Title: Effects of recombinant adeno-associated virus encoding leptin on body weight regulation and energy homeostasis
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00100764/00001
 Material Information
Title: Effects of recombinant adeno-associated virus encoding leptin on body weight regulation and energy homeostasis
Physical Description: Book
Language: English
Creator: Dhillon, Harveen, 1971-
Publisher: State University System of Florida
Place of Publication: Florida
Florida
Publication Date: 2000
Copyright Date: 2000
 Subjects
Subject: Neuroscience thesis, Ph. D   ( lcsh )
Dissertations, Academic -- Neuroscience -- UF   ( lcsh )
Genre: government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )
 Notes
Abstract: ABSTRACT: Leptin is a weight reducing hormone synthesized by white adipose tissue. Administration of leptin reduces body weight (BW) in obese and normal rodents. However, obesity is not a result of leptin deficiency, instead excess leptin accompanies human obesity suggesting resistance to leptin actions. We hypothesize that leptin resistance is due to insufficient availability of leptin at target sites within the brain. We employed recombinant adeno-associated virus encoding leptin (rAAV-leptin) gene therapy to enhance leptin production. An intravenous injection of rAAV-leptin to leptin deficient ob/ob mice increased blood leptin levels and reduced BW in a dose dependent manner. When administered intracerebroventricularly (icv), it increased leptin mRNA in the hypothalamus, and suppressed BW gain without decreasing food intake (FI) in adult lean male and female Sprague-Dawley rats. A single injection icv of rAAVleptin regulated BW for six months without any evidence of leptin resistance. Our data show a dose-dependent dichotomy in the response to icv rAAV-leptin. While rats receiving 5 X 10 to the 10th power particles of rAAV-leptin icv maintained their pre-injection BW, a 2 fold higher dose caused a 10-15% decrease in BW accompanied by a significant reduction in FI. UCP-1 mRNA in brown adipose tissue (BAT) was enhanced with rAAV-leptin with both the high and low dose, indicating that increased leptin production in the hypothalamus enhanced energy expenditure via increased thermogenesis.
Abstract: Analysis of body composition revealed a marked decrease in body fat without altering lean mass. Serum leptin and insulin levels were reduced, however, blood glucose levels were normal. Hypothalamic expression of the appetite regulating pro-opiomelanocortin (POMC) and Neuropeptide Y (NPY) genes were altered with the higher dose only. Thus, we show for the first time effective use of gene therapy for long term BW regulation. A single central injection of rAAV-leptin reduced BW without development of leptin resistance. Whereas lower levels of centrally produced leptin reduce BW by increasing energy expenditure, higher levels reduce BW both by increasing energy expenditure and by decreasing FI via an increase in hypothalamic POMC and decrease in NPY signaling.
Summary: KEYWORDS: obesity, leptin, gene therapy, AAV, hypothalamus
Thesis: Thesis (Ph. D.)--University of Florida, 2000.
Bibliography: Includes bibliographical references (p. 149-170).
System Details: System requirements: World Wide Web browser and PDF reader.
System Details: Mode of access: World Wide Web.
Statement of Responsibility: by Harveen Dhillon.
General Note: Title from first page of PDF file.
General Note: Document formatted into pages; contains xiii, 171 p.; also contains graphics.
General Note: Vita.
 Record Information
Bibliographic ID: UF00100764
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 78922739
alephbibnum - 002678721
notis - ANE5948

Downloads

This item has the following downloads:

dissertation ( PDF )


Full Text












EFFECTS OF RECOMBINANT ADENO-ASSOCIATED VIRUS ENCODING
LEPTIN ON BODY WEIGHT REGULATION AND ENERGY HOMEOSTASIS
















By


HARVEEN DHILLON


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


2000















ACKNOWLEDGMENTS

This dissertation has been made possible through the support and encouragement

I have received from faculty, friends and family here in Gainesville and around the world.

I first want to thank my mentor, Dr. Pushpa Kalra, for all the time she has made for me

over the years. Dr. Kalra has helped me become independent; I thank her for having faith

in me and taking me on as her graduate student. Dr. Kalra has been a perfect role model,

a mentor and a friend. I also want to acknowledge Dr. Satya Kalra for unofficially co-

mentoring me in my Ph.D. endeavors. He has been good humored and kind at all times,

going well out of his way to help me in times of personal crisis. My committee members

Dr. Streit and Dr. Schultz have assisted me to grow intellectually and I thank them. Dr.

Streit has also been part of the stress relief operation at the Market Street Pub when

occasion demanded.

I am very grateful to my colleagues in the lab; in particular Dr. Michael Dube,

who taught me the ropes when I started, and has always been ready with a witty remark

to lighten things up. Dr.s Elena Beretta and Michela Bagnasco, "the Italians" who have

been there when I just did not have enough hands, and who in addition to scientific input

introduced me to the pleasures of (almost) all things Italian. Erin Rhinehart, my fellow

graduate student, and dear friend, has been a calm rational voice of reason that helped me

keep things in perspective many times. I want to thank Laura Dixon, a new member of

the lab and a new friend, who has been extremely helpful and supportive in every way

possible.









I thank Drs. Zolotukhin, Scarpace, and Moldawer for allowing me access to their

laboratory facilities. I am grateful to Dr. Bill Farmerie for introducing me to the "fun" in

molecular biology, Drs. Gerry Shaw, Colin Sumners, Mohan Raizada, Charlie Wood for

their support over the years, and all my friends for lending me their ears all along.

I want to thank my husband from the bottom of my heart. Without his love,

support and chauffeuring I would never have reached this far and lastly I thank my

parents, my mother- in- law, my sister and brother who have been wonderful and

supportive at all times.















TABLE OF CONTENTS

page

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

LIST OF TABLES ................. ......................................... .............. .. viii

LIST OF FIGURES........... ......................... .. ... ..... ..............ix

A B S T R A C T ............. ......... .. ............. .. ...................... ............................. ...............x ii

CHAPTER

1. L IT E R A TU R E R E V IE W ................................................................ ..........................1

Neuroanatomy of Appetite Regulation.................................................... 1
M olecules Involved in the Regulation of Appetite ..................................... ............ .. 2
N europeptide Y ................................... .. ............. 4
O rex in s ........................................................................................................................ 4
M elan ocortin s ............. ..... ............. ....................... .... ................................ . . 5
L e p tin ........................ .................... .................................................................................. 6
Leptin Receptors ....................................... ............... 10
Leptin Interaction with Appetite Regulating Molecules ................. ....................... 12
Leptin and N PY ....................................... ................ .......... ... ..... ..... 12
Leptin and the M elanocortins................................................ .............. 13
L eptin and In sulin ........................................................................... . ........... 14
Fat tissue and Uncoupling Proteins in Energy Metabolism ..................................... 16
L eptin and U ncoupling Proteins......................................................... ....................... 18
Adenoassociated Virus and Recombinant Adenoassociated Virus......................................... 19
Study D esign and R ationale .......................................................................................... 2 1


2. G E N E R A L M E TH O D S ........................................................................ ..................26

Experim mental A nim als ............................................................... ...26
Third Ventricle Cannulation and Injection................ .................................... 26
B lood C collection ...................................................... ............. 27
Cerebro-spinal Fluid Collection .............. ..................................... .............. 27
Food Intake M easurem ent ...................... .... ......... ........................ .............. 28
U rine C ollection............................................ 28
O rg an C collection .............................................................. ......... .. ............ 2 8
Carcass Fat and Protein Estimation.................................. 29


iv









O xygen C onsum ption...................................................................... .............. 29
RNA Isolation....................................... ........ 30
Leptin mRNA Expression using RT-PCR........................................................ 30
Dot Blot Analysis for UCP-1 and UCP-3............. .............. .......... ............. 31
In Situ H ybridization (ISH ) ............................................... ........................... 33
C construction of R iboprobes.......................................................... .. .................. 33
Tissue Sectioning............... ......... .......... ......... 34
T issue e P ro cessin g ........... .... ...... ................................................... .... 34
A analysis of ISH data.................. ..... .... ........ .... .... ...... ................ ... 35
Immunohistochemistry for Green Flourescent Protein (GFP).............................. 35
R adioim m u n oassay s................................................................................... .. 36
Leptin............................................. .............. 36
In su lin ........................................................................................ 3 6
Norepinephrine ............................ ..... .............. 37
Thyroid Horm ones (T3 and T4)................................................ .................... 37
Free Fatty Acid Analysis............................. .............. 37
G lucose M easurem ent..................... .......................... .... ...... .. ............ 38
Recombinant AAV Production........................................ ......................... 38
Statistical A naly sis .............. ..... .... ......... .. .......................... ........................... 40


3. LEPTIN GENE THERAPY REVERSES OBESITY IN OB OB MICE ...................42

Introduction....................... ............... ..... ............. 42
M materials and M ethods............................................................ ................................. 43
A n im als ........................................................................... 4 3
Recom binant A A V Production............................................ ........... .............. 43
Study D design ................... ...... ..... ........... ...... ... ..... .... 45
D ata A nalyses............................ ............... ..... 46
Results ............. ............. ......... ..... .... ................. 46
D is c u s sio n ..................................................................................................................... 5 1


4. REGULATION OF BODY WEIGHT WITH LEPTIN GENE THERAPY IN LEAN
SPRAGUE-DAWLEY RATS .................. .................................... ................54

Introduction................................. .............. 54
M ethods.............................. .............. ...... 56
Study Design ...................................................................... ........ 56
Carcass Fat and Protein Estimation................................................. .... ........ 57
Immunohistochemistry for Green Flourescent Protein (GFP) ............................ 57
Leptin mRNA Expression using RT-PCR 5........... ......... ........... ... .......... .. 58
Statistical A n aly sis ............... ......... ...... ............... ............. ............... 59
Results ................... .. ................ .................. ....................... 60
Hypothalamic Leptin mRNA RT-PCR ......................................... 60
Immunohistochemical Localization of GFP ............ ..................................... .... 60
B ody W eight ......... .......... ...................................... .............. 6 1


v









Food Intake........................................................... 61
Serum Leptin Levels .............. ............................................ ............ .... .... ...... 61
C arcass Fat and Protein A nalysis.................................................... ... ................. 62
D discussion ................................................ 7 1


5. LONG TERM EFFECTS OF LEPTIN GENE THERAPY ........................................75

Introduction ............................ .............. ...... 75
M materials an d M eth od s ................................................................................................. 7 9
S tu d y D e sig n ....................................................................................................... 7 9
R adioim m unoassays................................................ .. .. .... .. .. .............. 79
L e p tin ..................................................... 7 9
Insulin......................................... 80
N orepinephrine ............................................ ........... ............ ..... 80
Thyroid horm ones (T3 and T4).................................................. ... ................. 80
Dot Blot Analysis for UCP-1 and UCP-3.......................... .. ..... ............. 81
Statistical A naly sis ........ .. .......... .. .......... ...................... .. ...... ........... 82
Results ............... ......... .......................... ................ 82
Body Weight ......... .... ..... ..................... 82
Food Intake................................... ............. 83
L eptin L levels ........................................ 83
Insulin Levels ....................................... ............. 84
G lucose L evels ................. ................... ............... ..... 84
Urinary Norepinephrine Levels...... ...... ......... .................... 85
Thyroid H orm one L evels...................................................................... ........ 85
UCP-1 mRNA Expression..................................................... 85
UCP-3 mRNA Expression..................................................... 85
Discussion ............................. .............. 94


6. DOSE DEPENDENT REGULATION OF BODY WEIGHT WITH rAAV-LEPTIN100

In tro d u ctio n ........................................................................................ 10 0
Materials and Methods......................................... 103
Study D design ............................................................................................... ......... 103
Experiment 1 ........................................... ............ .............. 103
E xperim ent 2 .................................................................. ............................ 103
In Situ Hybridization (ISH) ............................................. 104
Construction of riboprobes ........... ......................... 104
Tissue sectioning ................................ ..................................... 104
Tissue processing ..................................... 105
Analysis of ISH data .............................................................. ... ........ 105
L ep tin ..................................................... 10 6
In sulin ....................................................................................................... 106
Statistical Analysis ................................. ........................... ... ........ 107
R e su lts ........................................................................................... 1 0 7









B o d y W e ig h t ................................................................. .................................. 1 0 7
Food Intake................................. .............. 108
L eptin .................................................... 10 8
CSF Leptin and CSF:Serum Ratios................................................... 109
In su lin ........................................................................................................ 1 0 9
G lu co se ...............................1 09...............................
Thyroid Hormones .. ................. .................. ....... .......... .. 109
F ree F atty A cid L levels ............................ .............................. .... .............. 110
U C P -1 ........ ........................................................... 1 10
U C P 3 ........................... ...... ................... ..... ................... 110
Brain NPY, POMC, AGRP mRNA Expression............................. .............. 111
Discussion....................... ............... ...... .............. 126


7. G EN E R A L D ISC U SSIO N ........................................ ..........................................133

L IST O F R E FE R E N C E S ......................................................................... ...................149

B IO G R A PH ICA L SK ETCH ............................................................................ ........ 171
















LIST OF TABLES


Table Page

1-1. List of major appetite regulating peptides ............. ............................................3

2-1. Riboprobes for in situ hybridization...................... ..... ............................ 33

5-1. Time related changes in circulating leptin, insulin and glucose levels in female
Sprague-D aw ley rats. ..................... .. .... ....................................... 89

5-2. Time related changes in circulating leptin, insulin and glucose levels in male
Sprague-D aw ley rats. ..................... .......................... ......... ........... 89

6-1. Serum hormone and glucose profiles at week 6 post-injection. ...............................118


















LIST OF FIGURES


Figure Page


Fig. 3-1. Schematic diagrams of the rAAV vector constructs used in the study ..................48

Fig. 3-2. Dose dependent change in body weight in ob/ob mice post iv rAAV-leptin
inj section. ........... ............................... ............... 48

Fig. 3-3. Effect of different doses of rAAV-leptin on food intake in ob/ob mice ................49

Fig. 3-4. Serum leptin levels post iv injection of rAAV-leptin.................. ..................49

Fig. 3-5. R representative ob/ob m ice........................................................... ............... 50

Fig. 4-1. Effect of a single injection of rAAV-leptin on leptin mRNA expression in the
hypothalamus of female Sprague-Dawley rats................... ........... ............... 63

Fig. 4-2. Photomicrograph (4X) of representative hypothalamic sections showing GFP
positive cells around the site of icy rAAV-UF5 injection...............................64

Fig. 4-3. Photomicrograph (20 X) of a representative hypothalamic section showing GFP
immunoreactivity in neurons and fibres transduced by rAAV-UF5...................65

Fig. 4-4. Effect of rAAV-leptin on BW in female Sprague-Dawley rats.............................. 66

Fig. 4-5. Effect of a single injection of rAAV-leptin on BW in male Sprague-Dawley
rats.. ........................................................................... . . 6 6

Fig. 4-6. Twenty four hour food intake in female Sprague-Dawley rats injected icy
rAAV-UF5 (control) or rAAV-leptin ...... ......... ....................................... 67

Fig. 4-7. Twenty four hour food intake in male Sprague-Dawley rats injected icy with
rAAV-UF5 (control) or rAAV-leptin................................ ..................... 67

Fig. 4-8. Serum leptin levels in female Sprague-Dawley rats..............................................68

Fig. 4-9. Serum leptin levels in male Sprague-Dawley rats........................................ 68









Fig. 4-10. Body composition analysis post rAAV-leptin injection................... ............69

Fig. 4-11. Effect of rAAV-leptin on body fat. ............... ..................................................70

Fig. 5-1. Effect of long term leptin gene therapy on body weight in female Sprague-
D aw ley rats ........................................................................86

Fig. 5-2. Effect of long term leptin gene therapy on body weight in male Sprague-
D aw ley rats ........................................................................86

Fig. 5-3. Effect of rAAV-leptin on food intake in female Sprague-Dawley rats ................87

Fig. 5-4. Effect of rAAV-leptin on food intake in male Sprague-Dawley rats ...................87

Fig. 5-5. Effect of central rAAV-leptin injection on circulating leptin levels in female
Sprague-D aw ley rats.. ................. .. ....................................... .......... .... .. 88

Fig. 5-6. Effect of central rAAV-leptin injection on circulating leptin levels in male
Sprague-D aw ley rats...................... .. .... ........................................... 88

Fig. 5-7. Effect of central rAAV-leptin injection on urinary norepinephrine levels in
female Sprague-Dawley rats at week 16 post -injection................... ................90

Fig. 5-8. Effect of central rAAV-leptin injection on urinary norepinephrine levels in
male Sprague-Dawley rats at week 16 post -injection.............................90

Fig. 5-9. Effect of rAAV-leptin on thyroid hormones in female Sprague-Dawley rats at
16 w eeks post injection.. .................................... .........................91

Fig. 5-10. Effect of rAAV-leptin on thyroid hormones in male Sprague-Dawley rats at 16
w weeks post injection.. ........................................ ... ......... ................. 91

Fig. 5-11. Effect of rAAV-leptin on UCP-1 mRNA expression in BAT at 24 weeks post
injection in female Sprague-Dawley rats. ...................................................92

Fig. 5-12. Effect of rAAV-leptin on UCP-1 mRNA expression in BAT at 24 weeks post
injection in female Sprague-Dawley rats. ...................................................92

Fig. 5-13. Effect of rAAV-leptin on UCP-3 mRNA expression in skeletal muscle 24
weeks post injection in female Sprague-Dawley rats........................................93

Fig. 5-14. Effect of rAAV-leptin on UCP-3 mRNA expression in skeletal muscle at 24
weeks post injection in male Sprague-Dawley rats.........................................93

Fig. 6-1. Effect of low dose rAAV-leptin on body weight in lean female Sprague-Dawley
rats.. ............................................... ........... ............ ................. . 1 12

Fig. 6-2. Effect of high dose rAAV-leptin on body weight in lean female Sprague-
D aw ley rats............................................................................................. 112









Fig. 6-3. Effect of high dose of rAAV-leptin on food intake in female Sprague-Dawley
ra ts ............................................................................ .1 1 3

Fig. 6-4. Dose dependent effects of rAAV-leptin injection on serum leptin (A) Low dose
(B) High dose .................................................................... ......... ....... 114

Fig. 6-5. Dose effects of icy rAAV-leptin on serum insulin levels in female SD rats 6
w weeks post injection. .................................. ..... .......... .... .. .. ............ 115

Fig. 6-6. Effect of low dose rAAV-leptin on cerebro spinal fluid levels of leptin (A) and
C SF:serum leptin ratio (B ) ........................................................ ............. ..116

Fig. 6-7. Effect of high dose rAAV-leptin on cerebrospinal fluid levels of leptin (A) and
C SF:serum leptin ratio (B ) ........................................................ ............. ..117

Fig. 6-8. Effect of rAAV-leptin on serum free fatty acids..................... ...............19

Fig. 6-9. Change in UCP-1 mRNA in BAT with different doses of rAAV-leptin .............120

Fig. 6-10 Effect of rAAV-leptin on skeletal muscle UCP-3 expression ............................120

Fig. 6-11. Photomicrograph (10X) of the effects of high dose rAAV-leptin on POMC
mRNA expression in the Arcuate nucleus of the hypothalamus.......................121

Fig. 6-12. Photomicrograph (10X) of the effects of high dose rAAV-leptin on NPY
mRNA expression in the Arcuate nucleus of the hypothalamus....................... 122

Fig. 6-13. Photomicrograph (10X) of the effects of high dose rAAV-leptin on AGRP
mRNA expression in the Arcuate nucleus of the hypothalamus.......................... 123

Fig. 6-14. Effect of low dose rAAV-leptin on hypothalamic appetite regulating peptides.. 124

Fig. 6-15. Effect of the high dose of rAAV-leptin on hypothalamic appetite regulating
neuropeptides. .................................................................... 125

F ig 7-1. Sum m ary diagram ................................................................................ ........ 148















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

EFFECTS OF RECOMBINANT ADENO-ASSOCIATED VIRUS ENCODING
LEPTIN ON BODY WEIGHT REGULATION AND ENERGY HOMEOSTASIS

By

Harveen Dhillon

December 2000

Chair: Pushpa S Kalra
Major Department: Neuroscience

Leptin is a weight reducing hormone synthesized by white adipose tissue.

Administration of leptin reduces body weight (BW) in obese and normal rodents.

However, obesity is not a result of leptin deficiency, instead excess leptin accompanies

human obesity suggesting resistance to leptin actions. We hypothesize that leptin

resistance is due to insufficient availability of leptin at target sites within the brain.

We employed recombinant adeno-associated virus encoding leptin (rAAV-leptin)

gene therapy to enhance leptin production. An intravenous injection of rAAV-leptin to

leptin deficient ob/ob mice increased blood leptin levels and reduced BW in a dose

dependent manner. When administered intracerebroventricularly (icv), it increased leptin

mRNA in the hypothalamus, and suppressed BW gain without decreasing food intake

(FI) in adult lean male and female Sprague-Dawley rats. A single injection icy of rAAV-

leptin regulated BW for six months without any evidence of leptin resistance. Our data

show a dose-dependent dichotomy in the response to icy rAAV-leptin. While rats









receiving 5 X10 10 particles of rAAV-leptin icy maintained their pre-injection BW, a 2

fold higher dose caused a 10-15% decrease in BW accompanied by a significant

reduction in FI. UCP-1 mRNA in brown adipose tissue (BAT) was enhanced with

rAAV-leptin with both the high and low dose, indicating that increased leptin production

in the hypothalamus enhanced energy expenditure via increased thermogenesis. Analysis

of body composition revealed a marked decrease in body fat without altering lean mass.

Serum leptin and insulin levels were reduced, however, blood glucose levels were

normal. Hypothalamic expression of the appetite regulating pro-opiomelanocortin

(POMC) and Neuropeptide Y (NPY) genes were altered with the higher dose only.

Thus, we show for the first time effective use of gene therapy for long term BW

regulation. A single central injection of rAAV-leptin reduced BW without development

of leptin resistance. Whereas lower levels of centrally produced leptin reduce BW by

increasing energy expenditure, higher levels reduce BW both by increasing energy

expenditure and by decreasing FI via an increase in hypothalamic POMC and decrease in

NPY signaling.














CHAPTER 1
LITERATURE REVIEW

The survival of higher organisms, including mammals, depends upon the

maintenance of adequate BW. In order to maintain a stable BW over an extended period

of time it is essential that the amount of energy intake is matched by the energy

expenditure. Regulation of energy expenditure for all the energy requiring processes in

the body is a complex and tightly controlled process. The physiological control of

energy homeostasis though effective is not perfect often resulting in obesity which is on

the rise in the western world. In the US, obesity has reached epidemic proportions

(Wilding et al., 1998) with nearly half the adult population of the US being clinically

obese. It is now generally believed to be caused by dysregulation in the balance between

food intake and energy expenditure rather than a loss of control over the amount of food

ingested. Obesity is a major contributing factor for medical conditions such as non-

insulin dependent diabetes mellitus (NIDDM), infertility, hypertension and

cardiovascular disease. Failure to contain the obesity epidemic now will only further

burden the health care system in the US.





Neuroanatomy of Appetite Regulation

Studies involving twins, adoption studies, analysis of familial aggregation as well

as studies from several animal models of obesity, point towards obesity being a result of

both genetic and environmental factors (Stunkard et al., 1990; Coleman, 1978) and not









just a lack of will power. Current literature supports the hypothesis that both hyperphagia

and anorexia are brought about by the induction of environmental, genetic or hormonal

changes in neurochemical signaling within the hypothalamus (Kalra SP, 1997; Kalra SP

et al., 1998). Over half a century ago lesions of the ventromedial nucleus (VMH) of the

hypothalamus were shown to result in increased BW and obesity (Anand and Brobeck ,

1951). Based on the early lesion studies, the VMH was termed the "satiety "center. In

contrast, lesions of the lateral hypothalamic nucleus (LHA) lead to aphagia; this nucleus

was thus termed the "feeding" center (Hetherington and Ranson, 1940). It is now

known that there are discrete sites in the hypothalamus associated with appetite


There is an increasing awareness that expression of appetite is chemically coded

in the hypothalamus in the central nervous system (reviewed in: Kalra SP, 1997; Kalra

SP et al.,1999; Morley 1987). An array of neurotransmitters and neuromodulators

localized in the central nervous system (CNS) forms a distinct circuitry of pathways

comprised of both appetite stimulating (orexigenic) and appetite inhibiting anorexigenicc)

signals (Kalra SP 1997; Kalra SP and Kalra PS 1996). Appetite is thus regulated by a

complex network of interconnected over-lapping pathways of neuromodulators in the

brain, however, these central regulatory mechanisms are complex and are not yet fully

understood.


Molecules Involved in the Regulation of Appetite

CNS neurons involved in the production of orexigenic and anorexigenic signals

are regulated both by peripherally secreted hormones that cross the blood brain barrier,

and by other neurotransmitters within the brain. A change in availability of neural









orexigenic molecules such as Neuropeptide Y (NPY) and Agouti related peptide (AGRP)

or anorexigenic molecules such as alpha-melanocyte stimulating hormone (cMSH) and

cocaine and amphetamine regulated transcript (CART) precede the onset of feeding

(Kalra SP et al., 1998). Disruption of the neural microenvironment through either

changes in the amount or signaling capability of these molecules may lead to hyperphagia

and subsequent obesity on the one hand or to anorexia and weight loss on the other (Kalra

SP et al., 1998).


Table 1-1: List of major appetite regulating neuropeptides
Orexigenic peptides Anorexigenic peptides


Neuropeptide Y (NPY, Clark et al., 1984) Proopiomelamocortin (POMC), alpha

melanocyte stimulating hormone (cMSH,

Lu et al., 1994))


Agouti related peptide (AGRP, Ollmann et Cocaine and amphetamine regulated

al., 1997) transcript (CART, Kristensen et al., 1998)



Orexins (Sakurai et al., 1998) Leptin (Zhang et al., 1994)


Melanin Concentrating Hormone (MCH,

Qu et al., 1996)









Neuropeptide Y

NPY is the most potent stimulant of appetite and feeding behavior currently known

(Kalra SP,1997). Our laboratory has shown that NPY injected intracerebroventricularly

(icv) causes increased feeding in satiated rats (Clark et al.,1984). An important

physiologic role for NPY in the regulation of appetite was suggested by the

demonstrations that NPY mRNA is up-regulated in food deprived rats (Sahu et al., 1992),

NPY secretion increases in the paraventricular nucleus (PVN) of the hypothalamus in

association with increased appetite for food and NPY antibodies block feeding in fasted

rats (Kalra SP et al.,1991). NPY neurons are localized in highest density in the arcuate

nucleus (ARC) of the hypothalamus, and project to the PVN where NPY is released from

terminals into the extracellular compartment (Kalra SP et al., 1998). Five NPY receptors

Y1, Y2, Y4, Y5 and Y6 have been cloned (Larsen et al.,1993; Kanatani et al.,1990;

Schaffhauser et al., 1997; Hu et al., 1996) however, feeding receptor has not been

unequivocally identified. Both Y1 and Y5 receptors have been implicated in separate

studies as being involved in feeding behavior (Larsen et al., 1993; Kanatani et al., 1990,

Schaffhauser et al., 1997; Hu et al.,1996).

Orexins

Orexins, two recently discovered peptides localized in the lateral hypothalamus, also

stimulate feeding behavior (Sakurai et al., 1998). Both Orexin A and B stimulate food

intake, although Orexin A, a 33 amino acid peptide is the more active form (Sakurai et

al., 1998). There are reciprocal synaptic contacts between orexin producing cells in the

LHA and NPY neurons as well as POMC producing neurons in the ARC, thus, raising

the possibility that orexins may be involved in the regulation of feeding. Our laboratory

has identified the sites of action of Orexin A to be the LHA, PVN and perifomical area









(PFH) of the hypothalamus (Dube et al., 1998). Although not as potent as NPY, orexins

elicit an increase in food intake

Melanocortins

Melanocortins are peptides cleaved from the pro-opiomelanocortin (POMC)

precursor molecule that exert their effects by binding to members of a family of

melanocortin receptors. POMC is differentially processed in various neurons to produce

a variety of peptides including the orexigenic peptide, P-endorphin, and the anorexigenic

peptide, cMSH. Disruption of proopiomelanocortin processing leads to obesity

(Yaswen et al., 1999; Krude et al., 1998). cMSH, a melanocortin receptor 4 (MC-4R)

agonist, reduces food intake (Lu et al., 1994). Its activity is antagonized by Melanin

Concentrating Hormone (MCH, not derived from POMC), and by AGRP (Ollmann et al.,

1997; Considine and Caro, 1997). Targeted disruption of the MC-4 R in mice leads to

hyperphagia and subsequent obesity (Huszar et al., 1997).


AGRP is found primarily in the ARC nucleus of the hypothalamus and also acts at

the MC-4R to cause an increase in feeding. An icy injection of AGRP increases feeding

(Rossi et al., 1998) and over expression of AGRP leads to morbid obesity (Ollman et al.,

1997). Pharmacological blockade of the MC-4R reduces the anorectic effects of leptin,

suggesting that the melanocortin system is an important downstream target of leptin

action (Seeley et al., 1997). This concept is supported by the recent finding that in

humans obesity is strongly linked to a region of chromosome 2 near the POMC gene

locus (Comuzzie et al., 1997). Defect in MC-4R is the most common cause of

monogenic obesity seen in humans and is responsible for 3-4% of the cases of severe

early onset obesity (Farooqi et al., 2000).












Leptin

The presence of a weight reducing hormone had been suspected for many decades

based on the study of mouse models of genetic obesity. Many of these models are

characterized by single gene mutations as seen in the obese ob/ob mice, the diabetic

db/db mice, Zucker fatty fa/fa rats and tubby mice (reviewed by Spiegelman and Flier,

1996; Levine and Billington, 1998). The obese ob/ob mice are characterized by severe

obesity, hyperinsulinemeia and hyperglycemia that resemble diabetes mellitus, as are the

diabetic, obese db/db mice. Parabiosis experiments were conducted in the 1970's by

Coleman in which two strains of mice were surgically joined together in a manner such

that they shared their blood circulation. When an ob/ob mouse shared circulation with a

wild type mouse it became lean leading Colman to argue for the possibility of a blood

borne factor that was missing in the obese mouse. In similar experiments when the db/db

mouse shared circulation with a wild type mouse, it failed to become lean. The wild type

mouse in this experiment however, was aphagic, became hypoglycemic, lost weight and

died of starvation. Coleman also conducted a parabiosis experiment with ob/ob and

db/db mice. In this case the ob/ob mouse became lean without any effects seen in the

db/db mouse. Based on these experiments Coleman hypothesized that the db/db mice

produced a weight reducing factor but were unresponsive to it, possibly because they

lacked the receptor for it, while the ob/ob mice lacked the factor itself.

This factor was finally identified in 1994 and named leptin (Greek, leptos = thin).

Leptin was discovered by positional cloning, an approach not used before in the

discovery of a major hormone (Zhang et al., 1994). Leptin is the product of the obese









gene and is also referred to as the obese protein. Leptin is a 16 kD secreted protein

hormone produced primarily in white adipose tissue (Auwerx and Staels, 1998,

Campfield et al., 1996; Campfield et al.,1995; Caro et al., 1996). Other minor sites of

leptin production include the gastric mucosa (Bado et al, 1998), placenta (Senaris et al,

1997), kidney, mammary epithelium (Smith-Kirwin et al, 1998), skeletal muscle, brown

adipose tissue (Wang et al., 1998) and most recently leptin mRNA and protein have

recently been localized in the brain (Morash et al., 1999). The amount contributed from

each of these sites to the plasma leptin pool and their physiological significance awaits

further studies. One report suggests that the brain is a significant, non-adipose source of

leptin contributing upto 40 % of the circulating leptin in women (Wiesner et al., 1999).

These authors also suggest that there is a gender bias towards the release of leptin from

the brain, with female brains releasing more leptin than males.

Leptin has a circulating half life of approximately 30 minutes, is released in a

pulsatile manner from adipose tissue and demonstrates a circadian rhythm with a night-

time elevation in the circulating concentration (Caro et al., 1996; Lewis-Higgins et al.,

1996). In rodents as well as in humans, leptin circulates in the blood bound to several

plasma proteins. In obese subjects, there is a decrease in the amount of bound leptin and

an increase in free leptin in the circulation (Houseknecht et al., 1996). As with other

hormones, the potential functions of these binding proteins could be to change the rate of

clearance of leptin, increase or decrease its biological activity, as well as to affect the

transport of leptin.

Injection of recombinant leptin into obese ob/ob mice (Halaas et al., 1995), which

have now been identified to have a nonsense mutation at codon 105 of the obese gene









(Zhang et al., 1994), leads to BW loss due to decreased food intake and increased energy

expenditure (Pelleymounter et al., 1995). Leptin also decreases BW in normal mice and

in mice with diet induced obesity when administered at supra-physiological doses

(Campfield et al., 1996; Caro et al., 1996; Flier, 1997). Such observations have led

researchers to dub it the peripheral satiety factor.

Circulating leptin levels are positively correlated with body fat mass and are

elevated in several models of rodent and human obesity (Considine et al., 1995;

Friederich et al., 1995; Maffei et al., 1995). Steady state levels of leptin are elevated in a

variety of rodent models of obesity (Maffei et al., 1995; Mizuno et al., 1996; Friederich

et al., 1995). Fasting causes a dramatic down-regulation while increased caloric intake

results in up-regulation of leptin mRNA in adipocytes (Ahima et al., 1996; Friedman

1996; Mizuno et al., 1996; Saladin et al., 1995). These observations have led to the

proposal that leptin serves as an adipostat informing the brain of the status of energy

storage in the adipose tissue so that appropriate changes in appetite, metabolism and

nutrient partitioning can be signaled. Leptin crosses the blood brain barrier (BBB)

through a non-linear saturable transport mechanism, in a unidirectional manner (Banks et

al., 1996). It is transported intact across the BBB as well as through "leaks" at

circumventricular organs where the BBB is not as tight. Uptake of leptin is reported in

the choroid plexus, ARC region of the hypothalamus and in the median eminence (Banks

et al., 1996). It is not known how leptin gets access to specific areas in the brain that are

not directly peri-ventricular in location. Interestingly, diet induced obesity in rodents is

associated with a decrease in the amounts of leptin transported across the BBB (Burguera

et al., 2000).









Leptin reduces BW in ob/ob mice; however, the effects of leptin in normalizing

the metabolism of obese ob/ob mice do not stem solely from a decrease in caloric intake.

This is illustrated by the report that ob/ob mice that are pair fed to wild type lean

littermates lose less weight than do ob/ob mice administered leptin (Levin et al., 1996).

Leptin selectively reduces adipose tissue mass unlike that observed in times of starvation

when there is also a loss in lean body mass (Pelleymounter et al., 1995). Further, the

elevation in free fatty acids and ketones observed in times of energy crunch is not seen

following leptin treatment of wild type rats (Shimabukuro et al., 1997). An important

difference in leptin administration vs. decreased energy input mediated decrease in BW is

that leptin does not decrease energy expenditure as is observed with a decrease in food

intake. On the contrary, there is an increase in energy expenditure with leptin

administration has been reported (Halaas et al., 1997). Leptin thus exerts its larger role in

energy metabolism through increased energy expenditure as well as reduced caloric

intake.

Although the effects of leptin given intracerebroventricularly (icv) or peripherally

are similar (Halaas et al., 1997), icv leptin is more potent in reducing food intake and

enhancing metabolism, thus suggesting that leptin modulates its effects on energy balance

by action at the level of the CNS (Campfield et al., 1995; Halaas et al., 1997). How

leptin mediates its effects and through which modulators is not fully understood. The

primary site of leptin action is thought to be the hypothalamus and involves hypothalamic

nuclei associated with the regulation of feeding such as the ARC, VMN, PVN, etc. The

neural connections involved in the action of leptin on energy metabolism were elucidated

by analyzingfos like immunoreactivity as an indicator of neuronal activation. These









studies indicate that leptin stimulatesfos like immunoreactivity in the dorsomedial VMH,

the dorsomedial nucleus (DMN) of the hypothalamus, the ventral and parvicellular

subdivisions of the PVN, the premammillary nucleus as well as the superior lateral

parabrachial nucleus of the hypothalamus (Yokasuka et al., 1998; Elmquist et al., 1998,

1999). Tract tracing studies have revealed major anatomical links between the DMN and

PVN (Swanson and Swachenko, 1983). The PVN, an important nucleus for regulation of

appetitive behavior has descending axons to autonomic preganglionic neurons within the

spinal cord and the medulla. Thus, leptin activation of nuclear groups in the PVN and

ARC may regulate neuroendocrine function and energy balance possibly through the

sympathetic nervous system.


Leptin Receptors

Leptin affects BW and reproduction via binding to receptors in the CNS. The

leptin receptor was first isolated from mouse choroid plexus using an expression cloning

strategy (Tartaglia et al., 1995). The 894 a leptin receptor belongs to the class 1 cytokine

receptor family and is a single transmembrane spanning receptor (Tartaglia et al., 1995).

It is closely related to the gpl30 transduction unit of the cytokine receptors such as those

for interleukin-6 leukemia inhibitory factor, granulocyte colony stimulating factor and

cilliary neurotrophic factor. It has a short intracellular domain that contains a Janus

kinase (JAK) binding site; the extracellular domain binds its ligand leptin (Tartaglia et al

1995). Six different isoforms (ob-R a-f) of the leptin receptor exist (Tartaglia, 1997,

White et al., 1997, Reviewed in Ahima et al., 2000). These leptin receptor isoforms share

a common extracellular ligand binding domain at the amino terminus, but differ at the

intracellular carboxy-terminal domain. Only the long form, ob-Rb, contains all the









intracellular motifs needed for activation of the JAK/STAT signaling pathway. The

biological actions of leptin following binding to ob-Rb are mediated through activation of

the JAK/STAT signaling pathway (Tartaglia 1997; White et al., 1997). Ob-Ra, Ob-Rc,

Ob-Rd and ob-Rf can activate JAK but this activation is weak and does not led to the

activation of STAT, these isoforms thus are incapable of signaling (Ghilardi et al., 1996;

Bjorbaek et al., 1997) Ob-Re lacks both the transmembrane and the intracellular domain

and circulates as a soluble leptin receptor, the function of this receptor is not clear

(Halaas and Friedman, 1997). The short form of the receptor ob-Ra, has a truncated

intracellular domain and is not capable of signaling (Tartaglia, 1997). It is highly

localized in the choroid plexus and is speculated to play a role in the transport of leptin

across the BBB (Tartaglia, 1997).

There are genetic rodent models of obesity attributed to leptin receptor defects.

In db/db mice, there is a premature stop codon in the 3' end of the ob-Rb transcript,

resulting in the synthesis of a receptor with a truncated intracellular domain resembling

ob-Ra (Coleman, 1978; Chen et al., 1996; Tartaglia, 1997; Chua et al., 1996). The

db/db mice are completely insensitive to the effects of leptin and are morbidly obese. In

the Zuckerfa/fa rats there is a Gin-to-pro substitution at amino acid 269 in the

extracellular domain of the leptin receptor leading to a decreased cell surface expression

of the receptor and hence decreased leptin signaling (Phillips et al., 1996, da Silva et al.,

1998). The Zuckerfa/fa rats are consequently obese, but unlike the db/db mice, thefa/fa

rats are capable of responding to very high doses of leptin administered icy (Cusin et al.,

1996). Another model of rodent obesity associated with defective leptin receptors is the

obese Koletsky rat. A point mutation at amino acid 763 leads to a stop codon in the









extracellular domain of the receptor so that these rats do not express any leptin receptor

(Wu-peng et al., 1997; Takaya et al., 1996). In humans a few rare cases of obesity

associated with mutations in the leptin receptor have been reported (Clement, et al.,

1998). These patients are hyperphagic, obese and insensitive to exogenous leptin

administration as in the case of the db/db mice.

The long form of the leptin receptor has been localized in heart, lung, kidneys,

ovaries, uterus, testes, pancreas, adipose tissue and brain (Couce et al., 1997; Mercer et

al., 1996; Tartaglia, 1997; White et al., 1997; Zomorano et al., 1997) with the highest

levels of ob-Rb in the hypothalamus of the brain (Elmquist et al 1998). More specifically

ob-Rb has been found in the VMH, DMH, ARC, PVN, and LHA in the hypothalamus

(Elmquist et al., 1998). Leptin receptors are colocalized in NPY and POMC producing

neurons in the ARC, suggesting that the CNS actions of leptin include modulation of

orexigenic and anorexigenic molecules (Cheung et al., 1997; Kalra SP et al., 1998;

Leibowitz and Hoebel, 1998).


Leptin interaction with appetite regulating molecules



Leptin and NPY

NPY producing neurons are found in the ARC nucleus of the hypothalamus.

NPY mRNA levels are significantly elevated in ob/ob mice as well as in db/db mice

(Sanacora et al., 1990; Wilding et al., 1993). Administration of leptin to ob/ob mice

returns NPY mRNA levels to wild type levels and corrects the obese phenotype

(Stephens et al., 1995). It has also been shown that icv leptin decreases NPY mRNA

expression and release in the hypothalamus (Schwartz et al., 1996). However, NPY









knockout mice, which are normal in most aspects, are sensitive to the actions of leptin

(Baraban et al., 1997; Ericson et al., 1996; Ericson et al., 1996). In the absence of NPY

(NPY -/-) ob/ob mice were still responsive to exogenously administered leptin, although

the obese phenotype was not corrected. Thus, even though leptin and NPY have

reciprocal effects on feeding and leptin reduces NPY gene expression, it seems unlikely

that leptin induced decrease in feeding and increase in energy expenditure occurs solely

through the NPY system.

Leptin and the Melanocortins

Several lines of evidence suggest a connection between leptin and the

melanocorticergic pathways. The long form of the leptin receptor Ob-Rb and POMC

mRNA are co-expressed neurons in the ARC of the hypothalamus (Cheung et al., 1997).

POMC mRNA is reduced in leptin deficient animals, and leptin administration causes

increased expression of POMC mRNA (Mizuno et al., 1997; Thornton et al., 1997;

Schwartz et al., 1997). Leptin delivered icy increases POMC mRNA expression as does

fasting (Schwartz et al., 1997). uMSH, a product of the POMC gene, decreases feeding

when injected icy (Poggioli et al., 1986) by binding to the MC-4R (Lu et al., 1994).

THUS, aMSH is an important player in mediating leptin's effects on inhibiting appetite.

Obesity in another rodent model, the lethal yellow mouse (A y /a) mouse, is

caused by constitutive expression of the agouti peptide (Miller et al., 1993). Agouti is a

potent antagonist of the hypothalamic MC-4R and thus increases feeding in mice

(Ollmann et al., 1997). AGRP is homologous to agouti and antagonizes both MC-3R and

MC-4R (Fong et al., 1997). Mice lacking leptin have elevated hypothalamic AGRP

mRNA. Reinstatement of leptin in these mice decreases AGRP mRNA expression









(Mizuno et al., 1999). Further, the levels of AGRP are elevated 8-10 fold in leptin

receptor mutant db/db mice. (Ollmann et al., 1997). Administration of a synthetic brain

melanocortin receptor antagonist blocks the effect of leptin on feeding (Seeley et al.,

1997). In the fed state leptin suppresses AGRP expression, but during fasting this

restraint is lowered resulting in upregulation of AGRP and increased drive towards

ingestive behavior (Wilson et al., 1999). Because POMC derived melanocortins are MC-

4R agonists and AGRP is an antagonist, the net result of leptin action is decreased

signaling through the MC-4R. Thus, leptin may possibly mediate its weight reducing

actions by altering/modulating melanocortin neurochemistry within the hypothalamus.

Leptin and Insulin

The relationship between leptin and insulin levels in the circulation is very

complex. In the absence of leptin as seen in the leptin deficient ob/ob mice there is

hyperinsulinemia as well as hyperglycemia. Leptin deficiency causes severe insulin

resistance in these mice which is reversed upon administration of leptin. Exogenous

administration of leptin to ob/ob mice leads to a decrease in the plasma levels of insulin

as well as glucose to the normal range (Halaas et al., 1995; Campfield et al., 1995;

Weigle et al., 1995; Schwartz et al., 1996). Interestingly, these changes precede any

changes in food intake or BW. Leptin administration also reverses the severe insulin

resistance and hyperglycemia in lipodystrophic mice that lack white adipose tissue and

hence have very low levels of circulating leptin. Thus, leptin has effects on insulin

secretion and action independent of its actions on food intake. It is not known whether

leptin exerts its effects on insulin secretion directly in the periphery via its actions on the

pancreatic p-cells or through targets in the central nervous system. In the literature, data









on the exact influence of leptin on insulin secretion are controversial, although there is

some evidence of a negative effect of leptin on insulin secretion (reviewed in Casaneuva

and Dieguez, 1999).

Insulin, on the other hand, has a positive effect on the synthesis and secretion of

leptin. Blood leptin levels are elevated after peak insulin secretion during the course of

the feeding cycle ( Saladin et al., 1995; Sinha et al., 1996 ). Insulin directly stimulates

leptin mRNA synthesis in white adipocytes in vitro (Rentsch et al.,1996). Intravenous

injection of insulin in rodents leads to an increase in plasma leptin levels accompanied by

increases in adipose tissue leptin mRNA expression (Saladin et al., 1995 ). Slowing of

leptin production is rapidly reversed by insulin administration (McDougald et al., 1995).

Some reports suggest that in humans the levels of insulin may predict the levels of leptin.

In the plasma leptin levels increase with insulin administration, and conversely, the

lowered levels of insulin in the fasting state, coincide with lowered leptin in the plasma

(Boden et al., 1997; Segal et al., 1996; Boden et al., 1996). There is a positive

relationship between hyperleptinemia and insulin resistance (Cusin et al., 1995).

How insulin mediates its effects in energy regulation is not completely

understood. Like leptin, insulin acts on the hypothalamus to decrease food intake

(Woods SC et al., 1996). Ahima and Flier (2000) suggested that the effects of leptin on

nutrition are mediated in part by insulin. There is a prevalent view that leptin facilitates

the actions of insulin in the CNS. Leptin possibly plays an important role in enhancing

insulin sensitivity in the periphery. There is further debate as to whether it is the central

action of leptin that enhances insulin sensitivity or whether this is accomplished by the

direct action of leptin on target organs outside of the CNS.









Fat tissue and Uncoupling Proteins in Energy Metabolism

There are two different types of adipose tissue, white adipose tissue (WAT) and

brown adipose tissue (BAT). White adipose is the major source of stored energy in the

form of triglycerides and is found throughout the animal kingdom while BAT is restricted

to mammals. WAT is broken down and its triglyceride stores used in times of energy

deprivation. The mass of WAT in the periphery reflects the balance between energy

expenditure and intake (Klaus, 1996). It plays a key role in signaling energy availability

to the hypothalamus through production of leptin. WAT and BAT are localized in

anatomically distinct sites, with BAT found mainly in the interscapular, subscapular,

axillary and the suprastemal regions (Klaus 1996). BAT is well established as a

thermoeffector organ (Reviewed in Cannon et al., 1997). It is a source of non-shivering

thermogenesis and plays an important part in the regulation of body temperature in small

animals such as rodents. In humans, however, BAT is found only through early infancy.

BAT is extensively innervated by the sympathetic nervous system (SNS,

reviewed in Himms-Hagen, 1991; Bartness et al., 1999). In response to inputs such as

cold exposure, overfeeding and acute feeding the hypothalamus transmits signals to the

BAT via the SNS. The SNS acts on the BAT through 33 adrenergic receptors to elicit

heat production. Bartness et al., (1999) using pseudorabies virus, a transneural viral tract

tracer, showed a direct connection between magnocellular neurons of the PVN and

interscapular BAT (iBAT), as well as between some brainstem regions such as the caudal

raphe region of the brain and iBAT. Substantial innervations to iBAT from the SCN as

well as the MPOA are also seen but the functional implications for this innervation are

not known. BAT is diffused in adipose tissue of non-human primates but lacks 33









adrenergic receptors required for sympathetic activation. Mice with genetic ablation of

BAT are obese, have reduced energy expenditure and are prone to diet-induced obesity

(Hamman et al., 1998). These studies suggest that intact BAT is necessary for protection

from diet induced obesity.

BAT cells have a unique mitochondrial machinery that allows them to uncouple

oxidative phosphorylation from ATP synthesis. The thermogenic action of brown

adipocytes is due to the presence of a mitochondrial protein known as uncoupling protein

1 (UCP-1). UCP-1 is a proton translocator present in the inner mitochondrial membrane

and functions by diverting protons from ATP synthesis towards the dissipation of heat

(Klingenberg 1990; Garlid et al., 1998 ). Five types of UCP's (UCP-1-5) have been

identified. UCP-2 as well as UCP-3 can also uncouple mitochondrial respiration (Fleury

et al., 1997;Gong et al., 1997). UCP-1 is expressed only in BAT, and is primarily

responsible for cold induced thermogenesis (Boss et al., 1998 ). UCP-2 is expressed in

several tissues such as liver, skeletal muscle, white adipose tissue and brain areas

involved in homeostasis, specifically hypothalamic nuclei such as the SON, SCN, PVN

and ARC (Fleury et al 1997; Gimeno et al., 1997). UCP-3 is expressed primarily in

skeletal muscle and heart (Vidal-Puig et al., 1997). UCP-4 is a brain specific uncoupling

protein (Mao et al 1999), of yet unknown function.

UCP-1 levels are sensitive to stimuli such as fasting, cold exposure and thyroid

hormone levels in the blood in a tissue specific manner. Fasting decreases UCP-1 mRNA

expression in BAT while cold exposure leads to an increase in UCP-1 mRNA in this

tissue ( Boss et al., 1998) as do rising thyroid hormones (Nicholls and Rial 1999). The

primary stimulus for enhancing UCP-1 is norepinephrine action via the 31 and P 3









adrenergic receptors (Boss et al 1998). Targeted disruption of the UCP-1 gene leads to

mice that are cold intolerant but not obese mice (Enerback et al., 1997).

UCP-2 expression on the other hand is unchanged in BAT upon fasting but is

increased in muscle. Cold exposure increases UCP-2 expression in muscle, BAT and

heart. There is increased expression of UCP-2 in WAT in obesity resistant strains of

mice, (Collins et al., 1997; Surwitt et al., 1998) but it remains unchanged in diet induced

obese mice. Similarly, there is no change in BAT UCP-2 in response to a high fat diet.

Therefore, there are tissue specific differences in mechanisms regulating UCP-2

expression in response to increased dietary fat. Generally, UCP-2 is modulated by diet

but is not regulated by the SNS.

UCP-2 and 3 are 56% homologous to UCP-1. UCP-3 is 73% similar in structure

to UCP-2, but its functions are more closely related to UCP-1. Sympathetic innervation is

necessary to maintain basal mRNA levels of UCP-3. As with UCP-1, UCP3 is

upregulated by cold exposure, sympathetic stimulation and thyroid hormones. Genetic

knockouts for UCP-3 result in mice that are not obese (Vidal-Puig et al., 2000), while

mice over-expressing UCP-3 are lean yet surprisingly hyperphagic (Clapham et al.,

2000).



Leptin and Uncoupling Proteins

There are reports that support the view that BAT helps regulate BW after

hyperphagia with non-shivering thermogenesis (Himms-Hagen ,1990). Thermogenesis

by BAT may be one mechanism by which leptin regulates BW. Immunohistochemical

methods have localized leptin in BAT although to a much lesser degree than in WAT

(Cinti et al., 1997). Leptin increases sympathetic activity in ob/ob mice with increases in









BAT UCP-1 mRNA (Collins et al., 1996). Peripheral delivery of leptin leads to increased

energy expenditure as measured by oxygen consumption and UCP-1 mRNA expression

in BAT (Scarpace et al., 1997). Leptin given icy along with food restriction prevents the

food restriction induced decrease in UCP-1 and UCP-3 mRNA levels without an effect

on UCP-2 expression (Scarpace et al., 2000). Denervation of BAT prevents upregulation

of UCP1 by leptin (Scarpace et al., 1999). UCP-3 m RNA increases with leptin

treatment in skeletal muscle of ob/ob mice (Liu et al., 1998). There appears to be a strain

specific effect on UCP-2 induction by leptin in mice. Leptin induces UCP-2 mRNA in wt

mice (Zhou et al 1997). However, there is no effect ofleptin on UCP-2 mRNA in diet

induced obesity resistant strains of A/ J or B6 mice (Surwitt et al., 1998). In summary, it

is clear that leptin acts to modulate the different subtypes of UCP's in different ways

(Scarpace et al., 2000). Whereas UCP-1 is modulated via changes in sympathetic activity

in the BAT, UCP-2 and 3 are regulated by leptin independently of the SNS via unknown

mechanisms.




Adenoassociated Virus and Recombinant Adenoassociated Virus

Adenoassociated virus (AAV) is a single stranded DNA virus and a member of

the parvovirus family. AAV requires the presence of a helper virus for viable

transfection (Berns and Bohenzky, 1987). The most important feature for the use of

AAV is safety. AAV is non-pathogenic, non-immunogenic and thus ideal for use in vivo.

AAV exists as a latent infection in humans and is not associated with the etiology of any

disease (Berns and Bohenzky, 1987). Most significantly, this virus is capable of infecting

non-dividing cells (Flotte et al., 1995; Klein and Peel, 2000). The site of integration of









the wild type AAV is human chromosome 19 (Samulski, 1983). rAAV vectors are

derived from AAV but unlike their wild type parent AAV are incapable of site specific

integration and thus integrate randomly into the host genome in the presence of a helper

virus (Xiao et al., 1997 ). Recombinant AAV is made up of a simple capsid with a single

stranded DNA genome that has short viral inverted terminal repeats (ITR's) but no viral

coding sequences (Hermonat et al., 1984; McLaughlin et al., 1988). This removal of all

viral sequences except the ITR's further enhances the safety of AAV for in vivo studies.

This is achieved by eliminating the generation of wt helper virus as well as reducing the

probability of an immune response to rAAV delivery (Xiao et al., 1996). One limiting

factor in the use of this virus as a vehicle for gene delivery is the limitation of the insert

size, upto 4.7 kb. There are, however, recent publications that have "expanded" the size

of the insert with the use of heterodimer vectors (reviewed in Samulski, 2000).

rAAV vectors have great potential for delivery of genes to the CNS. AAV can

transduce non-dividing neurons over extended periods of time in different areas of the

adult rat brain (Kaplitt et al., 1994). Other tissues successfully transduced by rAAV

include the spinal cord, eye, muscle, lung, heart and liver without any detectable cellular

immune responses (reviewed in Xiao et al., 1997). Of all the different types of viral

vectors used in gene therapy, rAAV makes up only 1.1 %. This was mostly because in

the past it was difficult to produce high titre AAV vectors for in vivo delivery, and due to

the comparatively slower transduction and lower number of cells transduced. At the

University of Florida Gene Therapy Center and elsewhere there are now production

paradigms in place that have resulted in higher titre virus production. Due to its excellent









safety features and ability to transduce a variety of tissues, rAAV will likely be the virus

of choice for gene therapy applications in the future.


Study Design and Rationale

Leptin is a fundamental component of the energy regulatory system. Deficiency

of leptin or mutations in its receptor leads to morbid obesity in a variety of rodent models

(Halaas and Friedman, 2000). However, most human obesity is not genetic in origin and

does not occur due to mutations such as those described for the ob/ob or db/db mice

(Campfield et al., 1996). Human obesity is accompanied by hyperleptinemia and often

hyperinsulinemia. Leptin levels increase with increasing fat mass resulting in a positive

correlation between adiposity and serum leptin levels in humans (Caro et al., 1995). In

human obesity leptin gene expression and increased leptin secretion are increased

(Considine et al., 1996). Obese humans resemble db/db mice in that they have an

impaired response to the high leptin levels in their system. The principal site of action of

leptin is the brain, but in obese patients the brain does not respond to the peripherally

elevated leptin levels. This unresponsiveness to endogenous elevated leptin is termed

"leptin resistance".

The etiology of leptin resistance is little understood. There are several possible

causes of leptin resistance. The first hypothesis is that there is a defect in the transport of

leptin across the BBB (Van Heek et al., 1997), possibly due to a saturated transport

system across the BBB attributable perhaps to excess leptin availability as seen in

hyperleptinemia. This view is supported by experiments in obese rodents that were

insensitive to peripherally administered leptin but responded by losing weight when

leptin was injected icy (Van Heek et al 1997). Also, the CSF:serum ratio ofleptin is









greatly reduced in obese humans (Caro et al., 1996). Since the CSF leptin concentration

of obese humans is 30% higher than normal controls, it is also likely that a part of leptin

resistance seen in obesity could be due to "reduced leptin sensitivity within the CNS.

Defects in the leptin receptors implicated in facilitating the transport of leptin

across the BBB may be another means of disruption of leptin transport. However, this

possibility is refuted by the presence of leptin in the CSF of Koletsky rats which due to a

mutation in their leptin receptor gene lack any form of the leptin receptor (Wu-peng et

al., 1997) yet can transport leptin into the CSF.

Aging is another a factor associated with leptin resistance. Aged rats are leptin

resistant and have increased adiposity with an increase in serum leptin levels (Hua Li et

al., 1998). Fasting-induced regulation of serum leptin as well as leptin mRNA expression

are impaired in aged rats (Hua Li et al 1998).

Deciphering the nature of leptin resistance is of paramount importance to the

understanding of human obesity. Currently there is no direct evidence to support any of

the above mentioned possible causes of leptin resistance. In common forms of obesity

there are neither structural nor functional defects in leptin receptors in the hypothalamus,

thus in the absence of defects in either the availability of leptin in the CSF, its receptor

protein or the ability of its receptors to function correctly, we propose that leptin

resistance stems from insufficient availability of leptin at the site of its target neurons in

the hypothalamus. Leptin resistance is likely due to loss of leptin availability at specific

sites within the brain especially the appetite regulating nuclei within the hypothalamus.

The effects of elevated, site directed, levels of leptin in the hypothalamus is not known.









There are a few studies that have used viral vectors to deliver the leptin gene into

rodent systems. In 1996, Chen et al generated sustained hyperleptinemia for a period of

28 days in male Wistar rats.via carotid artery infusion of an adenovirus vector encoding

leptin. The animals lost BW and displayed reduced appetite in response to the 8ng/ml

plasma leptin achieved. In the same year Muzzin et al., injected an adenovirus encoding

leptin through an injection into the tail vein in ob/ob mice and corrected their obesity and

other associated symptoms such as hyperinsulinemia and diabetes. Similar results were

seen in another study (the only study reported with AAV) where leptin was delivered via

rAAV encoding leptin into the skeletal muscle. BW and food intake were reduced in

ob/ob mice for a period of 15 weeks post injection. There are no reports in the literature

that explore the long-term outcome of central delivery of leptin using gene therapy

methods.

The recent clinical trial (Heymsfield et al., 1999) to study the potential therapeutic

role of leptin in reducing obesity yielded maximum reductions in BW (7.1 kg in treated

humans vs. 1.7 kg in the placebo group) with recombinant leptin administered

subcutaneously at a dose of 0.3 mg/kg BW. Direct delivery of leptin into the CSF is

more potent than peripherally administered leptin. The main concern, however, has been

safety in the delivery of leptin into the CSF. With the rAAV encoding leptin the safety

concerns can be addressed. Since it is our hypothesis that leptin resistance stems from a

loss of leptin availability within the central nervous system, in the studies presented in

this dissertation I sought to understand the long term consequences and viability of

central leptin delivery using a rAAV vector encoding leptin in lean Sprague-Dawley rats.

The objective of these studies was to deliver leptin into the third ventricle of the brain in









order to achieve targeted increase at specific hypothalamic sites involved in the

regulation of energy balance and BW regulation. We propose that leptin resistance is not

due to chronic over-exposure at its site of action but conversely due to an insufficiency at

critical hypothalamic nuclei.

As summarized earlier, several neural and peripheral systems are implicated in

mediating the action of leptin. In order to elucidate which of these known modulating

systems is implicated, I examined the hypothalamic neuropeptidergic and blood

hormonal changes in response to central delivery of rAAV encoding leptin. This research

will likely provide information useful for developing targeted therapies for some forms of

human obesity by determining the efficacy and functioning of site specific increases in

leptin concentrations in the brain over the long term.














CHAPTER 2
GENERAL METHODS

Experimental Animals

Adult male and female Sprague-Dawley rats (200-250 g) were purchased from

Harlan (Indianapolis, IN). All rats were housed individually in an air-conditioned room

(22-250C) with lights on from 0500-1900 hrs. Food and water were available ad-libitum

to all animals. Rats were sacrificed at the end of the experiment by decapitation. Leptin

mutant ob/ob mice were purchased from Jackson Laboratories. Mice were housed four

per cage in a specific pathogen free environment. Mice were sacrificed by cervical

dislocation. The animal protocols were approved by the Institutional Animal Care and

Use Committee.

Third Ventricle Cannulation and Injection

Permanent cannulae were stereotaxically implanted in the third cerebroventricle

of male and female Sprague-Dawley rats, under anesthesia (Ketamine 100mg/kg BW +

Xylazine 15 mg/kg BW) according to the rat stereotaxic atlas (Palkovits and Brownstein,

1988). The nose bar was set 5 mm above the horizontal zero. The cannula was placed at

the midline, 6.4 mm anterior to the interaural line and 8 mm deep and cemented in place

with dental cement. The cannulae were observed for cerebrospinal fluid (CSF) flow and

then closed with a stylet. This flow of CSF served as an indicator of accuracy of the

placement of the cannulae. Rats were allowed to recover from surgery for 7-10 days

before the experiment was initiated. The stylets were removed from the cannulae 30 mins

before the time of injection. Injectors were constructed to fit the implanted cannulae. PE-









50 tubing was attached to the injector on one end and a Hamilton syringe on the other.

Vectors were injected in a 5 [Il volume to unanesthetized rats over a period of 30 seconds

using the Hamilton syringe. The injector was retained in the cannula for an additional 30

seconds in order to prevent backflow.



Blood Collection

Blood samples from rats were collected from the jugular vein at different time

points during the course of the long-term studies. Rats were anesthetized with Ketamine

(100 mg/kg BW) and Xylazine (15 mg/kg BW). The jugular vein was exposed by an

incision in the skin at the level of the jugular vein. A 20 gauge needle coated with

heparin was inserted through muscle into the vein and 1.0 ml blood was slowly

withdrawn into a heparinized syringe. The incision wound was closed with metal clips.

The blood was centrifuged at 3000 rpm for 3 mins, plasma was removed and stored at

-200C until analysis of blood hormone levels. Blood samples from mice were collected by

retro-orbital puncture and processed as for rats.



Cerebro-spinal Fluid Collection

Cerebro-spinal fluid (CSF) was collected as previously described (Stein et al.,

1983) using a stereotaxic apparatus. Rats were maximally ventroflexed, a 22 gauge

needle attached to PE 50 tubing mounted on the stereotaxic apparatus was then inserted

horizontally into the cisterna magna at 6.4-6.6 mm below the occipital crest. Flow was

initiated with gentle suction using a Hamilton syringe, following which CSF was

collected by gravity drainage. CSF was centrifuged to remove blood and debris; only









clear CSF was used for leptin analysis. CSF was collected on ice and stored at -20 C

until use.



Food Intake Measurement

Twenty four hour food consumption was monitored on a weekly basis for the

duration of each experiment. Pre-weighed food pellets were placed in specially designed

feeders that were placed inside the rat cages. Food was weighed at the end of the 24

hour period along with any spillage collected at the bottom of the feeders. Food intake

was calculated to the nearest 0.1 g as the difference between initial and final food weight

over a 24 hour period.



Urine Collection

Two 24 hour urine samples were collected on consecutive days after placing

animals in metabolic cages. The animals were placed in the metabolic cages for three

days prior to collection of samples. The average volume collected over a 24 hour period

was 15-20 ml. 10 ml of 0.1N HC1 was added to the urine sample post collection. The

samples were stored at -20 OC until assayed for norepinephrine.



Organ Collection

Rats were sacrificed by decapitation. Trunk blood was collected in 15 ml

polypropylene collection tubes. The brains were rapidly removed from the skull and snap

frozen in powdered dry ice. In experiments where hypothalami were used for analysis,

the brains were removed and the hypothalami were carefully dissected out and snap

frozen in 2 ml sterile RNase free tubes. White adipose tissue was removed from the









epididymal area and snap frozen directly in 2 ml RNase free tubes. Interscapular BAT

was dissected with surgical scissors. The BAT was deposited into petri dishes containing

cold normal saline. While in cold saline the WAT was trimmed off as well as any muscle

tissue. This was done as quickly as possible in order to maintain the integrity of the RNA

in BAT. Once cleaned the BAT was snap frozen and stored in 2 ml tubes. All collected

tissues were stored at -80 OC until used for analyses.



Carcass Fat and Protein Estimation

Carcass water, fat and fat-free dry mass were determined gravimetrically (Fong,

1989). Carcasses were weighed immediately after killing the rats, then frozen in liquid

nitrogen and pulverized with solid carbon dioxide in a commercial blender. Pulverized

carcasses were dried for 2-4 days to a constant mass at 800C. Lipid content was

determined by sequential chloroform-methanol (1:1), ethanol-acetone (1:1), and

petroleum ether extractions. Carcass protein content was measured from dried carcass

aliquots after NaOH extraction with a routine Bradford protein assay.



Oxygen Consumption

Oxygen consumption was assessed as previously reported (Scarpace et al., 1992).

Briefly, oxygen consumption was measured in three rats simultaneously with an Oxyscan

analyzer (OXS-4; Omnitech Electronics, Columbus, OH). Flow rates were 21/min with

a 30 second sampling time at 5 min intervals. The temperature was maintained in the

thermoneutral zone at 260 C. Results were normalized to BW and expressed as ml.min1.

Kg0.67
Kg7









RNA Isolation

Total RNA was isolated from tissue using an RNA isolation kit (STAT-60,

Teltest Inc, Friendswood, TX). This method is a modification of the Chomscynski and

Sacchi RNA isolation procedure. Briefly, 30-100 mg tissue was homogenized using a

Polytron homogenizer in STAT-60 solution. The samples were extracted with

chloroform and precipitated in isopropanol. The precipitate was washed in ethanol, and

reconstituted in DEPC treated water. The samples were read in a spectrophotometer at an

absorbency of 260 as well as 280 nm. Ratios of the two reading equal to 1.8 and not

greater than 2.0 was an indicator of good RNA yield. Integrity of the RNA samples was

verified by running aliquots of the sample on 1% agarose gels. All RNA samples were

stored at -80 OC until use.



Leptin mRNA Expression using RT-PCR

Leptin mRNA expression was analyzed using reverse transcriptase-PCR (RT-

PCR). Briefly, total RNA was extracted from hypothalami using the RNA STAT 60

RNA isolation kit (Tel test Inc, Friendswood, TX). First-strand cDNA was synthesized

using 1 ug total RNA with a RNA PCR kit. All reagents were purchased from PE

Biosystems, Foster City, CA. Primers were designed to the rat leptin gene to encompass

a 308 bp region of the coding sequence. (Gen bank Accession code D49653), Sense: 3'

CCC ATT CTG AGT TTG TCC, Antisense: 3' GCA TTC AGG GCT AAG GTC.

Primers were designed for cyclophilin (internal control) to generate a 470 bp product

(Gene bank accession code M19533). Sense: 3' GAC AAA GTT CCA AAG ACA GCA

GAA A, Antisense: 3' CTG AGC TAC AGA AGG AAT GGT TTG A. The PCR

products generated by these primers were sequenced and independently verified and









found to match rat leptin and rat cyclophilin completely. Linearity of the PCR was tested

by amplification for 20-45 cycles for leptin and cyclophilin. The linear range was found

to be between 25 and 40 cycles.

Five microliters of the first-strand cDNA was amplified for 30 cycles for leptin

and 26 cycles for cyclophilin. Each gene was amplified in a separate PCR reaction from

a single RT reaction by using the following parameters:

Leptin: Denaturation @ 95 OC, 1 min, annealing @ 56 OC, 1 min, extension @ 72

OC, 1 min, 30 cycles, 10 min final extension 72 OC.

Cyclophilin: Denaturation @ 94 OC for 50 s, annealing @ 55 OC for 45 sec,

extension @ 72 OC for 2 min, 26 cycles.

PCR products were analyzed using agarose gel electrophoresis. Twenty

microlitres of the PCR products were separated on a 2% agarose gel stained with

ethidium bromide and placed on an UV illuminator equipped with a camera connected to

a gel documentation system (BIORAD). The gel image was analyzed using an image

analysis program (Image Quant system BIORAD laboratories Inc). The relative

expression of the mRNA levels were derived from a comparison of the intensity of the

target and simultaneously run internal controls (cyclophilin). All PCR products were run

on a single gel in order to control for inter gel variation.



Dot blot analysis for UCP-1 and UCP-3

Total cellular RNA was extracted as described above. The integrity of the isolated

RNA was verified using 1 % agarose gels stained with ethidium bromide. The RNA was

quantified by spectrophotometric absorption at 260 nm as well as 280 nm using multiple

dilutions of each sample.









The full-length cDNA clone for uncoupling protein-1 (UCP1) was kindly

provided to Dr Phillip Scarpace by Dr. Leslie Kozak, Jackson Laboratory, Bar Harbor,

ME and verified by Northern analysis, as previously described (Scarpace et al., 1997).

Full length UCP- 3 cDNA .was kindly supplied by Dr Olivier Boss and used as

previously described (Boss et al 1997). All probes were random prime labeled using

Prime-A-Gene kit (Promega, Cat # U 1000) according to manufacturer's instructions.

The labeled probes were purified by filtering through a Nick Column (Pharmacia).

For dot-blot analysis, multiple concentrations of RNA were immobilized

on nylon membranes (Gene Screen Plus, Dupont, NEN) using a dot-blot apparatus (Bio-

Rad, Richmond, CA). The membranes were pre-wet in 20X SSC for 10 mins before the

diluted samples were applied. After applying the samples, the membranes were baked at

800C for 2 hours. The baked membranes were warmed in 400C water for 2 mins, and then

pre-hybridized for 30-60 mins at 650C while rotating in Hybaid Quikhyb solution. The

labeled probe was added in a concentration of 1.5 X 10 6 cpm/ml of the hybridization

solution. The membranes in hybridization solution were hybridized for 2 hours at 65 OC.

After hybridization, the membranes were washed in 2X SSC/0.1% SDS at 50 OC for 15

mins with two changes of solution. The membranes were further washed in 0. 1X

SSC/0.1% SDS for 15 mins. The blots were removed from the hybridization bottles,

wrapped in saran wrap and exposed to a phosphor imaging screen for 24-48 h. Care was

taken to minimize folds in the saran wrap. The latent image on the phosphor imager

screen was scanned using a Phosphor Imager (Molecular Dynamic, Sunnyvale, CA) and

analyzed by Image Quant Software (Molecular Dynamics). Intensities were calculated

per tg total RNA for each animal. Control as well as treated animal samples were applied









on the same blot to minimize variability. All samples from one experiment were run on

the same blot.



In Situ Hybridization (ISH)

Construction of Riboprobes

The POMC probe was constructed by cloning a 478 bp cDNA fragment (5' psn

220, 3' psn 697, GenBank Accession No J00759) into pGEM-T vector (Promega Corp.,

Madison, WI). The NPY probe was constructed using a plasmid containing a 511 bp rat

NPY fragment kindly provided by Dr S.L. Sabol (NIH, Bethesda, MD). The 396 bp

complete AGRP cDNA fragment (GenBank U89484) used to construct the probe, a

generous gift of Dr Roger Cone (Oregon Health Science University, Portland, OR), was

inserted into pBSK+/- vector.

Antisense riboprobes were transcribed in the presence of 35S-UTP (Amersham

Life Sciences Inc., Arlington Heights, IL) using T7 (AGRP, POMC) or T3 (NPY) RNA

polymerase.




Table 2-1. Riboprobes for in situ hybridization

Target Probe Sequence

AGRP 396 bp complete mouse AGRP cDNA

NPY 511 bp spanning 1-511 of rat preproNPY mRNA

POMC 478 bp spanning psnl50-608 of rat POMC mRNA











Tissue Sectioning

The brain tissue used for ISH was snap frozen in dry ice and stored at -800C until

used. The brains were coronally sectioned at -200C in a cryostat into 16 |jm sections.

The sections covered areas of the brain starting at the level of the anterior commissure

and ending past the ventral pre-mammillary nucleus of the hypothalamus, thus effectively

covering the entire hypothalamus. The sections were thaw-mounted on microscope slides

coated with Poly-L-Lysine. The sections were collected in four series; slides were stored

at -800C until use.

Tissue Processing

On the day of the experiment, one series of sections was removed from the freezer

and allowed to warm up to room temperature (RT). They were post-fixed in 3 %

paraformaldehyde for 10 min at RT followed by two washes in phosphate buffered saline

(PBS, pH 7.4) for 5 min each. Next, the slides were incubated in 0.1M triethanolamine

(pH 8.0) for 10 min at RT, dehydrated through 70%, 80%, 95% and 100 % alcohol for 2

min each and air-dried. 100 pl hybridization buffer containing 35S labeled antisense

probe containing 1 X 10 6 cpm was applied to each slide. The sections were covered with

parafilm and incubated at 500C overnight in a humidified oven.

On day two of the procedure, the parafilm was carefully removed from the

sections. The slides were washed in 2 X SSC for 5 min at RT with two changes of

solution, 0.2 X SSC for 30 min at 550C, 30 min in 0.1X SSC at 600C and 2X SSC at 370C

for 5 min. Next the sections were incubated in 20 [tg/ml RNase solution in 2 X SSC for

30 min at 370C while shaking. After another wash in 0.1 X SSC at 600C the slides were









air dried and exposed to Kodak Biomax MR autoradiography film for 3-5 days. Slides

were dipped in Kodak NTB2 emulsion, dried and stored at 40C for 2-4 weeks. The slides

were counterstained with 0.1 % cresyl violet after developing. Slides from control and

treated animals were processed together to eliminate any variability that might be

introduced by processing.

Analysis of ISH data

For semi-quantitative analysis, the relative optical density (ROD) calculated as

total target area multiplied by the integrated optical density for AGRP, NPY, and POMC

were estimated from autoradiograms with the MCID image analysis system (Imaging

Research, St. Catherine, Ontario, Canada). Twelve sections from each brain were

matched anatomically and analyzed. The background optical density in an area adjacent

to each target was subtracted from the target. The ROD of 12 sections in the same brain

were added and expressed relative to the average ROD from the control group.



Immunohistochemistry for Green Flourescent Protein (GFP)

Immunohistochemistry was performed as described earlier (Peel et al., 1997,

Klein et al., 1998). Briefly, animals were anesthetized with 100mg/kg sodium

pentobarbital ip and perfused with 300 ml 0.9% saline followed by 300-600 ml cold 4 %

paraformaldehyde in 1 X PBS. Brains were removed and kept in 4%

paraformaldehyde/1XPBS overnight, then in 20 % sucrose solution, followed by 30 %

sucrose until the tissue submerged. The brains were sectioned coronally in a cryostat

into 40 |tm slices. Immunohistochemistry was performed on floating sections. The

sections were blocked in 0.01M PBS/ 1% bovine serum albumin /0.3% Triton-X 100 for

one hour at room temperature, followed by incubation in a GFP polyclonal antibody









(Clontech Laboratories, Palo Alto, CA, 1:2000 dilution) at 40C for 24-48 hrs. After

rinsing in high salt buffer sections were incubated for 24 hrs at RT in secondary antibody

conjugated to biotin followed by incubation in Extravidin (Sigma) for 1 hour. Sections

were then stained in diaminobenzidine (Sigma), mounted on slides, dried overnight,

dehydrated through alcohol and coverslipped.



Radioimmunoassays

Leptin

Serum and plasma leptin was assayed in glass tubes using a rat leptin RIA kit

(Linco Research, Inc., St Charles, MO) according to manufacturer's instructions. The

sensitivity of this assay is 0.5 ng/ml and the range of detection is 0.5 ng/ml to 50 ng/ml.

All samples were assayed in duplicate to minimize variability. Leptin in the CSF was

measured in polystyrene tubes using a more sensitive rat/mouse leptin RIA kit (ALPCO,

Windham, NH). The sensitivity of this assay is 6 pg/ml and the range of detection is

12.5-800 pg/ml.



Insulin

Insulin was measured in polystyrene tubes with a rat insulin RIA kit (Linco

Research, Inc., St. Charles, MO) according to manufacturer's instructions. The sensitivity

of the assay is 0.1 ng/ml and the range of detection is 0.1 ng/ml to 10 ng/ml. All samples

from one experiment were analyzed in a single assay.









Norepinephrine

The NE RIA was performed in polystyrene tubes performed using a kit from

ALPCO (Windham, NH) according to manufacturer's instructions. In this assay NE is

first extracted from the urine sample using a cis-diol specific boronate affinity gel,

simultaneously acylated to N-acylnorepinephrine and then converted enzymatically into

N-acylnormetanephrine. The extracted sample is run in the RIA. The analytical

sensitivity of this assay is 135 pg/ml and the range of detection is 0.15 to 0.5 ng/ml.



Thyroid hormones (T3 and T4)

The thyroid hormones T3 (tri iodo-thyronine) and T4 tetraa iodo-thyronine) were

analyzed in serum and plasma samples using solid phaseT3 and T4 RIA kits (ICN

Pharmaceuticals, Inc. Costa Mesa, CA). The assays were conducted separately for the

two hormones in tubes coated with either T3 polyclonall) or T4 (monoclonal) antibody.

The analytical sensitivity of the T3 RIA is 6.7 ng/dL, while that of the T4 RIA is 0.76

ltg/dL.



Free Fatty Acid Analysis

Serum non-esterified fatty acid (FFA) levels were measured by an in vitro

enzymatic colorimetric method for quantitation of non-esterified fatty acids (free fatty

acids) using a NEFA-C kit (Wako Chemicals USA Inc. Richmond, VA) according to

manufacturer's instructions. In this assay there is acylation of coenyme A (CoA) by the

fatty acids in the presence of additional acyl-CoA synthatase (ACS). The acyl-Co-A thus

made is oxidized by the addition of acyl-CoA oxidase with generation of hydrogen

peroxide. Hydrogen peroxide in the presence of peroxidase leads to the oxidative









condensation of 3-methyl-N-ethyl-N-(p-hydoxyethyl)-aniline with 4-aminoantipyrine to

form a purple adduct that is measured colorimetrically at 550 nm.



Glucose Measurement

Serum/plasma glucose was measured using Sigma Diagnostics Glucose (Trinder)

reagent. This is a quantitative, enzymatic determination of glucose. It is a colorimetric

reaction, read at 505 nm.



Recombinant AAV Production

To produce a recombinant AAV, DNA from two plasmids are transfected into a

host cell in culture. One of these plasmids consists of an AAV vector containing the

AAV terminal repeat sequences (ITR's) and the promoter along with the gene of choice.

The other plasmid is a helper plasmid that contains both the rep and cap genes needed for

packaging the DNA flanked by the ITR's (Hauswirth et al., 2000) as well as Ad helper

genes required for AAV infection. A plasmid construct, pTR-leptin, containing 640 bp

rat leptin cDNA (a kind gift from Dr Roger Unger, Southwestern Medical center, Dallas,

TX) under the control of a hybrid chicken 3 actin (CBA) promoter with a

cytomegalovirus (CMV) enhancer and AAV terminal repeats was generated. To produce

rAAV-leptin, rAAV vector plasmid was co-transfected with pTR-leptin. The product was

then co-transfected with the helper plasmid pDG carrying the AAV rep and cap genes, as

well as Ad helper genes, required for rAAV replication/packaging. Plasmid DNA used in

the transfection was purified by conventional alkaline lysis/CsCl gradient protocol.

Before transfection the presence of the two flanking ITR's in the plasmid DNA isolated









from Escherischia coli was confirmed with digestion of 0.5 |tg plasmid DNA with Sma I.

The transfection was carried out by incubation of HEK-293 cells (low passage number,

passage 29-40) at 370C for 48 hrs in the presence of calcium phosphate precipitate of

rAAV and pDG plasmids. Forty-eight hr post-transfection, cells were harvested by

centrifugation for 10 min at 1,140 g. Cells were then lysed in 15 ml of 0.15 M NaCl 50

mM Tris HC1, pH 8.5 by 3 freeze/thaw cycles in dry ice-ethanol/370C baths. Benzonase

(Nycomed Pharma A/S, pure grade) was added to the mixture (50 tlg/ml final

concentration) and the lysate was incubated for 30 min at 370C. The crude lysate was

clarified by centrifugation at 3,700 g for 20 min and the virus-containing supernatant was

further purified by iodixanol (Nycomed) density gradient centrifugation in a Type 70 Ti

rotor at 350,000 g for 1 hr at 180C. Each gradient consisted of (from the bottom up): 60%,

40%, 25%, and 15% iodixanol. The last density step contained 1 MNaC1. After

centrifugation, the 40% iodixanol fraction containing virus was applied to heparin

agarose Type I column (Sigma, St Louis, MO) equilibrated with PBS-MK (IxPBS-1 mM

MgC2, 2.5 mM KC1). The rAAV was eluted with the same buffer containing 1 M NaC1.

The virus was concentrated by centrifugation through the BIOMAX 100 K filter

(Millipore) according to the manufacturer's instructions. The control vector, rAAV-UF5

was produced using a similar protocol after generation of a construct plasmid containing

green flourescent protein (GFP) cDNA The virus thus produced was titred using two

methods, the Infectious Center Assay and Quantitative PCR (Hauswirth et al., 2000).The

ratio of physical-to-infectious particles was less than 100. rAAV vectors, purified using

iodixanol gradient/heparin-affinity chromatography, were 99% pure as judged by the

PAAG/silver-stained gel electrophoresis (not shown). Since mini-Ad helper plasmid









pDG was used to produce the vectors, there was no Ad or wtAAV contamination in the

rAAV stocks used in these studies.





Statistical Analysis

Weekly BW and 24 hr food intake in rats were compared between groups using a

two way ANOVA with treatment and time as variables. The p value was set at p<0.05 to

attain significance. In experiments with three experimental groups, BW and FI were

compared using one way ANOVA followed by post hoc analysis using Neuman Keul's

multiple comparison test.

Circulating leptin, insulin, glucose,T3, T4, and NEFA and urinary NE levels

between two treatment groups were compared using Students 't' test with p<0.05

considered significant. Circulating serum/plasma leptin, insulin and glucose levels were

compared across time were done so using a one way ANOVA followed by Neuman

Keul's multiple comparison test post-hoc.

Body fat, body protein and total oxygen consumption were compared using

Students 't' test. UCP-1, UCP-3, POMC, AGRP, NPY and leptin mRNA levels were

compared using Students 't' test. p<0.05 was considered significant in all analyses.

In ob/ob mice average FI data and serum/plasma leptin levels for each group were

compared by ANOVA and Tukey's post hoc test. For BW data, least squares means for

each group were calculated for the change from initial values. These means were

compared by 2-way ANOVA with time and treatment as variables followed post hoc by

Tukey's multiple comparison test for comparisons within a group and among groups at

each time point. Level of significance was set at p<0.05.






41


All analyses were performed using Graph Pad Prism Software Version 3.00 for

windows (Graph Pad Software, San Diego, CA, USA, www.graphpad.com).














CHAPTER 3
LEPTIN GENE THERAPY REVERSES OBESITY IN OB/OB MICE


Introduction



Leptin, a 16-kD secreted protein, is secreted predominantly by the adipose tissue

(Zhang et al., 1994). Serum levels of leptin correlate directly with body-fat mass

(Considineet al., 1996). Leptin exerts its effects on appetite and thermogenesis via

receptors located in the central nervous system (Spiegelman and Flier, 1996) and is

thought to act mainly on the hypothalamus, although leptin receptors are also present in

peripheral tissues. There are several genetic models of obesity in rodents, one of which is

the ob/ob mouse which bears a mutation in the gene encoding leptin, resulting in lack of

leptin and increased food consumption. This is the underlying cause of the increased

mass of adipose tissue and development of a syndrome that resembles morbid obesity and

non insulin dependent diabetes mellitus in humans (Halaas et al., 1995). Additional

abnormalities associated with the ob mutation are hypothermia, lethargy, hyperglycemia,

glucose intolerance and hyperinsulinemia. These abnormalities are alleviated by

administration of recombinant leptin which reduces food intake, increases energy

expenditure and leads to a loss of BW and fat mass (Pelleymounter et al., 1995; Halaas et

al., 1995; Maffei et al., 1995). However, for leptin to be effective there is a need for

continuous administration of the recombinant leptin protein; withdrawal results in

reversal to the obese phenotype (Giese et al., 1996). The delivery of genes via gene









therapy, especially in diseases that are a result of missing or mutated genes, is especially

useful since it offers the potential for sustained delivery of the gene in question.

Adenoassociated virus is a non-immunogenic, non-pathogenic virus (Berns and

Bohenzky 1987). Long term expression of AAV vectors has been demonstrated in the

lung, liver, muscle, heart as well as the brain (Kaplitt et al., 1994; Flotte et al., 1996). In

order to achieve long-term sustained delivery of leptin in ob/ob mice we constructed a

recombinant AAV vector encoding leptin cDNA. There were three objectives of this

study (1) to enhance leptin levels in the blood for extended periods of time (2) to test the

efficacy of intravenously delivered rAAV-leptin on reducing BW and food intake (FI)

and (3) to determine whether enhanced leptin levels sustained over an extended period

would result in leptin resistance in obese leptin deficient ob/ob mice.


Materials and Methods

Animals

Leptin mutant ob/ob mice were purchased from Jackson Laboratories (Bar

Harbor, Maine). The animals were housed 4 per cage, in a temperature (230C) and light

controlled (14 hr light, 10 hr dark), specific pathogen free environment; standard mouse

chow (Teklad, Madison, WI) and water were available ad libitum. The animal protocols

were approved by the Institutional Animal Care and Use Committee.

Recombinant AAV Production

To produce a recombinant AAV, DNA from two plasmids are transfected into a

host cell in culture. One of these plasmids consists of an AAV vector containing the

AAV terminal repeat sequences (ITR's) and the promoter along with the gene of choice.

The other is a helper plasmid that contains both the rep and cap genes needed for









packaging the DNA is flanked by the ITR's (Hauswirth et al., 2000) as well as Ad helper

genes required for AAV infection. A plasmid construct, pTR-leptin, containing 640 bp

rat leptin cDNA (a kind gift from Dr Roger Unger, Southwestern Medical center, Dallas,

TX) under the control of a hybrid chicken 3 actin (CBA) promoter with a

cytomegalovirus (CMV) enhancer and AAV terminal repeats was generated (Fig. 3-1).

To produce rAAV-leptin, the rAAV vector plasmid was co-transfected with pTR-leptin.

The product was then co-transfected with the helper plasmid pDG carrying the AAV rep

and cap genes, as well as Ad helper genes required for rAAV replication/packaging.

Plasmid DNA used in the transfection was purified by conventional alkaline lysis/CsCl

gradient protocol. Before transfection the presence of the two flanking ITR's in the

plasmid DNA isolated from Eschereschia coli was confirmed with digestion of 0.5 |tg

plasmid DNA with Sma I. The transfection was carried out by incubation of HEK293

cells (low passage number, passage 29-40) at 370C for 48 hrs in the presence of calcium

phosphate precipitate of rAAV and pDG plasmids. Forty-eight hrs post-transfection,

cells were harvested by centrifugation for 10 min at 1,140 g. Cells were then lysed in 15

ml of 0.15 M NaCl 50 mM Tris HC1 pH 8.5 by 3 freeze/thaw cycles in dry ice-

ethanol/370C baths. Benzonase (Nycomed Pharma A/S, pure grade) was added to the

mixture (50 [tg/ml final concentration) and the lysate was incubated for 30 min at 370C.

The crude lysate was clarified by centrifugation at 3,700 g for 20 min and the virus-

containing supernatant was further purified by iodixanol (Nycomed) density gradient

centrifugation in a Type 70 Ti rotor at 350,000 g for 1 hr at 180C. Each gradient consisted

of (from the bottom up): 60%, 40%, 25%, and 15% iodixanol. The last density step

contained 1 M NaC1. After centrifugation, the 40% iodixanol fraction containing virus









was applied to heparin agarose Type I column (Sigma, St Louis, MO) equilibrated with

PBS-MK (IxPBS-1 mM MgCl2, 2.5 mM KC1). The rAAV was eluted with the same

buffer containing 1 M NaC1. The virus was concentrated by centrifugation through the

BIOMAX 100 K filter (Millipore) according to the manufacturer instructions. The

control vector, rAAV-UF5 was produced using a similar protocol after generation of a

construct plasmid containing green flourescent protein (GFP) cDNA (Fig. 3-1). The

virus thus produced was titred using two methods, the Infectious Center Assay and

Quantitative PCR (Hauswirth et al., 2000).

The titre of rAAV- leptin used in the study was 3x1013 physical particles/ml. The

ratio of physical-to-infectious particles was less than 100. rAAV vectors, purified using

iodixanol gradient/heparin-affinity chromatography, were 99% pure as judged by the

PAAG/silver-stained gel electrophoresis (not shown). Since mini-Ad helper plasmid

pDG was used to produce the vectors there was no Ad or wtAAV contamination in the

rAAV stocks used in this study.

Study Design

Male ob/ob mice were randomly divided into 4 groups (n=7-8 per group). Mice

were injected with one of 3 different doses (6 x109, 6 x1010, 6x1011 particles) of rAAV-

leptin or a single dose of rAAV- UF5 (control) in a 50 [tl volume via the tail vein. BW

was monitored weekly for the duration of the experiment (75 days). Average food

consumed per week by each mouse was calculated by placing pre-weighed food pellets in

the cages and re-weighing after 7 days. Food intake was first recorded at day 18 post-

injection when a difference in BW was apparent and was monitored weekly until Day 62

of the experiment. Blood samples were collected by retro-orbital puncture on Days 54









and 75 post-injection for analysis of serum leptin levels with a rat leptin

radioimmunoassay kit (Linco Research, Inc, St Louis, MO). On Day 75 post-injection,

animals were sacrificed by cervical dislocation.

Data Analyses

Average FI data and serum leptin levels for each group were compared by

ANOVA and Tukey's post hoc test. For BW data, least squares means for each group

were calculated for the change from initial values. These means were compared by 2-

way ANOVA with time and treatment as variables followed post hoc by Tukey's

multiple comparison test for comparisons within a group and among groups at each time

point. Level of significance was set at p<0.05.




Results

There was a clear dose dependent effect of rAAV-leptin on BW in ob/ob mice.

Analyses of the BW data showed significant effects of treatment and time (p<0.001).

Administration of 6x1011 particles/ml rAAV-leptin elicited a steady loss in BW

beginning on Day 19 until Day 75 post-injection (p<0.001 vs. initial BW, Fig.3-2). On

Day 75 this represented a loss of 32% from the pre-injection weight (Fig. 3-2) and was

44% lower than the BW of obese control mice at this time (p<0.001). This loss in BW

was accompanied by a 24% decrease in FI throughout the experiment (p<0.01 vs. rAAV-

UF5 treated controls, Fig 3-3). In contrast, injection of a one order lower magnitude titre

(6x1010) of rAAV-leptin prevented BW gain, resulting in maintenance of BW in the pre-

injection range for the duration of the experiment (Fig.3-2). This maintenance effect was

apparent from Day 26 post-injection when the BW of controls and 6x109 particles rAAV-









leptin injected animals displayed a progressive increase in BW. Interestingly, BW

maintenance in the 6x1010 particles-treated mice was not accompanied by any significant

change in FI. The BW and FI of mice injected with the lowest rAAV-leptin dose (6x109

particles) were indistinguishable from those of controls.

Plasma leptin levels assayed on Days 54 and 75 post-injection are shown in Fig.3-

4. Leptin levels in control, 6x109 and 6x1010 particles rAAV-leptin injected mice were at

the lower end of sensitivity of the rat leptin RIA. However, mice injected with 6x1011

rAAV-leptin particles, displayed detectable circulating leptin concentrations in the range

of 2-3ng/ml on both Days 54 and 75 post-injection.


























Fig. 3-1. Schematic diagrams of the rAAV vector constructs used in the study. TR:
145-bp AAV terminal repeat sequence; CMV-P-act: cytomegalovirus (CMV) enhancer
linked to chicken 3-actin promoter; pA: polyadenylation site, GFP: green fluorescent
protein, leptin: rat leptin cDNA, IVS: intervening sequence.








) --0- rAAV-UF5 1010particles
C 10
10 -- rAAV-leptin 109particles
S-V-rAAV-leptin 1010particles
S 0- -- rAAV-leptin 10"particles


1 -10-


-20 iI
0 15 30 45 60 75
Days Post-Injection




Fig. 3-2. Dose dependent change in body weight in ob/ob mice post iv rAAV-leptin
injection. Body weight was significantly (p<0.001) reduced by a single 50 [tl
intravenous injection of rAAV-leptin in obese male ob/ob mice in a dose dependent
manner, the highest dose was the most effective.


pTR-leptin TR CMV I-act IVS LEPTIN A



pTR-UF5 CMV -act GFP
pTR-UF5 TR CMV IP-act -- GFP (pA) TR











-- rAAV-UF5 10 10 particles
-- rAAV-leptin 10 9 particles
SrAAV-leptin 10 10 particles
SrAAV-leptin 10 11 particles


0 10 20 30 40


50 60 70


Days Post-Injection




Fig. 3-3. Effect of different doses of rAAV-leptin on food intake in ob/ob mice. Food
intake in ob/ob mice was decreased by the highest dose (p <0.01 vs. control). n = 6-8
per group.


E





1-



0-
l-


SrAAV-UF5 (control)
I rAAV-leptin 6 X 10 particles
I rAAV-leptin 6 X 101 particles
SrAAV-leptin 6 X 101 particles


Day 54


Day 75


Fig. 3-4. Serum leptin levels post iv injection of rAAV-leptin. Serum leptin levels in
ob/ob mice were elevated significantly after intravenous injection of 6 X 10 1 particles of
rAAV-leptin (p<0.05). n = 4-6 per group.


- 40-

-e 35-
--

*l 30-
0
0
t,-
L 25-


w 20-
-


-I'







































Fig. 3-5. Representative ob/ob mice. rAAV-UF5 (left) vs. rAAV-Leptin treated
(right) 6 weeks post injection.









Discussion

The present results demonstrate the successful transfer of leptin using a

recombinant AAV vector in a dose-dependent manner. These results confirm earlier

reports of the use of AAV for gene transfer with long lasting effects (Samulski et al.,

1997; Mandel et al., 1998; Murphy et al., 1997). Gene transfer with intravenous injection

of AAV has been demonstrated in liver, lung and muscle (Samulski et al, 1997). In the

studies presented in this chapter we did not examine the tissues transduced by AAV. A

single injection of rAAV-leptin to ob/ob mice via the tail vein resulted in an elevation in

blood leptin levels for over 10 weeks. In leptin deficient ob/ob mice, the high dose

rAAV-leptin produced circulating leptin levels of 2-3 ng/ml. Mice injected with the

lower dose had almost undetectable serum leptin levels. Basal leptin levels detected also

in the leptin deficient control vector injected ob/ob mice. These baseline values probably

represent non-specific interference in the plasma of ob/ob mice and has been reported by

other researchers previously (Murphy et al., 1997).

The effects of rAAV-leptin on BW were recorded for the entire duration of the

experiment. There was a dose dependent change in BW in the leptin deficient ob/ob mice

with rAAV-leptin injection. The highest dose of rAAV-leptin (6x 011 particles) caused a

significant decrease in BW vs. initial BW starting at 19 days post injection and was

maintained for the duration of the study. Mice given the middle dose (6x1010 particles)

maintained their initial BW and failed to display the BW gain characteristic of ob/ob

mice, suggesting that the small amounts of leptin measured in the circulation was

adequate to affect BW gain. There was no difference in the BW of the lowest dose of

rAAV-leptin (6x109 particles) injected mice as compared to that of controls.

Administration of the leptin protein for short periods leads to decreased BW through









reduction in caloric consumption accompanied by an increase in energy expenditure in

ob/ob mice (Zhang et al., 1994; Halaas et al., 1995; 1997; Pelleymounter et al., 1995;

Harris et al., 1998). Intriguingly, only the highest dose of the vector elicited a change in

food intake. While the weight loss in the high dose group can be attributed to decreased

energy intake, the observation that the middle dose blunted BW gain is the most

interesting. These data confirm the super-sensitivity of ob/ob mice in their response to

leptin, as previously reported (Ioffe et al., 1998). Although this study did not examine the

parameters to measure energy expenditure, it is well documented in the literature that

leptin enhances energy expenditure in ob/ob mice (Halaas et al., 1995, 1997;

Pelleymounter et al., 1995). This highlights a dichotomy in the action of leptin based on

the dose administered. Whereas, the high dose affects energy expenditure in conjunction

with energy intake thus affecting both aspects of the energy homeostasis equation, the

lower effective dose retards the rate of BW gain by affecting energy expenditure only.

Interestingly, reinstatement of leptin via rAAV-leptin injection in ob/ob mice did

not lead to development of leptin resistance during the 75 days of observation.

Resistance to leptin action is usually associated with the presence of higher than normal

peripheral levels of leptin. In this experiment, serum leptin levels even in the highest

dose group did not exceed those of normal wild type mice and do not constitute

hyperleptinemia as seen in obese rodents. This could be one possible basis for the lack of

resistance to leptin in these rAAV-leptin treated animals. Also, since ob/ob mice are

supersensitive to leptin administration (Pelleymounter et al., 1995; loffe et al., 1998)

much higher peripheral leptin levels would likely be necessary to induce resistance.






53


In summary, rAAV is able to effectively introduce leptin into leptin deficient

ob/ob mice. This reinstatement of leptin has a dose dependent effect on decreasing body

mass along with reduction in energy intake and apparent enhancement of energy

expenditure without induction of leptin resistance. This initial demonstration of the

successful use of rAAV to deliver leptin was extended in later chapters to study the

central effects of leptin in lean rodents.














CHAPTER 4
REGULATION OF BODY WEIGHT WITH LEPTIN GENE THERAPY IN LEAN
SPRAGUE-DAWLEY RATS


Introduction

Since its discovery in 1994 by Zhang et al., leptin's integral role in the

maintenance of energy homeostasis has been firmly established. Leptin is a 16 kD

protein secreted primarily by adipocytes (Zhang et al., 1994). Genetically obese ob/ob

mice lack functional leptin and are hyperphagic, hyperinsulinemic, and manifest a

syndrome resembling non insulin dependent diabetes mellitus (NIDDM, Zhang et al.,

1994). The obese phenotype in ob/ob mice can be reversed with recombinant leptin

treatment, along with normalization of the hyperinsulinemia, hyperglycemia and low

basal metabolic rate characteristic of this model (Pelleymounter et al., 1995; Halaas et al.,

1995; Campfield et al., 1995). In Chapter 3 I have reported correction of obesity in these

mice using viral vectors to deliver the leptin gene. Although the ob/ob mice are grossly

obese due to the lack of leptin, this is not the case in obese humans. Morbid obesity

associated with leptin deficiency has been identified in very few humans (Montague et

al., 1997). Most human obesity is not genetic in origin; it is in fact associated with

markedly elevated serum leptin levels that correlate positively with body mass

(Considine et al., 1996). These observations have led to the speculation that obesity may

be a consequence of resistance to the weight reducing actions of leptin. However, the site

and mechanism of leptin resistance is not known. A likely possibility to account for the

lack of leptin action despite high circulating levels is inadequate availability of leptin at









its site of action. This concept is supported by the demonstration of decreased

cerebrospinal fluid to blood leptin level ratio in obese humans which implies decreased

availability of leptin in the CNS of these patients (Caro et al., 1996). Diet induced

obesity in rodents is also linked to a decrease in the amount of leptin transported into the

CSF (Bruguera et al., 2000).

The localization of leptin receptors suggests that the primary site of action of

leptin is the hypothalamus. Leptin receptors are localized in critical nuclei of the

hypothalamus such as the ARC, PVN, VMH and the DMN that form the neural substrate

for the appetite regulating network (Elmquist et al., 1997). Recombinant leptin

administered intracerebroventricularly (icv) in rodent models of obesity is effective in

reducing food intake and enhancing metabolism (Halaas et al., 1997).

Evidence from rodent studies that diet induced obesity associated with peripheral

hyperleptinemia can be attenuated with central but not peripheral leptin treatment (Halaas

et al., 1997) and centrally administered leptin has a greater potency than either

intravenous infusion or intramuscular injection confirm that leptin's effects on energy

balance occur at the level of the CNS (Halaas et al., 1997; Campfield et al., 1995). These

results raise the question whether continuous elevation of leptin at its site of action would

also result in leptin resistance as seen with peripheral hyperleptinemia, or whether leptin

resistance is a consequence of insufficient availability at the site of action, as alluded to

above. In order to address this question we generated an adenoassociated viral vector

(rAAV) encoding leptin. AAV is ideally suited for use in in vivo experiments in that it is

non-immunogenic and non-pathogenic (Berns and Bohenzky 1987) and can transduce

non-dividing neurons in a variety of brain regions (Kaplitt et al., 1994). When delivered









into the CNS AAV transfects neurons preferentially and has been shown to have stable,

long-term expression lasting for upto 18 months post injection (Mandel et al., 1998). In

order to test the hypothesis that resistance to endogenously high leptin levels is not due to

resistance to leptin action within the CNS but rather to insufficiency at its target sites, we

examined the long-term effects of leptin delivery in the central nervous system at its site

of action in wild-type lean male and female rats. If there is central resistance to

hyperleptinemia then chronic enhancement of leptin levels within the CNS would be

expected to result in obesity. The data from this experiment support the hypothesis that

obesity may be due to inadequate availability of leptin at its target sites.




Methods

Study Design

Male and female lean Sprague-Dawley rats recieved permanently cannula in the

third ventricle and were randomly divided into two groups each (n= 6-8 per group) so

that the average BW of the groups was identical. One group each was injected icy with a

single injection of 6 X10 10 particles rAAV-leptin in a 5 [tl volume. Control groups

received 5 [tl or rAAV-UF5 icy. BW and 24 hour food intake were recorded weekly for

a period of 12 weeks (n= 6-8 per group). At 12 weeks post-injection the animals were

sacrificed by decapitation, blood was collected for subsequent analysis of leptin by RIA.

A second experiment was performed to characterize the expression of rAAV

post icy injection. Female Sprague-Dawley rats (9-12 per group) were implanted with

permanent third ventricle cannulae and injected with either rAAV-UF5 or rAAV-leptin,

as above. Animals were sacrificed 6 weeks post injection. Three animals per group were









used for body fat and protein determination. Three of the rAAV-UF5 treated animals

were perfused and their brains were collected and sectioned for immunohistochemical

localization of GFP. Hypothalami were rapidly excised from the brains of six rAAV-

leptin and six rAAV-UF5 injected rats for RT-PCR determination of leptin mRNA

expression.



Carcass Fat and Protein Estimation

Carcass water, fat and fat-free dry mass were determined gravimetrically (Fong,

1989). Carcasses were weighed immediately after killing the rats, then frozen in liquid

nitrogen and pulverized with solid carbon dioxide in a commercial blender. Pulverized

carcasses were dried for 2-4 days to a constant mass at 800C. Lipid content was

determined by sequential chloroform-methanol (1:1), ethanol-acetone (1:1), and

petroleum ether extractions. Carcass protein content was measured from dried carcass

aliquots after NaOH extraction with a routine Bradford protein assay.



Immunohistochemistry for Green Flourescent Protein (GFP)

Immunohistochemistry was performed as described earlier (Peel et al., 1997,

Klein et al., 1998). Briefly, animals were anesthetized with 100mg/kg sodium

pentobarbital ip and perfused with 300 ml 0.9% saline followed by 300-600 ml cold 4 %

paraformaldehyde in 1 X PBS. Brains were removed and kept in 4%

paraformaldehyde/1XPBS overnight, then in 20 % sucrose solution, followed by 30 %

sucrose until the tissue submerged. The brains were sectioned coronally in a cryostat

into 40 |tm slices. Immunohistochemistry was performed on floating sections. The

sections were blocked in 0.01M PBS/ 1% bovine serum albumin /0.3% Triton-X 100 for









one hour at room temperature, followed by incubation in a GFP polyclonal antibody

(Clontech Laboratories, Palo Alto, CA, 1:2000 dilution) at 40C for 24-48 hrs. After

rinsing in high salt buffer sections were incubated for 24 hrs at RT in secondary antibody

conjugated to biotin followed by incubation in Extravidin (Sigma) for 1 hour. Sections

were then stained in diaminobenzidine (Sigma), mounted on slides, dried overnight,

dehydrated through alcohol and coverslipped.



Leptin mRNA Expression using RT-PCR

Leptin mRNA expression was analyzed using reverse transcriptase-PCR (RT-

PCR). Briefly, total RNA was extracted from hypothalami using the RNA STAT 60

RNA isolation kit (Tel test Inc, Friendswood, TX). First-strand cDNA was synthesized

using 1 ug total RNA with a RNA PCR kit. All reagents were purchased from PE

Biosystems, Foster City, CA. Primers were designed to the rat leptin gene to encompass

a 308 bp region of the coding sequence. (Gen bank Accession code D49653), Sense: 3'

CCC ATT CTG AGT TTG TCC, Antisense: 3' GCA TTC AGG GCT AAG GTC.

Primers were designed for cyclophilin (internal control) to generate a 470 bp product

(Gene bank accession code M19533). Sense: 3' GAC AAA GTT CCA AAG ACA GCA

GAA A, Antisense: 3' CTG AGC TAC AGA AGG AAT GGT TTG A. The PCR

products generated by these primers were sequenced and independently verified and

found to match rat leptin and rat cyclophilin completely. Linearity of the PCR was tested

by amplification for 20-45 cycles for leptin and cyclophilin. The linear range was found

to be between 25 and 40 cycles.









Five microliters of the first-strand cDNA was amplified for 30 cycles for leptin

and 26 cycles for cyclophilin. Each gene was amplified in a separate PCR reaction from

a single RT reaction by using the following parameters:

Leptin: Denaturation @ 95 OC, 1 min, annealing @ 56 OC, 1 min, extension @ 72

OC, 1 min, 30 cycles, 10 min final extension 72 OC.

Cyclophilin: Denaturation @ 94 OC for 50 s, annealing @ 55 OC for 45 sec,

extension @ 72 OC for 2 min, 26 cycles.

PCR products were analyzed using agarose gel electrophoresis. Twenty

microlitres of the PCR products were separated on a 2% agarose gel stained with

ethidium bromide and placed on an UV illuminator equipped with a camera connected to

a gel documentation system (BIORAD). The gel image was analyzed using an image

analysis program (Image Quant system BIORAD laboratories Inc). The relative

expression of the mRNA levels were derived from a comparison of the intensity of the

target and simultaneously run internal controls (cyclophilin). All PCR products were run

on a single gel in order to control for inter gel variation.



Statistical Analysis

Weekly BW and food intake data were compared between groups using a two

way ANOVA with treatment and time as variables. Serum leptin levels, body fat content,

body protein content and hypothalamic leptin mRNA were compared between treatment

groups using Students 't' test. The p value was set at p<0.05 to attain significance.












Results

Hypothalamic leptin mRNA RT-PCR

Leptin gene expression in the hypothalamus of rAAV-leptin injected rats was

confirmed by relative RT-PCR. As shown in Fig 4-1, a faint band of leptin mRNA is

seen in rAAV-UF5 rats. In the rats injected with rAAV-leptin this band was very

prominent. Leptin mRNA levels expressed relative to cyclophillin mRNAwere

significantly higher in the hypothalami of rAAV-leptin treated rats ( p<0.05 Fig.4-1 ).



Immunohistochemical localization of GFP

Gene expression following viral vector therapy was verified by

immunohistochemical localization of GFP in the brain of rAAV-UF5 injected control

rats. No GFP positive cells were observed in the negative controls which were not

incubated with the GFP antibody. At 6 weeks post-injection, GFP-positive cells were

distributed in mid-line structures along the site of injection in the third cerebroventricle,

and the lateral ventricles extending from the anterior commissure to the posterior

hypothalamus (Fig. 4-2, 4-3). Almost all these cells displayed neuron-like morphology

with GFP immunoreactivity also evident in dendritic and axonal fibers. Prominent

clusters of immunopositive cells were localized in the bed nucleus of the anterior

commissure, ventrally in the preoptic area and in the suprachiasmatic nucleus. Caudally,

GFP-positive cells were seen in the anterior hypothalamic area, paraventricular nucleus,

dorsomedial hypothalamus and arcuate nucleus (Fig4-3). Occasionally, small groups of

GFP-positive cells were seen in remote areas near the hippocampus.









Body Weight

In female rats BW was significantly decreased (p<0.05) following the icy rAAV-

injection vs. the rAAV-UF5 treated control group (Fig 4-4). The first significant

decrease in BW was observed at 2 weeks post injection and was maintained until 7 weeks

with a return to initial BW from weeks 8 to 12 post-injection. These rats exhibited no

gain in BW over the 12 week period. In contrast, the control rats exhibited an overall

gain of 17 % over their initial BW so that the rAAV-leptin treated females weighed 17%

less than the controls at the end of the experiment.

The response of male rats to rAAV-leptin was different to that of females. The

rAAV-leptin treated male rats continued to gain weight, albeit at a slower rate than the

control animals (Fig 4-5, Controls 408.6 + 7.2 g vs.. 363.7 + 2.3 rAAV-leptin at the end

of 12 weeks). At the end of the experiment at 12 weeks post-injection the rAAV-UF5

treated control male rats had gained 30 % BW compared with 17% in the rAAV-leptin

treated group (p<0.05).

Food Intake

There was no difference in 24 hr food intake (FI) in either the female (overall

average, 19.5 + 1. 0 g controls vs. 18.6 + 1.2 g rAAV-leptin group, Fig 4-6) or male

(Overall average, 24.6 + 0.5 g controls vs. 24.4 + 0.5 g rAAV-leptin, Fig. 4-7) rats

administered rAAV-leptin vs. their respective control groups.



Serum Leptin Levels

As expected, blood leptin levels in female control rats were significantly higher

than in male control rats (p<0.05, Fig 4-8 and 4-9). Circulating levels of leptin in both

males and females were markedly suppressed when measured at 12 weeks following a









central rAAV-leptin injection. In rAAV-leptin treated female rats serum leptin levels

were significantly lower than in rAAV-UF5 treated female rats (0.8 + 0.2 ng/ml vs. 3.7+

0.6 ng/ml, p<0.05). Similarly, the rAAV-leptin treated male rats displayed significantly

lower leptin levels than the rAAV-UF5 treated group (0.5+ 0.1 vs. 1.98 + 0.2 ng/ml).



Carcass Fat and Protein Analysis

Body composition analysis at 6 weeks post-injection indicated that weight

reduction was due solely to a loss of fat depots. There was a significant depletion in fat

in rAAV-leptin treated rats (p<0.05). Total body fat wasl2.5% in controls vs. 5.4% in

the rAAV-leptin treated group (Fig 4-10 A and Fig 4-11) there was with no change in

lean mass with rAAV-leptin treatment. (Fig. 4-10, A and B,).
















rAAV-UF5 rAAV-leptin
II I
Leptin
Cycloph.

3.0

< 2.5
Z
W : 2.0-
E*
E- 1.5
4) 1-
I 1.0-


0.5

0.0rAAV-UF5 rAAV-eptin
rAAV-UF5 rAAV-leptin


Fig 4-1. Effect of a single injection of rAAV-leptin on leptin mRNA expression in the
hypothalamus of female Sprague-Dawley rats. There was a significant induction in
hypothalamic leptin mRNA expression in rAAV-leptin treated rats vs. controls (p<0.05).
n= 6 per group.









A. B.































Fig. 4-2. Photomicrograph (4X) of representative hypothalamic sections showing
GFP positive cells around the site of icy rAAV-UF5 injection The panel on the left
(A) represents a GFP antibody -ve control with no positively stained cells, the panel on
the right (B) is positively stained for GFP, transduction of neurons is visible along the
third ventricle. OC = optic chiasm 3V= third ventricle, AC = anterior commissure.






















































Fig. 4-3. Photomicrograph (20 X) of a representative hypothalamic section showing
GFP immunoreactivity in neurons and fibres transduced by rAAV-UF5. Groups of
GFP positive cells are seen in the ARC, extending along the third ventricle towards the
PVN. ARC= Arcuate nucleus of the hypothalamus, 3V= third ventricle, PH = posterior
hypothalamus.















275-


zo-
250-


g 225-

0
M 200-


175-


FEMALE


-0- rAAV-UF5
-*- rAAV-leptin


I I
4 8
Weeks post-injection


Fig 4-4. Effect of rAAV-leptin on BW in female Sprague-Dawley rats. BW was
significantly decreased in rAAV-leptin treated rats vs. control rAAV-UF5 treated rats
(p<0.05) over a period of 12 weeks following a single injection of rAAV-leptin. n = 6-8
per group.


rAAV-UF5


450-

- 425-
4o-
400-

*7 375-

S350-
-
O 325-

300-


-- rAAV-leptin


MALE


0 2 4 6 8 10 12
Weeks post-injection


Fig 4-5. Effect of a single injection of rAAV-leptin on BW in male Sprague-Dawley
rats. BW was significantly lower in rAAV-leptin treated rats vs. rAAV-UF5 treated
control rats (p<0.05) for the 12 week period of the study. n=6 per group.
















FEMALE


-0- rAAV-UF5
-- rAAV-leptin


U I I
0 4 8 12
Weeks post-injection



Fig 4-6. Twenty four hour food intake in female Sprague-Dawley rats injected icy
rAAV-UF5 (control) or rAAV-leptin. Food intake was unchanged with treatment. n =
6-8 per group.


40-

-
S30-

"1-
O 20-
O
LL
U-

S10-
C4


MALE

^S*^r^=


- rAAV-UF5
-- rAAV-leptin


U I
0


Weeks post-injection


Fig 4-7. Twenty four hour food intake in male Sprague-Dawley rats injected icy
with rAAV-UF5 (control) or rAAV-leptin. Food intake was unchanged with treatment.
n = 6 per group.

















4-
E




1-
1-
3-


0'


FEMALE

T


= rAAV-UF5
M rAAV-leptin


Fig 4-8. Serum leptin levels in female Sprague-Dawley rats. Leptin levels were
significantly decreased (p<0.05) in rAAV-leptin vs. the rAAV-UF5 treated control rats at
12 weeks post-injection. n=6-8 per group.


3-




2-
c,
-1-


0 1-1 1


MALE
TIIAE


= rAAV-UF5
M rAAV-leptin


*


Fig 4-9. Serum leptin levels in male Sprague-Dawley rats. Leptin levels were
significantly decreased (p<0.05) in rAAV-leptin vs. the rAAv-UF5 treated control rats at
12 weeks post injection. n= 6 per group.










40-


30-



20-


0-


3.0x10 M



2.0x10 o-



1.0x10 1-


0.0x10 .-


m rAAV-UF5
rAAV-leptin


Fig 4-10. Body composition analysis post rAAV-leptin injection. Body fat was
significantly decreased (A) while lean mass as depicted by protein composition was
unaltered (B) at 6 weeks after a single injection of rAAV-leptin.
































Fig. 4-11. Effect of rAAV-leptin on body fat. Abdominal fat is depleted in the rAAV-
leptin. Left = rAAV-UF5, Right = rAAV-leptin









Discussion

In order to test the hypothesis that leptin resistance is due to insufficient

availability at the site of action the objective of this study was to determine the long term

efficacy of sustained leptin elevation in the CNS. Leptin was delivered into the CSF of

wild type (wt) lean male and female rats via a rAAV viral vector encoding leptin and BW

and FI were measured for 12 weeks. We demonstrate here a successful transduction of

neurons in the CNS at the site of action of leptin. We observed several GFP positive cells

in and around the hypothalamic nuclei associated with the regulation of appetite. We

believe we have enhanced local leptin production in the hypothalamus by rAAV

mediated central leptin gene delivery. This idea is supported by the demonstration of

increased leptin mRNA in the hypothalamus in addition to the presence of several groups

of GFP positive neurons in the hypothalamus. A single icy injection of rAAV-leptin was

effective in maintaining BW in Sprague-Dawley rats for 12 weeks post-injection.. We

observed a significant decrease in BW with leptin treatment in both male as well as

female rats as compared to the control rats. However, there was a sex-specific difference

in the response to rAAV-leptin treatment. Whereas the males continued to grow, albeit at

a slower rate than the controls (17 % vs. 30 % total weight gain), the female leptin treated

rats maintained BW at or below the pre-injection level. One possible explanation for this

sex difference in BW response could be the effective dose of rAAV-leptin. Since both the

male and female rats received the same actual dose, the males with a 30 % higher initial

BW received a 30% lower relative dose. It is possible that increasing the relative dose in

male rats would produce the same pattern of changes in BW as displayed by the female

rats. It is also likely that this divergence may be due to the "fat melting" action of leptin

(Chen et al ., 1996). As females have a higher fat mass compared with males, thus, this









may account for part of the difference in response to rAAV-leptin. We observed a

striking reduction in body fat post icy rAAV-leptin injection without any change in

muscle mass. These results confirm previous observations that leptin acts to reduce BW

by specifically targeting the adipose mass in the body. Because peripheral leptin levels

were markedly suppressed with icv rAAV-leptin injection, this fat melting effect is likely

due to the central effect of leptin at its hypothalamic sites of action rather than peripheral

action directly at the level of the adipocytes..

A large body of experimental evidence shows that leptin acts on the hypothalamic

appetite regulating network to inhibit FI (Reviewed in Kalra SP et al., 1998). However,

contrary to these reports, in our study the reduction in BW was not accompanied by a

decrease in caloric consumption. FI in neither the male, nor the female, leptin treated

groups differed from their respective controls. This brings up an interesting question as to

how the weight loss is mediated by leptin in these animals and argues against the

hypothesis that leptin is predominantly an appetite suppressing hormone. Evidence

suggestive of a selective effect of leptin on energy expenditure, independent of an effect

on FI, has been previously reported (Harris et al., 1998; Breslow et al., 1999). Leptin has

been shown to actively stimulate the sympathetic nervous system to enhance energy

expenditure by way of increased thermogenesis by activation of uncoupling proteins in

BAT (Scarpace et al., 1997). Indeed, the location of GFP positive cells in this study

establish a neuroanatomical basis for links with the SNS as well as BAT. It is possible

that the loss or decrease in the rate of weight gain in the rAAV-leptin treated animals may

primarily be attributable to enhanced energy expenditure. This possibility will be

addressed in later chapters.









An interesting observation was the drastic decrease in peripheral leptin levels in

rats injected centrally with rAAV-leptin. The primary role of leptin in metabolic

homeostasis is to provide information to the hypothalamus on the status of body fat,

thereby modulating CNS functions that regulate energy balance (reviewed in Ahima et al

2000; Schwartz et al., 2000). Thus, it would be intuitive that direct injection of leptin

into the CNS falsely serves as a signal to the brain of the presence of adequate fat

reserves and thus initiate a negative feed-back loop that would deplete body fat as seen in

these rats and this would secondarily result in lower peripheral levels of leptin.

At the end of the 12 week experiment we did not observe the onset of leptin

resistance in either sex, although, in female rats there was a return to initial BW. None of

the rAAV-leptin injected animals displayed any inclination towards developing obesity.

This is an important observation in that it directly supports our hypothesis that leptin

resistance is not a result of hyperleptinemia or saturation of leptin receptors in the CNS,

but due to insufficiency of leptin at its target sites. Thus, leptin resistance may be due to

diminished availability of leptin in the interstitial fluid bathing the leptin receptors at

specific hypothalamic sites involved in the regulation of appetite and energy balance. By

sustained enhanced production of leptin with an AAV vector we have chronically

increased production and availability at the critical sites.

Cumulatively, these results document the efficacy of rAAV-leptin gene therapy in

transducing neurons and reveal a sex difference as well as a dichotomy in BW regulatory

mechanisms by leptin that is independent of caloric consumption. This issue is further

explored in later chapters. These data also highlight the use of rAAV as an effective gene

delivery vehicle intracerebroventricularly. Due to its excellent safety features and the






74


ability to transduce a wide variety of tissues AAV will likely be the virus of choice for

gene therapy applications in the future.














CHAPTER 5
LONG TERM EFFECTS OF LEPTIN GENE THERAPY


Introduction

Leptin, a 16 kD protein hormone is an afferent signal from the periphery to the

brain in a homeostatic feedback loop that regulates adipose tissue mass (Zhang et al.,

1994). Obese ob/ob mice that are leptin deficient have a reversal of the obese phenotype

when given recombinant leptin. Leading to the early hypothesis that obesity in humans

maybe a result of either absolute or a relative deficiency of leptin (Friedman and

Halaasl998; Rosenbaum N and Liebel 1998). However, a consistent feature of obesity

is the presence of high levels of circulating leptin (Friedman and Halaasl998). Recent

studies have suggested that reduced sensitivity to rising endogenous leptin levels, as seen

with increasing adiposity, may play a significant role in the development of obesity.

Both diet induced obesity as well as several genetic models of obesity appear to be

associated with resistance to the anorexic effects of leptin (Friedrich et al., 1996; Halaas

et al., 1997). Thus, while the experimental elevation of leptin, within physiological

levels, produces a transient decrease in food intake and weight loss (Halaas et al., 1997;

Ahima et al 1996), the normal physiological context in which leptin acts as a negative

adipostatic signal limiting weight gain in times of nutritional excess, remains to be

defined. The primary site of action of leptin has been determined to be the brain, where

its receptors are localized in critical nuclei of the hypothalamus such as the ARC, PVN,









VMH and the DMN (Schwartz et al., 1996; Elmquist et al 1997; Elmquist et al 1998).

The exact mechanism by which leptin mediates its weight reducing function is uncertain.

A decrease in the cerebrospinal fluid to serum leptin level ratio in obese

individuals, implies decreased availability of leptin in the CNS, the primary site of leptin

action (Considine et al 1995). Further, the fact that obese rodents respond to central

leptin in the face of resistance to peripheral leptin supports a role for reduced leptin

transport to the CNS (Van Heek et al., 1997). Thus, leptin resistance seems to be

associated with insufficiency of leptin at its main site of action, the brain. Data presented

in Chapter 4 of this dissertation support this assumption and demonstrate that leptin

resistance does not occur with chronic elevation of leptin at its site of action. A sustained

increase in leptin levels in the CNS with rAAV encoding leptin for a period of 12 weeks

prevented normal weight gain. The attenuation of normal weight gain, however, was not

accompanied by any decrease in food intake. An imbalance of metabolic energy (energy

absorbed and available for metabolism) compared with energy expenditure yields either a

loss or a gain of body mass. Therefore, one possible explanation for the attenuation in the

rate of weight gain, despite normal food intake could be enhanced energy expenditure.

The experiments discussed in this chapter are designed to explore the underlying

mechanism for retarded BW gain in the presence of normal food intake. In order to

assess energy expenditure in rats with sustained long term elevation of leptin in the CNS

we examined the hormones and other factors associated with enhanced energy

expenditure. While a number of neuroendocrine afferent signals are implicated in BW

homeostasis, the major efferent pathway is the sympathetic nervous system (SNS), which

affects both energy expenditure and substrate utilization. Leptin increases central









sympathetic outflow (Mantzoros et al., 1996). A single leptin injection icy increases

plasma norepinephrine (NE) in a dose dependent manner (Satoh et al., 1999). In rhesus

macaques, icv leptin increased circulating NE levels by approximately 50 % within an

hour of administration (Tang-Christensen et al., 1999). Circulating catecholamines affect

glucose and lipid metabolism (Nonogaki, 2000) and NE has been shown to increase

thermogenesis in BAT via activation of the 33 adrenergic receptors (Landsberg and

Young 1992). In skeletal muscle NE augments glycogenolysis and thereby promotes

energy utilization (Nonogaki, 2000).

Thyroid hormones have important thermogenic function (Krotkiewski 2000) and

leptin alters thyroid hormone levels. Thyroid hormones stimulate resting metabolic rate

and, therefore, enhance energy expenditure in rodents (Jekabsons et al., 1999).

Circulating thyroid hormone levels decrease during fasting and, leptin administration

prevents this drop in plasma triiodo-thyronine (T3) and tetraiodo-thyronine (T4) levels

(Ahima et al., 1996; legradi G et al., 1997). Recent data suggest that central leptin

stimulates T3 production via enhanced conversion of T4 to T3 (Cusin et al., 2000).

Thyroid hormones could thus be important mediators of the effect of leptin on energy

expenditure.

Other candidates involved in regulation of energy homeostasis include the

metabolic regulators, the uncoupling proteins (UCP's, reviewed in Adams 2000). UCP's

are found in the inner mitochondrial membrane and uncouple protons from ATP

synthesis leading to generation of heat as opposed to ATP (Lin and Klingenberg, 1980).

Uncoupling protein-1 (UCP-1) is abundant in BAT, an effector organ for adaptive

thermogenesis in rodents. In rodents, leptin administration upregulates UCP-1 mRNA









expression through sympathetic activation of P33 adrenergic receptors in BAT (Scarpace

et al., 1997). This increase in UCP-1 is an important means by which leptin may regulate

energy expenditure. In humans, the major thermogenic "organ" is the skeletal muscle

(Rolfe and Brand, 1986). UCP-3, a homologue of UCP-1, is expressed in skeletal muscle

and brown fat (Vidal-puig et al., 1997). UCP-3, like UCP-1, is a thermogenic protein

(Boss et al., 1996; Vidal-puig et al., 1997). Indeed UCP-3 expression in skeletal muscle

increases in response to thyroid hormone administration (Vidal-puig et al., 1998) and is

modulated by leptin (Scarpace et al, 1998). Interestingly mice overexpresing UCP-3 are

hyperphagic and lean (Clapham et al., 2000) and have a striking reduction in adipose

tissue mass. Thus it is likely that in our experimental rats with sustained elevation of

leptin in the CNS, body weight restraint without lowered food intake might be due to

activation of UCP-1.

Age is a contributing factor to the development of obesity. An age related increase

in body adiposity leads to increases in peripheral leptin and often times is accompanied

by obesity, possibly due to development of leptin resistance. In humans and rodents

visceral fat or deep abdominal fat levels increase with aging (Brazilai et al., 1998;

Shimokata et al., 1989). Leptin levels increase significantly with age in rodents

(Rasmussen et al., 1999; Wolden-Hanson et al., 1999) as does leptin mRNA (Maffei et

al., 1995). Another factor accompanying aging is the increase in plasma insulin levels

(Rasmussen et al., 1999; Wolden-Hanson et al., 1999), which may contribute to insulin

resistance and NIDDM. We propose that the obesity associated with increasing age is

not due to lack of peripheral leptin or insulin rather is due to decreased availability of

leptin at its target sites. Thus, sustained central availability of leptin should prevent the









adverse effects of aging on obesity, insulin and related factors. The long term studies

detailed in this chapter detail the biochemical basis of BW maintenance without a

reduction in food intake.


Materials and methods

Study Design

Male and female lean Sprague-Dawley rats were permanently cannulated in the

third ventricle, randomly divided into two groups each (n= 6-8 per group) so that the

average BW of the groups was identical. Each group was injected icy with a single

injection of either rAAV-leptin or rAAV-UF5 (controls), a dose of 6 X10 10 particles in

5 [Il. BW and food intake were recorded weekly for 24 weeks. An additional control

group of untreated un-operated females was simultaneously monitored. At 6 and 16

weeks post-injection blood samples were collected from the jugular vein for leptin,

insulin and glucose analyses. Urine samples were collected at week 16 for NE analysis.

At 24 weeks post-injection the animals were sacrificed by decapitation, blood, BAT and

skeletal muscle were collected for analyses.

Radioimmunoassays

Leptin

Serum and plasma leptin was assayed in glass tubes using a rat leptin RIA kit

(Linco Research, Inc., St Charles, MO) according to manufacturer's instructions. The

sensitivity of this assay is 0.5 ng/ml and the range of detection is 0.5 ng/ml to 50 ng/ml.

All samples were assayed in duplicate to minimize variability. Leptin in the CSF was

measured in polystyrene tubes using a more sensitive rat/mouse leptin RIA kit (ALPCO,









Windham, NH). The sensitivity of this assay is 6 pg/ml and the range of detection is

12.5-800 pg/ml.



Insulin

Insulin was measured in polystyrene tubes with a rat insulin RIA kit (Linco

Research, Inc., St. Charles, MO) according to manufacturer's instructions. The sensitivity

of the assay is 0.1 ng/ml and the range of detection is 0.1 ng/ml to 10 ng/ml. All samples

from one experiment were analyzed in a single assay.

Norepinephrine

The NE RIA was performed in polystyrene tubes performed using a kit from

ALPCO (Windham, NH) according to manufacturer's instructions. In this assay NE is

first extracted from the urine sample using a cis-diol specific boronate affinity gel,

simultaneously acylated to N-acylnorepinephrine and then converted enzymatically into

N-acylnormetanephrine. The extracted sample is run in the RIA. The analytical

sensitivity of this assay is 135 pg/ml and the range of detection is 0.15 to 0.5 ng/ml.

Thyroid hormones (T3 and T4)

The thyroid hormones T3 (tri iodo-thyronine) and T4 tetraa iodo-thyronine) were

analyzed in serum and plasma samples using solid phaseT3 and T4 RIA kits (ICN

Pharmaceuticals, Inc. Costa Mesa, CA). The assays were conducted separately for the

two hormones in tubes coated with either T3 polyclonall) or T4 (monoclonal) antibody.

The analytical sensitivity of the T3 RIA is 6.7 ng/dL, while that of the T4 RIA is 0.76

tg/dL.









Dot blot analysis for UCP-1 and UCP-3

Total cellular RNA was extracted as described above. The integrity of the isolated

RNA was verified using 1 % agarose gels stained with ethidium bromide. The RNA was

quantified by spectrophotometric absorption at 260 nm as well as 280 nm using multiple

dilutions of each sample.

The full-length cDNA clone for uncoupling protein-1 (UCP1) was kindly

provided to Dr Phillip Scarpace by Dr. Leslie Kozak, Jackson Laboratory, Bar Harbor,

ME and verified by Northern analysis, as previously described (Scarpace et al., 1997).

Full length UCP- 3 cDNA .was kindly supplied by Dr Olivier Boss and used as

previously described (Boss et al 1997). All probes were random prime labeled using

Prime-A-Gene kit (Promega, Cat # U 1000) according to manufacturer's instructions.

The labeled probes were purified by filtering through a Nick Column (Pharmacia).

For dot-blot analysis, multiple concentrations of RNA were immobilized

on nylon membranes (Gene Screen Plus, Dupont, NEN) using a dot-blot apparatus (Bio-

Rad, Richmond, CA). The membranes were pre-wet in 20X SSC for 10 mins before the

diluted samples were applied. After applying the samples, the membranes were baked at

800C for 2 hours. The baked membranes were warmed in 400C water for 2 mins, and then

pre-hybridized for 30-60 mins at 650C while rotating in Hybaid Quikhyb solution. The

labeled probe was added in a concentration of 1.5 X 10 6 cpm/ml of the hybridization

solution. The membranes in hybridization solution were hybridized for 2 hours at 65 OC.

After hybridization, the membranes were washed in 2X SSC/0.1% SDS at 50 OC for 15

mins with two changes of solution. The membranes were further washed in 0. 1X

SSC/0.1% SDS for 15 mins. The blots were removed from the hybridization bottles,

wrapped in saran wrap and exposed to a phosphor imaging screen for 24-48 h. Care was









taken to minimize folds in the saran wrap. The latent image on the phosphor imager

screen was scanned using a Phosphor Imager (Molecular Dynamic, Sunnyvale, CA) and

analyzed by Image Quant Software (Molecular Dynamics). Intensities were calculated

per tg total RNA for each animal. Control as well as treated animal samples were applied

on the same blot to minimize variability. All samples from one experiment were run on

the same blot.



Statistical Analysis

BW and food intake datawere compared with one way ANOVA, followed by

post hoc analysis using Neuman-Keuls test. The p value was set at p<0.05 to attain

significance. Longitudinal leptin and insulin levels were compared using one way

ANOVA as well. Norepinephrine, glucose, T3 and T4 levels were compared with

Students 't' test as was BAT UCP-1 mRNA expression and skeletal muscle UCP-3

mRNA expression in treated vs. control groups. p<0.05 was considered significant in all

analysis.




Results

Body Weight

There were significant increases in the BW of female SD rats treated with rAAV-

UF5 (control) and untreated un-operated control rats over the 6 month period of the

experiment (Fig. 5-1). The rAAV-UF5 treated group gained 25.6% BW at 24 weeks post

injection, this was identical to the BW increase (25.8%) of untreated un-operated

controls. On the other hand, there was no significant BW gain the in the rAAV-leptin









treated rats which maintained their pre-injection weight (<3% gain overall) over the six

month period. At the end of the experiment, control females weighed 315.4 + 7.2 g,

untreated females weighed 310.4 +10.5 and rAAV-leptin treated rats weighed 256.3 + 8.0

g this represented a difference of 22 % in leptin treated vs. control groups (average

initial BW of the three groups was 249.3+ 2.3 g ).

Similarly, there was a significant difference in BW at 24 weeks post injection in

rAAV-UF5 vs. rAAV-leptin treated male rats (controls, 433.6 + 15.4 g vs. 392.5 + 14.7 g

with rAAV-leptin treatment, Fig. 5-2). The initial BW of the male rats in this experiment

was 325.6 +11.2 g controls vs. 324.8 + 7.1 g rAAV-leptin. The control animals gained

32% vs. their initial BW as opposed to rAAV-leptin treated rats that gained 12.1% BW

during the six months post injection.

Food Intake

There were no significant differences in 24 hour food intake in the three groups

in female rats, 20.8 + 0.4 g in rAAV-UF5, 20.5 + 0.4 g in untreated, and 19.6 + 0.4 g in

rAAV-UF5 treated rats (Fig. 5-3). Similarly, amount of food consumed by male

Sprague-Dawley rats was not different between treatment groups. Food intake by both

rAAV-UF5 control male rats and rAAV-leptin treated male rats averaged 24.0 + 0.6 g

(Fig. 5-4).

Leptin levels

Leptin levels were determined at three time points over the 24 week course of the

study (Fig. 5-5). Age related increases were observed in both the untreated female

controls (1.3 + 0.1 wk 6 vs 2.6 0.5 wk 16 vs. 9.1 + 0.5 wk 24 untreated) as well as in

rAAV-UF5 treated females (1.9 0.1 wk 6 vs. 4.2 + 1.0 wk 16 vs. 8.6 1.1 wk 24).









The blood leptin levels of rAAV-leptin injected rats did not increase with age (0.9 + 0.1

wk 6 vs. 1.43 0.7 wk 16 vs. 1.79 0.8 wk 24) and were significantly attenuated vs.

controls at 6, 16 and 24 weeks post injection (p<0.05).

Similarly, in male rats treated with rAAV-leptin the age related increase in

circulating leptin levels vs. rAAV-UF5 controls was not seen (P<0.05, Fig. 5-6). The

circulating leptin levels of male rats treated with rAAV-leptin were significantly

decreased vs. controls at 8, 16 and 24 weeks post injection.

Insulin levels

Circulating insulin levels in rAAV-leptin treated females remained unchanged at

week 6 and 16 post injection. As expected untreated as well as rAAV-UF5 treated

controls exhibited an age-related increase in serum insulin levels which was apparent at

24 weeks (p<0.05). The rAAV-leptin treated rats failed to show this increase in insulin

levels at week 24 post-injection (Table 5-1).

Similarly, control rAAV-UF5 treated male rats had significantly elevated

circulating insulin levels at week 24 compared with week 16 or 8 post injection (p<0.05).

In rAAV-leptin treated rats there was an increase in serum insulin levels at 24 weeks, vs.

week 16 or 8 but these levels at week 24 were significantly lower than those of rAAV-

UF5 treated rats at the same time point (Table 5-2). Thus rAAV-leptin treatment

suppresses the age-related increase in insulin in both males as well as females.

Glucose levels

Despite differences in insulin levels the rAAV-leptin treatment in females (Table

5-1) as well as males (Table 5-2) displayed normoglycemia at all time points examined.

There were no significant differences in glucose levels between treatment groups.









Urinary Norepinephrine levels

NE levels were measured in urine collected at week 16 post injection in male and

female rats. There was no significant difference in rAAV-UF-5 treated vs. rAAV-leptin

treated female rats (15.9 + 4 vs. 16.2 + 4.6 ng/ml Fig. 5 -7). Similarly there were no

changes in urinary NE in male rats treated with rAAV-leptin (Fig. 5- 8).

Thyroid Hormone levels

Neither T3 nor T4 levels measured at 16 weeks post injection were altered by

rAAV-leptin treatment in female ( Fig. 5-9 ) or male rats (Fig. 5-10).

UCP-1 mRNA expression

UCP-1 mRNA expression was significantly elevated (p<0.05) in rAAV-leptin

treated females vs. rAAV-UF5 treated controls. rAAV-leptin treatment resulted in a 2

fold increase in BAT UCP-1 mRNA expression in females (Fig. 5-11 ). Similarly, a

significant increase in UCP-1 mRNA expression was observed in male rats treated with

rAAV-leptin vs. rAAV-UF5 controls at 24 weeks post injection (Fig. 5-12).

UCP-3 mRNA expression

Although there was a tendency for skeletal muscle UCP-3 mRNA expression to

decrease in rAAV-leptin treated male and female rats, these differences were statistically

not significant.(Fig's 5-13 and 5-14).










-A- Untreated
-0- rAAV-UF5
-*- rAAV-leptin


I I I I I I I
0 4 8 12 16 20 24
Weeks post-injection


Fig. 5-1. Effect of long term leptin gene therapy on body weight in female Sprague-
Dawley rats. There was a significant decrease in body weight (g) with rAAV-leptin
treatment (p<0.05) vs. untreated and rAAV-UF5 treated groups. (n=6-8 per group).


-0- rAAV-UF5
-*- rAAV-leptin


I I I I I I I
0 4 8 12 16 20 24
Weeks post-injection



Fig. 5-2. Effect of long term leptin gene therapy on body weight in male Sprague-
Dawley rats. There was a significant decrease in body weight (g) with rAAV-leptin
treatment (p<0.05) vs. rAAV-UF5 treated rats. (n = 6 per group).


325-

% 300-
-6-

275-

250-

0 225-
0
200-


450-



400-



350-



300-













FEMALE


i U
0)


< 25

-
C

S20


0

OC

10
10-


-A- Untreated
-0- rAAV-UF5
-*- rAAV-leptin


I I 1 1
12 16 20 24
post-injection


Fig. 5-3. Effect of rAAV-leptin on food intake in female Sprague-Dawley rats.
There was no difference in 24 hour food intake between the three treatment groups. n=6-
8 per group.




40 -0- rAAV-UF5
S40- MALE
-*- rAAV-leptin


30-

0
0


0
3 20


O
I


0 4 8

Weeks


12 16

post-injection


Fig. 5-4. Effect of rAAV-leptin on food intake in male Sprague-Dawley rats. There
was no difference in 24 hour food intake between the treatment groups. n=6-8 per group.


I I I
0 4 8
Weeks


20 24
20 24




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs