DISS. ETH Nr.

19107

Molecular mechanisms of adipogenesis in obesity and the metabolic syndrome

A dissertation submitted to the ETH Zurich for the degree of Doctor of Sciences Dr. sc. ETH Zrich presented by

Bettina Meissburger
Diplom-Biologin (t.o.), Universitt Stuttgart

born May 19th 1981 Citizen of Germany

Accepted on recommendation of Prof. Dr. Christian Wolfrum Prof. Dr. Markus Stoffel Prof. Dr. Christoph Meier

2010

Table of Contents
Abstract Zusammenfassung Chapter 1 General introduction 3 5 7

Chapter 2 Retinoid-related orphan receptor gamma controls adipocyte hyperplasia and insulin sensitivity in obesity Chapter 3 Tissue inhibitor of metalloproteinase 1 (TIMP1) inhibits adipocyte differentiation in obesity Chapter 4 Regulation of adipogenesis by paracrine factors from adipose stromal-vascular cells- a link to fat depot-specific differences Chapter 5 Conclusions and Outlook

33

66

91

121

Appendix Chapter 2 Appendix Chapter 4 Acknowledgements Curriculum vitae

125 132 142 143

Abstract

Abstract
Adipose tissue is a key player in whole body metabolism and excess adipose tissue poses a major risk factor for the development of metabolic disorders such as type 2 diabetes. In response to nutritional overload, de novo adipocyte differentiation can serve as an adaptive mechanism by increasing the storage capacity of adipose tissue and maintaining normal adipocyte function. This in turn prevents systemic lipid overload, which is a major cause for insulin resistance. This PhD thesis provides novel insights into the regulation of adipogenesis and its impact on glucose and lipid metabolism. Chapter 2 describes the transcription factor retinoid-related orphan receptor (ROR) as an important inhibitor of adipocyte differentiation through its direct target gene matrix-metalloproteinase 3 (MMP3). Evaluating ROR expression in human stromal-vascular fraction (SVF), which contains adipocyte precursor cells, shows that ROR is increased in obese compared to lean patients. Decreased expression of ROR in murine and human obese subjects is correlated with increased adipogenesis, smaller adipocytes and improved insulin sensitivity. Through its influence on adaptive adipose tissue plasticity ROR is an important regulator of lipid and glucose homeostasis in the state of obesity. Further studies will need to elucidate how MMP3 remodels the extracellular matrix of adipose tissue thereby regulating adipogenesis. As described above, extracellular matrix reorganization has an important impact on adipocyte differentiation. Preadipocytes need to remodel their cell shape during adipogenesis and the matrix scaffold serves as an anchor for growth factors, thus regulating their availability. The process of matrix degradation is controlled by matrixmetalloproteinases (MMPs) and the tissue inhibitors of MMPs (TIMPs). Chapter 3 investigates the regulation of TIMP1 in obesity and its impact on adipocyte differentiation. TIMP1 is upregulated in obese mouse models and inhibits adipocyte formation. TIMP1 injections in mice in combination with a high fat diet challenge lead to increased adipocyte size, impaired glycemic control and increased systemic fatty acid overload compared to PBS injected control mice. Thus, TIMP1 promotes insulin resistance in the state of obesity. Future experiments will need to address how TIMP1 3

Abstract regulates the activity of MMPs and thereby influences the extracellular matrix in adipose tissue. In addition to matrix regulation, body fat distribution plays a crucial role in the risk assessment of obesity-related metabolic complications. Reasons for this are functional differences in intra-abdominal adipose tissue compared to subcutaneous fat. Chapter 4 explores depot-specific differences in the adipogenic potential, which might contribute to dysregulated adipocyte function in visceral adipose tissue. We show that adipogenesis of murine subcutaneous SVF is enhanced compared to visceral SVF, which is reflected in a shift towards smaller adipocytes in subcutaneous adipose tissue. Experiments using conditioned media reveal that paracrine secreted factors from the SVF inhibit adipocyte differentiation; this effect is more pronounced in visceral compared to subcutaneous SVF. Applying a mass spectrometry approach and mRNA expression analysis, a set of secreted proteins, which are upregulated in visceral compared to subcutaneous SVF, has been identified. Functional analysis of these proteins in adipocyte differentiation have led to the identification of 7 novel secreted factors with a negative impact on adipogenesis, the majority of these proteins being associated with the extracellular matrix. Further studies will need to investigate their exact role in adipocyte differentiation. It will also be of interest to explore which cell population within the SVF is responsible for the secretion of these inhibitory factors and whether these factors are deregulated in adipose tissue in the state of obesity. Taken together, this PhD thesis identifies several novel factors that control adipogenesis and have an impact on metabolic homeostasis. These factors are tightly linked to extracellular matrix organization, it will therefore be of interest for the future to study the influence of altered extracellular matrix structure on adipocyte differentiation in the progression of obesity.

Zusammenfassung

Zusammenfassung
Das Fettgewebe hat einen groen Einfluss auf den Metabolismus des Krpers. bergewicht ist daher ein Hauptrisikofaktor fr metabolische Erkrankungen. Ein wichtiger adaptiver Mechanismus, der bei bergewicht die Speicherkapazitt des Fettgewebes erhht und dadurch normale Adipozytenfunktion erhlt, ist die Neubildung von Adipozyten durch Differenzierung. Dieser Vorgang verhindert eine Akkumulierung von Lipiden auerhalb des Fettgewebes, welches eine Hauptursache fr das Entstehen von Insulinresistenz darstellt. Diese Doktorarbeit liefert neue Erkenntnisse ber die Regulierung der Adipozytendifferenzierung und deren Einfluss auf den Zucker- und Lipid-Metabolismus. Kapitel 2 errtert, wie Retinoid-related orphan receptor (ROR) die Expression von Matrixmetalloproteinase 3 (MMP3) kontrolliert und dadurch die Adipogenese inhibiert. Expressionsanalysen in der Stromafraktion (SVF), die Adipozyten-Vorluferzellen enthlt, ergaben, dass ROR bei dickleibigen Patienten hochreguliert ist. Verminderte Expression von ROR in murinem und humanem Fett bei Dickleibigkeit korreliert mit erhhter Adipozytendifferenzierung, kleineren Adipozyten und verbesserter

Insulinsensitivitt. Durch seinen Einfluss auf die Plastizitt des Fettgewebes ist ROR daher ein Faktor, der die Lipid- und Glucose-Homeostase bei bergewicht beeinflusst. Die Rolle von MMP3 in extrazellulrer Matrix-Reorganisation und wie sich dies auf die Adipozytendifferenzierung auswirkt, wird fr die Zukunft Gegenstand weiterer Untersuchungen sein. Wie vorangehend beschrieben ist die dynamische Vernderung der extrazellulren Matrix ein wichtiger Bestandteil der Adipozytendifferenzierung. Preadipozyten mssen whrend der Differenzierung ihre Zellstruktur verndern und darber hinaus dient die extrazellulre Matrix als Anker fr Wachstumsfaktoren, die den Differenzierungsprozess beeinflussen. Die Remodellierung der extrazellulren Matrix wird durch Matrix-Metalloproteinasen (MMPs) und deren Inhibitoren (TIMPs) bewerkstelligt. Kapitel 3 untersucht die Regulation von TIMP1 in Adipositas und dessen Einfluss auf die Adipozytendifferenzierung. TIMP1 ist hochreguliert in dickleibigen Mausmodellen und inhibiert die Adipogenese. Injektionen von TIMP1 in Kombination mit einer 5

Zusammenfassung Hochfett-Dit fhrt zu vergrerten Adipozyten, verschlechterter Kontrolle des Blutzuckers und erhhten Blutfettwerten verglichen mit Kontrollmusen. Dadurch begnstigt TIMP1 Insulinresistenz bei bergewicht. Weitere Experimente werden die Rolle von TIMP1 in der Regulation der MMP-Aktivitt, und wie sich dies auf die Matrix des Fettgewebes auswirkt, untersuchen. Neben Matrixregulation spielt auch die Verteilung von Fettgewebe im Krper eine wichtige Rolle fr die Ausbildung von metabolischen Komplikationen. Funktionale Unterschiede in intra-abdominalem und subkutanem Fettgewebe sind der Grund dafr. Kapitel 4 vergleicht das Differenzierungspotential verschiedener Fettdepots. Die Adipogenese in subkutaner SVF ist signifikant erhht im Vergleich zu viszeraler SVF, was sich in kleineren Adipozyten im subkutanen Fettgewebe wiederspiegelt. Experimente mit konditioniertem berstand zeigen, dass parakrine Faktoren, sekretiert von der Stromafraktion, die Adipozytendifferenzierung inhibieren; dieser Effekt ist ausgeprgter in der viszeralen verglichen mit der subctuanen SVF. Mittels massenspektrometrischer und mRNA-Expressionsanalysen wurden Proteine ermittelt, die von der Stromafraktion sekretiert werden und in visceraler SVF hochreguliert sind. Eine funktionelle Analyse dieser Proteine fhrte zur Identifikation von sieben neuen, sekretierten Faktoren, die die Adipogenese unterdrcken. Die Mehrheit dieser identifizieren Proteine ist mit der extrazellulren Matrix assoziiert. Deren exakte Rolle im Differenzierungsprozess wird Gegenstand weiterer Studien sein. Darber hinaus wird es von Interesse sein zu untersuchen, welche Zellpopulationen in der Stromafraktion diese Proteine sekretieren und ob diese Proteine in Zusammenhang mit bergewicht dereguliert sind. Zusammenfassend beschreibt diese Arbeit mehrere neue Faktoren, die die Adipogenese beeinflussen und dadurch einen Einfluss auf den Metabolismus haben. All diese Faktoren hngen eng mit extrazellulrer Matrixorganisation zusammen. Fr die Zukunft wird es daher von Bedeutung sein zu untersuchen, wie eine vernderte extrazellulre Matrixstruktur im Fett bei bergewicht die Adipozytendifferenzierung beeinflusst.

Chapter 1

Chapter 1
Table of Contents

Obesity and the metabolic syndrome- a growing epidemic......................................... 8 Facts and numbers .................................................................................................... 8 Genetic predisposition .............................................................................................. 9 Drugs and treatments ............................................................................................. 10 Adipose tissue- a major player in metabolism ........................................................... 12 Adipose tissue function ........................................................................................... 12 Molecular mechanisms of nutrient management in the adipocyte ....................... 12 Fat depots within the body ..................................................................................... 14 Adipose tissue composition .................................................................................... 16 Adipose tissue - an endocrine organ ....................................................................... 16 Adipocyte differentiation and adipose tissue plasticity.......................................... 17 Adipocyte precursors- finding the needle in the hay stack .................................... 20 Adipose tissue in obesity ............................................................................................ 21 Dysregulated lipolysis- how free fatty acids impair metabolic homeostasis.......... 21 The role of changing adipocytokine profiles in obesity .......................................... 22 Mitochondrial dysfunction in adipocytes ............................................................... 23 Adipose tissue fibrosis............................................................................................. 23 Adipocyte cell size and adipogenesis - a link to adipose tissue function................ 23

Chapter 1

Obesity and the metabolic syndrome- a growing epidemic

Facts and numbers Obesity has emerged as one of the leading public-health issues in the past decades. According to the International Obesity Taskforce 1.1 billion people are overweight with a body mass index (BMI) over 25 kg/m2 and 312 million are classified as obese (BMI > 30 kg/m2)1. Whereas in the US the high prevalence of obesity (33.8% in 2007/082) has attracted broad public attention, overnutrition has also become an increasing problem in Europe and in less developed countries (Fig. 1a)3. According to a survey from 2008 the prevalence of obesity is 19% for men and 22% for women in Germany, whereas in Switzerland 15% of men and 10-11% of women are considered obese4. Though obesity itself is not a disease per se, it is a major risk factor for developing type II diabetes, cardiovascular disease and certain types of cancer at later ages5-8. In accordance with the increasing prevalence of obesity, the number of type II diabetes cases is expected to rise by 32% in Europe, 72% in the United States and over 100% in developing countries (Fig. 1b)3. In 2006, the International Diabetes Federation published the latest definition of the metabolic syndrome describing a cluster of factors associated with an increased risk for atherosclerotic cardiovascular disease (CVD) and diabetes. For a person to be diagnosed with the metabolic syndrome the following criteria have been defined: central obesity measured by waist circumference plus two additional factors such as raised triglycerides (>150 mg/dl), reduced HDL cholesterol (<40-50 mg/dl), raised blood pressure (>130 mm Hg systolic or >85 mmHg diastolic) or raised fasting plasma glucose (>100mg/dl)9-10. These defined criteria are important in order to diagnose people with metabolic syndrome early and to initiate lifestyle interventions and treatment before the development of diabetes and CVD.

Chapter 1
a

Figure 1 a) trends in adult obesity prevalence in Europe, adapted from International Association of Obesity b) Prospective development of diabetes in the future world wide3 Genetic predisposition The causes of obesity and insulin resistance are diverse. Although our modern sedentary life style, with reduced physical activity and an unlimited offer of food, supports metabolic diseases, it cannot be neglected that genetic predisposition also plays a major role11-12. Several monogenetic causes for obesity have been described, for example mutations in leptin13 and the leptin receptor14 or mutations in the

Chapter 1 melanocortin receptor 415 and pro-opiomelanocortin16 leading to pronounced hyperphagia. Likewise, monogenetic disorders of glucose metabolism are known covering essential genes in -cell development, insulin secretion and insulin signaling. One of the earliest examples is the discovery of mutations in the insulin receptor in 198817. Other monogenetic forms affect transcription factors such as HNF4, HNF1 and , PDX-1, NeuroD1 and KLF1118-24, which influence -cell development and function. These mutations follow an autosomal-dominant inheritance and lead to a type of early-onset diabetes known as maturity onset diabetes of the young (MODY). Mutations in glucokinase (MODY2), the sulphonylurea receptor or the Kir6.2 potassium channel also compromise -cell function25-27, whereas mutations in insulin itself or AKT228-29 can disrupt insulin signaling. However, monogenetic cases are rare, most people prone to obesity or diabetes carry poly-genetic risk factors. Using new and fast sequencing techniques, much effort has been undertaken in the past several years to identify common genetic variants associated with features of the metabolic syndrome. Genome wide association studies and meta analyses combining results from multiple studies30-32 have led to the identification of numerous single nucleotide polymorphisms and loci, such as the fto gene, which is strongly associated with body mass index33-34. Many of these identified variants lay within introns or even in noncoding regions. Therefore, one of the major challenges in the future will be to identify their function in the disease. Drugs and treatments The most effective treatment for people with metabolic syndrome is a drastic change in lifestyle, including weight loss, a healthy diet, regular physical activity and giving up smoking. A very effective option to reduce weight is bariatric surgery including gastroplasty, gastric banding, gastric bypass and biliopancreatic diversion35-36. Interestingly, gastric surgery has beneficial effects on glucose control and insulinresistance independent of weight loss within days of the surgery; the mechanisms for these effects are currently unclear37-38. Other than invasive surgery, medical treatment for different features of the metabolic syndrome is common practice39. A widely used drug against hyperglycemia is metformin, which lowers hepatic glucose output, partially through AMPK activation40. Insulin analogues and drugs increasing insulin 10

Chapter 1 secretion such as sulphonylureas can be used to overcome insulin resistance in the early stages of the disease. Despite certain side effects, thiazolidinediones- PPAR agonists- are also commonly prescribed in type II diabetes, as they indirectly enhance insulin sensitivity by modulating adipose tissue function41. Much hope has recently been raised by GLP-1 analogues, which stimulate insulin secretion, and DPP4-inhibtors that prolong GLP-1 action42. For the control of dyslipidemia, the most commonly used drugs are statins, which inhibit HMG-CoA reductase thereby blocking cholesterol synthesis43. Another popular class of drugs are the fibrates, a group of PPAR agonists that primarily enhance fatty acid oxidation44. Other possibilities are cholesterol-absorption blockers45 that lower LDL serum levels. For the treatment of obesity, most currently available drugs exert their effect on the central nervous system, for example inhibiting noradrenergic and serotonergic reuptake in the hypothalamus (sibutramine)46 or antagonizing cannabinoid receptors47. Another strategy to reduce weight pharmacologically is the use of fat absorption blockers, e.g. through the inhibition of gastric and pancreatic lipases (orlistat)48, which are most effective in combination with a reduced-calorie diet. As adipose tissue growth is largely dependent on nutrient supply, it is not surprising that angiogenesis plays an important role in fat expansion49-50. This may be a completely novel approach to inhibit excess adipose tissue accumulation51. Future studies are required to show whether this approach improves the metabolic status or on the contrary leads to adverse effects through ectopic lipid accumulation. Overall, it has to be noted that the development of new drugs targeting features of the metabolic syndrome has had limited success over the past years. The multifactorial nature of the metabolic syndrome poses the problem of polypharmacy and finding suitable drugs targets that reduce multiple metabolic risk factors at a time has been proven difficult39.

11

Chapter 1

Adipose tissue- a major player in metabolism

Adipose tissue function Even though excess adipose tissue predisposes to metabolic complications, adipose tissue also has essential functions within the body. Adipocytes are the main storage site for excess energy in the form of triglycerides. An average man of 70 kg has about 15 kg of adipose tissue (21%)52. In the case of severe forms of lipodystrophy, such as Berardinelli-Seip congenital lipodystrophy53, where the majority of normal adipose tissue depots are missing, fat is ectopically stored in extra-adipose tissue, leading to severe insulin resistance. Acquired lipodystrophy is also a side effect of antiretroviral drugs in HIV patients54. Besides its obvious storage function, adipose tissue has mechanical functions such as insulation and protection against mechanical forces. Moreover, adipose tissue functions as an endocrine organ. This was only discovered in 1994 with the description of leptin, which is secreted from adipose tissue and has systemic signaling capacity55. Molecular mechanisms of nutrient management in the adipocyte As mentioned earlier, adipose tissue stores and manages fatty acid pools within the body. Fatty acids reach the adipocyte via the circulation in three major forms: nonesterified fatty acids (FFA) are associated with albumin in the serum56, lipoprotein lipase (LPL) is associated with the outer membrane of the adipocyte via heparane sulphate proteoglycans57 and can hydrolyse FFA from triglyceride-rich chylomicrons58 and very low density lipoprotein particles (VLDL) can be taken up completely by the adipocyte via the VLDL receptor (VLDL-R) or low density lipoprotein receptor-related protein (LRP)59-60. Although FFAs can enter the cell via passive diffusion, several membrane proteins are also implicated in their uptake such as fatty acid transport protein 1 (FATP1) and CD36, especially for long-chain fatty acids61. Intracellularly, fatty acids are transported by fatty acid binding proteins (FABPs), especially A-FABP and to a minor extent E-FABP62. After entry into the cell fatty acids are esterified by acyl-coAsynthetase and used for triacylglyceride (TAG) synthesis on the endoplasmatic reticulum (Fig. 2)63. Apart from fatty acids, adipocytes take up glucose via GLUT4 transporters, which are translocated to the membrane upon insulin stimulation64. An

12

Chapter 1 intermediate metabolite of glycolysis is also needed for TAG synthesis, as dihydroxyacetone-phosphate is converted to glycerol-3-phosphate, the backbone of triglycerides, by glycerol-3-phosphate dehydrogenase (GPDH). Additionally, adipocytes use acetyl-CoA for de novo fatty acid synthesis via acetyl-CoA carboxylase (ACC) and the multifunctional enzyme fatty acid synthase (FAS)65. For storage, TAG are incorporated into lipid droplets, which form at the ER and are covered with PAT family proteins such as perilipin66. These anabolic pathways are mainly driven by insulin in the fed state, which in turn stimulates the major transcription factor for fatty acid and triglyceride synthesis, sterol regulatory element binding protein 1c (SREBP-1c)67. During fasting and exercise, when stored nutrients need to be mobilized, lipolysis is stimulated in adipocytes and leads to the release of free fatty acids and glycerol, which can be metabolized by other tissues68. Triglyceride hydrolysis is a three-step process driven by adipose triglyceride lipase (ATGL), hormone sensitive lipase (HSL) and monoacylglycerol lipase (MGL). The major signaling pathways activating lipolysis are adrenergic stimulation and the fasting hormone glucagon. Both lead to activation of adenylate cylase and the production of cAMP, which in turn activates protein kinase A (PKA). Subsequently, PKA phosphorylates perilipin and HSL for the mobilization of fatty acids (Fig. 2)69. Insulin suppresses lipolysis via phosphorylation of phosphodiesterase 3B (PDE3B) by Akt leading to hydrolysis of cAMP. Additionally, 2-adrenergic stimulation can activate an inhibitory G-protein thereby suppressing lipolysis70. Thus, circulating hormones and adrenergic innervation in the adipocyte reciprocally regulate anabolic and catabolic pathways.

13

Chapter 1

Figure 2 Lipid management in the adipocyte. Free fatty acids are released by lipoprotein lipase (LPL), taken up by the cell and incorporated into triglycerides. Lipid droplets coated with PAT family proteins (TIP47, ADRP, S3-12, perilipin) emerge from the ER. After -adrenergic stimulation, activation of proteinkinase A (PKA) leads to phosphorylation of perilipin (PER) and hormone sensitive lipase (HSL) and free fatty acids are released by lipolysis. ACS: acyl-CoA synthetase, GPAT: glycerol-3-phosphate acyltransferase, AGPAT: sn-1-acylglycerol-3-phosphate acyltransferase, PAP: phosphatidic-acid phosphohydrolase, MGAT: monoacylglycerolacyltransferase, DGAT: diacylglycerol acyltransferase from Shi et al.71 Fat depots within the body There are many different adipose tissue depots within the body. One special subclass is brown adipose tissue, which is especially developed in infants and small rodents, where it can be induced by cold exposure72. Brown adipocytes have many small lipid droplets, in contrast to white adipocytes, which have one large lipid core. The main function of brown adipocytes is the production of heat through expression of uncoupling-protein 1 (UCP1). It has long been believed that only children and rodents have brown adipose tissue, which is located in distinct pads adjacent to the shoulder blades. However, it has recently been shown that brown adipocytes exist in adult humans and are interspersed into the common white fat pads and might also be inducible by cold73-74. Inducing brown adipocytes by specific agents could therefore be an interesting novel therapeutic intervention to increase energy expenditure and treat obesity. 14

Chapter 1 On the other hand, white adipose tissue represents the majority of body fat. Based on the localization adipose tissue depots can be classified as subcutaneous adipose tissue underneath the dermis or visceral adipose tissue within the cavities of the body. Most visceral adipose tissue lies within the abdominal cavity with the gonadal, the mesenteric and the retroperitoneal fat compartments being the most prominent (Fig. 3)52. This distinction is not only of anatomical interest but also has functional implications. It has been shown that visceral and subcutaneous adipose tissue show distinct gene expression patterns75. Furthermore, visceral adipose tissue poses a greater risk for metabolic complications than subcutaneous adipose tissue76. The reasons for this are currently not clear. Potential explanations include that factors secreted from the visceral adipose tissue drain directly into the portal vein or that adipocyte function itself is altered in different depots, which will be discussed further below. Due to hormonal differences, overweight women tend to have more subcutaneous adipose tissue, whereas overweight men show increased central adiposity and are thus more prone to metabolic complications77. Moreover, treatment with glitazones leads, amongst other things, to redistribution of fat to the subcutaneous compartment, thereby ameliorating the metabolic profile without causing weight loss78.

Figure 3 CT scan of different adipose tissue depots in a mouse.

15

Chapter 1 Adipose tissue composition tion Adipose tissue is not solel olely composed of mature adipocytes, althou though these cells comprise the majority of f the space (Fig.4). Endothelial and smooth musc uscle cells as well as fibroblasts can be found und within adipose tissue. Another class of c cells present in adipose tissue are immune ne c cells such as macrophages, which play an n important im role in obesity, as discussed further ther down79. Finally, adipose tissue has the cap capacity to rebuild new fat cells with the hel help of fibroblast-like adipocyte precursor r ce cells, which can proliferate and differentiate tiate into mature adipocytes80. All these cells ls ca can be physically separated from the floating ting adipocytes by centrifugation and are theref erefore collectively referred to as the stromal-va vascular fraction (SVF). As a connective tissue sue, fat is also rich in extracellular matrix prot proteins with high amounts of collagen I, IV, , VI and laminin among others81-82.

Figure 4 Adipo dipose tissue composition. from Schffler et al.83

Adipose tissue - an endocrin crine organ Cells within the adipose e tis tissue secrete many important signaling mo molecules, which control whole body metabol abolism. The first adipokine to be discovered wa was leptin. Leptin is an 18 kDa protein primar marily secreted from adipocytes and signals s thr through different isoforms of the leptin recept ceptor84-85. Leptin is a major satiety signal and suppresses food

16

Chapter 1 intake through direct hypothalamic repression of the orexigenic peptides NPY and AgRP86. Leptin also enhances fatty acid oxidation and improves peripheral insulin sensitivity via indirect signaling over the central nervous system, as well as directly through peripheral activation of AMP kinase (AMPK)87. Another important adipokine is adiponectin, which signals as a multimer through its receptors adipoR1 and adipoR2 and has a strong insulin-sensitizing effect88. Two adipokines primarily secreted by visceral adipose tissue are visfatin and omentin89-90. Whereas visfatin activates glucose uptake via the insulin-receptor, omentin has insulin-sensitizing effects. Resistin, on the other hand, promotes glucose output from the liver, thereby increasing plasma glucose levels. Another adipocytokine secreted from adipocytes is retinol-binding protein 4 (RBP4), which primarily transports retinol in the blood but also strongly impairs insulin signaling in muscle and liver91. Finally, adipocytes and macrophages residing within the adipose tissue can also secrete many inflammatory cytokines such as TNF and IL-679.

Figure 5 Secreted adipocytokines from adipose tissue and their influence on glucose metabolism. from Rosen et al.85

Adipocyte differentiation and adipose tissue plasticity Adipose tissue has the remarkable capacity to remodel and adapt to the nutritional status of the body. As adipocytes continue storing lipids, they can change in size several fold92; this mechanism of growth is commonly referred to as hypertrophy. In addition, adipose tissue is able to regenerate fat cells by de novo adipogenesis from adipocyte precursor cells. This alternative mechanism of adipose tissue growth is

17

Chapter 1 called hyperplasia. In fact, adipocytes undergo a constant turn-over during life, making fat a highly dynamic tissue93. Many of the genetic processes involved in adipocyte differentiation have been revealed by cell culture experiments either using preadipocyte cell lines, such as 3T3L1 or 3T3-F442A, primary stromal-vascular fraction, embryonic fibroblasts or bone marrow stromal cells, all of which can be differentiated into adipocytes in vitro. One of the major regulators of adipogenesis is the transcription factor peroxisome proliferator-activated receptor PPAR294, which is mandatory for the adipocyte lineage as well as for the maintenance of the adipocyte phenotype. CCAAT-enhancer binding proteins (C/EBPs) also play an important role in adipogenesis, with C/EBP and C/EBP being expressed during early differentiation and inducing PPAR and C/EBP expression95. Not surprisingly, the family of Krppel-like factors (KLFs), which generally control cell proliferation and differentiation, also have a large impact on adipocyte differentiation. Whereas KLF5, KLF6 and KLF16 promote differentiation, other family members like KLF2 and 7 inhibit the process80. In contrast to proadipogenic factors, many factors are expressed in the preadipocyte to maintain the undifferentiated state; these factors are generally downregulated upon induction of differentiation. Examples include the surface protein preadipocyte factor1 (PREF1)96-97 and members of the GATA family, e.g. GATA2 and GATA398. Another general principle of regulation is the recruitment of co-activators and co-repressors, which often mediate histone acyltransferase (HAT) or histone deacetylase (HDAC) activity, respectively. Examples include the proadipogenic HATs CREB-binding protein (CBP) and p300 or the antiadipogenic nuclear receptor co-repressor NcoR99. In multipotential cells such as mouse embryonic fibroblasts, differentiation into different myogenic, osteogenic or adipogenic lineages reciprocally inhibit each other. One well-known case is the reciprocal inhibition of PPAR and RUNX2 deciding between an adipogenic versus an osteogenic fate100-101. Apart from transcriptional regulation, extracellular signaling is also an important factor for adipogenesis. Examples for secreted proteins that trigger inhibitory signaling pathways are Wnt10b102, sonic hedgehog (SHH)103 or TGF104. On the other hand, extracellular proteins promoting adipogenesis include several

18

Chapter 1 members of the bone morphogenic protein family such as BMP2 and 4105and different fibroblast growth factors (Fig. 6)106-107. Another important step in adipogenesis is extracellular matrix (ECM) remodeling108-109. During differentiation, the fibronectin-rich matrix of a preadipocyte needs to be converted to the typical basement membrane of a mature adipocyte, which includes laminin, nidogen/entacin, type-IV and -VI collagens110. Two main systems involved in reshaping the extracellular matrix are the plasminogen/plasmin cascade as well as the matrix metalloproteinase and tissue inhibitor of matrix metalloproteinases (MMP/TIMP) proteins. During adipogenesis plasmin is activated by kallikrein and degrades the fibronectin-rich preadipocyte stromal matrix110. The matrix

metalloproteinases and their inhibitors have diverse effects on adipogenesis and their expression is differentially regulated during the differentiation process. Whereas several MMPs such as MMP2, MMP9 and MT1-MMP promote adipogenesis (the effect of MT1-MMP being only evident in 3D culture111), others like MMP3, MMP11 and MMP19 have the inverse effect108,112. The TIMPs, binding and inhibiting a variety of different MMPs, add another complex layer of regulation to this system. Whether different components of the ECM modulate differentiation by physical force, by sequestering important growth factors or even mediating signals themselves needs to be further investigated. For extracellular matrix proteins, as well as for many other proteins involved in adipogenesis, it will be crucial not only to look at the process of forced in vitro differentiation in a 2D culture but also to develop novel in vivo techniques in order to shed light on the mechanisms of adipogenesis in vivo in a natural 3D environment.

19

Chapter 1

Figure 6 Transcription factors, extracellular signaling molecules and matrix alterations regulating adipocyte differentiation. green: pro-adipogenic, red: anti-adipogenic

Adipocyte precursors- finding the needle in the hay stack Whereas many molecular signaling pathways are known that control adipogenesis, identifying the pool of preadipocytes in vivo has proven to be difficult. During development, adipose tissue arises from the mesoderm and preadipocytes are thought to originate from multipotent mesenchymal stem cells. However, whether there are different stages of multipotent precursor cells or cells committed entirely to the adipocyte line in vivo remains elusive. Fact is that the stromal-vascular fraction of adipose tissue contains precursor cells and that adipocyte turnover in adipose tissue takes place throughout life92-93. Studies by Tang et al.113 have recently proposed that the adipose vasculature forms a niche for adipocyte precursor cells. This matches the well-established concept of a stem cell niche that has been shown for hematopoietic and other stem cells114. It has been debated whether cells from the bone marrow are able to repopulate the pool of precursor cells in adipose tissue. Experiments using GFPpositive bone marrow transplantations in mice have led to the conclusion that under

20

Chapter 1 chow diet conditions bone-marrow derived cells do not give rise to new adipocytes in the adipose tissue. However, homing of bone marrow cells to the fat and differentiation into adipocytes can be achieved by feeding a high fat diet or, to a much larger extent, through treatment with the PPAR activator rosiglitazone115-117. For a long time the only surface marker known for preadipocytes was preadipocyte factor 1 (PREF1/DLK1)96-97. Recently, several markers known from other stem cells have been proposed to define an adipocyte precursor cell as

CD29+CD34+Sca1+CD24+118. However, there has been some doubt about whether CD34 can be considered as an adipocyte precursor marker, as briefly cultured human CD34+ stroma cells have a high proliferation but low differentiation potential in vitro119. Furthermore Sca1+CD34+ cells from visceral and subcutaneous fat depots show different proliferation and differentiation capacities, indicating that these markers alone do not define adipocyte precursor cells sufficiently120. For hematopoietic stem cells, CD34 has been proposed as a marker distinguishing dormant from active selfrenewing stem cells114, a concept that has not been investigated for adipocyte precursor cells so far. These studies show that there is a great need for the identification of new markers, which describe subpopulations of adipocyte precursors. These markers will have to be thoroughly characterized in vivo without prior in vitro culture.

Adipose tissue in obesity

Dysregulated lipolysis- how free fatty acids impair metabolic homeostasis Excess adipose tissue increases the risk for a number of diverse conditions such as atherosclerosis, hypertension, insulin resistance and cancer. Much effort has been undertaken to understand the molecular changes in adipose tissue function in the context of obesity that lead to these secondary complications. One important malfunction of adipose tissue during obesity is a detrimental increase in lipolysis concomitant with an excess release of non-esterifed fatty acids. Free fatty acids are thought to be one of the major culprits for insulin resistance; they acutely inhibit insulin stimulated glucose uptake in muscle, as well as insulin-mediated suppression of hepatic glucose output121-122. The mechanisms involved in these processes are diverse. 21

Chapter 1 Primarily, elevated FFA levels lead to intracellular accumulation of fatty acid based signaling molecules such as diacylglycerol (DAG) or long-chain acyl-CoA. DAG activates protein kinase C (PKC), which can in turn inhibit insulin signaling by serinephosphorylation of insulin receptor substrate 1 and 2 (IRS 1/2)7. PKC also activates NADH oxidase, thus increasing reactive oxygen species and inhibiting NO production123. This is especially harmful in endothelial cells leading to decreased vasodilation and hypertension. Additionally, intracellular fatty acids lead to activation of inflammatory pathways, mainly via activation of nuclear factor B (NFB) and c-jun N-terminal kinase (JNK)124. Moreover, increased serum FFAs also provoke cardiomyopathy, increasing the risk for myocardial infarction125 and promote atherosclerosis, at least indirectly, via increased inflammation as well as elevated hepatic VLDL output126. In -cells, long term exposure with high FFA concentrations causes impaired glucose-stimulated insulin secretion, thus blocking compensatory insulin release127. The role of changing adipocytokine profiles in obesity Apart from increased lipolysis, the adipocytokine profile of adipose tissue also undergoes major changes in obesity. The release of most adipokines is elevated in the state of obesity. This is also true for the metabolically beneficial hormone leptin. However, leptin signaling is concomitantly impaired, making it an unattractive target for the treatment of obesity128. One exception from generally elevated adipokine levels is adiponectin, whose secretion is diminished in obesity129. Inflammatory cytokines, too, are highly upregulated in adipose tissue of obese subjects. For example, monocyte chemoattractant protein 1 (MCP-1) is secreted in high amounts from adipocytes leading to massive macrophage infiltration79. Other examples for proinflammatory cytokines either released from macrophages or from adipocytes directly are TNF, IL-1, IL-6 and C-reactive protein (CRP). Interestingly, many of the cytokines are secreted to a greater extent from visceral adipose tissue130-131, adding to the theory that visceral fat leads to more severe metabolic effects than subcutaneous fat. Cytokines released from adipose tissue induce a systemic low-grade inflammation promoting insulin resistance132. Moreover, TNF inhibits triglyceride synthesis through downregulation of PPAR and its target gene LPL as well as the glucose transporter 22

Chapter 1 GLUT4. At the same time, TNF inhibits insulin-mediated attenuation of lipolysis, increases the cAMP pool and downregulates the lipid droplet protein perilipin, all of which further enhances FFA release133. Mitochondrial dysfunction in adipocytes Though more discussed in the context of muscle function, mitochondrial dysfunction has also gotten some attention with regard to the adipocyte. Metabolizing fatty acids via -oxidation takes place in the mitochondria; impaired mitochondrial function therefore promotes systemic lipid overload and ectopic lipid accumulation. In muscle cells as well as in adipocytes reduced mitochondrial numbers and function are associated with obesity134-136. On the other hand, the insulin-sensitizing agent rosiglitazone promotes mitochondrial biogenesis136-137. However, whether reduced mitochondrial function primarily leads to accelerated accumulation of lipids or whether intracellular lipid overload and an increased rate of -oxidation ultimately damage mitochondria, as it has been suggested in muscle cells138, needs to be further analyzed in adipose tissue. Adipose tissue fibrosis Another aspect of adipose tissue dysfunction in the obese state that has been largely neglected concerns the adipose tissue matrix. It has been shown that the expression of matrix proteins is highly upregulated in adipose tissue of obese subjects, rendering the adipose tissue scaffold more rigid139. Surprisingly, ablation of collagen VI, a highly abundant extra-cellular matrix protein in the adipose tissue, leads to improvement of whole-body energy homeostasis in obese animals82. Additionally, a variety of matrix metalloproteinases, which shape the extra-cellular matrix, are deregulated in adipose tissue in the state of obesity140-141. Whereas to date the functional impact of an altered extracellular matrix on mature adipocytes remains elusive, many studies have established a link between extracellular matrix remodeling and adipogenesis, thus modulating adipose tissue plasticity, as it has been discussed earlier on. Adipocyte cell size and adipogenesis- a link to adipose tissue function Several studies have made the observation that subgroups of obese but otherwise metabolically healthy subjects exist142-144. This is accompanied by high insulin 23

Chapter 1 sensitivity, low levels of inflammation markers and decreased visceral obesity. The underlying mechanisms are not fully understood. One possible explanation for the difference in metabolic phenotypes can be the fact that total fat mass itself describes the functional status of adipose tissue insufficiently. This is true for differences in fat depots, i.e. subcutaneous versus visceral, but also for single cells within the adipose tissue. Over 40 years ago Salans et al. showed that insulin-responsiveness in isolated human and rat adipose tissue is related to adipocyte cell size, with larger cells being increasingly insulin-resistant145-146. Since then many studies have supported this

theory: large cells show a distinct gene expression profile with elevated expression of inflammatory markers147, the lipolytic rate is higher in hypertrophied cells and further stimulated by TNF148-149, and insulin-stimulated GLUT4 translocation for glucose uptake depends on cell size150. A recent cohort study supports the idea that adipocyte size, especially in the omental fat compartment, is a critical parameter describing metabolic health in obese subjects151. Adipocyte size in the adipose tissue is mainly determined by the ability of the mature adipocyte to grow in size (hypertrophy) versus the ability to build new fat cells through adipogenesis (hyperplasia). Thus, increased adipocyte differentiation in response to excess nutrient supply will ultimately lead to a larger number of smaller cells, which remain insulin sensitive and secrete less inflammatory signals (Fig. 7). Recently, it has been shown that visceral adipose tissue mainly grows via hypertrophy, whereas for subcutaneous adipose tissue a hyperplastic response is predominant in mice fed a high fat diet120. This observation could be another important explanation for the more detrimental effect of excess visceral adipose tissue. In human adipose tissue this effect is less evident. Judging by gene expression of differentiation markers in obese women, hyperplasia is predominant in the subcutaneous fat depot152. Van Harmelen et al. describe increased proliferation of human subcutaneous SVF but no difference in differentiation capacity between fat depots153. On the contrary, Tchkonia et al. report increased differentiation in subcutaneous compared to visceral SVF154. Age- and sexdependent differences could play an important role here, as preadipocyte pools might exhaust with age and increasing adiposity and sex steroids have opposing effects on adipocyte differentiation155-156. 24

Chapter 1

Figure 7 Adipose tissue growth via hyperplasia (adipocyte differentiation) or hypertrophy (increasing volume of existing adipocytes) and the resulting metabolic consequences. In conclusion, adipocyte differentiation is an important adaptive response to excess nutrition and influences the metabolic outcome of obesity. This PhD thesis explores novel pathways regulating adipogenesis, which have an impact on the metabolic status of obese subjects. This work also examines the mechanisms as to why an adaptive hyperplastic response is impaired in visceral depots compared to subcutaneous adipose tissue.

25

Chapter 1

References
1. 2. 3. 4. 5. 6. Haslam, D.W. & James, W.P. Obesity. Lancet 366, 1197-1209 (2005). Flegal, K.M., Carroll, M.D., Ogden, C.L. & Curtin, L.R. Prevalence and trends in obesity among US adults, 1999-2008. JAMA 303, 235-241 (2010). Hossain, P., Kawar, B. & El Nahas, M. Obesity and diabetes in the developing world--a growing challenge. N Engl J Med 356, 213-215 (2007). Berghofer, A., et al. Obesity prevalence from a European perspective: a systematic review. BMC Public Health 8, 200 (2008). Pothiwala, P., Jain, S.K. & Yaturu, S. Metabolic syndrome and cancer. Metab Syndr Relat Disord 7, 279-288 (2009). Calle, E.E., Rodriguez, C., Walker-Thurmond, K. & Thun, M.J. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med 348, 1625-1638 (2003). Kahn, S.E., Hull, R.L. & Utzschneider, K.M. Mechanisms linking obesity to insulin resistance and type 2 diabetes. Nature 444, 840-846 (2006). Van Gaal, L.F., Mertens, I.L. & De Block, C.E. Mechanisms linking obesity with cardiovascular disease. Nature 444, 875-880 (2006). Eddy, D.M., Schlessinger, L. & Heikes, K. The metabolic syndrome and cardiovascular risk: implications for clinical practice. Int J Obes (Lond) 32 Suppl 2, S5-10 (2008). Grundy, S.M., Brewer, H.B., Jr., Cleeman, J.I., Smith, S.C., Jr. & Lenfant, C. Definition of metabolic syndrome: report of the National Heart, Lung, and Blood Institute/American Heart Association conference on scientific issues related to definition. Arterioscler Thromb Vasc Biol 24, e13-18 (2004). Maes, H.H., Neale, M.C. & Eaves, L.J. Genetic and environmental factors in relative body weight and human adiposity. Behav Genet 27, 325-351 (1997). O'Rahilly, S. Human genetics illuminates the paths to metabolic disease. Nature 462, 307-314 (2009). Montague, C.T., et al. Congenital leptin deficiency is associated with severe early-onset obesity in humans. Nature 387, 903-908 (1997). Clement, K., et al. A mutation in the human leptin receptor gene causes obesity and pituitary dysfunction. Nature 392, 398-401 (1998). Vaisse, C., Clement, K., Guy-Grand, B. & Froguel, P. A frameshift mutation in human MC4R is associated with a dominant form of obesity. Nat Genet 20, 113-114 (1998). Krude, H., et al. Severe early-onset obesity, adrenal insufficiency and red hair pigmentation caused by POMC mutations in humans. Nat Genet 19, 155-157 (1998). Yoshimasa, Y., et al. Insulin-resistant diabetes due to a point mutation that prevents insulin proreceptor processing. Science 240, 784-787 (1988). Neve, B., et al. Role of transcription factor KLF11 and its diabetes-associated gene variants in pancreatic beta cell function. Proc Natl Acad Sci U S A 102, 4807-4812 (2005). Malecki, M.T., et al. Mutations in NEUROD1 are associated with the development of type 2 diabetes mellitus. Nat Genet 23, 323-328 (1999). Horikawa, Y., et al. Mutation in hepatocyte nuclear factor-1 beta gene (TCF2) associated with MODY. Nat Genet 17, 384-385 (1997). Yamagata, K., et al. Mutations in the hepatocyte nuclear factor-4alpha gene in maturity-onset diabetes of the young (MODY1). Nature 384, 458-460 (1996). Stoffers, D.A., Ferrer, J., Clarke, W.L. & Habener, J.F. Early-onset type-II diabetes mellitus (MODY4) linked to IPF1. Nat Genet 17, 138-139 (1997). Frayling, T.M., et al. Mutations in the hepatocyte nuclear factor-1alpha gene are a common cause of maturity-onset diabetes of the young in the U.K. Diabetes 46, 720-725 (1997). Yamagata, K., et al. Mutations in the hepatocyte nuclear factor-1alpha gene in maturity-onset diabetes of the young (MODY3). Nature 384, 455-458 (1996). Vionnet, N., et al. Nonsense mutation in the glucokinase gene causes early-onset non-insulindependent diabetes mellitus. Nature 356, 721-722 (1992). Gloyn, A.L., et al. Activating mutations in the gene encoding the ATP-sensitive potassiumchannel subunit Kir6.2 and permanent neonatal diabetes. N Engl J Med 350, 1838-1849 (2004).

7. 8. 9. 10.

11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.

26

Chapter 1
27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. Babenko, A.P., et al. Activating mutations in the ABCC8 gene in neonatal diabetes mellitus. N Engl J Med 355, 456-466 (2006). Stoy, J., et al. Insulin gene mutations as a cause of permanent neonatal diabetes. Proc Natl Acad Sci U S A 104, 15040-15044 (2007). George, S., et al. A family with severe insulin resistance and diabetes due to a mutation in AKT2. Science 304, 1325-1328 (2004). Hindorff, L.A., et al. Potential etiologic and functional implications of genome-wide association loci for human diseases and traits. Proc Natl Acad Sci U S A 106, 9362-9367 (2009). McCarthy, M.I. & Hirschhorn, J.N. Genome-wide association studies: potential next steps on a genetic journey. Hum Mol Genet 17, R156-165 (2008). McCarthy, M.I. & Zeggini, E. Genome-wide association studies in type 2 diabetes. Curr Diab Rep 9, 164-171 (2009). Dina, C., et al. Variation in FTO contributes to childhood obesity and severe adult obesity. Nat Genet 39, 724-726 (2007). Frayling, T.M., et al. A common variant in the FTO gene is associated with body mass index and predisposes to childhood and adult obesity. Science 316, 889-894 (2007). Buchwald, H., et al. Bariatric surgery: a systematic review and meta-analysis. JAMA 292, 17241737 (2004). Rubino, F., Moo, T.A., Rosen, D.J., Dakin, G.F. & Pomp, A. Diabetes surgery: a new approach to an old disease. Diabetes Care 32 Suppl 2, S368-372 (2009). Schauer, P.R., et al. Effect of laparoscopic Roux-en Y gastric bypass on type 2 diabetes mellitus. Ann Surg 238, 467-484; discussion 484-465 (2003). Scopinaro, N., Marinari, G.M., Camerini, G.B., Papadia, F.S. & Adami, G.F. Specific effects of biliopancreatic diversion on the major components of metabolic syndrome: a long-term followup study. Diabetes Care 28, 2406-2411 (2005). Grundy, S.M. Drug therapy of the metabolic syndrome: minimizing the emerging crisis in polypharmacy. Nat Rev Drug Discov 5, 295-309 (2006). Zhou, G., et al. Role of AMP-activated protein kinase in mechanism of metformin action. J Clin Invest 108, 1167-1174 (2001). Edgerton, D.S., Johnson, K.M. & Cherrington, A.D. Current strategies for the inhibition of hepatic glucose production in type 2 diabetes. Front Biosci 14, 1169-1181 (2009). Drucker, D.J. & Nauck, M.A. The incretin system: glucagon-like peptide-1 receptor agonists and dipeptidyl peptidase-4 inhibitors in type 2 diabetes. Lancet 368, 1696-1705 (2006). Robinson, J.G., Smith, B., Maheshwari, N. & Schrott, H. Pleiotropic effects of statins: benefit beyond cholesterol reduction? A meta-regression analysis. J Am Coll Cardiol 46, 1855-1862 (2005). Staels, B. & Fruchart, J.C. Therapeutic roles of peroxisome proliferator-activated receptor agonists. Diabetes 54, 2460-2470 (2005). Pearson, T.A., et al. A community-based, randomized trial of ezetimibe added to statin therapy to attain NCEP ATP III goals for LDL cholesterol in hypercholesterolemic patients: the ezetimibe add-on to statin for effectiveness (EASE) trial. Mayo Clin Proc 80, 587-595 (2005). Arterburn, D.E., Crane, P.K. & Veenstra, D.L. The efficacy and safety of sibutramine for weight loss: a systematic review. Arch Intern Med 164, 994-1003 (2004). Van Gaal, L.F., Rissanen, A.M., Scheen, A.J., Ziegler, O. & Rossner, S. Effects of the cannabinoid1 receptor blocker rimonabant on weight reduction and cardiovascular risk factors in overweight patients: 1-year experience from the RIO-Europe study. Lancet 365, 1389-1397 (2005). Curran, M.P. & Scott, L.J. Orlistat: a review of its use in the management of patients with obesity. Drugs 64, 2845-2864 (2004). Rupnick, M.A., et al. Adipose tissue mass can be regulated through the vasculature. Proc Natl Acad Sci U S A 99, 10730-10735 (2002). Brakenhielm, E., et al. Angiogenesis inhibitor, TNP-470, prevents diet-induced and genetic obesity in mice. Circ Res 94, 1579-1588 (2004). Cao, Y. Adipose tissue angiogenesis as a therapeutic target for obesity and metabolic diseases. Nat Rev Drug Discov 9, 107-115 (2010).

39. 40. 41. 42. 43.

44. 45.

46. 47.

48. 49. 50. 51.

27

Chapter 1
52. 53. 54. 55. 56. 57. Shen, W., et al. Adipose tissue quantification by imaging methods: A proposed classification. Obesity Research 11, 5-16 (2003). Van Maldergem, L., et al. Genotype-phenotype relationships in Berardinelli-Seip congenital lipodystrophy. J Med Genet 39, 722-733 (2002). Moreno, S., et al. Disorders of body fat distribution in HIV-1-infected patients. AIDS Rev 11, 126-134 (2009). Zhang, Y., et al. Positional cloning of the mouse obese gene and its human homologue. Nature 372, 425-432 (1994). Brodersen, R., Andersen, S., Vorum, H., Nielsen, S.U. & Pedersen, A.O. Multiple fatty acid binding to albumin in human blood plasma. Eur J Biochem 189, 343-349 (1990). Misra, K.B., Kim, K.C., Cho, S., Low, M.G. & Bensadoun, A. Purification and characterization of adipocyte heparan sulfate proteoglycans with affinity for lipoprotein lipase. J Biol Chem 269, 23838-23844 (1994). Brown, C.M. & Layman, D.K. Lipoprotein lipase activity and chylomicron clearance in rats fed a high fat diet. J Nutr 118, 1294-1298 (1988). Tacken, P.J., Hofker, M.H., Havekes, L.M. & van Dijk, K.W. Living up to a name: the role of the VLDL receptor in lipid metabolism. Curr Opin Lipidol 12, 275-279 (2001). Large, V., Peroni, O., Letexier, D., Ray, H. & Beylot, M. Metabolism of lipids in human white adipocyte. Diabetes Metab 30, 294-309 (2004). Wilsie, L.C., Chanchani, S., Navaratna, D. & Orlando, R.A. Cell surface heparan sulfate proteoglycans contribute to intracellular lipid accumulation in adipocytes. Lipids Health Dis 4, 2 (2005). Hertzel, A.V., et al. Lipid metabolism and adipokine levels in fatty acid-binding protein null and transgenic mice. American journal of physiology 290, E814-823 (2006). Coleman, R.A. & Lee, D.P. Enzymes of triacylglycerol synthesis and their regulation. Prog Lipid Res 43, 134-176 (2004). Olson, A.L. & Pessin, J.E. Structure, function, and regulation of the mammalian facilitative glucose transporter gene family. Annu Rev Nutr 16, 235-256 (1996). Stoops, J.K., et al. Presence of two polypeptide chains comprising fatty acid synthetase. Proc Natl Acad Sci U S A 72, 1940-1944 (1975). Martin, S. & Parton, R.G. Lipid droplets: a unified view of a dynamic organelle. Nat Rev Mol Cell Biol 7, 373-378 (2006). Azzout-Marniche, D., et al. Insulin effects on sterol regulatory-element-binding protein-1c (SREBP-1c) transcriptional activity in rat hepatocytes. Biochem J 350 Pt 2, 389-393 (2000). Arner, P. Human fat cell lipolysis: biochemistry, regulation and clinical role. Best Pract Res Clin Endocrinol Metab 19, 471-482 (2005). Carmen, G.Y. & Victor, S.M. Signalling mechanisms regulating lipolysis. Cell Signal 18, 401-408 (2006). Wright, E.E. & Simpson, E.R. Inhibition of the lipolytic action of beta-adrenergic agonists in human adipocytes by alpha-adrenergic agonists. J Lipid Res 22, 1265-1270 (1981). Shi, Y. & Burn, P. Lipid metabolic enzymes: emerging drug targets for the treatment of obesity. Nat Rev Drug Discov 3, 695-710 (2004). Cannon, B. & Nedergaard, J. Brown adipose tissue: function and physiological significance. Physiol Rev 84, 277-359 (2004). Cypess, A.M., et al. Identification and importance of brown adipose tissue in adult humans. N Engl J Med 360, 1509-1517 (2009). van Marken Lichtenbelt, W.D., et al. Cold-activated brown adipose tissue in healthy men. N Engl J Med 360, 1500-1508 (2009). Vohl, M.C., et al. A survey of genes differentially expressed in subcutaneous and visceral adipose tissue in men. Obesity Research 12, 1217-1222 (2004). Kissebah, A.H. & Krakower, G.R. Regional adiposity and morbidity. Physiol Rev 74, 761-811 (1994). Krotkiewski, M., Bjorntorp, P., Sjostrom, L. & Smith, U. Impact of obesity on metabolism in men and women. Importance of regional adipose tissue distribution. J Clin Invest 72, 1150-1162 (1983).

58. 59. 60. 61.

62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77.

28

Chapter 1
78. 79. 80. 81. Smith, S.R., et al. Effect of pioglitazone on body composition and energy expenditure: a randomized controlled trial. Metabolism 54, 24-32 (2005). Weisberg, S.P., et al. Obesity is associated with macrophage accumulation in adipose tissue. J Clin Invest 112, 1796-1808 (2003). Rosen, E.D. & MacDougald, O.A. Adipocyte differentiation from the inside out. Nature Reviews Molecular Cell Biology 7, 885-896 (2006). Kubo, Y., Kaidzu, S., Nakajima, I., Takenouchi, K. & Nakamura, F. Organization of extracellular matrix components during differentiation of adipocytes in long-term culture. In Vitro Cell Dev Biol Anim 36, 38-44 (2000). Khan, T., et al. Metabolic dysregulation and adipose tissue fibrosis: role of collagen VI. Mol Cell Biol 29, 1575-1591 (2009). Schaffler, A., Scholmerich, J. & Buchler, C. Mechanisms of disease: adipocytokines and visceral adipose tissue--emerging role in intestinal and mesenteric diseases. Nat Clin Pract Gastroenterol Hepatol 2, 103-111 (2005). Ahima, R.S. Adipose tissue as an endocrine organ. Obesity (Silver Spring) 14 Suppl 5, 242S-249S (2006). Rosen, E.D. & Spiegelman, B.M. Adipocytes as regulators of energy balance and glucose homeostasis. Nature 444, 847-853 (2006). Leshan, R.L., Bjornholm, M., Munzberg, H. & Myers, M.G., Jr. Leptin receptor signaling and action in the central nervous system. Obesity (Silver Spring) 14 Suppl 5, 208S-212S (2006). Minokoshi, Y., et al. Leptin stimulates fatty-acid oxidation by activating AMP-activated protein kinase. Nature 415, 339-343 (2002). Hotta, K., et al. Circulating concentrations of the adipocyte protein adiponectin are decreased in parallel with reduced insulin sensitivity during the progression to type 2 diabetes in rhesus monkeys. Diabetes 50, 1126-1133 (2001). Fukuhara, A., et al. Visfatin: a protein secreted by visceral fat that mimics the effects of insulin. Science 307, 426-430 (2005). Yang, R.Z., et al. Identification of omentin as a novel depot-specific adipokine in human adipose tissue: possible role in modulating insulin action. American journal of physiology 290, E12531261 (2006). Yang, Q., et al. Serum retinol binding protein 4 contributes to insulin resistance in obesity and type 2 diabetes. Nature 436, 356-362 (2005). Jo, J., et al. Hypertrophy and/or Hyperplasia: Dynamics of Adipose Tissue Growth. PLoS computational biology 5, e1000324 (2009). Spalding, K.L., et al. Dynamics of fat cell turnover in humans. Nature 453, 783-787 (2008). Tontonoz, P., Hu, E.D. & Spiegelman, B.M. Stimulation of Adipogenesis in Fibroblasts by PparGamma-2, a Lipid-Activated Transcription Factor. Cell 79, 1147-1156 (1994). Tanaka, T., Yoshida, N., Kishimoto, T. & Akira, S. Defective adipocyte differentiation in mice lacking the C/EBPbeta and/or C/EBPdelta gene. EMBO J 16, 7432-7443 (1997). Lee, K., et al. Inhibition of adipogenesis and development of glucose intolerance by soluble preadipocyte factor-1 (Pref-1). J Clin Invest 111, 453-461 (2003). Smas, C.M. & Sul, H.S. Molecular mechanisms of adipocyte differentiation and inhibitory action of pref-1. Crit Rev Eukaryot Gene Expr 7, 281-298 (1997). Tong, Q., et al. Function of GATA transcription factors in preadipocyte-adipocyte transition. Science 290, 134-138 (2000). Yoo, E.J., Chung, J.J., Choe, S.S., Kim, K.H. & Kim, J.B. Down-regulation of histone deacetylases stimulates adipocyte differentiation. J Biol Chem 281, 6608-6615 (2006). Akune, T., et al. PPARgamma insufficiency enhances osteogenesis through osteoblast formation from bone marrow progenitors. J Clin Invest 113, 846-855 (2004). Jeon, M.J., et al. Activation of peroxisome proliferator-activated receptor-gamma inhibits the Runx2-mediated transcription of osteocalcin in osteoblasts. J Biol Chem 278, 23270-23277 (2003). Longo, K.A., et al. Wnt10b inhibits development of white and brown adipose tissues. J Biol Chem 279, 35503-35509 (2004).

82. 83.

84. 85. 86. 87. 88.

89. 90.

91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101.

102.

29

Chapter 1
103. Spinella-Jaegle, S., et al. Sonic hedgehog increases the commitment of pluripotent mesenchymal cells into the osteoblastic lineage and abolishes adipocytic differentiation. J Cell Sci 114, 2085-2094 (2001). Choy, L., Skillington, J. & Derynck, R. Roles of autocrine TGF-beta receptor and Smad signaling in adipocyte differentiation. J Cell Biol 149, 667-682 (2000). Kang, Q., et al. A comprehensive analysis of the dual roles of BMPs in regulating adipogenic and osteogenic differentiation of mesenchymal progenitor cells. Stem Cells Dev 18, 545-559 (2009). Kawaguchi, N., et al. De novo adipogenesis in mice at the site of injection of basement membrane and basic fibroblast growth factor. Proc Natl Acad Sci U S A 95, 1062-1066 (1998). Sakaue, H., et al. Requirement of fibroblast growth factor 10 in development of white adipose tissue. Genes Dev 16, 908-912 (2002). Lilla, J., Stickens, D. & Werb, Z. Metalloproteases and adipogenesis: a weighty subject. The American journal of pathology 160, 1551-1554 (2002). Christiaens, V. & Lijnen, H.R. Role of the fibrinolytic and matrix metalloproteinase systems in development of adipose tissue. Arch Physiol Biochem 112, 254-259 (2006). Selvarajan, S., Lund, L.R., Takeuchi, T., Craik, C.S. & Werb, Z. A plasma kallikrein-dependent plasminogen cascade required for adipocyte differentiation. Nat Cell Biol 3, 267-275 (2001). Chun, T.H., et al. A pericellular collagenase directs the 3-dimensional development of white adipose tissue. Cell 125, 577-591 (2006). Andarawewa K. L., R.M.-C. The Cancer Degradome Proteases and Cancer Biology Chapter 19 "New Insights into MMP function in Adipogenesis". in Springer New York (2008). Tang, W., et al. White Fat Progenitor Cells Reside in the Adipose Vasculature. Science (2008). Wilson, A., et al. Dormant and self-renewing hematopoietic stem cells and their niches. Annals of the New York Academy of Sciences 1106, 64-75 (2007). Tomiyama, K., et al. Characterization of transplanted green fluorescent protein+ bone marrow cells into adipose tissue. Stem Cells 26, 330-338 (2008). Koh, Y.J., et al. Bone marrow-derived circulating progenitor cells fail to transdifferentiate into adipocytes in adult adipose tissues in mice. J Clin Invest 117, 3684-3695 (2007). Crossno, J.T., Jr., Majka, S.M., Grazia, T., Gill, R.G. & Klemm, D.J. Rosiglitazone promotes development of a novel adipocyte population from bone marrow-derived circulating progenitor cells. J Clin Invest 116, 3220-3228 (2006). Rodeheffer, M.S., Birsoy, K. & Friedman, J.M. Identification of white adipocyte progenitor cells in vivo. Cell 135, 240-249 (2008). Suga, H., et al. Functional implications of CD34 expression in human adipose-derived stem/progenitor cells. Stem Cells Dev 18, 1201-1210 (2009). Joe, A.W., Yi, L., Even, Y., Vogl, A.W. & Rossi, F.M. Depot-specific differences in adipogenic progenitor abundance and proliferative response to high-fat diet. Stem Cells 27, 2563-2570 (2009). Roden, M., et al. Mechanism of free fatty acid-induced insulin resistance in humans. J Clin Invest 97, 2859-2865 (1996). Lam, T.K., et al. Mechanisms of the free fatty acid-induced increase in hepatic glucose production. American journal of physiology 284, E863-873 (2003). Ray, R. & Shah, A.M. NADPH oxidase and endothelial cell function. Clin Sci (Lond) 109, 217-226 (2005). Boden, G. Obesity and free fatty acids. Endocrinol Metab Clin North Am 37, 635-646, viii-ix (2008). Pilz, S. & Marz, W. Free fatty acids as a cardiovascular risk factor. Clin Chem Lab Med 46, 429434 (2008). Bamba, V. & Rader, D.J. Obesity and atherogenic dyslipidemia. Gastroenterology 132, 21812190 (2007). Carpentier, A., et al. Acute enhancement of insulin secretion by FFA in humans is lost with prolonged FFA elevation. Am J Physiol 276, E1055-1066 (1999). Hukshorn, C.J., et al. The effect of pegylated recombinant human leptin (PEG-OB) on weight loss and inflammatory status in obese subjects. Int J Obes Relat Metab Disord 26, 504-509 (2002).

104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117.

118. 119. 120.

121. 122. 123. 124. 125. 126. 127. 128.

30

Chapter 1
129. 130. Yatagai, T., et al. Hypoadiponectinemia is associated with visceral fat accumulation and insulin resistance in Japanese men with type 2 diabetes mellitus. Metabolism 52, 1274-1278 (2003). Fain, J.N., Madan, A.K., Hiler, M.L., Cheema, P. & Bahouth, S.W. Comparison of the release of adipokines by adipose tissue, adipose tissue matrix, and adipocytes from visceral and subcutaneous abdominal adipose tissues of obese humans. Endocrinology 145, 2273-2282 (2004). Maury, E., et al. Adipokines oversecreted by omental adipose tissue in human obesity. American journal of physiology 293, E656-665 (2007). Hotamisligil, G.S. Inflammation and metabolic disorders. Nature 444, 860-867 (2006). Guilherme, A., Virbasius, J.V., Puri, V. & Czech, M.P. Adipocyte dysfunctions linking obesity to insulin resistance and type 2 diabetes. Nat Rev Mol Cell Biol 9, 367-377 (2008). Choo, H.J., et al. Mitochondria are impaired in the adipocytes of type 2 diabetic mice. Diabetologia 49, 784-791 (2006). Sparks, L.M., et al. A high-fat diet coordinately downregulates genes required for mitochondrial oxidative phosphorylation in skeletal muscle. Diabetes 54, 1926-1933 (2005). Wilson-Fritch, L., et al. Mitochondrial remodeling in adipose tissue associated with obesity and treatment with rosiglitazone. J Clin Invest 114, 1281-1289 (2004). Bogacka, I., Xie, H., Bray, G.A. & Smith, S.R. Pioglitazone induces mitochondrial biogenesis in human subcutaneous adipose tissue in vivo. Diabetes 54, 1392-1399 (2005). Koves, T.R., et al. Mitochondrial overload and incomplete fatty acid oxidation contribute to skeletal muscle insulin resistance. Cell Metab 7, 45-56 (2008). Henegar, C., et al. Adipose tissue transcriptomic signature highlights the pathological relevance of extracellular matrix in human obesity. Genome Biol 9, R14 (2008). Chavey, C., et al. Matrix metalloproteinases are differentially expressed in adipose tissue during obesity and modulate adipocyte differentiation. J Biol Chem 278, 11888-11896 (2003). Maquoi, E., Munaut, C., Colige, A., Collen, D. & Lijnen, H.R. Modulation of adipose tissue expression of murine matrix metalloproteinases and their tissue inhibitors with obesity. Diabetes 51, 1093-1101 (2002). Karelis, A.D., et al. The metabolically healthy but obese individual presents a favorable inflammation profile. J Clin Endocrinol Metab 90, 4145-4150 (2005). Stefan, N., et al. Identification and characterization of metabolically benign obesity in humans. Arch Intern Med 168, 1609-1616 (2008). Brochu, M., et al. What are the physical characteristics associated with a normal metabolic profile despite a high level of obesity in postmenopausal women? J Clin Endocrinol Metab 86, 1020-1025 (2001). Salans, L.B. & Doughert.Jw. Effect of Insulin Upon Glucose Metabolism by Adipose Cells of Different Size - Influence of Cell Lipid and Protein Content, Age, and Nutritional State. Journal of Clinical Investigation 50, 1399-& (1971). Salans, L.B., Knittle, J.L. & Hirsch, J. The role of adipose cell size and adipose tissue insulin sensitivity in the carbohydrate intolerance of human obesity. J Clin Invest 47, 153-165 (1968). Jernas, M., et al. Separation of human adipocytes by size: hypertrophic fat cells display distinct gene expression. Faseb Journal 20, 1540-+ (2006). Jacobsson, B. & Smith, U. Effect of cell size on lipolysis and antilipolytic action of insulin in human fat cells. J Lipid Res 13, 651-656 (1972). Langin, D. & Arner, P. Importance of TNFalpha and neutral lipases in human adipose tissue lipolysis. Trends Endocrinol Metab 17, 314-320 (2006). Franck, N., et al. Insulin-induced GLUT4 translocation to the plasma membrane is blunted in large compared with small primary fat cells isolated from the same individual. Diabetologia 50, 1716-1722 (2007). O'Connell, J., et al. The relationship of omental and subcutaneous adipocyte size to metabolic disease in severe obesity. PLoS One 5, e9997 (2010). Drolet, R., et al. Hypertrophy and hyperplasia of abdominal adipose tissues in women. Int J Obes (Lond) 32, 283-291 (2008). Van Harmelen, V., Rohrig, K. & Hauner, H. Comparison of proliferation and differentiation capacity of human adipocyte precursor cells from the omental and subcutaneous adipose tissue depot of obese subjects. Metabolism 53, 632-637 (2004).

131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141.

142. 143. 144.

145.

146. 147. 148. 149. 150.

151. 152. 153.

31

Chapter 1
154. 155. 156. Tchkonia, T., et al. Fat depot-specific characteristics are retained in strains derived from single human preadipocytes. Diabetes 55, 2571-2578 (2006). Kirkland, J.L., Tchkonia, T., Pirtskhalava, T., Han, J.R. & Karagiannides, I. Adipogenesis and aging: does aging make fat go MAD? Experimental Gerontology 37, 757-767 (2002). Dieudonne, M.N., Pecquery, R., Leneveu, M.C. & Giudicelli, Y. Opposite effects of androgens and estrogens on adipogenesis in rat preadipocytes: evidence for sex and site-related specificities and possible involvement of insulin-like growth factor 1 receptor and peroxisome proliferator-activated receptor gamma2. Endocrinology 141, 649-656 (2000).

32

Chapter 2

Chapter 2
Retinoid-related orphan receptor gamma controls adipocyte hyperplasia and insulin sensitivity in obesity
Bettina Meissburger1, Jozef Ukropec2, Eva Roeder3, Nigel Beaton1, Evangeline Chew1, Matthias Geiger1, Burcak Civan1, Daniel Teupser4, Jacquelien Hillebrand5, Nori Geary5, Iwar Klimes2, Dan Littman6, Peter P. Nawroth3, Daniela Gasperikova2, Gottfried Rudofsky3 and Christian Wolfrum1
Swiss Federal Institute of Technology, ETH Zrich, Institute for Molecular Systems Biology, Wolfgang-Pauli Str. 16, 8093 Zrich, Switzerland Institute of Experimental Endocrinology, Slovak Academy of Science, Slovak Academy of Sciences, Vlarska 3, 83306 Bratislava, Slovakia Department of Medicine I and Clinical Chemistry, University of Heidelberg, Im Neuenheimer Feld 410, 69120 Heidelberg, Germany
4 3 2 1

Universittsklinikum Leipzig, Institut fr Laboratoriumsmedizin Klinische Chemie und Molekulare Diagnostik, Liebigstrae 27, 04103 Leipzig, Germany

Swiss Federal Institute of Technology, ETH Zrich, Institute of Animal Science, Schorenstr. 16, 8603 Schwerzenbach, Switzerland

33

Chapter 2

Abstract
Obesity is a risk factor for the development of metabolic complications such as type 2 diabetes. However, it is currently unclear why only a subset of the obese population develops secondary metabolic disorders. Here we identify the transcription factor retinoid-related orphan receptor (ROR) as a regulator of adipogenesis, thereby controlling insulin sensitivity in obesity. ROR inhibits adipogenesis via its new target gene matrix metallopeptidase 3 (MMP3). In vivo differentiation of injected preadipocytes from ROR-deficient mice is enhanced compared to wildtype preadipocytes. Consequently, obese ROR-/- mice show adipocyte hyperplasia accompanied by decreased adipocyte sizes. These small adipocytes are highly insulin sensitive, leading to improved control of circulating free fatty acids, which are a major cause of insulin resistance. Ultimately, ROR-/- mice are protected from hyperglycemia and insulin resistance during obesity. In obese human subjects, ROR expression is inversely correlated with the adipogenic potential as well as insulin sensitivity. Taken together, we identify ROR as a novel regulator of adipocyte differentiation, thereby influencing insulin sensitivity in obesity.

34

Chapter 2

Introduction
Obesity is a major risk factor for the development of type 2 diabetes and the metabolic syndrome. Even though there is a clear link between obesity and insulin sensitivity, the degree of insulin resistance in obese subjects can vary enormously1-2. One possible reason is the variability of adipose tissue plasticity, which determines whether excess energy intake can be buffered safely in adipose tissue without leading to detrimental effects3. Adipose tissue mass increases during obesity via two distinct mechanisms, increasing cell number (hyperplasia) and/or increasing adipocyte size (hypertrophy)4-5. Hyperplasia occurs via de novo differentiation of preadipocytes, which reside in the stromal-vascular fraction (SVF) of adipose tissue6. The balance between hyperplasia and hypertrophy strongly influences the metabolic outcome of obesity. While smaller cells retain insulin sensitivity and normal function, large hypertrophic cells are insulin resistant and change their secretory profile towards pro-inflammatory adipocytokines710

. As a result, lipolytic release of free fatty acids is enhanced in large adipocytes

ultimately leading to cytotoxic fatty acid accumulation in extra-adipose tissue11. Shifting the balance towards hyperplasia replenishes the pool of small, functional adipocytes, thus ameliorating the metabolic consequences of obesity. A wellestablished pharmacological example for this mechanism is the clinical use of glitazones activating PPAR12-13, which leads to an increase in adipose tissue mass concomitant with a shift towards smaller, insulin sensitive adipocytes. An important regulator of the adipogenic process is the proteolytic remodeling of the extracellular matrix, as preadipocytes have to undergo cell shape transition during differentiation14. Several classes of metallopeptidases (MMPs) have intriguing and opposing effects in the adipogenic process, suggesting that they have distinct roles and cleave different substrates during fate decisions and/or terminal differentiation15. For example, the secretion of MMP2 and MMP9 increases during adipocyte differentiation in both human adipocytes and the 3T3-F442A cell line. Additionally, global inhibition of MMP activity inhibits differentiation of preadipocytes and adipose tissue development16-17. On the other hand, MMP3 seems to have an inhibitory effect on adipogenesis. MMP3-deficient mice show accelerated adipogenesis during mammary

35

Chapter 2

gland involution suggesting a negative regulatory role for MMP3 in adipocyte differentiation18. Retinoid-related orphan receptor (ROR) is a nuclear receptor with the typical conserved structure of an N-terminal (A/B) domain followed by the DNA-binding domain and a C-terminal ligand-binding domain, which are interconnected by a hinge domain. RORs bind as monomers to the consensus half-site PuGGTCA19. Recently, several ligands have been proposed to regulate ROR transcriptional activity such as oxysterols or the benzenesulfonamide T090131720-22. ROR has two isoforms, the main isoform being expressed in a variety of tissues such as adipose tissue, liver, muscle and kidney, whereas expression of RORt is restricted to cells specific to the immune system. RORt has been intensively studied in the context of IL-17 producing T-helper cells23-24. The role of ROR in other tissues, however, is less clear25. In cell culture studies, ROR regulates metabolic genes in skeletal muscle cells26. With regard to adipose tissue, single nucleotide polymorphisms of ROR have been linked to fat accumulation in cattle27. During in vitro adipogenesis ROR expression is upregulated in late stages of differentiation28, however, a functional role for ROR in adipose tissue plasticity has yet to be identified. In this study, we investigate the mechanistic role of ROR in adipocyte differentiation and show that ROR is an important negative regulator of adipogenesis, both in mice and humans. By controlling fat cell development ROR influences the progression of obesity-associated insulin-resistance.

36

Chapter 2

Materials and Methods

Animals All mouse models were maintained in the C57Bl/6J background and kept on a 12-h/12-h light/dark cycle in a pathogen-free animal facility. Groups of mice were fed a high-fat diet (Provimi Kliba AG) containing 60% fat for 8 weeks. O2, CO2, food and water intake were simultaneously determined for four mice per experiment in an Oxymax metabolic cage system (Columbus Instruments). ROR global knockout mice29 were crossed with C57Bl/6-Lepob or C57Bl/6-Tg(UBC-GFP) (Jackson). Plasmids ROR cDNA was amplified from murine adipocyte cDNA using the following primers: ROR-fw GCT AGC GAC AGG GCC CCA CAG AG ACA, ROR-rv TTG TCC TTC CTC CAG ATC AC. ROR containing an HA- and flag-tag was subcloned into pCR2.1 (Invitrogen) and subsequently into pcDNA3 (Invitrogen) for transfection or pCCL.sin.PPT.hPGK. IRES.GFP.PRE for lentiviral production. For all luciferase assays, the pGL3-enhancer vector containing firefly luciferase was used. Different parts of the MMP3promoter as described above or 3kb of the ROR promoter or 700bp of the PPAR promoter as described before30 were cloned from murine C57Bl/6J genomic DNA. For PPAR activity, a luciferase reporter with the PPAR response element PPRE was used as described before31. Single siRNA or siRNA pools were purchased from Dharmacon. Cell culture 3T3-L1 preadipocytes were cultured on collagen-coated plates in high-glucose Dulbeccos modified Eagles medium (Invitrogen) supplemented with 10% fetal bovine serum and penicillin/streptomycin. Cells were transfected using Fugene (Roche) according to the manufacturers protocol. For stable transfection cells were kept on geneticin (500 g/ml) for at least two weeks. Adipogenesis was induced by treating postconfluent 3T3-L1 preadipocytes with insulin, dexamethasone, isobutyl-

methylxanthine and rosiglitazone if indicated, as described previously32. Where indicated, the MMP3 inhibitor II (Merck) was added 2 days before and after induction of differentiation. Lipid incorporation was monitored by fluorescent staining.

37

Chapter 2

Automated analysis of adipocyte differentiation Differentiated cells were fixed with 5% formaldehyde prior to staining with BODIPY for lipid droplets, Hoechst for nuclei and Syto60 for cytosolic staining (all Invitrogen). 25 pictures per well were taken with an automated microscope imaging system (CellWorx). Pictures were analyzed using Cell Profiler Software (Suppl Fig. 2a). Lentiviral infection Lentiviral stocks were prepared essentially as described previously33. Briefly, HEK293T cells were transfected using lipofectamine with lentiviral overexpression or control vector and the packaging vectors pMD2.G and psPAX2. Virus containing medium was collected 24 h after changing transfected cells to high BSA (1.1g/100ml) D-MEM supplemented with 10% fetal bovine serum and penicillin/streptomycin. For functional assays of 3T3-L1 adipocytes, 3T3-L1 cells were infected for 24 h with virus containing medium supplemented with polybrene (8 g/ml) two days after induction of differentiation. Differentiation was continued until day 8. Western Blotting Whole cell lysates were subjected to 12% SDS-PAGE and western blotting as previously described32. Adipocyte fatty acid binding protein was detected with an antibody raised against recombinant mouse A-FABP (gift of N. Haunerland), ROR was detected using a mouse anti-HA-tag antibody (Covance) and cell extracts were blotted against -tubulin (mouse anti -tubulin antibody, Sigma) as a loading control. Luciferase Assay, Chromatin Immunoprecipitation Luciferase assays were performed as described before32. Briefly, 3T3-L1 preadipocytes at 60% confluence were transfected with the reporter genes, renilla luciferase for normalization and either pcDNA3-ROR or empty pcDNA3. 48 hours after transfection luciferase activity was measured using the Dual Luciferase Detection Kit (Promega). For ChIP Assays 3T3-L1 cells were infected with ROR or control lentivirus at 60% confluence. Day 6-8 after transfection chromatin immunoprecipitation was performed using the chromatin IP Assay Kit (LucernaChem) according to the manufacturers protocol. Precipitated DNA was subjected to real-time PCR using the following primers:

38

Chapter 2

RRE1-fw AGC CAC TAC CAG ATC CTT GC RRE1-rv GGA AAA ACG GGA GTG AAA GG, RRE2-fw TGC TAT GTG CCT TTC AGT CC,RRE2-rv CTA TCA TTC CCC GCT TTT CA Nuclear run-on assay The nuclear run on assay was performed according to the method described by Reeder and Roeder34. Real-Time PCR mRNA was isolated and transcribed into cDNA using the Multi-MACS cDNA kit (Miltenyi). mRNA expression was assessed by real-time PCR using SybrGreen (Invitrogen) according to the manufacturers protocol. mRNA expression was normalized to gapdh. The following primer pairs were used (tan= 60C): a-fabp-fw ATG AAA TCA CCG CAG ACG ACA G, a-fabp-rv GCC TTT CAT AAC ACA TTC CAC CAC, c/ebpfw GGT CGA GCG CAA CAA CAT C, c/ebp-rv CTC GGG CAG CTG CTT GAA CAA, c-fos-fw CAT CGG CAG AAG GGG CAA AGT A, c-fos-rv GTC AGC TCC CTC CTC CGA TTC C, c-jun-fw ACC GCC GCG CCA AGA ACT CG, c-jun-rv AAC TGG GTG GGG GGT CTG TGT AG, ror-fw GCT GTG GGG TAG ATG GGA TA, ror-rv CCC ATG GGG AAA TAC AAT GA, mmp3-fw ACA TGG AGA CTT TGT CCC TTT TG, ror-fw CTG GCC GTG GGG ATG TCT CG, ror-rv AGG CTG CTG CTG GTG GTC TCG, ror-fw GAT CGA TCG AAT TGC ACA GAA CAT, ror-rv GCC TCC CTG GAC TTG CTT TGA TAC mmp3-rv TTG GCT GAG TGG TAG AGT CCC, gapdhfw CTG ACG TGC CGC CTG GAG AAA, gapdh-rv CCG GCA TCG AAG GTG GAA GAG T For expression of Ror and Mmp3 in human samples TaqMan Gene Expression Assays were used according to the manufacturers protocol. Microarrays Total RNA was extracted using Trizol (Invitrogen) according to the manufacturers protocol. Genomic DNA was digested using the DNA-free kit (Ambion) and RNA was labeled with either Affimetrix labelling system (human samples) or with One-Colour RNA Spike-In Kit (Agilent) for mouse samples. Labelled cRNA was hybridized to human genome U133 arrays (Affimetrix) or 44k whole mouse genome arrays (Agilent).

39

Chapter 2

Glucose uptake and lipolysis Glucose uptake and lipolysis was determined essentially as described previously32,35. Briefly, adipocytes were preincubated with 20 nM (primary adipocytes) or 50 nM (3T3L1 adipocytes) insulin for 10 min. Cells were incubated for 1 h with 1mM glucose spiked with D-[1-14C]glucose in Krebs-Ringer buffer adding 1% BSA for primary adipocytes. Supernatant of 3T3-L1 adipocytes was collected, cells were washed, lysed and 14C incorporation as well as total protein content by BCA Assay (Sigma) was measured. Primary adipocytes were spun through dinoyl-phthalate oil, washed and the aqueous phase was collected. Osmium-tetroxide fixed primary adipocytes36 were counted for normalization using a Coulter Cell Counter. Glycerol release in the supernatant was measured to assess lipolysis using glycerol reagent (Sigma) according to the manufacturers protocol. In vivo differentiation of SVF Stromal-vascular fraction from subcutaneous and visceral fat was prepared as described before37. Briefly, fat tissue was dissected, minced and incubated in collagenase type II for 1 h at 37C. Pelleted stomal-vascular fraction was strained through a 40 m net, erythrocytes were lysed and 106 cells were resuspended in 100 l of Matrigel [BD]. Cells were injected subcutaneously into a skin fold of the neck. After 4-6 weeks Matrigel pads were excised and treated as described below for the determination of adipocyte size. From each pad pictures of 3 full sections were taken and adipocyte numbers as well as the number of nuclei were determined automatically with Cell Profiler Software. Adipocyte size Adipose tissue was fixed in 5% paraformaldehyde over night before paraffin embedding and sectioning to 10m slices. H&E staining of sections was performed according to standard procedures38. Microscopic pictures were taken and cell size was analyzed using Cell Profiler Software. Fat pads were either weighted after excision or measured with an animal CT-Scanner (LaTheta). At least 2000 adipocytes were measured per animal to determine adipocyte size.

40

Chapter 2

Serum Measurements Blood glucose was measured with a glycometer (Contour). Insulin was determined with the Rat/Mouse Insulin ELISA kit (Crystal Chem). Triglycerides were measured with the Trig/GB Kit against the C.f.a.s. standard (Roche/Hitachi) and free fatty acids were measured with the NEAFA-HR kit (Wako). For serum interleukins a micro-bead based assay system (Luminex) was used. Statistical Analysis Results are given as mean +/- standard deviation. Statistical analyses were performed using a two-tailed Students t test. Linear regression was calculated using Graph Pad Prism.

41

Chapter 2

Results
ROR regulates adipogenesis In order to identify candidate genes that regulate adipose tissue plasticity, we analyzed transcription profiles of visceral adipose SVF from lean and obese patients. The transcription factor Ror was found to be the highest upregulated gene in the state of obesity (Suppl. Tab. I). Increased expression of Ror was confirmed by quantitative mRNA analysis in visceral adipose tissue of obese patients versus lean patients (Fig. 1.1a). As this factor is highly expressed in the preadipocyte containing stromalvascular fraction (SVF), we analyzed whether ROR plays a role in regulating adipose tissue plasticity. To this end, we performed adipocyte differentiation assays using the preadipocyte cell line 3T3-L1. During differentiation mRNA expression of Ror is high in confluent 3T3-L1 preadipocytes. Upon induction of differentiation, Ror exhibits a sharp decline in expression suggesting that ROR might influence the process of differentiation (Figure 1.1b). We next expressed ROR in 3T3-L1 preadipocytes and analyzed the effect on adipogenesis. ROR overexpression in 3T3-L1 cells decreases the amount of differentiated adipocytes as well as expression of the mature adipocyte marker adipocyte fatty acid binding protein (A-FABP) (Fig. 1.2c,d, Suppl. Fig.1a,b). This diminished adipogenesis is retained, albeit to a lesser extent, when rosiglitazone is added, suggesting that ROR acts upstream of PPAR (Fig. 1.2c,d). Conversely, siRNA mediated knockdown of ROR increases the adipogenic potential of 3T3-L1 preadipocytes (Fig. 1.2e,f Suppl. Fig. 1c). No changes were found in the expression of closely related family members ROR and ROR, suggesting that the effect of ROR is not due to compensatory regulation of other family members (Suppl. Fig. 1d).

42

Chapter 2

a
mRNA expression

normalzied Ror

22.5 20.0 17.5 15.0 12.5 10.0 7.5 5.0 2.5 0.0 BMI <20

***

BMI >35

normalized mRNA expression

3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -48 -24 0 24 48 72 hours 96 120 144 168

Ror

Mmp3

A-Fabp

induction of differentiation

Figure 1.1 ROR inhibits early adipogenesis. a) Ror expression in visceral whole fat samples of lean (BMI<20) and obese (BMI>35) subjects (n = 9). b) Expression profile of Ror, Mmp3 and A-Fabp during 3T3-L1 adipogenesis.

43

Chapter 2

c
45

**

- rosiglitazone

% differentiation

40 35 30 25 20 15 10 5 0 pcDNA3 ROR - rosiglitazone pcDNA3 ROR + rosiglitazone

***
pcDNA3 + rosiglitazone ROR

pcDNA3

ROR

d A-FABP -Tubulin

pcDNA3

ROR A-FABP -Tubulin

pcDNA3

ROR

- rosiglitazone

+ rosiglitazone

e
70

**

**

% differentiation

60 50 40 30 20 10 0 scramble siRNA 1 siRNA2

siRNA scramble

siRNA2 ROR

siRNA1 ROR

f A-FABP -Tubulin

siRNA ctrl siRNA1 siRNA2

Figure 1.2 ROR inhibits early adipogenesis. c) Fluorescent staining (blue: nuclei, red: cytosol, green: lipid droplets, scale: 100 m) of 3T3-L1 adipocytes overexpressing ROR or empty vector at day 7 of differentiation with or without rosiglitazone. d) Whole cell extracts of differentiated 3T3-L1 adipocytes were blotted against the mature adipocyte marker adipocyte fatty acid binding protein (A-FABP) using -Tubulin as a loading control. e, f) Fluorescent staining and A-FABP western blot of differentiated 3T3-L1 cells transfected with control siRNA or siRNA against ROR three days before induction of differentiation. Values represent means +/- SD, *p 0.05, **p 0.01, ***p 0.001.
44

Chapter 2

To further pinpoint at which stage ROR interferes with the differentiation process, we next analyzed expression of key regulatory adipogenic genes. Early and late adipogenic genes such as c-Jun, c-Fos, C/ebp and , PPAR and A-Fabp were repressed by ROR demonstrating that ROR exerts its inhibitory effect during the initial steps of adipogenesis (Fig. 1.3g-l). To confirm that ROR acts upstream of PPAR we measured PPAR transcriptional activity in differentiating cells overexpressing ROR or control vector. PPAR activity in ROR transfected cells was suppressed in the first two days after induction of differentiation (Fig. 1.3m). On the other hand, activity of the ROR promoter was not affected by the PPAR agonist rosiglitazone suggesting that PPAR does not regulate ROR expression (Suppl. Fig. 2b). As Rev-erb is known to modulate transcription of PPAR30, we tested whether ROR directly influences PPAR expression. ROR did not repress PPAR promoter activity in a luciferase assay, whereas promoter activity of the known target gene IL-17 was induced by ROR (Fig. 1.3n). Taken together, these findings demonstrate that ROR tightly regulates adipocyte differentiation by regulating early phase genes, which in turn leads to an indirect reduction in PPAR expression and activity.

45

Chapter 2

normalized mRNA expression

normalized mRNA expression

6 5 4 3 2 1 0

c-Jun
control ROR

*
1.5

c-Fos
control

ROR

1.0

0.5

0.0 day 0 day 1 day 2 day 3 day 6

day 0

day 1

day 2

day 3

day 6

i
normalized mRNA expression

C/ebp
0.4 0.3

normalized mRNA expression

j
control ROR

C/ebp
4 3 2 1
control ROR

*
0.2 0.1 0.0 day 0 day 1 day 2 day 3 day 6

*
0 day 0 day 1

*
day 2 day 3 day 6

k
normalized mRNA expression

normalized mRNA expression

2.5 2.0 1.5 1.0 0.5 0.0 day 0

Ppar
control ROR

A-Fabp
2.0 1.5 1.0 0.5 0.0
control ROR

**

** *
day 1

* *
day 2 day 3 day 6

* * *
day 0

*
day 1 day 2 day 3 day 6

normalized luciferase activity

m
PPRE luciferase activity

25

control ROR

***

5 4 3 2 1 0

***

control ROR

15

***

**

day 0

day 1

day 2

IL-17

PPAR p rr

PPRE

Figure 1.3 ROR inhibits early adipogenesis. g-l) qPCR for adipogenic genes c-Jun, c-Fos, C/ebp, C/ebp, Ppar and A-Fabp normalized to Gapdh during differentiation of 3T3-L1 cells overexpressing ROR or empty vector. m) firefly luciferase activity of the PPAR response element (PPRE) normalized to renilla luciferase before, one and two days after induction of differentiation in 3T3-L1 cells overexpressing ROR or empty vector. n) firefly luciferase activity of the IL-17 promoter (positive control), PPAR promoter and PPRE (negative control) normalized to renilla luciferase in 3T3-L1 cells overexpressing ROR or empty vector (n=4). Values represent means +/- SD, *p 0.05, **p 0.01, ***p 0.001.

46

Chapter 2

The effect of ROR on adipogenesis is mediated by MMP3 To identify new target genes of ROR that are involved in the repression of adipogenesis we analyzed transcription profiles of 3T3-L1 preadipocytes stably or transiently overexpressing ROR (Suppl. Tab. II). Genes upregulated by ROR and highly expressed in murine SVF were chosen for a functional siRNA screen in order to test their potential impact on adipogenesis (Fig. 2.1a). Soley knockdown of matrixmetalloproteinase 3 (MMP3) was able to increase differentiation of 3T3-L1 preadipocytes. We confirmed Mmp3 mRNA to be upregulated by transient or lentiviral overexpression of ROR (Fig. 2.1b), suggesting that ROR controls adipogenesis by regulation of this gene. Interestingly, expression of Mmp3 during early 3T3-L1 adipogenesis closely follows the expression profile of Ror (Fig. 1.1b), indicating that Mmp3 expression is tightly controlled by ROR in preadipocytes. To elucidate whether MMP3 is a direct target gene of ROR we screened the MMP3 promoter for potential ROR response elements (RREs). The promoter sequence contains two potential RREs; the first RRE matches the consensus half-site PuGGTCA and is preceded by an AT-rich region, whereas the second RRE has the half-core motif PuGGTAA. Using different luciferase constructs we showed that ROR-induced transcriptional activation of the MMP3 promoter depends on the first but not on the second RRE and reaches levels similar to the known ROR target gene IL-17 (Fig. 2.2c). These findings were corroborated by ChIP assays demonstrating binding of ROR to the first MMP3 promoter site (Fig. 2.2d). Additionally, active transcription of the Mmp3 gene is enhanced through ROR overexpression in a nuclear run-on assay (Fig. 2.2e).

47

Chapter 2

a 90 80

***

***

% differentiation

70 60 50 40 30 20 10 0
bl R e o A r b A Ac a t rh c n g 1 A ap rh 3 0 ga C p6 xc l Eg 9 r3 Fa G p c G h1 cn G t1 pr 4 I fn M g fs M d4 m M p3 s N t1 r xp h Pb 3 p2 S l Rtp c2 3 S l 8a 2 c4 St a1 m n3 ra m

sc

***
normalized Mmp3 expression

5 4 3 2 1 0 control ROR
transie nt

**

control ROR
le ntiv irus

Figure 2.1 MMP3 is a target gene of ROR that inhibits adipogenesis. a) siRNA screen of potential ROR target genes. 3T3-L1 cells were transfected with siRNA pools, differentiated and fluorescently stained to assess the percentage of differentiated cells. b) Expression of Mmp3 in 3T3-L1 preadipocytes which either transiently overexpress ROR and empty vector or which are infected with ROR and control virus, respectively. Values represent means +/- SD, *p 0.05, **p 0.01, ***p 0.001.

48

Chapter 2
c

-2460 bp

RRE1

-1697 bp RRE2 RRE2

luciferase luciferase luciferase luciferase

II III

RRE2

IV
25

***

control

***

ROR

luciferase activity

20 15 10 5 0 MMP3prI MMP3prII MMP3prIII MMP3prIV IL17

d
35

***

DNA enrichment ROR

anti-HA Ab input control ROR

IgG ctrl. Ab ROR

30 25 20 15 10 5 0 control ROR

control

ROR Ror Mmp3 Lrh1 Ppar C/ebp Pref1 C/ebp Gapdh

Figure 2.2 MMP3 is a target gene of ROR that inhibits adipogenesis. c) MMP3 promoter constructs with potential ROR response elements (RREs) cloned into a luciferase reporter vector. Luciferase assay in 3T3-L1 cells overexpressing ROR or control vector 48 h after transfection normalized to renilla luciferase activity. (n=4) d) qPCR of RRE1 from 3T3-L1 genomic DNA which was chromatin-immunoprecipitated 6 days after infection with ROR or control lentivirus. (n=3) e) Nuclear run-on assay for transcriptional activity of genes in 3T3-L1 cells infected with control or ROR overexpressing virus. Values represent means +/- SD, ***p 0.001.

49

Chapter 2

To assess whether the effect of ROR on adipogenesis is mediated by MMP3, we added a specific MMP3 inhibitor during adipocyte differentiation. Inhibition of MMP3 enhances the number of differentiated adipocytes in 3T3-L1 adipogenesis (Suppl. Fig. 2c). Moreover, inhibition of differentiation by ROR can be partially rescued with the MMP3 inhibitor (Fig. 2.3f), suggesting that inhibition of adipogenesis by ROR is at least partly mediated through induction of MMP3. Taken together, these results demonstrate that MMP3 is a direct target of ROR and that regulation of adipogenesis by ROR is mediated through MMP3.

f ***
% normalized differentiation
125 100 75 50 25 0 control

* **

ROR

ROR + MMP3 inhib

control

ROR

ROR + MMP3 inhib

Figure 2.3 MMP3 is a target gene of ROR that inhibits adipogenesis. f) Fluorescent staining of differentiated 3T3-L1 adipocytes expressing ROR or control vector during differentiation treated with or without 20 M of MMP3 inhibitor 2 days before and after induction of differentiation. (n=4) Scale: 100 m. Values represent means +/- SD, *p 0.05, **p 0.01, ***p 0.001.

50

Chapter 2

ROR is a key factor regulating adipocyte differentiation in vivo To assess whether the effect of ROR on adipogenesis can be recapitulated in vivo, we implanted murine adipose SVF containing adipocyte precursor cells into subcutaneous fat pads. SVF from ubiquitously GFP-expressing mice resuspended in matrigel and injected into the subcutaneous layer of the neck could be retrieved as a distinct pad 4-6 weeks after transplantation. Cells in these pads stained positive for GFP and had a distinct adipocyte structure (Fig. 3.1a). Little or no differentiation into mature adipocytes was observed when acceptor mice were fed a normal chow diet (data not shown). On the contrary, injected pads from mice fed a high-fat diet contained mature adipocytes, which were positive for GFP as well as for the adipocyte marker A-FABP (Fig. 3.1a). Taken together, these results demonstrate that preadipocytes derived from a donor mouse can undergo adipogenesis in vivo without host cell infiltration under these conditions. To determine the effect of ROR on the process of adipogenesis we next injected GFP positive SVF of wildtype and ROR knockout animals into C57Bl/6J acceptor mice. After 4 weeks of high-fat diet the excised pads with ROR-/- cells contained more mature adipocytes than pads with cells from control animals, demonstrating that the absence of ROR potentiates adipocyte differentiation in a cell autonomous fashion in vivo (Fig. 3.1a, right panel). Quantification of adipocyte numbers by high content imaging demonstrated a 2.5 fold increase of de novo differentiated adipocytes in the absence of ROR (Fig. 3.1b).

51

Chapter 2

15

***

% adipocytes/nuclei

10

***
5

0 wildtype ROR -/-

Figure 3.1 Loss of ROR lead leads to increased adipocyte formation in vivo. a) Injection of SVF of different donor mice (as indicated) into the scapular subcutan taneous region of C57Bl/6J mice. Fat pads wer were excised after 6 or 4 weeks of HFD feeding ing and stained as indicated. b) Quantification ation of adipocyte differentiation in SVF tran transplants. (n=5) Values represent means +/+/ SD, ***p 0.001.

52

Chapter 2

Loss of ROR leads to increased formation of adipocytes in mouse models of obesity As we originally found ROR to be upregulated in obese human subjects, we next examined whether the altered differentiation capacity of ROR-/- preadipocytes affected fat cell size within the adipose tissue in the state of obesity. To accomplish this, we challenged ROR knockout mice either by feeding them a high fat diet (DIO, 60% calories derived from fat) or by crossing them into the leptin deficient ob/ob background. Mean fat cell size from adipose tissue sections was determined in ROR-/and control animals via automated image analysis (Fig. 3.2c). To estimate the number of adipocytes we also quantified whole fat volume using CT scanning and fat pad excision (Fig. 3.2d, Suppl. Fig. 3a). While adipocyte size was reduced up to two fold in subcutaneous and visceral adipose tissue of all mouse models, fat pad weight was only altered mildly in obese ROR-/- mice. This data demonstrates that loss of ROR in mice with diet-induced or genetic obesity leads to an increase in the number of adipocytes. To confirm that the observed changes in adipocyte size were caused by altered differentiation capacity rather than altered adipocyte function, functional assays were performed in 3T3-L1 adipocytes infected with a lentivirus overexpressing ROR. Neither glucose uptake nor lipolysis was altered by ROR (Suppl. Fig. 3b, data not shown). Differences in adipocyte size were not due to altered energy balance, as 12week old ROR-/- mice on a normal chow diet did not exhibit altered food intake, movement or oxygen consumption in metabolic cage experiments (data not shown). As MMP3 is a direct target gene of ROR and at least partially responsible for the alteration of adipogenesis, we compared mRNA expression of MMP3 in visceral SVF of 12 week old ROR-/- and control animals on a normal chow diet. MMP3 expression was reduced 2.4 fold in the knockout animals (Fig. 3.2e), demonstrating that MMP3 is regulated by ROR in vivo. Taken together, these studies show that ROR regulates adipogenesis in vivo, leading to an increased adipocyte number with reduced adipocyte cell size.

53

Chapter 2

mean adipocyte size [ m2]

20000

+/+ -/-

***
***

***
10000

**

***

***

t bc u su e ob / ob vi sc

al

bc u

bc u

al

er

sc

vi

su

su

vi

sc

D IO

IO

al

al

al

al

al

ob

fe

fe

subcutaneous

+/+

200 m

-/-

200 m

visceral

al

/o b

IO

IO

er

er

al

**
+/+ -/-

*
e
relative Mmp3 expression

fat pad weight [g]

5 4 3 2 1 0

**

0.5 0.4 0.3 0.2 0.1 0.0 wildtype ROR -/-

**

**

al

al

bc ut

bc ut

bc ut su e ob / ob

er

er

sc

su

su

sc

vi

vi

D IO

D IO

D IO

D IO

al

al

al

al

al

ob /

fe

fe

Figure 3.2 Loss of ROR leads to increased adipocyte formation in vivo. c) Mean adipocyte size of ROR-/- and wildtype mice crossed into the ob/ob background or fed a high fat diet, insets are fat sections from representative samples of wildtype and ROR-/- female DIO mice (n = 4-7). d) Fat pad weights of excised subcutaneous and visceral adipose tissue from ROR knockout and wildtype mice in the ob/ob background or fed a high fat diet for 8 weeks (n = 5-10). e) Mmp3 expression in SVF of ROR-/- mice. (n=5) Values represent means +/- SD, *p 0.05, **p 0.01, ***p 0.001.
54

al

ob

vi

sc

er

al

Chapter 2

Loss of ROR protects against obesity induced insulin resistance To examine the metabolic consequences of altered fat cell size in obese mouse models, we measured standard metabolic parameters in ROR-/- and control animals. Mean body weight was slightly reduced for male and female knockout animals on a high-fat diet (Fig. 4a). In all models ROR-/- mice were protected from obesity induced hyperglycemia (Fig. 4b) accompanied by significantly lower insulin levels in the fasted as well as in the fed state (Fig. 4c, Suppl. Fig. 4a). In an insulin tolerance test, male ROR-/- animals showed improved insulin sensitivity compared to control mice after 8 weeks of high fat diet (Fig. 4d). Besides whole body insulin sensitivity, we also tested insulin-sensitivity of isolated primary adipocytes from obese ROR-/- mice and control littermates. As expected from the observed differences in adipocyte size, insulinmediated suppression of lipolysis normalized to cell number was more pronounced in small adipocytes from ROR-/- mice in comparison to larger adipocytes from wildtype mice (Fig. 4e). In line with this, concentrations of circulating free fatty acids were decreased in obese ROR-/- mice (Fig. 4f, Suppl. Fig. 4b). In contrast, triglyceride concentrations were mildly altered only in male high fat diet animals (Suppl. Fig. 4c). As ROR has been implicated in immunological processes, we quantified different markers of inflammation associated with insulin resistance, including TNF, INF, IL-1, IL-2, IL-12 and IL-6. Serum inflammation makers as well as macrophage infiltration, measured by Emr-1 expression in the adipose tissue, were not significantly altered in these mice (Suppl. Fig. 4d, data not shown). Taken together, our results demonstrate that ROR leads to enhanced adipogenesis in vivo, causing an increased number of mature adipocytes. As a consequence, lipids are stored in small, insulin sensitive adipocytes with intact lipolytic control, which prevents systemic free fatty acid overload and diet-induced insulin resistance.

55

Chapter 2

50

fasting blood glucose [mM]

*
15

** *

weight [g]

40 30 20 10 0
+/ +

10

**

0
fe +/ m + al e D m IO al -/e ob /o b m +/ al + e ob /o b -/+/ + D IO D IO fe -/m al e D IO

-/ob /o b m +/ al + e ob /o b -/-

+/ +

IO

-/-

D IO

D IO

DI O

m al e

m al e

fe m

fe

c
9 8 7 6 5 4 3 2 1 0
+/ + D IO

al e

* *** ***

d
% inital blood glucose
125 100 75

m al e

m al e

al e

al e

+/+ -/** *

insulin [ng/ml]

50 25 0

+/ +

ob /o b m +/ al + e ob /o b -/-

-/-

-/-

30

60

90

120

150

D IO

D IO

m al e

m al e

al e

al e

e
% basal lipolysis/ cell number

m al e

fe m

fe

D IO

time [min]

basal
120 100 80 60 40 20 0

20nM insulin

f
2.0

***
free fatty acids [mM]

***
1.5 1.0 0.5 0.0

***

**

Figure 4 Loss of ROR protects against diet-induced insulin resistance by inducing adipocyte hyperplasia. a) Body weight of ROR-/- and wildtype mice (12 weeks) crossed into the ob/ob background or fed a high fat diet (DIO) for 8 weeks (n = 5-10). b, c) Fasting blood glucose and serum insulin of ob/ob and high fat diet ROR-/- and wildtype mice (n = 5-10). d) Insulin tolerance test with ROR-/- and control mice after 8 weeks of high fat diet injected i.p. with 0.75U/kg insulin (n = 5-10). e) In vitro lipolysis of isolated adipocytes from ob/ob ROR-/- and wildtype mice treated with or without insulin normalized to cell number and set to 100% for basal lipolysis (n = 4). f) Fasting free fatty acid serum concentrations of ob/ob and high fat diet ROR-/- and wildtype mice (n = 5-10). Values represent means +/- SD, *p 0.05, **p 0.01, ***p 0.001.

56

D fe IO m -/al e DI O fe +/ m + al e D m IO al -/e ob /o b m +/ al + e ob /o b -/-

+/+

-/m al e

D IO

m al e

+/ +

Chapter 2

ROR expression in obese subjects predicts adipocyte size and insulin sensitivity Since we have shown that ablation of ROR serves as a protective mechanism against the development of obesity-associated insulin resistance in mice, we next analyzed the link between Ror expression and insulin resistance in human obesity. To this end, we obtained subcutaneous fat biopsies from 21 obese patients (11 female, 10 male) with body mass indices (BMI) of 32-56 kg/m2. As shown in previous studies, we observed a significant link between mean fat cell size and whole body insulin resistance determined by HOMA-index10 (Fig. 5a). To quantify the hyperplastic response of adipose tissue we measured the differentiation potential of human SVF from subcutaneous adipose tissue biopsies using ex vivo differentiation assays, in which adipogenesis was quantified by high-throughput image analysis. An inverse correlation was found between mean fat cell size and the differentiation potential of the corresponding SVF, which shows that fat cell size is also partly determined by the ability to produce new fat cells (Fig. 5b). Similar to its function in mice, Ror expression correlated positively with adipocyte size and inversely with adipocyte differentiation potential (Fig. 5c,d). MMP3 regulation is also conserved in human patients, as Mmp3 expression in the SVF is correlated with both Ror expression and with adipocyte size (Fig. 5e,f). With regard to insulin resistance, Ror expression in the SVF of subjects with a HOMA index < 3.5 is decreased compared to highly insulin resistant subjects with a HOMA index > 3.5. Importantly, Ror expression is independent of BMI, suggesting that Ror is an independent predictor of insulin resistance (Fig. 5g).

57

Chapter 2

a
mean adipocyte size [ m2]
25000 23000 21000 19000 17000 0 10 20

b
mean adipoycte size [ m2] R2= 0.47 p=0.001
25000

22500

20000

R2=0.49 p=0.004
17500 0 1 2 3 4 5 6 7

30

normalized Ror expression

9 R2= 0.47 8 p=0.0008 7 6 5 4 3 2 1 0 17000 19000

HOMA

normalized Ror expression

9 8 7 6 5 4 3 2 1 0 0

% differentiation

R2=0.38 p=0.011
1 2 3 4 5 6 7

21000

23000

25000

9 R2= 0.34 8 p=0.005 7 6 5 4 3 2 1 0 0.15 0.25 0.35

normalized Mmp3 expression

adipocyte size [ m2]

% differentiation
0.6 0.5 0.4 0.3 0.2 0.1 0.0 17000 19000 21000 23000 25000

normalized Ror expression

R2= 0.35 p=0.006

0.45

0.55

0.65

normalized Mmp3 expression

adipocyte cell size [m2]

Ror expression BMI normalized Ror expression

7 6 5 4 3 2 1 0 HOMA <3.5 HOMA >3.5 30 20

50 40

BMI [kg/m2]

10 0

Figure 5 ROR expression in adipose SVF correlates with adipocyte size and insulin sensitivity in human obese patients. a) Subcutaneous mean adipocyte cell size plotted against HOMA of obese patients. (n=19) b) Subcutaneous mean adipocyte cell size plotted against percent in vitro differentiated cells of human subcutaneous SVF. (n=15) c) Ror mRNA expression normalized to Gapdh in human subcutaneous SVF of obese subjects plotted against mean adipocyte cell size. (n=20) d) Ror mRNA expression in human subcutaneous SVF of obese subjects plotted against percent in vitro differentiated cells of human subcutaneous SVF. (n=16) e) Ror mRNA expression plotted against Mmp3 expression in subcutaneous SVF of obese patients. (n=21) f) Mmp3 mRNA expression in human subcutaneous SVF of obese subjects plotted against adipocyte cell size. (n=20) g) Ror expression in subcutaneous SVF and BMI of subjects with a HOMA index above or below 3.5. (n=19) Values represent means +/- SD, *p 0.05.

58

Chapter 2

Discussion
Our results identify ROR as a new key factor regulating the hyperplastic response of adipose tissue. Through modulation of adipogenesis via MMP3, ROR controls adipocyte numbers and thus adipocyte size. This in turn leads to an improvement of adipose tissue insulin sensitivity both in mice and in obese human patients (Fig. 6). Our results support the concept that stimulation of adipogenesis improves whole body insulin sensitivity through reduced systemic free fatty acid overload, a major cause for insulin resistance. It has been suggested that this is one reason for the differences between metabolically healthy and metabolically unhealthy obese subjects39-40 . Our experiments suggest that varying ROR expression in adipose SVF can account for differences in insulin resistance among obese subjects. Thus, ROR expression in fat is a novel marker, which can be used to assess the risk of secondary metabolic complications of individual obese patients.

ROR-/-

HFD ob/ob

wildtype

insulin
FFA

insulin

FFA

glucose

glucose insulin sensitive insulin resistant

Figure 6 Improved metabolic status in obese ROR-/- animals through enhanced adipocyte differentiation. Animals lacking ROR show enhanced adipocyte differentiation leading to small, insulin sensitive adipocytes. As a consequence, glycimic and lipolytic control is improved compared to wildtype animals.

59

Chapter 2

In addition, ROR might be a novel target for the treatment of obesity-associated metabolic disorders. Inhibition of ROR would mirror the effect of glitazones in clinical use, where activation of PPAR leads to improved insulin sensitivity independent of a reduction in body mass41-42. Given the fact that glitazones induce weight gain through additional modulation of mature adipocyte function and that glitazone treatment poses certain risks12,43-44, ROR repression might be a completely novel approach to treat insulin resistance in obese patients by improving compensatory de novo adipocyte differentiation without affecting mature adipocyte function. Recent studies have identified new ROR ligands, i.e. 7-hydroxycholesterol, 24S-hydroxycholesterol or the benzenesulfonamide T0901317, which repress ROR transcriptional activity20-22. The structure of these ligands might serve as a starting point to develop new therapeutic agents that inhibit ROR activity, thus leading to improved insulin sensitivity. Interestingly, the 1q21-q23 region containing the human ROR locus has been associated with type 2 diabetes45. Our study linking ROR to adipocyte differentiation provides further mechanistic clues as to why this might be the case. With regard to the expression profile of Ror during adipogenesis, it has to be noted that Ror is also expressed in confluent 3T3-L1 preadipocytes before induction of differentiation and shows a sharp decline in expression immediately after induction (Figure 2B). It has previously been reported that Ror is induced upon adipocyte differentiation28. However, in this study the early time points of differentiation were not analyzed, i.e. 3T3-L1 cells were only compared in the pre-confluent state and after differentiation. The immediate repression of Ror expression upon induction of differentiation mirrors expression of other inhibitory markers such as preadipocytes factor-146 and suggests an inhibitory role for ROR in early differentiation events. This is corroborated by the fact that ROR inhibits transcriptional activity of PPAR, an early key event of adipogenesis. In contrast to Rev-erb, which directly modulates activity of the PPAR promoter30, we did not detect any changes in PPAR promoter activity by ROR. This is further corroborated by the fact that ROR mediated repression of adipogenesis cannot be reversed by the PPAR activator rosiglitazone. Therefore, it can

60

Chapter 2

be presumed that ROR inhibits adipogenesis at an early stage through a PPAR independent mechanism, which ultimately leads to a less pronounced induction of PPAR. This is similar to the role of ROR in adipogenesis, as ROR inhibits the transcriptional activity of C/EBP, thereby blocking the induction of both PPAR and C/EBP. Although our results demonstrate that ROR regulates adipogenesis at the level of the adipocyte precursor cells both in vitro and in vivo, it cannot be excluded that some of the effects of ROR are mediated by mature adipocytes. Even though ROR is also highly expressed in mature adipocytes, ROR does not affect mature adipocyte functions such as lipolysis or glucose uptake. As the novel target gene MMP3 is a secreted protein, it can be speculated that ROR is instead expressed in mature adipocytes to create a negative feedback loop inhibiting further differentiation through secretion of MMP3. It has indeed been shown before that mature adipocytes can secrete inhibitory factors that block adipogenesis47. Thus, MMP3 could function as a paracrine inhibitor of differentiation, which would be an additional explanation for the effect of ROR on the process of fat cell formation. The regulation of adipogenesis through MMP3 by ROR is of special interest, because MMPs in general have been implicated in the process of adipocyte development. MMPs are likely candidates for regulation of adipocyte differentiation, as fibroblastlike preadipocytes need to remodel their cell shape through matrix alterations during adipogenesis15. Whereas global inhibition of MMPs reduces differentiation48, MMP3-/mice show an approximately 35% accelerated adipocyte differentiation during mammary gland involution18 suggesting that MMP3 functions as an inhibitor of adipocyte differentiation. We could confirm that MMP3 is an inhibitor of adipogenesis in 3T3-L1 adipocytes by siRNA knockdown and pharmacological inhibition of MMP3. Given the fact that MMP3 expression is driven by ROR, we identify here a novel paracrine mechanism regulating adipogenesis. An interesting aspect is the enhanced expression of ROR in visceral adipose tissue of obese patients. Since visceral adipose tissue is an independent determinant for the development of obesity-associated metabolic complications49-50, it can be speculated

61

Chapter 2

that insufficient suppression of ROR expression in visceral adipose tissue of obese patients leads to increased insulin resistance51. This could be one factor explaining why visceral adipose tissue is a larger risk factor for the development of obesity-associated metabolic disorders than subcutaneous adipose tissue. Taken together, our studies reveal ROR as an important regulator of compensatory adipocyte differentiation in the progression of obesity. Enhanced ROR expression in the state of obesity leads to adipocyte hypertrophy, thereby promoting increased systemic fatty acid overload, which ultimately favours hyperglycemia and insulin resistance.

62

Chapter 2

References
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Desai, M.Y., et al. Association of body mass index, metabolic syndrome, and leukocyte count. American Journal of Cardiology 97, 835-838 (2006). Meigs, J.B., et al. Body mass index, metabolic syndrome, and risk of type 2 diabetes or cardiovascular disease. Journal of Clinical Endocrinology and Metabolism 91, 2906-2912 (2006). Sethi, J.K. & Vidal-Puig, A.J. Thematic review series: Adipocyte biology - Adipose tissue function and plasticity orchestrate nutritional adaptation. J. Lipid Res. 48, 1253-1262 (2007). Jo, J., et al. Hypertrophy and/or Hyperplasia: Dynamics of Adipose Tissue Growth. PLoS computational biology 5, e1000324 (2009). Spalding, K.L., et al. Dynamics of fat cell turnover in humans. Nature 453, 783-787 (2008). Gesta, S., Tseng, Y. & Kahn, C.R. Developmental origin of fat: tracking obesity to its source. Cell 131, 242-256 (2007). Goossens, G.H. The role of adipose tissue dysfunction in the pathogenesis of obesity-related insulin resistance. Physiology & behavior 94, 206-218 (2008). Tilg, H. & Moschen, A.R. Adipocytokines: mediators linking adipose tissue, inflammation and immunity. Nature reviews 6, 772-783 (2006). Paolisso, G., et al. A high concentration of fasting plasma non-esterified fatty acids is a risk factor for the development of NIDDM. Diabetologia 38, 1213-1217 (1995). Roberts, R., et al. Markers of de novo lipogenesis in adipose tissue: associations with small adipocytes and insulin sensitivity in humans. Diabetologia 52, 882-890 (2009). Unger, R.H. Minireview: weapons of lean body mass destruction: the role of ectopic lipids in the metabolic syndrome. Endocrinology 144, 5159-5165 (2003). Crossno, J.T., Jr., Majka, S.M., Grazia, T., Gill, R.G. & Klemm, D.J. Rosiglitazone promotes development of a novel adipocyte population from bone marrow-derived circulating progenitor cells. The Journal of clinical investigation 116, 3220-3228 (2006). Johnson, J.A., Trasino, S.E., Ferrante, A.W., Jr. & Vasselli, J.R. Prolonged decrease of adipocyte size after rosiglitazone treatment in high- and low-fat-fed rats. Obesity (Silver Spring, Md 15, 2653-2663 (2007). Kubo, Y., Kaidzu, S., Nakajima, I., Takenouchi, K. & Nakamura, F. Organization of extracellular matrix components during differentiation of adipocytes in long-term culture. In Vitro Cell Dev Biol Anim 36, 38-44 (2000). Lilla, J., Stickens, D. & Werb, Z. Metalloproteases and adipogenesis: a weighty subject. The American journal of pathology 160, 1551-1554 (2002). Bouloumie, A., Sengenes, C., Portolan, G., Galitzky, J. & Lafontan, M. Adipocyte produces matrix metalloproteinases 2 and 9: involvement in adipose differentiation. Diabetes 50, 2080-2086 (2001). Lijnen, H.R., et al. Matrix metalloproteinase inhibition impairs adipose tissue development in mice. Arterioscler Thromb Vasc Biol 22, 374-379 (2002). Alexander, C.M., Selvarajan, S., Mudgett, J. & Werb, Z. Stromelysin-1 regulates adipogenesis during mammary gland involution. The Journal of cell biology 152, 693-703 (2001). Jetten, A.M. Retinoid-related orphan receptors (RORs): critical roles in development, immunity, circadian rhythm, and cellular metabolism. Nucl Recept Signal 7, e003 (2009). Kumar, N., et al. The benzenesulfonamide T0901317 is a novel RORalpha/gamma Inverse Agonist. Molecular pharmacology (2009). Wang, Y., Kumar, N., Crumbley, C., Griffin, P.R. & Burris, T.P. A second class of nuclear receptors for oxysterols: Regulation of RORalpha and RORgamma activity by 24S-hydroxycholesterol (cerebrosterol). Biochim Biophys Acta (2010). Wang, Y., et al. Modulation of RORalpha and RORgamma activity by 7-oxygenated sterol ligands. The Journal of biological chemistry (2009). Zhang, F., Meng, G. & Strober, W. Interactions among the transcription factors Runx1, RORgammat and Foxp3 regulate the differentiation of interleukin 17-producing T cells. Nature immunology 9, 1297-1306 (2008). Zhou, L., et al. TGF-beta-induced Foxp3 inhibits T(H)17 cell differentiation by antagonizing RORgammat function. Nature 453, 236-240 (2008).

13.

14.

15. 16.

17. 18. 19. 20. 21.

22. 23.

24.

63

Chapter 2

25. 26.

27.

28. 29. 30. 31. 32. 33. 34. 35.

36. 37.

38. 39. 40. 41.

42.

43. 44. 45.

46. 47.

Meissburger, B. & Wolfrum, C. The role of retinoids and their receptors in metabolic disorders. European Journal of Lipid Science and Technology 110, 191-205 (2008). Raichur, S., Lau, P., Staels, B. & Muscat, G.E.O. Retinoid-related orphan receptor gamma regulates several genes that control metabolism in skeletal muscle cells: links to modulation of reactive oxygen species production. Journal of Molecular Endocrinology 39, 29-44 (2007). Barendse, W., Bunch, R.J., Kijas, J.W. & Thomas, M.B. The effect of genetic variation of the retinoic acid receptor-related orphan receptor C gene on fatness in cattle. Genetics 175, 843853 (2007). Austin, S., et al. Induction of the nuclear orphan receptor ROR gamma during adipocyte differentiation of D1 and 3T3-L1 cells. Cell Growth & Differentiation 9, 267-276 (1998). Sun, Z.M., et al. Requirement for ROR gamma in thymocyte survival and lymphoid organ development. Science 288, 2369-2373 (2000). Wang, J. & Lazar, M.A. Bifunctional role of Rev-erbalpha in adipocyte differentiation. Mol Cell Biol 28, 2213-2220 (2008). Kim, J.B., Wright, H.M., Wright, M. & Spiegelman, B.M. ADD1/SREBP1 activates PPARgamma through the production of endogenous ligand. Proc Natl Acad Sci U S A 95, 4333-4337 (1998). Wolfrum, C., et al. Role of Foxa-2 in adipocyte metabolism and differentiation. Journal of Clinical Investigation 112, 345-356 (2003). Naldini, L., et al. In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector. Science 272, 263-267 (1996). Reeder, R.H. & Roeder, R.G. Ribosomal RNA synthesis in isolated nuclei. J Mol Biol 67, 433-441 (1972). Urso, B., et al. Differences in signaling properties of the cytoplasmic domains of the insulin receptor and insulin-like growth factor receptor in 3T3-L1 adipocytes. Journal of Biological Chemistry 274, 30864-30873 (1999). Mersmann, H.J. & Macneil, M.D. Variables in Estimation of Adipocyte Size and Number with a Particle Counter. Journal of Animal Science 62, 980-991 (1986). Hansen, L.H., Madsen, B., Teisner, B., Nielsen, J.H. & Billestrup, N. Characterization of the inhibitory effect of growth hormone on primary preadipocyte differentiation. Molecular Endocrinology 12, 1140-1149 (1998). Chen, H.C. & Farese, R.V. Determination of adipocyte size by computer image analysis. J. Lipid Res. 43, 986-989 (2002). Lonn, M., Mehlig, K., Bengtsson, C. & Lissner, L. Adipocyte size predicts incidence of type 2 diabetes in women. FASEB J 24, 326-331 (2010). O'Connell, J., et al. The relationship of omental and subcutaneous adipocyte size to metabolic disease in severe obesity. PLoS One 5, e9997 (2010). Czoski-Murray, C., et al. Clinical effectiveness and cost-effectiveness of pioglitazone and rosiglitazone in the treatment of type 2 diabetes: a systematic review and economic evaluation. Health Technol Assess 8, iii, ix-x, 1-91 (2004). Zanchi, A., Dulloo, A.G., Perregaux, C., Montani, J.P. & Burnier, M. Telmisartan prevents the glitazone-induced weight gain without interfering with its insulin-sensitizing properties. Am J Physiol Endocrinol Metab 293, E91-95 (2007). Grey, A. Skeletal consequences of thiazolidinedione therapy. Osteoporos Int 19, 129-137 (2008). Wan, Y., Chong, L.W. & Evans, R.M. PPAR-gamma regulates osteoclastogenesis in mice. Nature medicine 13, 1496-1503 (2007). Wang, H., et al. Molecular screening and association studies of retinoid-related orphan receptor gamma (RORC): a positional and functional candidate for type 2 diabetes. Mol Genet Metab 79, 176-182 (2003). Smas, C.M. & Sul, H.S. Pref-1, a protein containing EGF-like repeats, inhibits adipocyte differentiation. Cell 73, 725-734 (1993). Tremblay, F., et al. Bidirectional modulation of adipogenesis by the secreted protein Ccdc80/DRO1/URB. The Journal of biological chemistry 284, 8136-8147 (2009).

64

Chapter 2
48. Chavey, C., et al. Matrix metalloproteinases are differentially expressed in adipose tissue during obesity and modulate adipocyte differentiation. The Journal of biological chemistry 278, 1188811896 (2003). Ravussin, E. & Smith, S.R. Increased fat intake, impaired fat oxidation, and failure of fat cell proliferation result in ectopic fat storage, insulin resistance, and type 2 diabetes mellitus. Ann N Y Acad Sci 967, 363-378 (2002). Heilbronn, L., Smith, S.R. & Ravussin, E. Failure of fat cell proliferation, mitochondrial function and fat oxidation results in ectopic fat storage, insulin resistance and type II diabetes mellitus. Int J Obes Relat Metab Disord 28 Suppl 4, S12-21 (2004). Garaulet, M., Hernandez-Morante, J.J., Lujan, J., Tebar, F.J. & Zamora, S. Relationship between fat cell size and number and fatty acid composition in adipose tissue from different fat depots in overweight/obese humans. Int J Obes (Lond) 30, 899-905 (2006). Acknowledgments: A-FABP antibodies were provided by N. Haunerland. IL-17 promoter construct was a kind gift of Takashi Kobayashi. These studies were supported by ERC-Adipodif (CW), SNF-3100A0-117979 (CW), FP7-LipidomicNet (CW, DG), EASD (GR, CW), BMBF (PN) and the Roche Research Foundation (BM).

49.

50.

51.

52.

65

Chapter 3

Chapter 3
Tissue inhibitor of metalloproteinase 1 (TIMP1) inhibits adipocyte differentiation in obesity

Bettina Meissburger, Lena Stachorski, Evangeline Chew-Siegle, Christian Wolfrum

Swiss Federal Institute of Technology, ETH Zrich, Institute for Molecular Systems Biology, Wolfgang-Pauli Str. 16, 8093 Zrich, Switzerland

66

Chapter 3

Abstract
Extracellular matrix reorganization is a crucial step of adipocyte differentiation and is controlled by the MMP/TIMP enzyme system. The aim of this study was to define the role of TIMP1 in adipogenesis and to elucidate, whether upregulation of TIMP1 in obesity has direct effects on adipocyte formation and adipocyte size distribution, thereby influencing metabolism in the state of obesity. We show here that circulating TIMP1 levels are increased in obese mouse models and that this increase is dependent on altered adipocyte expression of this protein. TIMP1 is an inhibitor of adipocyte differentiation in 3T3-L1 as well as in primary preadipocytes. Conversely, neutralizing TIMP1 with a specific antibody enhances adipocyte differentiation. Moreover, we demonstrate that inhibition of adipocyte differentiation by injection of recombinant murine TIMP1 in mice challenged with a high-fat diet leads to hypertrophied adipocytes and a decrease in adipocyte numbers. As a consequence, TIMP1-treated mice develop increased circulating free fatty acid levels, hepatic steatosis and accelerated insulin resistance. In summary, TIMP1 is an important regulator of adipogenesis in the obese state. Inhibition of adipogenesis by TIMP1 promotes detrimental metabolic consequences such as systemic fatty acid overload, hepatic lipid accumulation and insulin resistance.

67

Chapter 3

Introduction
Obesity is the main risk factor for type 2 diabetes and cardiovascular diseases and has become a growing health problem throughout the world in the last decades. Expansion of adipose tissue depots results from a chronic imbalance between caloric intake and energy expenditure. Secondary health complications caused by obesity partly depend on how well adipose tissue can adapt to an extended nutrient supply1-2. One way of buffering excess nutrients in the adipose tissue is the generation of fat cells through de novo differentiation. This hyperplastic response is preferable to a hypertrophic reaction, where nutrients are stored in pre-existing adipocytes. Adipocyte hypertrophy commonly leads to enlarged, insulin-resistant fat cells with a high lipolytic rate, thus promoting ectopic lipid accumulation and systemic insulin resistance3-6. New adipocytes can differentiate from precursor cells, which reside in the stromal-vascular fraction (SVF) of adipose tissue7-8. For a fibroblast-like preadipocyte to become a round, lipid-laden fat cell, major restructuring of the extracellular matrix (ECM) is indispensable. Therefore, ECM remodeling has an important impact on adipocyte differentiation9-10. In the course of differentiation, the fibronectin-rich matrix is degraded and replaced by the basement membrane of an adipocyte11. Moreover, collagen expression switches from type I and type III collagen to type IV and VI collagen12-14. The concomitant extracellular protein degradation is to a great extent mediated by matrix metalloproteinases (MMPs) and their regulatory counterparts, the tissue inhibitors of metalloproteinases (TIMPs)10. Altered expression of several MMPs has been observed in adipose tissue of obese mice or humans and expression of the MMP system is tightly regulated during in vitro adipogenesis15. Thus, it is not surprising that MMPs influence the process of adipogenesis, albeit in both a repressive or activating way depending on substrate specificity10. Notably, mice deficient for MMP3, MMMP11 or MMP19 exhibit increased adiposity during diet-induced obesity16-18. In female mice after lactation, when secretory epithelial cells are replaced by adipocytes, several MMPs also influence adipogenesis in the mammary gland. Loss of MMP3, for example, leads to increased fat accumulation19. Moreover, lack of plasminogen, a precursor of plasmin, which is one of

68

Chapter 3

the key enzymes activating MMPs, efficiently blocks adipocyte differentiation20. Finally, mice treated with a global MMP inhibitor display reduced adipose tissue weight concomitant with a lower total number of adipocytes when exposed to a high-fat diet21. This is in line with the fact that the global MMP inhibitor BB-94 inhibits differentiation in 3T3-L1 cells15. MMPs are inhibited in a 1:1 stoichiometry by the family of TIMP proteins consisting of four isoforms (TIMP1-4) in mammals. TIMP1, which is thought to be mainly synthesized by cells within the SVF, inhibits most MMPs with low selectivity except for MMP14, MMP16 and MMP1922. Expression analysis in adipose tissue of lean and obese mice and humans revealed significant upregulation of TIMP1 in obesity23. Conflicting data has been reported concerning the functional role of TIMP1 in adipose tissue. Whereas male TIMP1-deficient mice kept on a high fat diet develop less adiposity than their wildtype counterparts, female TIMP1-deficient mice develop more adipose tissue in diet-induced obesity due to hyperphagia24-25. On the other hand, systemic adenoviral overexpression of the human TIMP1 homologue in mice did not have a marked effect on adipose tissue mass26. Nor did human TIMP1 influence in vitro differentiation of murine 3T3-F442A cells into adipocytes, whereas in another study, human TIMP1 accelerated 3T3-L1 adipogenesis19,26. A certain drawback of the latter studies is the fact that human TIMP1 was used in mouse model systems. Thus, it is possible that cross species differences in affinity and activity disturb the delicate balance between MMPs and their inhibitors. As TIMP1 expression is modulated in obesity, this protein might play a role in the underlying pathophysiological conditions associated with this disease. Therefore, the aim of this study was to define the role of TIMP1 in adipogenesis and to elucidate, whether upregulation of TIMP1 in obesity has direct effects on adipocyte formation and adipocyte size distribution thereby influencing metabolism in the state of obesity.

69

Chapter 3

Experimental Procedures
Real-Time PCR mRNA was isolated and transcribed to cDNA using the Multi-MACS cDNA kit (Miltenyi). mRNA expression was assessed by real-time PCR using SybrGreen (Invitrogen) according to the manufacturers protocol. mRNA expression was normalized to Gapdh. The following primer pairs were used (tan= 60C): gapdh-fw CTG ACG TGC CGC CTG GAG AAA, gapdh-rv CCG GCA TCG AAG GTG GAA GAG T, timp1-fw GAT ATG CCC ACA AGT CCC AGA ACC, timp1-rv GCA CAC CCC ACA GCC AGC ACT AT, ppar-fw AGG CGA GGG CGA TCT TGA CAG, ppar-rv AAT TCG GAT GGC CAC CTC TTT G, a-fabp-fw ATG AAA TCA CCG CAG ACG ACA G, afabp-rv GCC TTT CAT AAC ACA TTC CAC CAC. TIMP1 secretion from SVF Stromal-vascular fraction was isolated as described previously27 using 2 mg/ml collagenase type II for 1 h, followed by a filter step and erythrocyte lysis. Cells were grown over night on collagen-coated plates in high-glucose Dulbeccos modified Eagles medium supplemented with 10% fetal bovine serum and

penicillin/streptomycin.

The next day, the medium was changed to Opti-MEM

(Invitrogen) and supernatant was collected after 24h. TIMP1 in the supernatnat was measured using the Quantikine Mouse TIMP1 ELISA kit (R&D Systems) according to the manufacturers manual. MMP9 activity assay mMMP9 and hMMP9 were purchased from Anaspec. Inhibition of MMP9 by hTimp1 [R&D Systems] was tested using the SensoLyte 570 Generic MMP Assay Kit Fluorimetric (Anaspec) according to the manufacturers protocol. Recombinant murine TIMP1 Murine Timp1 devoid of the signaling peptide was cloned from murine adipose tissue cDNA using the following primers: mTimp1-fw AAG CTT TTG TAG CTG TGC CCC ACC CCA mTimp1-rv CTC GAG TCG GGC CCC AAG GGA TCT CC. Using HindIII and XhoI restriction sites recTimp1 was cloned into the pSecTagA vector (Invitrogen). For production of mTIMP1 protein HEK293T cells were transfected with lipofectamine

70

Chapter 3

(Invitrogen) according to the manufacturers protocol. Opti-MEM supernatant was collected twice every two days. mTIMP1 was purified twice via the his-tag using a NiNTA column (GE Healthcare) connected to an automated FPLC system. mTIMP1 was dialyzed with Slide-A-Lyzer Dialysis Cassettes (Thermo Scientific) against PBS and analyzed by Western Blot using anti-his tag antibody (Cell Signalling) or Coommassiestained 12% SDS-polyacrylamide gel. Inhibition of mMMP9 by mTIMP1 was assessed using the SensoLyte Generic MMP Assay Kit Colorimetric (Anaspec). Adipocyte differentiation assays 3T3-L1 preadipocytes and primary cells from stromal-vascular fraction were cultured on collagen-coated black 96-well plates in high-glucose Dulbeccos modified Eagles medium supplemented with 10% fetal bovine serum and penicillin/streptomycin. Adipogenesis was induced treating 48h postconfluent 3T3-L1 preadipocytes or 80% confluent stromal-vascular cells with induction medium (1 g/ml insulin, 1 M dexamethasone, 115 g/ml isobutyl-methylxanthine) for 2 days, followed by 2 days of insulin medium (1 g/ml) and 2 days of normal medium. mTIMP1 at indicated concentrations, monoclonal rat anti-mouse TIMP1 antibody (R&D systems) or rat IgG1 isotype control antibody (MBL) at 0.5 g/ml was added one day before and 2 days after induction of differentiation. Differentiated cells were fixed with 5 % formaldehyde and stained with 4 M Hoechst (nuclei), 5 M Syto60 (cytosol) and 2 M Bodipy (lipid droplets) or Oil Red O. After a 2 minute wash with 1.5% fat free BSA, 9 pictures per well were taken with an automated microscope imaging system (CellWorx). Pictures were automatically analyzed using Cell Profiler Software. Animal studies Male C57BL/6J mice were kept on a 12-h/12-h light/dark cycle in a pathogen-free animal facility. Mice were fed a high-fat diet (Provimi Kliba AG) containing 60% kcal from fat for 3 weeks. During this period daily subcutaneous mTIMP1 injections (200 g/kg) into the scapular region were performed. For analysis of short-term effects, mice were injected with mTIMP1 and fed a high-fat diet for 3 consecutive days followed by metabolic cage analysis of VO2, VCO2, food intake and physical activity using an Oxymax metabolic cage system (Columbus Instruments).

71

Chapter 3

Serum measurements Blood glucose was measured with a glycometer (Contour). For the ITT, 0.75 U/kg body weight of insulin was injected after 3 weeks of high fat diet and mTIMP1 injections. Insulin was determined with the Rat/Mouse Insulin ELISA kit (Crystal Chem). Triglycerides were measured with the Trig/GB Kit against the C.f.a.s. standard (Roche/Hitachi) and free fatty acids were measured with the NEAFA-HR kit (Wako). TIMP1 serum levels were assessed with the Quantikine Mouse TIMP1 ELISA kit (R&D Systems). Liver triglycerides Lipids from 150 mg liver were extracted with 1 ml hexane:isopropanol (3:2) by homogenizing the tissue in a TissueLyser. Supernatant was taken off and the extraction was repeated with 0.5 ml hexane:isopropanol. 0.5 ml of 15% Na2SO4 was added and the upper organic phase was evaporated over night. Lipids were dissolved in 2 ml TritonX100:methanol:butanol (Roche/Hitachi)28. Adipocyte and fat pad size Adipose tissue was fixed in 5% paraformaldehyde over night before paraffin embedding and sectioning to 10m slices. H&E staining of sections was performed according to standard procedures29. Microscopic pictures were taken and cell size was analyzed using Cell Profiler Software. Fat pads were measured with an animal CTScanner (LaTheta). Statistical Analysis Results are given as mean +/- standard deviation. Statistical analyses were performed using a two-tailed Students t test. (1:1:3) and measured with the Trig/GB Kit

72

Chapter 3

Results
It is known that obesity leads to a decreased capacity of compensatory de novo adipogenesis30-31, which is in turn influenced by paracrine secreted factors from the adipose tissue. Therefore, we analyzed whether Timp1 mRNA expression in adipose tissue is deregulated in mouse models of obesity. We found Timp1 mRNA to be highly expressed in the stromal-vascular fraction (SVF), whereas expression in mature adipocytes was lower, suggesting that TIMP1 is predominantly secreted from cells within the connective tissue (Fig. 1.1a). We next analyzed fat depot specific differences as well as the effect of obesity on Timp1 mRNA expression. As reported previously, we found Timp1 mRNA to be higher in visceral compared to subcutaneous adipose tissue. Furthermore, an increase of Timp1 mRNA expression could be observed in mature adipocytes of obese animal models such as ob/ob mice or mice fed a high-fat diet. Timp1 mRNA expression in the SVF, however, did not change in obese mouse models (Fig. 1.1a). Similar to the high expression in SVF, high levels of Timp1 mRNA in the 3T3-L1 preadipocyte cell culture model were detected. Interestingly, in a time curve of preadipocyte differentiation, expression of Timp1 mRNA declined at the time of induction and was very low in postconfluent differentiating 3T3-L1 preadipocytes. However, Timp1 mRNA expression was upregulated at later stages several days after the induction of peroxisome proliferator activated receptor gamma (Ppar) or adipocyte fatty acid binding protein (A-Fabp), which is in line with the observed expression levels in primary adipocytes (Fig. 1.1b).

73

Chapter 3

normalized Timp1 expression

2.5 2.0 1.5 1.0 0.5 0.0

**

**

subcutaneous visceral
** *** ***

**

D IO

w ild ty pe

w ild ty pe

SVF

adipocytes

b
normalized mRNA expression
3.0 2.5 2.0 1.5 1.0 0.5 0.0 -48 -24 0 24 48 72 96 120 144 168 192 time [h]

Timp1

Ppar

ob /o b

ob /o b

A-Fabp

induction of differentiation

Figure 1.1 Expression of Timp1 in adipocytes and preadipocytes. a) mRNA expression of Timp1 in subcutaneous and visceral stromal-vascular fraction and adipocytes of lean or obese male C57BL/6J mice. Expression was normalized to Gapdh. (n=4) b) mRNA expression of Timp1, Ppar and A-Fabp normalized to Gapdh in the course of 3T3-L1 differentiation. (n=5)

74

D IO

Chapter 3

To assess whether SVF from lean and obese animals show alterations in TIMP1 secretion, we measured TIMP1 in the supernatant of SVF from subcutaneous and visceral tissue. In accordance with the mRNA profile, TIMP1 release from visceral SVF was markedly higher than from subcutaneous SVF. However, no changes were observed in the amount of secreted TIMP1 protein from the SVF of obese compared to lean mice (Fig. 1.2c). As TIMP1 is a circulating factor that can exert systemic functions, we also measured TIMP1 protein concentrations in the serum of lean and obese mice. In ob/ob mice and mice with diet-induced obesity TIMP1 serum concentrations were three to four fold upregulated compared to lean mice (Fig. 1.2d).
c
75
subcutaneous visceral

c [TIMP1] /c [total protein]

50

25

0 lean DIO
** **

d
2.5
TIMP1 serum conc. [ng/ml]

2.0 1.5 1.0 0.5 0.0 C57Bl6 ob/ob HFD

Figure 1.2 Expression of TIMP1 in adipocytes and preadipocytes. c) TIMP1 secretion from murine stromal-vascular fraction of lean and obese mice. Cells were incubated with Opti-MEM for 24 h. TIMP1 concentrations were measured by ELISA. (n=3) d) TIMP1 protein concentrations in the serum of lean and obese mice measured by ELISA. (n=5) Values represent mean SD *p<0.05, **p<0.01, ***p<0.001

75

Chapter 3

It has previously been reported that global loss of TIMP1 in the mouse leads to alterations in adipose tissue constitution. However, other reports using the human analogue of TIMP1 (hTIMP1) for overexpression in mice or in cell culture models of adipogenesis did not observe any differences in adipocyte development. Given the fact that human and mouse TIMP1 only share 72 % homology (Fig. 2a), it is possible that the human isoform of TIMP1 is not able to bind and inhibit mouse MMPs in an efficient manner. To prove this, we compared the ability of hTIMP1 to inhibit murine (mMMP9) or human MMP9 (hMMP9) in an MMP activity assay. As shown in Fig. 2b, hTIMP1 inhibited cleavage of a substrate peptide by hMMP9 much more efficiently than cleavage by mMMP9. This demonstrates that species divergences have a considerable effect on the formation of MMP/TIMP1 complexes and subsequently on the inhibition of proteolytic MMP activity. In order to test whether mTIMP1 plays a functional role in adipogenesis, we purified secreted recombinant mouse TIMP1 (mTIMP1) with a molecular weight of 34kDa from cell culture supernatant using his-tag affinity purification (Fig. 2c,d). mTIMP1 activity was tested by its ability to inhibit MMP9 in a peptide cleavage assay (data not shown).

76

Chapter 3
a

b
% MMP9 activity

125 100 75 50 25 0 0 0.31 0.63 1.25 *** ** ** *

mMMP9 hMMP9

hTIMP1 [g/ml]
kDa

2.50 control inhibitor

c
8g 4g 2g 1g 0.5g

d
8g 4g 2g 1g 0.5g 0.25g

Figure 2 Activity of human and murine recombinant TIMP1. a) alignment of murine and human TIMP1 b) inhibition of mMMP9 or hMMP9 by hTIMP1. MMP9 activity was measured using a fluorimetric MMP activity assay with a FRET substrate peptide. (n=3) c) Coomassie staining of purified recombinant mTIMP1. d) Westernblot of purified recombinant mTIMP1. mTIMP1 was detected using an anti his-Tag antibody. Values represent mean SD *p<0.05, **p<0.01, ***p<0.001 To test whether mTIMP1 has an effect on adipocyte differentiation, we used the recombinant protein in a 3T3-L1 differentiation assay. As seen in Fig. 3a, increasing amounts of mTIMP1 added to the cell culture media in the range of endogenous secreted TIMP1 from 3T3-L1 preadipocytes (1.27 +/- 0.14 g/ml after 24h of culture) led to a progressive decrease in adipocyte formation evidenced by high throughput image analysis of lipid droplet accumulation. We next tested whether this inhibitory effect can also be seen in primary cells from murine adipose SVF. Whereas addition of mTIMP1 did not change the differentiation capacity of visceral SVF, mTIMP1 at a concentration of 50 g/ml lowered the degree of differentiation in subcutaneous SVF (Fig. 3b,c). The opposite effect could be achieved using a neutralizing antibody against TIMP1, which increased adipocyte differentiation solely in subcutaneous SVF (Fig. 3d).

77

Chapter 3

a
% differentiated cells

35 30 25 20 15 10 5 0 0 3

**

**

***

13

25

50

100

g/ml mTIMP1 control 3T3-L1 cells 25 g/ml TIMP1 50 g/ml TIMP1 100 g/ml TIMP1

c)

b
% differentiated cells

**
35 30 25 20 15 10 5 0 0 1 10 50 0 1 10 50

subcutaneous visceral g/ml mTIMP1 c control subcutaneous stromal-vascular fraction 12.5 g/ml TIMP1 25 g/ml TIMP1

50 g/ml TIMP1

78

Chapter 3

d 35

***
30 25 20 15 10 5 0 subcutaneous visceral control Ab anti-TIMP1 Ab

subcutaneous SVF anti- TIMP1 control

% differentiated cells

visceral SVF control anti- TIMP1

Figure 3 TIMP1 inhibits adipocyte differentiation. a) 3T3-L1 preadipocytes were differentiated into adipocytes in the presence of different amounts of murine recombinant TIMP1. Lipid droplet formation was assessed by fluorescent staining with automated image acquisition and analysis. Example pictures show lipid droplets in green, cytosol in red and nuclei in blue (n=4). b) subcutaneous and visceral SVF from male C57BL/6J mice were differentiated in the presence of different amounts of mTIMP1 (n=4). c) Oil red O staining of differentiated subcutaneous SVF in the presence of different amounts of mTIMP1. d) subcutaneous and visceral SVF from male C57BL/6J mice were differentiated in the presence of anti-TIMP1 or control antibody (n=4). Examples pictures show lipid droplets in green, cytosol in red and nuclei in blue. All results shown are representative of three independent experiments. Values represent mean SD. Scale: 100 m *p<0.05, **p<0.01, ***p<0.001 To analyze whether mTIMP1 also affects adipocyte plasticity in vivo we treated male C57BL/6J mice fed a high-fat diet with mTIMP1. 200 g/kg body weight of mTIMP1 or PBS was injected daily into subcutaneous, scalpular adipose tissue for a time period of 3 weeks. Apart from local accumulation in the subcutaneous fat, this injection regime also led to stably increased circulating blood levels of TIMP1 approximately 3 fold over

79

Chapter 3

baseline levels over the whole period of injection (Fig. 4.1a). To assess the effect of TIMP1 injection on adipose tissue development we measured whole body weight as well as fat pad weight by CT scanning (Fig. 4.1b,c). Three weeks of high fat diet led to a similar weight gain for PBS injected and TIMP1 injected mice. mTIMP1 injections did not alter whole body mass or fat pad sizes. However, we could observe a significant increase in mean adipocyte size and a shift in the size distribution profile towards larger adipocytes within the adipose tissue of mice injected with mTIMP1 (Fig. 4.2d-f). This is in line with the observation that TIMP1 inhibits adipocyte formation, thus causing adipocyte hypertrophy.
a
TIMP1 serum conc. [ng/ml]
8 7 6 5 4 3 2 1 0 0 10 20 30 40 50 60 70
0 PBS TIMP1

b
30

start weight 3 weeks HFD

weight [g]

20

10

time [h]
injections

12.5

PBS TIMP1

fat pad weight [g]

10.0 7.5 5.0 2.5 0.0 subcutanous visceral

Figure 4.1 mTIMP1 leads to adipocyte hypertrophy in vivo. a) serum concentration of TIMP1 measured by ELISA during daily injections (n=5) b, c) body weight and fat pad weight measured by CT of male C57BL/6J mice daily injected with TIMP1 or PBS and fed a high-fat diet for 3 weeks (n=5).

80

Chapter 3
d

mean adipocyte area [ m 2]

4500 4000 3500 3000 2500 2000 1500 1000 500 0

PBS

TIMP1

**

subcutaneous

injected

visceral

e
40

% of adipocytes

*
30

subcut. injected adipose tissue

PBS TIMP1

*
20 10 0

* ** ** **
0 00 00

**
0 00 0 00 00 00 00 011 0 20 -9

*
0 00 014 0 00 013 -1 -1 000 014 00 15 00 0 15 00 0

0 -5 00 -6

00

00

00

-4

00

0 -7

-2

00

-3

00

-8 00

00

00

00

00

00

00

00

90

11 0

00

13

00

40

50

60

70

10

20

30

80

00

adipcyte area m

35

% of adipocytes

30 25 20 15 10 5 0

***

**

subcutaneous adipose tissue * ** *** ***

0 00 00

PBS TIMP1

***
0 00

**
0 00 -1

***
00 11 0

**
00 00 20 -1

**
0 00 13 0-

**
013 00 14 00 0

00

00

00

00

00

-2

-7

-3

-4

-8

00

00

00

00

00

00

00

00

-9

-5

-6

00

0-

10

20

60

70

80

30

40

50

00

90

11 0

10

adipcyte area m

Figure 4.2 mTIMP1 leads to adipocyte hypertrophy in vivo. d) mean adipocyte size of subcutaneous fat at the site of injection, ventral subcutaneous fat and omental fat in male C57BL/6J mice daily injected with TIMP1 or PBS for 3 weeks e-f) adipocte size distributions of subcutaneous fat pads after 3 weeks of high fat diet. (n=5) Values represent mean SD. *p<0.05, **p<0.01, ***p<0.001

12

00

00

14

10

12

81

Chapter 3

Along with an increase in adipocyte size, mTIMP1 injected mice showed significantly higher blood glucose and insulin levels as well as decreased insulin sensitivity in an insulin tolerance test compared to control animals, indicating increased insulin resistance in mTIMP1 injected mice (Fig. 5.1a-c). In line with this, free fatty acid serum concentrations as well as liver triglycerides were increased in mTIMP1 treated mice, suggesting a higher lipolytic rate in adipocytes of TIMP1-treated animals (Fig. 5.1d,e). Serum triglyceride levels on the other hand were unchanged (Fig. 5.1f).
*
a
9 8 7 6 5 4 3 2 1 0 PBS TIMP1

2.5

blood glucose [mM]

insulin [ng/ml]

2.0 1.5 1.0 0.5 0.0

c
150

d
3

PBS

TIMP1

**

glucose (% of initial)

TIMP1 PBS

FFA [mM]

100

**
50

**

0 0 30 60 90 120

0 PBS TIMP1

e
70

time [min]

f
100

**
triglycerides [mg/dl]
PBS TIMP1

mg TG /g wet liver

60 50 40 30 20 10 0

75 50 25 0 PBS TIMP1

Figure 5.1 mTIMP1 impairs glucose and fatty acid metabolism a, b) fasting blood glucose and insulin serum levels in mice injected with TIMP1 or PBS c) ITT of mice injected with TIMP1 or PBS and fed a high-fat diet for 3 weeks (0.75U insulin / kg body weight) (n=5). d-f) serum free fatty acids, liver triglyceride content and serum triglycerides of PBS or TIMP1 injected mice after 4 weeks of high-fat diet (n=5).

82

Chapter 3

In order to elucidate whether TIMP1 injections alter MMP levels in adipose tissue, we analyzed expression of MMP2 and MMP9, which are known modulators of adipogenesis32 in different adipose tissue depots. Though MMP levels varied between mice, a general trend towards upregulation of MMP expression with significantly higher expression of MMP9 and MMP2 in the injected adipose tissue could be seen (Fig. 5.2g-i).
g
MMP/ -tubulin intensity

2.5 2.0 1.5 1.0 0.5 0.0

PBS MMP9 MMP2 -tubulin

TIMP1
105 kDa 72 kDa 68 kDa 50 kDa

PBS TIMP1 *

p=0.07

subcut. injected adipose tissue

proMMP2 matureMMP2 proMMP9

MMP/ -tubulin intensity

h PBS MMP9 MMP2 -tubulin subcutaneous adipose tissue TIMP1

PBS TIMP1
1

proMMP2 matureMMP2 proMMP9

i MMP9 MMP2 -tubulin

PBS

TIMP1

MMP/ -tubulin intensity

PBS TIMP1

visceral adipose tissue

proMMP2 matureMMP2 proMMP9

Figure 5.2 mTIMP1 impairs glucose and fatty acid metabolism g-i) Western Blot for MMP2, MMP9 and -tubulin of protein extracts from subcutaneous, epididymal and subcutaneous, scalpular adipose tissue. The results shown are representative of two independent injection experiments with independent protein preparations. Values represent mean SD. *p<0.05, **p<0.01, ***p<0.001

83

Chapter 3

Since it has been reported that some of the observed effects of TIMP1 might be due to a central regulation of energy balance through modulation of energy expenditure or food intake, we tested respiration as well as activity of the mice in a metabolic cage system after short term injections of mTIMP1 (3 days) and after 3 weeks of mTIMP1 treatment. Neither short (data not shown) nor long term (Fig. 6a-c) mTIMP1 injections led to any alterations in oxygen consumption, feeding or physical activity of the mice, implying that modulation of the mTIMP1 circulating concentrations through injection does not affect regulatory mechanisms of energy homeostasis in the central nervous system.
a
2.5

b
35000

food intake [g]

2.0 1.5 1.0 0.5 0.0

PBS TIMP1
xtot [counts]

30000 25000 20000 15000 10000 5000 0

PBS TIMP1

day

night

day

night

c
VO2/VCO2 [ml/kg/h]

5500 5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0

PBS TIMP1

day

night

day

night

VO2

VCO2

Figure 6 mTIMP1 does not affect central energy regulation. a) food intake b) movement and c) respiration in male C57BL/6J mice injected with TIMP1 or PBS and fed a high-fat diet for 3 weeks. (n=5)

84

Chapter 3

In summary, we show here that TIMP1, which is a secreted factor from adipose tissue, is upregulated in obesity. Increased circulating TIMP1 levels inhibit adipocyte differentiation, thereby decreasing the hyperplastic response of adipose tissue under high fat diet. This in turn results in decreased numbers of adipocytes, which are enlarged in size leading to an impaired lipolytic and glycemic control.

85

Chapter 3

Discussion
Tight regulation of extracellular matrix remodeling by MMPs and their inhibitors such as TIMP1 is crucial for the adipogenic process. The fact that TIMP1 is highly upregulated in obesity in mice and humans led to the assumption that TIMP1 might influence adipose tissue plasticity in states of chronic over-nutrition. We could confirm here that TIMP1 is mainly expressed in SVF in comparison to adipocytes and that expression is higher in the visceral adipose compartment15,23,33-35. Surprisingly, even though circulating TIMP1 levels are elevated in obesity (our data,
23

), neither TIMP1

mRNA levels in the SVF nor TIMP1 secretion from SVF is altered when comparing obese and lean animals. Thus, the observed increase in circulating TIMP1 levels could be due to secretion from mature adipocytes, which is corroborated by the elevation of TIMP1 mRNA in obese vs. lean animal models. These data conflict to some extent with a report from Maury et al.35, which shows higher TIMP1 secretion from SVF and adipocytes using human omental adipose tissue from obese compared to lean patients. These differences might be due to differences between the human and the mouse system or might be dependent on different culture and preparation conditions and will need to be addressed in future studies. Alternatively, other tissues might contribute to elevated TIMP1 serum levels in obesity, as TIMP1 is widely expressed. As reported previously, TIMP1 expression is blunted at induction of adipocyte differentiation, hinting at an inhibitory role of TIMP1 in adipogenic matrix reorganization15. We could show that addition of murine recombinant TIMP1 to the medium represses adipogenesis in 3T3-L1 preadipocytes and in subcutaneous SVF, whereas neutralization of TIMP1 with a specific antibody leads to the opposite effect. As TIMP1 inhibits a broad spectrum of MMPs, these results go in line with the fact that global MMP inhibitors repress adipocyte differentiation in vitro15,32. In a previous study, Demeulemeester et al.26 reported no effect of overexpressed human TIMP1 on adipocyte differentiation in 3T3-F332A preadipocytes, whereas Alexander et al.19 reported enhanced adipogenesis using the MMP inhibitor GM6001 or hTIMP1. One crucial difference between the various reports is the fact that hTIMP1 was used in the context of a murine cell culture system. Murine and human TIMP1 share only 72%

86

Chapter 3

sequence identity and we could show that hTIMP1 inhbits mMMP9 less effeciently than its ortholouge, hMMP9. MMP9 and MMP2 are highly expressed during adipocyte differentiation and are supposed to promote adipocyte formation32,36, these MMPs are therefore important targets for TIMP1 with regard to adipogenesis. In contrast to subcutaneous SVF, no effect of mTIMP1 or TIMP1 neutralization could be seen in visceral SVF. It is known that subcutaneous and visceral adipose tissue show distinct gene expression patterns and display different risk factors for the development of insulin resistance8,37. As the differentiation capacity of visceral SVF is much lower per se (our data, 38), it is likely that other inhibitory factors apart from TIMP1 additionally control adipogenesis in this adipose tissue compartment. In vivo, global knockout of TIMP1 leads to gender-specific effects, giving rise to less fat mass and smaller adipocytes in male animals on a high-fat diet, whereas female mice with a null mutation in TIMP1 develop bigger fat pads and larger adipocytes in dietinduced obesity24-25. One important issue is the impact of TIMP1 ablation on the central nervous system, leading to hyperphagia in female mice, which might mask direct effects on adipose tissue development. On the other hand, male TIMP1 ko mice do not show altered food intake or activity but display lower body weights already at weaning, hinting at a role for TIMP1 in development. Stable overexpression of hTIMP1 by adenoviral gene transfer did not result in any effect on adipose tissue development in one study26, whereas Alexander et al.19 observed more, hypertrophied adipocytes in the mammary gland after overexpression of human TIMP1. Again, species incompatibility might play a crucial role here and adipocyte development in the mammary gland underlies most likely different regulatory mechanisms than in the adipose tissue. Here, we could show that injection of murine TIMP1 into subcutaneous fat during 3 weeks of high-fat diet leads to a decreased number of adipocytes in subcutaneous fat pads, which are larger in size, as formation of new adipocytes is inhibited by TIMP1. Importantly, the shift in adipocyte number and size is independent of body or fat pad weight. As neither food intake nor energy expenditure is affected in mTIMP1-injected mice, it can be ruled out that the observed effect on adipocyte size and cellularity is mediated by indirect central effects on feeding and energy consumption.
87

Chapter 3

Thus, inhibition of adipocyte differentiation by TIMP1, as shown in vitro, shifts the balance from a hyperplastic response to hypertrophy of existing adipocytes in the state of increased energy supply. It is known that hypertrophied adipocytes become insulin resistant and have a high lipolytic rate5-6,39-40. In line with this, serum levels of free fatty acids are enhanced, which leads to increased triglyceride accumulation in the liver promoting whole body insulin resistance. As a consequence, TIMP1 injected mice show increased blood glucose and insulin serum levels as well as impaired performance in insulin-stimulated glucose clearing. As TIMP1 inhibits a broad range of MMPs , the precise mechanism by which TIMP1 inhibits adipocyte differentiation is difficult to assess22. Looking at the class of gelatinases (MMP2/9), which are known to promote differentiation32,36, we could see a trend towards upregulation of those MMPs, hinting at a regulatory role of TIMP1 for these MMPs important in adipocyte differentiation. Further studies will need to clarify the precise contributions of different MMPs and their effect on organization of extracellular matrix proteins in adipose tissue. It is well known that proinflammatory cytokines and adipokines are upregulated in adipose tissue of obese subjects and contribute to the development of insulinresistance41-42. In 3T3-L1 adipocytes TIMP1 can be induced by cytokines such as TNF, IL-6 and IL1-43-44. It is therefore intriguing to speculate that increased TIMP1 expression in obesity is driven by inflammatory signals, thus inhibiting differentiation of new adipocytes. In fact, it has been shown that the differentiation capacity of SVF from obese mice is significantly impaired, which could in turn lead to adipocyte hypertrophy and increased insulin resistance45. In summary, elevated TIMP1 levels in obesity inhibit adipocyte differentiation, thus decreasing the hyperplastic response of adipose tissue under high fat diet. This in turn results in decreased numbers of adipocytes, which are enlarged in size leading to an impaired lipolytic and glycemic control.

88

Chapter 3

References
1. 2. 3. 4. 5. Goossens, G.H. The role of adipose tissue dysfunction in the pathogenesis of obesity-related insulin resistance. Physiol Behav 94, 206-218 (2008). Sethi, J.K. & Vidal-Puig, A.J. Thematic review series: adipocyte biology. Adipose tissue function and plasticity orchestrate nutritional adaptation. J Lipid Res 48, 1253-1262 (2007). Jo, J., et al. Hypertrophy and/or Hyperplasia: Dynamics of Adipose Tissue Growth. PLoS Comput Biol 5, e1000324 (2009). Spalding, K.L., et al. Dynamics of fat cell turnover in humans. Nature 453, 783-787 (2008). Salans, L.B. & Doughert.Jw. Effect of Insulin Upon Glucose Metabolism by Adipose Cells of Different Size - Influence of Cell Lipid and Protein Content, Age, and Nutritional State. Journal of Clinical Investigation 50, 1399-& (1971). Lundgren, M., et al. Fat cell enlargement is an independent marker of insulin resistance and 'hyperleptinaemia'. Diabetologia 50, 625-633 (2007). Gregoire, F.M. Adipocyte differentiation: from fibroblast to endocrine cell. Exp Biol Med (Maywood) 226, 997-1002 (2001). Gesta, S., Tseng, Y. & Kahn, C.R. Developmental origin of fat: tracking obesity to its source. Cell 131, 242-256 (2007). Chun, T.H., et al. A pericellular collagenase directs the 3-dimensional development of white adipose tissue. Cell 125, 577-591 (2006). Lilla, J., Stickens, D. & Werb, Z. Metalloproteases and adipogenesis: a weighty subject. The American journal of pathology 160, 1551-1554 (2002). Spiegelman, B.M. & Ginty, C.A. Fibronectin modulation of cell shape and lipogenic gene expression in 3T3-adipocytes. Cell 35, 657-666 (1983). Aratani, Y. & Kitagawa, Y. Enhanced synthesis and secretion of type IV collagen and entactin during adipose conversion of 3T3-L1 cells and production of unorthodox laminin complex. J Biol Chem 263, 16163-16169 (1988). Kubo, Y., Kaidzu, S., Nakajima, I., Takenouchi, K. & Nakamura, F. Organization of extracellular matrix components during differentiation of adipocytes in long-term culture. In Vitro Cell Dev Biol Anim 36, 38-44 (2000). Weiner, F.R., Shah, A., Smith, P.J., Rubin, C.S. & Zern, M.A. Regulation of collagen gene expression in 3T3-L1 cells. Effects of adipocyte differentiation and tumor necrosis factor alpha. Biochemistry 28, 4094-4099 (1989). Chavey, C., et al. Matrix metalloproteinases are differentially expressed in adipose tissue during obesity and modulate adipocyte differentiation. J Biol Chem 278, 11888-11896 (2003). Lijnen, H.R., Van, H.B., Frederix, L., Rio, M.C. & Collen, D. Adipocyte hypertrophy in stromelysin3 deficient mice with nutritionally induced obesity. Thromb Haemost 87, 530-535 (2002). Maquoi, E., Demeulemeester, D., Voros, G., Collen, D. & Lijnen, H.R. Enhanced nutritionally induced adipose tissue development in mice with stromelysin-1 gene inactivation. Thromb Haemost 89, 696-704 (2003). Pendas, A.M., et al. Diet-induced obesity and reduced skin cancer susceptibility in matrix metalloproteinase 19-deficient mice. Mol Cell Biol 24, 5304-5313 (2004). Alexander, C.M., Selvarajan, S., Mudgett, J. & Werb, Z. Stromelysin-1 regulates adipogenesis during mammary gland involution. The Journal of cell biology 152, 693-703 (2001). Selvarajan, S., Lund, L.R., Takeuchi, T., Craik, C.S. & Werb, Z. A plasma kallikrein-dependent plasminogen cascade required for adipocyte differentiation. Nat Cell Biol 3, 267-275 (2001). Lijnen, H.R., et al. Matrix metalloproteinase inhibition impairs adipose tissue development in mice. Arterioscler Thromb Vasc Biol 22, 374-379 (2002). Brew, K. & Nagase, H. The tissue inhibitors of metalloproteinases (TIMPs): an ancient family with structural and functional diversity. Biochim Biophys Acta 1803, 55-71 (2010). Kralisch, S., et al. Tissue inhibitor of metalloproteinase-1 predicts adiposity in humans. European journal of endocrinology 156, 257-261 (2007). Lijnen, H.R., Demeulemeester, D., Van Hoef, B., Collen, D. & Maquoi, E. Deficiency of TIMP-1 impairs nutritionally induced obesity in mice. Thromb Haemost 89, 249-255 (2003).

6. 7. 8. 9. 10. 11. 12.

13.

14.

15. 16. 17.

18. 19. 20. 21. 22. 23. 24.

89

Chapter 3

25. 26.

27.

28. 29. 30.

31. 32.

33.

34.

35. 36.

37. 38.

39.

40.

41. 42. 43. 44.

45.

Gerin, I., et al. Hyperphagia and obesity in female mice lacking tissue inhibitor of metalloproteinase-1. Endocrinology (2008). Demeulemeester, D., et al. Overexpression of tissue inhibitor of matrix metalloproteinases-1 (TIMP-1) in mice does not affect adipogenesis or adipose tissue development. Thromb Haemost 95, 1019-1024 (2006). Hansen, L.H., Madsen, B., Teisner, B., Nielsen, J.H. & Billestrup, N. Characterization of the inhibitory effect of growth hormone on primary preadipocyte differentiation. Molecular Endocrinology 12, 1140-1149 (1998). Hara, A. & Radin, N.S. Lipid extraction of tissues with a low-toxicity solvent. Anal Biochem 90, 420-426 (1978). Chen, H.C. & Farese, R.V., Jr. Determination of adipocyte size by computer image analysis. J Lipid Res 43, 986-989 (2002). Permana, P.A., et al. Subcutaneous abdominal preadipocyte differentiation in vitro inversely correlates with central obesity. American Journal of Physiology-Endocrinology and Metabolism 286, E958-E962 (2004). Wolfrum, C., Luca, E. & Stoffel, M. Role of foxa2 in the regulation of hepatic glucose and fatty acid metabolism in starvation and in diabetes. Diabetes 53, A5-A5 (2004). Bouloumie, A., Sengenes, C., Portolan, G., Galitzky, J. & Lafontan, M. Adipocyte produces matrix metalloproteinases 2 and 9: involvement in adipose differentiation. Diabetes 50, 2080-2086 (2001). Klimcakova, E., et al. Profiling of adipokines secreted from human subcutaneous adipose tissue in response to PPAR agonists. Biochemical and biophysical research communications 358, 897902 (2007). Maquoi, E., Munaut, C., Colige, A., Collen, D. & Lijnen, H.R. Modulation of adipose tissue expression of murine matrix metalloproteinases and their tissue inhibitors with obesity. Diabetes 51, 1093-1101 (2002). Maury, E., et al. Adipokines oversecreted by omental adipose tissue in human obesity. American journal of physiology 293, E656-665 (2007). Bourlier, V., et al. Protease inhibitor treatments reveal specific involvement of matrix metalloproteinase-9 in human adipocyte differentiation. J Pharmacol Exp Ther 312, 1272-1279 (2005). Wajchenberg, B.L. Subcutaneous and visceral adipose tissue: their relation to the metabolic syndrome. Endocr Rev 21, 697-738 (2000). Joe, A.W., Yi, L., Even, Y., Vogl, A.W. & Rossi, F.M. Depot-specific differences in adipogenic progenitor abundance and proliferative response to high-fat diet. Stem Cells 27, 2563-2570 (2009). Buren, J., Lindmark, S., Renstrom, F. & Eriksson, J.W. In vitro reversal of hyperglycemia normalizes insulin action in fat cells from type 2 diabetes patients: is cellular insulin resistance caused by glucotoxicity in vivo? Metabolism 52, 239-245 (2003). Weyer, C., Foley, J.E., Bogardus, C., Tataranni, P.A. & Pratley, R.E. Enlarged subcutaneous abdominal adipocyte size, but not obesity itself, predicts type II diabetes independent of insulin resistance. Diabetologia 43, 1498-1506 (2000). Stienstra, R., Duval, C., Muller, M. & Kersten, S. PPARs, Obesity, and Inflammation. PPAR Res 2007, 95974 (2007). Zeyda, M. & Stulnig, T.M. Obesity, inflammation, and insulin resistance--a mini-review. Gerontology 55, 379-386 (2009). Kralisch, S., et al. Proinflammatory adipocytokines induce TIMP-1 expression in 3T3-L1 adipocytes. FEBS letters 579, 6417-6422 (2005). Weise, S., et al. Tissue inhibitor of metalloproteinase-1 mRNA production and protein secretion are induced by interleukin-1 beta in 3T3-L1 adipocytes. The Journal of endocrinology 198, 169174 (2008). Wolfrum, C., et al. Role of Foxa-2 in adipocyte metabolism and differentiation. J Clin Invest 112, 345-356 (2003).

90

Chapter 4

Chapter 4
Regulation of adipogenesis by paracrine factors from adipose stromal-vascular fraction- a link to fat depot-specific differences
Bettina Meissburger, Hansjrg Mst, Aliki Perdikari, Matthias Geiger, Christan Wolfrum

Swiss Federal Institute of Technology, ETH Zrich, Institute for Molecular Systems Biology, Wolfgang-Pauli Str. 16, 8093 Zrich, Switzerland

91

Chapter 4

Abstract
Visceral and subcutaneous adipose tissue depots have distinct features and contribute differentially to metabolic disease. Therefore, the adipogenic potential of different fat depots was investigated and found to be higher in subcutaneous compared to visceral stromal-vascular fraction SVF, which contains adipocyte precursor cells. This increased differentiation capacity is not due to elevated numbers of Lin-Sca1+CD29+CD34+Pref1+ precursor cells, as the number of preadipocytes is higher in visceral than in subcutaneous SVF. Using conditioned media it as shown that secreted heat-sensitive factors from the SVF inhibit adipocyte differentiation to a higher extent in visceral than in subcutaneous SVF. In order to explore secreted proteins that potentially inhibit differentiation, the secretome of murine SVF was analyzed by mass spectrometry and identified 113 secreted proteins were identified with an overlap of 42% between subcutaneous and visceral SVF. Comparison of the mRNA expression in SVF from both depots revealed 16 transcripts that are significantly higher expressed in visceral than in subcutaneous SVF. A functional differentiation screen using adenoviral knockdown identified 7 potential inhibitory candidates, i.e. biglycan (Bgn), decorin (Dcn), bone morphogenic protein 1 (Bmp1), epidermal growth factor-containing fibulin-like extracellular matrix protein 2 (Efemp2), elastin microfibril interfacer 1 (Emilin1), matrix gla protein (Mgp) and Sparc-like 1 (Sparcl1). For further verification, murine recombinant Sparcl1 or Dcn was added to the media during the differentiation process leading to a dose-dependent decrease in adipogenesis. Further analysis will be necessary to assess the impact of the other candidates on adipocyte differentiation. Furthermore, it will also be of interest to explore, which cell population within the SVF secretes these inhibitory proteins and to investigate the underlying molecular mechanisms leading to the inhibition of differentiation.

92

Chapter 4

Introduction
Adipose tissue is an important regulator of lipid metabolism within the body. Adipocytes serve as a safe storage place for excess lipids in times of abundant nutrient supply. Vice versa, adipocytes release free fatty acids during fasting to supply other peripheral tissues with energy1-2. Apart from its role as a lipid reservoir, adipose tissue is also an endocrine organ secreting adipokines, cytokines and hormones, which exert paracrine effects on adipose tissue function as well as endocrine regulation of whole body energy homeostasis3-5. In response to fluctuations in energy supply, adipose tissue has evolved into a very dynamic tissue being able to expand and shrink in mass according to the nutritional status6-7. To do so, adipocytes have the unique capability to change their size several fold8-9. Additionally, adipocytes are thought to have a constant turnover rate10 with new adipocytes being formed by de novo differentiation from adipocyte precursor cells (APC)11-12, which contributes to adipose tissue plasticity. APCs are fibroblast-like cells that reside within the stromal-vascular fraction (SVF) of adipose tissue. Whereas it has long been known that cells within the SVF can be differentiated into adipocytes and other cell types in vitro using appropriate induction reagents13-15, the exact nature and identification of APCs within the SVF has proved difficult. Within the Lin- population, which excludes CD31+ (endothelial cells), CD45+ (hematopoietic cells) and Ter119 (erythroid cells), a combination of several stem cell markers such as CD29 (integrin 1), CD34 and stem cell antigen 1 (Sca1/Ly6a) has recently been proposed to characterize APCs16-18. Additionally, preadipocyte factor 1 (Pref1/Dlk), which needs to be downregulated upon differentiation, has been suggested as an APC marker within adipose tissue19-20. Differentiation is not only determined by adipocyte progenitor numbers but is also influenced by external stimuli within the extracellular adipose tissue milieu. Numerous examples of endocrine and paracrine factors are known to influence adipogenesis, such as insulin, fibroblast growth factor and glucocorticoids promoting differentiation or TNF, Wnt and lysophosphatidic acid repressing differentiation21-26. Extracellular matrix components also play an important role either directly in cell shape modulation

93

Chapter 4

or by sequestering important growth factors and possibly influencing the adipocyte progenitor niche27-29. These factors can either be secreted from mature adipocytes or from cells within the SVF or reach the adipose tissue via the circulation30-33. Whereas adipose tissue is very efficient in storing energy for times of scarce nutritional supply, the evolutionary more recent, constant oversupply of food rich in calories can pose major functional disturbances in adipose tissue. Excess adipose tissue is one of the key features of the metabolic syndrome increasing the risk to develop type 2 diabetes and cardiovascular disease34-36. Of major importance is the fact that hypertrophied adipocytes show dysregulated function with increased free fatty acid and pro-inflammatory cytokine release, thus promoting dyslipidemia and low-grade inflammation contributing to insulin resistance9,37-39. Therefore, fat cell size, which is determined by the balance between de novo adipogenesis and adipocyte hypertrophy, is an important factor that influences the metabolic outcome of obesity40-42. A key factor for metabolic secondary complications is the location of excess adipose tissue. Central adiposity with increased amounts of visceral fat poses a greater metabolic risk than excess subcutaneous adipose tissue6,43-44. Many reports support the idea that subcutaneous and visceral adipose tissue display distinct features such as different gene expression, higher lipolytic rate and decreased insulin sensitivity in visceral adipose tissue45-47. Part of an explanation for this could be differences in the adipogenic potential of SVF resulting in depot-specific cell size distributions and adipocyte function. This report links adipogenic potential of murine SVF to adipocyte size within different fat depots. To shed light on the underlying mechanisms, adipocyte progenitor numbers as well as paracrine factors secreted from SVF, which might affect the differentiation ability, are analyzed in subcutaneous and visceral SVF. Using a mass spectrometry approach, novel secreted factors from SVF are identified that inhibit adipogenesis in primary SVF culture, thereby potentially contributing to increased adipocyte hypertrophy.

94

Chapter 4

Methods
Animals C57Bl/6J mice aged 12-14 weeks were obtained from Jackson Laboratory and kept on a 12-h/12-h light/dark cycle in a pathogen-free animal facility. SVF isolation and adipocyte differentiation SVF from subcutaneous and visceral fat was prepared as described before48. Briefly, subcutaneous fat around the thighs and shoulders as well as epididymal fat pads were excised, minced and incubated in collagenase type II for 1 h at 37C. Pelleted SVF was strained through a 40 m net, erythrocytes were lysed and 9*105/cm2 cells were seeded on collagen-coated plates in D-MEM with 10% FBS penicillin/streptomycin. 3T3-L1 cells were seeded on collagen-coated plates at 6*104/cm2 and kept confluent for 48-72h before induction of differentiation. Differentiation was induced using 1 M dexamethasone, 1 g/ml insulin and 0.5 mM isobutylmethylxanthine for 2 days, followed by 2 days of insulin medium (1 g/ml) and 2 days of growth medium. For conditioned media experiments, SVF cells were seeded on collagen-coated 6-well plates. After cells had attached, fresh growth medium was added, collected after 48h and used for differentiation of subcutaneous and visceral SVF. As a control, medium was incubated at 37C in empty cell culture dishes for the same time. For heat inactivation, collected supernatant was heated to 95C for 10 min. Automated analysis of adipocyte differentiation Differentiated cells were fixed with 5% formaldehyde prior to staining with BODIPY (2 M) for lipid droplets, Hoechst (4 M) for nuclei and Syto60 (5 M) for cytosolic staining (all Invitrogen). 12 pictures per well were taken with an automated microscope imaging system (CellWorx). Pictures were analyzed using Cell Profiler Software. Adipocyte cell size Adipose tissue was fixed in 5% paraformaldehyde over night before paraffin embedding and sectioning to 10m slices. H&E staining of sections was performed

95

Chapter 4

according to standard procedures49. Microscopic pictures were taken and cell size was analyzed using Cell Profiler Software. Flow cytometry analysis of adipocyte precursor markers SVF was stained with Ter119-Alexa488, CD45-Alexa488, CD31-Alexa488 for Linselection, CD29-PE, CD34-Alexa647, Sca1-PE-Cy7 (all Biolegend) and Pref-1 antibody (MBL) conjugated with APC-Cy7 (lightning kit, Innova). 7-AAD was added for live cell staining and cells were analyzed on a FACSCanto II machine (BD). Data was analyzed using FlowJo software. Mass spectrometry analysis of supernatant from SVF SVF were seeded as described before on collagen or poly-lysine coated 6-well plates. After cells had attached, growth medium was replaced by Opti-MEM and collected after 48h. Collected supernatant was dialyzed against PBS, proteins were reduced with 5 mM Tris 2-carboxyethyl phosphine (TCEP) and alkylated with 10 mM iodoacetamide. Samples were digested with trypsin (Promega) in a 1:50 ratio for 12 h at room temperature. Peptides were desalted on Ultra MicroTIP Columns (The Nest Group, Southborough, MA, USA) and dried in a SpeedVac concentrator. Dried peptides were resolubilized in 20 L HPLC grade water containing 0.1% formic acid. Sample analysis was performed on a linear ion trap LTQ mass spectrometer (Thermo Electron, San Jose, CA) equipped with a nanoelectrospray ion source (Thermo Electron) coupled to an Agilent 1100 micro HPLC system. Peptides were loaded with a cooled Agilent autosampler on a 2 cm long pre-column filled with C18 resin (Magic C18 AQ 5 m; Michrom Bioresources, Auburn, CA, USA). A linear gradient of 80min from 5% to 40% acetonitrile in H2O with 0.1% formic acid was used to separate peptides on a 10 cm long fused silica emitter packed with C18 resin spraying directly into the mass spectrometer at a flow rate of 0.5 ul/min. The MS instrument was operated in positive ion mode. The data-dependent acquisition mode was set to acquire one MS scan followed by three collision induced dissociation MS/MS scans. The MS full scans were recorded over a mass range of 400-1600 m/z. Dynamic exclusion was enabled, the repeat count was set to 2 and the exclusion duration to 30s. Further MS conditions

96

Chapter 4

were set as following: spray voltage 1.95 kV, transfer capillary temperature 230 C, normalized collision energy 35%, activation q 0.25 and activation time 30 ms. The acquired raw files were converted to mzXML files using ReAdW with default settings and searched against the mouse IPI database version 3.26 with the Sequest search algorithm. The sequest search parameters contained the static modification of cysteine +57.02 Da, at least one tryptic terminus and one missed cleavage was allowed. The data were further processed using the using the Trans-Proteomic Pipeline TPP including PeptideProphet and ProteinProphet to estimate the false discovery rate in the datasets. A protein probability of 0.5 was set as a cutoff corresponding to a false discovery rate of approximately 5%. The protein list was annotated for secreted proteins using the algorithm SignalP and further manually curated for secreted proteins using UniProt database and literature search. Functional annotation was assigned using the PANTHER Classification system. Venndiagrams were created with Vennmaster. qPCR expression analysis mRNA from visceral and subcutaneous SVF as well as virus-infected 3T3-L1 cells was isolated and transcribed to cDNA using the MultiMACS cDNA synthesis kit (Miltenyi). mRNA expression was assessed by real-time PCR using SybrGreen qPCR reagent (Invitrogen). Expression was normalized to Gapdh and -actin. For primer sequences see Suppl. Table I. Adenovirus screen For adenoviral shRNA knockdown, shRNA-sequences were designed using

RNAidesigner (Invitrogen) and shRNA-oligos were cloned into pENTR_U6 containing a modified linker with EcoRI and Age I restriction sites (Suppl. Table II). pENTR_U6 vectors were recombined with pAdBLOCK-iT-DEST using LRII clonase (Invitrogen). Crude adenoviral stock was produced transfecting HEK293A cells with 1 g of PacI digested adenoviral plasmid according to the manufacturers protocol. Crude viral lysate was used to reinfect HEK293A cells and virus from lysed cells was purified with the Adeno MINI Purification ViraKit (Virapur).

97

Chapter 4

For infection, purified virus was diluted 1:10 in Opti-MEM with 0.0001% poly-L-lysine and incubated for 90 min at room temperature. Cells were then incubated with diluted virus at 37C and after 2 h an equal amount of medium was added. Virus-containing medium was changed after 24 h and cells were induced for differentiation 48 h after infection. Cells were fixed, stained and analyzed as described above. For the adenoviral shRNA screen 3 independent rounds of differentiation for each cell type were performed. Percent differentiation was normalized to mean differentiation of each round and cell type. For every shRNA a mean normalized differentiation over all rounds was calculated for every cell type. shRNAs enhancing or inhibiting differentiation compared to control shRNA were selected based on a two-tailed Students t-test. Protein overexpression For cloning of overexpression constructs, cDNA was amplified from murine fat or 3T3-L1 cDNA (for primer sequences see Suppl. Table III) and ligated into pSecTagA containing an Ig-leader sequence for high protein secretion and a c-myc/his-tag for protein purification. cDNA was also cloned into pENTR_CMV_MCS_TkPA (Invitrogen) adding a C-terminal HA-tag to produce adenoviral overexpression constructs. Westernblot analysis HEK293A cells were transfected with FuGENE (Roche) according to the manufacturers protocol. Medium was changed to OptiMEM after 12h. Cells were lysed and supernatant was collected after 48h and whole cell extract as well as supernatant was subjected to 12% SDS gel and Western blotting using mouse anti-HA antibody (Covance) or mouse anit-myc antibody (kind gift of Gudrun Christiansen). Statistical Analysis Results are given as mean +/- standard deviation. Statistical analyses were performed using a two-tailed Students t-test.

98

Chapter 4

Results
As cell size plays an important role for intact adipocyte function, we measured adipocyte size in visceral epididymal as well as in subcutaneous adipose tissue of C57Bl/6J mice aged 12 weeks. As shown in Fig. 1.1a, adipocytes in the epididymal fat pad were significantly larger than in subcutaneous adipose tissue.
a subcutaneous adipose tissue visceral adipose tissue

200 m

200 m

200 m

200 m

60

% of adipocytes

50 40 30 20 10 0

subcutaneous visceral * * *

00

00

00

00

00 -8 00 80 00

0 -9

00

00

10

00

-5

-6

-2

-3

0-

00

00

00

-4

00

00

50

60

00

-7

2 adipocyte area [m ]

Figure 1.1 a) Microscopic pictures of HE-stained paraffin sections from subcutaneous and visceral adipose tissue from 12 weeks old male C57Bl/6J mice. Values represent means SD, *p<0.05
99

70

10

20

30

40

00

Chapter 4

Since adipocyte size is determined by the balance between adipocyte hypertrophy versus hyperplasia, we measured the capacity of adipocyte precursors in the SVF to undergo de novo adipogenesis. Upon induction with a differentiation cocktail containing insulin, dexamethasone and isobutyl-methylxanthine, subcutaneous SVF had a higher potential to differentiate into lipid-droplet containing adipocytes than epididymal SVF (Fig. 1.2b).

***
b

subcutaneous SVF

visceral SVF

35

% differentiated cells

30 25 20 15 10 5 0
us eo an sc vi er al

Figure 1.2 b) In vitro differentiation of plated subcutaneous and visceral SVF induced with insulin, dexamethasone and IBMX and fixed on day 7 of differentiation. Pictures show differentiated SVF stained with BODIPY for lipid droplets (green), Cyto60 (red) and Hoechst (blue). (n=4) Scale: 100 m. Values represent means SD, ***p<0.001

100

su

bc

ut

Chapter 4

We hypothesized that this is due to an increased adipocyte precursor number within the subcutaneous SVF. In order to compare precursor numbers in the subcutaneous and visceral compartment, the Lin-CD29+CD34+Sca1+ cell population as well as preadipocyte factor 1 (Pref1) positive cells were analyzed by flow cytometry in different fat compartments. Surprisingly, both the Lin-CD29+CD34+Sca1+ population (sc 11.2 0.7%, vis 36.3 0.7% of live cells, p<0.001) and the Lin-Pref1+ population (sc 0.8 0.1%, vis 6.7 0.3% of live cells, p<0.01) as well as the population positive for all markers (sc 0.5 0.0%, vis 1.6 0.1% of live cells, p<0.01) showed a higher abundance in visceral (Fig. 2.1) than in subcutaneous (Fig. 2.2) adipose tissue. As adipocyte precursor numbers are not the cause of impaired differentiation capacity in visceral adipose tissue, we next tested whether paracrine factors secreted from the SVF modulate adipocyte differentiation. To do so, differentiation assays with conditioned media that had been incubated with subcutaneous or visceral SVF beforehand and therefore contained secreted factors from SVF, were performed. Differentiation in conditioned media from subcutaneous and visceral SVF lowered adipocyte numbers in subcutaneous SVF to 24 % and 12 % of the original differentiation capacity, respectively. On the other hand, differentiation of visceral SVF was again low and not further affected by conditioned media (Fig. 3.1a, black bars). As a first characterization of the nature of inhibitory factors present in the supernatant, cells were additionally treated with conditioned media that had been heat inactivated for 10 min at 95C. Heat-inactivation of unconditioned media did not significantly alter the differentiation capacity of SVF. Heat-inactivation of conditioned media, however, reversed the inhibitory effect completely for the supernatant from subcutaneous SVF and to a large extent for the supernatant from visceral SVF (Fig. 3.1a, grey bars). We therefore hypothesized that heat-sensitive factors, such as proteins secreted from the SVF, exert an inhibitory effect on adipogenesis.

101

Chapter 4

Figure 2.1 a) Fluorescent t sta staining of visceral SVF (n=4) with 7-aad for r liv live cell staining, Ter119-Alexa488, CD45-Ale Alexa488, and CD31-Alexa488 for Lin selec election, CD29-PE, CD34-Alexa647, Sca1-PE-Cy7 Cy7 and Pref1-APC-Cy7. Shown are example e sca scatter plots with gating on FSC/SSC and 7-aa aad- cells (all-live), staining with all markers kers on the all-live population and sequential tial gating on LinCD29+CD34+Sca1+Pref1+ cells cell b) % parent populations of LinCD29 CD2 +CD34+Sca1+Pref1+ and LinPref1+ cells cel in visceral + + + + SVF c) % all live populations ons of Lin CD29 CD34 Sca1 Pref1 in visceral SVF SV
102

Chapter 4

Figure 2.2 d) Fluorescent t st staining of subcutaneous SVF (n=4) e) % parent pare populations + + + + + of Lin CD29 CD34 Sca1 Pref1 Pr and Lin Pref1 cells in subcutaneous SVF f) % all-live + + population of Lin CD29 CD34 CD3 Sca1+Pref1+ in subcutaneous SVF

103

Chapter 4

*** ** *** ** *** *

30
% differentiated cells

non-treated heat inactivation

20
***

10

Figure 3.1 a) Differentiation of subcutaneous and visceral SVF using normal medium and conditioned medium from subcutaneous/visceral SVF. The medium was either used directly or heat-inactivated at 95C for 10 min. (n=4) *p<0.05 **p<0.01 ***p<0.001 Scale: 100 m, SN: supernatant

104

co nt ro lm ed iu m su bc ut .S N vi sc er co al nt SN ro lm ed iu m su bc ut .S N vi sc er al SN

subcut. SVF

visceral SVF

Chapter 4

In order to identify secreted proteins from the SVF, we analyzed supernatant from cultured subcutaneous and visceral SVF by mass spectrometry. The majority of identified peptides was derived from secreted proteins. Out of 89 secreted proteins in the supernatant of subcutaneous SVF and 72 proteins in the visceral supernatant, 48 proteins were found to be secreted from both compartments corresponding to an overlap of 42% (Fig. 3.2b, Suppl. Table IV and V). Using the panther classification system on all identified proteins, the biggest group of proteins was classified as extracellular matrix proteins followed by the functional class of proteases (Fig. 3.2c).

visceral

24 48 41

subcutaneous
c

Figure 3.2 b) Venn-diagram of secreted proteins identified by mass spectrometry in the supernatant of subcutaneous and visceral SVF. c) Functional Panther classification of all identified proteins in the supernatant of subcutaneous and visceral SVF.

105

Chapter 4

To narrow down the list of potential candidates, proteins that had previously been shown to promote adipocyte differentiation such as insulin growth factor I (IGF-I)50 were discarded from the candidate list. Similarly, all types of collagens, which have already been studied intensively in preadipocyte/adipocyte conversion51 were neglected. To further reduce the number of interesting candidates that potentially lead to a strong inhibition of adipogenesis in visceral SVF, we next analyzed expression levels of the remaining 45 proteins in subcutaneous and visceral SVF by qPCR. Normalization to -actin or Gapdh essentially led to the same results. mRNA levels of 16 proteins were significantly higher expressed in visceral compared to subcutaneous SVF and were included in a secondary functional screen (Fig. 3.3d). Given the fact that protein activity can be regulated by other means than only mRNA expression, 6 further candidates were included into the adipocyte differentiation screen that had previously been described to modulate differentiation processes (e.g. colony-stimulating factor 152 and cadherin353). For each of these proteins, three different shRNAs were designed and corresponding adenoviruses were produced. Additionally, adenovirus expressing scrambled shRNA (negative control), GFP-overexpressing adenovirus (transfection control) and adenovirus expressing shRNA against Timp1 (positive control) were generated. Purified adenoviruses were used for infection of subcutaneous and visceral SVF as well as 3T3-L1 preadipocytes and three independent differentiation experiments were performed for each cell type. After normalization, candidate proteins were selected, for which differentiation was significantly enhanced by at least 2 shRNAs in visceral SVF (5 proteins) or by 1 shRNA in visceral SVF and at least 2 shRNAs in subcutaneous SVF or 3T3-L1 cells (2 proteins) (Suppl. Table VI). For these 7 proteins biglycan (Bgn), decorin (Dcn), bone morphogenic protein 1 (Bmp1), epidermal growth factor-containing fibulin-like extracellular matrix protein 2 (Efemp2), elastin microfibril interfacer 1 (Emilin1), matrix gla protein (Mgp) and SPARC-like 1 (Sparcl1), mRNA knockdown was verified by qPCR in 3T3-L1 cells (Fig. 4). >50% knockdown could be achieved with the majority of shRNA constructs, except for Bmp1.

106

50

subcutaneous visceral

40

***

***

30

20

10

* * *
p=0.07

* * ** * ** * * * * *

mRNA expression relative to -a -actin

e 9 n 1 8 3 f1 n 1 1 2 1 p 1 n 1 f1 3 4 7 ra l1 p 2 m p 0 2 e tn p rf 2 3 n f1 1 t3 l1 3 r3 2 1 2 0 Ac am1 Bg mp Cc1 Cdh Cs Dc cm mp mp ilin Enp Fn Gs Htra Ig fbp fbp fbp l15 Il1r Lb Lcn Lu Mg p1 mp colc os Psa Ptp rres Saa na2 pin ing Sli arc ock gfb hbs imp imp em2 i r p m E fe fe m B Ig Ig Ig I M M P P E E E rp Se Ser Sp Sp T T T T Tm Ra Ad e S

Chapter 4

107

108
normalized mRNA expression
0.0 0.5 1.0 1.5 0.00 0.25 0.50 0.75 1.00 1.25 0.0 0.5
G sh FP RN A co nt ro l sh RN A co nt ro l G FP sh RN A1 sh RN A1 sh RN A2 sh RN A3

Chapter 4

normalized mRNA expression

0.75 1.00 1.25
G sh FP RN A co nt ro l sh RN A1

normalized mRNA expression

normalized mRNA expression

1.0 1.5

0.00

0.25

0.50

G sh FP RN A co nt ro l

sh RN A1

Dcn

Bgn

Emilin1

Sparcl1

*
sh RN A2 sh RN A2 sh RN A3 sh RN A3

sh RN A2

* *
normalized mRNA expression
0 0 1 2
G sh FP RN A co nt ro l sh RN A1 G sh FP RN A co nt ro l

sh RN A3

*** *
sh RN A1

normalized mRNA expression

1 2
G sh FP RN A co nt ro l sh RN A1

normalized mRNA expression

0 1 2

Bmp1

Efemp2

Mgp

* * *

sh RN A2

sh RN A2 sh RN A3

sh RN A2 sh RN A3

Figure 4 qPCR analysis of adenoviral knockdown 48h after infection of 3T3-L1 cells for selected candidate proteins as indicated. Results are normalized to Gapdh (n=3). Values represent means SD, *p<0.05 **p<0.01 ***p<0.001
sh RN A3

Chapter 4

To confirm the potential role of these proteins in adipogenesis, overexpression constructs were cloned in order to purify recombinant protein from cell culture supernatant or to use adenoviral overexpression in primary SVF culture. Using the secretion vector pSecTagA with an Ig-leader sequence only Emilin1 and Pro-Bgn were secreted into the supernatant in sufficient amounts (Fig. 5a). First purification attempts of Emilin1 led to insufficient protein concentrations and purity and will be repeated in a larger scale (data not shown). All proteins except for Sparcl1 and Bmp1 could also be expressed using an adenoviral entry vector, which will allow adenoviral overexpression for in vitro and in vivo differentiation assays (Fig. 5b). Murine recombinant Dcn and Sparcl1 are commercially available and led to a strong inhibition of adipocyte differentiation in subcutaneous and visceral SVF (Fig. 5c,d) complementing the knockdown results. Further experiments will need to confirm the inhibitory potential of the other identified proteins in vitro and in vivo.

109

Chapter 4

pSecTagA Pro-Bgn Bgn Bmp1 Efemp2 Emilin1 Mgp wce SN wce SN wce SN wce SN wce SN wce SN

pENTR Bgn Bmp1 Dcn Efemp2 Emilin1 Mgp Sparcl1 wce SN wce SN wce SN wce SN wce SN wce SN wce SN

45kDa 43kDa 110kDa 55kDa 130kDa

17kDa

45kDa

43kDa

55kDa

130kDa 17kDa

c
30

d
% differentiation

30

% differentiation

20

20

*** *** *** ***

10

10

*** *** *** *** *** *** ***

0

*** ***
0

***
5

3. 8

7. 5

7. 5

30

15

15

30

3 1.

2.

0.

1.

subcutaneous SVF Decorin

visceral SVF [g/ml]

subcutaneous SVF

visceral SVF

Sparcl1 [g/ml]

subcut.

visceral

subcut.

visceral

n.d.

Decorin

Figure 5 Westernblots of protein overexpression in the whole cell extract (wce) or supernatant (SN) of transiently transfected HEK293A cells (observed molecular weights are indicated) a) pSecTagA constructs detected by anti-myc Ab b) pENTR constructs detected by anti-HA Ab c, d) differentiation of subcutaneous and visceral SVF in the presence of murine recombinant Dcn or Sparcl1. Protein was added one day before and 2 days after induction of differentiation. Cells were fixed and stained at day 7 of differentiation (n=3) Scale: 100 m, *p<0.05 **p<0.01 ***p<0.001
110

Sparcl1

2.

n.d.

Chapter 4

Discussion
Adipocyte precursor cells are responsible for de novo adipocyte formation in adipose tissue growth. We and others show that the capacity of adipose tissue to generate new adipocytes is dependent on the fat depot with murine subcutaneous fat having a greater adipogenic potential than visceral adipose tissue16. In male adult mice, the impaired differentiation capacity of intraabdominal adipose tissue is accompanied by an increased adipocyte size in epididymal compared to subcutaneous fat depots. Interestingly, Tchkonia et al.54 also observed differences in adipogenic potential in a small cohort of human female and male obese subjects. In this study, subcutaneous adipocyte precursor cells show highest adipogenicity followed by SVF from mesenteric and omental fat with much lower adipogenic potential, thus mirroring the results in mice. Adding to the theory of depot-specific differences, Permana et al. show in a cohort of obese Pima Indians that subcutaneous differentiation is inversely correlated with central obesity55. Furthermore, treatment with glitazones leads to a beneficial redistribution of fat towards subcutaneous depots56. It can therefore be speculated that a high differentiation capacity in subcutaneous adipose tissue can buffer excess energy and prevent accumulation of intraabdominal fat, which poses a greater risk factor for metabolic complications43-44,57. When looking at regional fat distribution, sex differences play an important role. Female mice show a higher differentiation capacity in subcutaneous inguinal adipose tissue than male mice58 and the ratio of subcutaneous to visceral adipose tissue is higher in women compared to men59, which is in part explainable by the effect of sex steroids on adipose tissue development60-62. The capacity to rebuild new adipocytes also seems to be age dependent, as epidydimal SVF of young mice show a rate of in vitro adipocyte differentiation similar to that of subcutaneous SVF63 (own observations). In prepubertal children, high differentiation rates can be detected in subcutaneous and visceral SVF64 and adipocyte cell numbers increase rapidly in lean and even more pronounced in obese children65. Furthermore, young women have a higher total fat cell number compared to older women within the same BMI range66, thereby supporting the theory of declining adipocyte differentiation with age67. When

111

Chapter 4

studying adipogenic processes it will therefore be informative to investigate fat depotspecific differences also in the context of age and sex. It has been shown that secreted factors from adipose tissue modulate adipocyte differentiation. The majority of studies investigated the influence of secreted factors from isolated adipocytes on adipocyte differentiation, which promote or inhibit adipogenesis depending on the species and culturing conditions30-32. Secreted factors from the stromal-vascular cell fraction of adipose tissue, however, have so far largely been neglected. Two studies address the impact of endothelial cells on adipogenesis of human or rat SVF with different outcome, potentially due to species differences or different culture conditions33,68. Here we show that conditioned medium from murine SVF contains heat-sensitive factors that inhibit adipogenesis in primary adipocyte precursor cells. This effect is more pronounced in visceral compared to subcutaneous SVF, which could be part of the reason for the enhanced adipogenic potential of subcutaneous compared to visceral SVF. In fact, in the hematopoietic stem cell field it is a well-established concept that stem cells can switch from a dormant to an active self-renewing state dependent on their microenvironment69. This concept might be true in a similar way for adipocyte progenitor cells, especially given the fact that progenitor cell numbers are not higher in subcutaneous than in visceral SVF. Under the assumption that secreted proteins are responsible for the inhibitory effect, we performed mass spectrometry analysis of supernatant from subcutaneous and visceral SVF. Many studies explore protein secretion from adipose tissue, often with a focus on adipocytes or whole adipose tissue explants70-73. The secretome of isolated human SVF or 3T3-L1 preadipocytes has also been analyzed in several reports74-77. However, these studies are focused on identifying differentially secreted proteins during adipocyte differentiation, none of the identified proteins has been characterized further with regard to their function in adipogenesis. Here, we combine an unbiased identification approach of secreted proteins from murine SVF with a functional analysis in adipocyte differentiation. Of all identified proteins 22 were chosen for a functional characterization based on an enhanced mRNA expression in visceral compared to subcutaneous SVF or prior literature reports about an implication

112

Chapter 4

in differentiation processes. Knockdown of seven proteins led to enhanced adipocyte differentiation in primary SVF and preliminary results confirm an inhibitory effect of Dcn and Sparcl1 protein on adipogenesis. Whereas for none of the identified proteins a direct influence on adipocyte differentiation has been shown so far, some are described to be regulated during the differentiation process. Furthermore, many of the identified proteins are functionally interconnected, potentially forming a cluster of proteins regulating adipogenesis (Fig. 6). Decorin and biglycan are extracellular small leucine-rich proteoglycans that bind to a variety of collagens78-81 and to transforming growth factor (TGF)82. Both proteoglycans are highly expressed in 3T3-L1 preadipocytes and are readily downregulated upon induction of adipocyte differentiation, hinting at an inhibitory role in differentiation83-85. Their expression is also upregulated in adipose tissue of obese mice and differentially expressed in subcutaneous and visceral fat depots86. Decorin modulates attachment of dermal fibroblasts to fibronectin87 and is implicated in collagen fibril formation88. Given the fact that fibronectin and different types of collagen are important modulators of adipocyte differentiation89-91, this is a likely pathway by which decorin inhibits adipocyte differentiation. Both decorin and biglycan are secreted with a propeptide that can be cleaved off by bone morphogenic protein 1 (Bmp1), another protein that has been identified in the functional screen92-93. In contrast to other bone morphogenic proteins, which mostly induce adipocyte differentiation94-95, Bmp1 functions as a metalloproteinase, which can process different types of collagens96 and might thereby influence adipocyte differentiation. Matrix gla protein (Mgp) is a small, ubiquitous extracellular protein that modulates osteogenic and chondrogenic differentiation97-98. Direct interaction of Mgp with Bmp2 thereby inhibiting Bmp2 activity has been shown97,99. As Bmp2 is a positive modulator of adipocyte differentiation100-101, Mgp might sequester Bmp2 in adipose tissue thereby controlling adipogenic differentiation. Secreted protein acidic and rich in cysteine like-1 (Sparcl1/hevin) is a calciummodulated protein that binds to collagen type I102. In vitro, Sparcl1 blocks cell proliferation and has been suggested as a tumor suppressor gene103. Interestingly,

113

Chapter 4

Sparcl1 modulates collagen fibril formation and stimulates decorin secretion in the dermis104, a mechanism that has not been investigated in adipose tissue so far. Moreover, the closely related protein Sparc is highly expressed in adipose tissue and negatively regulates adipogenesis and fat accumulation105-106. EGF-containing fibulin-like extracellular matrix protein 2 (Efemp2/fibulin4) and elastin microfibril interfacer 1 (Emilin1) are two proteins associated with elastic fibers. Efemp2 knockout mice die around birth because of aneurysm rupture107. At an ultrastructural level, elastin crosslinks are diminished and the elastic lamellae are fragmented by loss of Efemp2. Emilin1 is located at the interface between the amorphous core and the surrounding microfibrils of elastic fibers. In embryonic fibroblasts lacking Emilin1, formation of elastic fibers is abnormal and anchorage of endothelial and smooth muscle cells to elastic fibers is disrupted108. So far, the role of Efemp2 and Emilin1 and the influence of elastic fibers on adipogenesis has not been studied in adipose tissue, but might play a crucial role in adipose tissue development. Further overexpression studies in vitro and in vivo will be necessary to prove the role of this list of identified proteins in adipogenesis. Given the close interaction with extracellular matrix structures, it will be instructive to study the remodeling processes during adipogenesis in greater detail in order to shed light on the mechanistic processes, by which these proteins inhibit adipogenesis. Kawaguchi et al.109 describe in detail, how intracellular F-actin, extracellular fibronectin and activated cell surface 1integrin, which links extra- and intracellular structures, change during differentiation dependent on ADAM12. Similar experiments linked to the list of identified proteins will give further mechanistic insights. Lastly, it will also be of interest to subdivide the SVF into different cell types by cell sorting in order to analyze, which cell type expresses these inhibitory factors.

114

Chapter 4

Emilin1

microfibril (Fibrillin, Efemp2 etc.) survival differentiation

amorphous core (Elastin) elastic fiber

survival proliferation differentiation preadipocyte collagen fiber Sparcl1

growth factors Bgn Dcn + Bmp1 Bmp1

Bmp1

Dcn, Bgn expression adipocyte differentiation

Figure 6 Hypothetical mechanisms by which the identified proteins inhibit adipocyte differentiation. Emilin1 and Efemp2 are components of the elastic fiber and Emilin1 mediates cell adhesion to the elastic fiber. Mgp binds to and inhibits Bmp2, which is known to promote differentiation. Dcn and Bgn bind to collagen fibers and sequester growth factors. Both are upregulated in obesity and downregulated upon induction of adipocyte differentiation. Sparcl1 stimulates Dcn secretion in dermal fibroblasts. Bmp1 has metallo-proteinase activity and cleaves the pro-peptides of Dcn and Bgn as well as different types of collagen.

115

Chapter 4

References
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. Shi, Y. & Burn, P. Lipid metabolic enzymes: emerging drug targets for the treatment of obesity. Nat Rev Drug Discov 3, 695-710 (2004). Carmen, G.Y. & Victor, S.M. Signalling mechanisms regulating lipolysis. Cell Signal 18, 401-408 (2006). Ahima, R.S. Adipose tissue as an endocrine organ. Obesity (Silver Spring) 14 Suppl 5, 242S-249S (2006). Calabro, P. & Yeh, E.T. Obesity, inflammation, and vascular disease: the role of the adipose tissue as an endocrine organ. Sub-cellular biochemistry 42, 63-91 (2007). Rosen, E.D. & Spiegelman, B.M. Adipocytes as regulators of energy balance and glucose homeostasis. Nature 444, 847-853 (2006). Wajchenberg, B.L. Subcutaneous and visceral adipose tissue: their relation to the metabolic syndrome. Endocr Rev 21, 697-738 (2000). Jo, J., et al. Hypertrophy and/or Hyperplasia: Dynamics of Adipose Tissue Growth. PLoS computational biology 5, e1000324 (2009). Guo, W., et al. New scanning electron microscopic method for determination of adipocyte size in humans and mice. Obesity (Silver Spring) 15, 1657-1665 (2007). Jernas, M., et al. Separation of human adipocytes by size: hypertrophic fat cells display distinct gene expression. FASEB J 20, 1540-1542 (2006). Spalding, K.L., et al. Dynamics of fat cell turnover in humans. Nature 453, 783-787 (2008). Gesta, S., Tseng, Y. & Kahn, C.R. Developmental origin of fat: tracking obesity to its source. Cell 131, 242-256 (2007). Rosen, E.D. & MacDougald, O.A. Adipocyte differentiation from the inside out. Nature Reviews Molecular Cell Biology 7, 885-896 (2006). Zuk, P.A., et al. Multilineage cells from human adipose tissue: implications for cell-based therapies. Tissue Eng 7, 211-228 (2001). Hauner, H., et al. Promoting effect of glucocorticoids on the differentiation of human adipocyte precursor cells cultured in a chemically defined medium. J Clin Invest 84, 1663-1670 (1989). Bjorntorp, P., et al. Isolation and characterization of cells from rat adipose tissue developing into adipocytes. J Lipid Res 19, 316-324 (1978). Joe, A.W., Yi, L., Even, Y., Vogl, A.W. & Rossi, F.M. Depot-specific differences in adipogenic progenitor abundance and proliferative response to high-fat diet. Stem Cells 27, 2563-2570 (2009). Suga, H., et al. Functional implications of CD34 expression in human adipose-derived stem/progenitor cells. Stem Cells Dev 18, 1201-1210 (2009). Rodeheffer, M.S., Birsoy, K. & Friedman, J.M. Identification of white adipocyte progenitor cells in vivo. Cell 135, 240-249 (2008). Garces, C., Ruiz-Hidalgo, M.J., Bonvini, E., Goldstein, J. & Laborda, J. Adipocyte differentiation is modulated by secreted delta-like (dlk) variants and requires the expression of membraneassociated dlk. Differentiation 64, 103-114 (1999). Moon, Y.S., et al. Mice lacking paternally expressed Pref-1/Dlk1 display growth retardation and accelerated adiposity. Molecular and cellular biology 22, 5585-5592 (2002). Ross, S.E., et al. Inhibition of adipogenesis by Wnt signaling. Science 289, 950-953 (2000). Chapman, A.B., Knight, D.M. & Ringold, G.M. Glucocorticoid regulation of adipocyte differentiation: hormonal triggering of the developmental program and induction of a differentiation-dependent gene. J Cell Biol 101, 1227-1235 (1985). Klemm, D.J., et al. Insulin-induced adipocyte differentiation. Activation of CREB rescues adipogenesis from the arrest caused by inhibition of prenylation. J Biol Chem 276, 28430-28435 (2001). Simon, M.F., et al. Lysophosphatidic acid inhibits adipocyte differentiation via lysophosphatidic acid 1 receptor-dependent down-regulation of peroxisome proliferator-activated receptor gamma2. J Biol Chem 280, 14656-14662 (2005).

17. 18. 19.

20. 21. 22.

23.

24.

116

Chapter 4
25. Xing, H., et al. TNF alpha-mediated inhibition and reversal of adipocyte differentiation is accompanied by suppressed expression of PPARgamma without effects on Pref-1 expression. Endocrinology 138, 2776-2783 (1997). Sakaue, H., et al. Requirement of fibroblast growth factor 10 in development of white adipose tissue. Genes Dev 16, 908-912 (2002). Chavey, C., et al. Matrix metalloproteinases are differentially expressed in adipose tissue during obesity and modulate adipocyte differentiation. J Biol Chem 278, 11888-11896 (2003). Selvarajan, S., Lund, L.R., Takeuchi, T., Craik, C.S. & Werb, Z. A plasma kallikrein-dependent plasminogen cascade required for adipocyte differentiation. Nat Cell Biol 3, 267-275 (2001). Bi, Y., et al. Extracellular matrix proteoglycans control the fate of bone marrow stromal cells. J Biol Chem 280, 30481-30489 (2005). Shillabeer, G., Forden, J.M. & Lau, D.C. Induction of preadipocyte differentiation by mature fat cells in the rat. J Clin Invest 84, 381-387 (1989). Marques, B.G., Hausman, D.B. & Martin, R.J. Association of fat cell size and paracrine growth factors in development of hyperplastic obesity. Am J Physiol 275, R1898-1908 (1998). Janke, J., Engeli, S., Gorzelniak, K., Luft, F.C. & Sharma, A.M. Mature adipocytes inhibit in vitro differentiation of human preadipocytes via angiotensin type 1 receptors. Diabetes 51, 16991707 (2002). Lau, D.C., Shillabeer, G., Wong, K.L., Tough, S.C. & Russell, J.C. Influence of paracrine factors on preadipocyte replication and differentiation. Int J Obes 14 Suppl 3, 193-201 (1990). Guilherme, A., Virbasius, J.V., Puri, V. & Czech, M.P. Adipocyte dysfunctions linking obesity to insulin resistance and type 2 diabetes. Nat Rev Mol Cell Biol 9, 367-377 (2008). Kahn, S.E., Hull, R.L. & Utzschneider, K.M. Mechanisms linking obesity to insulin resistance and type 2 diabetes. Nature 444, 840-846 (2006). Van Gaal, L.F., Mertens, I.L. & De Block, C.E. Mechanisms linking obesity with cardiovascular disease. Nature 444, 875-880 (2006). Skurk, T., Alberti-Huber, C., Herder, C. & Hauner, H. Relationship between adipocyte size and adipokine expression and secretion. J Clin Endocrinol Metab 92, 1023-1033 (2007). Salans, L.B. & Doughert.Jw. Effect of Insulin Upon Glucose Metabolism by Adipose Cells of Different Size - Influence of Cell Lipid and Protein Content, Age, and Nutritional State. Journal of Clinical Investigation 50, 1399-& (1971). Livingston, J.N., Cuatrecasa, P. & Lockwood, D.H. Insulin insensitivity of large fat cells. Science 177, 626-628 (1972). Slawik, M. & Vidal-Puig, A.J. Adipose tissue expandability and the metabolic syndrome. Genes Nutr 2, 41-45 (2007). Tan, C.Y. & Vidal-Puig, A. Adipose tissue expandability: the metabolic problems of obesity may arise from the inability to become more obese. Biochem Soc Trans 36, 935-940 (2008). Virtue, S. & Vidal-Puig, A. Adipose tissue expandability, lipotoxicity and the Metabolic Syndrome--an allostatic perspective. Biochim Biophys Acta 1801, 338-349 (2010). Kissebah, A.H. & Krakower, G.R. Regional adiposity and morbidity. Physiol Rev 74, 761-811 (1994). Despres, J.P. & Lemieux, I. Abdominal obesity and metabolic syndrome. Nature 444, 881-887 (2006). Gesta, S., et al. Evidence for a role of developmental genes in the origin of obesity and body fat distribution. Proc Natl Acad Sci U S A 103, 6676-6681 (2006). Vohl, M.C., et al. A survey of genes differentially expressed in subcutaneous and visceral adipose tissue in men. Obesity Research 12, 1217-1222 (2004). Wajchenberg, B.L., Giannella-Neto, D., da Silva, M.E. & Santos, R.F. Depot-specific hormonal characteristics of subcutaneous and visceral adipose tissue and their relation to the metabolic syndrome. Horm Metab Res 34, 616-621 (2002). Hansen, L.H., Madsen, B., Teisner, B., Nielsen, J.H. & Billestrup, N. Characterization of the inhibitory effect of growth hormone on primary preadipocyte differentiation. Molecular Endocrinology 12, 1140-1149 (1998). Chen, H.C. & Farese, R.V. Determination of adipocyte size by computer image analysis. Journal of Lipid Research 43, 986-989 (2002).

26. 27. 28. 29. 30. 31. 32.

33. 34. 35. 36. 37. 38.

39. 40. 41. 42. 43. 44. 45. 46. 47.

48.

49.

117

Chapter 4

50.

51. 52.

53. 54. 55.

56. 57. 58.

59. 60.

61. 62. 63. 64. 65.

66. 67. 68.

69. 70. 71.

Schmidt, W., Poll-Jordan, G. & Loffler, G. Adipose conversion of 3T3-L1 cells in a serum-free culture system depends on epidermal growth factor, insulin-like growth factor I, corticosterone, and cyclic AMP. J Biol Chem 265, 15489-15495 (1990). Gregoire, F.M., Smas, C.M. & Sul, H.S. Understanding adipocyte differentiation. Physiological Reviews 78, 783-809 (1998). Umezawa, A., et al. Colony-stimulating factor 1 expression is down-regulated during the adipocyte differentiation of H-1/A marrow stromal cells and induced by cachectin/tumor necrosis factor. Molecular and cellular biology 11, 920-927 (1991). Castro, C.H., et al. Targeted expression of a dominant-negative N-cadherin in vivo delays peak bone mass and increases adipogenesis. J Cell Sci 117, 2853-2864 (2004). Tchkonia, T., et al. Fat depot origin affects adipogenesis in primary cultured and cloned human preadipocytes. Am J Physiol Regul Integr Comp Physiol 282, R1286-1296 (2002). Permana, P.A., et al. Subcutaneous abdominal preadipocyte differentiation in vitro inversely correlates with central obesity. American Journal of Physiology-Endocrinology and Metabolism 286, E958-E962 (2004). Smith, S.R., et al. Effect of pioglitazone on body composition and energy expenditure: a randomized controlled trial. Metabolism 54, 24-32 (2005). O'Connell, J., et al. The relationship of omental and subcutaneous adipocyte size to metabolic disease in severe obesity. PLoS One 5, e9997 (2010). Ogawa, R., et al. Adipogenic differentiation by adipose-derived stem cells harvested from GFP transgenic mice-including relationship of sex differences. Biochem Biophys Res Commun 319, 511-517 (2004). Enzi, G., et al. Subcutaneous and visceral fat distribution according to sex, age, and overweight, evaluated by computed tomography. Am J Clin Nutr 44, 739-746 (1986). Heine, P.A., Taylor, J.A., Iwamoto, G.A., Lubahn, D.B. & Cooke, P.S. Increased adipose tissue in male and female estrogen receptor-alpha knockout mice. Proc Natl Acad Sci U S A 97, 1272912734 (2000). Krakower, G.R., et al. Regional adipocyte precursors in the female rat. Influence of ovarian factors. J Clin Invest 81, 641-648 (1988). Zhang, Y., et al. Response of the adipose tissue transcriptome to dihydrotestosterone in mice. Physiol Genomics 35, 254-261 (2008). Prunet-Marcassus, B., et al. From heterogeneity to plasticity in adipose tissues: Site-specific differences. Experimental Cell Research 312, 727-736 (2006). Grohmann, M., et al. Characterization of differentiated subcutaneous and visceral adipose tissue from children: the influences of TNF-alpha and IGF-I. J Lipid Res 46, 93-103 (2005). Knittle, J.L., Timmers, K., Ginsberg-Fellner, F., Brown, R.E. & Katz, D.P. The growth of adipose tissue in children and adolescents. Cross-sectional and longitudinal studies of adipose cell number and size. J Clin Invest 63, 239-246 (1979). Ktotkiewski, M., Sjostrom, L., Bjorntorp, P. & Smith, U. Regional adipose tissue cellularity in relation to metabolism in young and middle-aged women. Metabolism 24, 703-710 (1975). Kirkland, J.L. & Dobson, D.E. Preadipocyte function and aging: links between age-related changes in cell dynamics and altered fat tissue function. J Am Geriatr Soc 45, 959-967 (1997). Rajashekhar, G., et al. IFATS collection: Adipose stromal cell differentiation is reduced by endothelial cell contact and paracrine communication: role of canonical Wnt signaling. Stem Cells 26, 2674-2681 (2008). Wilson, A., et al. Dormant and self-renewing hematopoietic stem cells and their niches. Annals of the New York Academy of Sciences 1106, 64-75 (2007). Alvarez-Llamas, G., et al. Characterization of the human visceral adipose tissue secretome. Mol Cell Proteomics 6, 589-600 (2007). Celis, J.E., et al. Identification of extracellular and intracellular signaling components of the mammary adipose tissue and its interstitial fluid in high risk breast cancer patients: toward dissecting the molecular circuitry of epithelial-adipocyte stromal cell interactions. Mol Cell Proteomics 4, 492-522 (2005).

118

Chapter 4
72. Chen, X., Cushman, S.W., Pannell, L.K. & Hess, S. Quantitative proteomic analysis of the secretory proteins from rat adipose cells using a 2D liquid chromatography-MS/MS approach. J Proteome Res 4, 570-577 (2005). Zhou, H., et al. Quantitative analysis of secretome from adipocytes regulated by insulin. Acta Biochim Biophys Sin (Shanghai) 41, 910-921 (2009). Kim, J., et al. Comparative analysis of the secretory proteome of human adipose stromal vascular fraction cells during adipogenesis. Proteomics 10, 394-405 (2010). Maury, E., et al. Adipokines oversecreted by omental adipose tissue in human obesity. Am J Physiol Endocrinol Metab 293, E656-665 (2007). Zvonic, S., et al. Secretome of primary cultures of human adipose-derived stem cells: modulation of serpins by adipogenesis. Mol Cell Proteomics 6, 18-28 (2007). Kratchmarova, I., et al. A proteomic approach for identification of secreted proteins during the differentiation of 3T3-L1 preadipocytes to adipocytes. Mol Cell Proteomics 1, 213-222 (2002). Vogel, K.G. & Trotter, J.A. The effect of proteoglycans on the morphology of collagen fibrils formed in vitro. Coll Relat Res 7, 105-114 (1987). Scott, J.E. & Orford, C.R. Dermatan sulphate-rich proteoglycan associates with rat tail-tendon collagen at the d band in the gap region. Biochem J 197, 213-216 (1981). Sugars, R.V., et al. Molecular interaction of recombinant decorin and biglycan with type I collagen influences crystal growth. Connect Tissue Res 44 Suppl 1, 189-195 (2003). Wiberg, C., Heinegard, D., Wenglen, C., Timpl, R. & Morgelin, M. Biglycan organizes collagen VI into hexagonal-like networks resembling tissue structures. J Biol Chem 277, 49120-49126 (2002). Yamaguchi, Y., Mann, D.M. & Ruoslahti, E. Negative regulation of transforming growth factorbeta by the proteoglycan decorin. Nature 346, 281-284 (1990). Ajuwon, K.M., Cruz, M. & Buhman, K. Changes in ECM proteins, decorin and biglycan, during adipogenesis in 3T3-L1 cells and in adipose tissue of mice on a high fat diet. FASEB J. 23, 1022.1012- (2009). Musil, K.J., Malmstrom, A. & Donner, J. Alteration of proteoglycan metabolism during the differentiation of 3T3-L1 fibroblasts into adipocytes. J Cell Biol 114, 821-826 (1991). Urs, S., et al. Gene expression profiling in human preadipocytes and adipocytes by microarray analysis. J Nutr 134, 762-770 (2004). Ward, M.G., Adapala, J., Adedokun, S. & Ajuwon, K. Dynamic changes in extracellular matrix gene regulation in adipose tissue and impact on preadipocyte proliferation and apoptosis. FASEB J. 24, 341.341- (2010). Winnemoller, M., Schmidt, G. & Kresse, H. Influence of decorin on fibroblast adhesion to fibronectin. Eur J Cell Biol 54, 10-17 (1991). Danielson, K.G., et al. Targeted disruption of decorin leads to abnormal collagen fibril morphology and skin fragility. J Cell Biol 136, 729-743 (1997). Spiegelman, B.M. & Ginty, C.A. Fibronectin modulation of cell shape and lipogenic gene expression in 3T3-adipocytes. Cell 35, 657-666 (1983). Bortell, R., Owen, T.A., Ignotz, R., Stein, G.S. & Stein, J.L. TGF beta 1 prevents the downregulation of type I procollagen, fibronectin, and TGF beta 1 gene expression associated with 3T3-L1 pre-adipocyte differentiation. J Cell Biochem 54, 256-263 (1994). Ibrahimi, A., Bonino, F., Bardon, S., Ailhaud, G. & Dani, C. Essential role of collagens for terminal differentiation of preadipocytes. Biochem Biophys Res Commun 187, 1314-1322 (1992). Scott, I.C., et al. Bone morphogenetic protein-1 processes probiglycan. J Biol Chem 275, 3050430511 (2000). von Marschall, Z. & Fisher, L.W. Decorin is processed by three isoforms of bone morphogenetic protein-1 (BMP1). Biochem Biophys Res Commun 391, 1374-1378 (2010). Kang, Q., et al. A comprehensive analysis of the dual roles of BMPs in regulating adipogenic and osteogenic differentiation of mesenchymal progenitor cells. Stem Cells Dev 18, 545-559 (2009). Schulz, T.J. & Tseng, Y.H. Emerging role of bone morphogenetic proteins in adipogenesis and energy metabolism. Cytokine Growth Factor Rev 20, 523-531 (2009). Medeck, R.J., Sosa, S., Morris, N. & Oxford, J.T. BMP-1-mediated proteolytic processing of alternatively spliced isoforms of collagen type XI. Biochem J 376, 361-368 (2003).

73. 74. 75. 76. 77. 78. 79. 80. 81.

82. 83.

84. 85. 86.

87. 88. 89. 90.

91. 92. 93. 94. 95. 96.

119

Chapter 4

97.

98.

99. 100.

101. 102. 103. 104. 105. 106. 107. 108. 109.

Bostrom, K., Tsao, D., Shen, S., Wang, Y. & Demer, L.L. Matrix GLA protein modulates differentiation induced by bone morphogenetic protein-2 in C3H10T1/2 cells. J Biol Chem 276, 14044-14052 (2001). Yagami, K., et al. Matrix GLA protein is a developmental regulator of chondrocyte mineralization and, when constitutively expressed, blocks endochondral and intramembranous ossification in the limb. J Cell Biol 147, 1097-1108 (1999). Urist, M.R., et al. Purification of bovine bone morphogenetic protein by hydroxyapatite chromatography. Proc Natl Acad Sci U S A 81, 371-375 (1984). Sottile, V. & Seuwen, K. Bone morphogenetic protein-2 stimulates adipogenic differentiation of mesenchymal precursor cells in synergy with BRL 49653 (rosiglitazone). FEBS Lett 475, 201-204 (2000). Sun, F., et al. Contrary Effects of BMP-2 and ATRA on Adipogenesis in Mouse Mesenchymal Fibroblasts. Biochem Genet (2009). Hambrock, H.O., et al. SC1/hevin. An extracellular calcium-modulated protein that binds collagen I. J Biol Chem 278, 11351-11358 (2003). Claeskens, A., et al. Hevin is down-regulated in many cancers and is a negative regulator of cell growth and proliferation. Br J Cancer 82, 1123-1130 (2000). Sullivan, M.M., et al. Matricellular hevin regulates decorin production and collagen assembly. J Biol Chem 281, 27621-27632 (2006). Nie, J. & Sage, E.H. SPARC inhibits adipogenesis by its enhancement of beta-catenin signaling. J Biol Chem 284, 1279-1290 (2009). Nie, J. & Sage, E.H. SPARC inhibits adipogenesis. Matrix Biology 25, S86-S86 (2006). McLaughlin, P.J., et al. Targeted disruption of fibulin-4 abolishes elastogenesis and causes perinatal lethality in mice. Molecular and cellular biology 26, 1700-1709 (2006). Zanetti, M., et al. EMILIN-1 deficiency induces elastogenesis and vascular cell defects. Molecular and cellular biology 24, 638-650 (2004). Kawaguchi, N., et al. ADAM12 induces actin cytoskeleton and extracellular matrix reorganization during early adipocyte differentiation by regulating beta1 integrin function. J Cell Sci 116, 3893-3904 (2003).

120

Chapter 5

Chapter 5
Conclusion and Outlook
Adipose tissue is a key player in whole body metabolism and undergoes profound changes in endocrine and metabolic function during obesity, thereby contributing to obesity-related metabolic diseases. It has become more and more evident that adipose tissue is a dynamic tissue, where adipocytes undergo a high turnover rate and new adipocytes are formed by de novo differentiation from adipocyte precursor cells. The capacity to generate new adipocytes in the state of nutritional oversupply, thereby building up a safe sink for excess energy, can tip the balance between metabolically healthy and metabolically unhealthy obese subjects1. This PhD thesis studies novel molecular mechanisms determining adipocyte differentiation and explores their influence on metabolic health in the state of obesity. Exploratory data analysis of gene expression profiles and protein expression by mass spectrometry have led to the identification of new key players in adipogenesis. Mechanistic in vitro data is thereby integrated with functional in vivo data using mouse models and human patient samples. Transcription factors play a crucial role in activating the transcriptional cascade of differentiation2. Chapter 2 describes the role of retinoid-related orphan receptor (ROR) in adipogenesis. ROR triggers an inhibitory signal by stimulating expression of matrix-metalloproteinase 3 (MMP3), which traps adipocyte precursor cells in an undifferentiated state. Deactivation of ROR leads to enhanced adipogenesis and consequently to improved glycemic and lipolytic control in obese mice and humans. Future studies will need to confirm that this effect is solely adipose tissue dependent using targeted deletion of ROR in fat. ROR is a ligand-dependent transcription factor and several recent studies have proposed oxidized sterols as ligands repressing ROR activity3-6. In the future it will be of interest to further characterize endogenous ROR ligands using an unbiased lipidomics identification approach and explore whether these ligands affect adipocyte differentiation through ROR. This will open up the

121

Chapter 5

possibility for pharmacological inhibition of ROR, thereby potentially improving the metabolic status. Adipocyte differentiation is tightly controlled by para- and endocrine extracellular signals either secreted from adipose tissue itself or reaching the preadipocyte via the circulation2. Chapters 3 and 4 explore factors in the extracellular space that modulate adipogenesis. Of pivotal importance for adipogenesis is the extracellular matrix and its dynamic remodeling7 for the following reasons: - During differentiation preadipocytes need to undergo a structural metamorphosis changing from a long stretched fibroblast-like cell into a large, round adipocyte. Remodeling of structural matrix proteins and intracellular cytoskeleton is indispensable for this process8. - The extracellular matrix is a scaffold that traps and releases growth factors and other signaling molecules, it therefore controls the availability of important signaling molecules influencing adipocyte differentiation9. - Thinking of preadipocytes as stem cells, these cells will probably sit in a defined niche10 (it has been proposed that this preadipocyte niche is located near the vasculature11), which is determined by the 3D scaffold of the extracellular matrix. Preadipocytes are likely to be attached to this scaffold, which will influence their proliferation, migration and differentiation capacity. Extracellular matrix remodeling is driven by a large family of matrix-metalloproteinases (MMP) and their inhibitors (TIMPs). Whereas in chapter 2 MMP3 was shown to inhibit adipogenesis, chapter 3 adds an additional level to the regulation of matrix alteration by exploring tissue inhibitor of metalloproteinase 1 (TIMP1) in the context of adipogenesis. TIMP1 inhibits adipocyte formation in vitro and in vivo thereby promoting detrimental adipocyte hypertrophy in response to a high fat diet stimulus. Future studies will need to assess in greater detail, how the activity of different matrixmetalloproteinases in adipose tissue is influenced by TIMP1 and which consequences arise for the organization of the extracellular matrix. The risk to develop obesity-related metabolic complications is associated with body fat distribution; i.e. intraabdominal or visceral adipose tissue poses a much greater risk
122

Chapter 5

factor than subcutaneous fat depots12. Chapter 4 explores depot-specific differences in adipogenic potential, which might contribute to dysregulated adipocyte function in visceral adipose tissue. The differentiation potential of murine visceral stromalvascular cells is greatly reduced compared to subcutaneous depots, which is mirrored in an increased visceral adipocyte size. Further studies will need to conclusively show that this holds true under in vivo conditions without forced stimulation of adipogenesis. Using conditioned media experiments we could show that inhibitory factors are secreted from cells within the stromal-vascular fraction. An unbiased proteomics approach revealed secreted factors from the SVF that influence adipogenesis of primary adipocyte precursor cells. Interestingly, many of the identified factors are extracellular matrix proteins such as the proteoglycans decorin and biglycan binding to collagen fibrils or EGF-containing fibulin-like extracellular matrix protein 2 (Efemp2) and elastin microfibril interfacter 1 (Emilin1), which are associated with elastic fibers. This supports the important role of extracellular matrix proteins in adipogenesis as discussed above. Future studies will have to investigate the precise role of these proteins in extracellular matrix remodeling during adipocyte differentiation in vitro and in vivo. With regard to differentiation capacity and adipocyte cell size it will be interesting to study these features depending on sex, age or type of diet to get a more complete picture of adipose tissue plasticity. Further research will need to place the newly identified factors into this broader context. Another big challenge for the future in the field of adipocyte differentiation will be to better characterize adipocyte precursor cells. To date little consensus exists about appropriate cell surface markers that make it possible to sort out or specifically target the preadipocyte. There is a great need for unbiased approaches to identify new markers, which better characterize cells with adipogenic potential. New markers will need to be carefully analyzed with regard to in vivo adipogenesis using reporter mouse lines that enable in vivo lineage tracing11,13. Considering the importance of extracellular matrix scaffolds, it will also be of importance for the future to study adipocyte differentiation not only in a 2D culture dish but under more physiological conditions in a 3D environment.

123

Chapter 5

References
1. 2. 3. 4. O'Connell, J., et al. The relationship of omental and subcutaneous adipocyte size to metabolic disease in severe obesity. PLoS One 5, e9997 (2010). Rosen, E.D. & MacDougald, O.A. Adipocyte differentiation from the inside out. Nature Reviews Molecular Cell Biology 7, 885-896 (2006). Kumar, N., et al. The benzenesulfonamide T0901317 is a novel RORalpha/gamma Inverse Agonist. Mol Pharmacol (2009). Solt, L.A., Griffin, P.R. & Burris, T.P. Ligand regulation of retinoic acid receptor-related orphan receptors: implications for development of novel therapeutics. Curr Opin Lipidol 21, 204-211 (2010). Wang, Y., Kumar, N., Crumbley, C., Griffin, P.R. & Burris, T.P. A second class of nuclear receptors for oxysterols: Regulation of RORalpha and RORgamma activity by 24S-hydroxycholesterol (cerebrosterol). Biochim Biophys Acta (2010). Wang, Y., et al. Modulation of ROR{alpha} and RORgamma activity by 7-oxygenated sterol ligands. The Journal of biological chemistry (2009). Lilla, J., Stickens, D. & Werb, Z. Metalloproteases and adipogenesis: a weighty subject. The American journal of pathology 160, 1551-1554 (2002). Selvarajan, S., Lund, L.R., Takeuchi, T., Craik, C.S. & Werb, Z. A plasma kallikrein-dependent plasminogen cascade required for adipocyte differentiation. Nat Cell Biol 3, 267-275 (2001). Taipale, J. & Keski-Oja, J. Growth factors in the extracellular matrix. FASEB J 11, 51-59 (1997). Wilson, A., et al. Dormant and self-renewing hematopoietic stem cells and their niches. Annals of the New York Academy of Sciences 1106, 64-75 (2007). Tang, W., et al. White Fat Progenitor Cells Reside in the Adipose Vasculature. Science (New York, N.Y (2008). Wajchenberg, B.L. Subcutaneous and visceral adipose tissue: their relation to the metabolic syndrome. Endocr Rev 21, 697-738 (2000). Carlone, D.L. Functional analysis of adult stem cells using Cre-mediated lineage tracing. Curr Protoc Stem Cell Biol Chapter 5, Unit 5A 2 (2009).

5.

6. 7. 8. 9. 10. 11. 12. 13.

124

Appendix Chapter 2

Appendix Chapter 2
Suppl. Table I Differentially expressed genes in adipose SVF during obesity. Human whole genome microarrays were performed with visceral SVF of obese (BMI 35-41) or lean (BMI < 25) patients. Data was filtered for regulated genes with p<0.01 and at least 2 fold change in expression.
probe name A_23_P324107 A_24_P279060 A_32_P226918 A_32_P73452 A_32_P150928 A_24_P337657 A_23_P50773 A_24_P40757 A_23_P141699 A_24_P938313 A_23_P90369 A_23_P312863 A_24_P44916 A_24_P416961 A_24_P238402 A_24_P14367 A_32_P420563 A_24_P204269 A_23_P331348 A_23_P119141 A_24_P365180 A_24_P154573 A_24_P268357 A_23_P125771 A_23_P346309 A_23_P150129 A_24_P363953 A_23_P169558 A_24_P269101 A_23_P15621 A_32_P121978 A_23_P108501 A_32_P137632 A_32_P162076 A_32_P64461 gene symbol RORC HCN4 PKN2 ANO8 HS2ST1 SRF CRTC1 RPL12P8 PARD6G FHIT C19orf6 TUBB2C CDC42EP5 ARVCF ZNF644 PTBP1 RNF215 MLL3 DOCK7 KEAP1 DSEL ZNF509 SMPD4 HCFC1 BAX SAPS3 RAB11FIP2 CSPP1 NEUROG1 PRAC SLC8A1 EPHA4 FBXL17 AL663074.1 AL138743.2 fold change 5.9 5.9 5.2 4.6 4.5 4.3 3.7 3.1 3.1 3.0 2.7 2.5 2.4 2.3 2.3 2.3 2.2 2.2 2.2 2.1 2.1 2.1 2.1 2.1 2.0 2.0 -2.0 -2.1 -2.1 -2.2 -2.2 -2.8 -3.7 -3.9 -7.0 p-value 2.64E-05 0.002143 0.008378 0.005497 0.009215 0.008989 0.002551 0.004547 0.005569 0.006487 0.00715 0.004121 0.008756 3.59E-05 0.005806 0.009519 0.000882 0.003224 0.005897 0.007043 0.008475 0.008692 0.009585 0.009926 0.000317 0.006681 0.009375 0.000878 0.005232 0.006322 0.008659 0.008421 0.008048 0.006292 0.004655

125

Appendix Chapter 2

Suppl. Table II Identification of ROR target genes. Murine whole genome microarrays were performed with 3T3-L1 cells stably or transiently overexpressing ROR. Data was filtered for regulated genes with p<0.05 and at least 2 fold change.
agilent probe A_52_P325265 A_52_P222113 A_52_P771377 A_52_P338266 A_52_P216820 A_51_P358894 A_51_P122481 A_52_P940338 A_52_P510387 A_51_P334876 A_52_P604243 A_51_P259765 A_52_P971776 gene description RIKEN cDNA 1110007M04 gene similar to RIKEN cDNA 1700029H17 7 days neonate cerebellum cDNA, RIKEN full-length enriched library, clone:A730052E04 product:unclassifiable, full insert sequence RIKEN cDNA 2410076I21 gene receptor transporter protein 2 tetratricopeptide repeat domain 9B PREDICTED: hypothetical protein LOC69333 [Mus musculus], mRNA sequence [XM_355963] similar to Cytochrome P450 11B1, mitochondrial precursor (CYPXIB1) (P450C11) (Steroid 11-beta-hydroxylase) (P450(11 beta)-DS) fibroblast activation protein amiloride-sensitive cation channel 1, neuronal (degenerin) protein phosphatase 1, regulatory (inhibitor) subunit 14c RIKEN cDNA 4930564B18 gene solute carrier family 25 (mitochondrial carrier, Aralar), member 12 Mus musculus 12 days embryo embryonic body between diaphragm region and neck cDNA, RIKEN full-length enriched library, clone:9430091L05 product:unclassifiable, full insert sequence. PREDICTED: Mus musculus secretoglobin, family 1C, member 1 (Scgb1c1), mRNA [XM_355976] Mus musculus adult male testis cDNA, RIKEN full-length enriched library, clone:4933404M01 product:unclassifiable, full insert sequence. T-cell receptor beta, variable 13 matrix metallopeptidase 10 olfactory receptor 450 lipocalin 6 olfactory receptor 1303 interferon gamma predicted gene, EG328264 receptor transporter protein 3 RIKEN cDNA 9030421J09 gene Rho GTPase activating protein 30 RIKEN cDNA 5830406J20 gene cholinergic receptor, nicotinic, alpha polypeptide 1 (muscle) Mus musculus 12 days embryo embryonic body between diaphragm region and neck cDNA, RIKEN full-length enriched library, clone:9430087A22 product:unclassifiable, full insert sequence. RIKEN cDNA E430014L09 gene phosphatidylethanolamine binding protein 2 kallikrein 1-related peptidase b9 RIKEN cDNA 4930478A21 gene keratin associated protein 4-7 glucosaminyl (N-acetyl) transferase 1, core 2 olfactory receptor 357 leucine rich repeat protein 3, neuronal hypothetical protein A630043P06 NHS-like 1 LIM and senescent cell antigen-like domains 1

A_51_P416356 A_52_P156775 A_52_P76104 A_52_P102846 A_51_P120830 A_52_P468639 A_51_P380330 A_52_P74960 A_52_P68893 A_52_P557395 A_52_P134023 A_52_P239086 A_52_P274960 A_51_P208132 A_52_P533707

A_52_P916310 A_51_P107752 A_51_P324450 A_51_P221753 A_51_P439531 A_51_P335694 A_52_P21550 A_51_P340525 A_52_P146719 A_52_P643674 A_51_P186421 A_52_P206836 A_52_P559378

126

Appendix Chapter 2

A_52_P178651 A_51_P392377 A_52_P542419 A_51_P185275 A_52_P678837 A_51_P265444 A_51_P462708 A_51_P254262 A_51_P461665 A_52_P128691 A_52_P456243 A_52_P440836 A_52_P1090043 A_51_P196972 A_51_P322851 A_51_P255699 A_52_P87804 A_52_P194500 A_52_P62489 A_52_P1188322 A_51_P215841 A_52_P680827 A_51_P421249 A_52_P430411 A_51_P451498 A_52_P662134 A_52_P1083688 A_51_P374592 A_51_P135920 A_52_P241834 A_51_P309556 A_51_P110672 A_51_P408199 A_51_P122762 A_52_P298002 A_52_P591760 A_51_P140690 A_51_P441231 A_51_P179293 A_52_P7221 A_52_P137710 A_52_P292481 A_52_P561746 A_51_P405638 A_52_P396312 A_51_P286436

solute carrier family 1 (glial high affinity glutamate transporter), member 2 protein tyrosine phosphatase, receptor type, f polypeptide (PTPRF), interacting protein (liprin), alpha 2 casein kinase 1, epsilon olfactory receptor 1518 rotatin solute carrier family 28 (sodium-coupled nucleoside transporter), member 2 WAP four-disulfide core domain 15B neurexophilin 3 chemokine (C-X-C motif) ligand 9 G protein-coupled receptor 4 RIKEN cDNA 2900079G21 gene early growth response 3 solute carrier family 4 (anion exchanger), member 1 olfactory receptor 622 matrix metallopeptidase 3 major facilitator superfamily domain containing 4 RIKEN cDNA D630042P16 gene Mus musculus 12 days embryo spinal ganglion cDNA, RIKEN full-length enriched library, clone:D130027L03 product:unclassifiable, full insert sequence. Adult male spinal cord cDNA, RIKEN full-length enriched library, clone:A330023K10 product:unclassifiable, full insert sequence 4-aminobutyrate aminotransferase 13 days embryo lung cDNA, RIKEN full-length enriched library, clone:D430008B02 product:unclassifiable, full insert sequence RIKEN cDNA 1700029F12 gene Mus musculus 12 days embryo spinal ganglion cDNA, RIKEN full-length enriched library, clone:D130060I10 product:unclassifiable, full insert sequence. neuron navigator 3 Mus musculus adult male diencephalon cDNA, RIKEN full-length enriched library, clone:9330184N06 product:unclassifiable, full insert sequence. 602916129F1 NCI_CGAP_Lu29 Mus musculus cDNA clone IMAGE:5066599 5', mRNA sequence. Mus musculus 2 days neonate thymus thymic cells cDNA, RIKEN full-length enriched library, clone:E430034J06 product:unclassifiable, full insert sequence. Hypothetical gene supported by AK083927 RIKEN cDNA 4933425L06 gene macrophage stimulating 1 receptor (c-met-related tyrosine kinase) RIKEN cDNA 1110033F04 gene zinc fingerprotein 618 GTP cyclohydrolase 1 RS25_HUMAN (P62851) 40S ribosomal protein S25, complete [TC1411675] stathmin-like 3 RIKEN cDNA 2310002L13 gene olfactory receptor 64 zinc finger with KRAB and SCAN domains 1 late cornified envelope 1H Rho GTPase activating protein 6 Ig heavy chain V region cadherin 17 olfactory receptor 434

127

Appendix Chapter 2

pcDNA3

ROR

ROR

DAPI

merge

c ROR -Tubulin d

scramble

siRNA1 siRNA2

14

**
control ROR

normalized expression

13 12 11 0.4 0.3 0.2 0.1 0.0

Ror

Ror

Ror

Suppl. Figure 1 a) Westernblot of ROR overexpression using anti HA-tag antibody. 3T3-L1 preadipocytes were transfected with pcDNA3 or ROR 2 days before cell lysis. b) Nuclear staining of 3T3-L1 preadipocytes transfected with ROR against HA-tagged ROR and with DAPI for nuclear localization. c) Western blot of 3T3-L1 preadipocytes stably overexpressing ROR two days after transfection with control siRNA or siRNA against ROR. d) mRNA expression of Ror, Ror and Ror relative to Gapdh in 3T3-L1 preadipocytes overexpressing ROR measured by qPCR. Values represent means +/SD, *p 0.05, **p 0.01, ***p 0.001.

128

Appendix Chapter 2
a

cytosol Cyto60

nuclei Hoechst

lipid droplets BODIPY

b
2.5 - rosiglitazone + rosiglitazone

**

luciferase activity

2.0 1.5 1.0 0.5 0.0

ROR pr

PPRE

% normalized differentiation

400

*
300 200 100 0 0

**

**

***

2.5

10

20

MMP3 inhibitor [ M]

control

20 M MMp3 inhibitor

Suppl. Figure 2 a) Automated image analysis of differentiated adipocytes by CellProfiler. Cells are stained with BODIPY for lipid droplets, Hoechst for nuclear detection and Syto60 for cytosolic staining. b) Luciferase assay of 3T3-L1 cells transfected with ROR-promoter or PPRE luciferase construct and renilla luciferase for normalization and treated with rosiglitazone 8 h prior to cell lysis. c) Fluorescent staining of differentiated 3T3-L1 preadipocytes in the presence of different concentrations of MMP3 inhibitor two days before and after induction of differentiation. Scale: 100 m. Values represent means +/- SD, *p 0.05, **p 0.01, ***p 0.001.

129

Appendix Chapter 2

a
4

fat pad weight [g]

3 2 1 0

+/+ -/-

+/+

al e fe m

fe m

al e

su bc DI ut O vi sc er al

D IO

-/b
15000

control
counts / mg protein

ROR
10000

5000

0 basal insulin

Suppl. Figure 3 a) Total fat pad weights of subcutaneous and visceral adipose tissue from female ROR knockout and wildtype mice after 7 weeks of high fat diet assessed by CT measurements (n = 7). b) Glucose uptake of 3T3-L1 adipocytes infected with control virus or virus for ROR overexpression normalized to protein content. Values represent means +/- SD

130

Appendix Chapter 2

25 20

b
free fatty acids [mM]

+/+ -/-

2.0

insulin [ng/ml]

1.5

+/+ -/-

15 10 5 0

* *
fasted refed
125

1.0

***

0.5

0.0

fasted

refed

c
triglycerides [mg/dl]
100 75 50 25 0

d
concentration [pg/ml]

400 300 200 100 0

+/+ -/-

Suppl. Figure 4 a,b) Serum insulin and free fatty acid concentrations after fasting and 6 h refeeding of ob/ob ROR-/- and wildtype mice (n = 4-7). c) Fasting triglyceride serum levels of ROR-/- and wildtype mice crossed into the ob/ob background or fed a high fat diet for 8 weeks (n = 5-10). d) Serum inflammation markers measured by multiplex micro-beads array system (n = 5-10). Values represent means +/- SD, *p 0.05, **p 0.01, ***p 0.001.

H F H D FD m a ob fem le /o a b le m al e H FD H FD m a ob fem le /o a b le m al e H FD H FD m a ob fem le /o a b le m al e H FD H FD m a ob fem le /o a b le m al e H F H D FD m a ob fem le /o a b le m al e H F H D FD m a ob fem le /o a b le m al e

IL-12

IL-1

IL-2

D IO fe +/ m + al e D m IO al -/e ob /o b m +/ al + e ob /o b -/-

+/ + D IO m al e

m al e

fe m al e

D IO

-/-

IL-6

IFN

TNF

131

Appendix Chapter 4

Appendix Chapter 4
Suppl Table I primer sequences for qPCR analysis
Ace - fwd Ace - rev Adam19 - fwd Adam19 - rev Bgn - fwd Bgn - rev Bmp1 - fwd Bmp1 - rev Ccl8 - fwd Ccl8 - rev Cdh3 - fwd Cdh3 - rev Csf1 - fwd Csf1 - rev Dcn - fwd Dcn - rev Ecm1 - fwd Ecm1 - rev Efemp1 - fwd Efemp1 - rev Efemp2 - fwd Efemp2 - rev Emilin1 - fwd Emilin1 - rev Enpp2 - fwd Enpp2 - rev Fn1 - fwd Fn1 - rev Gsn - fwd Gsn - rev Htra1 - fwd Htra1 - rev Igf1 - fwd Igf1 - rev Igfbp3 - fwd Igfbp3 - rev Igfbp4 - fwd Igfbp4 - rev Igfbp7 - fwd Igfbp7 - rev Il15ra - fwd Il15ra - rev Il1lr1 - fwd Il1lr1 - rev Lbp - fwd Lbp - rev Lcn2 - fwd Lcn2 - rev TGGGGACAAATACGTCAATCTCA GGGAAAGGCACTACCATGTCG TCAGTGGCGGACTTCAGAAAG GCAAAAAGGTGCTCGTTCTTC TGCCATGTGTCCTTTCGGTT CAGGTCTAGCAGTGTGGTGTC TTGTACGCGAGAACATACAGC CTGAGTCGGGTCCTTTGGC TCTACGCAGTGCTTCTTTGCC AAGGGGGATCTTCAGCTTTAGTA CTGGAGCCGAGCCAAGTTC GGAGTGCATCGCATCCTTCC CCGGGCATCATCCTAGTCTTG CGCCCCACAGAAGAATCCA TCTTGGGCTGGACCATTTGAA CATCGGTAGGGGCACATAGA TGTGGGAGGATGCAATGACC TTCCTTCTGGAAGCAAGAGAATC GCGCTGGTCAAGTCACAGTA AAGCATCTGGGACAATGTCAC CGGACAGCTACACGGAATG CGAGGCAGACACAAATAACCC TGTGCCTACGTGGTGACTC CGGTACATGATACTTCGGGAAC TTTGCACTATGCCAACAATCGG GGAGGCACTTTAGTCCTGTACTT GCAGTGACCACCATTCCTG GGTAGCCAGTGAGCTGAACAC ATGGCTCCGTACCGCTCTT GCCTCAGACACCCGACTTT GCAATGCGTACTGCCATTCG TGTAGGTCTTGGCGTCGCTA GTGAGCCAAAGACACACCCA ACCTCTGATTTTCCGAGTTGC CCAGGAAACATCAGTGAGTCC GGATGGAACTTGGAATCGGTCA AGAAGCCCCTGCGTACATTG TGTCCCCACGATCTTCATCTT CTGGTGCCAAGGTGTTCTTGA CTCCAGAGTGATCCCTTTTTACC ACTCCAGGGAGAGGTATGTCT CCTGTAGTTCCTTTTGCAGAGG TGACACCTTACAAAACCCGGA AGGTCTCTCCCATAAATGCACA GATCACCGACAAGGGCCTG GGCTATGAAACTCGTACTGCC TGGCCCTGAGTGTCATGTG CTCTTGTAGCTCATAGATGGTGC Lum - fwd Lum - rev Mgp - fwd Mgp - rev Mmp10 - fwd Mmp10 - rev Mmp2 - fwd Mmp2 - rev Pcolce - fwd Pcolce - rev Postn - fwd Postn - rev Psap - fwd Psap - rev Ptprf - fwd Ptprf - rev Rarres2 - fwd Rarres2 - rev Saa3 - fwd Saa3 - rev Serpina2n - fwd Serpina2n - rev Serpinf1 - fwd Serpinf1 - rev Serping1 - fwd Serping1 - rev Slit3 - fwd Slit3 - rev Sparcl1 - fwd Sparcl1 - rev Spock3 - fwd Spock3 - rev Tgfbr3 - fwd Tgfbr3 - rev Thbs2 - fwd Thbs2 - rev Timp1 - fwd Timp1 - rev Timp2 - fwd Timp2 - rev Tmem20 - fwd Tmem20 - rev Gapdh-fwd Gapdh-rv -actin-fwd -actin-rv CTCTTGCCTTGGCATTAGTCG GGGGGCAGTTACATTCTGGTG GGCAACCCTGTGCTACGAAT CCTGGACTCTCTTTTGGGCTTTA GAGCCACTAGCCATCCTGG CTGAGCAAGATCCATGCTTGG CAAGTTCCCCGGCGATGTC TTCTGGTCAAGGTCACCTGTC GCCAGACCCCCAACTACAC CCGTAATTGTCCAGATGCACTT CCTGCCCTTATATGCTCTGCT AAACATGGTCAATAGGCATCACT CCTGTCCAAGACCCGAAGAC CAAGGAAGGGATTTCGCTGTG CTGCTCTCGTGATGCTTGGTT ATCCACGTAATTCGAGGCTTG GCTGATCTCCCTAGCCCTATG CCAATCACACCACTAACCACTTC TGCCATCATTCTTTGCATCTTGA CCGTGAACTTCTGAACAGCCT ATTTGTCCCAATGTCTGCGAA TGGCTATCTTGGCTATAAAGGGG GCCCTGGTGCTACTCCTCT CGGATCTCAGGCGGTACAG TAGAGCCTTCTCAGATCCCGA ACTCGTTGGCTACTTTACCCA TGCCCCACCAAGTGTACCT GGCCAGCGAAGTCCATTTTG GGCAATCCCGACAAGTACAAG TGGTTTTCTATGTCTGCTGTAGC CTCAAGGTGTCAGCCTTATTGT AGATTGTCGTGAGCCATTGTTT GGTGTGAACTGTCACCGATCA GTTTAGGATGTGAACCTCCCTTG CTGGGCATAGGGCCAAGAG GCTTGACAATCCTGTTGAGATCA GCAACTCGGACCTGGTCATAA CGGCCCGTGATGAGAAACT TCAGAGCCAAAGCAGTGAGC GCCGTGTAGATAAACTCGATGTC CCCCGGACTTGGCTTGTTTTA AACGCACTGATCTCTACAGCA CTG ACG TGC CGC CTG GAG AAA CCG GCA TCG AAG GTG GAA GAG T CTA AGG CCA ACC GTG AAA AG ACC AGA GGC ATA CAG GGA CA

132

Appendix Chapter 4

Suppl Table II oligonucleotide sequences for adenoviral shRNA cloned into pENTR_U6 (green: sense, red: antisense, black: loop and overhang for ligation)
Ace-1fwd Ace-1rev Ace-2fwd Ace-2rev Ace-3fwd Ace-3rev Bgn-1fwd Bgn-1rev Bgn-2fwd Bgn-2rev Bgn-3fwd Bgn-3rev Bmp1-1fwd Bmp1-1rev Bmp1-2fwd Bmp1-2rev Bmp1-3fwd Bmp1-3rev Cdh3-1fwd Cdh3-1rev Cdh3-2fwd Cdh3-2rev Cdh3-3fwd Cdh3-3rev Csf1-1fwd Csf1-1rev Csf1-2fwd Csf1-2rev Csf1-3fwd Csf1-3rev Dcn-1fwd Dcn-1rev Dcn-2fwd Dcn-2rev Dcn-3fwd Dcn-3rev CCGGGCAGTACAACTCTCTGCTAAGCGAACTTAGCAGAGAG TTGTACTGCTTTTTG AATTCAAAAAGCAGTACAACTCTCTGCTAAGTTCGCTTAGCA GAGAGTTGTACTGC CCGGGCAACATGAGCAGAATCTACTCGAAAGTAGATTCTGC TCATGTTGCTTTTTG AATTCAAAAAGCAACATGAGCAGAATCTACTTTCGAGTAGA TTCTGCTCATGTTGC CCGGGGAACATATCTACCACCAACTCGAAAGTTGGTGGTAG ATATGTTCCTTTTTG AATTCAAAAAGGAACATATCTACCACCAACTTTCGAGTTGGT GGTAGATATGTTCC CCGGGGACTTCACCTTGGATGATGGCGAACCATCATCCAAG GTGAAGTCCTTTTTG AATTCAAAAAGGACTTCACCTTGGATGATGGTTCGCCATCAT CCAAGGTGAAGTCC CCGGGCTTTGAACCAGGAGCCTTTGCGAACAAAGGCTCCTG GTTCAAAGCTTTTTG AATTCAAAAAGCTTTGAACCAGGAGCCTTTGTTCGCAAAGG CTCCTGGTTCAAAGC CCGGGGTTGGGCTTAGGTCACAATCCGAAGATTGTGACCTA AGCCCAACCTTTTTG AATTCAAAAAGGTTGGGCTTAGGTCACAATCTTCGGATTGT GACCTAAGCCCAACC CCGGGGGTCATCCCGTTTGTGATTGCGAACAATCACAAACG GGATGACCCTTTTTG AATTCAAAAAGGGTCATCCCGTTTGTGATTGTTCGCAATCAC AAACGGGATGACCC CCGGGCACAGATGAGGACAGCTATACGAATATAGCTGTCCT CATCTGTGCTTTTTG AATTCAAAAAGCACAGATGAGGACAGCTATATTCGTATAGC TGTCCTCATCTGTGC CCGGGCTATATTGTATTCACCTACCCGAAGGTAGGTGAATA CAATATAGCTTTTTG AATTCAAAAAGCTATATTGTATTCACCTACCTTCGGGTAGGT GAATACAATATAGC CCGGGCATGCACCATGCAGACAATGCGAACATTGTCTGCAT GGTGCATGCTTTTTG AATTCAAAAAGCATGCACCATGCAGACAATGTTCGCATTGT CTGCATGGTGCATGC CCGGGCATCTTAAGGAGACGAAAGACGAATCTTTCGTCTCC TTAAGATGCTTTTTG AATTCAAAAAGCATCTTAAGGAGACGAAAGATTCGTCTTTC GTCTCCTTAAGATGC CCGGGCTGGCTGTTGTTGCATATGCCGAAGCATATGCAACA ACAGCCAGCTTTTTG AATTCAAAAAGCTGGCTGTTGTTGCATATGCTTCGGCATATG CAACAACAGCCAGC CCGGGCATCATCCTAGTCTTGCTGACGAATCAGCAAGACTA GGATGATGCTTTTTG AATTCAAAAAGCATCATCCTAGTCTTGCTGATTCGTCAGCAA GACTAGGATGATGC CCGGGCCTCCTGTTCTACAAGTGGACGAATCCACTTGTAGA ACAGGAGGCTTTTTG AATTCAAAAAGCCTCCTGTTCTACAAGTGGATTCGTCCACTT GTAGAACAGGAGGC CCGGGACCCTCAGACATTGGATTCTCGAAAGAATCCAATGT CTGAGGGTCTTTTTG AATTCAAAAAGACCCTCAGACATTGGATTCTTTCGAGAATCC AATGTCTGAGGGTC CCGGGCTTCTGGCATAATCCCTTATCGAAATAAGGGATTAT GCCAGAAGCTTTTTG AATTCAAAAAGCTTCTGGCATAATCCCTTATTTCGATAAGGG ATTATGCCAGAAGC CCGGGCGGAAATCCGACTTCAATGGCGAACCATTGAAGTCG GATTTCCGCTTTTTG AATTCAAAAAGCGGAAATCCGACTTCAATGGTTCGCCATTG AAGTCGGATTTCCGC CCGGGCATCTCAGACACCAACATAACGAATTATGTTGGTGT CTGAGATGCTTTTTG AATTCAAAAAGCATCTCAGACACCAACATAATTCGTTATGTT GGTGTCTGAGATGC Fn1-3fwd Fn1-3rev Gsn-1fwd Gsn-1rev Gsn-2fwd Gsn-2rev Gsn-3fwd Gsn-3rev Igf1-1fwd Igf1-1rev Igf1-2fwd Igf1-2rev Igf1-3fwd Igf1-3rev Igfbp3-1fwd Igfbp3-1rev Igfbp3-2fwd Igfbp3-2rev Igfbp3-3fwd Igfbp3-3rev Lbp-1fwd Lbp-1rev Lbp-2fwd Lbp-2rev Lbp-3fwd Lbp-3rev Mgp-1fwd Mgp-1rev Mgp-2fwd Mgp-2rev Mgp-3fwd Mgp-3rev MMP2-1fwd MMP2-1rev MMP2-2fwd MMP2-2rev CCGGGCTAAACTCTTCCACCATTATCGAAATAATGG TGGAAGAGTTTAGCTTTTTG AATTCAAAAAGCTAAACTCTTCCACCATTATTTCGAT AATGGTGGAAGAGTTTAGC CCGGGCAGCTGAGGAATGGGAATCTCGAAAGATTC CCATTCCTCAGCTGCTTTTTG AATTCAAAAAGCAGCTGAGGAATGGGAATCTTTCG AGATTCCCATTCCTCAGCTGC CCGGGCAGTATGACCTCCACTATTGCGAACAATAGT GGAGGTCATACTGCTTTTTG AATTCAAAAAGCAGTATGACCTCCACTATTGTTCGC AATAGTGGAGGTCATACTGC CCGGGCTTTGAGTCGTCCACCTTCTCGAAAGAAGGT GGACGACTCAAAGCTTTTTG AATTCAAAAAGCTTTGAGTCGTCCACCTTCTTTCGAG AAGGTGGACGACTCAAAGC CCGGGCAGCCTTCCAACTCAATTATCGAAATAATTG AGTTGGAAGGCTGCTTTTTG AATTCAAAAAGCAGCCTTCCAACTCAATTATTTCGAT AATTGAGTTGGAAGGCTGC CCGGGCCTCTGTGACTTCTTGAAGACGAATCTTCAA GAAGTCACAGAGGCTTTTTG AATTCAAAAAGCCTCTGTGACTTCTTGAAGATTCGTC TTCAAGAAGTCACAGAGGC CCGGGGCATTGTGGATGAGTGTTGCCGAAGCAACA CTCATCCACAATGCCTTTTTG AATTCAAAAAGGCATTGTGGATGAGTGTTGCTTCGG CAACACTCATCCACAATGCC CCGGGCGCTACAAAGTTGACTATGACGAATCATAGT CAACTTTGTAGCGCTTTTTG AATTCAAAAAGCGCTACAAAGTTGACTATGATTCGT CATAGTCAACTTTGTAGCGC CCGGGCACAGACACCCAGAACTTCTCGAAAGAAGT TCTGGGTGTCTGTGCTTTTTG AATTCAAAAAGCACAGACACCCAGAACTTCTTTCGA GAAGTTCTGGGTGTCTGTGC CCGGGGAGGACACACTGAATCATCTCGAAAGATGA TTCAGTGTGTCCTCCTTTTTG AATTCAAAAAGGAGGACACACTGAATCATCTTTCGA GATGATTCAGTGTGTCCTCC CCGGGGACCTCTGCTCTCTACATTGCGAACAATGTA GAGAGCAGAGGTCCTTTTTG AATTCAAAAAGGACCTCTGCTCTCTACATTGTTCGCA ATGTAGAGAGCAGAGGTCC CCGGGCTGTACAAGATCACACTACCCGAAGGTAGT GTGATCTTGTACAGCTTTTTG AATTCAAAAAGCTGTACAAGATCACACTACCTTCGG GTAGTGTGATCTTGTACAGC CCGGGCAAATGGAAGGTGCGCAAATCGAAATTTGC GCACCTTCCATTTGCTTTTTG AATTCAAAAAGCAAATGGAAGGTGCGCAAATTTCG ATTTGCGCACCTTCCATTTGC CCGGGCAACCCTGTGCTACGAATCTCGAAAGATTCG TAGCACAGGGTTGCTTTTTG AATTCAAAAAGCAACCCTGTGCTACGAATCTTTCGA GATTCGTAGCACAGGGTTGC CCGGGCATGGAGTCCTATGAAATCACGAATGATTTC ATAGGACTCCATGCTTTTTG AATTCAAAAAGCATGGAGTCCTATGAAATCATTCGT GATTTCATAGGACTCCATGC CCGGGGAGAAATGCCAACACCTTTACGAATAAAGG TGTTGGCATTTCTCCTTTTTG AATTCAAAAAGGAGAAATGCCAACACCTTTATTCGT AAAGGTGTTGGCATTTCTCC CCGGGGGAGATTCTCACTTTGATGACGAATCATCAA AGTGAGAATCTCCCTTTTTG AATTCAAAAAGGGAGATTCTCACTTTGATGATTCGT CATCAAAGTGAGAATCTCCC CCGGGCTCCACCACATACAACTTTGCGAACAAAGTT GTATGTGGTGGAGCTTTTTG AATTCAAAAAGCTCCACCACATACAACTTTGTTCGCA AAGTTGTATGTGGTGGAGC

133

Appendix Chapter 4
CCGGGCAATGCACCGATGGATATGACGAATCATATCCATCG GTGCATTGCTTTTTG AATTCAAAAAGCAATGCACCGATGGATATGATTCGTCATATC CATCGGTGCATTGC CCGGGCCGAAATAACTTTGTCATCCCGAAGGATGACAAAGT TATTTCGGCTTTTTG AATTCAAAAAGCCGAAATAACTTTGTCATCCTTCGGGATGAC AAAGTTATTTCGGC CCGGGCACCTCAGGGACTCACAATTCGAAAATTGTGAGTCC CTGAGGTGCTTTTTG AATTCAAAAAGCACCTCAGGGACTCACAATTTTCGAATTGTG AGTCCCTGAGGTGC CCGGGCCCATGCTCAACAACCAAACCGAAGTTTGGTTGTTG AGCATGGGCTTTTTG AATTCAAAAAGCCCATGCTCAACAACCAAACTTCGGTTTGGT TGTTGAGCATGGGC CCGGGCTCAACAACCAAACCCTTGCCGAAGCAAGGGTTTGG TTGTTGAGCTTTTTG AATTCAAAAAGCTCAACAACCAAACCCTTGCTTCGGCAAGG GTTTGGTTGTTGAGC CCGGGCTTCCAGTTGGGACCTAACACGAATGTTAGGTCCCA ACTGGAAGCTTTTTG AATTCAAAAAGCTTCCAGTTGGGACCTAACATTCGTGTTAG GTCCCAACTGGAAGC CCGGGCTGTGGACACAGTCTTTAATCGAAATTAAAGACTGT GTCCACAGCTTTTTG AATTCAAAAAGCTGTGGACACAGTCTTTAATTTCGATTAAAG ACTGTGTCCACAGC CCGGGGGCCACCTTAACAACCATCACGAATGATGGTTGTTA AGGTGGCCCTTTTTG AATTCAAAAAGGGCCACCTTAACAACCATCATTCGTGATGG TTGTTAAGGTGGCCC CCGGGCGAACCCTGGAGAAGCTAATCGAAATTAGCTTCTCC AGGGTTCGCTTTTTG AATTCAAAAAGCGAACCCTGGAGAAGCTAATTTCGATTAGC TTCTCCAGGGTTCGC CCGGGCAGCAAGGTTATGCCCAACACGAATGTTGGGCATAA CCTTGCTGCTTTTTG AATTCAAAAAGCAGCAAGGTTATGCCCAACATTCGTGTTGG GCATAACCTTGCTGC CCGGGCTGGCCACTGGTTTATATCCCGAAGGATATAAACCA GTGGCCAGCTTTTTG AATTCAAAAAGCTGGCCACTGGTTTATATCCTTCGGGATATA AACCAGTGGCCAGC CCGGGCGAGAGAAGTTTAACCATAGCGAACTATGGTTAAAC TTCTCTCGCTTTTTG AATTCAAAAAGCGAGAGAAGTTTAACCATAGTTCGCTATGG TTAAACTTCTCTCGC CCGGGCTGAACTGTACCTGCTTTGGCGAACCAAAGCAGGTA CAGTTCAGCTTTTTG AATTCAAAAAGCTGAACTGTACCTGCTTTGGTTCGCCAAAGC AGGTACAGTTCAGC CCGGGCACCTATCACAGGGTATAGACGAATCTATACCCTGT GATAGGTGCTTTTTG AATTCAAAAAGCACCTATCACAGGGTATAGATTCGTCTATAC CCTGTGATAGGTGC CCGGGCAAGTAGATGCTGCCTTTAACGAATTAAAG GCAGCATCTACTTGCTTTTTG AATTCAAAAAGCAAGTAGATGCTGCCTTTAATTCGT TAAAGGCAGCATCTACTTGC CCGGGCTGTTCCTGTGTGATATTAACGAATTAATAT CACACAGGAACAGCTTTTTG AATTCAAAAAGCTGTTCCTGTGTGATATTAATTCGTT AATATCACACAGGAACAGC CCGGGGAGAACAATGTCAATGTTGACGAATCAACA TTGACATTGTTCTCCTTTTTG AATTCAAAAAGGAGAACAATGTCAATGTTGATTCGT CAACATTGACATTGTTCTCC CCGGGCATGGTTATTCCTTCAATGTCGAAACATTGA AGGAATAACCATGCTTTTTG AATTCAAAAAGCATGGTTATTCCTTCAATGTTTCGAC ATTGAAGGAATAACCATGC CCGGGCAGGAGGAGATCCTTCATTACGAATAATGA AGGATCTCCTCCTGCTTTTTG AATTCAAAAAGCAGGAGGAGATCCTTCATTATTCGT AATGAAGGATCTCCTCCTGC CCGGGCAAGAAACTGGTCCTCTATTCGAAAATAGA GGACCAGTTTCTTGCTTTTTG AATTCAAAAAGCAAGAAACTGGTCCTCTATTTTCGA ATAGAGGACCAGTTTCTTGC CCGGGGTCCTCTATTTGGAACATAACGAATTATGTT CCAAATAGAGGACCTTTTTG AATTCAAAAAGGTCCTCTATTTGGAACATAATTCGTT ATGTTCCAAATAGAGGACC CCGGCCTAAGGTTAAGTCGCCCTCGCGAACGAGGG CGACTTAACCTTAGGTTTTTG AATTCAAAAACCTAAGGTTAAGTCGCCCTCGTTCGC GAGGGCGACTTAACCTTAGG CCGGGGATAGAGCCTTCTCAGATCCCGAAGGATCT GAGAAGGCTCTATCCTTTTTG AATTCAAAAAGGATAGAGCCTTCTCAGATCCTTCGG GATCTGAGAAGGCTCTATCC CCGGGCAAAGCTCTCAGAGGCTTTGCGAACAAAGC CTCTGAGAGCTTTGCTTTTTG AATTCAAAAAGCAAAGCTCTCAGAGGCTTTGTTCGC AAAGCCTCTGAGAGCTTTGC CCGGGCCCATGATGAGTAGCGTAAACGAATTTACG CTACTCATCATGGGCTTTTTG AATTCAAAAAGCCCATGATGAGTAGCGTAAATTCGT TTACGCTACTCATCATGGGC CCGGGCCTTGAAGCTATTGGCAACCCGAAGGTTGC CAATAGCTTCAAGGCTTTTTG AATTCAAAAAGCCTTGAAGCTATTGGCAACCTTCGG GTTGCCAATAGCTTCAAGGC CCGGGCAGCTTTATGAACCAAATCCCGAAGGATTTG GTTCATAAAGCTGCTTTTTG AATTCAAAAAGCAGCTTTATGAACCAAATCCTTCGG GATTTGGTTCATAAAGCTGC CCGGGCAATCCCGACAAGTACAAGGCGAACCTTGT ACTTGTCGGGATTGCTTTTTG AATTCAAAAAGCAATCCCGACAAGTACAAGGTTCGC CTTGTACTTGTCGGGATTGC

Efemp1-1fwd Efemp1-1rev Efemp1-2fwd Efemp1-2rev Efemp1-3fwd Efemp1-3rev Efemp2-1fwd Efemp2-1rev Efemp2-2fwd Efemp2-2rev Efemp2-3fwd Efemp2-3rev Emilin1-1fwd Emilin1-1rev Emilin1-2fwd Emilin1-2rev Emilin1-3fwd Emilin1-3rev Enpp2-1fwd Enpp2-1rev Enpp2-2fwd Enpp2-2rev Enpp2-3fwd Enpp2-3rev Fn1-1fwd Fn1-1rev Fn1-2fwd Fn1-2rev

MMP2-3fwd MMP2-3rev Postn-1fwd Postn-1rev Postn-2fwd Postn-2rev Postn-3fwd Postn-3rev Psap-1fwd Psap-1rev Psap-2fwd Psap-2rev Psap-3fwd Psap-3rev scramble-fw scramble-rv Serping1-1fwd Serping1-1rev Serping1-2fwd Serping1-2rev Serping1-3fwd Serping1-3rev Sparcl1-1fwd Sparcl1-1rev Sparcl1-2fwd Sparcl1-2rev Sparcl1-3fwd Sparcl1-3rev

134

Appendix Chapter 4

Suppl Table III primer sequences for PCR amplification of candidate proteins and ligation into pSecTagA and pENTR_CMV_MCS_TkPA

primer for pSecTagA Bgn-fw-long Bgn-fw-short Bgn-rv Efemp2-fw Efemp2-rv Mgp-fw Mgp-rv Emilin1-fw Emilin1-rv Bmp1-fw Bmp1-rv AAGCTTTGAGCAGAAGGGTTTCTGGGA AAGCTTTGATGAGGAGGCTTCAGGTTC TCTAGACTTCTTATAATTTCCAAATTGGATGGCC AAGCTTTTCCCCACAGGATCCCGAG TCTAGAGAAGGTATAGGCTCCCACAAAGACC AAGCTTTTACGAATCTCACGAAAGCATGGAGTC TCTAGA ATATTTGGCTCCTCGGCGCT AAGCTTTTACCCTCCTCGAGGTTACAG TCTAGACACCTGTTCAAGCTCTGTGT AAGCTTTAGATGGAGAGGCAGGCCTCG TCTAGACTTCCTGCTGTGGAGTGTGT

primer for pENTR_CMV_MCS_TkPA Bgn-HindIII-Kozak-fw Bgn-XhoI-HAtag-rv Bmp1-HindIII-Kozak-fw Bmp1-XhoI-HAtag-rv Dcn-HindIII-Kozak-fw Dcn-XhoI-HAtag-rv Efemp2-HindIII-Kozak-fw Efemp2-XhoI-HAtag-rv Emilin1-EcoRI-HAtag Emilin1-HindIII-Kozak-fw Mgp-HindIII-Kozak-fw Mgp-XhoI-HAtag-rv Sparcl1-HindIII-Kozak-fw Sparcl1-XhoI-HAtag-rv NNNAAGCTTGCCGCCGCCATGTGTCCCCTGTGGCTACT NNNCTCGAGTCAAGCGTAATCGGGCACGTCATAAGGGTACTTCTTATAATTTCCAAATTGGATG NNNAAGCTTGCCGCCGCCATGCCCGGCGTGGCCCG NNNCTCGAGTCAAGCGTAATCGGGCACGTCATAAGGGTACTTCCTGCTGTGGAGTGTGT NNNAAGCTTGCCGCCGCCATGAAGGCAACTCTCATCTTCTT NNNCTCGAGTCAAGCGTAATCGGGCACGTCATAAGGGTACTTGTAGTTTCCAAGTTGAATGG NNNAAGCTTGCCGCCGCCATGCTCCCTTTTGCCTCCTG NNNCTCGAGTCAAGCGTAATCGGGCACGTCATAAGGGTAGAAGGTATAGGCTCCCACAA NNNGAATTCTCAAGCGTAATCGGGCACGTCATAAGGGTACACCTGTTCAAGCTCTGTGT NNNAAGCTTGCCGCCGCCATGGCCCCCAGAGCCCTCTG NNNAAGCTTGCCGCCGCCATGAAGAGCCTGCTCCCTCT NNNCTCGAGTCAAGCGTAATCGGGCACGTCATAAGGGTAATATTTGGCTCCTCGGCGCT NNNAAGCTTGCCGCCGCCATGAAGGCTGTGCTTCTCCT NNNCTCGAGTCAAGCGTAATCGGGCACGTCATAAGGGTAAAAGAGGAGGTTTTCATCTATATC

135

Appendix Chapter 4

Suppl. Table IV secreted proteins from subcutaneous SVF (grey: qPCR analysis, black: functional screen), TM: transmembrane
Entrez ID 11421 11568 11816 12111 12153 50909 50909 100048401 12263 12266 625018 100045680 12279 20307 14962 21922 12842 12843 12824 12825 12826 12827 12831 12832 12833 12834 12835 13010 19025 13030 13033 13036 13039 64138 13179 13601 216616 30008 14114 14115 14118 100047082 14268 14314 75612 227753 15530 56213 16010 242050 MGI symbol Ace Aebp1 Apoe Bgn Bmp1 C1ra C1rb C1s C2 C3 C4a C4b C9 Ccl8 Cfb Clec3b Col1a1 Col1a2 Col2a1 Col3a1 Col4a1 Col4a2 Col5a1 Col5a2 Col6a1 Col6a2 Col6a3 Cst3 Ctsa Ctsb Ctsd Ctsh Ctsl Ctsz Dcn Ecm1 Efemp1 Efemp2 Fbln1 Fbln2 Fbn1 Fbn2 Fn1 Fstl1 Gns Gsn Hspg2 Htra1 Igfbp4 Igsf10 MGI Description angiotensin I converting enzyme (peptidyl-dipeptidase A) 1 AE binding protein 1 apolipoprotein E biglycan bone morphogenetic protein 1 complement component 1, r subcomponent A complement component 1, r subcomponent B complement component 1, s subcomponent complement component 2 (within H-2S) complement component 3 complement component 4A (Rodgers blood group) complement component 4B (Childo blood group) complement component 9 chemokine (C-C motif) ligand 8 complement factor B C-type lectin domain family 3, member b collagen, type I, alpha 1 collagen, type I, alpha 2 collagen, type II, alpha 1 collagen, type III, alpha 1 collagen, type IV, alpha 1 collagen, type IV, alpha 2 collagen, type V, alpha 1 collagen, type V, alpha 2 collagen, type VI, alpha 1 collagen, type VI, alpha 2 collagen, type VI, alpha 3 cystatin C cathepsin A cathepsin B cathepsin D cathepsin H cathepsin L cathepsin Z decorin extracellular matrix protein 1 epidermal growth factor-containing fibulin-like extracellular matrix epidermal growth factor-containing fibulin-like extracellular matrix protein fibulin 12 fibulin 2 fibrillin 1 fibrillin 2 fibronectin 1 follistatin-like 1 glucosamine (N-acetyl)-6-sulfatase gelsolin perlecan (heparan sulfate proteoglycan 2) HtrA serine peptidase 1 insulin-like growth factor binding protein 4 immunoglobulin superfamily, member 10

136

Appendix Chapter 4
16169 17082 16333 16678 16775 16777 226519 16803 19039 228357 100604 17022 17390 17392 381204 18542 50706 96875 19128 19156 71660 58809 20210 20716 20317 12258 20564 20657 100046740 13602 72902 21814 21826 21858 21923 12751 22041 22359 497097 Il15ra Il1rl1 Ins1 Krt1 Lama4 Lamb1-1 Lamc1 Lbp Lgals3bp Lrp4 Lrrc8c Lum Mmp2 Mmp3 Naaladl1 Pcolce Postn Prg4 Pros1 Psap Rarres2 Rnase4 Saa3 Serpina3n Serpinf1 Serping1 Slit3 Sod3 Sparc Sparcl1 Spock3 Tgfbr3 Thbs2 Timp2 Tnc Tpp1 Trf Vldlr Xkr4 interleukin 15 receptor, alpha chain interleukin 1 receptor-like 1 insulin I keratin 1 laminin, alpha 4 laminin B1 subunit 1 laminin, gamma 1 lipopolysaccharide binding protein lectin, galactoside-binding, soluble, 3 binding protein low density lipoprotein receptor-related protein 4 (TM) leucine rich repeat containing 8 family, member C (TM) lumican matrix metallopeptidase 2 matrix metallopeptidase 3 N-acetylated alpha-linked acidic dipeptidase-like 1 (TM) procollagen C-endopeptidase enhancer protein periostin, osteoblast specific factor proteoglycan 4 (megakaryocyte stimulating factor, articular superficial protein S (alpha) prosaposin retinoic acid receptor responder (tazarotene induced) 2 ribonuclease, RNase A family 4 serum amyloid A 3 serine (or cysteine) peptidase inhibitor, clade A, member 3N serine (or cysteine) peptidase inhibitor, clade F, member 1 serine (or cysteine) peptidase inhibitor, clade G, member 1 slit homolog 3 (Drosophila) superoxide dismutase 3, extracellular secreted acidic cysteine rich glycoprotein SPARC-like 1 sparc/osteonectin, cwcv and kazal-like domains proteoglycan 3 transforming growth factor, beta receptor III thrombospondin 2 tissue inhibitor of metalloproteinase 2 tenascin C tripeptidyl peptidase I transferrin very low density lipoprotein receptor (TM) X Kell blood group precursor related family member 4 (TM)

137

Appendix Chapter 4

Suppl. Table V secreted proteins from visceral SVF (grey: qPCR analysis, black: functional screen), TM: transmembrane
Entrez ID 100045780 11568 11657 11816 12111 12153 100048401 12263 12266 100045680 100045680 12560 14962 12814 12815 12842 12843 12825 12826 12827 12831 12832 53867 12833 12835 12977 13010 13030 13039 64138 13179 13601 216616 30008 100952 18606 227358 14115 14118 14268 14314 227753 15439 15530 16000 16009 16010 29817 16333 16775 MGI Adam19 Aebp1 Alb Apoe Bgn Bmp1 C1s C2 C3 C4a C4b Cdh3 Cfb Col11a1 Col11a2 Col1a1 Col1a2 Col3a1 Col4a1 Col4a2 Col5a1 Col5a2 Col5a3 Col6a1 Col6a3 Csf1 Cst3 Ctsb Ctsl Ctsz Dcn Ecm1 Efemp1 Efemp2 Emilin1 Enpp2 Fam132b Fbln2 Fbn1 Fn1 Fstl1 Gsn Hp Hspg2 Igf1 Igfbp3 Igfbp4 Igfbp7 Ins1 Lama4 MGI Description a disintegrin and metallopeptidase domain 19 (meltrin beta) (TM) AE binding protein 1 albumin apolipoprotein E biglycan bone morphogenic protein 1 complement component 1, s subcomponent complement component 2 (within H-2S) complement component 3 complement component 4A (Rodgers blood group) complement component 4B (Childo blood group) cadherin 3 complement factor B collagen, type XI, alpha 1 collagen, type XI, alpha 2 collagen, type I, alpha 1 collagen, type I, alpha 2 collagen, type III, alpha 1 collagen, type IV, alpha 1 collagen, type IV, alpha 2 collagen, type V, alpha 1 collagen, type V, alpha 2 collagen, type V, alpha 3 collagen, type VI, alpha 1 collagen, type VI, alpha 3 colony stimulating factor 1 (macrophage) cystatin C cathepsin B cathepsin L cathepsin Z decorin extracellular matrix protein 1 epidermal growth factor-containing fibulin-like extracellular matrix epidermal growth factor-containing fibulin-like extracellular matrix protein 2 elastin microfibril interfacer 1 ectonucleotide pyrophosphatase/phosphodiesterase 2 family with sequence similarity 132, member B fibulin 2 fibrillin 1 fibronectin 1 follistatin-like 1 gelsolin haptoglobin perlecan (heparan sulfate proteoglycan 2) insulin-like growth factor 1 insulin-like growth factor binding protein 3 insulin-like growth factor binding protein 4 insulin-like growth factor binding protein 7 insulin I laminin, alpha 4

138

Appendix Chapter 4
16819 269878 17313 17384 17390 17392 630776 18074 18484 18542 50706 19268 20210 20317 12258 20657 100046740 21857 21858 240660 22041 497097 Lcn2 Megf8 Mgp Mmp10 Mmp2 Mmp3 Nid1 Nid2 Pam Pcolce Postn Ptprf Saa3 Serpinf1 Serping1 Sod3 Sparc Timp1 Timp2 Tmem20 Trf Xkr4 lipocalin 2 multiple EGF-like-domains 8 (TM) matrix Gla protein matrix metallopeptidase 10 matrix metallopeptidase 2 matrix metallopeptidase 3 nidogen 1 nidogen 2 peptidylglycine alpha-amidating monooxygenase procollagen C-endopeptidase enhancer protein periostin, osteoblast specific factor protein tyrosine phosphatase, receptor type, F serum amyloid A 3 serine (or cysteine) peptidase inhibitor, clade F, member 1 serine (or cysteine) peptidase inhibitor, clade G, member 1 superoxide dismutase 3, extracellular secreted acidic cysteine rich glycoprotein tissue inhibitor of metalloproteinase 1 tissue inhibitor of metalloproteinase 2 transmembrane protein 20 transferrin X Kell blood group precursor related family member 4 (TM)

139

Appendix Chapter 4

Suppl. Table VI adenoviral shRNA screen of adipocyte differentiation in visceral, subcutaneous SVF and 3T3-L1 cells. shRNAs enhancing or inhibiting differentiation compared to control shRNA are listed. Numbers indicate different shRNAs, selected candidate proteins are marked in grey shRNAs enhancing differentiation visceral subcut. 3T3 Bgn 1 Bgn 2 Bgn 2 Bmp1 2 Bmp1 2 Bmp1 Bmp1 3 Cdh3 1 Cdh3 2 Csf1 3 Dcn 3 Dcn 3 Dcn Dcn 2 Dcn Efemp1 2 Efemp2 1 Efemp2 1 Efemp2 2 Emilin1 2 Emilin1 Emilin1 3 Enpp2 3 Enpp2 2 Fn1 2 Gsn Igf1 2 Igf1 3 Igf1 Igfbp3 Lbp 2 Mgp 3 Mgp 3 Mgp Mgp 1 Mgp 1 Mmp2 2 Postn 1 Psap 1 Psap Serping1 2 Sparcl1 1 Sparcl1 Sparcl1 2

3 2

2 2 2 3

3 3

140

Appendix Chapter 4

shRNAs inhibiting differentiation visceral subcut. 3T3 Ace 3 Ace 2 Ace Ace 1 Bgn 3 Bgn Bgn Bmp1 1 Bmp1 1 Bmp1 Cdh3 3 Cdh3 Csf1 2 Csf1 2 Csf1 Csf1 3 Csf1 1 Csf1 Csf1 1 Csf1 Dcn Efemp1 3 Efemp1 Efemp1 1 Efemp1 1 Efemp1 Efemp2 3 Efemp2 3 Efemp2 Emilin1 1 Emilin1 Enpp2 1 Enpp2 1 Enpp2 2 Fn1 1 Fn1 3 Fn1 Fn1 3 Gsn 3 Gsn Gsn Igf1 1 Igf1 1 Igf1 Igf1 3 Igfbp3 3 Igfbp3 3 Igfbp3 Igfbp3 2 Igfbp3 2 Lbp 3 Lbp Lbp 2 Lbp Mgp Mgp Mmp2 3 Mmp2 3 Mmp2 Mmp2 Postn 3 Postn 2 Postn Postn 2 Postn Postn Psap 3 Psap Psap Serping1 2 Serping1 Serping1 Serping1 Sparcl1 2 Sparcl1

2 1 3

1 1 3 1 2 1 3 2 3 2

1 1 3 1 3 3 2 1 2 3 1 3 2 1 2 1 1 2 3 2

141

Acknowledgements

Acknowledgements
I would like to thank Christian Wolfrum for his constant support and help throughout my PhD thesis. Thanks to Markus Stoffel, Christoph Meier and Isabelle Dugail for helpful input at my committee meetings. Evangeline Chew-Siegle and Matthias Geiger have been of great help with timeconsuming tasks such as making fat sections, genotyping mice and preparing adenoviruses. Thanks also to my semester students Burak Civan (ROR project), Lena Stachorski (Timp1 project) and Aliki Perdikari (paracrine factor project) for their work and commitment. I would like to say thank you to all current and former members of the Wolfrum and Stoffel-group for a great working atmosphere and fruitful discussions. Thanks to Nigel, Matthias and Hansjrg for proof-reading my thesis. Special thanks to my parents for their constant support and belief in me and to Thomas for his love and support over the past years.

Il faut avoir dj beaucoup appris de choses pour savoir demander ce qu'on ne sait pas. Man muss viel gelernt haben, um ber das, was man nicht weiss, fragen zu knnen. [Jean-Jacques Rousseau, Julie ou La Nouvelle Hlose]

142

Curriculum Vitae

Curriculum Vitae
Name Nationality Address Date/place of birth Education 04/2007 present 10/2000 01/2007 09/2003 07/2004 1991 2000 Research since 04/2007 PhD thesis, Institute of Molecular Systems Biology, Swiss Federal Institute of Technology (ETH) Supervisor: Prof. Dr. Christian Wolfrum Molecular mechanisms of adipogenesis in obesity and the metabolic syndrome diploma thesis at the Institute for Immunology and Cell Biology, University of Stuttgart, Prof. Dr. R. Kontermann Bispecific Antibodies for Effector Cell Targeting in Tumor Therapy study thesis at ProImmune Ltd in Oxford, England Supervisor: Prof. Dr. P. Scheurich, University of Stuttgart Development of an Epitope Discovery Assay using MHCPentamers for Fluorescent T-cell Staining in Flow Cytometry research assistant at the Institute for Technical Biochemistry, University of Stuttgart research assistant in the Department of Biophysics, University of Stuttgart PhD course Systems Biology of Complex Diseases, Swiss Federal Institute of Technology (ETH), Zurich, Switzerland Studies of Technical Biology, University of Stuttgart Degree: Diplom-Biologin (technisch orientiert) Matrise in Cell Biology and Physiology, Universit Louis Pasteur Strasbourg Gymnasium Zitadelle, Jlich Bettina Meissburger German Regensbergstrasse 190, 8050 Zrich, Switzerland 19.05.1981, Dren (Germany)

03/2006 12/2006

08/2005 02/2006

2002 / 2003 2001

143

Curriculum Vitae

Scholarships and Awards 2007/2008 2003/2004 06/2000 scholarship of the Roche Research Foundation one year DAAD scholarship for studies in France high school award of the German Physics Society

Student Activities since 2007 2004/2005 2002-2006 Publications Meissburger, B., Ukropec, J., Roeder E., Beaton N., Chew E., Geiger M., Civan B., Teupser D., Hillebrand J., Geary N., Klimes I., Littman D., Nawroth P., Gasperikova D., Rudofsky G. and Wolfrum C. Retinoid-related orphan receptor gamma controls adipocyte hyperplasia and insulin sensitivity in obesity. (under review) Meissburger, B., Stachorski L., Chew-Siegle E., Wolfrum C. Tissue inhibitor of metalloproteinase 1 (TIMP1) inhibits adipocyte differentiation in obesity. (under review) Meissburger B., Mst H., Perdikari A., Geiger M., Wolfrum C. Regulation of adipogenesis by paracrine factors from adipose stromal-vascular fraction- a link to fat depot-specific differences. (in preparation) Meissburger, B. and Wolfrum, C. (2008) The role of retinoids and their receptors in metabolic disorders. EJLST 110, 191-205. Mller D., Karle A., Meissburger B., Hfig I., Stork R. and Kontermann RE. (2007) Improved pharmacokinetics of recombinant bispecific antibody molecules by fusion to human serum albumin. JBC 282, 12650-60. member of the university orchestra AOZ tutor of the botanical microscopy course instructor of the university sports course trampoline

144