role of wnt signaling and glycogen synthase kinase-3 beta in adipocyte biology

Trang 1

BOSTON UNIVERSITY SCHOOL OF MEDICINE

Dissertation

ROLE OF WNT SIGNALING AND GLYCOGEN SYNTHASE KINASE-3 BETA IN ADIPOCYTE BIOLOGY

by

DAVID A SILVA

B.Sc., University of Massachusetts, Amherst, MA, 1994

Submitted in partial fulfillment of the

requirements for the degree of Doctor of Philosophy

Trang 2

UMI Number: 3232922

INFORMATION TO USERS

The quality of this reproduction is dependent upon the quality of the copy submitted Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper

alignment can adversely affect reproduction

In the unlikely event that the author did not send a complete manuscript

and there are missing pages, these will be noted Also, if unauthorized copyright material had to be removed, a note will indicate the deletion

® UMI UMI Microform 3232922

Copyright 2006 by ProQuest Information and Learning Company

All rights reserved This microform edition is protected against

unauthorized copying under Title 17, United States Code

Trang 4

ACKNOWLEDGMENTS

| would like to extend my thanks and appreciation to Dr Stephen Farmer for his considerable support and encouragement throughout my time as a

graduate student in his laboratory In particular, | would like to thank him for encouraging scientific curiosity and discussion as well as for understanding that at all times his students are individuals with many interests who now and again

have life events outside of the laboratory and for supporting us nonetheless

| | would also like to thank the numerous members of the Farmer laboratory whom | have had the opportunity to work with over the years Working with you has challenged my knowledge and understanding of science, taught me many laboratory techniques, and exposed me to other cultures and perspectives all of which enriched my personal experience and often helped to make graduate student life more enjoyable

It goes without saying that it takes the efforts of many to shepherd and

nurture a student through graduate school That being said, | sincerely

appreciate the efforts that the members of my thesis committee have exerted on my behalf Your advice both in formal committee settings and in less formal surroundings was always helpful and appreciated | would like to thank Dr Paul

Pilch for being my second reader and reviewing my thesis | would also like to

Trang 5

in committee meetings, for offering all of their helpful advice, and for always

being so accessible

There have been many post-docs, graduate students, and technicians

who have provided advice, reagents, or equipment over the years as well | would especially like to thank those in the Pilch, Symes, Xiao, Kandror, and Nugent laboratories for all of their help

In addition, | would like to thank my family for all of the support that they have provided | am grateful to my mother, Kathleen, who always encouraged me to pursue my dreams, and my father, Richard, who has always been a role model

for me Also, | would like to thank my brother and sisters, who have always been

there for me

Lastly and most importantly, | would like to thank my wife, Margaretta, without whom | may never have gone to graduate school and who supported me

throughout Maggie, your encouragement and sacrifice enabled me _ to accomplish this and for that | cannot thank you enough Not only have you enabled this chapter of my life, but you have also provided me with three

Trang 6

ROLE OF WNT SIGNALING AND GLYCOGEN SYNTHASE KINASE-3 BETA IN ADIPOCYTE BIOLOGY

(Order No ) DAVID A SILVA

Boston University School of Medicine, 2007

Major Professor: Stephen R Farmer, Ph.D., Professor of Biochemistry

ABSTRACT

Adipocyte differentiation from fibroblasts into fat cells (adipogenesis) is regulated by two families of transcription factors: CCAAT/enhancer binding proteins (C/EBPs), members of the basic leucine zipper family, and peroxisome proliferator-activated receptors (PPARs), which belong to the nuclear hormone receptor family Adipocytes contribute to energy homeostasis with their ability to

undergo lipogenesis and lipolysis, insulin-stimulated glucose transport, and

secretion of hormones such as leptin and adiponectin Alterations in adipocyte function play an important role in pathophysiological states, including diabetes

and cardiovascular disease Drosophila wingless (wg) and mouse mammary

Trang 7

abundance of f-catenin, which is then able to interact with the lymphoid enhancer factor/T cell factor (LEF/Tcf) family of transcription factors, increasing their transcriptional activity Although wnt/B-catenin signaling is involved in the differentiation of several cell types, a role for wnt or B-catenin signaling had not been established in adipocytes This research demonstrated that GSK-38 activity was required during the early stages of adipogenesis and that the expression of B-catenin decreased during differentiation It was also observed that ectopic expression of wnt-1 prevented adipogenesis Furthermore, the ability of wnt signaling to inhibit adipogenesis was at least partially dependent on wnt’s ability to suppress the expression and activity of PPARy Conversely, factors that regulated the induction of PPARy during adipogenesis negatively influenced B- catenin expression Lastly, it was observed that GSK-38 activity was required to maintain PPARy expression in mature adipocytes Further examination revealed that chronic inhibition of GSK-38 activity diminished insulin-sensitive glucose uptake Western blot analysis demonstrated a defect in insulin signaling due to

decreased insulin receptora expression Furthermore, it was observed that

decreased perilipin expression correlated with a diminished ability of these cells

to undergo hormone-sensitive lipolysis Taken together, these data demonstrated

an important role for wnt and GSK-38 signaling in regulating PPARy expression

Trang 8

TABLE OF CONTENTS

Page

Od t= (© | — i

Readers’ Approval Page .cccsces ese ecesee eee eee nea nee ease HH HH HH nh khe kh kh ii AcknowledgmerIs .QQQQQ HH HH KH nh nhà kh kh ili ADSIrACT EEE nà Ki nàn Ki Ki kh V II) oáto ao “ddiddd ee eee nee eee ee eet eee tensa eee nna ees vii List of Tabl©S - ng HS TH KĐT KĐT Ko nhi KT kh ki n kh kế x

Imo án - T86 a4 xi

List of AbbreviafiOnS ee teeter ee ner ee teeta eee nen tenses eter XIV INTRODUCTION 0Q Hà Hee

Energy homeosfaSis QQQQQQQnnn HH TH HH HH nà nh hy 1 Role of the adipocyte in disease c.QQnnnn nn nhe nhe 1

Adipose fiSSU@ LH HH HH HT TH HH kh kh ky 4 Preadipocyte cell culture models c 5 CCAAT/enhancer binding proteins 6

Peroxisome-proliferator-activated recepfors 22

Other transcription factors involved in adipogenesis 32

Whit signaling -.-.cQQQQnn HH n ke nn kh kh nh chư ren 35

A ce 45

Trang 9

AntibDOdies .ccccccc cee ccccccueeceeceesuavereeeseeeenvgeaeenererntnanvanantenenns 47 PlasMids 0.0.00 nn nh nh nh kh nh kh 48 Cell lines and cell culfUr@ . . - SH SH» kh kh ky 48 Production of stable cell lines using retroviral gene transfer 49

Preparation of total protein exfrac†S c cà cà 51

lImmunobilotting and SDS-PAGE analysis 52 Preparation of nuclear profeins cà nàn se 53 Gel-electrophorectic mobility shift analysis (GEMSAI 54 RNA isolafion .-.ccQQQQn TH nh KHE nhà kh kh kh nhe 54 Northem blot analySis -. - nọ SH HH nh, 55

Reverse transcriptase PCR (RT-PCR) à 57 Staining of neutral lipid droplets with Oil red O 58 H]-2-deoxyglucose uptakes - -.c cà Q22 2222211 nn ngày 58

Glycerol release aSSay QQQ TQ nn nnn He nHn nh nh kh ven 59 Stable transfections using TOPflash .- 59 Luciferase aSSaY QQQQQQQnn nnnn TH» HT nh nh Hàn kh kh xu 61 Calculation of error bafS -Q QQQnnn TH nh The nh rxy 61 CHAPTER 1: ROLE OF B-CATENIN AND GLYCOGEN SYNTHASE

KINASE 3 IN THE EARLY STAGES OF ADIPOGENESIS 62 CHAPTER 2: REGULATION OF ADIPOGENESIS BY ECTOPIC

EXPRESSION 0 0.0 EE EE OnE ee nent ne tne 93 CHAPTER 3: NEGATIVE REGULATION OF INSULIN-RESPONSIVE

GLUCOSE UPTAKE AND LIPOLYSIS BY CHRONIC ADMINISTRATION

Trang 11

LIST OF TABLES

CHAPTER 3

1, SB-216763 is a specific inhibitor of glycogen synthase kinase-3 (GSK-3)

and not other kinases such as protein

Trang 12

mitogen-LIST OF FIGURES

INTRODUCTION

1 Transcriptional cascade model of adipogenesis .cccccccccceseeeeeeteeeeees 15 2 Simplified diagram for the canonical wnt signaling pathway 39

CHAPTER 1

3 Expression oƒ B-catenin decreases during adipogenesis 69 4 B-catenin reporter activity is altered during differentiation of 3T3-L1 cells 71

5 Loss of nuclear B-catenin coincides with increased peroxisome

proliferator-activated receptor y (PPARy) Expression ccccccceececeereeeeennees 73

6 Expression of B-catenin mRNA is not altered during adipocyte

eÏ1—1/-05111-)119 0 ddÖd343ẢẢẢ 75

7 Expression of B-catenin is controlled at the level of its stability 77

8 Treatment of mature adipocytes with protease inhibitors increases the

nuclear localization Of B-ca†enin -‹cc các cv 79

9 Method of regulating glycogen synthase kinase-3 (GSK-3) activity using

0¬ 81

10 Pharmacological inhibition of glycogen synthase kinase-38 (GSK-3B)

prevents adipocyte differentiation 0 00.:0cccccssseeeseceeseeeeeeeesseseeesessseeeeeeeres 83

11 LiCl treatment prevents the expression of markers of adipocyte

lêlÌJÍ=14=14111-1419)X&caadaaaiddaaaẳiẳaảaiiiỎồồồẮẮ 85 12 LiCl treatment inhibits adipogenesis in a dose-dependent

mann6ï cọ và 87

Trang 13

14 Inhibition of glycogen synthase kinase-3B (GSK-38) with LiCl during the early stages of adipogenesis suppresses the expression of peroxisome proliferator- activated receptor y2 (PPARy2), adipocyte fatty acid binding protein (aP2), and perilÏDif TQ nọ TS TT TT nàn TT ng kh kg tin kh kh rã 91

CHAPTER 2

15 Schematic diagram of retrovial infecfios cài se 98

16 Ectopic expression of wnt-1 blocks differentiation of 3T3-L1

17 Ectopic expression of wnt-1 in 3T3-L1 cells prevents the expression of

adipogenic markers and induces the expression of b-catenin target genes 102 18 Wnt cells undergo growth arrest at the same rate as vector cells 104

19 Wnt signaling fails to alter the DNA binding activity of the early adipogenic

transcription factors CCAAT/enhancer binding proteins (C/EBPs) B and 6 106 20 Ectopic expression of peroxisome proliferator-activated receptory2

(PPARy2) in wnt-1-expressing 3T3-L1 cells and addition of an exogenous

PPARy ligand reverse the block in adipogenesis 108 21 Dexamethasone in combination with either insulin or 3-isobuty-1-methyl- xanthine is able to induce morphological differentiation in 3T3-L1

9I le|i9s 24: cee 110

22 Dexamethasone suppresses j-catenin expression during adipocyte

Cifferentiation 00 cece cece cece cee n nen nn nen nà ke Kế KT ng nh ty ke 112

CHAPTER 3

23 Induction of insulin sensitivity is a late event during adipogenesis 122 24 Treatment of mature adipocytes with LiCl is unable to reduce cytoplasmic

Ijlslse hố 6 ( .4đAđă:a ai c 124

25 Treatment of mature adipocytes with LiCl suppresses peroxisome

Trang 14

26 SB-216763 inhibits glycogen synthase kinase-38 (GSK-3) activity 130

27 SB-216763 suppresses the expression of peroxisome proliferator-activated

receptor y (PPARy) in mature adipocyfes co co nen 132

28 SB-216763 inhibits peroxisome proliferator-activated receptor y (PPARy) expression in a dose-dependent manner 134 29 SB-216763 inhibits insulin responsive glucose uptakes in mature adipocytes

c1 11kg g0 1k TT E001 1 K00 111111 k0 1411111155 tạ 136 30 SB-216763 fails to downregulate the expression of glucose transporter 4 (GLUT4) after chronic treatmenf .c.ccQQ Q0 0Q Q HH HH HH vu 138 31 Chronic treatment of 3T3-L1 cells with SB-216763 inhibits insulin

"s10 ` 140

32 Treatment of 3T3-L1 cells with SB-216763 diminishes the expression of

insulin recepfor œ (Í.F.() cc cọ ha 142 33 SB-216763 suppresses catecholamine-stimulated lipolysis 144

34 SB-216763 suppresses the expression of perilipin 146

DISCUSSION

35 Regulation of adipocyte differentiation by LiC| 158 36 Wnt signaling inhibits the expression of CCAAT/enhancer binding protein œ

(C/EBPơ) and peroxisome proliferator-activated receptor y (PPARy) 160

37 Wnt signaling affects eroxisome proliferator-activated receptor y

(PPARy) on multiple levelS HT TT ng nh nen g0 1111k kg 162 38 Attenuation of insulin signaling by glycogen synthase kinase-3

Trang 15

ACRP30 ADD Akt APC aP2 AP-2œ APC A.U BAT BCA BMI BRG1 BSA cAMP CBP CDK C/EBP C/EBPa C/EBPB C/EBPð cGMP LIST OF ABBREVIATIONS

adipocyte compliment related protein 30 Kda adipocyte determination and differentiation factor

protein kinase B

adenomatous polyposis coli

adipocyte fatty acid-binding protein activator protein2a

adenomatous polyposis coli

arbitrary unit

brown adipose tissue

bicinchoninic acid body mass index

Brahma-related gene 1

bovine serum albumin

cyclic adenosine monophosphate

CREB binding protein cyclin-dependent kinase

CCAAT/enhancer binding protein CCAAT/enhancer binding protein o

CCAAT/enhancer binding protein B (also known as NF-IL6)

CCAAT/enhancer binding protein 6

Trang 16

CHOP CK CREB cs CUP Dck DMEM DM! DMIT DMSO DP Dvl E2F ECL EDTA EGF ERK ES FAM FAS

C/EBP homologous protein casein kinase

cAMP-response element binding protein

calf serum

C/EBPa-undifferentiated protein

Dickopf

Dulbecco’s modified Eagle’s medium

dexamethasone, 1-methyl-3-isobutyl xanthine, and insulin dexamethasone, 1-methyl-3-isobuty! xanthine, insulin, and troglitazone

dimethylsulfoxide

E2F dimerization partner

dishevelled

early-region-2 transcription factor enhanced chemilluminescence ethylenediaminetetraacetic acid epidermal growth factor

extracellular signal-regulated kinase

embryonic stem

fat facets (faf) gene-associated molecule

Trang 17

FFA FGF GAG GAPDH GEMSA GLUT GSK HBP 1 HDAC HEK293 HEPES HNF HSL IGF Kda KH;PO¿ KRH LAP

free fatty acid

fibroblast growth factor glycosaminoglycan

glyceraldehyde-3-phosphate dehydrogenase gel-electrophoretic mobility shift analysis glucose transporter

glycogen synthase kinase

HMG box protein 1

histone deacetylase

human embryonic kidney 293 cells

4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid hepatocyte nuclear factor

hormone-sensitive lipase

insulin-like growth factor

insulin receptor

insulin-responsive aminopeptidase insulin receptor substrate

potassium chloride Kilo daltons

potassium phosphate

Krebs-Ringers HEPES buffer

Trang 18

LDR LEF/Tcf LiCI LIF LIP LRP MAP kinase MIX mom MSC mTOR NaCl Na;HPO¿ NCoR NIDDM Nkd pAkt PBS PBST PCNA PDGF

low-density lipoprotein receptor

lymphoid enhancer factor/T cell factor lithium chloride

leukemia inhibitory factor

liver inhibitory protein

low-density lipoprotein receptor-related protein mitogen-activated protein kinase

1-methyl-3-isobutyl-xanthine

more mesoderm

mesenchymal stem cell

mammalian target of rapamycin sodium chloride

sodium phosphate

nuclear receptor corepressor

non-insulin-dependent diabetes mellitus

naked cuticle

phosphorylated-Akt

phosphate-buffered saline PBS with Tween 20

Trang 19

PGC-1 PI PI-3K PKA PKB PKC PKR PMSF polybrene PPAR PP2A PPRE PVDF RAR Rb RDM RLU RT-PCR RXR SC SDS-PAGE sFRP PPAR y coactivator-1 phosphoinositide phosphoinositide-3-kinase

(cAMP-dependent) protein kinase A

protein kinase B protein kinase C

dsRNA-activated serine/threonine protein kinase phenyl methylsulfonyfluoride

1,5-dimethyl-1,5-diazaundecamethylene polymethobromide, hexadimethrine bromide peroxisome proliferator-activated receptor protein phosphatase 2A

PPAR response element polyvinylidene fluoride retinoic acid receptor retinoblastoma protein repressor domain

relative luciferase unit

reverse transcriptase polymerase chain reaction retinoid X receptor

synergy control motif

sodium dodecyl sulfate-polyacrylamide gel electrophoresis secreted frizzled-related protein

Trang 20

SMRT Sp SREBP SSC STAT SWI/SNF TAD TLS-CHOP TrCP TNFa TZD UV WAT WIF wnt

silencing mediator of retinoid and thyroid hormone receptors

stimulatory protein

sterol-response element binding protein

sodium chloride-sodium citrate buffer

signal transducer and activator of transcription

switch/sucrose nonfermenting transcriptional activator domain

translocation liposarcoma-C/EBP homologous protein

transducin repeat-containing protein

tumor necrosis factor a thiazolidenedione ultraviolet

white adipose tissue wnt inhibitory factor

wingless (wg) and mouse mammary tumor virus integration

Trang 21

INTRODUCTION Energy homeostasis

The need to regulate the balance between energy intake and expenditure (“energy homeostasis”) is a trait common to all organisms (Jang and Sheen

1994; Teleman, Maitra et al 2006; Flier and Maratos-Flier 2000) and is a crucial factor in maintaining proper body weight (Flier and Maratos-Flier 2000) Energy

homeostasis is controlled by three related processes: caloric intake, caloric expenditure, and metabolic activity (Flier and Maratos-Flier 2000)

Failure to properly regulate energy balance leads to obesity (Spiegelman and Flier 2001; Webber 2003) and has been associated with pathophysiological conditions such as non-insulin-dependent diabetes mellitus (Herberg 1991; Bray

1992) and cardiovascular disease (Poirier Giles et al 2006; Rashid, Fuentes et

al 2003) In order to maintain proper body weight, metabolic parameters, and physiology, higher organisms have developed complex mechanisms to regulate

energy homeostasis, including the ability to suppress or stimulate appetite, to store or mobilize energy reserves, and to increase or decrease energy utilization Role of the adipocyte in disease

According to data from the United States Centers for Disease Control and

Prevention, the prevalence of obesity in the United States is on the rise (Ogden,

Fryar et al 2004) Although the association between obesity and many diseases

Trang 22

disease or a result of an unappreciated pathophysiological state? If obesity itself

contributes directly to the onset of disease, how does it do it? Is it through the

aberrant secretion of hormones (Steppan, Bailey et al 2001) and cytokines

(Hotamisligil, Shargill et al 1993), or is it through alterations in the adipocyte’s ability to produce and secrete lipids (Jensen, Caruso et al 1989; Sztalryd and

Kraemer 1995; Large and Arner 1998)? Although many of these possibilities are

the subject of intense speculation and research, the exact contribution of the adipocyte remains elusive What has become clear is that a large number of patients are able to improve their clinical outcomes by modifying their diets (Tuomilehto, Lindstrom et al 2001) or increasing their activity (Santeusanio, Di Loreto et al 2003; Sigal, Kenny et al 2004), both of which result in a decrease in

adipose mass

Epidemiologically, obesity has been identified as a risk factor for a number

of pathophysiologies, including hepatic steatosis (Scheen and Luyckx 2002: Clain and Lefkowitch 1987; Diehl 2005), insulin resistance (Ludvik, Nolan et al

1995), type-ll diabetes mellitus (Herberg 1991; Bray 1992), hypertension

(Lamounier-Zepter, Bornstein et al 2004), coronary artery disease (Brochu and Poehlman 2000), cancer (Calle and Thun 2004; Kuriyama, Tsubono et al 2005), impairment of the immune system (Marti, Marcos et al 2001), and osteoarthritis

Trang 23

With the prevalence of obesity throughout our society (Ogden, Fryar et al 2004) and its role in a number of pathological conditions, it is imperative to find effective treatments Current treatment methods for obesity include surgery

(O'Brien and Brown 2005), lifestyle modifications (Tuomilehto, Lindstrom et al 2001), and pharmacological intervention (Clapham 2004; Halford 2001; Jandacek and Woods 2004) Examples of pharmacologic therapeutics include sibutramine (Ryan and Kaiser 1995), orlistat (Drent and Larsson et al.1995), and phentermine, which block serotonin and dopamine reuptake, intestinal fat absorption, or appetite, respectively (Ryan and Kaiser 1995; Drent, Larsson et al 1995; Jandacek and Woods 2004) In addition to drugs that reduce obesity, there

are also therapeutics for some of obesity’s consequences, such as the thiazolidenediones and metformin, which improve insulin sensitivity and

hyperglycemia (Hofmann, Lorenz et al 1991; Antonucci, Whitcomb et al 1997: Guthrie 1997); sulfonylureas, which increase insulin secretion from the B-cells in the pancreas (Kolterman, Prince 1983); and statins and fibrates, which reduce serum cholesterol levels (Lo, Noviasky et al 2005; Shepherd 2005; Robins

2003)

Given the complications associated with many of the treatments for obesity, such as thiazolidenedione-induced non-alcoholic fatty liver disease (Forman, Simmons 2000; Gitlin, Julie 1998), it is becoming more important to find

novel and more effective ways of treating obesity and its consequences Undoubtedly, a deeper understanding of factors positively or negatively

Trang 24

regulating fat-cell development will yield new therapeutic targets for treating obesity and treatments with fewer complications than the treatments currently

available Likewise, further research into factors regulating adipocyte function could help to explain and treat the aberrant pathophysiological behavior of

adipocytes in diseases such as diabetes and cardiovascular disease

Adipose tissue

As in other mammals, in humans white adipose tissue (WAT) begins to develop prenatally and is linked with the development of the vasculature (Poissonnet et al 1983) Post-natal studies on adult rats fed high-carbohydrate

diets suggest that increases in adipose mass are the result of both hypertrophy and hyperplasia of adipocytes (Faust, Johnson et al 1978)

Hypertrophy plays a role in the onset of obesity, and it most likely contributes to mild obesity The contribution of hypertrophy to morbid obesity is probably limited by physical constraints associated with the cell’s ability to increase its volume relative to its surface area The ability of preadipose cells to undergo hyperplasia and differentiate into mature adipocytes is a major factor in

the onset of morbid obesity (Hirsch and Batchelor 1976; Johnson, Stern 1978) A greater understanding of factors that regulate adipocyte differentiation is

Trang 25

Preadipocyte cell culture models

Because of the complexity of in vivo systems, most of our understanding

of adipogenesis is derived primarily from in vitro cell culture systems The most widely used preadipocyte cell lines are the Swiss 3T3-L1 and 3T3-F442As These cell lines were originally developed by Howard Green and colleagues

(Green and Kehinde, 1976; Green and Meuth, 1974) by disaggregating

seventeen-day-old mouse embryos and clonally isolating cells with a predisposition to accumulate cytoplasmic lipid droplets

Initially, these cells took several weeks to spontaneously differentiate in culture Further research then demonstrated that insulin (Green and Kehinde, 1975), 1-methyl-3-isobutyl-xanthine (MIX), Russell and Ho 1976), or a

combination of MIX and dexamethasone (Rubin, Hirsch 1978) could induce more robust and rapid differentiation Eventually, it was determined that the combination of insulin, dexamethasone, and MIX was optimal for adipocyte differentiation (Coleman and Bell 1980; Student, Hsu et al.1980) This protocol

has now been adopted as the standard method for differentiating 3T3-L1

preadipocytes (Figure 1)

In vitro differentiated adipocytes possess nearly indistinguishable ultrastructural characteristics compared with in vivo adipocytes (Novikoff,

Novikoff et al 1980) Injection of nude mice with preadipocyte cell lines has been demonstrated to lead to the formation of fat pads that are nearly indistinguishable from endogenous fat pads (Green and Kehinde, 1979; Vannier, Gaillard et al

Trang 26

1985) Cell culture models also effectively mimic the induction of many

adipogenic processes such as lipogenesis (Mackall, Student et al 1976) and glucose transport (Rosen, Smith et al 1978) as well as the expression of the

adipogenic markers adipocyte fatty acid-binding protein (aP2) (Bernlohr et a 1984), perilipin (Greenberg et al 1993), leptin (MacDougald et al 1995), and

glucose transporter 4 (GLUT4) (De Herreros A, Birnbaum 1989; Kaestner,

Christy et al 1989)

Our knowledge of the signaling pathways, transcriptional networks, and

cytoskeletal-extracellular matrix factors involved in adipocyte differentiation has been greatly enhanced by studying in vivo models of adipocyte differentiation To date, many of the factors have been identified that regulate adipocyte differentiation, including the CCAAT/enhancer binding protein and peroxisome proliferator-activated receptor (PPAR) families of transcription factors, signaling pathways such insulin/insulin-like growth factor (IGF), and extracellular matrix factors such as integrins and fibronectin

CCAAT/enhancer binding proteins

CCAAT/enhancer binding proteins (C/EBPs) were originally identified in

rat liver as proteins that could bind to viral enhancer DNA (Johnson, Landschulz et al 1987; Ryden and Beemon 1989) C/EBPs are members of the leucine

Trang 27

6, , and y (Davydov, Bohmann et al 1995) Three members of the C/EBP family are expressed during adipogenesis (Cao, Umek et al 1991): C/EBP a, B, and 6

The C/EBPs, particularly C/EBPa and £, express multiple isoforms (Descombes and Schibler 1991) that are thought to play differing roles in

regulating C/EBP target genes (Lin, MacDougald et al 1993; Ossipow, Descombes et al 1993) It was originally postulated that the generation of the

multiple C/EBP isoforms was the result of leaky ribosomes scanning past the first start codon in their mRNA (Ossipow, Descombes et al 1993) Subsequent

evidence was presented suggesting that these isoforms were the result of

proteolytic cleavage (Welm, Timchenko et al 1999) More recent evidence suggests that the alternate translation of C/EBP a and 8 is controlled at least in part by the activity of dsRNA-activated serine/threonine protein kinase (PKR) and mammalian target of rapamycin (mTOR) and their ability to regulate the activity of the translation factors elF-2a and elF4E (Calkhoven, Muller et al 2000) This

regulatory event is important for adipocyte differentiation (Calkhoven, Muller et al

2000) Alternate translation products for C /EBPa or B either contain or lack transcriptional activator domains (TADs), which are necessary for transactivating

the promoters of C/EBP target genes In effect, the generation of TAD-less C/EBPs creates C/EBP isoforms that function as endogenous inhibitors of C/EBP function

C/EBP activity is regulated by post-translational modifications such as

sumoylation and phosphorylation A sumoylation site lies within a conserved

Trang 28

repressor domain (RDM) common among ail of the C/EBP family members (Kim, Cantwell et al 2002), which was separately identified as the synergy control motif (SC) in C/EBPa (Subramanian, Benson et al 2003) Furthermore, mutation of the sumoylation site lysine 173 in full-length C/EBP§ results in derepression of

C/EBPB on the cyclin D1 promoter (Eaton and Sealy 2003) Taken together, these data suggest a role for sumoylation in negatively regulating the

transcriptional activity of C/EBP family members

C/EBPB along with C/EBP8, is expressed early during adipocyte

differentiation (Cao, Umek et al 1991) C/EBPB, also known as NF-IL6, is induced during in vitro adipocyte differentiation by 1-methyl-3-isobutyl-xanthine

(Yeh, Cao et al 1995) and reaches maximum expression during the first twenty-

four hours of differentiation (Cao, Umek et al 1991) C/EBPB induction appears

to be mediated at least in part by the activity of CREB (cAMP-response element

binding protein) (Zhang, Klemm et al 2004) Recent studies have identified a CREB binding site in the murine C/EBPB promoter that is required for full promoter activity in reporter assays (Zhang, Klemm et al 2004) In addition, CREB activity has been reported to be essential for adipogenesis (Reusch, Colton et al 2000) Additional evidence suggests a role for the transcription

Trang 29

adipogenesis in vitro (Cao, Umek et al 1991) C/EBPB was subsequently demonstrated to have the ability to induce adipogenesis in non-precursor

fibroblasts (Yeh, Cao et al 1995; Wu, Zie et al 1995) and to relieve the need for

MIX as an inducer for adipogenesis (Yeh, Cao et al 1995) Notably, ectopic expression of C/EBPB was demonstrated to play a role in inducing the expression of the transcription factor PPARy (Wu, Zie et al 1995), which has

been demonstrated to play a central role in regulating the expression of adipogenic genes such as aP2 PPARy expression in many non-precursor fibroblast cell lines is dependent on both the expression of C/EBPB and the presence of an exogenous ligand for the glucocorticoid receptor (Wu, Zie et al 1995; Wu, Bucher et al 1996) Conditions that do not favor the expression of C/EBPB8 during adipogenesis (Yeh, Cao et al 1995; Hamm, Park et al 2001) or that inhibit the function of endogenous C/EBP£$ (Hamm, Park et al 2001) prevent

adipogenesis and the expression of PPARy2

C/EBP8 subcellular nuclear localization is thought to be regulated by

mitogen-activated protein kinase (MAP kinase) (Piwien Pilipuk, Galigniana et al

2003) and subsequently glycogen synthase kinase-3 (GSK-3)-mediated

phosphorylation (Tang, Gronborg et al 2005) Mutation of the MAP kinase and

GSK-3 phosphorylation sites has been demonstrated to inhibit C/EBP®B DNA binding activity (Tang, Gronborg et al 2005) as well as C/EBP8 ability to induce

Trang 30

GSK-3 leads to the dephosphorylation of C/EBPB and decreased DNA binding activity for liver activating protein (LAP) but increased DNA binding activity for the

inhibitory form of C/EBP§, liver inhibitory protein (LIP) (Piwien Pilipuk, Van Mater

et al 2001)

C/EBP8 DNA binding activity is also thought to be positively regulated by other phosphorylation events, including those mediated by protein kinase A (PKA) or C (PKC) on serine 240 (Trautwein, van der Geer et al 1994) Mutation

of serine 240 to alanine results in diminished DNA binding by C/EBP8 (Trautwein, van der Geer et al 1994) PKA signaling also regulates C/EBPB by

controlling its ability to translocate into the nucleus This observation is supported

by mutation of the PKA phosphorylation site serine 299 to alanine, which

prevents C/EBPB from entering the nucleus (Chinery, Brockman et al 1997)

There are also data suggesting a critical role for p38 MAP kinase signaling

in regulating C/EBPB function Inhibition of p38 MAP kinase prevents phosphorylation of C/EBPB and blocks C/EBP ’s ability to induce adipogenesis and PPARy2 expression (Engelman, Lisanti et al 1998)

Trang 31

ai 2003), and expression of dominant negative C/EBPB has been demonstrated to inhibit clonal expansion and terminal adipogenesis (Zhang, Tang et al 2004)

C/EBPB activity is regulated by its interaction with transcriptional co- repressors Recent studies have demonstrated that the transcription co-repressor

ETO/MTG8 is expressed early during adipogenesis (Rochford, Semple et al

2004) and ectopic expression of wild-type ETO/MTG8 leads to the suppression

of adipogenesis (Rochford, Semple et al 2004) Regulation of adipogenesis by

ETO/MTG8 appears to be mediated by its interaction with C/EBPB and

ETO/MTG8’s ability to inhibit C/EBPB DNA binding activity (Rochford, Semple et

al 2004)

C/EBP§ transcriptional activity is thought to be mediated in part by its

interaction with transcriptional coactivators such as CREB binding protein

(CBP)/p300 (Mink, Haenig et al 1997) and chromatin remodeling factors

(Kowenz-Leutz and Leutz 1999; Pedersen, Kowenz-Leutz et al 2001) The

interaction between C/EBPB and CBP/p300 has been demonstrated to be

negatively regulated by insulin/protein kinase B (Akt) signaling and results in suppression of C/EBP§’s ability to transactivate C/EBP reporter genes (Guo,

Cichy et al 2001)

Among the more important C/EBPB target genes in adipocytes are

C/EBPa (Yeh, Cao et al 1995; Tang 2004), PPARy2 (Wu, Xie et al 1995; Tang 2004), and adiponectin (Park, Qiang et al 2004; Kita, Yamasaki et al 2005) Other potential C/EBPB target genes include acyl-CoA carboxylase (Tae, Zhang

Trang 32

et al 1995) and phosphoenolpyruvate carboxy kinase (PEPCK) (Eubank, Williams et al 2001)

Ectopic expression of C/EBPB in non-precursor fibroblasts has been

demonstrated to induce adipocyte differentiation (Yeh, Cao et al 1995) as well as PPARy expression (Wu, Xie et al 1995), whereas expression of the dominant negative isoform of C/EBP® inhibits PPARy2 and C/EBPa expression as well as adipocyte differentiation (Hamm, Park et al 2001) However, a direct role for

C/EBPPB in regulating the PPARy2 promoter has yet to be demonstrated (Clarke,

Robinson et al 1997; Elberg, Gimble et al 2000)

Similar to C/EBPB, C/EBPS is expressed early during differentiation (Yeh,

Cao et al 1995) In preadipocytes, C/EBP6 expression has been demonstrated

to be induced by the addition of dexamethasone (Yeh, Cao et al 1995), suggesting a role for glucocorticoid receptor signaling in the induction of C/EBP%S More recent work also suggests a role for leukemia inhibitory factor (LIF) in inducing C/EBPs expression by increasing the DNA binding activity of signal

transducers and activators of transcription (STATs) 1 and 3 to regulatory

elements in the C/EBP6 promoter (Hogan and Stephens 2005) The exact role

Trang 33

It is also possible that C/EBPS can compensate to a degree for C/EBP8 function during the early stages of adipogenesis, since mice that are homozygous null for C/EBPB can still undergo morphologic differentiation and express adipogenic genes (Tanaka, Yoshida et al 1997) Unlike C/EBPa and B, C/EBP6 does not

appear to undergo alternate translation Loss-of-function studies provide

compelling evidence that C/EBPs play an important role in adipocyte differentiation Mice that are homozygous null for both C/EBPB and6 have a diminished ability to differentiate both in vitro and in vivo, even though in vivo

adipocytes still express PPARy2 and C/EBPa mRNA (Tanaka, Yoshida et al 1997) This suggests a more complicated relationship between C/EBPB and 6 and PPARy2 and C/EBP«a than one in which C/EBPB and 6 are solely expressed

so as to induce PPARy2 and C/EBPa mRNA expression Further evidence for the role of C/EBP £8 and 6 in adipocyte differentiation is provided by the

observation that preadipocytes, which overexpress the C/EBP inhibitory protein, translocation liposarcoma-C/EBP homologous protein (TLS-CHOP), fail to

undergo adipocyte differentiation (Adelmant, Gilbert et al 1998) This effect appears to be mediated at least in part to C/EBP homologous protein’s (CHOP’s) ability to inhibit C/EBPB DNA binding activity (Adelmant, Gilbert et al 1998)

Trang 34

that are homozygous for C/EBPB expression from the C/EBPa locus fail to develop normal levels of epididymal fat or to express leptin or adipsin (Chen, Chen et al 2000) Curiously, they still appear to express normal levels of PPARy,

GLUT4, and aP2 mRNA and also have metabolically identical profiles as their littermates (Chen, Chen et al 2000) It is interesting that C/EBPB appears to compensate for C/EBPa function in liver (Chen, Chen et al 2000), suggesting that this phenotype is an adipose tissue-specific effect and that the ability of

C/EBP family members to compensate for each other is dependent on the

inherent characteristics of the tissue in question This may be due to differences in C/EBP a and §'s ability to interact with transcriptional co-regulators or to

recognize cis regulatory elements in the promoters of target genes

C/EBPa was the first member of the C/EBP family cloned (Landschulz, Johnson et al 1988) and was shown to be expressed in a number of tissues,

including liver, fat, intestine, lung, adrenal gland, and placenta (Birkenmeier,

Gwynn et al 1989; Antonson and Xanthopoulos 1995) In liver, adipose tissue, and granulocytes, the expression of C/EBPca is restricted to growth-arrested, fully

Trang 35

Figure 1 Transcriptional cascade model of adipogenesis Schematic diagram demonstrating the key contributions of the various adipogenic inducers: the CCAAT/enhancer binding protein (C/EBP) family members and peroxisome proliferator-activated receptor y (PPARy) to adipogenesis The top-bottom

schematic indicates the relative temporal expression pattern of adipogenic transcription factors, the inducers that regulate their expression, and the subsets of the adipogenic gene program that they regulate aP2 = adipocyte fatty acid-

Trang 36

Dexamethasone, Insulin, MIX and FBS

|| Glucocorticoid Receptor CIEBPB PPARy C/EBPa

Lipogenesis Insulin Sensitivity

Lipolysis GLUT4

Perilipin Insulin Receptor aP2 I.R.S.-†

Trang 37

(Landschulz, Johnson et al 1988; Birkenmeier, Gwynn et al 1989; Cao, Umek et al 1991; Mischoulon, Rana et al 1992; Flodby, Antonson et al 1993; Radomska,

Huettner et al 1998) Temporally, C/EBPa’s expressed after peak expression of C/EBPB and 8 during adipocyte differentiation (Cao, Umek et al 1991) but prior

to growth arrest (Timchenko, Wilde et al 1996) and the expression of terminal

adipogenic markers such as GLUT4 (Wu, Bucher et al 1996) and adipsin (Wu, Bucher et al 1996) These observations suggest a role for C/EBPa in inducing terminal adipocyte differentiation

C/EBPa plays a critical role in arresting proliferation and inducing

terminal differentiation in a number of tissues, such as the liver and adipocytes

(Tao and Umek 2000) In adipocytes, growth arrest is thought to be a

requirement for differentiation (Shao and Lazar 1997), although more recent data using ectopically-expressing E7 to prevent growth arrest in adipocytes cast doubt on the suggestion that C/EBPa-mediated growth arrest is a requirement for differentiation (Muller, Alunni-Fabbroni et al 1999) Despite the controversy

regarding the role of growth arrest in adipocyte differentiation, there is abundant

evidence that C/EBPa can regulate both processes For example, ectopic expression of C/EBPa in adipoblasts induces growth arrest and terminal

differentiation (Umek, Friedman et al 1991; Lin and Lane 1994), whereas loss of C/EBPa expression in 3T3-L1 cells using an antisense C/EBPa transgene

Trang 38

In addition, C/EBPa has the ability to interact with the retinoblastoma

protein (Rb) (Puigserver, Ribot et al 1998; Wang and Timchenko 2005) and,

depending on Rb’s phosphorylation state, to inhibit Rb activity and increase proliferation (Wang and Timchenko 2005) C/EBPa also has the ability to interfere with early-region-2 transcription factor (E2F) transcriptional activity (Slomiany, D'Arigo et al 2000) and to suppress the expression of E2F-target genes C/EBPa has been demonstrated to disrupt E2F-p107 and to induce E2F- p130 complexes as well (Timchenko, Wilde et al 1999) This activity is

dependent on a direct interaction of C/EBPa with p107-E2F and also on C/EBPa’s ability to induce p21 Curiously, Rb is required for adipogenesis and

has been demonstrated to interact directly with other C/EBP family members (Chen, Riley et al 1996)

There is also evidence that C/EBPa may play a role in inducing the cyclin dependent kinase (CDK) inhibitor p21 (Timchenko, Wilde et al 1996) and also in interacting with p21, CDK2, and CDK4 to directly inhibit CDK activity

(Harris, Albrecht et al 2001; Wang, lakova et al 2001) This allows CEBPa the

ability to regulate Rb by regulating CDK activity

Trang 39

expressed with PPARy2, could induce adipogenesis in myoblasts (Hu, Tontonoz

et al 1995)

Antisense knockdown of C/EBPa in_ preadipocytes prevents

adipogenesis and provides compelling evidence that C/EBPa is required for adipocyte differentiation (Samuelsson, Stromberg et al 1991; Lin and Lane 1992) Additional evidence was provided by the observation that cytokines such

as tumor necrosis factor « (TNFa) that have been demonstrated to reverse

adipogenesis do so in part by suppressing the expression of C/EBPa (Stephens,

Pekala 1991; Ron, Brasier et al 1992)

The critical role for C/EBPa in regulating terminal adipogenesis is further

supported by observations in transgenic mice that are homozygous null for C/EBPa These mice fail to develop adipose tissue in vivo and die shortly after birth from defects in hepatic glycogen synthesis (Wang, Feingold et al 1995)

Rescue of C/EBPa homozygous null mice by expressing a C/EBPa transgene

using the albumin promoter improves the survival of these mice and demonstrates the adipose-depot specific requirement for C/EBP« in vivo (Linhart, Ishimura-Oka et al 2001) These mice demonstrate diminished WAT mass in the subcutaneous and epididymal depots without any obvious effects on brown adipose tissue (BAT) or mammary adipose tissue development (Linhart,

Trang 40

however, failed to address the issue of whether C/EBPa was required to maintain

terminal adipogenesis or solely to induce the development of adipose tissue This issue was addressed by the generation of a conditional knockout mouse whereby C/EBPa expression could be suppressed postnatally (Yang, Croniger et al 2005) Postnatal inhibition of C/EBPa expression led to hypophagia and

decreased WAT mass without apparent alterations in BAT (Yang, Croniger et al

2005)

C/EBPa activity is critical for the expression of a number of adipocyte-

enriched genes, including the B3-adrenergic receptor (Dixon, Daniel et al 2001),

acyl-CoA carboxylase (Tae, Luo et al 1994), leptin (He, Chen et al 1995; MacDougald, Hwang et al 1995), GLUT4 (Kaestner, Christy et al 1990), aP2 (Christy, Yang et al 1989; Kaestner, Christy et al 1990), lipoprotein lipase

(Previato, Parrott et al 1991), stearoyl-CoA desaturase (Christy, Yang et al

1989), and PPARy2 (Elberg, Gimble et al 2000) These observations provide compelling evidence that C/EBPa plays an important role in regulating a number

of physiological processes in adipocytes

C/EBPa activity is thought to be regulated by numerous post-

translational events, including PKC phosphorylation (Mahoney, Shuman et al 1992), which appears to negatively regulate C/EBPa DNA binding activity