). p values below are results of post-hoc analysis following 2-way ANOVA with dietary group and vector as factors. a p<0.0001 versus CHOW-Con b p<0.001 versus CHOW-Con c p<0.01 versus DR-Con CHOWDIODR 0.0 0.5 1.0 1.5 rAAV-Con rAAV-Leptin † †Arbitrary Units Relative to 18S rRNA Figure 4-8: Hypothalamic lep tin receptor expression 30 da ys post-vector delivery. Values represent mean SEM. By 2-Way ANOVA, significance was found only for dietary group main effect (F=6.55, p<0.01). By post hoc analysis, (†) p<0.05 for difference between CHOW-c on and both DIO-con and DR-con.

PAGE 110

93 CHOWDIODR 0.0 0.5 1.0 1.5 rAAV-Con rAAV-Leptin **† ‡ P-STAT3 (Arbitrary Units) Figure 4-9: STAT3 phos phorylation in the hypothalamus at sacrifice. Values represent mean percent of control SEM. Arb itrary units are per unit hypothalamic protein. By 2-Way ANOVA, signifi cance was found for dietary group main effect (F=10.22, p<0.001) and vector (rAAV-leptin) main effect (F=15.19, p<0.001). The interaction between main effects closely approached significance (F=3.10, p=0.06). The eff ect of rAAV-leptin in each dietary group was then analyzed separately, with the Bonferr oni correction for multiple comparisons applied to the erro r rate. p<0.01 for effect of rAAVleptin on STAT3 phosphorylation in CHOW p<0.05 for effect in DR. By post-hoc analysis of di etary group effects, (‡) p<0.01 for difference between CHOW-con and DIO-con; (†) p<0.05 for difference between CHOW-con and DR-con.

PAGE 111

94 Table 4-4: Hypothalamic POMC NPY, AGRP and SOCS3 expression at sacrifice. CHOW-Con CHOW-Lep DIO-Con DIO-Lep DR-Con DR-Lep POMC 1005.5 1234.2 a 11913.2 98.02.2 1109.1 14314.3b NPY 10010.1 74.54.4c 74.59.9 95.98.0 85.44.2 91.44.4 AGRP 10017 63.912 95.321 85.419 12613 96.19.7 SOCS3 1002.5 1051.8 1082.5 96.86.6 93.93.6 112.77.9 Food was withdrawn a minimum of 2 hours prior to collecting tissue samples. Samples were collected between 10:00 and 14:00. Data represent the mean SE M. All mRNA data is expressed in arbitrary units and was standardized so that the mean value in C HOW-Con animals was set at 100. POMC, NPY, and SOCS3 were measured by RNA dot blot. AGRP was measured by relative quantitative PCR relative to 18S rRNA. By 2-Way ANOVA analysis of POMC, significance was found only for interaction between dietary group and vector (F=4.61, p<0.05). Similarly, by 2-Way ANOVA analysis of NPY, significance was found only for interaction between dietary group an d vector (F=6.18, p<0.01). Therefore, we evaluated the effect of rAAV-leptin on NP Y and POMC independently in each dietary group. 2-Way ANOVA analysis of AGRP and SOCS3 did not reveal any sign ificant main effects or interactions, although the vector main effect approach ed significance for AGRP (F=3.88, p=.059). p values below are results of posthoc analysis following 2-way ANOVA with dietary group and vector as factors. a p<0.01 versus CHOW-Con b p<0.05 versus DR-Con c p<0.05 versus CHOW-Con CHOWDIODR 0.0 2.5 5.0 7.5 10.0 rAAV-Con rAAV-Leptin *† †Arbitrary Units/mg Tissue Figure 4-10: UCP-1 concentration in BAT at sacrifice. Values represent mean percent of control SEM. By 2-Way ANOVA, si gnificance was found for main effects of vector (F=10.46, p<0.01) and dietary group (F=5.91, p<0.01). By post-hoc analysis, p<0.05 for effect of leptin on UCP-1 in both CHOW and DR. (†) p<0.05 for difference between CH OW-con and DIO-con; p<0.05 for difference between CHOW-con and DR-con.

PAGE 112

95 CHOWDIODR 0 1 2 3 rAAV-Con rAAV-Leptin † †*Arbitrary Units/Total IBAT Pad Figure 4-11: UCP-1 per total in terscapular brown adipose tis sue (IBAT) at sacrifice. Values represent mean percent of control SEM. By 2-Way ANOVA, significance was found only for main effect of dietary group (F=6.99, p<0.01). By post-hoc analysis (pairwise comp arisons with Bonferroni correction applied to -level), (†) p<0.05 for difference between CHOW-Con and both DIO-Con and DR-Con; p<0.05 for effect of rAAV-leptin in CHOW only. related to their reduced caloric intake a nd perhaps sub-optimal protein intake, or a product of metabolic and endocrine changes on the fat and sucrose rich diet. For example, increased cortisol and insulin coupled with suppr essed growth hormone and sex steroids in DR could promote increased vi sceral adiposity and decreased lean mass (Bjorntorp, 1995). Regardless of the mechanism, DR animals evidently do not completely escape the negative effects of ch ronic high fat/high sucrose feeding despite their resistance to excess weight gain. It was recently reported by Dube et al. (2002) that an i.c.v. injection of rAAVleptin can prevent diet induced obesity for 9 weeks when administered before the start of HF-feeding. The authors made the following conclusions:

PAGE 113

96 . insufficiency of central leptin, ra ther than impaired leptin receptor or postreceptor signal tran sduction, contributes to the loss of weight regulation Thus, one can assume that by experimentally increasing leptin expression in the hypothalamus, it is possible to reinstate we ight control for extended periods, even when rats consume HFD [High Fat Diet ]” (Dube et al., 2002, page 1734). Our present data is inconsistent with these conclusions. In the Dube study discussed above (Dube et al., 2002), animals were not leptin resistant at the time that rAAV-leptin was administered. Theref ore, the ability of rAAV-leptin to reinstate normal body weight and adiposity regulation co uld not be properly evaluated. The present study demonstrates th at if animals are allowed to become obese first, rAAVleptin is ineffective. Since central overexpr ession of leptin did not overcome this leptin resistance, it is clear that cen tral insufficiency of leptin is not the only factor in DIOrelated leptin resistance. Rather, there is a CNS component to leptin resistance. Phosphorylated STAT3 is a second messenge r for the long form leptin receptor, which is widely expressed in the hypotha lamus (Hakansson and Meister, 1998). rAAVleptin caused a two-fold increase in hypotha lamic P-STAT3 in CHOW, yet had no effect in DIO. However, basal levels of STAT3 activation in the hyperleptinemic obese state appear equivalent to the activ ation achieved as a result of rAAV-leptin transgene in the highly responsive CHOW animals. Notably, there was no effect of obesity (DIO vs. CHOW) on the expression of any of the downstream ne uropeptides (NPY, POMC, AGRP) that are normally regulated by lept in. Thus, DIO animals are unresponsive to their elevated endogenous leptin despite persistently elevated leptin signal transduction. This suggests that the signaling defect in DIO animals lies, at least in part, downstream of leptin receptor signal transduction. Hypothalamic expression of anorectic and orexigenic signals were appropriately increased and decreased, respectively, by rAAVleptin in CHOW animals but not DIO.

PAGE 114

97 Expression of POMC, a pro-pept ide for the anorexic signal MSH, was increased in CHOW but not in DIO in response to rAAVleptin. Similarly, rAAV-leptin reduced hypothalamic expression of the orexigenic signal NPY in CHOW but not DIO. Endorgan responses to rAAV-leptin mirrored hypot halamic signaling responses. Namely, rAAV-leptin increased UCP-1 concentration in BAT 3-fold in CHOW but had no effect in DIO. We conclude th at DIO animals were completely unresponsive to a dose of rAAV-leptin that caused an 80% reduction in visceral adiposity in CHOW-fed controls over 30 days. In contrast to both CHOW and DIO, the DR animals were partially responsive to rAAV-leptin, experiencing a 57% reduction in adiposity an d significant anorexia and weight loss. Since the DR controls had 2-fo ld greater adiposity than CHOW-controls, the magnitude of visceral fat loss with rAAV-leptin was actually slightly greater in the DR group (Figure 4-6). DR animals also had a reduction in food intake and significant weight loss in response to rAAV-leptin. Ba sal STAT3 phosphorylation was increased in DR with respect to CHOW, yet still appeared less than DI O (Figure 4-9). rAAV-leptin resulted in a significant increase in P-STAT 3 in DR, suggesting that an induction in PSTAT3 is a predictor of physiological re sponsiveness to exogenous leptin. The magnitude of P-STAT3 induction was certainly less in DR than in CHOW, consistent with the overall reduced rAAVleptin responsiveness in DR with respect to CHOW. POMC expression was enhanced in response to rAAV-leptin in DR, but we did not observe a significant decrease in NPY in these animals. This latter finding my not be surprising in light of the food intake data. While rAAV-leptin initially caused anorexia in DR, this response rapidly attenuated and wa s no longer significant by day 14 post-vector

PAGE 115

98 delivery. Such an attenuation was not observed in CHOW animals, and only CHOW animals had reduced NPY expression in re sponse to rAAV-leptin on day 30 post-vector delivery. Thus, we may conclude that DR an imals are prone to a faster onset of leptin resistance compared to CHOW animals. The reason for this rapid onset of leptin resistance may be a function of greater star ting adiposity, serum leptin, or both. This warrants further investigation. In addition to potential impairments downstr eam of leptin signal transduction, the reduced leptin receptor (Ob-Rb) expression in the hypothalamus of all HF-fat fed animals with respect to CHOW-fed controls may be another contributor to the leptin resistance/reduced responsiveness in these anim als. At sacrifice, both DIO and DR animals were hyperleptinemic and had signi ficantly reduced Ob-Rb expression with respect to CHOW-fed controls. rAAV-lep tin tended to suppress Ob-Rb expression in CHOW-fed rats, but this effect did not reac h statistical significance. The sum of these findings suggests that leptin, either in the form of obesity-related chronic hyperleptinemia or centrally overexpressed transgene, may be a negative regulator of leptin receptor expression. Consistent with this are the re sults of Chapter 3, where we demonstrated a significant increase in Ob-Rb expression after silencing the leptin transgene. In the present study, we believe that central overexpression of lep tin in CHOW-fed animals via rAAV-leptin caused elevated leptin in the vi cinity of the hypothalamic receptors and this was sufficient to cause the observed trend of leptin receptor expre ssion down regulation. However, we speculate with caution give n the statistic al non-significance of rAAVleptin’s effect in CHOW. Nevertheless, the effect of highfat feeding on Ob-Rb expression was significant, in cluding a 50% reduction in hypothalamic Ob-Rb mRNA in

PAGE 116

99 DIO (Figure 4-8). This receptor down-regulation may limit the ceiling of Ob-Rbmediated STAT3 phosphorylation in response to exogenous leptin, thereby preventing an increase in P-STAT3 over the already elevated basal levels in the DIO animals. Indeed, P-STAT3 induction by rAAV-leptin was absent in DIO and reduced in magnitude in DR. Future experiments are required to be tter understand the relationship between hyperleptinemia, leptin receptor expression, receptor signaling capacity, and leptin resistance. At first glance, the effect of rAAV-lep tin on BAT UCP-1 protein concentration in DR rats was surprising. Similar to DIO animals, basal BAT UCP-1 concentration was increased more than 2-fold in DR. Unlike in DIO animals, however, rAAV-leptin caused a further 2-fold increase in BAT UCP-1 concentr ation in DR (Figure 4-10). This implies that a UCP-1-inducing signal is released in response to rAAV-leptin in DR but not in DIO. Such a signal would likely be enhanced sympathetic nerve activity since this is known to be the primary inducer of UCP1 expression (Scarpace and Matheny, 1998). Consistent with this, DIO animals have been shown to have impaired induction of lumbar sympathetic nerve activity in re sponse to leptin (Lu et al., 19 98). The apparent increase in UCP-1 levels in DR but not DIO suggests that DIO rats still ha ve the capacity to increase UCP-1 levels if they receive the appr opriate periphera l signal. This prompts us to speculate that leptin re sistance in DIO animals is primarily a phenomenon of the central nervous system, not one of re duced end organ responsiveness. However, the picture changes slightly wh en UCP-1 is expressed per total BAT pad (Figure 4-11). By comparing this expressi on to the UCP-1 concentration data discussed above, it becomes clear that the increase in BAT UCP-1 concentration following rAAV-

PAGE 117

100 leptin treatment in DR was, in fact, due to a decrease in BAT tissue mass coupled with a maintenance of total UCP-1 protein. This analysis suggests that rAAV-leptin increased UCP-1 activity and lipolysis but not total protein levels in DR. This is not the case in CHOW, where rAAV-le ptin increased both UCP-1 concentration in BAT and total UCP1 per BAT pad. rAAV-leptin had no effect on UCP-1 protein or activ ity (i.e., lipolysis in BAT) in DIO. In summary, these results show that central overexpression of leptin cannot surmount the leptin resistance in a DIO model. Thus, we conclude that central delivery of leptin gene is not a viable strategy for ove rcoming leptin resistance. Leptin resistance in DIO is characterized by impaired central nervous system response to leptin, and this CNS defect must be corrected or bypassed in order to harness the power of leptin (or leptin-mediated signals) to reverse obesity.

PAGE 118

101 CHAPTER 5 CALORIC RESTRICTION REVERSES IM PAIRMENTS IN LEPTIN RECEPTOR EXPRESSION AND MAXIMAL SIGNALI NG CAPACITY IN DIET-INDUCED OBESE ANIMALS Introduction Common human obesity is associated with hyperleptinemia and leptin resistance. Clinical trials with leptin have been di sappointing due to this phenomenon of leptin resistance in the obese state (Gura, 1999) and in terest in leptin to treat obesity has waned. Although it is a subject of ri gorous investigation, the pr ecise mechanism of leptin resistance in obesity is unknown. While tr ansport of leptin acro ss the blood:brain barrier appears to be saturated at serum leptin leve ls observed in obesity (Banks et al., 1999), deficient blood:brain barrier tran sport is clearly not the only factor in leptin resistance. Both we and others have reported convincing evidence for a central component to leptin resistance in obese animal models including obese Ay mice (Halaas et al., 1997), dietinduced obese (DIO) C57BL/6J mice (El-Haschimi et al ., 2000), DIO Sprague Dawley rats (Levin and Dunn-Meynell, 2002b), a nd aged-obese F344xBN rats (Shek and Scarpace, 2000; Scarpace et al., 2001; S carpace and Tumer, 2001). As discussed in Chapter 4, we have demonstr ated that DIO F344xBN rats are completely unresponsive to an intracerebroventricular (i.c .v.) injection of rAAV-encodi ng-leptin (rAAV-leptin). A similar dose of rAAV-leptin administered to aged-matched chow-fed animals causes potent anorexia sustained for at least 150 days thermogenesis, and a near complete loss of visceral white adipose tissue (Scarpace et al., 2002b; Scarpace et al., 2003, Chapter 4).

PAGE 119

102 It has been previously dem onstrated that various obese animal models have reduced leptin receptor expression and/ or protein in the hypothalamus. This includes high fat-fed Osborne-Mendel rats (Madiehe et al., 2000), DIO C57 mice (Lin et al., 2000), and agedobese Wistar and F344XBN rats (FernandezGalaz et al., 2002; S carpace et al., 2001). Similarly, in Chapter 4 I demonstrated th at our DIO F344xBN rat model also exhibits reduced leptin receptor expr ession. We hypothesized that al terations in Ob-Rb expression in the hypothalamus may direc tly contribute to leptin sensitivity/r esistance. Our objective was to determine if DIO animals have reduced leptin receptor-mediated JAK/STAT3 signaling capacity as meas ured by maximal leptin-induced STAT3 phosphorylation in the hypothalamus. We reas oned that if DIO animals had both reduced Ob-Rb expression and reduced leptin receptor-mediated STAT3 phosphorylation capacity, then reduced Ob-Rb expression ma y directly contribute to the reduced responsiveness to pharm acological leptin observed in th ese animals. A final objective was to test whether 30 days of caloric restriction could reverse any potential receptor expression and signaling defi cits observed in DIO. Methods and Materials Animals Three-month old male Fischer 344 x Brow n Norway rats were obtained form Harlan Sprague-Dawley (Indianapolis, IN). Upon arrival, rats were examined and remained quarantined for one week. Animal s were individually caged with a 12:12 hour light:dark cycle (07:00 to 19:00 hr). Animal s were cared for in accordance with the principles of the NIH Guide to the Care and Use of Experimental Animals.

PAGE 120

103 Experimental Design Experimental design 1: leptin signaling in diet-induced obese All animals were maintained on standard rat chow (diet 2018 from Harlan Teklad, Madison, WI) from weaning until two weeks after arriving in our laboratory, at which point animals were approximately 3 months old. This CHOW diet provides 3.3 kcal/g of digestible energy and 15% of ener gy as fat. At this point, 65 animals were switched to a high fat/high sucrose (HF) diet (F3282 from Bi oServ, Frenchtown, NJ, USA). This HF diet provides 5.3 kcal/g and 59.4% of energy as fat. Over several weeks, the HF-fed animals spontaneously divided into two distin ct groups: those that were becoming obese on the HF diet (Diet Induced Obese, DIO) a nd those that were not gaining extra weight on the HF diet (Diet Resistant, DR). The t op 40% of weight gainer s on the HF diet were designated as DIO whereas the rest were de signated DR and removed from the study. An additional group of animals were continuously maintained on CHOW to serve as the “lean” control group. DIO and CHOW an imals remained on their respective diets through the conclusion of the study (when acute leptin sensitivity was evaluated) or until the start of calorie restriction (when remaini ng DIO animals were switched to chow diet). After approximately 105 days of HF feeding in DIO, groups of DIO and CHOW animals were prepared for stereotaxic surgery as desc ribed below. Animals were given a single injection into the third cerebroventricle containing 2 g of leptin peptide or 4 L ACSF vehicle. Sixty minutes after leptin de livery, animals were euthenized for the measurement of hypothalamic STAT3 phosphorylat ion and leptin receptor expression. This time frame was chosen because we previously demonstrated that leptin-induced STAT3 phosphorylation peaks at ~60 minutes (Scarpace et al., 2001).

PAGE 121

104 Experimental design 2: leptin sign aling following caloric restriction At this point (greater than 115 days of HF-feeding in remaining DIO), sub-groups of both DIO and CHOW were switched to a calo ric restriction (CR) diet. This CR diet provided 60% of mean, basal ad libitum caloric intake in the form of standard chow diet. Ad libitum caloric intake did not differ be tween CHOW and DIO and averaged approximately 62.4 Kcal/day in both groups in the week prior to commencing CR. Sixty percent of this is 37.4 Kcal/day, or 11. 34 grams of chow diet per day. Thus, approximately 11.34 grams of chow were pr ovided to each calorie restricted DIO and CHOW animal (DIO-CR and CHOW-CR) for 30 days. Food was placed in the stainless steel food baskets in the cage of each rat in the early evening (between 17:00 and 18:00). After 30 days of CR, these animals were tested for acute responsiveness to 2 g of i.c.v. leptin exactly as were the ad libitum DIO and CHOW animals (see Experimental Design 1). Leptin Administration Rats were anesthetized with 60 mg/kg pe ntobarbital and heads were prepared for surgery. Animals were placed into a stereotaxic frame and a small incision (1.5 cm) was made over the midline of the skull to expos e the landmarks of the cranium (Bregma and Lamda). The following coordinates were used for injection into 3rd cerebroventricle: 1.3 mm posterior to Bregma and 9.4 mm ventra l from the skull surface on the midline (medial fissure), with the nose bar set at 3. 3 mm below the ear bars (below zero) and the canula set at 20 posterior from vertical. A small hole was drilled through the skull and a 23-gauge stainless steel guide canula was lowered to the 3rd cerebroventricle. This was followed by an injection canula a ttached to a 10uL syringe. Two g of leptin dissolved

PAGE 122

105 in 4 uL ACSF (or ACSF vehicle alone) we re slowly injected (approximately 0.5 uL/min.) to minimize tissue damage. Tissue Harvesting Anesthetized rats were sacrificed by cervi cal dislocation 60 minutes after leptin or ACSF delivery. Blood was collected by cardiac puncture and serum was harvested by a 10 minute centrifugation in serum separator tube s. The circulatory system was perfused with 20 mL of cold saline. Perirenal a nd retroperitoneal white adipose tissue and hypothalami were excised, weighed, and imme diately frozen in liquid nitrogen. The hypothalamus was removed by making an incision medial to piriform lobes, caudal to the optic chiasm, and anterior to the cerebral crus to a depth of 2-3 mm. Tissues were stored at –80 C until analysis. Real-Time PCR We designed primers and a Taqman probe specific for the long-form leptin receptor (Ob-Rb) using Primer Express soft ware, version 1.5 (Perkin-Elmer Applied Biosystems, Inc., Foster City, CA). The se quences for the Ob-Rb primers were forward primer: 5’-GGGAACCTGTGAGGATGAG TGT-3’, reverse primer: 5’TTTCCACTGTTTTCACGTTGCT-3’. The fluor escent probe sequence was: 6FAMAGAGTCAACCCTCAGTTAAATATGCAACG CTG-TAMRA. Optimization experiments showed that 300 nM of forwar d primer, 900 nM of reverse primer, and 50 nM Taqman probe gave the most reproducib le results and maximally efficient PCR (i.e., lowest threshold cycle [CT] values). Total RNA (6 g ) was treated with RNase-free DNase using a DNA-free kit (Ambion). Fi rst-strand cDNA was ge nerated from 1.6 g RNA in a 40 L volume using random primers (Gib co BRL) containing 200 units of M-

PAGE 123

106 MLV reverse transcriptase (Gibco BRL). Real-time PCR for Ob-Rb was performed on 100 ng cDNA template in a 50 L total vo lume including Taqman RT-PCR Master Mix (Applied Biosystems, Inc., Foster City CA) using an ABI Prism GeneAmp 5700 Sequence Detection System (Applied Biosys tems). Ob-Rb expression was quantified using an 18S rRNA standard (Applied Biosystems) and the CT method (Bustin, 2000). The mean CT in the control group (CHOW) was chosen as the calibrator for CT calculation. STAT3/Phospo-STAT3 Assay These methods were described in detail previously (Scarp ace et al., 2000b). Briefly, hypothalamus was sonicated in 10 mM Tris-HCL, pH 6.8, 2% SDS, and 0.08 g/mL okadaic acid plus prot ease inhibitors (PMFS, benzam idine, and leupeptin) [an aliquot of this sonicate was frozen for RNA analysis]. Sonicate was diluted and quantified for protein using a detergent compa tible Bradford Assay. Samples were boiled and separated on an SDS-PAGE gel and electr otransferred to nitrocellulose membrane. Immunoreactivity was assessed with an an tibody specific to phosphorylated-STAT3 (antibody kit from New England Biolabs, Beverly MA). Immunoreactivity was visualized by chemiluminescents detection (Amersham Life Sciences, Piscataway, NJ) and quantified by video densitometry (BioRa d, Hercules, CA). Following P-STAT3 quantification, membranes were stripp ed of antibody with Immunopure (Pierce, Rockford, IL) and immunoreactivity was re -assessed using a total STAT3 antibody. STAT3 phosphorylation is expressed as PSTAT3/Total STAT3 in each sample. A single 26-lane Criterion gel was used to directly compare maximal leptin-induced PSTAT3 in CHOW and DIO, alt hough only representative lanes are shown in Figure 5-5.

PAGE 124

107 Leptin mRNA Levels in White Adipose Tissue Retroperitoneal white adipos e tissue (300 mg/sample) wa s sonicated in guanidine buffer, phenol extracted, and isopropanol prec ipitated using a modification of the method of Chomczynski (Chomczynski and Sacchi, 1987). Isolated RNA was re-suspended in ribonuclease-free water and quan tified by spectrophotometry. Inte grity was verified using 1% agarose gels stained with ethidium br omide. For dot blot analysis, multiple concentrations of RNA were immobilized on nylon membranes using a dot blot apparatus (BioRad, Richmond, CA). Membranes were baked in a UV crosslinking apparatus. Membranes were then prehybridized in 10 mL Quickhyb (Stratagene, LaJolla, CA) for 30 minutes followed by hybridization in the pres ence of a labeled probe for leptin mRNA and 100 ug salmon sperm DNA. After hybridization for 2 hours at 65C, the membranes were washed and exposed to a phosphor imag ing screen for 72 hours. The screen was then scanned using a Phosphor Imager (Molecular Dynamic, Sunnyvale, CA) and analyzed by Image Quant Software (Molecular D ynamics). Data is expressed as leptin mRNA per total RTWAT pad. Serum Leptin Serum leptin was measured using a mouse leptin ELISA Kit from Crystal Chem. Inc. (Chicago, IL) on blood harveste d at sacrifice by cardiac puncture. Statistical Analysis All data are expressed as mean standard error of mean. level was set at 0.05 for all analyses. Comparisons of mean food intake and body weight gain in DIO vs. CHOW were made by a 2-Way AN OVA with time and diet as factors. Comparisons of absolute body weights of CHOW versus DIO were made by student’s t-test. Comparisons in weight gain during the caloric restriction experiment (CHOW (ad lib),

PAGE 125

108 DIO-CR, and CHOW-CR groups) were made by 1-Way ANOVA with Tukeys post-hoc. All other comparisons were made by 2Way ANOVA with diet ary group (DIO vs. CHOW) and treatment (Leptin versus ACSF or ad libitum verses caloric restriction) as factors. Note that the effects of leptin and caloric restriction on STAT3 phosphorylation in DIO and CHOW were analyzed with se parate 2-Way ANOVAS, although the data is represented in a single graph. When only main effects were significant, relevant pairwise comparisons were made using the Bonferroni Multiple Comparison method with the error rate corrected for the number of contrasts (Rao, 1998). When there was an interaction, factors were separated and a further 1Way ANOVA was applied with a Bonferroni Multiple Comparison post-hoc. When sepa ration of factors resulted in only two population means to compare, the 1-WAY AN OVA was replaced with st udent’s t-test. GraphPad Prism software version 3.0 (San Diego, CA) was used for all statistical analysis and graphing. GraphPad QuickC alc (graphpad.com) was used for post-hoc analysis of all 2-WAY ANOVAS. Unless ot herwise noted: = p<0.05, ** = p<0.01, *** = p<0.001. Results Food Intake and Body Weight At the start of HF-feedi ng, there was no difference in body mass between DIO and CHOW (286.2 4.37 versus 285.64 6.03 g, respectively). Through 105 days of HFfeeding, DIO animals gained 39.3% more ma ss than chow-fed controls (CHOW) (Figure 5-1), translating to 12% greater weight at this point (467. 4 5.17 versus 415.7 8.86 g, respectively, p<0.0001). Alt hough there was a transient incr ease in caloric intake immediately after switching pre-DIO animals to the HF diet (data not shown), energy intake returned to that of CHOW animals w ithin 5 days, after which point caloric intake

PAGE 126

109 did not differ between CHOW and DIO through th e conclusion of the experiment (59.7 1.40 Kcal/day in CHOW, 61.1 0.65 Kcal/day in DIO). After approximately 115 days of HF-feed ing in DIO, sub-groups of DIO and CHOW were put on a calorie rest ricted diet, providing 60% of ad libitum caloric intake. Note that the CR diet that was provided to both CHOW-CR and DIOCR was in the form of standard chow as descri bed in Methods. By the end of the 30 day CR period, DIO-CR lost 61.9 3.46 g, eliminating the diff erence in body weights between DIO-CR and CHOW (ad libitum) (Figure 5-2). CHOW-CR lost 64.8 4.11 g during caloric restriction, leaving this group with 20.4% lower mean body weights than continuously ad libitum fed controls (Figure 5-2). Adiposity Visceral adiposity (sum of retroperit oneal and perirenal white adipose tissue depots) was elevated 2.46-fold in DIO with re spect to CHOW (Figure 5-3). Thirty days of caloric restriction reduced visceral adiposity by 55% in CHOW and by 35% in DIO (Figure 5-3). However, visceral adiposity af ter caloric restriction in DIO remained 60% above that of continuously ad libitum fed CHOW animals (Figure 5-3). Although CHOW experienced a slightly greater lo ss in visceral adi posity during CR when expressed as a percent of st arting adiposity, the magnitude of visceral fat loss was actually greater in DIO (mean visceral fat loss was 3.05 grams in CHOW and 4.79 grams in DIO). Leptin Receptor Expression in the Hypothalamus DIO animals had a 22% reduction in long fo rm leptin receptor (Ob-Rb) expression in the hypothalamus with respec t to CHOW (Figure 5-4). Ca loric restriction resulted in a highly significant increase in Ob-Rb expressi on in both groups of animals (Figure 5-4).

PAGE 127

110 This included a 43% increase in Ob-Rb e xpression in CHOW-CR versus CHOW and a 58% increase in DIO-CR versus DIO. This left DIO-CR with 24% greater hypothalamic Ob-Rb expression compared to CHOW, reversing the deficit observed in ad libitum fed DIO compared to their leaner chow -fed counterparts (Figure 5-4). Hypothalamic STAT3 Phosphorylation Basal (unstimulated) STAT3 phosphorylation was elevated approximately 3-fold in DIO compared to CHOW (Figure 5-5). Ho wever, we suspected that this does not represent maximal STAT3 phosphorylation. T hus, we administered a very high central dose of leptin that would cause maximal leptin-induced STAT3 phosphorylation at 60 minutes. The dose of leptin used in this study (2 g) is termed “supramaximal leptin” because it is >8 times the i.c.v. dose required to achieve maximal STAT3 phosphorylation in both lean and obese F334XBN rats in our laborat ory (Scarpace et al., 2001). A single i.c.v. injecti on containing supramaximal leptin caused a 6.1 fold increase in STAT3 phosphorylation in CHOW, yet cau sed only a 1.7 fold induction in STAT3 phosphorylation in DIO (Figure 55). After taking into account differences in basal STAT3 activation, there was still a signi ficant 16 percent decrease in maximally stimulated STAT3 phosphorylation capacity in DIO versus CHOW (p<0.05, Figure 5-5). Maximal STAT3 activation was then examined after 30 days of calor ic restriction. CR caused a 45% increase in maximal leptin -induced hypothalamic STAT3 phosphorylation capacity in CHOW and a dramatic 85% incr ease in DIO, completely reversing the obesity-associated impairment in signaling capacity (Figure 5-5). Similar to the pattern observed with Ob-Rb expression, DIO-CR disp layed 56% greater leptin-induced STAT3 phosphorylation capacity comp ared to the leaner ad libitum CHOW animals (Figure 5-5).

PAGE 128

111 Total STAT3 in the hypothalamus was not affect ed by diet or acute leptin. As described in Methods, STAT3 phosphorylation is normalized to total STAT3. Leptin Expression in White Adipose Tissue Leptin expression per RTWAT pad wa s elevated nearly 2.5 fold in ad libitum fed DIO compared to ad libitum fed CHOW (Figure 5-6). Thirty days of caloric restriction caused a 78% decrease in RTWAT leptin ex pression in CHOW and a 63% decrease in DIO (Figure 5-6). The effect of CR in DI O brought RTWAT leptin expression to a level comparable to that in CHOW (ad libitum) (Figure 5-6). Serum Leptin Consistent with the adiposity data, seru m leptin was elevated approximately 2.4fold in DIO compared to CHOW (Figure 57). Although we expected CR to lower serum leptin in both CHOW and DIO, an accurate measure of endogenous serum leptin in both CR groups (CHOW-CR and DIO-CR) could not be obtained due to leakage of the supramaximal leptin injection from the CSF. Thus, this raw data is difficult to interpret. Therefore, a factor that represents the leak age of supramaximal leptin (SML) from the CSF to the serum was calculated by delivering SML to 20 animals. Sixty minutes after leptin delivery, the effect of central SML on serum leptin (compared to diet-matched controls) was evaluated. SM L administration was found to increase serum leptin by an average of 11.88 ng/mL. For comparison, th is leakage factor was subtracted from CR animals, all of whom received SML (Figure 58). After this correc tion, it appears that CR decreased serum leptin in both CHOW and DIO (Figure 5-8).

PAGE 129

112 0 20 40 60 80 100 120 0 50 100 150 200 CHOW DIO DayCumulative Wieght Gain (Grams) Figure 5-1: Body mass during high fat feeding. Values represent means SEM. By 2Way ANOVA, significance was found for both the time (F=307.94, p<0.0001) and dietary group (F=857.85, p<0.0001) main effects. The interaction between time and diet was also significant (F=10.19, p<0.0001). 0 5 10 15 20 25 30 300 350 400 450 500 550 DIO-CR Lean-CR CHOW (Ad lib) DayMass (grams) Body Mass -80 -60 -40 -20 0 20 40 CHOW (Ad Lib) DIO-CR CHOW-CR Body Mass (Grams)*** *** Figure 5-2: Body mass during calo rie restriction. Values represent means SEM. Inset: Delta BM during CR period. By 1Way ANOVA Tukey post-hoc analysis, p<0.0001 for difference in total Body Mass between CHOW-CR and CHOW (ad lib) group, p<0.0001 for difference in Body Mass between DIOCR and CHOW (ad lib).

PAGE 130

113 CHOWDIO 0 5 10 15 Ad lib CR *** *** † ‡RTWAT+PWAT (g) Figure 5-3: Visceral adiposity (sum of retroperitoneal wh ite adipose tissue [RTWAT] and perirenal white adipose tissue [PWAT]) at sacrifice. Values represent mean SEM. By 2-Way ANOVA, significan ce was found for both the CR (F=79.01, p<0.0001) and dietary group (F=267.42, p< 0.0001) main effects. By post-hoc analysis, adiposity was significantly greater in DIO compared to CHOW (†p<0.001). By post-hoc analysis, effect of CR was significant both in CHOW (p<0.001) and DIO (p<0.001). Visceral adiposity in DIO-CR remained above that of CHOW (ad lib) (‡p<0.001). CHOWDIO 0.0 0.5 1.0 1.5 2.0 Ad lib CR † *** ***ObRb/18S rRNA (Arbitrary Units) Figure 5-4: Ob-Rb expression in the hypothalamus. Values represent mean SEM. By 2-Way ANOVA, significance was found fo r both the CR (F=39.85, p<0.0001) and dietary group (F=8.81, p<0.01) main effects. By post-hoc analysis, basal Ob-Rb expression was significantly redu ced in DIO (ad lib) versus CHOW (ad lib) (†)p<0.05. Effect of CR was si gnificant in both CHOW (p<0.001) and DIO (p<0.001).

PAGE 131

114 Figure 5-5: Maximal leptin -induced STAT3 phosphorylation capacity is reduced in DIO and increased by caloric restriction. (A) STAT3 Phosphoryl ation 1 hour after i.c.v. leptin (2 ug) or ACSF administration in ad libitum fed animals; STAT3 phosphorylation 1 hour after i.c.v. leptin in CR animals. Values represent mean SEM. By 2-Way ANOVA with di etary group and leptin as factors, leptin main effect was significant (F =84.40, p<0.0001), as was interaction between dietary group and leptin (F=14.1, p<0.001), but diet main effect was not significant (F=1.75). By a sec ond 2-Way ANOVA with diet and caloric restriction (CR) as factor s, significance was found only for the CR main effect (F=60.75, p<0.0001). By post-hoc analysis (a) p<0.05 for difference in basal STAT3 phosphorylation in CHOW-AC SF and DIO-ACSF (b) p<0.0001 for effect of leptin in CHOW (CHOW-ACSF vs. CHOW-Leptin) and p<0.01 for effect of leptin in DIO (DIO-ACSF vs. DIO-Leptin). (c) p<0.05 for difference in maximally stimulated STAT3 phosphorylation in CHOW-Leptin and DIO-Leptin. (d) p<0.001 and p<0.001, respectively, for effect of CR on STAT3 phosphorylation capacity in CHOW and DIO. (e) STAT3 phosphorylation capacity was elevated in DIO-CR vs. CHOW (ad lib). (B) Representative gels of P-STAT3 in the hypothalamus. TOP: lanes 1,3: CHOW animals administered 2 ug i.c.v. leptin 60 minutes prior to sacrifice (see text). lanes 2,4: CHOW-i.c.v. ACSF. lanes 5,7: DIO-i.c.v. leptin. lanes 6,8: DIO-i.c.v. ACSF. (~45 sec. expo sure to X-Ray Film). BOTTOM: All animals represented on this image were gi ven i.c.v. leptin to evaluate max. leptin signaling capacity. lanes 1,3, 5, and 7: CHOW. lanes 2,4: CHOW-CR. lanes 6,8: DIO-CR. (~20 sec. exposure). Note that densito metry of for each P-STAT3 band was divided by densitometr y for total STAT3 for that sample after stripping and reprobing nitrocellulose membrane. CHOWDIO 0.0 2.5 5.0 7.5 10.0 Ad lib (ACSF) Ad lib (Leptin) CR (Leptin) b,c a b d d,eP-STAT3/Total STAT3(Arbitrary Units) 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 P-STAT3 P-STAT3

PAGE 132

115 CHOWDIO 0 1 2 3 Ad lib CR †*** ***Leptin Expression (Arbitrary Units/Total RTWAT) Figure 5-6: Leptin expre ssion in retroperitoneal white adipose tissue (RTWAT) at sacrifice. Values represent mean SEM. By 2-Way AN OVA, both diet and CR main effects were significant (F =126.75, p<0.0001; F=153.79, p<0.0001, respectively), as was the interaction between the main effects (F=16.54, p<0.001). By post-hoc analysis, (†)p<0.0001 for difference in basal RTWAT leptin expression in CHOW (ad lib) and DIO (ad lib); p<0.0001 for effect of CR in both CHOW and DIO. CHOWDIO 0 10 20 30 Ad lib CR (plus Leptin) *** *†Serum Leptin (ng/mL) Figure 5-7: Serum leptin at sacrifice. Values represent mean SEM. By 2-Way ANOVA, both diet and (CR+leptin) main effects were significant (F=47.54, p<0.0001; F=25.73, p<0.0001, respectively). Basal serum leptin was significantly elevated in DIO (ad lib) versus CHOW (ad lib) †(p<0.01).

PAGE 133

116 CHOWDIO 0 10 20 Ad lib CR *†**Serum Leptin (Corrected For Leakage) (ng/mL) Figure 5-8: Serum leptin corrected for leakag e of infused leptin from the CSF. Values represent mean SEM. Since the effects of CR are confounded by SML (leptin) administration, it is difficult to in terpret this raw data. A factor that represents the leakage of SML from the CSF to the serum was calculated by delivering SML to 20 animals and measuri ng effect on serum leptin at 1 hour versus diet-matched controls. This factor averaged 11.88 ug/mL. After correcting for this factor: By 2-Way ANOVA, both diet a nd CR main effects were significant (F=47.54, p<0.0001; F= 15.46, p<0.01, respectively). CR significantly reduced this measure of serum leptin in both CHOW and DIO (p<0.01, p<0.05, respectively). Discussion The major objective of the present study was to determine if reduced leptin receptor expression in DIO animals is associated with a reduction in maximal leptin-induced STAT3 phosphorylation and, second, to determin e if such a deficit can be reversed by caloric restriction. Here, we report th at DIO animals have a small but significant reduction in hypothalamic Ob-Rb expression (-22%) coupled with a 3-fold elevation in basal STAT3 phosphorylation. However, when fully stimulated by leptin, maximal STAT3 phophorylation capacity in DIO is diminished by 16% compared to CHOW. Moreover, we observed a 58% increase in hypothalamic Ob-Rb expression and an

PAGE 134

117 impressive 85% increase in maximal leptin -induced STAT3 phosphorylation capacity in DIO following 30 days of caloric restriction (Figures 5-4 a nd 5-5). CR caused similar increases in hypothalamic Ob-Rb expressi on and STAT3 phosphorylation capacity in lean chow-fed animals (Figures 5-4 and 55). Although we did not measure receptor protein levels in the present study, both we and others have previ ously reported reduced Ob-Rb protein in obese animals, including bot h DIO and aged-obese (Fernandez-Galaz et al., 2002; Lin et al., 2000; Madiehe et al., 2000 ; Scarpace et al., 2001). While Madiehe et al. (2000) reported a decrease in Ob-Rb protein but not mRNA after HF-feeding, to the best of our knowledge a d ecrease in Ob-Rb mRNA but not protein has never been reported. Although not definitively pr oven in the present inves tigation, the data are suggestive of a cause-and-effect relationshi p between changes in Ob-Rb expression and changes in STAT3 phosphorylation capacity in the hypothalamus. Indeed, changes in Ob-Rb expression, including the decrease in DIO and increase with CR, were accompanied by quantitatively similar change s in STAT3 phosphorylation capacity. It has been well documented that the Ob -Rb catalyzes STAT3 phosphorylation via receptor-bound JAK2, the latter of which docks on the intracellular domain of Ob-Rb homodimers upon leptin activation of th e receptor complex (Sweeney, 2002). We believe that a decrease in Ob-Rb expression, and presumably receptor number, decreases the available amount of Ob-Rb for JAK2 binding and activation. This, in turn, decreases maximal-leptin-induced STAT3 phosphorylation capacity. The mechanism behind reduced Ob-Rb expression in DIO and enhanced Ob-Rb expression following CR remains in question. However, we believe it may involve the

PAGE 135

118 ability of leptin itself to regulate leptin r eceptor expression. Martin et al. reported a decrease in both Ob-Rb expression and prot ein following 28 days of leptin infusion (Martin et al., 2000), and we observed a similar trend following 30 days of central leptin gene therapy (Chapter 4). Moreover, Ferna ndez-Galaz et al. showed that approximately 1 month of food restriction (leading to a 40% reduction in serum leptin) reverses the reduction in Ob-Rb expression observed in hype rleptinemic aged-obese Wistar rats. These data, coupled with what we report here suggest that leptin may be a negative regulator of Ob-Rb expression. Such ligandmediated negative regulation of receptor expression is a common phenomenon. For ex ample, both the insulin receptor (Leibush and Lappova, 1995; Okabayashi et al., 1989) and IGF-1 receptor (Schillaci et al., 1998) have been shown to downregulate in response to exposure to their respective ligands. In the cytokine receptor family (t he same family as Ob-Rb), it has been demonstrated that stimulation of intestinal epith elial cells with IL-1 or TNFalpha decreases the expression of mRNA for the IL-1 receptor (McGee et al., 1996). We believe that the Ob-Rb receptor falls into this general category of receptor s whose expression and/or sensitivity is downregulated by chronic activation. If the mechanism for reduced Ob-Rb expr ession and signaling capacity in DIO and enhanced Ob-Rb expression and signaling capa city after CR involves leptin regulation of leptin receptor expression, then there must be elevated leptin in obesity and reduced leptin following CR. An elevation in serum leptin in the obese state has been well established (Levin and Dunn-Meynell, 2002a; Li n et al., 1998; Lin et al., 2000; Scarpace et al., 2001). Our present data is consistent w ith this as we report an approximate 2.4-fold increase in serum leptin in DIO compared to age-matched controls (Figure 5-7).

PAGE 136

119 Although transport of leptin across the BBB becomes saturate d at high leptin levels (Banks et al., 1999; Kastin and Pan, 2000) hyperleptinemia in the obese state is nonetheless accompanied by elevated CSF leptin (Nam et al., 2001). Moreover, the arcuate nucleus, perhaps the most critical site of leptin-responsive neurons with respect to energy balance, may effectively lie outside the BBB (Williams et al., 2001). This argument has been made because of the proxim ity of the arcuate to capillaries of the median eminence, which lack tight juncti ons. Thus, serum leptin may be a valid indicator of the arcuate’ s exposure to leptin. Short and long-term caloric restriction ha s been shown to dramatically reduced serum leptin, usually to a much greater exte nt than would be pred icted based on the loss of adiposity ( Miyawaki et al., 2002; Shimokawa and Higami, 2001). In the present study, we report significant redu ctions in leptin expression in white fat following 30 days of caloric restriction (-78% in CHOW and -63% in DIO, Figure 5-6). We were unable to obtain a reliable measure of endogenous serum lep tin in our calorie re stricted animals due to leakage of the i.c.v. supramaximal leptin from the CSF. Nevertheless, the leptin expression data strongly suggest our animal model is not an exception to the rule of reduced serum leptin following food restriction. Furtherm ore, we derived a correction factor that takes into account the leakage of supramaximal leptin from the CSF into the serum and then applied this factor to the raw data. Caloric restriction caused a significant reduction in this corrected measure of seru m leptin in both CHOW and DIO. Taken together, the changes in peripheral leptin expression following high-fat feeding and caloric restriction are cons istent with our hypothesis that leptin is regulating hypothalamic Ob-Rb expression.

PAGE 137

120 While P-STAT3 transcription factor binding was not measur ed in this study due to a shortage of hypothalamic tissue, we have observed a strong correlation between STAT3 phosphorlyation and P-STAT3 DNA binding in hypothalamic nuclear extracts from F344XBN male rats of various ages and adiposities [(S carpace et al., 2001), unpublished data]. In an earlier inves tigation by our group, i.c.v. supramaximal leptin was found to cause an approximate 6-fold increase in STAT 3 phosphorylation and an 8.6-fold increase in P-STAT3 transcription fact or binding in young adult F344x BN males (Scarpace et al., 2001). Moreover, we previously reported an approximate 36% reduction in STAT3 phosphorlation capacity in aged-obese animals and a 54% reduction in maximal leptininduced transcription factor binding in aged -obese with respect to young adults (Scarpace et al., 2001). Thus, we are confident th at the changes observed here in STAT3 phosphorylation represent quantitatively simila r changes in STAT3 transcription factor binding. We have argued for a potential causal rela tionship between reduced leptin receptor expression and signaling capacity in DIO. Nonetheless, it is possible that there are other contributors to the reduction in leptin-induced STAT3 activation in DIO independent of reduced Ob-Rb expression. For example, recent cell culture experiments suggest that prolonged exposure to high levels of leptin may promote Ob-Rb inte rnalization (Barr et al., 1999; Uotani et al., 1999). Perhaps the hyperl eptinemia in DIO is sufficient to cause Ob-Rb internalization in vivo, thereby reducing the number of leptin receptors on the surface of leptin-respons ive cells in the hypothalamus. Such a mechanism would not be detectable by measuring Ob-Rb expression or protein levels in a whole hypothalamus homogenate. In addition, it is possible that intracellular inhibitors of STAT3

PAGE 138

121 phosphorylation are up-regulated in DIO. One candidate inhi bitor molecule is SOCS3, which inhibits leptin-induced tyrosine phos phorylation of JAK2 (Bjorbaek et al., 1999). However, there is disagreement in the lite rature as to whether or not SOCS3 is upregulated in obesity (Peiser et al., 2000; Wang et al., 2000). As mentioned above, we previously de monstrated a reduction in both leptininduced hypothalamic STAT3 phosphorylation ca pacity and Ob-Rb prot ein in aged-obese F344xBN male rats (Scarpace et al., 2001). Fe rnandez-Galaz et al. reported both reduced Ob-Rb expression and receptor num ber in the hypothalami of aged-obese Wistar rats, as well as blunted physiological leptin respons iveness (Fernandez-Gal az et al., 2002). These deficits in aged-obese Wistar rats were co mpletely reversed by 3 months of moderate (~80% ad libitum) calorie restriction (Fernandez-Ga laz et al., 2002). Our present data suggests that similar defects in Ob-Rb expr ession and signaling cap acity are present in DIO and, moreover, that Ob-Rb expression and STAT3 phosphorylation capacity can be dramatically increased in both DIO and thei r CHOW-fed lean counterparts by 30 days of caloric restriction. It se ems plausible that a similar m echanism may be involved in the changes in Ob-Rb expression and leptin sens itivity observed in bot h aged-obese and DIO animals. Namely, the reductions in Ob-Rb expression and signaling capacity in aged animals may be due to chroni c hyperleptinemia, and, similarl y, the restoration of Ob-Rb expression and leptin sensitivity following CR in the Fernandez-Galaz study (FernandezGalaz et al., 2002) may be secondary to reduced leptin levels. In conclusion, reduced leptin receptor e xpression in DIO is associated with attenuated maximal leptin-induced STAT3 pho sphorylation capacity. Caloric restriction reverses these deficits, increasing both Ob -Rb expression and signa ling capacity in DIO

PAGE 139

122 to levels at or above those observed in the highly leptin responsive CHOW animals. Thus, short-term calorie restriction may be a vi able strategy to restore leptin sensitivity in previously leptin-resistant models of acquired obesity.

PAGE 140

123 CHAPTER 6 GENERAL DISCUSSION AND CONCLUSIONS At the onset of this dissert ation research, the main obj ective was to gain a better understanding of the molecular basis of obesity In particular, we were interested in defective pathways in the hypot halamic feeding/satiety cente rs that may contribute to obesity or complicate its clinical treatment. A key hormone involved in the regulation of energy balance is leptin, and we have used both leptin gene and peptide delivery to study these hypothalamic pathways. In doing so, we learned new information about leptin’s regulation of leptin receptor expression a nd leptin signal trans duction via STAT3. Moreover, we made some novel observations on impairments in leptin signaling in models of acquired obesity and how these defici ts may be reversed. In this chapter, I will highlight the major contributions this rese arch has made to the knowledge base in the field of molecular obesity research, and how it fits into the existing literature. Major Findings Physiological and Biochemical Effects of Leptin Gene Therapy are Reversible In Chapter 3, we demonstrated the dramatic weight and adiposity reducing effects of constitutively active leptin gene delivery. This was consistent with our previous findings, where 46 days of leptin transgene expr ession resulted in persistent anorexia and fat loss with no apparent attenuation of le ptin-induced STAT3 si gnaling or downstream neuropeptide regulation (Scar pace et al., 2002b). Howeve r, it was not known with certainty if these effects of leptin gene therapy were reversible or if there was a permanent changes induced in the central ne rvous system by this vector. Thus, we

PAGE 141

124 placed the leptin transgene under the control of the tetracycline transactivator and operon (TET-Ob). With this system, we demonstrated co mplete silencing of the leptin transgene when we removed doxycycline from the drinki ng water of animals. Moreover, we demonstrated that both the biochemical and physiological effects of leptin gene therapy are completely reversible. Specifically, th e effects of the leptin transgene on leptin signal transduction, POMC e xpression, UCP-1 expression in the BAT, serum leptin, bodyweight, and adiposity were all eliminated within 32 days of doxy withdrawal. A system incorporating regulation such as the TET-Ob system studied here has obvious clinical advantages, particularly in terms of managing side-effects. Leptin May Regulate Leptin Receptor Gene Expression In the TET-Ob study discussed in Chapter 3, we made a serendipitous finding. This was that leptin receptor expression in the hypothalamus appeared to be reduced by chronic leptin transgene expression in the CNS and, moreover, that this effect was reversed upon transgene silencing. In f act, hypothalamic leptin receptor expression was significantly greater in TET-Ob-OFF versus TET-Ob-ON 32 days after transgene inactivation. In Chapter 4 and 5, we then examined hypothalamic leptin receptor expression in hyperleptinemic diet-induced obese animals. These obese animals were also found to have reduced Ob-Rb expression. When we reversed th e hyperleptinemia in these animals via caloric rest riction, leptin receptor expr ession increased. Taken together, these results strongly suggest leptin is a negative regulator of leptin receptor expression. Moreover, our data with the TET-Ob system suggest that this is a direct effect of elevated leptin in the central nervous system, not an indirect effect of peripheral leptin.

PAGE 142

125 There is a Central Nervous System Component to Leptin Resistance Transport of leptin across the blood:brain ba rrier has been shown to be impaired in obese animal models (Banks et al., 1999), a nd obese animals appear to respond better to central versus periphera l leptin (Halaas et al., 1997; Van Heek et al., 1997; Niimi et al., 1999). As such, it has been reasoned that defective transport across the blood-brain barrier is a major factor in leptin resistance. Recently, it was stated that leptin resistance is a phenomenon of deficient leptin in the CNS (Dube et al., 2002). . insufficiency of central leptin, ra ther than impaired leptin receptor or postreceptor signal transducti on, contributes to the loss of weight regulation [in obesity] . by experimentally increasing leptin expression in the hypothalamus, it is possible to reinstate weight contro l for extended periods, even when rats consume HFD [a high fat diet, as in diet -induced obesity] (Dube et al., 2002, page 1734). We directly tested the hypot hesis that leptin resistance in diet-induced obesity is a phenomenon of central leptin insufficiency in Chapter 4. Diet-induced obese animals were completely non-responsive to the nor mally potent physiological and biochemical actions of central leptin gene delivery. T hus, we demonstrated that there is a central nervous system component to le ptin resistance in DIO and th at this leptin resistance cannot be overcome by central leptin overexpre ssion. This disproves the hypothesis that impaired blood:brain barrier transport is key factor in DIO-related leptin resistance. Rather, there is a CNS component to leptin resistance that must be bypassed or somehow overcome if we are to capture the power of leptin (or leptin-mediated signals) to reverse obesity.

PAGE 143

126 Diet-Induced Obese Animals Have Impa ired Leptin Signaling Capacity In Chapter 4, we demonstrated that DI O animals have reduced long form leptin receptor (Ob-Rb) expression in the hypothalamus and we repeated this finding in Chapter 5. We hypothesized that a reduction in Ob-Rb expression and, presumably, receptor number would be associated with a re duction in maximal leptin-induced STAT3 signaling capacity. In Chapter 5, we demons trated that this is indeed the case. DIO animals had reduced maximal STAT3 signaling response to a supra-maximal intracerebroventricular injection of leptin pep tide. While not definitively proven in this dissertation, we believe ther e may be a cause-and-effect relationship between reduced Ob-Rb expression and reduced sign aling capacity. Other resu lts of Chapter 5 (discussed in next paragraph) are consis tent with this belief. Impairments in Leptin Receptor Expression and Signaling in Diet-Induced Obese are Reversed by Caloric Restriction Reductions in both leptin receptor ex pression and Ob-Rb-mediated STAT3 phosphorylation capacity in DIO were comple tely reversed by 30 days of caloric restriction. In fact, ca loric restriction increased Ob-Rb signaling capacity in DIO animals to levels above those of the highly le ptin-responsive CH OW animals. It is not known what the relationship is betw een acute Ob-Rb signaling capacity and physiological leptin resistance. This certainly warrants further study. Nonetheless, it is possible that this augmented leptin receptor signaling following caloric restriction may be accompanied by enhanced physiological respon ses to pharmacological leptin. Thus, short-term calorie restriction may be a viable strategy to restore leptin sensitivity in previously leptin-resistant models of acquired obesity.

PAGE 144

127 New Insights on Leptin Resistance Evidence for Role of Lept in Signal Transduction While the phenomenon of leptin resistance wa s established in the literature at the onset of these experiments, the results presented in th is dissertation provide new information about the molecular basis of this phenomenon. Specifically, we have demonstrated that diet-induced obese, leptin resistant animals have reduced long-form leptin receptor expression in the hypothalamu s. Importantly, our data suggest that this reduced leptin receptor expression is asso ciated with a reduction in leptin-induced STAT3 phosphorylation capacity. Evidence Against Major Role for Blood:Brain Barrier Although the literature is in consistent, there is eviden ce of defective blood:brain barrier transport in the obese st ate (Banks et al., 1999), and so me believe that this is the primary cause of leptin resistance (Dube et al ., 2002). The experiments described in this dissertation demonstrate that this is not the ca se. Rather, we have shown that there is a central nervous system component to lept in resistance that cannot be overcome by overexpression of leptin in the brain. While we do not refute the potential satura tion of blood:brain barrier transport in the obese state, the results discussed in this document challenge the importance of this phenomenon as a cause of obesity and leptin non-responsiveness. Consistent with this, it has been demonstrated that CSF leptin levels are elevated in the obese state—just not to the same magnitude as serum leptin ( Nam et al., 2001; Unger, 2000). Moreover, in Chapter 4 we showed that diet-induced obese animals are completely non-responsive to the normally potent biochemical and physiological effects central leptin overexpression. These data are inconsistent w ith the model of “central leptin insufficiency” as a cause for

PAGE 145

128 obesity and resistance to exoge nous leptin. In light of th ese findings, we feel future research on leptin resistance should focus on the central nervous system.. Evidence for Down-Stream Contributors to Leptin Resistance This dissertation focused on the proximal segment of the leptin signaling pathway, namely the leptin receptor and signal transduc tion. However, some of this data is suggestive of one or more downs tream contributors to leptin re sistance. For example, in Chapter 4 rAAV-leptin caused a two-fold in crease in hypothalamic P-STAT3 in CHOW, yet had no effect in DIO. Yet basal levels of STAT3 activation in the obese state appear equivalent to the activation achieved as a re sult of rAAV-leptin tr ansgene in the CHOW animals. If the degree of STAT3 phosphoryl ation is the sole predictor of leptin responsiveness, than the DIO animals shoul d have had biochemical and physiological changes equivalent to those observed in CH OW animals given rAAV-leptin. Of course, this was not the caseDIO animals steadily ga ined weight and adiposity despite their elevation in hypothalamic P-STAT3. This suggests that the signaling defect in DIO animals lies, at least in part, downstream of leptin receptor signal transduction. Consistent with the belief that component s of leptin resistance lie downstream of the receptor, the changes in leptin receptor expression and signaling capacity in DIO reported in Chapter 5 were relatively small in magnitude (Figures 5-4 and 5-5). Although it is possible that ~20% reduction in leptin receptor expression and signal transduction capacity may contribute to leptin resistance, it is hard to believe that it is the singular factor leading to the complete leptin non-responsiveness we repor ted in DIO (Chapter 4). Although it is out of the scope of this disse rtation, there are several steps in the anorectic/thermogenic cascade activated by leptin that may be defective in obesity or leptin resistance. This cascade was discussed in detail in Chapter 1. One possibility is

PAGE 146

129 that there is a defect between leptin signal transduction and the appropriate regulation of neuropeptide synthesis and rele ase in first-order neurons in the arcuate. Specifically, POMC/ -MSH and NPY may not be upregulated and downregulated, respectively, despite elevations in leptin receptor second messengers including P-STAT3. The results discussed in Chapter 4 are consistent with this hypothesis. Specifically, hypothalamic STAT3 phosphorylation was persistently elevated in DIO versus CHOW, yet no differences in downstream neuropeptid e expression were detected. As evidence of a defect in leptin re gulation of melanocorti ns, certain obese, hyperleptinemic animal models have been s hown to have both reduced POMC expression in the arcuate and deficient -MSH peptide in the paraventri cular nucleus (PVN) (Kim et al., 2000a). Hansen et al. (2001) reported a decrease in -MSH in the PVN of hyperleptinemic, diet-induced obese Sprague Da wley rats. Given the hyperleptinemia in obese animals, an increase in POMC and -MSH would be expected if central leptin signaling was functioning normally. Interesti ngly, obese animals have also been shown to have enhanced MC4R expression in the hypothalamus, suggesting there is an upregulation of this important -MSH receptor in response to a decrease in melanocortin tone in the obese state (Harrold et al., 1999). This suggests that there may be enhanced sensitivity of the melanocortin pathway in lep tin resistant animals due to this homeostatic increase in receptor expression. Consistent with this, both obese Zucker rats and dietinduced obese rats have been shown to be hyper-responsive to -MSH or an -MSH analogue (Cettour-Rose and R ohner-Jeanrenaud, 2002; Hansen et al., 2001; Hwa et al., 2001). Scarpace et al. recently demonstrated th at leptin-induced leptin resistant rats are

PAGE 147

130 hyper-responsive to an MC4R agonist (Scarpace et al., 200 3). Thus, the MC4R may prove to be a fruitful target in attempts to circumvent leptin resistance in humans. There is also evidence suggesting involvement of a defect between leptin signal transduction and NPY downregulation. For example, in Chapter 4 we demonstrated an increase in STAT3 activati on following leptin gene therapy in lean, chow-fed animals and this was associated with a significant decrease in NPY expression. Scarpace et al. (Scarpace et al., 2002b) demonstrated the same pattern in lean animals both 9 days and 46 days after leptin gene delivery. However, in the Scarpace study, rAAV-leptin failed to significantly affect NP Y in leptin resistant aged-obese animals (Scarpace et al., 2002b). This was despite in elevation in STAT3 activation in the aged animals at both time points. The precise defect between Ob-Rb signal transduction and the appropriate regulation of both NPY and melanocorti n expression is now under rigorous investigation. Conclusion Obesity is a serious and growing public hea lth issue. The percent of U.S. adults that are obese has doubled in the last 20 years, and a similar pattern has been recorded in both children and teenagers (Marx, 2003). Obesity is a risk factor for diabetes, cardiovascular disease, stroke, and certain cancers. Far from a simple cosmetic concern, obesity is a legitimate kill er, claiming an estimated 300,000 Americans each year. Thus, acquiring an understanding of the molecular basi s of obesity and how to overcome it is an important goal of biomedical research. In the research discussed in this disserta tion, I studied how the brains response to the hormone leptin is impaired in obes ity. I demonstrated that leptin receptor expression and signaling is impa ired in the brains of obese animals. Moreover, I showed

PAGE 148

131 that these defects can be reve rsed by thirty days of caloric restriction to 60% of normal (ad libitum) food intake. I also part icipated in the developm ent and testing of a novel leptin gene delivery system that allows for external regulation of the leptin transgene expression. Such a system is both a useful research tool and a model for safer gene therapy systems that may one day be used clinically. I hope that the results discussed in this dissertation have made important contri butions to the field of molecular obesity research.

PAGE 149

132 LIST OF REFERENCES Ahima RS and Flier JS (2000a) Adipos e tissue as an endocrine organ. Trends Endocrinol Metab 11:327-332. Ahima RS and Flier JS (2000b) Leptin. Annu Rev Physiol 62:413-437. Akabayashi A, Zaia CT, Gabriel SM, Silv a I, Cheung WK and Leibowitz SF (1994) Intracerebroventricular injection of dibutyryl cyclic adenosine 3',5'monophosphate increases hypothalamic levels of neuropeptide Y. Brain Res 660:323-328. Arase K, York DA, Shimizu H, Shargill N and Bray GA (1988) Eff ects of corticotropinreleasing factor on food intake and brow n adipose tissue thermogenesis in rats. Am J Physiol 255:E255-259. Ashford ML, Boden PR and Treherne JM (1990) Glucose-induced excitation of hypothalamic neurones is mediated by ATPsensitive K+ channels. Pflugers Arch 415:479-483. Asnicar MA, Smith DP, Yang DD, Heiman ML, Fox N, Chen YF, Hsiung HM and Koster A (2001) Absence of cocaineand amphetamine-regulated transcript results in obesity in mice fed a high caloric diet. Endocrinology 142:4394-4400. Astrup A (2000) Thermogenic drugs as a strategy for treatment of obesity. Endocrine 13:207-212. Bagdade JD, Bierman EL and Porte D, Jr. (1967) The significance of basal insulin levels in the evaluation of the insulin response to glucose in diabetic and nondiabetic subjects. J Clin Invest 46:1549-1557. Banks WA, DiPalma CR and Farrell CL (1999) Im paired transport of leptin across the blood-brain barrier in obesity. Peptides 20:1341-1345. Barr VA, Lane K and Taylor SI (1999) Subcel lular localization and internalization of the four human leptin receptor isoforms. J Biol Chem 274:21416-21424. Bates SH, Stearns WH, Dundon TA, Schubert M, Tso AW, Wang Y, Banks AS, Lavery HJ, Haq AK, Maratos-Flier E, Neel BG, Schwartz MW and Myers MG, Jr. (2003) STAT3 signalling is required for leptin regulation of ener gy balance but not reproduction. Nature 421:856-859.

PAGE 150

133 Baumann H, Morella KK, White DW, Dembski M, Bailon PS, Kim H, Lai CF and Tartaglia LA (1996) The full-length leptin receptor has signaling capabilities of interleukin 6-type cytokine receptors. Proc Natl Acad Sci U S A 93:8374-8378. Bell AC, West AG and Felsenfeld G (1999) The protein CTCF is required for the enhancer blocking activity of vertebrate insulators. Cell 98:387-396. Bergen HT, Monkman N and Mobbs CV (1996) Injection with gold thioglucose impairs sensitivity to glucose: evidence that gl ucose-responsive neurons are important for long-term regulation of body weight. Brain Res 734:332-336. Berns KI and Bohenzky RA (1987) Ade no-associated viruses: an update. Adv Virus Res 32:243-306. Bjorbaek C, Buchholz RM, Davis SM, Bates SH, Pierroz DD, Gu H, Neel BG, Myers MG, Jr. and Flier JS (2001) Divergent roles of SHP-2 in ERK activation by leptin receptors. J Biol Chem 276:4747-4755. Bjorbaek C, El-Haschimi K, Frantz JD and F lier JS (1999) The role of SOCS-3 in leptin signaling and leptin resistance. J Biol Chem 274:30059-30065. Bjorbaek C, Elmquist JK, Fr antz JD, Shoelson SE and Flie r JS (1998) Identification of SOCS-3 as a potential mediator of central leptin resistance. Mol Cell 1:619-625. Bjorntorp P (1995) Endocrin e abnormalities of obesity. Metabolism 44:21-23. Blacklow NR, Hoggan MD and Rowe WP (196 7) Isolation of ad enovirus-associated viruses from man. Proc Natl Acad Sci U S A 58:1410-1415. Blacklow NR, Hoggan MD and Rowe WP (1968) Serologic evidence for human infection with adenovirus-associated viruses. J Natl Cancer Inst 40:319-327. Boss O, Samec S, Paoloni-Giacobino A, Ro ssier C, Dulloo A, Seydoux J, Muzzin P and Giacobino JP (1997) Uncoupli ng protein-3: a new member of the mitochondrial carrier family with tissue-specific expression. FEBS Lett 408:39-42. Brabant G, Horn R, von zur Muhlen A, Mayr B, Wurster U, Heidenreich F, Schnabel D, Gruters-Kieslich A, Zimmermann-Belsing T and Feldt-Rasmussen U (2000) Free and protein bound leptin are di stinct and independently co ntrolled factors in energy regulation. Diabetologia 43:438-442. Bray GA (2000) Reciprocal relation of food intake and sympathetic activity: experimental observations a nd clinical implications. Int J Obes Relat Metab Disord 24 Suppl 2:S8-17. Bustin SA (2000) Absolute quantification of mRNA using real-time reverse transcription polymerase chain reaction assays. J Mol Endocrinol 25:169-193.

PAGE 151

134 Cettour-Rose P and Rohner-Jeanren aud F (2002) The leptin-like effects of 3-d peripheral administration of a melanocortin agonist are more marked in genetically obese Zucker (fa/fa) than in lean rats. Endocrinology 143:2277-2283. Chen G, Koyama K, Yuan X, Lee Y, Z hou YT, O'Doherty R, Newgard CB and Unger RH (1996) Disappearance of body fat in normal rats induced by adenovirusmediated leptin gene therapy. Proc Natl Acad Sci U S A 93:14795-14799. Chomczynski P and Sacchi N (1987) Singl e-step method of R NA isolation by acid guanidinium thiocyanatephe nol-chloroform extraction. Anal Biochem 162:156159. Christophe J (1998) Is there appe tite after GLP-1 and PACAP? Ann N Y Acad Sci 865:323-335. Clapham JC, Arch JR, Chapman H, Haynes A, Lister C, Moor e GB, Piercy V, Carter SA, Lehner I, Smith SA, Beeley LJ, Godden RJ Herrity N, Skehel M, Changani KK, Hockings PD, Reid DG, Squires SM, Hatche r J, Trail B, Latcham J, Rastan S, Harper AJ, Cadenas S, Buckingham JA Brand MD and Abuin A (2000) Mice overexpressing human uncoupling protein-3 in skeletal muscle are hyperphagic and lean. Nature 406:415-418. Commins SP, Watson PM, Frampton IC a nd Gettys TW (2001) Leptin selectively reduces white adipose tissue in mice via a UCP1dependent mechanism in brown adipose tissue. Am J Physiol Endocrinol Metab 280:E372-377. Cone RD (1999) The central melanocorti n system and energy homeostasis. Trends Endocrinol Metab 10:211-216. Conway JE, Rhys CM, Zolotukhin I, Zolot ukhin S, Muzyczka N, Hayward GS and Byrne BJ (1999) High-titer recombinant adeno-a ssociated virus production utilizing a recombinant herpes simplex virus type I vector expressing AAV-2 Rep and Cap. Gene Ther 6:986-993. de Lecea L, Kilduff TS, Peyron C, Gao X, Foye PE, Danielson PE, Fukuhara C, Battenberg EL, Gautvik VT, Bartlett FS, 2nd, Frankel WN, van den Pol AN, Bloom FE, Gautvik KM a nd Sutcliffe JG (1998) Th e hypocretins: hypothalamusspecific peptides with neuroexcitatory activity. Proc Natl Acad Sci U S A 95:322327. Devos R, Guisez Y, Van der Heyden J, White DW, Kalai M, Fountoulakis M and Plaetinck G (1997) Ligand-independent dimeri zation of the extrac ellular domain of the leptin receptor and determination of the stoichiometry of leptin binding. J Biol Chem 272:18304-18310.

PAGE 152

135 Dhillon H, Kalra SP, Prima V, Zolotukhin S, Scarpace PJ, Moldawer LL, Muzyczka N and Kalra PS (2001) Central leptin gene therapy suppresses body weight gain, adiposity and serum insulin without aff ecting food consumption in normal rats: a longterm study. Regul Pept 99:69-77. Douglass J, McKinzie AA and Couceyro P ( 1995) PCR differential display identifies a rat brain mRNA that is transcriptionally regulated by cocaine and amphetamine. J Neurosci 15:2471-2481. Dube MG, Beretta E, Dhillon H, Ueno N, Ka lra PS and Kalra SP ( 2002) Central leptin gene therapy blocks high-fat diet-in duced weight gain, hyperleptinemia, and hyperinsulinemia: increase in serum ghrelin levels. Diabetes 51:1729-1736. Ducy P, Amling M, Takeda S, Priemel M, Schilling AF, Beil FT, Shen J, Vinson C, Rueger JM and Karsenty G (2000) Lep tin inhibits bone formation through a hypothalamic relay: a central control of bone mass. Cell 100:197-207. El-Haschimi K, Pierroz DD, Hileman SM, Bj orbaek C and Flier JS (2000) Two defects contribute to hypothalamic leptin resist ance in mice with diet-induced obesity. J Clin Invest 105:1827-1832. Elias CF, Aschkenasi C, Lee C, Kelly J, Ah ima RS, Bjorbaek C, Flier JS, Saper CB and Elmquist JK (1999) Leptin differentia lly regulates NPY and POMC neurons projecting to the late ral hypothalamic area. Neuron 23:775-786. Elmquist JK, Maratos-Flier E, Saper CB a nd Flier JS (1998) Unraveling the central nervous system pathways underlying responses to leptin. Nat Neurosci 1:445-450. Fan W, Boston BA, Kesterson RA, Hruby VJ and Cone RD (1997) Role of melanocortinergic neurons in feedin g and the agouti obesity syndrome. Nature 385:165-168. Fernandez-Galaz C, Fernandez-Agullo T, Pe rez C, Peralta S, Arribas C, Andres A, Carrascosa JM and Ros M (2002) Long-te rm food restriction prevents ageingassociated central leptin resistance in wistar rats. Diabetologia 45:997-1003. Fisher KJ, Jooss K, Alston J, Yang Y, Haeck er SE, High K, Pathak R, Raper SE and Wilson JM (1997) Recombinant adeno-associ ated virus for muscle directed gene therapy. Nat Med 3:306-312. Fong TM, Huang RR, Tota MR, Mao C, Smith T, Varnerin J, Karpitskiy VV, Krause JE and Van der Ploeg LH (1998) Localization of leptin binding doma in in the leptin receptor. Mol Pharmacol 53:234-240. Friedman J (2003) A war on obesity, not the obese. Science 299: 856-858

PAGE 153

136 Fruhbeck G, Gomez-Ambrosi J, Muruzabal FJ and Burrell MA (2001) The adipocyte: a model for integration of endocrine and metabolic signaling in energy metabolism regulation. Am J Physiol Endocrinol Metab 280:E827-847. Gainsford T, Willson TA, Metcalf D, Handma n E, McFarlane C, Ng A, Nicola NA, Alexander WS and Hilton DJ (1996) Leptin can induce proliferation, differentiation, and functional act ivation of hemopoietic cells. Proc Natl Acad Sci U S A 93:14564-14568. Georg-Fries B, Biederlack S, Wolf J and zur Hausen H ( 1984) Analysis of proteins, helper dependence, and seroepidem iology of a new human parvovirus. Virology 134:64-71. Ghilardi N, Ziegler S, Wiestner A, Stoffe l R, Heim MH and Skoda RC (1996) Defective STAT signaling by the leptin receptor in diabetic mice. Proc Natl Acad Sci U S A 93:6231-6235. Greenhalgh CJ and Hilton DJ (2001) Negativ e regulation of cytokine signaling. J Leukoc Biol 70:348-356. Grill HJ, Ginsberg AB, Seeley RJ and Kapl an JM (1998) Brainstem application of melanocortin receptor ligands produces l onglasting effect s on feeding and body weight. J Neurosci 18:10128-10135. Grimm D, Kern A, Rittner K and Kleinschmi dt JA (1998) Novel tools for production and purification of recombinant ad enoassociated virus vectors. Hum Gene Ther 9:27452760. Gura T (1999) Obesity research. Leptin not impressive in clinical trial. Science 286:881882. Gura T (2003) Obesity drug pipeline not so fat. Science 299: 849-852 Hagen T, Zhang CY, Vianna CR and Lowell BB (2000) Uncoupling proteins 1 and 3 are regulated differently. Biochemistry 39:5845-5851. Hakansson ML and Meister B (1998) Transcripti on factor STAT3 in leptin target neurons of the rat hypothalamus. Neuroendocrinology 68:420-427. Halaas JL, Boozer C, Blair-West J, Fida husein N, Denton DA and Friedman JM (1997) Physiological response to long-term periphe ral and central leptin infusion in lean and obese mice. Proc Natl Acad Sci U S A 94:8878-8883. Hansen MJ, Ball MJ and Morris MJ (2001) E nhanced inhibitory feeding response to alpha-melanocyte stimulating hormone in the diet-induced obese rat. Brain Res 892:130-137.

PAGE 154

137 Harrold JA, Widdowson PS and Williams G (1999) Altered energy balance causes selective changes in melanocortin-4(MC4R), but not melanocortin-3 (MC3-R), receptors in specific hypothalamic regions : further evidence that activation of MC4-R is a physiological inhibitor of feeding. Diabetes 48:267-271. Hauswirth WW, Lewin AS, Zolotukhin S and Muzyczka N (2000) Production and purification of recombinant adeno-associated virus. Methods Enzymol 316:743-761. Haynes AC, Jackson B, Overend P, Buck ingham RE, Wilson S, Tadayyon M and Arch JR (1999) Effects of single and chronic intracerebrovent ricular administration of the orexins on feeding in the rat. Peptides 20:1099-1105. Hermonat PL, Labow MA, Wright R, Bern s KI and Muzyczka N (1984) Genetics of adeno-associated virus: isolation and preliminary characterization of adenoassociated virus type 2 mutants. J Virol 51:329-339. Heymsfield SB, Greenberg AS, Fujioka K, Dixon RM, Kushner R, Hunt T, Lubina JA, Patane J, Self B, Hunt P and McCamish M (1999) Recombinant leptin for weight loss in obese and lean adults: a randomi zed, controlled, dose-escalation trial. Jama 282:1568-1575. Hillebrand JJ, de Wied D and Adan RA (2002) Neuropeptides, food intake and body weight regulation: a hypothalamic focus. Peptides 23:2283-2306. Huang CJ, Spinella F, Nazarian R, Lee MM, Dopp JM and de Vellis J (1999) Expression of green fluorescent protein in oligodendr ocytes in a timeand level-controllable fashion with a tetracycline-regulated system. Mol Med 5:129-137. Hwa JJ, Ghibaudi L, Gao J and Parker EM (2001) Central melanocortin system modulates energy intake and expenditure of obese and lean Zucker rats. Am J Physiol Regul Integr Comp Physiol 281:R444-451. Kamegai J, Tamura H, Shimizu T, Ishii S, Sugihara H and Wakabayashi I (2001) Chronic central infusion of ghrelin increase s hypothalamic neuropeptide Y and Agoutirelated protein mRNA levels and body weight in rats. Diabetes 50:2438-2443. Kanai F, Ito K, Todaka M, Hayashi H, Kam ohara S, Ishii K, Okada T, Hazeki O, Ui M and Ebina Y (1993) Insulin-stimulated GLUT 4 translocation is relevant to the phosphorylation of IRS-1 and th e activity of PI3-kinase. Biochem Biophys Res Commun 195:762-768. Kaplitt MG, Leone P, Samulski RJ, Xiao X, Pfaff DW, O'Malley KL and During MJ (1994) Long-term gene expression an d phenotypic correction using adenoassociated virus vectors in the mammalian brain. Nat Genet 8:148-154. Kastin AJ and Pan W (2000) D ynamic regulation of leptin entry into brain by the bloodbrain barrier. Regul Pept 92:37-43.

PAGE 155

138 Kearns WG, Afione SA, Fulmer SB, Pang MC, Erikson D, Egan M, Landrum MJ, Flotte TR and Cutting GR (1996) Recombinant adeno-associated virus (AAV-CFTR) vectors do not integrate in a site-specific fashion in an immortalized epithelial cell line. Gene Ther 3:748-755. Kelner K and Helmuth L (2003) Obesity--What is to be done? Science 299:845. Khaodhiar L, McCowen KC and Blackburn GL (1999) Obesity and its comorbid conditions. Clin Cornerstone 2:17-31. Kim EM, O'Hare E, Grace MK, Welch CC, Billington CJ and Levine AS (2000a) ARC POMC mRNA and PVN alpha-MSH are lower in obese relative to lean zucker rats. Brain Res 862:11-16. Kim YB, Uotani S, Pierroz DD, Flier JS and Kahn BB (2000b) In vivo administration of leptin activates signal transduction direct ly in insulin-sensitive tissues: overlapping but distinct pathways from insulin. Endocrinology 141:2328-2339. Klein RL, Muir D, King MA, Peel AL, Zolo tukhin S, Moller JC, Kruttgen A, Heymach JV, Jr., Muzyczka N and Meyer EM (1999) Long-term actions of vector-derived nerve growth factor or brainderi ved neurotrophic factor on choline acetyltransferase and Trk receptor levels in the adult rat basal forebrain. Neuroscience 90:815-821. Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H and Kangawa K (1999) Ghrelin is a growth-hormone-releasing acylated peptide from stomach. Nature 402:656-660. Kopelman PG (2000) Obesity as a medical problem. Nature 404:635-643. Koyama K, Chen G, Wang MY, Lee Y, Sh imabukuro M, Newgard CB and Unger RH (1997) beta-cell function in normal rats made chronically hyperleptinemic by adenovirus-leptin gene therapy. Diabetes 46:1276-1280. Koyama K, Shimabukuro M, Chen G, Wang MY Lee Y, Kalra PS, Dube MG, Kalra SP, Newgard CB and Unger RH (1998) Re sistance to adenovirally induced hyperleptinemia in rats. Comparison of ventromedial hypothalamic lesions and mutated leptin receptors. J Clin Invest 102:728-733. Kristensen P, Judge ME, Thim L, Ribel U, Christjansen KN, Wulff BS, Clausen JT, Jensen PB, Madsen OD, Vrang N, Larsen PJ and Hastrup S (1998) Hypothalamic CART is a new anorectic peptide regulated by leptin. Nature 393:72-76. Kuhar MJ and Dall Vechia SE (1999) CART peptides: novel addi ctionand feedingrelated neuropeptides. Trends Neurosci 22:316-320. Lambert PD, Couceyro PR, McGirr KM, Dall Vechia SE, Smith Y and Kuhar MJ (1998) CART peptides in the centra l control of feeding and inte ractions with neuropeptide Y. Synapse 29:293-298.

PAGE 156

139 Larsen PJ, Vrang N, Petersen PC and Kristensen P (2000) Chronic intracerebroventricular administration of recombinant CART(4289) peptide inhibits and causes weight loss in l ean and obese Zucker (fa/fa) rats. Obes Res 8:590-596. Leibush BN and Lappova YL (1995) Insulin receptor downregulation in isolated hepatocytes of river lampre y (Lampetra fluviatilis). Gen Comp Endocrinol 100:1017. Levin BE and Dunn-Meynell AA (2002a) Defe nse of body weight depends on dietary composition and palatability in rats with diet-induced obesity. Am J Physiol Regul Integr Comp Physiol 282:R46-54. Levin BE and Dunn-Meynell AA (2002b) Reduced cen tral leptin sensitivity in rats with diet-induced obesity. Am J Physiol Regul Integr Comp Physiol 283:R941-948. Levin BE, Dunn-Meynell AA, Balkan B and Keesey RE (1997) Selective breeding for diet-induced obesity and resistan ce in SpragueDawley rats. Am J Physiol 273:R725-730. Levin BE and Keesey RE (1998) Defense of differing body weight se t points in dietinduced obese and resistant rats. Am J Physiol 274:R412-419. Levin BE, Triscari J, Hogan S and Sullivan AC (1987) Resistance to diet-induced obesity: food intake, pancreatic sympathetic tone, and insulin. Am J Physiol 252:R471-478. Lin S, Storlien LH and Huang XF (2000) Leptin receptor, NPY, POMC mRNA expression in the diet-in duced obese mouse brain. Brain Res 875:89-95. Lin X, Chavez MR, Bruch RC, Kilroy GE, Si mmons LA, Lin L, Braymer HD, Bray GA and York DA (1998) The effects of a high fat diet on leptin mRNA, serum leptin and the response to leptin are not altered in a rat strain susceptible to high fat dietinduced obesity. J Nutr 128:1606-1613. Linden RM and Berns KI (2000) Molecular biology of adeno-associated viruses. Contrib Microbiol 4:68-84. Liu B, Liao J, Rao X, Kushner SA, Chung CD, Chang DD and Shuai K (1998) Inhibition of Stat1-mediated gene activation by PIAS1. Proc Natl Acad Sci U S A 95:1062610631. Loeb JE, Cordier WS, Harris ME, Weitzman MD and Hope TJ (1999) Enhanced expression of transgenes from adeno-asso ciated virus vectors with the woodchuck hepatitis virus posttranscr iptional regulatory element: implications for gene therapy. Hum Gene Ther 10:2295-2305.

PAGE 157

140 Lu H, Duanmu Z, Houck C, Jen KL, Buison A and Dunbar JC (1998) Obesity due to high fat diet decreases the sympathetic ner vous and cardiovascular responses to intracerebroventricula r leptin in rats. Brain Res Bull 47:331-335. Lundberg C, Jungles SJ and Mulligan RC ( 2001) Direct delivery of leptin to the hypothalamus using recombinant adenoa ssociated virus vectors results in increased therapeutic efficacy. Nat Biotechnol 19:169-172. Lundin A, Rondahl H, Walu m E and Wilcke M (2000) E xpression and intracellular localization of leptin recept or long isoform-GFP chimera. Biochim Biophys Acta 1499:130-138. Madiehe AM, Schaffhauser AO, Brayme r DH, Bray GA and York DA (2000) Differential expression of leptin receptor in highand low-fat-fed Osborne-Mendel and S5B/Pl rats. Obes Res 8:467-474. Martin RL, Perez E, He YJ, Dawson R, Jr. and Millard WJ (2000) Leptin resistance is associated with hypothalamic leptin receptor mRNA and protein downregulation. Metabolism 49:1479-1484. Marx J (2003) Cellular warriors at the battle of the bulge. Science 299:846-849. Matarese G (2000) Leptin and the immune sy stem: how nutritional status influences the immune response. Eur Cytokine Netw 11:7-14. Mayer J (1955) Regulation of the energy in take and the body wei ght: the glucostatic theory and the lipostatic hypothesis. Ann N Y Acad Sci 63:15-43. McGee DW, Vitkus SJ and Lee P (1996) The effect of cytokine stimulation on IL-1 receptor mRNA expression by intestinal epithelial cells. Cell Immunol 168:276280. Meneses PI (1999) Analysis of adeno-associat ated viurs integrati on, and evaluation of recombinant angiostatin for gene therapy. Cornell University Doctoral Dissertation. Miyawaki T, Masuzaki H, Ogawa Y, Hosoda K, Nishimura H, Azuma N, Sugawara A, Masuda I, Murata M, Matsuo T, Hayash i T, Inoue G, Yoshimasa Y and Nakao K (2002) Clinical implications of leptin and its potential humoral re gulators in longterm low-calorie diet therapy for obese humans. Eur J Clin Nutr 56:593-600. Morsy MA, Gu MC, Zhao JZ, Holder DJ, R ogers IT, Pouch WJ, Motzel SL, Klein HJ, Gupta SK, Liang X, Tota MR, Rosenblum CI and Caskey CT (1998) Leptin gene therapy and daily protein ad ministration: a comparative study in the ob/ob mouse. Gene Ther 5:8-18. Morton NM, Emilsson V, de Groot P, Palle tt AL and Cawthorne MA (1999) Leptin signalling in pancreatic islets and clonal insulin-secreting cells. J Mol Endocrinol 22:173-184.

PAGE 158

141 Morton NM, Emilsson V, Liu YL and Cawthorne MA (1998) Leptin action in intestinal cells. J Biol Chem 273:26194-26201. Murphy JE, Zhou S, Giese K, Williams LT, Escobedo JA and Dwarki VJ (1997) Longterm correction of obesity and diabetes in genetically obese mice by a single intramuscular injection of recombinant adeno-associated virus encoding mouse leptin. Proc Natl Acad Sci U S A 94:13921-13926. Muzyczka N (1992) Use of adeno-associated vi rus as a general transduction vector for mammalian cells. Curr Top Microbiol Immunol 158:97-129. Muzzin P, Cusin I, Charnay Y and Rohner-Jeanrenaud F (2000) Single intracerebroventricular bolu s injection of a recombinant adenovirus expressing leptin results in re duction of food intake and body we ight in both lean and obese Zucker fa/fa rats. Regul Pept 92:57-64. Nakashima K, Narazaki M and Taga T (1997) Leptin receptor (OB-R) oligomerizes with itself but not with its closely relate d cytokine signal transducer gp130. FEBS Lett 403:79-82. Nam SY, Kratzsch J, Kim KW, Kim KR, Li m SK and Marcus C (2001) Cerebrospinal fluid and plasma concentrations of lep tin, NPY, and alphaMSH in obese women and their relationship to negative energy balance. J Clin Endocrinol Metab 86:4849-4853. Niimi M, Sato M, Yokote R, Tada S and Takahara J (1999) Effects of central and peripheral injection of leptin on food intake and on brain Fos expression in the Otsuka Long-Evans Tokushima Fa tty rat with hyperleptinaemia. J Neuroendocrinol 11:605-611. Niswender KD, Morton GJ, St earns WH, Rhodes CJ, Myers MG, Jr. and Schwartz MW (2001) Intracellular sign alling. Key enzyme in leptin-induced anorexia. Nature 413:794-795. Niswender KD and Schwartz MW (2003) Insulin and leptin revisited: adiposity signals with overlapping physiological and in tracellular signali ng capabilities. Front Neuroendocrinol 24:1-10. Okabayashi Y, Maddux BA, McDonald AR, L ogsdon CD, Williams JA and Goldfine ID (1989) Mechanisms of insulin-induced in sulin-receptor downregulation. Decrease of receptor biosynthesis and mRNA levels. Diabetes 38:182-187. Otto B, Cuntz U, Fruehauf E, Wawarta R, Folwaczny C, Riepl RL, Heiman ML, Lehnert P, Fichter M and Tschop M ( 2001) Weight gain decreases elevated plasma ghrelin concentrations of patients with anorexia nervosa. Eur J Endocrinol 145:669-673.

PAGE 159

142 Peel AL, Zolotukhin S, Schrimsher GW, Mu zyczka N and Reier PJ (1997) Efficient transduction of green fluorescent protein in spinal cord neurons using adenoassociated virus vectors containi ng cell typespecific promoters. Gene Ther 4:1624. Peiser C, McGregor GP and Lang RE (2000) Leptin receptor expression and suppressor of cytokine signaling transcript levels in high-fat-fed rats. Life Sci 67:2971-2981. Peyron C, Tighe DK, van den Pol AN, de L ecea L, Heller HC, Sutcliffe JG and Kilduff TS (1998) Neurons containing hypocretin (o rexin) project to multiple neuronal systems. J Neurosci 18:9996-10015. Pierroz DD, Ziotopoulou M, Ungsunan L, Mosc hos S, Flier JS and Mantzoros CS (2002) Effects of acute and chroni c administration of the melanocortin agonist MTII in mice with diet-induced obesity. Diabetes 51:1337-1345. Porter RK (2001) Mitochondria l proton leak: a role for uncoupling proteins 2 and 3? Biochim Biophys Acta 1504:120-127. Qiu J, Ogus S, Lu R and Chehab FF (2001) Transgenic mice overexpressing leptin accumulate adipose mass at an older, but not younger, age. Endocrinology 142:348-358. Rao PV (1998) Statistical Research Methods in the Life Sciences. Pacific Grove, CA: Brooks/Cole Publishing Company. Riediger T, Traebert M, Schmid HA, Scheel C, Lutz TA and Scharrer E (2003) Sitespecific effects of ghrelin on the neuronal activity in the hypothalamic arcuate nucleus. Neurosci Lett 341:151-155. Rohner-Jeanrenaud F, Craft LS, Bridwell J, Su ter TM, Tinsley FC, Smiley DL, Burkhart DR, Statnick MA, Heiman ML, Ravussin E and Caro JF (2002) Chronic central infusion of cocaineand amphetamine-regula ted transcript (CAR T 55-102): effects on body weight homeostasis in lean and high-fat-fed obese rats. Int J Obes Relat Metab Disord 26:143-149. Sahu A (2002) Resistance to th e satiety action of leptin fo llowing chronic central leptin infusion is associated with the development of leptin resistance in neuropeptide Y neurones. J Neuroendocrinol 14:796-804. Scapinelli R (1975) [Obesity and arti cular tropho-mechanic disorders]. Clin Ortop 26:97108. Scarpace PJ and Matheny M (1998) Leptin induction of UCP1 gene expression is dependent on sympathetic innervation. Am J Physiol 275:E259-264.

PAGE 160

143 Scarpace PJ, Matheny M, Borst S and Tumer N (1994) Thermoregulation with age: role of thermogenesis and uncoupling protei n expression in brown adipose tissue. Proc Soc Exp Biol Med 205:154-161. Scarpace PJ, Matheny M, Moore RL and Tumer N (2000a) Impaired leptin responsiveness in aged rats. Diabetes 49:431-435. Scarpace PJ, Matheny M, Pollock BH and Tumer N (1997) Leptin increases uncoupling protein expression and energy expenditure. Am J Physiol 273:E226-230. Scarpace PJ, Matheny M and Shek EW (2000b) Impaired leptin signal transduction with age-related obesity. Neuropharmacology 39:1872-1879. Scarpace PJ, Matheny M and Tumer N (2001) Hypothalamic leptin resistance is associated with impaired leptin signal transduction in aged obese rats. Neuroscience 104:1111-1117. Scarpace PJ, Matheny M, Zhang Y, Shek EW, Prima V, Zolotukhin S and Tumer N (2002a) Leptin-induced leptin resistance reve als separate roles for the anorexic and thermogenic responses in weight maintenance. Endocrinology 143:3026-3035. Scarpace PJ, Matheny M, Zhang Y, Tumer N, Frase CD, Shek EW, Hong B, Prima V and Zolotukhin S (2002b) Central le ptin gene delivery evokes persistent leptin signal transduction in young and aged-obese rats but physiological responses become attenuated over time in aged-obese rats. Neuropharmacology 42:548-561. Scarpace PJ, Matheny M, Zolotukhin S, Tumer N and Zhang Y (2003) Leptin-induced leptin resistant rats exhib it enhanced responses to th e melanocortin agonist MTII. Neuropharmacology (in press). Scarpace PJ, Matheny, M., Zhang, Y, Tumer, N, Frase, C, Hong, B, Prima, V, and Zolotukhin, S. (2002c) Central leptin gene delivery evokes persistent leptin signal transduction in young and aged-obese rats but physiological responses become attenuated over time in aged-obese rats. Neuropharmacology 42:549-562. Scarpace PJ, Nicolson M and Matheny M (1998) UCP2, UCP3 and leptin gene expression: modulation by f ood restriction and leptin. J Endocrinol 159:349-357. Scarpace PJ and Tumer N (2001) Peripheral and hypothalamic leptin resistance with agerelated obesity. Physiol Behav 74:721-727. Schillaci R, Brocardo MG, Ga leano A and Roldan A (1998) Downregulation of insulinlike growth factor-1 receptor (IGF-1 R) expression in human T lymphocyte activation. Cell Immunol 183:157-161.

PAGE 161

144 Shek EW, Kim Y, Wilsey JT, Li G and Scarpace PJ (2002) The metabolic responses to chronic i.c.v. infusion of MC3/4 receptor agonist are mediated by enhanced oxygen consumption, palmitate oxidation, and uncoupling protein levels. Endocrine Society Annual Meeting June 19-22: San Francisc o, CA. Abstract P2-247. Shek EW and Scarpace PJ (2000) Resistance to the anorexic and thermogenic effects of centrally administrated leptin in obese aged rats. Regul Pept 92:65-71. Shepherd PR, Withers DJ and Siddle K ( 1998) Phosphoinositide 3-kinase: the key switch mechanism in insulin signalling. Biochem J 333:471-490. Shimokawa I and Higami Y (2001) Leptin and anti-aging action of cal oric restriction. J Nutr Health Aging 5:43-48. Siegrist-Kaiser CA, Pauli V, Juge-Aubry CE Boss O, Pernin A, Chin WW, Cusin I, Rohner-Jeanrenaud F, Burger AG, Zapf J a nd Meier CA (1997) Direct effects of leptin on brown and white adipose tissue. J Clin Invest 100:2858-2864. Sierra-Honigmann MR, Nath AK, Murakami C, Garcia-Cardena G, Papapetropoulos A, Sessa WC, Madge LA, Schechner JS, Sc hwabb MB, Polverini PJ and FloresRiveros JR (1998) Biologi cal action of leptin as an angiogenic factor. Science 281:1683-1686. Sipols AJ, Baskin DG and Schwar tz MW (1995) Effect of in tracerebroventricular insulin infusion on diabetic hyperphagia and hypotha lamic neuropeptide gene expression. Diabetes 44:147-151. Spiegelman BM and Flier JS (2001) Obesity and the regulation of energy balance. Cell 104:531-543. Stanley BG, Kyrkouli SE, Lampert S and Leibowitz SF (1986) Neuropeptide Y chronically injected into the hypothalamu s: a powerful neurochemical inducer of hyperphagia and obesity. Peptides 7:1189-1192. Stellar E (1954) The physiology of motivation. Psychol Rev 61:5. Stryer L (1995) Biochemistry. W. H. Freeman and Co., New York. Sweeney G (2002) Leptin signalling. Cell Signal 14:655-663. Szanto I and Kahn CR (2000) Selective interac tion between leptin a nd insulin signaling pathways in a hepatic cell line. Proc Natl Acad Sci U S A 97:2355-2360. Takekoshi K, Motooka M, Isobe K, Nomura F, Manmoku T, Ishii K and Nakai T (1999) Leptin directly stimulates catecholamine s ecretion and synthesis in cultured porcine adrenal medullary chromaffin cells. Biochem Biophys Res Commun 261:426-431.

PAGE 162

145 Tartaglia LA, Dembski M, Weng X, Deng N, Culpepper J, Devos R, Richards GJ, Campfield LA, Clark FT, Deeds J Muir C, Sanker S, Moriarty A, Moore KJ, Smutko JS, Mays GG, Woolf EA, Sele nt-Munro C, and Tepper RI (1995) Identification and expression clonin g of a leptin receptor, OB-R. Cell 83:12631271. Thompson D, Edelsberg J, Kinsey KL and Oster G (1998) Estimated economic costs of obesity to U.S. business. Am J Health Promot 13:120-127. Toshinai K, Mondal MS, Nakazato M, Date Y, Murakami N, Kojima M, Kangawa K and Matsukura S (2001) Upregulation of Gh relin expression in the stomach upon fasting, insulininduced hypoglycem ia, and leptin administration. Biochem Biophys Res Commun 281:1220-1225. Travers JB, Travers SP and Norgren R (1987) Gustatory neural processing in the hindbrain. Annu Rev Neurosci 10:595-632. Troiano RP and Flegal KM (1999) Overweight prevalen ce among youth in the United States: why so many different numbers? Int J Obes Relat Metab Disord 23 Suppl 2:S22-27. Tschop M, Smiley DL and Heiman ML (2000) Ghrelin induces adi posity in rodents. Nature 407:908-913. Unger RH (2000) Leptin physiology: A second look. Regul Pept 92:87-95. Uotani S, Bjorbaek C, Tornoe J and Flier JS (1999) Functional properties of leptin receptor isoforms: internalization and de gradation of leptin and ligand-induced receptor downregulation. Diabetes 48:279-286. Vaisse C, Halaas JL, Horvath CM, Darnell JE, Jr., Stoffel M and Friedman JM (1996) Leptin activation of Stat3 in the hypothalamus of wild-type and ob/ob mice but not db/db mice. Nat Genet 14:95-97. Van Heek M, Compton DS, France CF, Tede sco RP, Fawzi AB, Graziano MP, Sybertz EJ, Strader CD and Davis HR, Jr. (1 997) Diet-induced obese mice develop peripheral, but not central, resistance to leptin. J Clin Invest 99:385-390. Vickers SP, Clifton PG, Dourish CT and Teco tt LH (1999) Reduced satiating effect of dfenfluramine in serotonin 5-HT(2C) receptor mutant mice. Psychopharmacology (Berl) 143:309-314. Wang Z, Zhou YT, Kakuma T, Lee Y, Kalra SP, Kalra PS, Pan W and Unger RH (2000) Leptin resistance of adipoc ytes in obesity: role of suppresso rs of cytokine signaling. Biochem Biophys Res Commun 277:20-26.

PAGE 163

146 Wang ZW, Zhou YT, Kakuma T, Lee Y, Higa M, Kalra SP, Dube MG, Kalra PS and Unger RH (1999) Comparing the hypothala mic and extrahypothalamic actions of endogenous hyperleptinemia. Proc Natl Acad Sci U S A 96:10373-10378. White DW and Tartaglia LA (1999) Evidence for ligand-independent homooligomerization of leptin receptor (OB-R) isoforms: a proposed mechanism permitting productive long-form signaling in the presence of excess short-form expression. J Cell Biochem 73:278-288. Williams G, Bing C, Cai X, Harrold JA, King P and Liu X (2001) The hypothlamus and the control of energy homeostasis: Di fferent circuits, different purposes. Physiol Behav 74:683-701. Wilsey JT, Zolotukhin, S., Prima, V., and Scar pace, P.J. (2003) Leptin resistance in dietinduced obese rats is associated with reduced leptin receptor expression and impaired signaling in the hypothalamus. Experimental Biology April 11-15: San Diego, California.:Abstract 652.653. Woods SC, Lotter EC, McKay LD a nd Porte D, Jr. (1979) Chronic intracerebroventricular infusion of insulin reduces food intake and body weight of baboons. Nature 282:503-505. Yamanaka A, Sakurai T, Katsumoto T, Yanagisawa M and Goto K (1999) Chronic intracerebroventricular administration of orex in-A to rats increases food intake in daytime, but has no effect on body weight. Brain Res 849:248-252. Yang Y, Nunes FA, Berencsi K, Furth EE, Gonczol E and Wilson JM (1994) Cellular immunity to viral antigens limits E1-d eleted adenoviruses for gene therapy. Proc Natl Acad Sci U S A 91:4407-4411. York D and Bouchard C (2000) How obesity develops: Insights from the new biology. Endocrine 13:143-154. York DA (1999) Peripheral and central mechanisms regulating food intake and macronutrient selection. Obes Surg 9:471-479. Zarjevski N, Cusin I, Vettor R, Rohner-J eanrenaud F and Jeanre naud B (1993) Chronic intracerebroventricular neuropeptide-Y ad ministration to normal rats mimics hormonal and metabolic changes of obesity. Endocrinology 133:1753-1758. Zhang Y, Proenca R, Maffei M, Barone M, Leopold L and Friedman JM (1994) Positional cloning of the mouse obese gene and its human homologue. Nature 372:425-432. Zhang Y, Wilsey, J., Frase, C., Matheny, M ., Bender, B., Zolotukhin, S., Prima, V. & Scarpace, P. J. (2002) Peripheral bu t not central leptin prevents the immunosuppression associated with hypoleptinemia in rats. J Endocrinol 174:455461.

PAGE 164

147 Zhao AZ, Huan JN, Gupta S, Pal R and Sahu A (2002) A phosphatidylinositol 3-kinase phosphodiesterase 3B-cyclic AMP pathway in hypothalamic action of leptin on feeding. Nat Neurosci 5:727-728. Zolotukhin S, Potter M, Zolotukhin I, Saka i Y, Loiler S, Fraites TJ, Jr., Chiodo VA, Phillipsberg T, Muzyczka N, Hauswirth WW, Flotte TR, Byrne BJ and Snyder RO (2002) Production and purification of sero type 1, 2, and 5 recombinant adenoassociated viral vectors. Methods 28:158-167.

PAGE 165

148 BIOGRAPHICAL SKETCH Jared Wilsey was born in Cooperstown, New York, on 28th of September, 1975, to Christine and Timothy Wilsey. Timothy and Ch ristine had one other son, Darren, 4 years earlier. Darren, who inherited the talent in the family, went on to become a successful musician. Jared spent most of his chil dhood and adolescence in the small town of Oneonta, New York. An avid athlete, Jared was a sprinter on the track team and a record setting running back on his high school footba ll team. It was not until his undergraduate years at Cornell University in Ithaca, New York, that he took a genuine interest in science. After receiving his B.Sc. (with distinction) in nutritional biochemistry, Jared entered the graduate program in exercise phys iology at the University of Florida. He reasoned that this would allow him to comb ine his new found love of science with his first love of athletics and human performan ce. Seeking new academic challenges, Jared entered the Interdisciplinary Program in Bi omedical Science at the UF College of Medicine in 1999 and, shortly thereafter, join ed the laboratory of Philip J. Scarpace to study the molecular basis of obesity. He spent several productive and rewarding years working with Dr. Scarpace, making several fr iends along the way. In the future, Jared hopes to have a successful career in biomedical research while still enjoying his athletic hobbies.


Permanent Link: http://ufdc.ufl.edu/UFE0000970/00001

Material Information

Title: Potent Anorexic and Lipopenic Effects of Central Leptin Gene Therapy Are Blocked by Diet-Induced Obesity: Evidence for Impaired Leptin Receptor Expression/Signal Transduction in Obesity and Reversal by Caloric Restriction
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0000970:00001

Permanent Link: http://ufdc.ufl.edu/UFE0000970/00001

Material Information

Title: Potent Anorexic and Lipopenic Effects of Central Leptin Gene Therapy Are Blocked by Diet-Induced Obesity: Evidence for Impaired Leptin Receptor Expression/Signal Transduction in Obesity and Reversal by Caloric Restriction
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0000970:00001


This item has the following downloads:


Full Text













POTENT ANOREXIC AND LIPOPENIC EFFECTS OF CENTRAL LEPTIN GENE
THERAPY ARE BLOCKED BY DIET-INDUCED OBESITY: EVIDENCE FOR
IMPAIRED LEPTIN RECEPTOR EXPRESSION/SIGNAL TRANSDUCTION IN
OBESITY AND REVERSAL BY CALORIC RESTRICTION















By

JARED TIMOTHY WILSEY


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2003

































Copyright 2003

by

Jared Timothy Wilsey

































This dissertation is dedicated my mother, Christine.















ACKNOWLEDGMENTS

The research described in this dissertation was made possible through the help and

support of University of Florida faculty, fellow graduate students, post-docs, scientists,

and other collaborators of the Philip J. Scarpace laboratory. First, I would like to thank

Dr. Scarpace for both allowing me to conduct a rotation project in his lab in the fall of

1999, and then inviting me to join his laboratory group in May 2000. Dr. Scarpace has

been a pleasure to work for. He has encouraged independence, which has greatly

enhanced my development as a student and scientist. Whenever needed, his advice and

direction on experimental design and the development of novel research hypotheses has

been invaluable. In addition, his counsel has enhanced my ability to both prepare

manuscripts and, more importantly, to appropriately respond to reviewers.

All members of Dr. Scarpace's lab, both past and present, have contributed to my

education and these projects. I thank Michael Matheny, who provided invaluable

technical advice on a daily basis despite his own busy research schedule. Mike has also

been a lot of fun to work with, and we are all lucky to have him around. Dr. Yi "Edi"

Zhang has been enormously helpful with both guidance on scientific issues and with

helping to keep me caught up on the literature. My fellow graduate student, Dr. Gang Li,

has likewise given me some great advice over the years and has calmed me down on

numerous occasions when I was convinced I mined an assay or experiment. All of Dr.

Scarpace's past technicians have been of assistance as well including "Thomas" Kit-Yan









Tang (the best rat weigher in the business), Min Li, and Constanza Diana Victoria Frase

(yes, that is all one name).

Several faculty members have helped me over the years. Dr. Nihal Tumer, who

frequently collaborates with Dr. Scarpace, has been a great positive influence in the

laboratories and is a truly kind woman. She also makes a deliciously potent cup of

Turkish coffee. I thank my other committee members for their great suggestions and

comments over the last couple of years, including Dr. Charles Wood, Dr. Daniel Driscoll,

and Dr. Steven Borst. Dr. Wood, along with Dr. David Weiner in the VA Hospital, were

very helpful in getting me started on real-time PCR when I had to quickly repeat all of

my expression data. Dr. Borst was one of the first College of Medicine Faculty I met,

way back in 1998 he gave me a lesson on rat brain excision and anatomy. Dr. Driscoll

has enlightened me as to the clinical significance of obesity research, particularly in the

treatment of the severe, early-onset obesity caused by specific genetic defects. He

explained that in such cases, gene delivery-even into the central nervous system-may

prove to be a reasonable strategy. Dr. Scarpace's collaboration with Dr. Sergei

Zolotukhin and Dr. Victor Prima was of critical importance in all of the gene therapy

experiments. I thank them both for their generous provision of vectors and related

materials, as well as their intellectual contributions to these projects.

Most importantly, I would like to thank my friends and family for giving me moral

support and making me laugh over the years. I thank my mom and brother Darren for

being sources of first-class advice and positive words. My mother, especially, has been

there for me through it all: track meets, football games, 40 yard dash time trials ad

nauseum, weightlifting and powerlifting competitions-she really is a Super-mom and a









great friend. I also thank my dad, my grandparents, aunts and uncles, and my little half-

sisters. I thank Amie Dirks, my best friend during the graduate school years and a

wonderful person. I thank my "industry contact" and friend, Mike Ferguson. Last but

not least, I thank all my friends and training partners from Gainesville Gym, including

Anthony (an ever-aspiring Olympian) and Josh.
















TABLE OF CONTENTS
page

A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES ....................................................... ............ .............. .. xii

L IST O F FIG U R E S .............. ............................ ............. ........... ... ........ xiii

ABSTRACT ........ .............. ............. ...... ...................... xvi

CHAPTER

1 BACKGROUND AND SIGNIFICANCE...................................................................

L e p tin ...................................... ...................................... ................. .
The Leptin R eceptor ....................................... ................ .. ........ .. ..
Isoform s ................... ... ..... ... ................................ .......... .. ...... 3
Leptin Binding to Leptin Receptor.......... .......... ......................... ............... 4
Leptin Receptor Signal Transduction ............................................ .. ............. 5
JAK/STAT Pathway ........................................ ............ .. ...... ................ .5
E R K -c-fos Pathw ay .............. ............. .......... ............................ .............
PIA 3 K inase and Insulin-Like Signaling.................................... .....................6
Negative Regulators of Leptin Signaling ...........................................................
Down-Stream Leptin Signaling and Neuropeptide Regulation in the Hypothalamus..9
A rcuate N ucleu s ....................................................... 9
Projections from the Arcuate.. ... ............................................................ ............... 12
Extra-Hypothalamic Signaling and Energy Balance ..........................................13
Other Molecules Regulating Energy Balance..........................................................14
Insulin ............ ................... ... ................................. ......... 14
G lu c o se .....................................................................................1 6
Orexins/Hypocretins ............... .......... .. ......................... .... ............... 16
Ghrelin .......................................................17
U ncoupling Proteins ............................. ........... .......... ...... .. ... 18
L eptin R esistance................................ .. ......... ................................................. 20
Adeno-Associated Virus and Gene Therapy ................................... .................21
W ild-Type A A V ................................... .... ............. .... ...... ...... 21
Recombinant Adeno-Associated Virus and Gene Delivery .............................23
Experim ental Design and Rational ....................................... ........................ 24
Incorporating Regulation into Leptin Gene Therapy .......................................24
Central Leptin Over-Expression in Diet-Induced Obese Animals....................25









Effects of Diet-Induced Obesity and Caloric Restriction on Leptin Receptor
Expression and Signal Transduction Capacity .............................................26
Chapter Sum m ary and Conclusions....................................... ......................... 26

2 GENERAL METHODS AND MATERIALS................................. .....................28

E xperim mental A nim als ............... ...................................................... .....................28
Construction of rAAV Vector Plasmid .............................................. ...............28
P ackaging of rA A V V ectors.......................................................................... ...... 29
Stereotaxic Injections .................. ................................................ 29
Third C erebroventricle ............................................... ............................. 29
H y p oth alam ic ............................................................... 3 0
O oxygen C onsum option ................................................................. .. ............... 30
Tissue Harvesting ................................ ...... .......................... ........ 31
Serum M easurem ents .................. .............................. ...... .. .......... .... 31
Leptin ................... .... ................... ........................ 31
In sulin ............. ........................................................................ 3 1
Glucose ............... ......... ........................31
F ree F atty A cid s .............................................................3 1
C SF L eptin................................................... 32
Probes .............. ............................................ ........ 32
RNA Isolation and RNA Dot Blot................................... ...... ........ 33
Relative-Quantitative RT-PCR Using 18S Competimers ................ .......... 33
Real-time RT-PCR for Leptin Receptor ...................................................... 34
STAT3/Phospo-STAT3 Assay ..................................................35
UCP-1 Protein in Brown Adipose Tissue ................. .......................... 36
Statistical A nalysis................................................... 36

3 INCORPORATING REGULATION INTO LEPTIN GENE THERAPY ................38

In tro d u ctio n ........................................................................................................... 3 8
M eth od s an d M materials ......................................................................................... 39
A n im a ls .......................................................................................................... 3 9
Construction of rAAV Vector Plasmid ........................................................40
Packaging of rA A V vectors .............................................................. ... 41
V ector A dm inistration ............................................................. ..............41
Third ventricle inj section ............................................................ ........ 41
Hypothalamic injection ................................. .................. ........ ... 46
E x p erim ental D esign ..................................................................................... 4 6
E x p erim ent 1 ............................................................4 6
E x p erim ent 2 ............................................................4 7
Tissue Harvesting .................................................................. ........ 47
Serum Leptin, FFA, Insulin, and Glucose .......................................................... 47
Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR) .......................48
STA T3/Phospo-STA T3 A ssay ................................................... .....................48
Probes ................. ........................................... .... .............. 49
mRNA Levels (Dot Blot Analysis)............................... ......................... 49









Statistical A analysis ...................................... .............................50
R e su lts ...........................................................................................5 1
Experim ent 1 ........................................51
Food consum ption and body m ass ........................................ ............... 51
Serum leptin and adiposity .................................... ............. ............... 51
Experim ent 2 ........................................52
Food consumption and body mass .................................... ............... 52
Serum leptin and adiposity ..................................... ............ ............... 53
Serum free fatty acids, insulin, and glucose.............................................. 53
L eptin expression ........................................................... .. ...... .. 53
Signal transduction in hypothalamus ................................. ............... 54
Leptin receptor expression in the hypothalamus ................ ... ............ 54
C SF leptin ................................................ ........ ..... 54
B row n adipose tissue.......... ............................ ........ .......... ............. 55
D iscu ssio n ........................ ... ............ ... ..........................................5 5

4 CENTRAL LEPTIN GENE THERAPY FAILS TO OVERCOME THE LEPTIN
RESISTANCE ASSOCIATED WITH DIET-INDUCED OBESITY .....................72

In tro du ctio n ...................................... ................................................ 7 2
M methods and M materials ............................. .................................. .... ...... ...... 73
A n im a ls .......................................................................................................... 7 3
E x p erim ental D esign ........................................ ............................................73
B lood C collection ......... ................................................................ ..... .... .. 74
O xygen C onsum ption ..................................................................... ............... 75
Construction of rAAV Vector Plasmid ............................................................75
Packaging of rAAV Vectors..................................................... 75
V ector A dm inistration ............................. ............................. ............. .76
Tissue H arvesting ........ .......... ............... ................... .... 76
Serum Leptin and FFA ....................... ..... .................................. ............... 77
Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR) ........................77
STA T 3/Phospo-STA T 3 A ssay ................................................. .....................78
P rob es .............................. ................................................. 7 8
mRNA Levels (D ot Blot Analysis) .......................................... ............... 79
Statistical A analysis .......................... .......... ............... .... ..... .. 79
R results .......... ....... .. ........ ................................80
Part I: High-Fat Feeding ........... ......... ............... ... ............... 80
Food consum ption and body m ass ........................................ ............... 80
Oxygen consumption....................... ..... ............................. 81
Serum leptin and free fatty acids....................................... ............... 81
Part II: Post rA A V -Leptin D delivery ........................................ .....................83
Food consum ption and body m ass ........................................ ............... 83
O xygen consum ption......................................................... ............... 84
A diposity ................................. ........... ............... ............ 85
Serum leptin and free fatty acids....................................... ............... 85
Leptin transgene expression in the hypothalamus.......................................86
Leptin receptor expression and signal transduction in the hypothalamus....86









Downstream neuropeptide regulation in the hypothalamus .........................86
UCP-1 in brown adipose tissue ..... ...................... ..............87
D iscu ssion ......... ........... .......... ................ ............................88

5 CALORIC RESTRICTION REVERSES IMPAIRMENTS IN LEPTIN RECEPTOR
EXPRESSION AND MAXIMAL SIGNALING CAPACITY IN DIET-INDUCED
O B E SE A N IM A L S ....................................................................... ..................... 10 1

Introduction ...................................... ............................... ......... ...... 10 1
M methods and M materials ........................................... ....................................... 102
A n im a ls .................................................... ................ 1 0 2
E xperim ental D esign .................................................................. ................ ... 103
Experimental design 1: leptin signaling in diet-induced obese................03
Experimental design 2: leptin signaling following caloric restriction .......104
L eptin A dm inistration ............................................... ............................ 104
Tissue H harvesting .................. .......................... .... .... ... ........ .... 105
Real-Tim e PCR .................................. .. ........ ...............105
STAT3/Phospo-STAT3 Assay ..... ........................ ...............106
Leptin mRNA Levels in W hite Adipose Tissue...............................................107
S erum L eptin ...............................................................10 7
Statistical A naly sis .......................... ...... ................ ............ .. .............. 107
R results ................... ......... ......... ..... ..................... 108
Food Intake and Body Weight..................................................108
Adiposity ...................... .......... ................ 109
Leptin Receptor Expression in the Hypothalamus ................. ... ............. 109
Hypothalamic STAT3 Phosphorylation ................................ .................110
Leptin Expression in W hite Adipose Tissue ..................................................111
S erum L eptin ................................................................... ................ 111
D iscu ssion ................................................................................................ ..... 1 16

6 GENERAL DISCUSSION AND CONCLUSIONS .............................................123

M aj o r F in d in g s ................ ........ ... .......... ........................................... 12 3
Physiological and Biochemical Effects of Leptin Gene Therapy are
R eversible .................................. .... ............ ........ ......... 123
Leptin May Regulate Leptin Receptor Gene Expression...............................124
There is a Central Nervous System Component to Leptin Resistance............125
Diet-Induced Obese Animals Have Impaired Leptin Signaling Capacity.........126
Impairments in Leptin Receptor Expression and Signaling in Diet-Induced
Obese are Reversed by Caloric Restriction .............. .................................. 126
N ew Insights on Leptin R esistance.................................... .................................... 127
Evidence for Role of Leptin Signal Transduction........................................127
Evidence Against Major Role for Blood:Brain Barrier................................... 127
Evidence for Down-Stream Contributors to Leptin Resistance ......................128
Conclusion ..................................... ................................. ......... 130









L IST O F R E FE R E N C E S ......................................................................... ................... 132

BIOGRAPHICAL SKETCH ............................................................. ..................148
















LIST OF TABLES


Table pge

3-1 Brown adipose tissue parameters, serum free fatty acids, insulin, and glucose,
and C SF leptin at sacrifice ............................................... ............................ 65

4-1 Serum leptin and free fatty acids after 75 days of high-fat feeding
(DIO and DR).......................... ..... .............. ..................... 83

4-2 Oxygen consumption and 24 hour caloric intake on day 7 following rAAV-leptin
or control v ector deliv ery .............................................................. .....................90

4-3 Serum leptin and FFA at sacrifice............. ..................... ....................92

4-4 Hypothalamic POMC NPY, AGRP, and SOCS3 expression at sacrifice .............94
















LIST OF FIGURES


Figure page

1-1 M odel of leptin receptor signal transduction. ........................................ ...............7

3-1 pTR-3ObW pTR-3ObW encodes rat leptin cDNA ............................ ..............42

3-2 The TET-responsive promotor/leptin plasmid construct pTR-tetR-Ob (top), and
the assessory plasmid construct pTR-rtTA/tTS (bottom). ...................................43

3-3 Schematic representation of leptin transgene regulation by doxycycline in dual
vector "TET-Ob" system, "OFF" state. ...................................... ...............44

3-4 Schematic representation of leptin transgene regulation by doxycycline in dual
vector "TET-Ob" system, "ON" state........ ......................................... 45

3-5 Body mass following intracerebroventricular administration of rAAV-leptin......56

3-6 Food intake following intracerebroventricular administration of rAAV-leptin ....56

3-7 Disappearance of visceral adipose tissue in 4 month old F344xBN rats
adm inistered rA A V -leptin. ............................................ ............................ 57

3-8 Leptin is undetectable in serum of animals following central leptin gene
d eliv ery ............................................................................ 5 7

3-9 Experiment 2 design, TET-Ob gene delivery and regulation. ............................58

3-10 Body mass following TET-Ob or control vector delivery..................................59

3-11 Daily food consumption following TET-Ob or control vector delivery................59

3-12 Serum leptin following TET-Ob or control vector delivery................................60

3-13 Visceral adiposity following TET-Ob or control vector delivery........................60

3-14 Hypothalamic leptin expression 66 days after TET-Ob or control vector
delivery ............. ....................... ........... ...... ........ ........ 61

3-15 Hypothalamic P-STAT3 66 days following TET-Ob or control vector delivery.
Values represent means + SEM (Bar Graph, Top). ............. .............. 62









3-16 POMC expression 66 days following TET-Ob or control vector delivery............63

3-17 SOCS3 expression 66 days following TET-Ob or control vector delivery. .........63

3-18 Long-form leptin receptor (Ob-Rb) expression in the hypothalamus 66 days
following TET-Ob or control vector delivery............................................ 64

3-19 UCP-1 protein in BAT 66 days following TET-Ob or control vector delivery.....64

4-1 Body mass during high fat feeding (pre-vector delivery)................................82

4-2 Caloric intake during high fat feeding (pre-vector delivery). ............................82

4-3 Oxygen consumption on day 30 and on day 70 after commencing HF-feeding
(pre-vector delivery). ....................................... ...........................83

4-4 Changes in body mass during 29 days post-vector delivery ..................................89

4-5 Caloric intake during 29 days post-vector delivery. ...................... ...............89

4-6 Visceral adiposity 30 days post-vector delivery. ................................................90

4-7 Leptin transgene expression in the hypothalamus 30 days post-vector
deliv ery ................................................................................ 9 1

4-8 Hypothalamic leptin receptor expression 30 days post-vector delivery ...............92

4-9 STAT3 phosphorylation in the hypothalamus at sacrifice................................93

4-10 UCP-1 concentration in BAT at sacrifice. ...................... .... ...............94

4-11 UCP-1 per total interscapular brown adipose tissue (BAT) at sacrifice .............95

5-1 Body mass during high fat feeding. .................................................................112

5-2 Body m ass during calorie restriction. .. ............ .......................... ...............112

5-3 V isceral adiposity at sacrifice. ................................................................. ....... 113

5-4 Ob-Rb expression in the hypothalamus ....................................... .................113

5-5 Maximal leptin-induced STAT3 phosphorylation capacity is reduced in DIO
and increased by caloric restriction. ............. .................. .......... .............. 114

5-6 Leptin expression in retroperitoneal white adipose tissue (RTWAT) at
sacrifice .............................................................................................. 115









5-7 Serum leptin at sacrifice.......................................................................... ....... 115

5-8 Serum leptin corrected for leakage of infused leptin from the CSF ....................116















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

POTENT ANOREXIC AND LIPOPENIC EFFECTS OF CENTRAL LEPTIN GENE
THERAPY ARE BLOCKED BY DIET-INDUCED OBESITY: EVIDENCE FOR
IMPAIRED LEPTIN RECEPTOR EXPRESSION/SIGNAL TRANSDUCTION IN
OBESITY AND REVERSAL BY CALORIC RESTRICTION

By

Jared Timothy Wilsey

August 2003

Chair: Philip J. Scarpace
Major Department: Pharmacology and Therapeutics

Obesity is estimated to result in 300,000 deaths each year in the United States

alone. The purpose of this dissertation was to study the molecular basis of obesity.

Specifically, we studied how the brain's response to the hormone "leptin" is impaired in

obese animals.

First, we demonstrated that young adult F344XBN rats are highly responsive to a

single intracerebroventricular injection of adeno-associated virus encoding leptin (rAAV-

leptin). Within 1 week of leptin gene delivery, animals exhibited a significant anorectic

response iithnit the concomitant decrease in resting energy expenditure that is normally

recorded during anorexia. This anorectic response persisted for more than two months.

As a result of this negative shift in energy balance, rAAV-leptin treated animals lost

significant weight with respect to controls. More impressively, there was a near complete

disappearance of visceral white fat following rAAV-leptin treatment.









We then incorporated regulation into rAAV-leptin by placing leptin under the

control of the tetracycline transactivator (rtTA) and operon (tetR). Using this system, we

showed that all of the physiological and biochemical responses to leptin gene therapy are

completely reversible. We also demonstrated that central leptin may negatively regulate

the expression of leptin receptor in the hypothalamus, and that this is reversed upon

transgene inactivation. This was the first ever experiment to use an externally regulated

transgene to control energy balance.

Next, we showed that diet-induced obese (DIO) animals are non-responsive to the

anorectic, lipopenic, and biochemical effects of rAAV-leptin whereas high-fat fed

animals that did not become obese (diet resistant) retained leptin responsiveness. It was

found that the leptin resistance in the DIO animals was associated with reduced

hypothalamic leptin receptor expression along with an absence of signal transduction

response to rAAV-leptin. DIO animals were then tested with an acute supramaximal

bolus of i.c.v. leptin, and were found to have reduced leptin receptor signal transduction

capacity compared to non-obese controls. Finally, the deficits in leptin receptor

expression and signaling capacity in DIO were completely reversed by 30 days of caloric

restriction. Thus, short-term calorie restriction may be a viable strategy to restore leptin

sensitivity in previously leptin-resistant models of acquired obesity.














CHAPTER 1
BACKGROUND AND SIGNIFICANCE

One in three Americans is obese (Spiegelman and Flier, 2001). It is estimated that

obesity results in 300,000 deaths per year in the Unites States (Kopelman, 2000), and the

World Health Organization now recognizes obesity as one of the top 10 global health

problems (Kelner and Helmuth, 2003). Obesity promotes diabetes, hypertension,

cardiovascular disease, stroke, and cancer (Khaodhiar et al., 1999). Being overweight

can decrease quality of life by triggering osteoarthritis, joint pain and degeneration, gout,

and sleep apnea (Scapinelli, 1975; Khaodhiar et al., 1999). In excess of 5% of U.S.

national health expenditures are directed at costs associated with obesity (Khaodhiar et

al., 1999; Thompson et al. 1998). Moreover, the incidence of obesity in children has

escalated dramatically (Troiano and Flegal, 1999), foreshadowing greater harm to come.

Leptin

Leptin, the product of the Ob gene cloned in 1994 (Zhang et al., 1994), is a 16 KD

peptide hormone secreted primarily from white adipose tissue (WAT). When leptin was

found to rapidly reverse obesity in the ob/ob mutant mouse, excitement soared in the

scientific community and the popular press alike for what could be a cure for the ever

growing obesity epidemic in Western societies. Before research began to accumulate

suggesting otherwise, it was even referred to as the "anti-obesity hormone" (Elmquist et

al., 1998). Leptin acts on satiety centers in the hypothalamus to both decrease food intake

and increase energy expenditure (Ahima and Flier, 2000b). These effects can combine

to cause dramatic lipopenia, or loss of adipose tissue, in many animal models. In some









cases, a complete disappearance of visible adipose tissue can be achieved following

pharmacological leptin infusion or Ob gene delivery (Chen et al., 1996; Halaas et al.,

1997; Scarpace et al., 2002b).

There are documented cases of leptin deficiency leading to obesity in humans,

analogous to the ob/ob mouse. One such case is described below.

... a 200-pound 9-year-old English girl, whose legs were so large she could barely
walk, was found to lack the weight-regulating hormone leptin. Treatment with
leptin dramatically reduced her food intake, and that of her similarly affected
cousin, to the point where they both now have body weights within the normal
range for their age and live normal lives. Before leptin treatment, the younger child
consumed in excess of 1100 calories at a single meal, which is approximately half
the average daily intake of an adult. With only a few leptin injections, this was
reduced by 84% to 180 calories, the typical intake of a normal child (Friedman,
2003, page 856).

Unfortunately, such a miraculous transformation following leptin treatment is

atypical. Common human obesity is a more complex disease than the monogenic leptin

deficiency described above. The prevailing obesity is believed to involve multiple genes

that participate in the intricate, and often redundant, feeding and metabolic pathways-

systems that are essential for survival. This complexity has greatly impeded the search

for safe and effective treatments for obesity, a search that has been largely unsuccessful

to date (Gura 2003). Most regrettably, human obesity is normally accompanied by high

levels of endogenous leptin and resistance to the weight-reducing effects of the hormone.

Leptin resistance will be discussed in detail later in this chapter as well as Chapters 4-6 of

this dissertation.

Leptin has diverse biological functions in addition to its role in energy balance.

These include roles in reproduction, hematopoiesis, angiogenesis, immune

responsiveness, blood pressure regulation, and bone formation (Fruhbeck et al., 2001).

Leptin stimulates the differentiation of macrophages and the production of cytokines by









macrophages (Gainsford et al., 1996). Leptin accelerates the onset of puberty in wild-

type rodents and appears to increase reproductive behavior (Ahima and Flier, 2000b). In

fact, leptin may be the signal that acts on the gonadotrophin releasing hormone system in

the hypothalamus at the time of puberty (Elmquist et al., 1998). Leptin stimulates the

growth of blood vessels and increases the rate of wound healing (Sierra-Honigmann et

al., 1998), and yet inhibits bone formation (Ducy et al., 2000). The effect of

pharmacogical leptin on these systems likely depends on the route of administration. For

example, chronic central delivery of the hormone results in low peripheral leptin. Thus,

direct administration of leptin into the central nervous system is expected to inhibit any

direct actions of leptin in the periphery.

The Leptin Receptor

Isoforms

The leptin receptor was first cloned in 1995 (Tartaglia et al., 1995). It is a member

of the class I cytokine receptor family that includes the IL-6, IL-11, IL-12, LIF, and

CNTF receptors (Baumann et al., 1996). The absence of a functional leptin receptor (as

in the db/db mouse) results in obesity and diabetes-like symptoms. Alternative splicing

of a single gene yields at least six isoforms of the leptin receptor, named Ob-Ra, Ob-Rb,

Ob-Rc, Ob-Rd, Ob-Re, and Ob-Rf. Ob-Rb is the "long form" of the leptin receptor and

is found primarily in the brain, although limited Ob-Rb expression has been detected

elsewhere including the adrenal gland (Takekoshi et al., 1999), the intestine (Morton et

al., 1998), and both brown and white adipose tissue (Siegrist-Kaiser et al., 1997). Ob-Rb

is the only form of the leptin receptor believed to be capable of full intracellular signaling

capacity (Spiegelman and Flier, 2001). Ob-Re is a soluble form of the leptin receptor

and may negatively regulate the bioavailability of circulating leptin by acting as a binding









protein (Brabant et al., 2000; Fruhbeck et al., 2001). The remaining isoforms (Ob-Ra,

Ob-Rc, Ob-Rd, and ObRf) have shortened intracellular domains. Ob-Ra appears to be

involved in transport across the blood-brain barrier (Kastin and Pan, 2000).

The intracellular domain of all leptin receptor isoforms contains an identical 29

amino acid "Box 1" Janus-family kinase (JAK) binding domain in the juxtamembrane

region. Only the long-form receptor Ob-Rb has a "Box 2" motif with signal transducer

and activator of transcription (STAT) binding sites. The intracellular domain of Ob-Rb

contains approximately 306 amino acids whereas the intracellular domain of the

shortened forms range from 32-40 amino acids (Tartaglia et al., 1995).

Leptin Binding to Leptin Receptor

Both Ob-Rb and short forms of the leptin receptor exist as homodimers in the

absence of ligand, and the formation of homodimers does not seem to be affected by the

presence of leptin (Nakashima et al., 1997; Sweeney, 2002; White and Tartaglia, 1999).

Each leptin receptor in a homodimer unit binds leptin in a 1:1 stoichiometry, hence

forming tetrameric complexes with 2 leptin molecules bound by the receptor dimer

(Devos et al., 1997). A conformational change in the receptor takes place upon the

formation of this tetramer that is a critical first step in leptin signaling (Fong et al., 1998).

It appears that these leptin-Ob-Rb complexes are internalized subsequent to ligand

binding via clatherin-coated vesicles (Sweeney, 2002; Lundin et al., 2000). Similar to

other members of the cytokine receptor family, it is believed that internalized receptors

are targeted for degradation or recycled back to the cell surface (Sweeney, 2002). It has

been recently estimated that 25% or less of leptin receptors are located at the cell surface,

with the majority sequestered in intracellular pools (Barr et al., 1999). While leptin

exposure may promote receptor internalization, it is presently unclear what other factors









may influence transport of receptors between intracellular compartments and the cell

surface.

Leptin Receptor Signal Transduction

JAK/STAT Pathway

Like all members of the class I cytokine receptor family, Ob-Rb operates through a

JAK/STAT signal transduction pathway (Vaisse et al., 1996). Activated Ob-Rb can

phosphorylate STAT-3, 5, and 6 (Baumann et al., 1996; Ghilardi et al., 1996), but most

research to date has focused on the STAT-3 pathway (Spiegelman and Flier, 2001). The

naked leptin receptor does not have intrinsic tyrosine kinase activity, so STAT signaling

is dependent upon Ob-Rb association with kinases such as JAK2 (Figure 1-1) (Vaisse et

al., 1996). Docking of JAK2 occurs following ligand binding and formation of the

leptin:receptor tetramer discussed above (Morton et al., 1999). JAK2 then activates the

Ob-Rb by phosphorylation of Tyr 985 and Tyr 1138, the latter of which is critical for the

recruitment and activation of STAT3 (see below). Once Tyr 1138 on Ob-Rb is

phophorylated, STAT3 is recruited to the receptor via SH2 domains where it is tyrosine

phophorylated by the receptor-associated JAK2 (Sweeney, 2002). Activated STAT3 (P-

STAT3) then dissociates from the receptor and forms homo- or heterodimers. These

activated STAT dimers travel to the nucleus where they act as transcription factors

(Sweeney, 2002).

While other signaling pathways are employed by the leptin receptor (discussed

below), the JAK/STAT pathway is required for leptin's effects on energy balance and

bodyweight regulation. Targeted disruption of Tyr 1138, the critical residue for STAT3

signaling by the leptin receptor, results in both hyperphagia and morbid obesity. This is

despite the fact that other leptin receptor-mediated signaling pathways appear to remain









intact (Bates et al., 2003). STAT3 independent signaling, on the other hand, seems to

play an important role in reproduction, linear growth, and glucose homeostasis (Bates et

al., 2003). A simplified diagram of Ob-Rb signal transduction, including JAK-STAT

signaling, is presented in Figure 1-1.

ERK-c-fos Pathway

Leptin has been shown to activate a signaling cascade via extracellular regulated

kinase (ERK) (Figure 1-1) (Kim et al., 2000b). ERKs are serine/threonine kinases and

belong to the MAPK family. Activation of the ERK pathway is apparently mediated by

the phosphorylation of Tyr 985 on the leptin receptor by JAK2 (see above), although

there may also be an ERK pathway independent of Tyr 985 (Bjorbaek et al., 2001). The

physiological significance of the ERC-c-fos pathway with respect to leptin signaling is

still under investigation, but it appears to regulate the expression of a completely different

set of genes than the STAT3 pathway (Figure 1-1) (Sweeney, 2002).

PIA3 Kinase and Insulin-Like Signaling

Furthermore, leptin can act through some components of the insulin signal

transduction pathway (Figure 1-1) (Kim et al., 2000b). Via JAK2, leptin can activate

insulin-receptor substrates IRS-1 and IRS-2 which, in turn, can activate PI3-kinase

(PI3K) (Figure 1-1) (Szanto and Kahn, 2000). Activation of PI3K occurs when SH2

domains of the regulatory subunit bind to phosphorylated tyrosine residues of IRS

molecules. PI3K then catalyzes the phosphorylation of phosphatidylinositol (4,5)

bisphosphate (PIP2) to phosphatidylinositol (3,4,5) trisphosphate (PIP3). Subsequently,

PIP3 activates a variety of downstream signaling molecules that contain pleckstrin









LepinAcie Signallig Pahwy


PTP1B--*


Membrane


Y1PTPase 8 Y113 NPY ARP
I SOCS-3


ERK1/2 PIAS3



1
Gene Expression POMC

Figure 1-1. Model of leptin receptor signal transduction.

homology domains, which recognize PIP3 (Niswender and Schwartz, 2003). Some

examples of molecules activated by PIP3 are serine-threonine kinases, tyrosine kinases,

and GTPases (Shepherd et al., 1998). Perhaps the most studied downstream target of

IRS-induced PIP3 is protein kinase B (PKB), also known as Akt (Niswender and

Schwartz, 2003). Based on knowledge of insulin signal transduction, it is believed that

PIP3 activates PKB via a phosphatidyl-inositol dependent kinase (PDK) (Shepherd et al.,

1998). PKB then activates phosphodiesterase 3B (PDE3B) which, in turn, decreases

intracellular cAMP. It has been demonstrated that this decrease in cAMP levels may

enhance leptin-dependent STAT3 phosphorylation (Niswender et al., 2001; Zhao et al.,









2002). This PDE3B-mediated decrease in cAMP may also be a critical step in the leptin-

dependent suppression of NPY (Akabayashi et al., 1994).

Activation of this IRS-PI3K signaling cascade is required for several physiological

effects normally associated with insulin, including glucose transporter (GLUT)

recruitment to the cell membrane and glycogenesis (Kanai et al., 1993). Because of

overlap between insulin and leptin signaling, leptin can modify certain insulin-induced

changes in gene expression in vivo (Kim et al., 2000b; Szanto and Kahn, 2000).

Negative Regulators of Leptin Signaling

SOCS3. Leptin signaling activates negative-feedback messages within the cell.

The most studied inhibitory signal is suppressor ofcytokine signaling-3 (SOCS-3), which

is so reliably activated by leptin that it serves as a marker for cells responding to the

hormone (Bjorbaek et al., 1998; Bjorbaek et al., 1999). It is believed that the

transcription factor P-STAT3, activated by leptin, directly upregulates SOCS-3

expression (Niswender and Schwartz, 2003). SOCS-3 inhibits leptin signaling in part by

suppressing the activity of JAK kinase (Bjorbaek et al., 1999). SOCS-3 is also a negative

regulator of insulin signaling (Fruhbeck et al., 2001). Thus, SOCS-3 induction in a

hyperleptinemic state could be one mechanism linking obesity and insulin resistance (see

discussion of leptin resistance below).

PIAS. Another negative regulator of leptin signaling is protein inhibitor of

activated STATS (PIAS) (Liu et al., 1998). This family of peptides was discovered by a

yeast two-hybrid screen for molecules that interact with STAT. PIAS proteins can inhibit

STAT-induced transcriptional activation, presumably by binding to activated STAT

dimers and blocking their passage into the nucleus (Greenhalgh and Hilton, 2001).

However, unlike SOCS-3, expression of PIAS family members does not appear to be up-









regulated by cytokine signaling, including leptin signaling (unpublished observations by

Philip Scarpace's Laboratory). Rather, PIAS family members (specifically PIAS-1 and

PIAS-3) are constitutively expressed in many cell types (Greenhalgh and Hilton, 2001).

Although not proven at this time, it is possible the PIAS activity increases in response to

leptin signaling, thus allowing it to act as a negative feedback effector.

Down-Stream Leptin Signaling and Neuropeptide Regulation in the Hypothalamus

Leptin signaling in the hypothalamus is the first step in a complex pathway

involving multiple neuropeptides and various intra-hypothalamic as well as extra-

hypothalamic centers involved in the regulation of energy balance. This section will

discuss both the neuroanatomy of this pathway within the hypothalamus and the key

neuropeptides and neurotransmitters that participate.

Arcuate Nucleus

The arcuate nucleus (ARC) is a critical area for leptin signaling with respect to the

regulation of energy balance. The ARC sits just above the median eminence, at the base

of the third ventricle (Williams et al., 2001). Leptin acts on two distinct populations of

neurons in the arcuate nucleus via the Ob-Rb receptor (as described above). One

population of neurons co-expresses the orexigenic (appetite-stimulating) peptides NPY

and AGRP, and leptin signaling reduces their expression (York and Bouchard, 2000).

The second population co-expresses the anorectic peptides a-MSH (derived from POMC

processing) and CART, and leptin up-regulates their expression (Elias et al., 1999; York,

1999).

a-MSH and AGRP. a-MSH and AGRP are the endogenous agonist and

antagonist, respectively, for a common receptor, the melanocortin 4 receptor (MC4R),









which is expressed primarily in the brain (Cone, 1999). Activation of MC4R by a-MSH

reduces food intake while suppression of MC4R activity by AGRP or a pharmacological

antagonist increases food intake and diminishes the response to leptin (Fan et al., 1997).

Signaling through the MC4R can also increase energy expenditure (Pierroz et al., 2002).

Another melanocortin receptor, MC3R, is also implicated in energy homeostasis, but the

MC4R appears to be the most critical for the hypophagic action of a-MSH and its

analogues whereas both receptors may be involved in regulating energy expenditure

(Williams et al., 2001). The mechanism of enhanced metabolic rate following the

delivery of an MC3/4 agonist appears to involve both enhanced uncoupling protein

expression in the brown adipose tissue and in increased capacity of skeletal muscle to

oxidize fatty acids (Shek et al., 2002).

NPY. NPY is a particularly potent stimulator of feeding in the hypothalamus.

Acute NPY injections into the brain can increase food intake several fold and,

surprisingly, chronic NPY infusion continues to stimulate feeding (Stanley et al., 1986).

Thus, NPY can apparently override short-term and long-term satiety signals. NPY can

also decrease energy expenditure. One mechanism behind this is an NPY-induced

reduction in thermogenesis in brown adipose tissue and, presumably, a reduction in

uncoupling protein expression and activity (Zarjevski et al., 1993). The activity of

NPY/AGRP neurons in the ARC is increased during fasting and in states of negative

energy balance in an apparent homeostatic attempt to limit or reverse the loss of body fat

stores (Williams et al., 2001).

CART. Cocaine- and amphetamine-regulated transcript (CART) is an anorectic

neuropeptide expressed by first-order leptin responsive neurons in the ARC. Alternative









splicing of the CART mRNA results in two peptide products consisting of 116 and 129

amino acids in the rat (Douglass et al., 1995). These pro-peptides are then tissue-

specifically processed into smaller peptides and packaged into vesicles as

neurotransmitters (Hillebrand et al., 2002). Several CART peptide fragments have been

identified in the hypothalamus and other parts of the brain, and many of these have

proven to be biologically active (Kuhar and Dall Vechia, 1999). CART was initially

identified in the striatum, where its expression is potently induced by various

psychoactive drugs (Douglass et al., 1995). However, CART is highly expressed in parts

of the hypothalamus, including the ARC, paraventricular nucleus, dorsomedial

hypothalamic nucleus, and lateral hypothalamus. As mentioned above, in the ARC it is

co-expressed with POMC in the anorexic class of first order neurons.

Central CART delivery causes anorexia and can block the orexigenic effect of NPY

(Kristensen et al., 1998; Lambert et al., 1998). Moreover, i.c.v. CART can increase

uncoupling protein (discussed later in Chapter 1) expression in brown and white adipose

tissue and muscle, suggesting that CART may promote increased energy expenditure.

CART knockouts have been shown to have increased susceptibility to diet-induced

obesity (Asnicar et al., 2001). Recent data also suggest that CART infusion may

promote lipid oxidation, particularly in high-fat fed animals (Rohner-Jeanrenaud et al.,

2002). However, animals appear to begin desensitizing to the effects of chronic CART

infusion, including the anorexic effect, within 6 days (Larsen et al., 2000). The

mechanism behind this rapid desensitization is unknown, but it has undoubtedly

contributed to the lack of interest in the CART pathway as a target for the

pharmacological treatment of obesity.









Projections from the Arcuate

The ARC has extensive afferent and efferent connections to "second order

neurons" in other parts of the hypothalamus, including the lateral hypothalamus (LH), the

paraventricular nucleus (PVN), the dorsomedial hypothalamic nucleus (DMH), and the

ventromedial hypothalamus (VMH) (Williams et al., 2001). Most of these centers have

functional Ob-Rb and, thus, may respond to leptin directly or indirectly via projections

from the arcuate.

Lateral hypothalamus. The lateral hypothalamus was classically described as the

"feeding center" since stimulaton of the LH increases food intake and its destruction

leads to profound anorexia, sometimes resulting in starvation. Neurons in the LH

express peptides that potently stimulate food intake, including orexins/hypocretins and

melanin-concentrating hormone (MCH). The LH is also richly enervated with NPY

terminals (originating in the ARC) and is densely populated with NPY-Y5 receptors

(Williams et al., 2001). It is believed that the anorectic signal a-MSH acts, in part, via

projections from the arcuate to the LH where the expression of MCH and orexins are

suppressed (Elias et al., 1999).

Ventromedial hypothalamus. In opposition to the LH, the VMH was classically

viewed as the "satiety center." It has been known for half a century that stimulation of

the VMH leads to anorexia and weight loss whereas destruction of the VMH causes over-

feeeding and obesity (Stellar, 1954). The VMH has connections with other key centers

affecting energy balance in the hypothalamus, including the PVN, LH, and DMH. In

addition, the VMH may be a direct target of leptin as it has been shown to be richly

populated with Ob-Rb (Williams et al., 2001).









Dorsomedial hypothalamus. The DMH is located immediately above the VMH.

It has reciprocal connections with the PVN and LH, and also projects to

extrahypothalamic areas including the brainstem. ARC-NPY/AGRP neurons terminate in

the DMH, and the DMH may act in concert with the PVN to help initiate and maintain

food intake (Christophe, 1998). The DMH is rich in both Ob-Rb and insulin receptors.

Paraventricular nucleus. The PVN is located in the anterior hypothalamus. If

functions as an "integration center" (Williams et al., 2001). The PVN receives

projections from the ARC and the LH. The PVN has abundant nerve-terminals secreting

neurotransmitters that potently modify appetite, including a-MSH and NPY from the

ARC, serotonin, galanin, norepinephrine, and endorphins. Corticotrophin releasing factor

(CRF) is synthesized in the PVN and then released via axonal projections in the median

eminence, where it may inhibit NPY/AGRP neurons (Williams et al., 2001; Arase et al.,

1998). In addition to contributing to an anorectic response, arcuate melanocortin nerve

terminals in the PVN may regulate pituitary hormone release and autonomic nervous

system activity (Spiegelman and Flier, 2001).

Extra-Hypothalamic Signaling and Energy Balance

Signals originating in the hypothalamus project to other parts of the nervous system

to affect both food intake and energy expenditure. For example, CART neurons in the

arcuate also project to autonomic centers in the spinal cord, thus providing a pathway for

the CART-induced increase in sympathetic activity and energy expenditure (Elias et al.,

1999). Melanocortin signaling originating in the hypothalamus may also enhance

sympathetic activity and, ergo, energy expenditure (Spiegelman and Flier, 2001).









Moreover, signals from other parts of the body may travel afferently to various

parts of the brain involved in energy homeostasis. For example, messages from the

digestive tract are conveyed via the vagus nerve to the nucleus of the tractus solitarius

(NTS) in the medulla (Travers et al., 1987). These messages include information about

taste, gastric distension, and portal vein glucose levels (Travers et al., 1987).

Cholecystokinin (CCK), an intestinal satiety signal, also signals to the NTS via CCK

receptors on the vagus. CCK is believed to be involved in meal termination, and its

release is potently stimulated by fatty meals. The NTS possesses receptors for leptin and

a-MSH, and it expresses POMC. Administration of an MC4 receptor agonist (i.e., an a-

MSH analogue) adjacent to the NTS results in potent anorexia, suggesting that this may

be an important extra-hypothalamic feeding/satiety center (Grill et al., 1998).

Certain signals originating in other parts of the brain may project to the

feeding/satiety centers in the hypothalamus. For example, the NTS has projections to the

LH. Neurons originating in the raphe nuclei of the caudal brain stem project adjacent to

the ARC and PVN, where they release serotonin (Williams et al. 2001). Serotonergic

signaling causes feelings of satiety and may also increase energy expenditure (Bray,

2000). As such, it has been the target of numerous weight loss drugs, including the

infamous dexfenfluramine (a main ingredient of"Fen-Phen" or Redux ) (Vickers et al.,

1999) and Meridia (Gura 2003).

Other Molecules Regulating Energy Balance

Insulin

As discussed previously, leptin and insulin share a common signal transduction

pathway through IRSs and PI3Kinase. Their physiological effects, especially in the

central nervous system, are also similar. Like leptin, intracerebroventricular infusion of









insulin causes potent anorexia and weight loss (Woods et al., 1979). Insulin receptors are

found in areas of the brain known to be key hunger/satiety centers, including the arcuate

nucleus of the hypothalamus (Niswender and Schwartz, 2003). Similar to what occurs in

the leptin-deficient ob/ob mouse, absolute insulin deficiency is associated with sustained

hyperphagia known as "diabetic hyperphagia" (Sipols et al., 1995). Twenty-four hour

integrated and fasting serum insulin levels correlate with total adiposity (Bagdade et al.,

1967), much like what is observed with leptin. Finally it is believed that both leptin and

insulin can activate POMC neurons and suppress the activity of NPY/AGRP neurons in

the arcuate nucleus (Niswender and Schwartz, 2003).

Despite the signaling overlap, there are some key differences between the

physiological actions of leptin and insulin. One example is that insulin facilitates fatty

acid and triglyceride deposition in adipose tissue whereas leptin promotes lipolysis via

both central signaling and direct effects on white fat (Wang et al., 1999). Thus, from a

whole-organism perspective, insulin is an anabolic hormone while leptin tends to be

catabolic in nature. However, it is important to point out that the anabolic actions of

peripheral insulin may be different from the specific effects of central insulin signaling,

which, as discussed above, cause anorexia and weight loss.

Obesity is a powerful risk factor for Type II diabetes, which is characterized by

insulin resistance. As we will discuss below, common obesity is typified by

hyperleptinemia and leptin resistance. The relationship between insulin and leptin, both

peripherally and in the central nervous system, is likely complex and not fully understood

at this time. However, given the common signaling pathway and the apparent synergy of









leptin and insulin action in the CNS, it is plausible that leptin resistance may be one

factor linking obesity and insulin resistance.

Glucose

In 1955, Mayer first published his theory of the "glucostat" hypothesis on the

regulation of food intake and body weight (Mayer, 1955). This hypothesis proposed that

special glucose sensing neurons helped to stimulate feeding at the end of a fast. Taken

alone, this theory does not appear to explain feeding behavior very well. For example,

Type II diabetics, who are extremely hyperglycemic, are also hyperphagic. Still, glucose

does appear to be one participant in the complex regulation of feeding. Hypoglycemia or

the blockade of neuronal glucose metabolism stimulates acute feeding behavior

(Williams et al., 2001). Several hypothalamic areas critical to the regulation of energy

homeostasis have been shown to have glucose-sensing neurons including the ARC,

DMH, PVN, VMH, and LH (Williams et al., 2001). Extrahypothalamic areas also have

glucose sensing neurons, including the NTS, substantial nigra, locus coeruleus, neocortex,

and hippocampus (Williams et al., 2001). Selective destruction of glucose-sensing

neurons in the VMH of mice leads to obesity, demonstrating that these neurons are, in

fact, involved in energy homeostasis (Bergen et al., 1996). The mechanism by which

these neurons respond to changes in glucose concentration appears to involve ATP-

sensitive K+ channels (Ashford et al., 1990).

Orexins/Hypocretins

Orexin A, orexin B, and hypocretins 1 and 2 are all derived from a common

precursor peptide known as "prepro-orexin" or "prepro-hypocretin" (Williams et al.,

2001). These four peptides, collectively referred to as orexins/hypocretins, are closely

related structurally and functionally. As mentioned previously, orexins/hypocretins are









expressed by neurons in the lateral hypothalamus. They are also expressed in the

perifomical nucleus and dorsal area of the hypothalamus (Williams et al., 2001).

Orexin/hypocretin neurons project to the PVN, ARC, and the NTS as well as the dorsal

motor nucleus of the vagus nerve (de Lecea et al., 1998; Peyron et al., 1998).

As their name would suggest, these peptides are stimulators of feeding

(orexigenic). Prepro-orexin expression is stimulated by fasting and hypoglycemia and

suppressed by food in the gut (gastric distension) (Williams et al., 2001). Orexins/

hypocretins appear to be short-term regulators of feeding, acting on a meal-to-meal basis.

Consistent with this, acute i.c.v. administration of orexin-A stimulates acute feeding, but

does not alter 24 hour energy intake (de Lecea et al., 1998; Haynes et al., 1999).

Moreover, chronic infusion of orexin-A does not cause weight gain (Yamanaka et al.,

1999).

Ghrelin

Ghrelin is an orexigenic peptide first isolated from the stomach, although it is

expressed in lesser amounts by the pancreas, kidney, ARC, and pituitary gland

(Hillebrand et al., 2002). Ghrelin release is stimulated in states of negative energy

balance and suppressed in conditions positive energy balance or obesity (Otto et al.,

2001; Toshinai et al., 2001). Ghrelin can enhance hypothalamic NPY and AgRP

expression in the hypothalamus (Kamegai et al., 2001), and chronic central delivery of

ghrelin results in increased food intake, bodyweight, and adiposity (Hillebrand et al.,

2002; Tschop et al. 2000). Moreover, it appears that ghrelin may inhibit CART/POMC

neurons in the ARC (Riediger et al., 2003). As such, ghrelin opposes many of the effects

of hypothalamic leptin signaling. Central ghrelin can increase ACTH release and

decrease thyroid stimulating hormone (TSH) levels, perhaps contributing to the reduction









in energy expenditure following ghrelin administration (Hillebrand et al., 2002). Ghrelin

increases growth hormone levels (Kojima et al., 1999), but the physiological significance

of this is still under investigation.

Uncoupling Proteins

Aerobic organisms synthesize most of their ATP through the process of oxidative

phosphorylation. This process starts with NADH and FADH2, high energy molecules

formed during glycolysis, fat oxidation, and the citric acid cycle. NADH and FADH2

pass their electrons to oxygen via a series of electron carriers in the mitochondria, and

this process leads to the pumping of protons out of the mitochondrial matrix. This

ultimately creates a proton gradient. Protons flow through the protein channel "ATP-

synthase," which uses the energy released when the protons flow down their

concentration gradient to phosphorylate ADP to ATP (Stryer, 1995). However,

mammalian metabolism is not perfectly efficient. In fact, it has been estimated that

proton leak across the inner mitochondrial membrane accounts for -25% of basal

metabolic rate in mammals (Porter, 2001).

Specialized proteins called "uncoupling proteins" contribute to the inefficiency of

oxidative phosphorylation. The first uncoupling protein discovered was UCP-1, which

carries electrons from the cytosol into the mitochondrial matrix. UCP-1 is found in

brown adipose tissue (BAT), where it allows the energy liberated from fat oxidation to be

dissipated as heat. As such, BAT UCP-1 is important for thermoregulation in mammals

and is rapidly activated during exposure to cold environments (Scarpace et al., 1994).

These proteins may also play a key role in determining resting metabolic rate. Indeed,

mutations in uncoupling proteins have been linked to reduced basal metabolism in

humans (Porter, 2001).









UCP-1 is one peripheral mediator of the increase in energy expenditure caused by

leptin (Scarpace et al., 1997). In a recent study by Scarpace et al. (2002b), leptin gene

therapy induced a dramatic 34-fold increase in UCP-1 protein levels in BAT. According

to one study, UCP-1 is an absolute requirement for leptin's lipolytic effects in mice

(Commins et al., 2001), but it is unclear if this is universally true in other mammals. It is

clear that central leptin signaling powerfully induces UCP-1 expression and protein levels

in the BAT pads of rats and this is associated with thermogenesis and lipolysis (Scarpace

et al., 1997).

Two other UCP's have been discovered, since named UCP-2 and UCP-3. UCP-2 is

widely distributed, expressed in spleen, lung, intestine, white adipose tissue (WAT),

BAT, uterus, kidneys, testes, brain, and heart with low levels detected in muscle and liver

(Porter, 2001). UCP-3 is found in BAT and skeletal muscle (Boss et al., 1997). The

physiological roles of UCP-2 and -3 are somewhat controversial, but there is evidence

that they can uncouple mitochondrial respiration and potentially increase energy

expenditure. Both UCP-2 and -3 have been shown to decrease resting membrane

potential in transfected yeast, and yeast mitochondria transfected with UCP-3 show

increased oxygen consumption and heat production (Hagen et al., 2000; Porter, 2001).

Moreover, mice overexpressing UCP-3 are hyperphagic and leaner than wild-type mice

(Clapham et al., 2000). Such a phenotype is consistent with increased energy expenditure

due to mitochondrial uncoupling. Leptin has been shown to upregulate UCP-2 in WAT

(Scarpace et al., 1998; Commins et al., 2001) and UCP-3 in BAT (Scarpace et al., 1998),

but the effect is less potent than what is observed with UCP-1. Nonetheless, UCP-2 and -

3 may play an important role in energy homeostasis.









Leptin Resistance

Human obesity is associated with hyperleptinemia and leptin resistance

(Heymsfield et al., 1999). As such, research that can describe the molecular basis of

leptin resistance and how to overcome it could have a profound impact on the clinical

treatment of obesity. Several animal models have been described that mimic the common

human pattern of hyperleptinemia in the obese state. Like obese humans in recent

clinical trials (Heymsfield et al., 1999), these animals are resistant to exogenous leptin.

Aged-obese rats are partially resistant to both injected leptin peptide (Scarpace et al.,

2000a) and constitutively active rAAV-leptin gene therapy (Scarpace et al., 2002b).

This includes impaired signal transduction response (Scarpace et al., 2000b) and impaired

anorectic, thermogenic, and lipolytic responses to leptin (Shek and Scarpace, 2000).

Data from this laboratory (Scarpace et al., 2002b) demonstrate both a reduced maximal

response and a complete attenuation of response to rAAV-leptin in aged-obese but not

young male rats observed for 46 days post-transfection. It is not known if the impaired

responses to leptin in these aged-obese animals are due to age itself, obesity, or a

combination of the two. However, there is evidence that diet-induced obesity can reduce

leptin responsiveness in young animals. Friedman et al. demonstrated that diet-induced

obese mice are less responsive to the lipolytic and anorectic effects of chronic leptin

infusion as compared to lean age-matched controls (Halaas et al., 1997). In a similar

study, 16 days on a high fat diet resulted in resistance to the anorectic effects of

peripheral leptin and this resistance increased in severity by day 56 of continued high fat

feeding (Van Heek et al., 1997). It was also demonstrated that diet-induced obesity

prevents the normal increase in lumbar sympathetic nerve activity in response to

pharmacologically administered leptin in rats (Lu et al., 1998). Increased sympathetic









drive is a key component to leptin's thermogenic and lipolytic effects (Astrup, 2000;

Scarpace and Matheny, 1998). These results suggest that obesity may play a significant

role in leptin desensitization in our aged-obese model, but it does not eliminate the

possibility that age also plays a role. As of the commencement of the research described

in this dissertation, it was not known if the desensitization and rapid attenuation of

response to rAAV leptin gene therapy observed in our aged-obese model could be

duplicated in a young animal via diet-induced obesity; nor was it known if leptin

responsiveness could be restored in an aged-obese model by a period of caloric restriction

that reduces adiposity.

Although there is evidence that obesity plays some role in leptin resistance

independent of aging, the mechanism of this obesity-induced leptin resistance is

unknown. One possibility is that the copious visceral and/or subcutaneous fat may be

secreting one or more substances that inhibit leptin's thermogenic and anorectic effects.

Another possibility is that when adipose depots expand, they stop producing a substance

that is required for normal leptin action. Indeed, discoveries in the past decade have

changed our view of the adipocyte from a simple storage depot to a complex endocrine

organ that plays a critical role in metabolic regulation. Hormones secreted from white

adipose tissue include leptin, TNF-a, TGF-B, IL-6, adipophilin, adipsin, ASP, MIF, IGF-

1, and the recently identified adiponectin (Fruhbeck et al., 2001). A deficiency in this

latter peptide could be one factor linking obesity and both leptin and insulin resistance.

Adeno-Associated Virus and Gene Therapy

Wild-Type AAV

Adeno-associated viruses (AAV) are members of the Parvoviridae family

(Meneses, 1999). AAV virions range from 20-30 nm with icosahedral symmetry









(Meneses, 1999). The AAV genome is a single stranded DNA (ssDNA) molecule

containing -4680 nucleotides (Meneses, 1999). The genome contains only two open

reading frames (ORFs), which are flanked by inverted terminal repeats (ITRs) (Linden

and Berns, 2000). These ITRs can serve as origins of replication (Meneses, 1999). The

5' OPR encodes the "Rep" proteins which, as their name implies, are involved in the

replication of AAV and are also believed to play a critical role in the site-specific

integration of AAV into human chromosome 19 (specifically, locus 19ql3.3-qter)

(Linden and Berns, 2000). The 3' OPF contains the capsid (cap) gene which encodes the

structural proteins of the virion. This cap gene expresses the three cap proteins (Vpl,

Vp2, and Vp2) via alternative splicing (Meneses, 1999).

Approximately 80% of adult humans are seropositive for antibodies against AAV,

with seroconversion typically occurring by age 8 (Berns and Bohenzky, 1987). Despite

this apparent wide spread AAV infection in humans, AAV appears to be benign as it has

never been associated with any disease (Berns and Bohenzky, 1987). In the absence of

co-infection or super-infection with other viruses such as herpes or adenovirus (Ad),

AAV enters a latent pathway, lying dormant in chromosome 19. There are 5 known

serotypes in humans: AAV-1 and AAV-2 are thought to be of simian origin, AAV-3 and

AAV-4 were first isolated from throat swabs of humans suffering from adenovirus

infection, and AAV-5 was initially isolated from a genital condyloma (Blacklow et al.,

1967; Blacklow et al., 1968; Georg-Fries et al., 1984; Meneses, 1999). The adeno-

associated virus used for gene delivery in this dissertation is a recombinant form of AAV

serotype-2.









Recombinant Adeno-Associated Virus and Gene Delivery

The recombinant adeno-associated virus (rAAV) typically used for gene delivery

has been modified by removal of viral coding sequences (both the rep and cap encoding

OTRs), thereby preventing replication (Meneses, 1999; Hermonat et al., 1984). Since the

rep proteins are missing, rAAV is not capable of site-specific integration (Kearns et al.,

1996). Nonetheless, long term expression oftransgene has been observed and there is

evidence of viral integration, albeit at random locations (Kearns et al., 1996). Most AAV

vectors, including those used in this dissertation, consists of the two ITRs flanking the

transgene of interest under the control of a non-AAV promoter such as chicken p-actin

(CBA) (Hauswirth et al., 2000; Zolotukhin et al., 2002). rAAV is a useful vector for

stable gene delivery because it is capable of transfecting non-dividing cells, including

neuronal tissue. This makes it an ideal tool for overexpressing peptides in the central

nervous system. rAAV has been successfully employed as a vector for gene delivery, or

"gene therapy," at the University of Florida and elsewhere (Muzyczka, 1992). For

example, erythropoietin has been produced by the skeletal muscle of mice after

transfection with a rAAV vector encoding the human epo gene (Fisher et al., 1997).

rAAV has been used to over-express tyrosine hydroxylase and aromatic amino acid

decarboxylase in the CNS of animal models of Parkinson's Disease, including non-

human primates (Kaplitt et al., 1994). Our laboratory has successfully used rAAV to

over-express leptin in both the hypothalamus and cells lining the cerebroventricles. We

have also used rAAV to overexpress pro-opiomelanocortin (POMC) in the central

nervous system. rAAV transfection of non-dividing neuronal tissue allows for very

stable transgene expression. In a recent studies by our group, no attenuation of leptin









transgene expression has been noted up to 300 days after intracerebroventricular gene

delivery (Scarpace et al., 2002a; Scarpace et al., 2002b; Scarpace et al., 2003).

Experimental Design and Rational

One purpose of the studies described in this doctoral dissertation was to test the

efficacy of leptin gene delivery as a strategy to reverse acquired obesity. Acquired

obesity refers to obesity that is not due to a specific mutant gene (as in the ob/ob or db/db

mouse). Rather, acquired obesity forms over time as a result of environmental conditions

interacting with many genes. This form of obesity more closely mimics common,

polygenic human obesity. Specifically, we used diet-induced obese animal models,

where obesity gradually occurs due to excessive consumption of a calorically dense diet.

A second purpose was to use central leptin gene and peptide delivery as tools to

study leptin resistance. Diet-induced obese animals were used as a model of leptin

resistance, and various physiological and biochemical responses to leptin were compared

to those in lean cohorts. The general experimental design and objectives of each major

experiment included in this dissertation are described below.

Incorporating Regulation into Leptin Gene Therapy

Constitutively active leptin gene therapy has been shown to have potent anorexic,

thermogenic, and lipopenic affects in both leptin deficient and genetically normal animal

models (Chen et al., 1996; Murphy et al., 1997; Scarpace et al., 2002b). One

disadvantage of all leptin gene delivery systems studied to date is lack of post-

transfectional control. This is undesirable clinically as it would be difficult or

impossible to reverse the progression of deleterious side effects as they appeared.

Perhaps more importantly in the immediate context of this research, incorporating

regulation into the rAAV-encoding-leptin could create an excellent research tool. For









example, an on/off leptin gene therapy system may be useful for studying leptin's long-

term regulation of hypothalamic signal transduction, neuropeptide expression, and leptin

receptor expression. Moreover, it would allow us to determine whether the various

biochemical and physiological changes attributed to leptin transgene overexpression are

completely reversible. Thus, the objective of this experiment was to create a leptin gene

therapy system that can be externally regulated and to then test and study this system in

vivo.

Central Leptin Over-Expression in Diet-Induced Obese Animals

Unlike the ob/ob mouse, diet-induced obese (DIO) animals have an obesity

characterized by hyperleptinemia and leptin resistance. DIO occurs when animals are fed

a high-fat, high-energy diet for several weeks to months. Most animal models, including

those with no known metabolic abnormalities, are susceptible to varying degrees of DIO.

It has been hypothesized that deficient transport of leptin across the blood:brain barrier

(BBB) plays a key role in leptin resistance and, as such, central over-expression of leptin

may overcome this leptin resistance (Dhillon et al., 2001). In a very recent report,

central leptin gene therapy was shown to prevent diet-induced obesity when the gene was

delivered prior to commencing high-fat feeding (Dube et al., 2002). However, since the

leptin transgene was delivered prior to high-fat feeding, the ability of central leptin gene

delivery to overcome leptin resistance and obesity could not be evaluated. Thus, the first

objective of this set of experiments was to determine if intracerebroventricular delivery of

rAAV-encoding-leptin can reverse obesity in genetically normal DIO rats. A second, yet

critical, objective of this investigation was to use the results of this study to further

understand leptin resistance in obese animals.









Effects of Diet-Induced Obesity and Caloric Restriction on Leptin Receptor
Expression and Signal Transduction Capacity

This set of experiments spawned from the results of the rAAV-leptin delivery in

DIO study (previous paragraph). We hypothesized that reduced leptin receptor

expression in DIO animals was the result of chronic hyperleptinemia leading to

chronically elevated leptin in the central nervous system, i.e., negative regulation of

leptin receptor expression by leptin. We further hypothesized that this downregulation in

leptin receptor expression would reduce maximal leptin-induced STAT3 phosphorylation

capacity. Finally, we hypothesized that deficits in leptin receptor expression and

signaling capacity could be reversed if we lowered leptin levels via caloric restriction.

Chapter Summary and Conclusions

In this chapter, we have discussed the many pathways involving a multitude of

hormones, neuropeptides, and neurotransmitters that contribute to the regulation of

energy balance. Given the escalating rates of obesity in modem societies, it appears that

these homeostatic pathways did not evolve to effectively defend against the over-

abundance of highly palatable, over-processed foods and sedentary lifestyles of our era.

While the regulation of energy balance is complex, leptin appears to be a key

player in the process. Indeed, many of the signals participating in energy homeostasis

including melanocortins, NPY, AGRP, CART, and uncoupling proteins occur down-

stream of leptin signaling. Put another way, leptin regulates the expression and activity

of most of the critical molecules involved in the regulation of food intake and energy

expenditure. Moreover, it is clear that the most common forms of human obesity are

typified by leptin resistance. Thus, therapeutic strategies that can overcome or

circumvent this leptin resistance could have an enormous clinical impact in treating not









only obesity, but its many associated disorders including diabetes and cardiovascular

disease.

The major objective of this doctoral dissertation was to explore the molecular

mechanisms of leptin resistance in obese animal models, with a particular emphasis on

leptin signal transduction in the hypothalamus. It is hoped that the results of this

dissertation will make a meaningful contribution to our understanding of leptin resistance

in the obese state. Ultimately, a better understanding of leptin resistance will likely lead

to more effective treatments for obesity and other metabolic disorders.














CHAPTER 2
GENERAL METHODS AND MATERIALS

Experimental Animals

Young adult (age 2-4 months) male Fischer 344 x Brown Norway rats were

obtained form Harlan Sprague-Dawley (Indianapolis, IN) for all experiments described in

this dissertation. Upon arrival, rats were examined and remained quarantined for one

week. Animals were individually caged with a 12:12 hour light:dark cycle (07:00 to

19:00 hr). Animals were cared for in accordance with the principles of the NIH Guide to

the Care and Use of Experimental Animals.

Construction of rAAV Vector Plasmid

pTR-3ObW encodes rat leptin cDNA (a kind gift from Roger Unger (Chen et al.,

1996) and green fluorescent protein (GFP) reporter gene cDNA under the control of

chicken 1-actin promoter linked to CMV enhancer (CBA). Parenthetically, this vector

was referred to as "rAAV-leptin" after it was packaged into recombinant adeno-

associated virus. The woodchuck hepatitis virus posttranscriptional regulatory element

(WPRE) was placed downstream to enhance the expression of the transgenes (Loeb et al.,

1999). The control vector (referred to as "rAAV-con" subsequent to packaging) encodes

GFP driven by a CBA. Vectors contain AAV terminal repeats at both sides of the

cassette to mediate replication and packaging of the vector (Bell et al., 1999). A vector

system with an inducible promoter, "TET-Ob", was also prepared. Like pTR-3ObW

described above, pTR-tetR-Ob encodes rat leptin cDNA and GFP cDNA. However, in

this case both genes are under the control of a tet-inducible promoter (tetR). This tetR









promoter is activated by the product of the accessory vector, pTR-rtTA/tTS, expressing

mutually exclusive reverse transactivator rtTA (Tet-On) and transcriptional silencer

(tTS). In the accessory vector, the rtTA and tTS transgenes are linked within dicistronic

cassettes through an IRES element for coordinate expression. Construction of rAAV-

vector plasmids, including the TET-Ob system (Chapter 3) and constitutively active

rAAV-leptin (Chapters 3 and 4), was done via a collaborative arrangement with Drs.

Sergei Zolotukhin and Victor Prima of the University of Florida's Department of

Molecular Genetics.

Packaging of rAAV Vectors

Vectors were packaged, purified, concentrated, and titered as described previously

(Conway et al., 1999). The titer ofrAAV-Ob was 2.3E13 physical particles/mL. A

mini-adenovirus helper plasmid (pDG) (Grimm et al., 1998) was used to produce rAAV

vectors with no detectable adenovirus or wild type AAV contamination. rAAV vectors

were purified using iodixanol gradient/heparin-affinity chromatography and were more

than 99% pure as judged by PAAG/silver-stained gel electrophoresis (not shown).

Packaging of rAAV viral vectors, was done via a collaborative arrangement with Drs.

Sergei Zolotukhin and Victor Prima of the University of Florida's Department of

Molecular Genetics.

Stereotaxic Injections

Third Cerebroventricle

Rats were anesthetized with 60 mg/kg pentobarbital and heads were prepared for

surgery. Animals were placed into a stereotaxic frame and a small incision (1.5 cm) was

made over the midline of the skull to expose the landmarks of the cranium (Bregma and

Lamda). The following coordinates were used for injection into 3rd cerebroventricle: 1.3









mm posterior to Bregma and 9.4 mm ventral from the skull surface on the midline

(medial fissure), with the nose bar set at 3.3 mm below the ear bars (below zero) and the

canula set at 200 posterior from vertical. A small hole was drilled through the skull and a

23-gauge stainless steel guide canula was lowered to the 3rd cerebroventricle. This was

followed by an injection canula attached to a 10uL syringe.

Hypothalamic

Rats were anesthetized with 60mg/kg pentobarbital and heads were prepared for

surgery. Animals were placed into a stereotaxic frame and a small incision (1.5 cm) was

made over the midline of the skull to expose the landmarks of the cranium (Bregma and

Lamda). The following coordinates were used for direct hypothalamic injection: 1.8 mm

posterior to Bregma, 0.8 mm right of midline (Medial Fissure), and 9.0 mm ventral from

the skull surface. The nose bar was set at zero (on same plane with ear bars) and the

canula was set vertically. A small hole was drilled through the skull and a 23-gauge

stainless steel guide canula was lowered to the hypothalamus. This was followed by an

injection canula attached to a 10uL syringe.

Oxygen Consumption

02 consumption was assessed in up to four rats simultaneously with an Oxyscan

analyzer (OXS-4; Omnitech Electronics, Columbus, OH) as described previously

(Scarpace et al., 1997). Flow rates were 2 L/min with a 30-s sampling time at 5-min

intervals. The rats were placed into the chamber for 150 min with the lowest 6

consecutive 02 consumption values during this period used in the calculations (basal

resting VO2). Food was not available. Animal rooms were free of human activity and

kept as quiet as possible during measurements. All measurements were made between









09:00 and 14:00 hrs. Results were expressed as 02 consumption relative to metabolic

body size (ml min-1 kg2/3).

Tissue Harvesting

Anesthetized rats (85 mg/kg pentobarbital) were sacrificed by cervical dislocation.

Blood was collected by cardiac puncture and serum was harvested by a 10 minute

centrifugation in serum separator tubes. The circulatory system was perfused with 20 mL

of cold saline. Perirenal and retroperitoneal white adipose tissue and hypothalami were

excised, weighed, and immediately frozen in liquid nitrogen. The hypothalamus was

removed by making an incision medial to piriform lobes, caudal to the optic chiasm, and

anterior to the cerebral crus to a depth of 2-3 mm. Tissues were stored at -80 C until

analysis.

Serum Measurements

Leptin

Serum leptin was measured using a rat leptin radioimmunoassay kit (Linco

Research, St. Charles, MO) or a rodent leptin ELISA kit (Crystal Chem, Chicago IL).

Insulin

Serum insulin was measured using a rat insulin radioimmunoassay kit (Linco

Research, St. Charles, MO).

Glucose

Serum glucose was via a colormetric reaction with Trinder, the Sigma Diagnostics

Glucose reagent (Sigma, St. Louis MO).

Free Fatty Acids

Serum free fatty acids were measured using the NEFA C colorimetric kit from

WAKO Chemicals GmbH (Neuss, Germany).









CSF Leptin

CSF leptin was measured by a modification to the Crystal Chem leptin ELISA.

Standards were prepared ranging from 100 down to 3.1 pg/mL. All modifications were

applied to "reaction 1" of the ELISA, where leptin from the biological sample (or

standards) binds to solid phase antibody on the bottom surface of the wells within the

microplate. These modifications are as follows: 50 tiL of standard or CSF were used in

reaction 1 along with 50 [tL of "guinea pig anti-mouse leptin serum" (provided in kit).

"Sample diluent" was omitted from reaction 1 in an attempt to enhance the detection limit

of the kit. In the manufacturer's protocol, 5 tiL of serum or plasma is used in reaction 1

along with 45 [iL of sample diluent.. "Reaction 2," which consists of the binding of

enzyme-linked secondary antibody to the primary antibody-bound leptin, was performed

as described in the manufacturer's instructions. Using this modification, we have a

detection limit of approximately 6 pg/mL CSF. The final absorbance of our 6.25 pg/uL

standard is significantly greater than that of a blank.

Probes

Leptin mRNA was detected using a 33-mer antisense oligonucleotide (5'-

GGTCTGAGGCAGGGAGCAGCTCTTGGAGAAGGC-3') probe. POMC mRNA was

detected using a 24-mer antisense oligonucleotide probe (5'-

CYYGCCCACCGGCTTGCCCCAGCG-3'). Oligonucleotide probes were end labeled

by terminal deoxynucleotidyl transferase (Promega). The UCP-1 probe is a full length

cDNA clone and was obtained from Dr. Leslie Kozak, Pennington Research Center,

Baton Rouge, LA. The AgRP cDNA probe was provided by Dr. Michael Schwartz

(University of Washington). The rat pre pro NPY cDNA was provided by Janet Allen

(University of Glasgow, UK). SOCS3 cDNA was a gift from Christian Bjorbaek









(Harvard University). The cDNA probes were labeled using a random primer kit (Prime-

a-Gene, Promega, Madison, WI). Probes were purified with Nick columns (Pharmacia)

and, except for oligonucleotide probes, were heat-denatured for 2 minutes. All probes

have been verified to hybridized to the corresponding specific mRNAs by Northern

Analysis prior to use in Dot Blot assay (below).

RNA Isolation and RNA Dot Blot

Tissue was sonicated in guanidine buffer, phenol extracted, and isopropanol

precipitated using a modification of the method of Chomczynski and Sacchi (1987).

Isolated RNA was quantified by spectrophotometry and integrity is verified using 1%

agarose gels stained with ethidium bromide. For dot blot analysis, multiple

concentrations of RNA were immobilized on nylon membranes using a dot blot apparatus

(BioRad, Richmond, CA). Membranes were baked in a UV crosslinking apparatus.

Membranes were then prehybridized in 10 mL Quickhyb (Stratagene, LaJolla, CA) for 30

minutes followed by hybridization in the presence of a labeled probe and 100 ug salmon

sperm DNA. After hybridization for 2 hours at 65 C, the membranes were washed and

exposed to a phosphor imaging screen for 24-72 hours (depending on anticipated strength

of signal). The screen was then scanned using a Phosphor Imager (Molecular Dynamic,

Sunnyvale, CA) and analyzed by Image Quant Software (Molecular Dynamics).

Expression data was obtained for the following transcripts using dot blot analysis: UCP-

1, NPY, POMC, SOCS3, and leptin.

Relative-Quantitative RT-PCR Using 18S Competimers

Relative quantitative RT-PCR was performed using QuantumRNA 18s Internal

Standards kit (Ambion, Austin, Tx). Total RNA (3 ig) was treated with RNase-free

DNase using a DNA-free kit (Ambion), and first-strand cDNA synthesis generated from









1 |jg RNA in a 20 pl volume using random primers (Gibco BRL) containing 200 units of

M-MLV reverse transcriptase (Gibco BRL). Relative PCR was performed by

multiplexing target gene primers and 18s primers and coamplifing for a number of cycles

found to be in the linear range of the target. Linearity for the leptin amplicon, for

example, was determined to be 25-30 cycles. The optimum ratio of 18s primer to

competimer is 1:9. PCR was performed at 940C denaturation for 120 sec, 59C

annealing temperature for 60 sec, and 720C elongation temperature for 120 sec for 27

cycles. The PCR product was electrophoresed on a 5% acrylamide gel and stained with

SYBR green (Molecular Probes, Eugene, OR.) Gels were scanned using a STORM

fluorescent scanner and digitized data analyzed using imagequant. (Molecular

Dynamics).

Real-time RT-PCR for Leptin Receptor

Real-time RT-PCR was used to quantify the effects of obesity and caloric

restriction on leptin receptor expression. We designed primers and a Taqman probe

specific for the long-form leptin receptor (Ob-Rb) using Primer Express software, version

1.5 (Perkin-Elmer Applied Biosystems, Inc., Foster City, CA). The sequences for the

ObRb primers were forward primer: 5'-GGGAACCTGTGAGGATGAGTGT-3', reverse

primer: 5'-TTTCCACTGTTTTCACGTTGCT-3'. The fluorescent probe sequence was:

6FAM-AGAGTCAACCCTCAGTTAAATATGCAACGCTG-TAMRA. Optimization

experiments showed that 300 nM of forward primer, 900 nM of reverse primer, and 50

nM Taqman probe gave the most reproducible results and maximally efficient PCR

(i.e., lowest threshold cycle (CT) values). Total RNA (6 Gg) was treated with RNase-free

DNase using a DNA-free kit (Ambion). First-strand cDNA was generated from 1.6 |tg









RNA in a 40 ptL volume using random primers (Gibco BRL) containing 200 units of M-

MLV reverse transcriptase (Gibco BRL). Real-time PCR for Ob-Rb was performed on

100 ng cDNA template in a 50 [tL total volume including Taqman RT-PCR Master Mix

(Applied Biosystems, Inc., Foster City, CA) using an ABI Prism GeneAmp 5700

Sequence Detection System (Applied Biosystems). Ob-Rb expression was quantified

using an 18S rRNA standard (Applied Biosystems) and the AACT method (Bustin, 2000).

The mean ACT in the control group (CHOW) was chosen as the calibrator for AACT

calculation.

STAT3/Phospo-STAT3 Assay

These methods were described in detail previously (Scarpace et al., 2000b). Briefly,

hypothalamus was sonicated in 10 mM Tris-HCL, pH 6.8, 2% SDS, and 0.08 ug/mL

okadaic acid plus protease inhibitors (PMFS, benzamidine, and leupeptin) [an aliquot of

this sonicate was frozen for RNA analysis]. Sonicate was diluted and quantified for

protein using a detergent compatible Bradford Assay. Samples were boiled and separated

on an SDS-PAGE gel and electrotransferred to nitrocellulose membrane.

Immunoreactivity was assessed with an antibody specific to tyrosine-705-

phosphorylated-STAT3 (antibody kit from New England Biolabs, Beverly, MA).

Immunoreactivity was visualized by chemiluminescents detection (Amersham Life

Sciences, Piscataway, NJ) and quantified by video densitometry (BioRad, Hercules, CA).

Following P-STAT3 quantification, membranes were stripped of antibody with

Immunopure (Pierce, Rockford, IL) and immunoreactivity was re-assessed using a total

STAT3 antibody. STAT3 phosphorylation is expressed as P-STAT3/Total STAT3 in

each sample.









UCP-1 Protein in Brown Adipose Tissue

Approximately 30 mg of interscapular brown adipose tissue was homogenized in

300 uL 10 mM Tris-HCL, pH 6.8, 2% SDS, and 0.08 ug/mL okadaic acid plus protease

inhibitors (PMFS, benzamidine, and leupeptin). Samples were boiled for 5 minutes to

and an aliquot of this homogenate was withdrawn and diluted for detergent compatible

Bradford protein analysis. Samples were boiled and separated on an SDS-PAGE gel (20

[g protein/lane) and electrotransferred to nitrocellulose membrane. Immunoreactivity

was assessed with an antibody specific to UCP-1 (Linco Research, St. Charles, MO).

Immunoreactivity was visualized by chemiluminescents detection (Amersham Life

Sciences, Piscataway, NJ) and quantified by video densitometry (BioRad, Hercules, CA).

Statistical Analysis

All data are expressed as mean + standard error of mean. ca level was set at 0.05

for all analyses. Data were analyzed by 2-Way ANOVA, 1-Way ANOVA, or student's t-

test, as appropriate. A Tukey's post-hoc was used for 1-Way ANOVAS post-hoc

analysis or a bonferoni multiple comparison test correcting for the number of contrasts.

The post-hoc analysis of 2-Way ANOVAS depended on the presence or absence of

interactions. When only 2-Way ANOVA main effects were significant, relevant pairwise

comparisons were made using the Bonferroni Multiple Comparison method with the error

rate corrected for the number of contrasts. When there was an interaction, factors were

separated and a further 1-Way ANOVA was applied with a Bonferroni Multiple

Comparison post-hoc. When separation of factors resulted in only two population means

to compare, the 1-WAY ANOVA was replaced with student's t-test. GraphPad Prism

software version 3.0 (San Diego, CA) was used for all statistical analysis and graphing.






37


GraphPad QuickCalc (graphpad.com) was used for post-hoc analysis of all 2-WAY

ANOVAS.














CHAPTER 3
INCORPORATING REGULATION INTO LEPTIN GENE THERAPY

Introduction

Leptin, the product of the Ob gene, acts on satiety centers in the hypothalamus to

both decrease food intake and increase energy expenditure. Virus-mediated leptin gene

delivery has been shown to cause a rapid and complete disappearance of white adipose

tissue in genetically normal animals (Chen et al., 1996). Leptin gene delivery is

substantially more effective in correcting genetic obesity than daily leptin infusion

despite higher peak serum leptin levels with the latter (Morsy et al., 1998). Peripheral

administration of virally-encoded leptin requires that the leptin cross the blood-brain

barrier (BBB) to reach its primary hypothalamic targets, but this transport system is

saturated at serum leptin levels seen in obese animals (Banks et al., 1999). Since several

obese, leptin resistant models respond better to central versus peripheral administration of

recombinant leptin (Halaas et al., 1997; Niimi et al., 1999; Van Heek et al., 1997), it has

been reasoned that central delivery of the leptin gene would be superior to peripheral

delivery (Lundberg et al., 2001). Indeed, central delivery of virally-encoded leptin causes

dramatic lipopenia while avoiding systemic hyperleptinemia and the limitations of BBB

transport (Lundberg et al., 2001; Scarpace et al., 2002b). In Experiment 1 of this

chapter, I administered rAAV-encoding leptin under the control of a constitutively active

CBA promoter (Figure 3-1) in an attempt to duplicate our previous finding of potent

weight and fat loss in with this vector.









The main limitation of viral-mediated leptin-gene therapy systems studied to date,

including the vector used in Experiment 1, is the lack of post-transfectional control. This

is undesirable clinically as it would be difficult or impossible to reverse the progression

of deleterious side effects as they appeared. Given leptin's diverse biological functions,

central overexpression of leptin (coupled with a precipitous fall in serum leptin) could

compromise the immune (Matarese, 2000) and reproductive systems (Ahima and Flier,

2000a). Central leptin signaling is also known to decrease bone mass (Ducy et al., 2000).

Peripheral overexpession of leptin, on the other hand, leads to pancreatic beta cell

disfunction (Koyama et al., 1997) and upregulates suppressor of cytokine signaling,

which may interfere with insulin signaling (Spiegelman and Flier, 2001).

We have attempted to solve these problems of uninhibited transgene expression by

placing leptin under the control of the tetracycline transactivator (rtTA) and operon

(tetR). In this dual-vector system, one vector encodes the tetR promoter and the leptin

gene while an assessory vector endodes rtTA and a transcriptional silencer (tTS) [Figure

3-2]. This dual vector "TET-Ob" system allows for the control leptin transgene

expression via doxycycline (doxy) in drinking water. In Experiment 2, we administered

the TET-Ob dual-vector system into the hypothalami of young adult, non-obese rodents,

and examined the effects on anorectic signaling in the brain, peripheral thermogenesis,

and adiposity. Moreover, we determined if these effects can be completely reversed if

the transgene is silenced.

Methods and Materials

Animals

Young adult (age -4 months) male Fischer 344 x Brown Norway rats were

obtained form Harlan Sprague-Dawley (Indianapolis, IN). Upon arrival, rats were









examined and remained quarantined for one week. Animals were individually caged

with a 12:12 hour light:dark cycle (07:00 to 19:00 hr). Animals were cared for in

accordance with the principles of the NIH Guide to the Care and Use of Experimental

Animals.

Construction of rAAV Vector Plasmid

Constitutively active leptin and control vector. pTR-8ObW (Fig 3-1) encodes

rat leptin cDNA (a gift from Guoxun Chen (Chen et al., 1996) and green fluorescent

protein (GFP) reporter gene cDNA under the control of chicken 8-actin promoter linked

to CMV enhancer (CBA). The woodchuck hepatitis virus posttranscriptional regulatory

element (WPRE) was placed downstream to enhance the expression of the transgenes

(Loeb et al., 1999). Control vector (pTR-control) is identical to pTR-BObW but without

the leptin cDNA and IRES element. The control vector encodes GFP alone.

TET-Ob vector system with inducible promoter. Like pTR-BObW described

above, pTR-tetR-Ob (Figure 3-2, top) encodes rat leptin cDNA and GFP cDNA.

However, in this case both genes are under the control of a tet-inducible promoter (tetR).

The woodchuck hepatitis virus posttranscriptional regulatory element (WPRE) was

placed downstream to enhance the expression of the transgenes (Loeb et al., 1999). This

tetR promoter is activated by the product of the accessory vector, pTR-rtTA/tTS (Figure

3-2, bottom), expressing mutually exclusive reverse transactivator rtTA (Tet-On) and

transcriptional silencer (tTS). In the accessory vector, the rtTA and tTS transgenes are

linked within dicistronic cassettes through an IRES element for coordinate expression.

All vectors contain AAV terminal repeats at both sides of the cassette to mediate

replication and packaging of the vector (Bell et al., 1999).









Packaging of rAAV Vectors

Vectors were packaged, purified, concentrated, and titered as described previously

(Conway et al., 1999). The titers of rAAV-tetR-Ob and rAAV-AS (accessory virus) used

were 6.6E12 and 1.37E12 infectious particles/mL, respectively. The titer of the

constitutively active rAAV-leptin was 6.0E9 infectious particles/mL. The physical

particle titer is approximately 100 fold greater than the infectious titer in all cases. A

mini-adenovirus helper plasmid (pDG) (Grimm et al., 1998) was used to produce rAAV

vectors with no detectable adenovirus or wild type AAV contamination. rAAV vectors

were purified using iodixanol gradient/heparin-affinity chromatoagraphy and were more

than 99% pure as judged by PAAG/silver-stained gel electrophoresis (not shown).

Vector Administration

Third ventricle injection

Vectors were delivered into the 3rd cerebroventricle in Experiment 1. Rats were

anesthetized with 60 mg/kg pentobarbital and heads were prepared for surgery. Animals

were placed into a stereotaxic frame and a small incision (1.5 cm) was made over the

midline of the skull to expose the landmarks of the cranium (Bregma and Lamda). The

following coordinates were used for injection into 3rd cerebroventricle: 1.3 mm posterior

to Bregma and 9.4 mm ventral from the skull surface on the midline (medial fissure),

with the nose bar set at 3.3 mm below the ear bars (below zero) and the canula set at 200

posterior from vertical. A small hole was drilled through the skull and a 23-gauge stain-

less steel guide canula was lowered to the 3rd cerebroventricle. This was followed by an

injection canula attached to a 10uL syringe. I injected 3.0 uL containing 1.8E9 infectious











fl(+) origin TR
CMV ie enhancer
vChicken p~-actin
SExon1
SChimeric intron
ApR Ob cDNA, rat
pTR-betaObW



ColE1 ori IRES

TR
bGH poly(A) hGFP F64L/S65T
WPRE

Figure 3-1: pTR-BObW. pTR-BObW encodes rat leptin cDNA (and green fluorescent
protein (GFP) reporter gene cDNA under the control of a constitutively active
chicken B-actin promoter linked to CMV enhancer (CBA). Control vector
(pTR-control) is identical to pTR-3ObW but without the leptin cDNA and
IRES element. The control vector encodes GFP. The cassette flanked by the
AAV terminal repeats (TR) in each plasmid were packaged into rAAV to
yield the rAAV-leptin vector.










fl (+) origin TR
STet responsive promoter


& Ob cDNA, rat


pTR-tetR-Ob
6198 bp


IRES


ColE1 ori


TR
fl(+) origin


GFP (F64L, S65"


SV40 poly(A)


\CMV ie enhancer
1 Chicken P-Actin
Exonl


pTR-rtTA/tTS
7157 bp


ColE1 ori


TR
bGH poly(A) /
SV40 poly(A)


Chimeric intron


rtTA


IRES


Figure 3-2: The TET-responsive promotor/leptin plasmid construct pTR-tetR-Ob (top),
and the assessory plasmid construct pTR-rtTA/tTS (bottom).


ApR











I r CC<


Accessory
Vector


CBA Promoter rtTA reverse
Tet
S/v Reaulatable


S

U


Trans-
A -+ ..-.+, -


tTS Tet
Regulatable
Transcriptional
Silencer


vaorUI


P $C


V
.4'"


TRE-Leptin
Vector


Responsive
Promoter


Rat Leptin
cDNA


Figure 3-3: Schematic representation of leptin transgene regulation by doxycycline in
dual vector "TET-Ob" system, "OFF" state. In this dual vector system, the
accessory vector (top) encodes a transcriptional activator (rtTA, green ovals)
and a transcriptional silencer (tTS, blue diamonds), both of which are under
the control of a constitutively active chicken p-actin (CBA) promoter. The
TET-leptin vector (bottom) encodes leptin under the control of a Tet
responsive promoter. In the default state, the transcriptional silencer binds the
Tet Responsive Promoter, preventing transcriptional activator binding to the
promoter and blocking leptin expression. Thus, in the absence of tetracycline
analogue such as doxycyline, this system is in the "OFF" state.


No DOXY













SIRPF


CBA Promoter

0 W


rtTA reverse
Tet
Regulatable
Trans-
Activator


tTS Tet
Regulatable
Transcriptional
Silencer


0 io


TRE-Leptin
Vector


Tet
Responsive
Promoter


Figure 3-4: Schematic representation of leptin transgene regulation by doxycycline in
dual vector "TET-Ob" system, "ON" state. In presence of doxycycline
(yellow circles), the relative affinities of the transcriptional activator (green
ovals) and the transcriptional silencer (blue diamonds) for the Tet Responsive
Promoter are reversed. The transcriptional silencer is displaced from the
promoter and the transcriptional activator binds to the promoter, activating the
expression of leptin transgene. Thus, in the presence of doxycyline, this
system is in the "ON" state.


Accessory
Vector


DOXY


M


Rat Leptin
cDNA









viral particles dissolved in Ringer's solution at approximately 0.25 uL/minute. Animals

received either the constitutively active rAAV-leptin vector (n=7) or rAAV-con encoding

GFP (n=3).

Hypothalamic injection

The dual vector TET-Ob system was micro-injected into the center of the right

hypothalamus to maximize co-infection of cells. As with the 3rd cerebroventricle

injections, rats were anesthetized with 60mg/kg pentobarbital and heads were prepared

for surgery. Animals were placed into a stereotaxic frame and a small incision (1.5 cm)

was made over the midline of the skull to expose the landmarks of the cranium (Bregma

and Lamda). The following coordinates were used for direct hypothalamic injection: 1.8

mm posterior to Bregma, 0.8 mm right of midline (Medial Fissure), and 9.0 mm ventral

from the skull surface. The nose bar was set at zero (on same plane with ear bars) and the

canula was set vertically. A small hole was drilled through the skull and a 23-gauge

stainless steel guide canula was lowered to the hypothalamus. This was followed by an

injection canula attached to a 10uL syringe. I injected 5 uL of viral particles in Ringer's

solution at approximately 0.25 uL/minute. Animals received either the two-vector TET-

Ob system containing equal amounts (6E9 infectious particles) of rAAV-TetR-Ob and

accessory vector (n=14) or control virus encoding GFP (n=6).

Experimental Design

Experiment 1

The aim of Experiment 1 was to study the effects of constitutively active rAAV-

leptin on body weight, food intake, and adiposity. I injected rAAV-leptin (n=7) or

control virus encoding GFP (rAAV-con) (n=3) into the 3rd ventricle of 4 month old male









F344xBN rats using a stereotaxic apparatus and monitored food intake and body weight

for -60 days, at which point animals were euthenized for tissue analysis.

Experiment 2

The primary aim of Experiment 2 was to test our ability to regulate leptin transgene

expression using the TET-Ob system. Experiment 2 is sub-divided into two stages.

During STAGE 1, doxycycline hydrochloride (Sigma, St. Louis, MO) (400 ug/mL) and

0.1% sacharrine (Sigma) were provided in the drinking water of all animals beginning

day 9 post-transfection. All rats received doxy for a total of 34 days. At the end of

STAGE 1, doxy was withdrawn from half of the TET-Ob rats for 32 days (TET-Ob-OFF,

n=7) while half continued to receive Doxy (TET-Ob-ON, n=7). This second phase of the

study is designated as STAGE 2. Controls received doxy throughout the study (i.e., both

STAGE 1 and STAGE 2).

Tissue Harvesting

Rats were anesthetized with 85 mg/kg pentobarbital and sacrificed by cervical

dislocation. Blood was collected by cardiac puncture and serum was harvested by a 10

minute centrifugation in serum separator tubes. The circulatory system was perfused

with 20 mL of cold saline. Inguinal, perirenal, and retroperioneal white adipose tissue,

brown adipose tissue, and hypothalami were excised, weighed, and immediately frozen in

liquid nitrogen. Tissues were stored at -80 C until analysis.

Serum Leptin, FFA, Insulin, and Glucose

Serum leptin and insulin were measured using rat radioimmunoassay kits (Linco

Research, St. Charles, MO). Serum free fatty acids were measured using the NEFA C

colorimetric kit from WAKO Chemicals GmbH (Neuss Germany). Serum glucose was









measured via a colormetric reaction with Trinder, the Sigma Diagnostics Glucose reagent

(Sigma, St. Louis MO).

Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR)

Hypothalamic leptin transgene expression and Ob-Rb expression were evaluated

by using relative quantitative RT-PCR through the use of QuantumRNA 18s Internal

Standards kit (Ambion, Austin, Tx). Total RNA (3 ig) was treated with RNase-free

DNase using a DNA-free kit (Ambion), and first-strand cDNA synthesis was generated

from 1 |tg RNA in a 20 [tl volume using random primers (Gibco BRL) containing 200

units of M-MLV reverse transcriptase (Gibco BRL). Relative PCR was performed by

multiplexing leptin primers (sense: 5'GGAGGAATCC-CTGCTCCAGC; antisense: 5'

CCTCTCCTGAGGATACCTGG) and 18s primers and coamplifing for a number of

cycles found to be in the linear range for leptin. Linearity for the leptin amplicon was

determined to be 25-30 cycles. The optimum ratio of 18s primer to competimer was 1:9.

PCR was performed at 940C denaturation for 120 sec, 59C annealing temperature for 60

sec, and 720C elongation temperature for 120 sec; for 27 cycles. Similarly, the number

of cycles found to be at the mid point of the linear range was 26 cycles for Ob-Rb. The

sequence for Ob-Rb primers were sense: GGGAACCTGTGAGGATGAGTG; antisense:

TAGCCCCTTGCTCTTCATCAG. The PCR products were electrophoresed on a 5%

acrylamide gel and stained with SYBR green (Molecular Probes, Eugene, OR.) Gels

were scanned using a STORM fluorescent scanner and digitized data analyzed using

imagequant. (Molecular Dynamics).

STAT3/Phospo-STAT3 Assay

These methods were described in detail previously (Scarpace et al., 2000b).

Briefly, hypothalamus was sonicated in 10 mM Tris-HCL, pH 6.8, 2% SDS, and 0.08









[g/mL okadaic acid plus protease inhibitors (PMFS, benzamidine, and leupeptin) [an

aliquot of this sonicate was frozen for RNA analysis]. Sonicate was diluted and

quantified for protein using a detergent compatible Bradford Assay. Samples were boiled

and separated on an SDS-PAGE gel and electrotransferred to nitrocellulose membrane.

Immunoreactivity was assessed with an antibody specific to phosphorylated-STAT3

(antibody kit from New England Biolabs, Beverly MA). Immunoreactivity was

visualized by chemiluminescents detection (Amersham Life Sciences) and quantified by

video densitometry (BioRad). Following P-STAT3 quantification, membranes were

stripped of antibody (Pierce) and immunoreactivity was re-assessed using a total STAT-3

antibody .

Probes

The UCP-1 probe is a full length cDNA clone and was obtained from Dr. Leslie

Kozak, Pennington Research Center, Baton Rouge, LA. The rat pre pro NPY cDNA

was provided by Janet Allen (University of Glasgow, UK). SOCS3 cDNA was a gift

from Christian Bjorbaek (Harvard University). The cDNA probes were labeled using a

random primer kit (Prime-a-Gene, Promega, Madison, WI). Probes were purified with

Nick columns (Pharmacia) and, except for oligonucleotide probes, were heat-denatured

for 2 minutes. All probes have been verified to hybridized to the corresponding specific

mRNAs by Northern Analysis prior to use in Dot Blot assay (below).

mRNA Levels (Dot Blot Analysis)

Tissue was sonicated in guanidine buffer, phenol extracted, and isopropanol

precipitated using a modification of the method of Chomczynski (Chomczynski and

Sacchi, 1987). Isolated RNA was quantified by spectrophotometry and integrity was

verified using 1% agarose gels stained with ethidium bromide. For dot blot analysis,









multiple concentrations of RNA were immobilized on nylon membranes using a dot blot

apparatus (BioRad, Richmond, CA). Membranes were baked in a UV crosslinking

apparatus. Membranes were then prehybridized in 10 mL Quickhyb (Stratagene, LaJolla,

CA) for 30 minutes followed by hybridization in the presence of a labeled probe and 100

tg salmon sperm DNA. After hybridization for 2 hours at 65 C, the membranes were

washed and exposed to a phosphor imaging screen for 24-72 hours (depending on

anticipated strength of signal). The screen was then scanned using a Phosphor Imager

(Molecular Dynamic, Sunnyvale, CA) and analyzed by Image Quant Software

(Molecular Dynamics).

Statistical Analysis

Food intake and bodyweight comparisons were made by a 2-Way ANOVA with

rAAV vector and time as factors. Comparisons of absolute body weights, mean daily

food intake, adiposity, and serum lepin in rAAV-leptin versus rAAV-con were made

using unpaired t-test. Comparisons between control and TET-Ob during stage 1 were

made using unpaired t tests with a one-tailed p value. For stage 2 and endpoint statistics,

relevant pairwise comparisons were made using unpaired t tests with the Bonferroni

correction for multiple comparisons applied to the a level, which was set at 0.05. We

chose this test over a 1-WAY ANOVA because we did not have 3 independent groups

receiving 3 treatments during this experiment as assumed in a 1-way ANOVA. TET-Ob-

ON and TET-Ob-OFF were one group (TET-Ob) during the first half of the experiment

and then were split at the midpoint. The controls, on the other hand, were always

independent of the other two groups. Moreover, we are not testing the single null

hypothesis that Control = Ob-ON = Ob-OFF for each of our experimental parameters, as

in the ANOVA. Rather, we our testing two separate null hypothesis at the conclusion of









the experiment: (1) Ob-ON = Control, and (2) Ob-ON = Ob-OFF. The first test is to

determine whether or not the TET-Ob system had an effect. The second test is to

determine if we were able to de-activate the transgene. In the case of leptin expression,

we tested a third null hypothesis, Ob-OFF = Control, and adjusted the alpha level

accordingly (i.e., a=0.05/3). Rejection of the null hypothesis in this latter test would

indicate some leaky expression of transgene after withdrawing doxycycline. Since we

corrected for the error rate associated with multiple pairwise comparisons, this method

does not increase our chance of a Type I error. GraphPad Prism software (San Diego,

CA) was used for all statistical analysis and graphing.

Results

Experiment 1

Food consumption and body mass

On the day of vector delivery, body mass did not differ between controls and

rAAV-leptin treated animals (252.711.3 vs 260.91.7 g, respectively) [Fig 3-5].

Following gene delivery, control (rAAV-con) rats steadily gained mass while rAAV-

leptin rats had a slight decrease in body mass such that rAAV-leptin rats weighed 31.5%

less than controls 61 days after gene delivery (329.817.2 vs. 250.9+11.0 g, p<0.0001)

[Fig 3-5]. rAAV-leptin treatment also resulted in a significant anorectic effect versus

controls beginning on day 6 post-transfection [Figure 3-6]. Between days 6 and 61,

mean daily food intake was 20.1% less in rAAV-leptin treated rats (17.280.19 vs.

13.800.30 g/day, p<0.0001) [Figure 3-6].

Serum leptin and adiposity

The suppression in age-associated weight gain in the rAAV-leptin group was

accompanied by the complete disappearance of white adipose tissue [Figure 3-7]. The









sum of two major visceral white fat depots (retroperitoneal white adipose [RTWAT] and

perirenal white adipose [PWAT] depots) was 3.08 + 0.29 g in the control rats -61 days

after gene delivery, whereas visceral fat pads were absent in rAAV-leptin treated group.

We know from other experiments in this laboratory that this complete catabolism of

white adipose depots takes place within 9 days (Scarpace et al., 2002b). Moreover, this

loss of white fat is accompanied by an apparent preservation of lean body mass (Dhillon

et al., 2001). Neither caloric restriction nor any drug known to the author can replicate

this dramatic, rapid and selective loss of white fat. Consistent with the loss of adipose

tissue in rAAV-leptin treated rats, serum leptin (an index of adiposity) was undetectable

in these animals [Figure 3-8].

Experiment 2

Food consumption and body mass

Beginning day 9 post-transfection, doxy was provided in the drinking water of all

animals for 34 days (STAGE 1), presumably activating the leptin transgene in all TET-

Ob treated animals [Figure 3-9]. During STAGE 1, TET-Ob rats gained 51.7 % less

mass [Figure 3-10, between arrows] and ate 11.4 % less food [Figure 3-11] than controls.

To begin STAGE 2, the TET-Ob group was divided into two subgroups. Doxy was

withdrawn from half of the TET-Ob treated animals for 32 days (TET-Ob-OFF) while

half continued to receive doxy (TET-Ob-ON) [Figure 3-9]. During STAGE 2, TET-Ob-

ON rats gained 44.8 % less mass than TET-Ob-OFF rats [Figure 3-10], and ate

significantly less food than both TET-Ob-OFF and controls [Figure 3-11]. Rates of body

mass gain were not different between TET-Ob-ON and controls during STAGE 2. The

same degree of anorexia was maintained in TET-Ob-ON animals throughout both stages

of the study. Doxy/saccharin did not affect fluid intake [data not shown].









Serum leptin and adiposity

Consistent with leptin-dependent catabolism of adipose tissue in TET-Ob rats,

serum leptin fell to 22.5% of control values during STAGE 1 [Figure 3-12]. During

STAGE 2, serum leptin increased to 83.4% of control values in the TET-Ob-OFF animals

but remained very low (21.9% of controls) in the TET-Ob-ON group [Figure 3-12]. At

sacrifice, average visceral adiposity (sum RTWAT and PWAT) was 14.6% of controls in

TET-Ob-ON animals [Figure 3-13], and three of the seven TET-Ob-ON had no visible

intra-abdominal adipose tissue [similar to Figure 3-7, right]. Visceral adiposity was

76.9% of controls in TET-Ob-OFF at sacrifice, suggesting these animals recovered

adipose tissue during STAGE 2 [Figure 3-13].

Serum free fatty acids, insulin, and glucose

At sacrifice, average serum free fatty acids were 15.2% lower in the TET-Ob-ON

group compared to the TET-Ob-OFF group and were 12.5% lower in TET-Ob-ON versus

controls [Table 3-1]. Neither of these findings was statistically significant after

employing the Bonferroni correction for multiple comparisons. Serum insulin was

significantly reduced in the TET-Ob-ON as compared to both controls and TET-Ob-OFF

[Table 3-1]. Serum glucose was unaffected by treatment [Table 3-1].

Leptin expression

Leptin expression in the hypothalamus of TET-Ob-ON animals was increased 8.5

fold versus controls [Figure 3-14]. The level of hypothalamic leptin expression did not

differ between TET-Ob-OFF and control, indicating that the transgene was silenced when

doxy was withdrawn [Figure 3-14].









Signal transduction in hypothalamus

Hypothalamic phosphorylated STAT3 (P-STAT3), a component of the leptin-

signaling cascade, was elevated by 46% in TET-Ob-ON group but not TET-Ob-OFF

[Figure 3-15]. Total STAT3 was not affected [data not shown]. It is believed that an

increase in POMC expression (the precursor to c-MSH) contributes to the anorectic

effect of leptin (Elias et al., 1999). In this study, hypothalamic POMC mRNA was found

to be elevated by 41.2% in TET-Ob-ON animals versus controls and by 16.4 % versus

TET-Ob-OFF [Figure 3-16]. Leptin is also known to up-regulate expression of

suppressor of cytokine signaling (SOCS3) (Wang et al., 2000; Scarpace, 2002). SOCS3

mRNA was increased by 67.3% in TET-Ob-ON versus controls and by 34.9% versus

TET-Ob-OFF [Figure 3-17]. Finally, leptin signaling is known to depress NPY

expression in arcuate neurons (Spiegelman and Flier, 2001). However, we did not

detect any differences in NPY expression (100.0 12.4 arbitrary units in control,

94.919.7 in TET-Ob-ON, 115.810.0 in TET-Ob-OFF).

Leptin receptor expression in the hypothalamus

Hypothalamic leptin receptor expression in TET-Ob-ON was 24% lower than that

in control and 31% lower than TET-Ob-OFF [Figure 3-18]. However, only the

difference between TET-Ob-ON and TET-Ob-OFF reached statistical significance.

Nevertheless, this novel data suggests that central overexpression of leptin may

negatively regulate leptin receptor expression. Moreover, this effect appears to be

reversed upon transgene silencing.

CSF leptin

Data from Dr. Scarpace's laboratory demonstrates an increase in CSF leptin

concentrations when a constitutively active leptin gene encoded by rAAV is delivered









i.c.v. (Scarpace et al., 2002b). In the present study, I gave the TET-Ob system directly

into the hypothalamus. In this case, I detected a significant decrease in CSF leptin in the

TET-Ob-ON but not TET-Ob-OFF animals as compared to controls [Table 3-1].

Brown adipose tissue

TET-Ob-ON animals had a 3-fold increase in UCP-1 protein levels per unit of BAT

compared to both controls and TET-Ob-OFF [Figure 3-19]. Furthermore, TET-Ob-ON

animals tended to have elevated UCP-1 mRNA in BAT compared to TET-Ob-OFF and

controls [Table 3-1]. Consistent with activated BAT in the TET-Ob-ON group, their

BAT appeared dark red (stimulated) whereas the BAT extracted from both controls and

TET-Ob-OFF animals appeared pale brown (dormant). Total BAT tissue mass declined

in the TET-Ob-ON animals while the protein concentration (per unit BAT) increased

[Table 3-1], suggesting that the decrease in mass was due to lipolysis. Even when the

decrease in BAT mass is accounted for, there was still a 2-fold increase in UCP-1 protein

per total interscapular BAT pad in TET-Ob-ON compared to both controls and TET-Ob-

OFF [Table 3-1].

Discussion

A single injection of rAAV encoding leptin under the control of a constitutively

active promoter has been shown to cause sustained (>2 months) anorexia and severe

reductions in adiposity, and we repeated these findings in Experiment 1. We then

hypothesized that a direct hypothalamic injection of rAAV encoding leptin under control

of the tetracycline transactivator and operon (TET-Ob) would allow us to regulate leptin

transgene expression and the subsequent anorexic and thermogenic effects (Experiment

2). In animals given the TET-Ob system, we demonstrated a reversible suppression in












375-
350-
325-
300-
275-
250-
225-


200-
-10


-o- rAAV-con
-- rAAV-leptin


I I I I I I I I7
0 10 20 30 40 50 60 70


Day

Figure 3-5: Body mass following intracerebroventricular administration of rAAV-leptin.
Values represent mean SEM. By 2-Way ANOVA, significance was found
for effects of vector (F=83.91, p<0.0001), time (F=5.40, p<0.0001), and the
interaction between vector and time (F=6.52, p<0.0001).


22.5

20.0

17.5

15.0

12.5

10.0-


-10 0 10


S-o- rAAV-con
-*- rAAV-leptin

20 30 40 50 60 70


Day


Figure 3-6: Food intake following intracerebroventricular administration of rAAV-leptin
or rAAV-con. Values represent mean SEM. By 2-Way ANOVA,
significance was found for effects of vector (F=77.50, p<0.0001), time
(F=8.21, p<0.0001), and the interaction of vector and time (F=3.54,
p<0.0001).











rAAV-con


rAAV-leptin


Figure 3-7: Disappearance of visceral adipose tissue in 4 month old F344xBN rats
administered rAAV-leptin. The animal on the left received an
intracerebroventicular injection of control rAAV encoding GFP. The animal
on the right received a similar injection of rAAV-encoding-leptin. These
pictures were taken 46 days after gene delivery, but no recovery in adiposity
was observed at 61 days.


rAAV-con


W rAAV-con
SrAAV-leptin



<0.05 ng/mL


rAAV-leptin


Figure 3-8: Leptin is undetectable in serum of animals following central leptin gene
delivery, consistent with severe reduction in adiposity. Values represent mean
SEM. Measurements taken 61 days after vector delivery


4-

E
3-

- 2
E-
0
O


I I I _


T












STAGE 1 STAGE 2

34 days 32days

000
Sacrifice
r 0 0 0 TET-Ob-ON S
rAAV-TET-Ob
1----- -- for
TET-Ob-OFF f

0 0 0 Tissue

r 0 0 0 0 0 An0 alyis
rAAV-GFP 0



Figure 3-9: Experiment 2 design, TET-Ob gene delivery and regulation. Experiment 2 is
sub-divided into two stages. Animals were given a direct hypothalamic
injection containing the dual-vector rAAV-TET-Ob system (top, n=14) or
control rAAV encoding GFP (bottom, n=6). During STAGE 1, which began
9 days after gene delivery, doxycycline hydrochloride was provided in the
drinking water of all animals, presumably activating the leptin transgene in the
TET-Ob treated rats. STAGE 1 lasted 34 days. At the end of STAGE 1, doxy
was withdrawn from half of the TET-Ob rats for 32 days (TET-Ob-OFF, n=7)
while half continued to receive Doxy (TET-Ob-ON, n=7). This second phase
of the study is designated as STAGE 2. Controls received doxy throughout
the study (i.e., both STAGE 1 and STAGE 2). During STAGE 2, the leptin
transgene should be silenced in the TET-Ob-OFF group, yet remain active in
the TET-Ob-ON.











380-

360- STAGE 1

340-

' 320-

m 300-

280-
-10 0 10 20 30 40


-o- Control
TET-Ob-ON
-*- TET-Ob-OFF

50 60 70


Day


Figure 3-10: Body mass following TET-Ob or control vector delivery. rAAV-TET-Ob
or control was administered on day (-)9. Left arrow represents start of
STAGE 1. Doxy was withdrawn from half of the TET-Ob treated animals on
day 34 (right arrow, start of STAGE 2). Values represent means SEM.
p<0.001 for difference in body mass gained during STAGE 1 in TET-Ob and
control. p<0.01 for difference in body mass gained during STAGE 2 in Ob-
ON and Ob-OFF.


20-

>19-
-

"0
o
018

o 17
U-
16
16-


151 I I M
STAGE 1


STAGE 2


III Control
- TET-Ob(ON)
SOb-OFF


Figure 3-11: Daily food consumption following TET-Ob or control vector delivery.
Values represent means SEM.. TET-Ob was divided into TET-Ob-ON and
TET-Ob-OFF to begin STAGE 2. p<0.01 during STAGE 1 for difference
between TET-Ob and control. p<0.01 during STAGE 2 for difference
between TET-Ob-ON and both control and TET-Ob-OFF.









I Control


I-.-
c


TET-Ob(ON)
SOb-OFF


Pre-DOXY STAGE 1 STAGE 2


Figure 3-12: Serum leptin following TET-Ob or control vector delivery. Values represent
means SEM. TET-Ob was divided into TET-Ob-ON and TET-Ob-OFF to
begin STAGE 2. p<0.0001 for difference between TET-Ob and control at the
end of STAGE 1. By the end of STAGE 2, serum leptin had increased to
control levels in TET-Ob-OFF but remained low in TET-Ob-ON (p<0.0001
vs. control and TET-Ob-OFF).






4-

3-
*0


C 2


= 1 ***



Control Ob-ON Ob-OFF

Figure 3-13: Visceral adiposity (sum of retroperitoneal [RTWAT] and perirenal white
adipose tissue [PWAT] depots) following TET-Ob or control vector delivery.
Values represent means SEM. p<0.0001 for difference between TET-Ob-
ON and controls and p<0.001 for difference between TET-Ob-ON and TET-
Ob-OFF.






61




c 8- **
.oz



.LL
4-. 4-

r-
j2-


0
Control Ob-ON Ob-OFF

1 2 3 4 5 6


Leptin W




18S rRNA

Figure 3-14: Hypothalamic leptin expression 66 days after TET-Ob or control vector
delivery. Values represent means + SEM (Bar Graph, Top). p<0.001 for
difference between TET-Ob-ON and control, p<0.01 for difference between
TET-Ob-ON and TET-Ob-OFF. Image (below) is 5% acryilamide gel of
representative relative quantitative PCR product. Top band is leptin amplicon,
bottom band is 18S rRNA control amplicon for same sample, produced in
same PCR reaction multiplexingg). Lanes 1, 4: Control; lanes 2, 5: TET-Ob-
ON; lanes 3, 6: TET-Ob-OFF.









200-

o ***
150-
Ioo-



50-
000
0

Control Ob-ON Ob-OFF


1 2 3 4 5 6 7 8 9
P-STAT3 --..

Figure 3-15: Hypothalamic P-STAT3 66 days following TET-Ob or control vector
delivery. Values represent means SEM (Bar Graph, Top). p<0.001 for
difference between TET-Ob-ON and both control and TET-Ob-OFF. Image
(Below) is of representative Western Blot for P-STAT3. Primary antibody is
specific for tyrosine-phosphorylated STAT3. Lanes 1, 3, 5: Control; lanes 2,
4, 7, 9: TET-Ob-ON; lanes 6, 8: TET-Ob-OFF. No differences across groups
were detected for total hypothalamic STAT3 (image not shown).









200-

150-

100-

50-


*


01 I1
Control Ob-ON


Ob-OFF


Figure 3-16: POMC expression 66 days following TET-Ob or control vector delivery.
Values represent means SEM. p<0.01 for difference between TET-Ob-ON
and control, p<0.025 for difference between TET-Ob-ON and TET-Ob-OFF.


200-

z 150-
E
O 100-
0
C 50-

0


-F


-F


Control


Ob-ON Ob-OFF


Figure 3-17: SOCS3 expression 66 days following TET-Ob or control vector delivery.
Values represent means SEM. p<0.025 difference between TET-Ob-ON
and control, p<0.025 for difference between TET-Ob-ON and TET-Ob-OFF.











1.25-
z
rr 1.00-
C)
co


0 0.50-
0

0.25-

0.00
Control Ob-ON Ob-OFF

Figure 3-18: Long-form leptin receptor (Ob-Rb) expression in the hypothalamus 66 days
following TET-Ob or control vector delivery. Values represent means +
SEM. p<0.05 for difference between TET-Ob-ON and TET-Ob-OFF.
Difference between TET-Ob-ON and control did not reach statistical
significance (p=0.0582).

400-

300

0a 200-

a 100-

0
Control Ob-ON Ob-OFF

1 2 3 4 5 6 7 8

UCP-1 --

Figure 3-19: UCP-1 protein in BAT 66 days following TET-Ob or control vector
delivery. Values represent means + SEM (Bar Graph, Top). p<0.001 for
difference between TET-Ob-ON and both control and TET-Ob-OFF. Image
(Below) is of representative Western Blot for UCP-1. Lanes 1, 3: Control;
lanes 2, 4, 6, 8: TET-Ob-ON; lanes 5, 7: TET-Ob-OFF. Note that lanes 1-4
and lanes 5-8 came from separate gels.









Table 3-1: Brown adipose tissue parameters, serum free fatty acids, insulin, and glucose,
and CSF leptin at sacrifice
Parameter Control TET-Ob-ON TET-Ob-OFF
BAT weight (mg) 30224 21118* 29023

BAT protein 102.77. 133.1+12.2f 101.46.6
(mg/g tissue)
UCP-1 mRNA 100.06.0 119.58..9 92.69.7
(arbitrary units/pg
RNA)
UCP-1 protein 100+12 208+27** 99+15
(arbitrary units/total
BAT)
Serum free fatty 0.442.0.038 0.3930.024 0.4640.019
acids (nMol/L)
Serum insulin 27.85.5 11.63.8 46.67.2
([U/mL)
Serum glucose 169.1+12.7 154.96.0 173.68.7
(mg/dL)
CSF leptin (pg/mL) 11.62.47 5.18+0.55* 6.49 1.61


Food was withdrawn 2 hours prior to collecting serum and tissue samples. Samples were collected between
10:00 and 13:00. Data represent the mean + SEM. *p<0.025 vs. controls and TET-Ob-OFF; **p<0.01 vs.
controls and TET-Ob-OFF, ***p<0.0001 vs. controls and TET-Ob-OFF, fp<0.025 vs. TET-Ob-OFF only,
p<0.025 vs. controls and p<0.01 vs. TET-Ob-OFF.

the rate of body mass gain and reversible anorexia. There was a dramatic 85.4%

reduction in visceral adiposity in TET-Ob animals continuously administered

doxycycline for 66 days (TET-Ob-ON). In a sub-population of TET-Ob animals from

which we withdrew Doxy (TET-Ob-OFF), visceral adiposity was indistinguishable from

controls. We believe this represents a recovery of adiposity in TET-Ob-OFF animals

during STAGE 2. Consistent with the adiposity data, serum leptin was 22.5% of control

levels in TET-Ob at the end of STAGE 1. By the end of STAGE 2, serum leptin in the

TET-Ob-OFF group was similar to controls and nearly quadruple the serum leptin in the

TET-Ob-ON. Hypothalamic leptin expression was increased 8.5 fold in the TET-Ob-ON

animals, and this was reversed in TET-Ob-OFF. Leptin signal transduction and changes

in the expression of POMC and SOCS3 in the hypothalamus were also consistent with a









reversible increase in central leptin activity. UCP-1 protein levels per unit BAT tripled in

TET-Ob-ON animals and this increase was completely reversed in TET-Ob-OFF.

Finally, serum insulin was reduced in the TET-Ob-ON but not TET-Ob-OFF group.

In the first leptin gene therapy study, Unger et al. demonstrated that an intravenous

infusion of adenovirus (Ad) encoding leptin maintained hyperleptinemia for 28 days and

resulted in a complete disappearance of white adipose tissue in Wistar rats (Chen et al.,

1996). Pairfed animals experienced a modest reduction in adiposity despite similar

weight loss, thus revealing that leptin gene therapy triggers an unprecedented, selective

catabolism of white fat (Chen et al., 1996). Given that uptake of leptin across the

blood:brain barrier (BBB) appears to be a saturable process, it recently has been

hypothesized that central delivery of the leptin gene may be more therapeutically

efficacious in models resistant to peripheral leptin (Lundberg et al., 2001). In a recent

investigation, central delivery of the leptin gene using an Ad vector resulted in both

anorexia and weight loss in the normally leptin-resistant obese fa/fa rat (Muzzin et al.,

2000). However, leptin transgene expression appeared to decrease dramatically between

day 7 and 14 post-transfection, as did physiological responses. This transient transgene

expression could be the biproduct of a host-immune reaction that appears to be mounted

in response to adenoviral antigens (Yang et al., 1994).

Recombinant adeno-associated virus appears to circumvent the limitations of Ad

(Conway et al., 1999). The rAAV vector lacks virally encoded genes, thus eliminating

the risk of immune response to viral-specific antigens. Moreover, rAAV-mediated gene

delivery results in unabated transgene expression for at least 6 months (Klein et al.,

1999), whereas transgene expression appears to decline within days after Ad vector









delivery (Koyama et al., 1998; Muzzin et al., 2000). Further, rAAV gene delivery is

suitable for delivery directly into neuronal tissue (Peel et al., 1997). This allows for

transgene expression in the CNS, thus bypassing potential limitations of BBB transport.

For all of these reasons, rAAV appears to be the superior vector for leptin gene delivery.

In the present study, the leptin gene was delivered centrally by rAAV, thereby

avoiding both the problem of saturable BBB transport and the problem of unpredictable

silencing of transgene with Ad. Moreover, this is the first study to incorporate regulation

into leptin gene therapy-presumably a desirable feature for any gene delivery system to

be studied in humans. The importance of post-transfectional control becomes clear when

one considers the potential side effects of leptin transgene overexpression. For example,

a single intracerebroventricular (i.c.v) injection of constitutively active rAAV-leptin was

recently shown to impair T-lymphocyte-mediated immunity in rats (Zhang, 2002). As

discussed previously, leptin overexpression could also have negative effects on the

reproductive system (Ahima and Flier, 2000a) and skeletal system (Ducy et al., 2000).

Should such side effects appear, transgene expression could be reduced or silenced if the

gene delivery system incorporates regulation. This is the principle advantage of the TET-

Ob system described here. Without such regulation, one would have to resort to

complicated (and likely less effective) pharmacological strategies or antisense delivery to

reduce the severity of side effects.

Recently, it has been suggested that long-term over-expression of leptin can lead to

leptin resistance even in the absence of obesity (Qiu et al., 2001; Scarpace et al., 2002a).

In one study, transgenic mice that over-expressed leptin displayed minimal adiposity and

low body weights early in life, yet eventually stopped responding to the transgene despite









continuous expression (Qiu et al., 2001). Both published (Scarpace et al., 2002a;

Scarpace et al., 2003) and unpublished data from our laboratory suggest that the same is

true of leptin gene therapy. Namely, chronically elevated levels of leptin in or around the

hypothalamus following rAAV-leptin gene delivery lead to leptin resistance, even in non-

obese animals that are fully leptin responsive at the onset of gene therapy. These

observations bring to light another potential advantage of a system that incorporates

regulation. We speculate that by periodically silencing the leptin transgene, this

phenomenon of acquired leptin-induced-leptin resistance may be circumvented. After a

brief period of silencing the leptin transgene, the defective step(s) in the leptin signaling

cascade may be allowed to re-sensitize prior to another period of transgene activation.

One particularly novel finding was that leptin overexpression appears to reduce

hypothalamic leptin receptor expression, and this attenuation is reversed within 32 days

of silencing the transgene. Expression of the long-form leptin receptor, Ob-Rb, was

reduced by 30.5% in TET-Ob-ON versus TET-Ob-OFF [Figure 3-14]. This suggests that

leptin is a negative regulator of leptin receptor expression and that silencing the transgene

restores receptor expression by reducing leptin concentration in the vicinity of the

hypothalamic Ob-Rb receptors. Long term peripheral hyperleptinemia following leptin

infusion has previously been associated reduced Ob-Rb expression in the hypothalamus

(Martin et al., 2000), but our present data is the first to demonstrate that this is a specific

effect of elevated leptin in the brain and, moreover, that this effect can be reversed. It is

unknown what role, if any, this leptin-induced decrease in leptin receptor expression

plays in leptin resistance. However, these results prompted us to further study the role of

hyperleptinemia in Ob-Rb expression and function. Experiments exploring leptin









receptor expression and signaling capacity in hyperleptimemic and hypoleptinemic states

will be discussed in Chapter 5.

In the present study, hypothalamic P-STAT3, a primary Ob-Rb second messenger,

remained elevated 66 days after beginning doxy administration in the TET-Ob-ON group.

These results are consistent with those of an earlier investigation by Dr. Scarpaces's

group using constitutively active leptin gene delivery, where P-STAT3 was elevated to

the same extent 9 and 46 days post-transfection (Scarpace et al., 2002b). Taken together,

these results suggest that there is no leptin-dependent desensitization of leptin signal

transduction over the time frame of these studies. Moreover, the elevation in STAT3

phosphorylation in TET-Ob-ON in this study persisted despite an increase in expression

of SOCS3, a putative negative regulator of leptin signaling in the hypothalamus. No

elevation in hypothalamic P-STAT3 was detected in the TET-Ob-OFF group, confirming

that the amplified leptin signal transduction in the TET-Ob-ON group was dependent on

the externally regulated leptin transgene expression. Moreover, POMC mRNA levels in

the hypothalamus and UCP-1 in BAT mirrored hypothalamic P-STAT3 levels, implying

that there was no desensitization anywhere downstream of the leptin receptor. The

physiological data confirms that the TET-Ob-ON animals remained responsive to the

transgene as anorexia continued through the final days of the study.

One surprising finding was that NPY expression was not significantly depressed in

the TET-Ob-ON group at sacrifice. This is particularly perplexing in light of the

increased POMC expression in TET-Ob-ON, as leptin signaling has been shown to have

opposite and simultaneous effects on POMC and NPY expressing neurons (Elias et al.,

1999; York, 1999). In previous studies, we have reliably detected reduced hypothalamic









NPY mRNA following central leptin peptide and leptin gene delivery (Shek and

Scarpace, 2000; Scarpace, 2002). One possibility is that the TET-Ob-mediated enhanced

leptin signaling did, indeed, suppress NPY expression, but to an extent that was too small

for us to detect. A second possibility is that sometime before day 66, compensatory

pathways reversed the effects of the transgene on NPY expression. Our previous

constitutively active leptin gene therapy trial was 20 days shorter, which may not have

been long enough for this compensatory pathway to override leptin's effect on NPY

expression (Scarpace, 2002). Sahu (2002) recently demonstrated that one of the first

steps in leptin-induced-leptin resistance is desensitization of NPY neurons. Thus, our

NPY data may be a preliminary sign of acquired leptin resistance in the TET-Ob-ON

animals. The relationship between the desensitization of NPY neurons and physiological

leptin resistance will require further investigation. However, the results of the present

study suggest that the desensitization of NPY neurons may precede the loss of

physiological responses to leptin.

The reduction in CSF leptin concentrations in the TET-Ob-ON rats versus both

controls and TET-Ob-Off animals was also unexpected. Apparently, the amount of leptin

leaking (if any) from the transfected hypothalamic cells was insufficient to compensate

for the reduced amount of leptin coming across the blood brain barrier. Indeed, there was

an approximate 80% reduction in serum leptin levels in the TET-Ob-ON group at

sacrifice. Peripheral leptin, not hypothalamic leptin transgene, appears to be the

dominant factor in determining CSF leptin levels in this study. Nevertheless, the

hypothalamic leptin expression and leptin signal transduction data confirm that local

levels of leptin were elevated.









In summary, constitutively active leptin gene delivery by rAAV ("rAAV-leptin")

has been shown to cause dramatic reductions in adiposity in young adult animal models,

and we repeat these observations here (Experiment 1). Our major objective was to

incorporate regulation into this potent rAAV-leptin system (Experiment 2), and the

results of this chapter demonstrate that we achieved this goal. The TET-Ob system

allows for post-transfection control of leptin transgene expression, and ergo leptin

signaling and associated biochemical and physiological responses. Although this

experiment examined only an "on" and "off" state, it is known that expression of genes

under the control of the TET operon occurs in a doxy-dose dependent manner (Huang et

al., 1999). This suggests that, by adjusting the dose of doxy, this system could be used to

titer leptin expression to achieve optimal levels of adiposity while minimizing side

effects.














CHAPTER 4
CENTRAL LEPTIN GENE THERAPY FAILS TO OVERCOME THE LEPTIN
RESISTANCE ASSOCIATED WITH DIET-INDUCED OBESITY

Introduction

When a few injections of leptin were found to rapidly reverse obesity in the ob/ob

mouse, excitement soared over a potential cure for the growing obesity epidemic.

Unfortunately, the ob/ob mouse did not turn out to be a model of common obesity.

Typical obese humans and animals are hyperleptinemic and resistant to exogenous leptin.

Clinical trials with leptin have been disappointing due to this phenomenon of leptin

resistance in the obese state and interest in pharmacological leptin to treat obesity has

waned. Since several obese, leptin resistant models respond better to central versus

peripheral administration of recombinant leptin (Halaas et al., 1997; Niimi et al., 1999;

Van Heek et al., 1997), it has been reasoned that leptin is not reaching its hypothalamic

targets. Transport of leptin across the blood:brain barrier appears to be saturated at serum

leptin levels observed in obesity (Banks et al., 1999), suggesting that deficient

blood:brain barrier transport may play a role in leptin resistance. Thus, central delivery

of leptin may be effective in cases of peripheral leptin resistance in obese humans and

animal models.

A single intracerebroventricular injection of rAAV-leptin has been shown to cause

a rapid and complete disappearance of white adipose tissue in genetically normal young

adult F344XBN male rats (Scarpace, 2002). In a recent report, a single injection of

rAAV encoding leptin prevented diet induced obesity in young male Sprague-Dawley









rats when administered before commencing high-fat (HF) feeding (Dube et al., 2002). In

the clinical treatment of obesity, however, a more typical objective is to reverse obesity in

an already obese subject. The purpose of the present investigation was to see if a single

intracerebroventricular injection of rAAV encoding leptin could reverse the obesity

caused by 100 days of high fat feeding. To this end, we administered rAAV-leptin or

control vector to high fat fed obese and diet resistant animals as well as lean, chow fed

animals. Physiological responses to the rAAV-leptin versus control vector were

measured in all dietary groups, including anorexia, weight loss, and whole body energy

expenditure via indirect calorimetry. After sacrifice, biochemical responses to rAAV-

leptin were evaluated, including leptin signal transduction and neuropeptide expression in

the hypothalamus, and uncoupling protein concentration in brown adipose tissue.

Methods and Materials

Animals

Three-month old male Fischer 344 x Brown Norway rats were obtained form

Harlan Sprague-Dawley (Indianapolis, IN). Upon arrival, rats were examined and

remained quarantined for one week. Animals were individually caged with a 12:12 hour

light:dark cycle (07:00 to 19:00 hr). Animals were cared for in accordance with the

principles of the NIH Guide to the Care and Use of Experimental Animals.

Experimental Design

All animals were maintained on standard rat chow (diet 2018 from Harlan Teklad,

Madison, WI) from weaning until two weeks after arriving in our laboratory, at which

point animals were approximately 3 months old. This CHOW diet provides 3.3 kcal/g of

digestible energy and 15% of energy as fat. At this point, 22 animals were switched to a

high fat/high sucrose (HF) diet (F3282 from BioServ, Frenchtown, NJ, USA) while 10









continued to receive chow. This HF diet provides 5.3 kcal/g and 59.4% of energy as fat.

Animals were maintained on these respective diets through the conclusion of the

experiment. Over several weeks, the HF fed animals spontaneously divided into two

distinct groups: those that were becoming obese on the HF diet (Diet Induced Obese,

DIO) and those that were not gaining extra weight on the HF diet (Diet Resistant, DR).

After 4 weeks of HF-feeding, the top 40% of weight gainers were designated as "DIO"

and the rest of the animals on the HF-fed diet were designated as "DR". This is similar to

the designation system used previously by Levin and Keesey (1998), who defined the top

37.5 % of weight gainers on a high energy diet as DIO. DIO and DR groups were

subsequently analyzed separately. After approximately 100 days of HF feeding, all

animals (including CHOW-fed) were prepared for stereotaxis surgery as described below.

Animals were given a single injection into the third cerebroventricle of rAAV-encoding-

leptin (rAAV-leptin) or control vector encoding green fluorescents protein (rAAV-con).

This yielded 6 groups of animals: an rAAV-leptin and an rAAV-con subgroup in each of

the three original dietary groups (CHOW, DIO, and DR). These 6 groups are abbreviated

CHOW-con, CHOW-leptin, DIO-con, DIO-leptin, DR-con, and DR-leptin. Animals

were monitored for another 30 days and then sacrificed for tissue analysis.

Blood Collection

Blood was harvested from all animals on day 75. Animals were placed under

temporary anesthesia by ethrane inhalation. Using a sterile, razor sharp scapula, a small

piece (-2 mm) of the tip of the tail was excised. Then, using a gentle "milking" motion,

0.5 mL of blood was collected from the tail into a sterile microfuge tube. The blood was

allowed to clot, and then was immediately spun at 1300 G for 10 minutes. The top serum

layer was placed in a fresh tube and stored at -20 C.









Oxygen Consumption

02 consumption was assessed in up to four rats simultaneously with an Oxyscan

analyzer (OXS-4; Omnitech Electronics, Columbus, OH) as described previously

(Scarpace et al., 1997). 02 consumption is used to estimate energy expenditure in a

procedure known as "indirect calorimetry." Flow rates were 2 L/min with a 30-s

sampling time at 5-min intervals. The rats were placed into the chamber for 120 min

with the lowest 6 consecutive 02 consumption values during this period used in the

calculations (basal resting VO2). Food was withdrawn 2 hours prior to commencing

measurements. All measurements were made between 09:00 and 14:00 hrs. Results

were expressed as 02 consumption relative to metabolic body size (ml min -1 kg2/3) or

/,L/Kcal food intake.

Construction of rAAV Vector Plasmid

The vector used in this study is identical to that used in Experiment 1 of Chapter 3.

pTR-Ob encodes rat leptin cDNA [a kind gift from Roger Unger (Chen et al., 1996)] and

green fluorescent protein (GFP) reporter gene cDNA. The woodchuck hepatitis virus

posttranscriptional regulatory element (WPRE) was placed downstream to enhance the

expression of the transgenes (Loeb et al., 1999). The control vector encodes GFP driven

by a CBA. Vectors contain AAV terminal repeats at both sides of the cassette to mediate

replication and packaging of the vector (Bell et al., 1999).

Packaging of rAAV Vectors

Vectors were packaged, purified, concentrated, and titered as described previously

(Conway et al., 1999). The titer of rAAV-leptin was 2.3E13 physical particles/mL. A

mini-adenovirus helper plasmid (pDG) (Grimm et al., 1998) was used to produce rAAV









vectors with no detectable adenovirus or wild type AAV contamination. rAAV vectors

were purified using iodixanol gradient/heparin-affinity chromatography and were more

than 99% pure as judged by PAAG/silver-stained gel electrophoresis (not shown).

Vector Administration

Rats were anesthetized with 60 mg/kg pentobarbital and heads were prepared for

surgery. Animals were placed into a stereotaxic frame and a small incision (1.5 cm) was

made over the midline of the skull to expose the landmarks of the cranium (Bregma and

Lamda). The following coordinates were used for injection into 3rd cerebroventricle:

1.3 mm posterior to Bregma and 9.4 mm ventral from the skull surface on the midline

(medial fissure), with the nose bar set at 3.3 mm below the ear bars (below zero) and the

canula set at 200 posterior from vertical. A small hole was drilled through the skull and a

23-gauge stainless steel guide canula was lowered to the 3rd cerebroventricle. This was

followed by an injection canula attached to a 10uL syringe. We injected 2.5 uL of viral

particles dissolved in Ringer's solution at approximately 0.25 uL/minute. Animals

received either the rAAV-leptin (n=5 CHOW, n=5 DIO, and n=5 DR) or control virus

encoding GFP (n=5 CHOW, n=4 DIO, and n=8 DR).

Tissue Harvesting

Rats were anesthetized with 85 mg/kg pentobarbital and sacrificed by cervical

dislocation. Blood was collected by cardiac puncture and serum was harvested by a 10

minute centrifugation (1300 G) in serum separator tubes. The circulatory system was

perfused with 20 mL of cold saline. Inguinal, perirenal, and retroperitoneal white adipose

tissue, brown adipose tissue, and hypothalami were excised, weighed, and immediately

frozen in liquid nitrogen. The hypothalamus was removed by making an incision medial









to piriform lobes, caudal to the optic chiasm, and anterior to the cerebral crus to a depth

of 2-3 mm. Tissues were stored at -80 C until analysis.

Serum Leptin and FFA

Serum leptin was measured using a rat radioimmunoassay kit (Linco Research, St.

Charles, MO). Serum free fatty acids were measured using the NEFA C colorimetric kit

from WAKO Chemicals GmbH (Neuss Germany). Day 75 data was derived from serum

harvested from tail bleeding. Endpoint measurements were made on serum harvested

from a cardiac puncture at sacrifice.

Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR)

Leptin transgene expression and long-form leptin receptor (Ob-Rb) expression

were evaluated by using relative quantitative RT-PCR through the use of QuantumRNA

18s Internal Standards kit (Ambion, Austin, Tx). Total RNA (3 Gg) was treated with

RNase-free DNase using a DNA-free kit (Ambion), and first-strand cDNA synthesis was

generated from 1 [tg RNA in a 20 [tL volume using random primers (Gibco BRL)

containing 200 units of M-MLV reverse transcriptase (Gibco BRL). Relative quantitative

PCR for rAAV-leptin expression was performed by multiplexing rAAV-leptin specific

primers (sense: 5'GGCTCTGACTGACCGCGTTA; antisense: 5'

CTGCCAGGGTCTGGTCCATC) and 18s primers and coamplifing for 28 cycles, the

midpoint of the linear range for signal intensity versus number of cycles. The optimum

ratio of 18s primer to competimer was 1:9. PCR was performed at 940C denaturation for

60 sec, 59C annealing temperature for 45 sec, and 720C elongation temperature for 120

sec. Similarly, the number of cycles found to be at the mid point of the linear range was

26 cycles for ObRb. The sequence for Ob-Rb primers were sense:

GGGAACCTGTGAGGATGAGTG; antisense: TAGCCCCTTGCTCTTCATCAG. The









PCR product was electrophoresed on a 5% acrylamide TBE gel and stained with SYBR

green (Molecular Probes, Eugene, OR). Gels were scanned using a STORM fluorescent

scanner and digitized data were analyzed using Imagequant software (Molecular

Dynamics, Sunnyvale, CA).

STAT3/Phospo-STAT3 Assay

These methods were described in detail previously (Scarpace et al., 2000b).

Briefly, hypothalamus was sonicated in 10 mM Tris-HCL, pH 6.8, 2% SDS, and 0.08

ug/mL okadaic acid plus protease inhibitors (PMFS, benzamidine, and leupeptin) [an

aliquot of this sonicate was frozen for RNA analysis]. Sonicate was diluted and

quantified for protein using a detergent compatible Bradford Assay. Samples were boiled

and separated on an SDS-PAGE gel and electrotransferred to nitrocellulose membrane.

Immunoreactivity was assessed with an antibody specific to phosphorylated-STAT3

(antibody kit from New England Biolabs, Beverly MA). Immunoreactivity was

visualized by chemiluminescents detection (Amersham Life Sciences, Piscataway, NJ)

and quantified by video densitometry (BioRad, Hercules, CA). Following P-STAT3

quantification, membranes were stripped of antibody using Immunopure reagent (Pierce,

Rockford, IL) and immunoreactivity was re-assessed using a total STAT-3 antibody.

Probes

POMC mRNA is detected using a 24-mer antisense oligonucleotide probe (5'-

CYYGCCCACCGGCTTGCCCCAGCG-3'). The oligonucleotide probe was end labeled

by terminal deoxynucleotidyl transferase (Promega). The AgRP cDNA probe was

provided by Dr. Michael Schwartz (University of Washington). The rat pre pro NPY

cDNA was provided by Janet Allen (University of Glasgow, UK). SOCS3 cDNA was a

gift from Christian Bjorbaek (Harvard University). The cDNA probes were labeled using









a random primer kit (Prime-a-Gene, Promega, Madison, WI). Probes were purified with

Nick columns (Pharmacia) and, except for oligonucleotide probes, were heat-denatured

for 2 minutes. All probes have been verified to hybridize to the corresponding specific

mRNAs by Northern Analysis prior to use in Dot Blot assay (below).

mRNA Levels (Dot Blot Analysis)

Tissue was sonicated in guanidine buffer, phenol extracted, and isopropanol

precipitated using a modification of the method of Chomczynski (Chomczynski and

Sacchi, 1987). Isolated RNA was quantified by spectrophotometry and integrity was

verified using 1% agarose gels stained with ethidium bromide. For dot blot analysis,

multiple concentrations of RNA were immobilized on nylon membranes using a dot blot

apparatus (BioRad, Richmond, CA). Membranes were baked in a UV crosslinking

apparatus. Membranes were then prehybridized in 10 mL Quickhyb (Stratagene, LaJolla,

CA) for 30 minutes followed by hybridization in the presence of a labeled probe and 100

tg salmon sperm DNA. After hybridization for 2 hours at 650C, the membranes were

washed and exposed to a phosphor imaging screen for 24-72 hours (depending on

anticipated strength of signal). The screen was then scanned using a Phosphor Imager

(Molecular Dynamic, Sunnyvale, CA) and analyzed by Image Quant Software

(Molecular Dynamics).

Statistical Analysis

All data are expressed as mean + standard error of mean. Body mass and food

intake during the HF-feeding period were compared by 2-WAY ANOVA, with dietary

group and time serving as factors. Total weight gain through day 75 was compared by 1-

WAY ANOVA with Tukey's post-hoc. Comparisons in cumulative caloric intake









(through day 5) and average daily caloric intake (day 6-100) were made by 1-way

ANOVA. A 1-way ANOVA was used to compare serum leptin, serum free fatty acids,

and oxygen consumption in the three dietary groups (CHOW, DIO, and DR) prior to

vector delivery. When 1-way ANOVA was significant, Tukey post-hoc was used to

evaluate pairwise comparisons. After vector delivery, comparisons were made by 2-Way

ANOVA on the now 6 groups (each dietary group was divided into rAAV-leptin and

control subgroups) with dietary group and vector as factors. When only 2-Way ANOVA

main effects were significant, relevant pairwise comparisons were made using the

Bonferroni Multiple Comparison method with the error rate corrected for the number of

contrasts. When there was an interaction, factors were separated and a further 1-Way

ANOVA was applied with a Bonferroni Multiple Comparison post-hoc. When separation

of factors resulted in only two population means to compare, the 1-WAY ANOVA was

replaced with student's t-test. a-level was set at 0.05 for all analyses. GraphPad Prism

software (San Diego, CA) was used for all statistical analysis and graphing. Post-hoc

analysis of 2-Way ANOVAS was done using GraphPad QuickCalc (GraphPad.com).

Results

Part I: High-Fat Feeding

Food consumption and body mass

Male F344XBN rats, age 3 months, were either maintained on standard chow diet

(CHOW) or a high-fat high sucrose (HF) diet as described in the Methods section. The

HF-fed animals spontaneously divided into two distinct groups: those that were

becoming obese on the HF diet (Diet Induced Obese, DIO) and those that were not

gaining extra weight on the HF diet (Diet Resistant, DR). DIO animals gained mass at a

greater rate than DR and CHOW, and this trend became significant by day 25 (Figure 4-









1). By day 75, DIO animals had gained over 25% more mass than both CHOW and DR

[p<0.001] (Figure 4-1). During the first week of high fat feeding, acute hyperphagia

was observed in all HF-fed animals (both DIO and DR). Cumulative caloric intake was

nearly 30% greater in HF- fed animals than CHOW-fed animals during the first 5 days of

HF-feeding (p<0.001), but this acute hyperphagia attenuated by day 7 (Figure 4-2).

After this acute phase, caloric intake was similar in DIO and CHOW while DR animals

consumed significantly less calories than both DIO and CHOW during this phase of the

experiment (p<0.01, Figure 4-2).

Oxygen consumption

Oxygen consumption (VO2) was first measured 30 days after beginning HF-

feeding. At this point, oxygen consumption (mL/min/kg2/3) was significantly reduced in

DIO compared to both CHOW and DR (p<0.01), suggesting that reduced energy

expenditure in DIO contributes to their accelerated rate of weight gain during this phase.

By day 70, there was no longer a difference in VO2 in the three groups (Figure 4-3).

Serum leptin and free fatty acids

At day 75, serum leptin was significantly greater in DIO compared to DR (p<0.01,

Table 4-1). CHOW animals had lower serum leptin than both DIO and DR (p<0.001,

p<0.01, respectively). Serum FFA did not differ across groups at this point in the study

(Table 4-1).




















0 25


-- CHOW
-o- DIO
- DR


50 75 100


Day

Figure 4-1: Body mass during high fat feeding (pre-vector delivery). Values represent
means SEM. By 2-Way ANOVA, significance was found for effects of
dietary group (F=46.38, p<0.0001), time (F=263.00, p<0.0001), and the
interaction between group and time (F=1.71, p<0.05).

100-
90-- CHOW
90-
-o-DIO
S80- DR

70


0 25 50 75 100


Day


Figure 4-2: Caloric intake during high fat feeding (pre-vector delivery). Values represent
means + SEM. By 2-Way ANOVA, significance was found for effects of
dietary group (F=31.59, p<0.0001), time (F=12.00, p<0.0001), and the
interaction between group and time (F=2.62, p<0.0001). Note the biphasic
pattern of caloric intake in all HF-fed (DIO and DR) animals demonstrated
acute hyperphagia after being switched to the high fat diet. After this transient
hyperphagic response to the diet subsided (day 7), average caloric intake did
not differ between DIO and CHOW for the rest of the study. Mean daily
caloric intake was significantly lower in DR compared to both DIO and
CHOW after day 7 (p<0.01).











120-

4! 110-

0.. 100-

So 90- -- CHOW
-o-DIO
80 DR
-*- DR
70-
20 30 40 50 60 70 80
Day

Figure 4-3: Oxygen consumption on day 30 and on day 70 after commencing HF-feeding
(pre-vector delivery). Values represent mean percent of control + SEM.
p<0.01 for difference between DIO and CHOW on day 30 and difference
between DIO and DR on day 30.



Table 4-1: Serum leptin and free fatty acids after 75 days of high-fat feeding (DIO and
DR).
CHOW DIO DR
Serum Leptin 4.900.52 13.291.43a 9.050.58
(ng/mL)
Serum FFA 0.550.059 0.440.045 0.640.055
(meq/L)
Data represent meanSEM. p values represent results of post-hoc analysis following 1-WAY ANOVA.
a p<0.001 versus CHOW, p<0.01 versus DR
b p<0.01 versus CHOW and DIO

Part II: Post rAAV-Leptin Delivery

Food consumption and body mass

After approximately 100 days of HF-feeding in DIO and DR, animals in all dietary

groups (i.e., CHOW, DIO, DR) were given a single i.c.v. injection containing 5.75E10

physical particles of rAAV encoding leptin (rAAV-leptin) or control virus (rAAV-con).

CHOW-leptin had a robust response to rAAV-leptin, losing 6.6% of their body mass

while CHOW-con animals increased their body mass by 3.2% during the same 29 days of