Probing the Role of Pparα in the Small Intestine

Probing the role of PPARα in the small intestine

A functional nutrigenomics approach

Meike Bünger Promotor Prof. dr. Michael Müller Hoogleraar Nutrition, Metabolism and Genomics Humane Voeding, Wageningen Universiteit

Co-promotor Dr. Guido J.E.J. Hooiveld Universitair docent Humane Voeding, Wageningen Universiteit

Promotiecommissie Prof. dr. Jaap Keijer Wageningen University

Prof. dr. Ulrich Beuers University of Amsterdam

Prof. dr. Hannelore Daniel Technical University of Munich

Prof. dr. Ivonne Rietjens Wageningen University

Dit onderzoek is uitgevoerd binnen de onderzoeksschool VLAG. Probing the role of PPARα in the small intestine

A functional nutrigenomics approach

Meike Bünger

Proefschrift ter verkrijging van de graad van doctor op gezag van de rector magniﬁcus van Wageningen Universiteit, Prof. dr. M.J. Kropff, in het openbaar te verdedigen op vrijdag 12 september 2008 des namiddags te vier uur in de Aula. Meike Bünger. Probing the role of PPARα in the small intestine: A functional nutrigenomics approach.

PhD Thesis. Wageningen University and Research Centre, The Netherlands, 2008. With summaries in English and Dutch.

ISBN 978-90-8504973-9 Abstract

Background The peroxisome proliferator-activated receptor alpha (PPARα) is a ligand- activated transcription factor known for its control of metabolism in response to diet. Although functionally best characterized in liver, PPARα is also abundantly expressed in small intestine, the organ by which nutrients, including lipids, enter the body. Dietary fatty acids, formed during the digestion of triacylglycerols, are able to profoundly inﬂuence gene expression by activating PPARα. Since the average Western diet contains a high amount of PPARα ligands, knowledge on the regulatory and physiological role of PPARα in the small intestine is of particular interest.

Aim In this thesis the function of PPARα in the small intestine was studied using a combination of functional genomics experiments, advanced bioinformatics tools, and dietary intervention studies.

Results Detailed analyses on the expression of PPARα in small intestine showed that PPARα is most prominently expressed in villus cells of the jejunum, coinciding with the main anatomical location where fatty acids are digested and absorbed. Genome-wide transcriptome analysis in combination with feeding studies using the synthetic agonist WY14643 and several nutritional PPARα agonists revealed that PPARα controls processes ranging from fatty acid oxidation and cholesterol-, glucose- and bile acid metabolism to apoptosis and cell cycle. In addition, we connected PPARα with the intestinal immune system. In a more focussed study we showed that PPARα controls the barrier function of the intestine. By comparing the intestinal and hepatic PPARα transcriptome we found that PPARα controls in these two organs the expression of two distinct, but overlapping sets of genes. Finally, by performing a range of functional studies deduced from the transcriptome analysis, we demonstrated that PPARα controls intestinal lipid absorption.

Conclusion By maximally utilizing the unique possibilities offered in the post-genome era, the studies described in this thesis reported on the function of PPARα in small intestine. We conclude that intestinal PPARα plays an important role, is relevant for nutrition, and its effects are distinguishable from the hepatic PPARα response. Our results provide a better understanding of normal intestinal physiology, and may be of particular importance for the development of fortiﬁed foods, and prevention and therapies for treating obesity and inﬂammatory bowel diseases.

Aim and outline of this thesis

The small intestine plays a critical role in nutrition, since it is the primary site of food digestion and nutrient absorption. Another function is to prevent the translocation of bacteria and foreign antigens to extra-intestinal sites by forming a selective barrier. It has been demonstrated that the transcription factor PPARα can be activated by natural fatty acids and their activated derivatives (acyl-CoA esters) [1-4]. PPARα is expressed in a variety of tissues including the small intestine [5, 6]; however, its function has been almost exclusively studied in liver. Little is known about PPARα and PPARα target genes in non-hepatic tissues. Knowledge on the regulatory and physiological function of PPARα in the small intestine is of particular interest, since the average Western diet contains a high amount of triacylglycerols [7] that are hydrolyzed to monoacylglycerol and free fatty acids before entering the enterocyte [8]. Consequently the small intestine is frequently exposed to high levels of PPARα agonists, and therefore an important functional role of this transcription factor may be envisioned.

The aim of the research described in this thesis was therefore to characterize the function of PPARα in the small intestine, with special emphasis on nutritional relevance, utilizing a nutrigenomics approach.

In chapter 1 an overview of the literature is given concerning the gastrointestinal tract, the concept of nutrigenomics research, the peroxisome-proliferator-activated receptors (PPARs), and the technology used in this thesis. The results of published microarray studies on PPARα- dependent gene regulation performed in different organs are also summarized in this ﬁrst chapter. In chapter 2 the genome-wide effect of PPARα activation in the small intestine is reported, with focus on intestinal epithelial cells. The commonalities and differences of PPARα dependent gene regulation in small intestine and liver was the subject of the studies described in chapter 3. In chapter 4 and 5 we focus on two main functions of the small intestine, the nutrient absorption (chapter 5) and the selective barrier function (chapter 4). In chapter 4 the effect of PPARα- activation by three different fatty acids and the synthetic compound WY14643 is examined, with emphasis on effects on intestinal transporters and phase I/II metabolic enzymes. In chapter 5 the effect of PPARα activation on intestinal dietary lipid absorption is investigated. Finally, in chapter 6 the general discussion, conclusions and recommendations are presented.

Contents

Abstract 5

Aim and outline of this thesis 7

1 General introduction 10

2 Genome-wide analysis of PPARα activation in murine small intestine 32

3 Organ-speciﬁc function of PPARα as revealed by gene expression proﬁling 58

4 PPARα-mediated effects of dietary lipids on intestinal barrier gene expression 82

5 PPARα regulates intestinal lipid absorption 102

6 General discussion 126

Appendix 132

References 134

Summary of this thesis 144

Samenvatting 148

Dankwoord 154

Curriculum Vitae 160

List of publications 162

Educational programme 164

1 General introduction

Apated from: Meike Bünger, Guido Hooiveld, Sander Kersten and Michael Müller, “Exploration of PPAR functions by microarray technology – a paradigm for nutrigenomics”,Biochimica et Biophysica Acta - Molecular and Cell Biology of Lipids, Volume 1771, Issue 8, pages 1046-1064. PMID: 17632033

The gastrointestinal tract: function and anatomy

The gastrointestinal tract is the system of organs which serves two main functions — assimilation of nutrients and elimination of waste. The gut anatomy is organized to serve these functions. A schematic overview of the gastrointestinal tract and accessory organs is given in Figure 1. The gastrointestinal tract starts with the mouth, followed by pharynx, esophagus, stomach, small intestine, colon and ends with the rectum (Figure 1). Hence, anatomically the small intestine is localized in-between stomach and colon and is directly connected with the accessory digestive organs, pancreas and liver via small ducts attached to the upper part of the small intestine.

Figure 1. The digestive tract. (Adapted from Campbell NA, 1987, Biol- ogy, the Benjamin/Cum- mings Publishing compa- ny, Inc, Redwood City.)

The physical and chemical digestion of foods starts in the mouth by chewing and the salivary enzyme amylase. This mixture is conducted to the stomach via the esophagus, where it is mixed with the acidic gastric juice for further degradation. The acid environment in the stomach also kills most bacteria that might have entered together with the food. Proteins are hydrolyzed into 12 polypeptides by the protease pepsin in the stomach. Most of the enzymatic hydrolysis of lipids and other food components occurs in the small intestine. However, pancreas and liver both contribute to this process. The enzymes secreted by the exocrine pancreatic tissue help break down carbohydrates, fats, and proteins in the small intestine. The liver has a wide variety of functions, among others to produce bile to assist in the digestion of food. Both pancreas and liver release their contents via the greater duodenal papilla (comprised of the ampulla of Vater and the sphincter of Oddi) into the upper small intestine, the duodenum. Next to digestion and nutrient absorption a healthy gut incorporates another vital function. By forming a selective barrier the translocation of bacteria and other foreign antigens to extra-intestinal sites is prevented. 1 General introduction

This barrier function is part of the innate immune system. Together with the host response, which is triggered through the activation of speciﬁc transcription factors controlling chemokine and cytokine expression, both innate and adaptive defense mechanisms protect the body against patho- gens. An overview of small intestinal anatomy is shown in Figure 2. The mucosa is the inner lining of the small intestine and consists of three layers: the epithelium, the lamina propria, and the muscularis mucosae. To enlarge the absorptive area of the small intestine, the mucosa and submucosa is arranged into high folds, called plicae circulars (valves of Kerckring).

Figure 2. The anatomy of the small intestine. (Adapted from Marieb EN 2004, the Digestive Sys- tem in Human Anatomy & Physiology. Pearson Benjamin Cummings, San Francisco, USA.)

Numerous villi, which are ﬁnger like structures, further increase this area (Figure 2a). A villus consists of a single line of epithelial cells. This epithelium undergoes perpetual renewal, fueled by a population of multipotential stem cells located at the base of the crypts of Lieberkühn. There are four main cell lineages the multipotential stem cells differentiate into: (1) enterocytes, (2) goblet cells, (3) enteroendocrine and (4) paneth cells. Enterocytes are the dominant cell popula- 13 tion of small intestinal villi. They are polarized cells with a basal nucleus and an apical brush border called microvilli. These cells are responsible for the (selective) absorptive properties of the small intestine, thus are essential for nutrient and drug uptake (Figure 2c). Goblet cells are mucin-secreting cells, protecting and lubricating the mucosa. The enteroendocrine cells form a deceptively large population and are suggested to “communi- cate” with the intestinal lumen. They contain numerous basally-sited, dense core neurosecretory granules, which contain secreted peptide hormones. Enteroendocrine cells are scattered throughout the whole intestinal epithelium, unlike the Paneth cells, which are found exclusively at the crypt base. Paneth cells contain large secretory granules, and express a number of proteins, including lysozymes and cryptins with antibacterial properties. Cellular differentiation in the small intestine is found along two distinct axes, the proximal-distal (also termed longitudinal or cephalocaudal axis) and the crypt-villus axis. Along the proximal- distal axis, the small intestine can be anatomically and functionally divided into three parts, the duodenum, jejunum, and the ileum. Cytological changes along this proximal-distal axis occur, in that (a) the goblet cell number increases (b) villi become more ﬁnger-like and decrease in length, (c) lymphoid tissue increases and (d) plicae circulares diminish. The crypt-villus axis is a vertical axis with epithelial cells extending bidirectional upwards and downwards from the stem cells anchored adjacent to the crypt base. Enterocytes, mucous produc- ing goblet cells, and enteroendocrine cells migrate out of the crypt towards the villus, while Pa- neth cells migrate towards the crypt-base. Structural differentiation and functional specialization of each of the four cell lineages occurs along this crypt-villus axis. The cytological and functional changes translate into differential gene expression along the proximal-distal and crypt-villus axis. Differential gene expression along these axes and between the four different cell types has been examined in some studies [1-3].

14 1 General introduction

Nutrigenomics: Molecular nutrition + genomics

Traditionally, nutritional science was mainly concentrated on nutrient deficiencies and their effects on health and disease. However, over the past few decades, research emphasis has gradually shifted to the link between (over)-nutrition and chronic diseases, including cancer, obesity, cardiovascular disease, and diabetes [4-8]. In parallel, there has been increasing interest in the molecular mechanisms underlying the beneficial or adverse effects of foods and food components. More recently, driven by the continuing and accelerating discoveries in omics technology, unique possibilities have emerged to investigate the genome-wide effects of nutrients at the molecular level. This research field of gene-nutrient interactions is called nutritional genomics or nutrigenomics, and encompasses the fields of biotechnology, genomics, molecular medicine and human nutrition [5, 7, 8]. It is widely recognized that nutrigenomics has the potential to increase our understanding of how nutrition influences metabolic pathways and homeostasis, how this regulation is disturbed in a diet-related disease, and to what extent individual genotypes contribute to such diseases [4-8]. However, as opposed to pharmacological research, nutritional research needs to take into account specific problems inherent to nutritional interventions. The rather small deviations from homeostasis that characterizes the early phase of a metabolic disease [9, 10], and the complexity as well as the variability of consumed foods in general are just two examples. The body has to handle a variety of nutrients at the same time, each of which can have numerous targets with different affinities and specificities. This contrasts starkly with pharmacology, where single agents are used at low concentrations and act with relatively high affinity and selectivity for a limited number of biological targets. The challenge of nutrigenomics research is to dissect the important but complex research problems into small and feasible projects. This requires simplification of our model systems in order to obtain correct and clear answers to the research question addressed. A fruitful strategy is to borrow methods that are well established in medical or pharmacological research but are rather new in the field of nutritional research. For example, in analogy to pharmacology, nutrients can be considered as signaling molecules that are recognized by specific cellular sensing mechanisms [5]. Since the property that allows nutrients to activate 15 specific signaling pathways is carried in their molecular structure, small changes in structure can have a profound influence on which sensor pathways are activated. PPARs and nutrition

Nutrients impact gene expression mainly by activating or suppressing specific transcription factors [5, 11]. The most important group of transcription factors involved in mediating the effect of nutrients and their metabolites on gene transcription is the superfamily of nuclear receptors, which consists of 48 members in the human genome [12]. This superfamily is subdivided into six families [13], of which the NR1 family is most relevant to nutrition. Nuclear receptors govern gene expression via several distinct mechanisms that involve both activation and repression of DNA transcription. After site-specific DNA binding, their final transcriptional activity depends on physical interactions with a set of associated proteins, the so-called coactivators and corepressors. These coregulators are not exclusive to nuclear receptors and are recruited in a similar manner by numerous other DNA-binding transcription factors [13-15]. One important group of receptors that mediates the effects of dietary fatty acids on gene expression are the Peroxisome Proliferator Activated Receptors (PPARs, NR1C) [13, 16, 17]. Three PPAR isotypes, α (NR1C1), δ (also called β) (NR1C2), and γ (NR1C3) can be distinguished and characterized by different biological roles. Transcriptional regulation by PPARs requires heterodimerization with the retinoid X receptor (RXR; NR2B) [18], which is also part of the nuclear receptor superfamily [13, 19]. When activated by an agonist, the PPAR/RXR heterodimer stimulates transcription via binding to DNA response elements (PPRE) present in and around the promoter of target genes. Besides upregulating gene expression, PPARs are also able to repress transcription by directly interacting with other transcription factors and interfere with their signaling pathways, a mechanism commonly referred to as transrepression [20]. Although much is already known about PPARs, gaps in our knowledge remain. In so far as the biological role of a particular PPAR is directly coupled to the function of its target genes, probing PPAR-regulated genes via the application of genomics tools can greatly improve our understanding of PPAR function. By combining transgenic animal models with elaborate microarray analyses, a comprehensive understanding of the in vivo role of PPARs can be obtained. See e.g. Michalik et al [18] for a current overview of PPAR agonists and models. However, it should be realized that 16 transcriptome profiling does not unequivocally demonstrate that differentially expressed genes are direct PPAR targets. Direct regulation has to be shown through additional methods, such as chromatin immunoprecipitation, yeast one-hybrid and transactivation assays. 1 General introduction

Fundamentals of microarray technology

One of the most powerful technologies to emerge from the age of genome sequencing are DNA microarray slides, which enable the comparatively scanning of genome-wide patterns of gene expression for any organism with a sequenced genome. Nowadays applications of microarrays fall into three main categories: studies of genomic structure, gene expression profiling, and profiling of protein-DNA interactions. The application of arrays for genomic studies primarily involves the search for single-nucleotide polymorphisms and DNA copy-number variation, which may have considerable importance regardless of whether they cause an overt disease [21- 23]. Gene expression profiling, or transcriptomics, is extensively used to study how cells respond to certain stimuli in order to identify altered molecular pathways or to diagnose and predict clinical outcomes [24-26]. Protein-DNA interaction-profiling, or ChIP-on-chip [chromatin immunoprecipitation-on-chip] or location analysis (LA), is an emerging technique that enables the systematic investigation of the precise location of genomic protein-binding sites [27, 28]. Here, we will focus on gene expression profiling since the other applications of array technology have yet to be applied in PPAR research.

Microarray technology, analyses and interpretation of data Microarray experiments are, in principle and practice, extensions of hybridization-based methods that have been used for decades to identify and quantify nucleic acids in biological samples [29- 31]. Microarray technology utilizes gene-specific probes that represent individual genes which are arrayed on an inert substrate. Several types of microarray platforms are available [32, 33], but currently the most commonly used arrays are manufactured by companies such as Agilent [34] or Affymetrix [35]. The experimental procedure involves extraction of RNA from biological samples, followed by labeling with a detectable marker (typically a fluorescent dye). After hybridization to the array and subsequent washing, an image of the array is acquired by determining the extent of hybridization to each gene-specific probe [26, 36]. The data are then normalized to facilitate the comparison between the experimental samples. Next, a set of objective criteria is applied, for example the elimination of genes with minimal variance between the samples. 17 The most common aim of transcriptome analysis is to find genes that are differentially expressed between the various experimental samples. Although early microarray papers used a simple ‘fold change’ approach to generate lists of differentially expressed genes, most analyses now rely on more sound statistical tests to identify differences in expression between groups [37]. Unfor- tunately, there are no standards at the level of data filtering, which is often done according to personal preference and experience. This leads to discrepancies and prevents a high degree of reproducibility. It should also be emphasized that statistical significance is not the same as biological significance. Moreover, the classical approach of treating genes as independent entities is increasingly being criticized. The main reasons are first the arbitrariness of the chosen cut-off (e.g. false discovery rate, fold change) and second the disregards of the broader context in which gene products function. Although by design this approach will enable the identification of genes that show large changes in gene expression, it might not reveal small yet coordinated changes in gene expression in a set of related genes, which is often the case in nutrigenomics research. In response, testing for gene classes is becoming increasingly popular. Gene classes are traditionally based on Gene Ontology categories [38], but recently also relationships based on e.g. metabolic pathways or signal transduction routes are taken into account [39]. Gene class testing thus improves the identification of affected biological processes in microarray data sets, promoting greater understanding of the underlying mechanisms driving the observed differences between samples.

Reproducibility, standardization and accessibility of microarray studies Doubt has often been cast on the reliability, reproducibility and cross-platform concordance of DNA microarrays [40-43]. This has led to the launch of the microarray quality control (MAQC) project, a large effort involving over 125 participants representing academia, industry and gov- ernment, which carefully scrutinized the microarray technology and its use. The ﬁrst set of results from this project has recently been published, addressing inter- and intra-platform agreement and reproducibility [44]. The results revealed that the distinct platforms and test sites perform comparably, generating similar lists of differentially expressed genes, undeniably establishing the robustness of the technology, provided the proper statistical analyses are applied [44]. Given the plethora of platforms and protocols for class-comparison, prediction or discovery, it is often hard to reconstruct the experimental methods in a study from the published paper, often making it impossible to enable other researchers to verify conclusions based on array data [45]. The Microarray Gene Expression Data (MGED) society therefore developed the concept of reporting minimum information about a microarray experiment (MIAME) to reduce the widespread confu- sion [46, 47]. The adoption of these guidelines by many journals has prompted, among others, the submission of detailed protocols and microarray data sets the major public repositories Gene Expression Omnibus [48] and ArrayExpress [49]. However, as we also experienced during the preparation of this manuscript, a huge point of concern is that not all data submissions provide 18 adequate detail for reanalysis, or even worse, have not been submitted to public databases [50]. 1 General introduction

PPARα biology and microarray analysis

PPARα was first described as a receptor that is activated by peroxisome proliferators [51]. Shortly after this discovery, it was found that PPARα could be activated by natural fatty acids [52]. Tissue expression patterns indicated that PPARα was highly expressed in organs that carry out significant catabolism of fatty acids such as the liver, brown adipose tissue, heart, intestine and kidney [52-54]. It is therefore not surprising that the identification of PPARα target genes has concentrated mainly on cellular lipid metabolism in the context of the hepatocyte. Indeed, the first PPARα target gene identified was acyl-coenzyme A oxidase [55] which is involved in peroxisomal fatty acid β-oxidation. This discovery was soon followed by the identification of many more PPARα target genes involved in many key functions of lipid metabolism, such as transport and cellular uptake of fatty acids, intracellular fatty acid binding and activation, microsomal ω-oxidation, peroxisomal β-oxidation, mitochondrial β-oxidation and ketogenesis, synthesis of lipoproteins, and glycerol metabolism. In addition, it was demonstrated that PPARα activation attenuated inflammatory responses [52-54]. More recently, the availability of microarrays, specific agonists and transgenic animal models has opened the possibility to comprehensively study the functional role of PPARα in other, less obvious processes in liver and other tissues.

Liver Several studies have been performed aiming to identify genes regulated by PPARα other than those connected with fatty acid catabolism. In one of the first studies that applied array technology to discover novel PPARα-regulated pathways, Kersten et al. [56] used Affymetrix GeneChips to show that PPARα influenced the expression of several genes involved in trans- and deamination of amino acids, and urea synthesis. Activation of PPARα using the synthetic ligand WY14643 decreased mRNA levels of these genes in wild type but not in PPARα-null mice, suggesting that PPARα is directly implicated in the regulation of their expression. Consistent with these data, plasma urea concentrations were modulated by PPARα in vivo. Thus, in addition to oxidation of fatty acids, PPARα also regulates metabolism of amino acids. In a similar study, Patsouris et al. [57] compared mRNA of livers of fed and fasted PPARα-null 19 and wild-type mice on Affymetrix GeneChips. As was expected, a fasting-induced increase in expression of fatty acid oxidative and ketogenic genes was observed that was PPARα-dependent. A similar type of regulation was observed for cytosolic and mitochondrial glycerol 3-phosphate dehydrogenase, which are involved in the conversion of glycerol to glucose. A combination of molecular biological and physiological follow-up studies was applied to functionally confirm these data, demonstrating that PPARα directly stimulates hepatic glycerol metabolism and, via this and other mechanisms, importantly influences hepatic glucose production during fasting. This effect of PPARα may account for the pronounced hypoglycemia observed in fasted PPARα- null mice. Combined with another study [58], the results from both papers underscore the key function of PPARα in controlling hepatic intermediary metabolism during fasting. While PPARs are primarily studied because of their pharmacological relevance, it should be realized that PPARs likely evolved as dietary lipid sensors. Accordingly, it can be hypothesized that Reference

[56] [60] [74] [75] [61] Year

2001 2001 2002 2002 2002

GEO id GEO 5 α Conclusion regulates many genes not associated represses amino acid metabolism and α α PPAR PPAR 12 genes differentially expressed between vehicle WT and DEHP- in role DEHP a play may lactoferrin, e.g. proteins, treated mice. Iron mediated binding liver growth, but not in peroxisome Changes in gene expression by WY14643 fenofibrate were almost identical. and Serum amyloid A2 was repressed; metallothionein-1 and -2 were induced by WY14643, but not fenofibrate. Changes in gene expression distinguishable were similar, for but the non-related fibrates, compound whereas phenobarbital the altered is a key controller of intermediary metabolism with peroxisomes. Delayed onset expression of of some enhanced genes may be the result of metabolic events occurring secondary to PPAR during fasting. activation and alterations in lipid metabolism. proliferation. expression of a distinct set genes. - -Liver α PPAR

Mice were fasted for Design 0h or 24h weeks Mice were treated with 1150mg/kg/day diethylhexylphthalate (DEHP) for 2 days Mice were treated with 100 mg/kg fenofibrate, or 30 mg/kg or 100 WY14643 for 2 or 3 days Animals were dosed orally via gavage with 250 mg/kg/day of clofibrate, 250 mg/ kg/day of WY14643, 100 mg/kg/day of gemfibrozil, or 120 mg/kg/day of pheno barbital for 24h or 2 weeks Mice were with treated w) 0.125% WY14643 for (w/ 2

Platform 4 cDNA cDNA Incyte, UniGene mouse cDNA microarrays TB NIEHS, Rat Chip 20 Mu6500 AF, U74v2 AF,

Strain α

-null -null 3 α α albino +

WT(SV129), PPAR WT(SV129), PPAR WT(SV129) PPAR P null WT(C57BL/ 6N) and CD-1 Sprague- Dawley VAF

Species 2

M M M M R

Model 1 IV IV IV IV IV apoptosis -dependently α agonists. α Research question Identification of genes that genes of Identification are PPAR regulated during fasting. (toxicogenomics study). (toxicogenomics study). (toxicogenomics study). Study mechanisms the responsible for molecular effects the peroxisome of proliferator a pleiotropic Study potent the peroxisome which through mechanisms proliferators hepatocyte and proliferation regulate Examine whether expression gene is WY14643 by ways altered different in or fenofibrate; genes whose to expression has not find been reported to previously be affected by PPAR To arrays evaluate can distinguish be used whether the to effects peroxisome of between proliferators different Table 1: Papers that used microarray technology to investigate (parts) of PPARα function in various tissues. technology to investigate (parts) of PPARα 1: Papers that used microarray Table 1 General introduction [76] [77] [65] [57] [64] 2003 2003 2003 2004 2004 GSE335 α was specific for α directly governs glycerol α -regulated genes and α α -dependently regulated genes Several known PPAR Kupffer cells are WY14643. PPAR not required for effects of Genes involved and apoptosis were in up- resp. downregulated in rodent, peroxisome but not proliferation human of hepatocyte nuclear hepatocytes. factor 1 Induction fatty of expression in increase fasting-induced The acid oxidative and ketogenic genes was PPAR A set of 19 genes, regulated in at least three of the seven compounds, was found to be capable in predicting the triglyceride-lowering capacity of transcription. human hepatocytes metabolism in liver each of these drugs. dependent. In addition, a similar type of regulation of type similar a addition, In dependent. was observed for glycerol cytosolic 3-phosphate dehydrogenase, which are and mitochondrial involved in the conversion of glycerol to glucose. Combined with additional was shown that PPAR (functional) data, it are involved in control carcinogenesis (ubiquitin and COOH-terminal cell hydrolase 37, T1) cyclin cycle some potentially new targets, as well as indirectly as well as targets, new potentially some affected non-target Affected genes, cellular sugar metabolism, were cell processes growth and proliferation, were stress-, identified. immune lipid and and inflammatory response, α M clofibrate µ M weeks Immortalized cells were incubated with WY14643 for 6h for 72h 0 or 24 h agonists for 4 days Rats were with treated ciprofibrate 50mg/kg/day for 8 Primary mouse rat hepatocytes (n=3), 6 human liver and donors (3 female, 3 incubated were male) 250 with Mice were fasted for Rats different received 8 PPAR HM O AF MOE430A HM AF, AF, 21 ) ) +/+ -/- -null -null α α F Immort. hepatocytes from WT (MuSHa & PPAR mice (MuSHa WT(SV129), PPAR SD R M M, H, R M R IV CL PC IV IV - - - -null α . α agonists α agonists and the α using immortalized hepa Identification of ciprofibrate of Identification related genes important for liver that growth; are to examine effects ciprofibrate on metabolic systems of liver the in diverse rats study). (toxicogenomics study). (toxicogenomics study). gands. tential of PPAR tocyte cell lines generated from WT and PPAR mice (toxicogenomics Examine carcinogenic po Compare clofibrate effect expression on primary profiles of hepatocytes rat, mRNA in mouse, of and human pathways novel identify To regulated by PPAR Elucidate the link between the gene PPAR regulation by concomitant changes on plasma lipid parameters; to identify smaller sets gene that of are the function of these li predictive - [78] [69] [79] [80] [81] 2004 2005 2006 2006 2006 see table legend 6 GSE2853 . - α α δ resulted α α , namely κ B families) α -null mice. α -null mice. agonism. This agonism. α and other xenobiotic agonists identified key α α plays a vital role in inducing various α regulation were identified. α angiopoietin-like protein 4, forkhead C2, hypoxia- angiopoietin-like forkhead C2, protein 4, Both 1. phosphatase MAPK and factor-1a inducible proteins may participate in signaling cascades of nuclear receptors detoxification system for plant compounds. (CAR, PXR) to induce a Genes related to biosynthesis were ribosome upregulated; and genes downregulated were in the proteasome complement and coagulation cascades; no damage transcriptional signal or transcriptional for oxidative signals were DNA stress consistent anti-proliferative with and was an pro-apoptotic effects. regulatory observed; Key genes (JUN, MYC, NF siRNA-mediated knockdown of PPAR Only sesame seeds killed PPAR small set of genes can be used to quantitatively into falls compound a which to degree the assess Combining the profiles from fenofibrate, mice the specificity of treated the RNAi proximal genes with response candidate new and confirmed was to PPAR Transcriptome Transcriptome analyses on identified core Affymetrix gene PPAR and AhR signatures for genes arrays 11 consisting resp. of 8 CITED2 was identified as coregulator for PPAR was identifiedcoregulator for CITED2 as Microarray analysis of cells CITED2-overexpressing exposed to PPAR genes that affected activity of PPAR were downregulated, which rodents. the Finally, magnitude of induction is in in contrast to hypoxic response and angiogenesis. rodents than primates. a certain mechanistic toxicological class. PPAR agonist alone could not enzymes, induce it most is of suggested there these crosstalk is among an essential PPAR in a transcript profile and comparable metabolic phenotype to those of PPAR xenobiotic metabolizing enzymes. Since a PPAR a Since enzymes. metabolizing xenobiotic oxidation pathways was substantially greater in - -null α -siRNA, α agonists. α agonists α Monkeys ciprofibrate, either for received 4 days day) (400mg/kg/ or 15 days 400mg/kg/ 150 or 30, (3, trol transfected cells were treated PPAR with day). Primary hepatocytes treated with 15 were AhR or PPAR mice fed fenofibrate (200mg/kg/day) for 1 or 7 days. one week. CITED2+ vs. con Mice were with injected PPAR Mice were control diet, fed or same a the diet enriched with various seeds and grains for plant control buffer; siRNA were or compared with responses WT and PPAR AF, AF, O RAE230A, TLDA HM, ~23,000 AG, whole mouse genomic oligo microarray 22 AF, HG-U95Av2 -null -null α α cells overexpressing CITED2 Cynomolgus monkey WT(SV129), PPAR WT(SV129), T(C57BL6), PPAR H P R M M CL IV IC IV IV - - - -based may be α in vitro agonists (toxi agonists α -siRNA α -siRNA using function. -/- mice fed various fed mice -/- α α hydrodynamic tail injection would vein result functional in target inhibition gene expression in mouse liver. of plant seeds and because PPAR grains, Identification regulators of required PPAR for co- primates. Developing cogenomics). Study ciprofibrate the treatment in effect of Determine of if PPAR delivery Examine gene expression profiles in wild type PPAR and gene expression assays for two prototypical cological toxi classes: aryl hy drocarbon receptor (AhR) PPAR and important detoxification systems. for inducing 1: Model: CL=Cell line, IV= in vivo, PC= primary isolated cells, 2: Species: H=Human, M=Mouse, R=Rat, P=Primate, 3: W=Wistar, SD=Sprague-Dawley, =Fish er, er, CDIGS= Charles River Laboratory CD (SD) IGS BR rats, 4: AF=Affymetrix, AG=Agilent, SA=Sigma-Aldrich, O=Operon, CT=Clontech, I=Incyte, AB=Amersham CG=Compugen, Biosciences, HM=home made, RG=Research Genetics, TB=ToxBlot, DSB= Display 6: http://www.jbc.org/cgi/conten/full/M401489200/DC1 Systems Genomic Research 5: GEO=Gene Expression Omnibus, http://www.ncbi.nlm.nih.gov/geo/, Biotech, SupA=SuperArray, TIGR=The Institute for 1 General introduction

PPARα is activated by changes in dietary fat load, for example by high fat feeding. However, because the effects of dietary fat on (PPARα-dependent) hepatic gene expression are minor, it was shown that changes in gene expression only appear significant after comprehensive analysis of gene expression using microarrays. Indeed, comparative microarray analysis of PPARα-dependent gene regulation by a synthetic PPARα agonist, by prolonged fasting, and by high fat feeding indicated that although all treatments caused activation of PPARα, pronounced differences in the magnitude of PPARα-target gene induction between these pharmacological, physiological and nutritional stimuli was observed [59]. Ideally, genome-wide analyses should include gene knockouts to allow unambiguous interpretation of the role of PPARα in regulating differentially expressed genes. However, this has not always been the case. Using Incyte cDNA arrays, Cherkaoui-Malki et al. [60] profiled changes in gene expression in livers of mice exposed to WY14643 for two weeks. PPARα activation increased expression of genes involved in lipid and glucose metabolism and genes associated with peroxisome biogenesis, cell surface recognition function, transcription, cell cycle, and apoptosis. The last two processes may be mechanisms underlying PPARα ligand-induced toxicity. The authors did not study nor mention the genes that were repressed by WY14643. Hamadeh et al. [61] compared the gene expression profiles elicited by three PPARα agonists (clofibrate, gemfibrozil, WY14643) and one unrelated compound (phenobarbital) in livers of rats after exposure to the respective compounds for 24 hours or 2 weeks. Not surprisingly, they found a greater similarity in gene expression profile among the PPARα agonists than between the PPARα agonists and phenobarbital. Approximately 25% of the probes present on their rat cDNA array were significantly altered by at least one of the PPARα agonists for at least one time point. Unfortunately, no data on differentially expressed genes were provided for each agonist separately. Overall, genes that were significantly altered by PPARα agonist treatment were involved in fatty acid transport, fatty acid synthesis and catabolism, cell cycle, cell proliferation, and the acute phase response. Based on the analysis of early (24hrs), late (2wks) and time-independently regulated genes, a relationship between expression profiles and gene function was proposed. 23 Most of early regulated transcripts corresponded to signaling related genes, whereas transcripts regulated at the latter time point predominantly reported on adaptation events, or were related to the observed hypertrophy. Remarkably, no genes were identified that were commonly regulated by all three agonists. Another comparative analysis between different PPARα agonists was performed using primary mouse hepatocytes [62]. Hepatocytes were exposed to multiple concentrations (10, 20 and 100µM) of bezafibrate, fenofibrate or WY14643 for 24 hours, followed by gene expression profiling on Agilent whole mouse genome arrays. Treatments with the highest concentration resulted in 151, 149, and 145 differentially expressed genes for bezafibrate, fenofibrate or WY14643, respectively. Hierarchical clustering analysis showed that expression profiles clustered according to dosage rather than specific drugs, indicating that a common effect exists across this class of compounds. Gene function analysis of 121 genes regulated by at least two out of three fibrates showed that the majority of these genes were involved in lipid catabolism. The authors speculate that this may lead to elevated generation of hydrogen peroxide and the production of reactive oxygen species, which may be linked to the development of liver cancer. Clinically, fibrate drugs are used mainly for their ability to lower plasma triglycerides and increase plasma HDL levels [63]. Although these effects are common to all fibrates, it has been shown that the various fibrates display different clinical efficacy towards the various lipoprotein classes [63], which likely originates in differential effects on hepatic gene expression. According- ly, Frederiksen et al. [64] studied the link between changes in gene expression elicited by various PPARα agonists, and the concomitant physiological changes, measured as plasma lipid profiles. In an animal model of dyslipidemia, treatment with different PPARα agonists caused a significant reduction in plasma total cholesterol and triglyceride level for all compounds studied [64]. In addition, four of the seven compounds showed an increase in plasma HDL concentration. Microar- ray analysis of liver samples led to the identification of a set of 19 genes that was regulated in at least three of the seven compounds, and that was capable of predicting the triglyceride-lowering activity of each of these drugs. Many of the genes in this set were involved in lipid metabolism and have previously been shown to be regulated by PPARα. A good linear correlation was found between the quantitative regulation of gene expression, measured as fold-changes, and the transactivation efficacy of the specific drugs assayed in vitro. A different approach to study PPARα function is to make a distinction between transcriptional responses in different species, triggered by specific PPARα activation. Aiming at comparing ex- pressional responses in humans and rodents using Affymetrix GeneChips, primary human, rat and mouse hepatocytes were treated with clofibric acid for 72hrs [65]. For human hepatocytes the difference between donors gave rise to a larger variation on transcriptional profile than the effect of clofibrate treatment. In mice or rats the inter-animal variability of the response towards PPARα activation was relatively low. The differentially expressed genes in each of the species could be grouped into three main pathways: fatty acid transport and metabolism, xenobiotic metabolism, 24 and cell/organelle proliferation and cell death. Whereas genes of the cytosolic, microsomal, and mitochondrial pathways involved in fatty acid transport and metabolism were upregulated across all species, genes of the peroxisomal pathway were only upregulated in rodents. The induction of the hepatocyte nuclear factor-1α by clofibrate was observed exclusively in humans, suggesting that this transcription factor may play a key role in the regulation of fatty acid metabolism in human liver and possibly as well as in the non-responsiveness of human liver to clofibrate-induced regulation of cell proliferation and apoptosis. This is functionally in line with the well-known observed toxic side-effects of artificial PPARα ligands, such as peroxisome proliferation, hepa- tomegaly and hepatocarcinoma, occurring in rodents only [66], and seem to be dependent on intrinsic differences between human and rodent PPARα [67, 68]. Another study investigated the transcriptional effect in primate livers of a four and ten days treatment with ciprofibrate at various concentrations (3, 30, 150, 400mg/kg/day) [69]. Pathway analysis of the Affymetrix GeneChip data revealed that although fatty acid metabolism was the most 1 General introduction

upregulated process in primates, the magnitude in terms of fold-induction in e.g. the ß-oxidation pathway was substantially greater in rodent liver. A pathway very strongly downregulated was the complement and coagulation pathway. This correlated with human clinical studies in which fibrates have been shown to have an effect on coagulation and fibrinolysis, and reduced human plasma fibrinogen levels by 12–15% [69]. Unlike the ß-oxidation genes, the magnitude of the response in the primate and the rat liver appear similar, in that the downregulation in both species is modest, often not exceeding two-fold. Moreover, it was found that genes related to ribosome and proteasome biosynthesis were significantly upregulated upon ciprofibrate treatment, whereas a number of key regulatory genes, including members of the Jun oncogene, c-myc proto-oncogene, and nuclear factor kappa B families were downregulated. The latter result may different in rodents, as Jun oncogene and c-myc proto-oncogene are reported to be upregulated after PPARα agonist treatment. Although data from fenofibrate-treated animals were not presented, the authors reported that the transcriptional response for ciprofibrate was more robust than fenofibrate, and that the fenofibrate dataset appeared to be largely a subset of the ciprofibrate dataset [69].

Small intestine Even though the small intestine expresses PPARα at high level and is frequently exposed to high levels of PPARα agonists via the diet, the role of PPARα in this organ was not investigated until recently [70]. Gene expression profiling on Affymetrix GeneChips of small intestines from wild- type and PPARα-null mice fed WY14643 for five days revealed that in addition to genes involved in fatty acid and triglyceride metabolism, transcription factors and enzymes connected to sterol and bile acid metabolism, including farnesoid x receptor and sterol regulatory element binding factor (SREBP)-1, were induced. In contrast, genes involved in cell cycle and differentiation, apoptosis, and host defense were repressed by PPARα activation, which morphologically resulted in a 22% increase in villus height and a 34% increase in villus area of wild-type animals treated with WY14643 [70]. Besides providing a comprehensive overview on PPARα-dependent gene regulation in small intestine, this study also pointed towards organ-specific functions of PPARα. 25 Reference

[82] [70] [72] [73] Year

2001 2007 2005 2006

GEO id GEO 5 GSE5475 GSE5100 on glucose α ß -oxidation, which in was time-dependently α not only governs diverse α -null mice were protected from protected were mice α -null Conclusion mice exhibited increased fatty mice developed glucose intolerance glucose developed mice α α gene contains a DNA consensus motif consensus DNA a contains gene α induced enzymes in lipid metabolism; α activation signature was identified that α MCK-PPAR diet-induced insulin resistance. In skeletal muscle, skeletal In resistance. insulin diet-induced MCK-PPAR Expression of PPAR PPAR A PPAR downregulated. Subsequent studies revealed that revealed studies Subsequent downregulated. PPAR the aspects of lipid metabolism but may also play a major role in maintaining epithelial integrity. despite being protected from diet-induced obesity. obesity. diet-induced from protected being despite PPAR Conversely, acid oxidation rates, diminished AMP-activated protein kinase stimulated activity, glucose uptake without and alterations insulin-signaling in key of status phosphorylation the reduced insulin- proteins. Pharmacologic inhibition of oxidation fatty or acid mitochondrial respiratory prevented coupling the effects of PPAR was evident in type I, muscle but fibers. not type The II, of skeletal fiber-type–selective this response nature was consistent with fatty increased acid uptake and for the hypoxia-inducible factor 1 (HIF-1). homeostasis. muscle toxicity remains elusive. muscle is and triglycerides reported plasma in concentration lowered to be increased Adverse insulin interactions sensitivity. associated with are a not likely to occur at the mechanism cellular level. The by which fibrates may produce repressed host apoptosis, enhanced cell differentiation, resulting defense, in an cell increased number enterocytes. growth of PPAR mature absorptive and , α -null mice α -null -Skeletal muscle α -Gastrointestinal tract α PPAR

Design T84cells for 6 or 18h. days. in gastrocnemius muscle were compared between MCK-PPAR WT, (25 mg/kg/day). and PPAR and T84 cells were epithelial-like exposed to ambient hypoxia Mice control diet, were or supplemented diet 0.1% WY14643 with for fed 5 Expression profiles Rats were daily for 2, 3, 4, 5, 6, dosed 9, and 16 days with fenofibrate (400 mg/ kg/day), (100 WY14643 bezafibrate (250 mg/ mg/kg/day), kg/day, rosiglitazone and (100 mg/kg/day), all-trans retinoic acid

PPAR

Platform 4 AF MOE430A U34A AG AF, AF,

26 AF,

Strain -null ), -null 3 α α α α

PPAR T84 WT(SV129) PPAR rexpressing PPAR in muscle (MCK- PPAR mice CDIGS WT, mice WT,

Species 2

H M M R

Model 1 CL IV IV IV – α in α function in α in response to α Research question Examine the expression of PPAR ambient hypoxia the small intestine Characterize the pattern of gene regulatory events in overexpressing animals PPAR muscle. specifically (toxicogenomics). activation in soleus (type I) and quadriceps femoris (type II) muscle fibers Study PPAR Identify the fiber-type selective nature of PPAR Table 1 (continued): Papers that used microarray technology to investigate (parts) of PPARα function in various tissues. technology to investigate (parts) of PPARα 1 (continued): Papers that used microarray Table 1 General introduction - [83] 2005 GSE2131 α , indicating that can be a key regulator of inflammatory α PPAR The agonist induced PPAR ß -oxidation in impact adipocytes its and could inflammation be WAT an limiting of effective means on systemic insulin sensitivity. signaling during lipolytic stress and that expanding that and stress lipolytic during signaling - -White adipose tissue α PPAR neally two times with 30 nmol agonist at 4- h intervals. Mice received osmotic via minipumps vehicle (control) agonists at or a rate of 3 nmol/h for 1, 3, or 6 days. For 8 treatment, mice were h of injected intraperito U74v2 AF, AF, 27 -null α WT(SV129), PPAR M IV . α 3-adrenergic δ Study whether adipose induced tissue-remodeling by a receptor agonist is mediated through PPAR 1: Model: CL=Cell line, IV= in vivo, PC= primary isolated cells, 2: Species: H=Human, M=Mouse, R=Rat, P=Primate, 3: W=Wistar, SD=Sprague-Dawley, =Fish er, er, CDIGS= Charles River Laboratory CD (SD) IGS BR rats, 4: AF=Affymetrix, AG=Agilent, SA=Sigma-Aldrich, O=Operon, CT=Clontech, I=Incyte, AB=Amersham CG=Compugen, Biosciences, HM=home made, RG=Research Genetics, TB=ToxBlot, DSB= Display 6: http://www.jbc.org/cgi/conten/full/M401489200/DC1 Systems Genomic Research 5: GEO=Gene Expression Omnibus, http://www.ncbi.nlm.nih.gov/geo/, Biotech, SupA=SuperArray, TIGR=The Institute for White adipose tissue To our knowledge, there are no published reports that have employed specific agonists or knockout mice to study PPARα function in adipose tissue. However, despite the fact that expression of PPARα in white adipose tissue is much lower compared to PPARγ, evidence abounds that PPARα may also influence adipose tissue remodeling. Affymetrix GeneChip analysis revealed that chronic activation of β3-adrenergic signaling expanded the oxidative capacity of adipocytes in white adipose tissue by inducing mitochondrial biogenesis and by upregulating PPARα and genes involved in fatty acid oxidation and mitochondrial electron transport activity [71]. Analy- sis of gene expression patterns also indicated that inflammation and adipocyte-specific gene expression were reciprocally related over time. Combined, the data suggested that upregulation of catabolic activity is critical to suppressing inflammation and restoring phenotypic expression in adipocytes during continuous β3-adrenergic stimulation, which may be mediated by PPARα.

Skeletal muscle The potential role of PPARα in the development of muscle insulin-resistance was explored using a gain- and loss-of-function approach combined with global analysis of gene expression [72]. Transgenic mice that overexpress PPARα in skeletal muscle were used to generate a PPARα- dependent transcriptome signature via Affymetrix GeneChip analysis. A prominent metabolic reprogramming in muscle fibers was observed, characterized by a switch from glucose utilization to fatty acid oxidation pathways, leading to muscle glucose intolerance and insulin resistance. Reciprocal outcomes from studies with PPARα-null mice corroborated these observations, suggesting that by altering fuel sources, PPARα has a major impact on skeletal muscle insulin resistance. In light of the fact that fibrates have been implicated in skeletal muscle toxicity, the effect of PPARα agonists on skeletal muscle gene transcription was studied in rats using Agilent arrays [73]. A separate analysis was performed for a muscle containing predominantly slow oxidative type I fibers and a muscle containing mainly fast glycolytic type II fibers. By comparing the transcriptional responses of different PPAR agonists using an advanced statistical analytical strategy, 28 a PPARα activation signature was identified that was specific for type I, but not type II, skeletal muscle fibers. The fiber-type–selective nature of this response was consistent with increased fatty acid uptake and β-oxidation, yet failed to reveal any obvious off-target pathways that may drive the reported adverse effects. 1 General introduction

Conclusions

Based on the published studies it is clear that transcriptomics has been primarily used to study two aspects of PPARα activation. One is the characterization of the physiological functions of PPARα in tissues by means of identification of PPARα-regulated genes and processes. However, almost all of these papers used transcriptome analysis merely as a screening tool to identify and further investigate a single gene or pathway other than fatty acid metabolism, thereby avoiding analyses of the whole genome. This is likely because there is an inherent bias among researchers and journal editors in favor of a focused approach that involves extensive functional validation of the data. While this type of approach has proven its effectiveness, exemplified by the findings that PPARα regulates hepatic glycerol and amino acid metabolism, by disregarding the bulk of the data the broader potential of the transcriptome analysis remains unexplored. Without down- playing the importance of proving the functional relevance of PPARα-dependent gene regulation, it should be acknowledged that transcriptome analysis can be valuable beyond study of functional significance. For example, while transcriptome analysis does not provide information about the precise molecular mechanisms underlying SPPARMs, it is probably the most effective tool to substantiate the concept in terms of differential gene regulation. In addition, it is difficult to imagine a method more suitable to systematically compare organ specific PPARα-dependent gene regulation. Numerous other useful applications can be envisioned that can rely or benefit from transcriptomics, including ligand screening, mutational analysis, and, in combination with ChIP on CHIP analysis, a whole genome search for PPAR target genes.

The other bulk of papers focused on the toxicological aspects of the peroxisome-proliferator class of compounds to explore the usefulness of transcriptomics for toxicogenomics research, a discipline aiming to identify and characterize mechanisms of action of known and suspected toxicants. Consequently, these studies did not thoroughly address the physiological functions of PPARα. In addition, an ideally performed genome-wide analysis of PPARα function should be a balanced combination of organ-specific gene knockouts, highly specific PPAR-agonists, and whole genome arrays. Unfortunately, examination of the literature reveals that this has rarely 29 been the case. This is mostly because of a lack of (organ-specific) PPARα-null models, the lack of data mining bioinformatics tools but also the high prices of arrays.

Nonetheless, the application of transcriptomics has clearly provided new insights on PPARα function. Although the most obvious ‘core-functions’ of the PPARα isoform was identified by targeted approaches shortly after its discovery, array studies revealed, for example, that this nuclear receptor modulates the innate immune system of the small intestine. It is expected that the exhaustive characterization and comparison of organ-specific PPAR transcriptomes will ultimately allow a comprehensive description of the whole body biology that is under control of PPARs, and finally their physiological relevance. Crucial for this is the proper and accessible storage of data allowing additional analyses when improved algorithms and bioinformatics tools become available.

Finally, only a few studies have investigated functions of PPARα using natural agonists, i.e. eicosanoids, unsaturated as well as long-chain fatty acids, and their activated derivatives (acyl- CoA esters). Instead, speciﬁc pharmacological PPARα-agonists have been employed. As nutrition researchers, we have to demonstrate beyond doubt that transcriptomics studies performed with synthetic agonists are of relevance for our understanding of nutrient-mediated gene regulation. This is one of the fascinating research topics in the ﬁeld of nutrigenomics. Notwithstanding, at present the use of synthetic agonists is the most practical and effective strategy in nutrigenomics research to magnify the subtle effects of nutrients.

30 1 General introduction

31 2 Genome-wide analysis of PPARα activation in murine small intestine

Meike Bünger, Heleen van den Bosch, Jolanda van der Meijde, Sander Kersten, Guido Hooiveld, Michael Müller

Published in Physiological Genomics, 2007 July 18;30(2):192-204 PMID: 17426115 Abstract

The peroxisome proliferator-activated receptor alpha is a fatty acid-activated transcription factor that governs a variety of biological processes. Little is known about the role of PPARα in the small intestine. Since this organ is frequently exposed to high levels of PPARα ligands via the diet, we set out to characterize the function of PPARα in small intestine using functional genomics experiments and bioinformatics tools.

PPARα was expressed at high levels in both human and murine small intestine. Detailed analyses showed that PPARα was expressed highest in villus cells of proximal jejunum. Microarray analyses of total tissue samples revealed, that in addition to genes involved in fatty acid and triacylglycerol metabolism, transcription factors and enzymes connected to sterol and bile acid metabolism, including FXR and SREBP1, were speciﬁcally induced. In contrast, genes involved in cell cycle and differentiation, apoptosis, and host defense were repressed by PPARα activation. Additional analyses showed that intestinal PPARα dependent gene regulation occurred in villus cells. Functional implications of array results were corroborated by morphometric data. The repression of genes involved in proliferation and apoptosis was accompanied by a 22% increase in villus height, and a 34% increase in villus area of wild-type animals treated with WY14643.

This is the ﬁrst report providing a comprehensive overview of processes under control of PPARα in the small intestine. We show that PPARα is an important transcriptional regulator in small intestine, which may be of importance for the development of novel foods and therapies for obesity and inﬂammatory bowel diseases.

34 2 Genome-wide analysis of PPARα activation in murine small intestine

Introduction

The peroxisome proliferator-activated receptor alpha (PPARα) is a ligand-activated transcription factor with diverse functions and is activated by a variety of synthetic compounds, including the lipid lowering fibrate drugs [53, 54]. High affinity natural ligands include eicosanoids, unsaturated as well as long-chain fatty acids, and their activated derivatives (acyl-CoA esters) [84-87]. In analogy with other nuclear receptors, PPARα forms obligate heterodimers with the retinoid X receptor and stimulates gene expression by binding to peroxisome proliferator response elements (PPREs) located in the regulatory domain of genes [53]. PPARα is expressed in a variety of tissues including the small intestine [88, 89], however, its function has been almost exclusively studied in liver. In liver PPARα is critical for the coordinate transcriptional activation of genes involved in lipid catabolism, including cellular fatty acid uptake and activation, mitochondrial β-oxidation, peroxisomal fatty acid oxidation, ketone body synthesis, fatty acid elongation and desaturation, and apolipoprotein synthesis [53, 54]. In addition, PPARα is an important regulator of the hepatic acute phase response. While the function of PPARα in liver is well studied, little is known about PPARα and PPARα target genes in non-hepatic tissues. This is especially true with respect to the role of PPARα in the small intestine, which has only been addressed in few studies [90, 91]. Knowledge on the regulatory and physiological function of PPARα in the small intestine is of particular interest, since the average Western diet contains a high amount of triacylglycerols [92] that are hydrolyzed to monoacylglycerol and free fatty acids before entering the enterocyte [93]. Consequently the small intestine is frequently exposed to high levels of PPARα ligands. Therefore we set out to determine the role of PPARα in the small intestine. We first analyzed in detail the expression of PPARα throughout the small intestine and then evaluated the outcome of specific PPARα activation on small intestinal gene expression using microarrays and bioinformatics tools. This allowed the genome-wide identification of intestinal PPARα target genes and corresponding processes. We conclude that PPARα plays an important role in the regulation of intestinal func- 35 tion by governing diverse processes ranging from numerous metabolic pathways to the control of apoptosis and cell cycle. Materials and Methods

Animals Pure bred wild-type (129S1/SvImJ) and PPARα-null (129S4/SvJae) mice [94] were purchased from Jackson Laboratories (Bar Harbor, ME) and bred at the animal facility of Wageningen University. Mice were housed in a light- and temperature-controlled facility and had free access to water and standard laboratory chow (RMH-B, Hope Farms, Woerden, the Netherlands). All animal studies were approved by the Local Committee for Care and Use of Laboratory Animals.

Experimental design and tissue handling Four independent studies were performed. In all studies 4-5 month old male wild-type and PPARα-null mice were used. Study A: Mice were fed chow, or chow supplemented with 0.1% WY14643 (Chemsyn, Lenexa, KA) for five days (n=6 mice per group). On the sixth day, mice were anaesthetized with a mixture of isofluorane (1.5%), nitrous oxide (70%) and oxygen (30%). Small intestines were excised and flushed with ice-cold PBS and all subsequent tissue handlings were performed on ice. Remaining fat and pancreatic tissue was carefully removed, and RNA was isolated from the complete full-length small intestine for microarray analysis. Study B: The above described experiment was repeated (n=3 mice per group), except that after removal the small intestine was divided into 10 equal parts to study gene expression along the proximal-distal axis. Study C: Study A was repeated (n=3-4 mice per group), except that after removal the small intestine was inverted on a 0.75 mm-diameter rod, washed in ice-cold PBS, and divided into segments of 1 cm. Before continuing with the cell isolation protocol, segments of all animals within each experimental group were pooled. Fractions enriched in crypt or villus cells were isolated as described by Flint et al [95]. This isolation protocol was repeated a week later for the control group. Cell fractions were used for RNA isolation. Study D: The feeding experiment was repeated as described, except that in addition to WY14643 mice were fed chow supplemented with fenofibrate (0.1% w/w) (Sigma, St. Louis, MO) for five days (n=5 mice per group). RNA was isolated from the complete full-length small intestine for quantitative reverse-transcription 36 PCR analysis.

RNA isolation and quality control Total RNA was isolated from small intestinal samples using TRIzol reagent (Invitrogen, Breda, the Netherlands) according to the manufacturer’s instructions. RNA was treated with DNAse and puriﬁed using the SV total RNA isolation system (Promega, Leiden, the Netherlands). Concentrations and purity of RNA samples were determined on a NanoDrop ND-1000 spectrophotometer (Isogen, Maarssen, the Netherlands). RNA integrity was checked on an Agilent 2100 bioanalyzer (Agilent Technologies, Amsterdam, the Netherlands) with 6000 Nano Chips according to the manufacturer’s instructions. RNA was judged as suitable for array hybridization only if samples exhibited intact bands corresponding to the 18S and 28S ribosomal RNA subunits, and displayed no chromosomal peaks or RNA degradation products. Total RNA from human tissues (FirstChoice Human Total RNA Survey Panel) was obtained from Ambion (Austin, TX). Each tissue pool comprises RNA from three or four donors. 2 Genome-wide analysis of PPARα activation in murine small intestine

Affymetrix GeneChip oligoarray hybridization and scanning For microarray analyses, we used RNA isolated from the full-length small intestine. RNA was hybridized on an Affymetrix GeneChip Mouse Genome 430A array. This array detects 22,626 transcripts that represent approximately 13,700 known genes. For each experimental group, three biological replicated were hybridized, thus in total 12 arrays were used. Detailed methods for the labeling and subsequent hybridizations to the arrays are described in the eukaryotic section of the GeneChip Expression Analysis Technical Manual, Revision 3, from Affymetrix (Santa Clara, CA). Array data have been submitted to the Gene Expression Omnibus, accession number GSE5475.

Analyses and functional interpretation of microarray data Scans of the Affymetrix arrays were processed using packages from the Bioconductor project [96]. Expression levels of probe sets were summarized using GCRMA [97], where after differentially expressed probe sets were identified using Limma [98]. P-values were corrected for multiple testing using a false discovery rate method [99]. Probe sets that satisfied the criterion of FDR < 5% (q-value < 0.05) were considered to be significantly regulated. Of these, probe sets that were also >1.5 fold changed in wild-type mice upon WY14643 treatment, but were not changed in treated PPARα-knockout mice, were designated PPARα regulated. Three complementary methods were applied to relate changes in gene expression to functional changes. One method is based on overrepresentation of Gene Ontology (GO) terms [100]. Another approach, gene set enrichment analysis (GSEA), takes into account the broader context in which gene products function, namely in physically interacting networks, such as biochemical, metabolic or signal transduction routes [39]. Both applied methods have the advantage that it is unbiased, because no gene selection step is used, and a score is computed based on all genes in a GO term or gene set. In addition, biological interaction networks among PPARα regulated genes were identified using Ingenuity Pathways Analysis (IPA) (Ingenuity Systems, Redwood City, CA). Detailed descriptions of the applied methods are available in the supplemental text (supplemental_1). 37

Quantitative reverse-transcription polymerase chain reaction Single-stranded complementary DNA (cDNA) was synthesized from 1 µg total RNA using the reverse-transcription system from Promega (Leiden, the Netherlands) according to the supplier’s protocol. Quantitative reverse-transcription polymerase chain reaction (qRT-PCR) was performed on a MyIQ thermal cycler (BioRad, Veenendaal, the Netherlands) using Platinum Taq DNA polymerase (Invitrogen, Breda, the Netherlands) and SYBR green (Molecular Probes, Leiden, the Netherlands). Most of the primer sequences were obtained from the PrimerBank at Harvard University [101]. Primer sequences are listed in table 1 of the supplemental data (supplemental_ 3). Samples were analyzed in duplicate and standardized to either cyclophilin or 18S expression. Expression levels in isolated villus cells were standardized to villin. Histology For histology studies a fifth, independent experiment was performed, exactly as described in study A. After removal, intestines were divided into 3 equal parts, which are referred to as duodenum, jejunum and ileum, respectively. Each section was prepared using a ‘Swiss role’ technique [102] to evaluate the entire longitudinal section on one slide. Tissues were fixed by immersion in 4% PBS-buffered formaldehyde, processed in an automatic tissue processor, embedded in paraffin, sectioned at 5µm, and stained with haematoxylin and eosin (H&E). Sections were examined on a CKX41 microscope (Olympus, Zoeterwoude, the Netherlands), equipped with calibrated DP- software, version 3.2 (Olympus). This software was used to measure villus height, crypt depth and villus area of 50 villi per section for each animal. Statistical analysis between groups was performed using ANOVA, followed by the Least Significant difference (LSD) post-hoc test.

38 2 Genome-wide analysis of PPARα activation in murine small intestine

Results

PPARα expression in human and murine small intestine To ascertain whether PPARα may be functionally relevant in the small intestine, the expression of PPARα was measured in twenty human tissues by quantitative RT-PCR (qRT-PCR). In humans the highest expression levels of PPARα were observed in kidney, followed by heart, small intestine and liver (Figure 1A). In mice, the expression of PPARα was slightly higher in liver compared to small intestine of SV129 mice, whereas RXRα expression was comparable between both tissues (Figure 1B). These data suggest that PPARα may be functionally relevant in the small intestine.

Figure 1: PPARα is expressed at high levels in human and murine small intestine. (A) Expression levels of PPARα in various human tissues. The top 10 out of the 20 tissues analyzed are presented, PPARα expression in kidney was arbitrarily set to 100%. (B) Murine PPARα and RXR expression levels in small intestine and liver. Data are presented as mean ± standard deviation, n=5. The small intestine was arbitrarily set to 100%. For both human and mouse samples, expression levels were standardized to 18S.

39 Next, the expression levels of PPARα and selected genes along the crypt-villus and proximal- distal axes of the small intestine were examined. As expected, mRNA levels of intestinal alkaline phosphatase (IAP) and villin, two markers for differentiated absorptive epithelial cells [103, 104] were maximal in fraction 1 (Figure 2). Conversely, expression of pancreatic lipase-related protein-2 (PNLIPRP2), a marker for paneth cells located at the base of villi [105], peaked in fractions 6 to 8. Importantly, expression of PPARα declined from villus to crypt cell-enriched fractions, which was mimicked by FATP4 and CD36/FAT, two proteins known to be highly expressed in enterocytes and involved in fatty acid uptake [106, 107]. These data demonstrate that PPARα is predominantly expressed in differentiated enterocytes, and co-localizes with other genes involved in fatty acid metabolism.

Figure 2: PPARα is predominantly expressed in differentiated enterocytes. qRT-PCR was used to determine relative expression levels of PPARα and marker genes in fractions enriched in villus or crypts cells isolated from intestines from adult 129Sv mice. Cell fractions were isolated in two independent experiments, using 3 or 4 mice per isolation. Messenger RNA levels were standardized to cyclophilin; fraction 8 was arbitrarily set to 1. Expression levels of intestinal alkaline phosphatase (IAP), villin, pancreatic lipase-related protein 40 2 (PNLIPRP2), PPARα, fatty acid transport protein 4 (FATP4, SLC27A4), and CD36/FAT. Data are presented as mean ± standard deviation. 2 Genome-wide analysis of PPARα activation in murine small intestine

A similar relationship was observed along the proximal-distal axis. PPARα expression gradually increased from the duodenum throughout the distal jejunum, and then decreased in ileum (Figure 3). The same pattern of expression was observed for FATP4. Expression of L-FABP peaked more proximally, whereas I-FABP expression was highest in the distal jejunum. As expected, the expression of IAP (Figure 3A) and ASBT (Figure 3B) was restricted either to the duodenum or terminal ileum, respectively [108, 109]. Combined, these data demonstrate that PPARαexpression is highest in jejunal villus cells.

Figure 3: PPARα is expressed at highest levels in jejunum. qRT-PCR was used to determine relative expression levels of PPARα and marker genes in sections isolated along the proximal-distal axis of the small intestine from adult 129Sv mice (n=3). The small intestine was divided into 10 equal parts; part 1 refers to the most proximal part (duodenum), part 10 refers to the most distal (terminal ileum). Messenger RNA levels were standardized to cyclophilin; part 10 was arbitrarily set to 1. Expression levels of intestinal alkaline phosphatase (IAP), apical sodium de- 41 pendent bile salt transporter (ASBT), liver-type fatty acid binding protein (L-FABP), intestine-type fatty acid binding protein (I-FABP), PPARα, and fatty acid transport protein 4 (FATP4, SLC27A4). Data are presented as mean ± standard deviation. Comparable results were observed when 18S ribosomal RNA was used as reference. Function of intestinal PPARα as assessed by transcriptome analyses To study the function of PPARα in the small intestine, wild-type and PPARα-null mice were treated with the synthetic PPARα agonist WY14643, followed by analyses of changes in global gene expression using Affymetrix MOE430A arrays. Results on the number of significantly regulated genes are summarized in Figure 4. A complete list of regulated genes is available in supplemental_2. Additional qRT-PCR analyses were performed for selected genes, which confirmed the array results (Figure 8 and supplemental_3). Under control conditions, expression levels of only 21 genes out of the approximately 13,700 genes analyzed were significantly different between wild-type and PPARα- null mice (FC >1,5; FDR<0.05).

-/- vs PPARα control WTcontrol WTWY-14,643 vs WTcontrol (21) (1138)

7 14 1122 0 0 2 0

-/- -/- PPARα WY-14,643 vs PPARα control (2)

Figure 4: Identification of PPARα target genes in small intestine. Affymetrix MOE430A arrays (n=3 per group) were hybridized with RNA isolated from intestines from control and WY14643-treated wild-type and PPARα-null mice. Groups were compared and genes that satisfied the criteria of fold change >1.5 and FDR<0.05 were designated significantly changed genes. Numbers of genes are summarized in a Venn plot. 42 PPARα-/-control: PPARα-null mice that received the control diet; PPARα-/-WY14643: PPARα null mice that received the control diet supplemented with 0.1% WY14643 for five days; WTcontrol: wild-type mice that received the control diet; WTWY14643: wild-type mice that received the control diet supplemented with 0.1% WY14643 for five days. 2 Genome-wide analysis of PPARα activation in murine small intestine

Of these 21 genes, 16 genes were expressed at lower levels and ﬁve genes at higher levels in PPARα-null mice (Figure 5). Most of these genes are known to be involved in lipid metabolism, and several have been identiﬁed as direct PPARα target genes in other tissues [53]. In wild-type mice, activation of PPARα resulted in differential expression of 1,138 genes, of which only 2, BC018473 (hitchhiker, Entrez Gene (EG) ID: 193217) and Abcb1a (Mdr1a, EG ID: 18671) were also altered in the PPARα-null mice upon WY14643 treatment (Figure 4). Thus, in total 1,136 genes were PPARα-dependently regulated in the small intestine; PPARα activation resulted in increased mRNA levels of 567 genes whereas 569 were repressed.

Figure 5: Twenty-one genes are differentially expressed in PPARα-null mice under basal conditions. Heat maps representing the expression levels of 21 genes that were differentially expressed in PPARα-null mice compared to wild type mice under control conditions. (A) Five genes significantly higher expressed in PPARα- null mice, (B) 16 genes significantly lower expressed in PPARα-null mice. *: genes that were also regulated upon PPARα activation. Signal intensities were log2 transformed and sub- jected to hierarchical clustering in Spotfire.

To gain insight into the underlying biological phenomena affected by PPARα activation, a scoring-based resampling method was applied to identify significantly overrepresented Gene Ontology (GO) classes [100]. As input the more than twenty-two-thousand t-test p-values from the probe set comparisons across the two diets in wild-type mice were used. Classes of genes that changed most significantly are listed in Table 1. With respect to the concept Biological Process, terms on this top list were mainly descriptors for fatty acid and lipid metabolism. Other overrepresented GO classes included descriptors for immune system, cell proliferation and differentiation, and programmed cell death (Table 1), suggesting that PPARα is involved in the regulation of these processes in the small intestine. A parallel gene set enrichment analysis (GSEA) was used to focus on groups of genes that comprise specific biochemical, metabolic or signal transduction routes [39]. This method allows the identification of up- or down-regulated processes (Table 2). However, due to overlap in the source databases, several functions are represented multiple times. The outcome of GSEA was similar to that of the GO-based analysis. Remarkable, almost all increased gene sets correspond to metabolic processes, including fatty acids catabolism, mitochondrial oxidative metabolism, and several pathways that feed intermediates into these processes. Other processes of interest that were up-regulated included genes related to steroid- and bile acid metabolism. The cellular responses represented by the down-regulated gene sets were much more diverse, and did not include metabolic pathways. Various pleiotropic signal transduction routes were suppressed and the functional outcomes were summarized as acting on immune system, cell proliferation, migration and differentiation, and apoptosis.

Table 1: Gene Ontology classes overrepresented upon PPARα activation.

Genes in Raw GO ID GO class FDR class Score 1 GO:0006637 Acyl-CoA metabolism 14 5.96 5.5E-11 2 GO:0048005 Antigen presentation, exogenous peptide antigen 8 5.30 1.3E-10 3 GO:0019886 Antigen processing, exogenous antigen via MHC class II 13 4.99 3.1E-11 4 GO:0042591 Antigen presentation, exogenous antigen via MHC class II 16 4.36 1.2E-10 5 GO:0006732 Coenzyme metabolism 35 4.09 3.9E-11 6 GO:0007031 Peroxisome organization and biogenesis 15 4.01 5.0E-11 7 GO:0019395 Fatty acid oxidation 17 3.74 4.5E-11 8 GO:0006631 Fatty acid metabolism 114 3.33 2.9E-11 9 GO:0030333 Antigen processing 38 3.27 3.6E-11 10 GO:0019884 Antigen presentation, exogenous antigen 22 3.18 5.9E-11 11 GO:0019752 Carboxylic acid metabolism 105 2.98 6.7E-11 12 GO:0006725 Aromatic compound metabolism 29 2.74 3.8E-11 13 GO:0019882 Antigen presentation 48 2.60 3.0E-11 14 GO:0006800 Oxygen and reactive oxygen species metabolism 49 2.22 1.6E-10 15 GO:0008203 Cholesterol metabolism 52 1.95 7.8E-11 16 GO:0050776 Regulation of immune response 39 1.95 5.2E-11 17 GO:0006869 Lipid transport 46 1.93 3.5E-11 18 GO:0016064 Humoral defense mechanism (sensu Vertebrata) 54 1.88 9.4E-10 19 GO:0016125 Sterol metabolism 49 1.80 2.9E-11 44 20 GO:0006959 Humoral immune response 59 1.72 1.9E-10 21 GO:0000087 M phase of mitotic cell cycle 95 1.68 6.3E-11 22 GO:0006092 Main pathways of carbohydrate metabolism 89 1.61 4.7E-10 23 GO:0009058 Biosynthesis 120 1.60 4.7E-11 24 GO:0007067 Mitosis 104 1.60 9.4E-11 25 GO:0008610 Lipid biosynthesis 96 1.59 2.4E-10 26 GO:0012502 Induction of programmed cell death 73 1.56 4.3E-11 27 GO:0030036 Actin cytoskeleton organization and biogenesis 96 1.50 3.1E-10 28 GO:0008202 Steroid metabolism 96 1.50 3.4E-11 29 GO:0006814 Sodium ion transport 81 1.50 3.3E-11 30 GO:0043065 Positive regulation of apoptosis 106 1.46 4.1E-11 31 GO:0006917 Induction of apoptosis 97 1.39 8.6E-11 32 GO:0001525 Angiogenesis 89 1.39 2.8E-11 33 GO:0048514 Blood vessel morphogenesis 93 1.38 7.2E-11 34 GO:0016481 Negative regulation of transcription 97 1.35 2.7E-11 35 GO:0006954 Inflammatory response 123 1.25 1.1E-10 A scoring-based resampling method was used to identify significantly overrepresented GO classes upon PPARα activation. As input the more than twenty-two-thousand t-test p-values from the probe set comparisons across the two diets in wild-type mice were used. The analysis was performed using the tool ErmineJ [100]. Only classes for the concept “biological process” with a FDR <0.0001 are shown. For the analysis only classes containing 8 through 125 genes were taken into account. 2 Genome-wide analysis of PPARα activation in murine small intestine FDR 3.1E-05 4.6E-05 5.3E-05 5.6E-05 7.6E-05 8.7E-05 1.0E-04 4.8E-04 6.3E-03 6.6E-03 6.6E-03 6.8E-03 1.3E-02 1.4E-02 1.5E-02 1.6E-02 1.9E-02 2.0E-02 2.1E-02 2.4E-02 2.4E-02 2.4E-02 2.4E-02 2.8E-02 2.8E-02 2.9E-02 2.9E-02 4.2E-02 6.3E-02 6.7E-02 6.9E-02 7.3E-02 7.4E-02 7.9E-02 9.0E-02 9.2E-02 9.7E-02 1.0E-01 1.1E-01 1.2E-01 2.40 2.34 2.33 2.24 2.12 2.12 2.13 2.14 2.04 1.91 1.90 1.90 1.91 1.85 1.84 1.84 1.83 1.81 1.80 1.79 1.77 1.77 1.78 1.78 1.76 1.75 1.74 1.75 1.71 1.67 1.66 1.65 1.64 1.64 1.63 1.61 1.61 1.59 1.59 1.56 NES ES 0.78 0.86 0.84 0.94 0.74 0.86 0.84 0.75 0.62 0.87 0.97 0.79 0.65 0.64 0.69 0.68 0.83 0.67 0.93 0.71 0.73 0.93 0.77 0.59 0.71 0.86 0.85 0.82 0.89 0.60 0.85 0.56 0.60 0.79 0.78 0.48 0.61 0.76 0.73 0.43 9 6 6 5 8 7 8 6 6 7 8 8 9 N 11 50 26 31 15 31 17 20 67 14 36 36 22 23 25 21 16 13 38 18 26 37 27 76 21 46 127 1 1,2 3 3 3 3 2 3 2 1 3 3 1 2 2 3 3 3 2 2 2 2 3 3 2 3 3 3 3 3 3 3 3 2 3 2 2 2 3 2 Up-regulated cellular processes -arrestins in GPCR Desensitization Fatty Acid Beta Oxidation-1 Fatty Acid Beta Oxidation Meta Fatty Toxicity Nuclear Receptors in Lipid Metabolism and biosynthesis Pantothenate and CoA Acid Synthesis Fatty Chain Transport Electron Acid Beta Oxidation Unsaturated Fatty leucine and isoleucine degradation Valine, Acid Beta Oxidation-2 Fatty Synthesis Triacylglyceride metabolism Tyrosine Carbon fixation Acid Beta Oxidation-3 Fatty control of lipid synthesis SREBP Cycle Krebs-TCA Fatty acid metabolism Mitochondrial fatty acid betaoxidation ß Alkaloid biosynthesis II Glycerolipid metabolism 1- and2-Methylnaphthalene degradation Androgen and estrogen metabolism Caprolactam degradation Bile acid biosynthesis Porphyrin and chlorophyll metabolism Benzoate degradation via hydroxylation Steroid Biosynthesis Styrene degradation Fatty acid biosynthesis (path2) FXR and LXR Regulation of Cholesterol Metabolism Galactose metabolism Butanoate metabolism Fluorene degradation Pentose and glucuronate interconversions Oxidative phosphorylation Pentose phosphate pathway Irinotecan pathway PharmGKB Heme Biosynthesis Glycolysis/Gluconeogenesis Adipogenesis FDR 2.5E-03 2.5E-02 2.9E-02 3.2E-02 6.6E-02 7.3E-02 7.5E-02 7.9E-02 8.1E-02 8.3E-02 8.3E-02 8.5E-02 8.6E-02 1.0E-01 1.0E-01 1.1E-01 1.1E-01 1.1E-01 1.1E-01 1.1E-01 1.1E-01 1.1E-01 1.1E-01 1.1E-01 1.1E-01 1.1E-01 1.1E-01 1.1E-01 1.1E-01 1.1E-01 1.1E-01 1.1E-01 1.1E-01 1.1E-01 1.2E-01 1.2E-01 1.2E-01 1.2E-01 1.2E-01 1.2E-01 2.05 1.86 1.87 1.89 1.80 1.78 1.76 1.74 1.75 1.74 1.72 1.76 1.73 1.70 1.67 1.67 1.69 1.64 1.65 1.64 1.64 1.64 1.67 1.63 1.62 1.65 1.61 1.62 1.62 1.66 1.61 1.68 1.63 1.63 1.65 1.65 1.68 1.60 1.68 1.60 NES ES 0.86 0.89 0.49 0.83 0.86 0.63 0.82 0.84 0.57 0.52 0.62 0.76 0.65 0.59 0.69 0.66 0.58 0.69 0.43 0.76 0.66 0.59 0.62 0.54 0.77 0.54 0.80 0.71 0.69 0.80 0.83 0.70 0.75 0.43 0.56 0.71 0.44 0.71 0.44 0.70 8 8 9 8 9 9 7 9 6 N 11 14 12 33 42 76 27 13 25 32 14 20 35 15 17 29 26 41 48 12 13 14 10 35 13 12 155 155 155 174 155 45 1 1 1 1 1 1 1 1 1 1 1 1 2 2 1 3 1 1 5 3 4 2 3 1 3 5 4 3 5 1 2 1 2 1 1 5 2 1 Repressed cellular processes 3 1 Granzyme A mediated Apoptosis Pathway mediated A Granzyme Inflammatory Response Pathway signaling pathway ( CD95 ) FAS Transduction Cell Signal T Apoptosis Caspase Cascade in Pathway Wnt Ca2 cyclic GMP mediated ligand-induced downregulation of EGF CBL TNF of Fas and Nef: negative effector HIV-I fragmentation and tissue homeostasis Apoptotic DNA Regulation Stress Induction of HSP Activation TCR Lck and Fyn tyrosine kinases in initiation of cytokine connection MMP Antigen Receptor B Cell Cytotoxic Cell Surface Molecules T Complement Activation Classical Complement Classical Complement Pathway Smooth muscle contraction Complement Pathway Lectin Induced Complement Pathway G alpha i Pathway Eph Kinases and ephrins support platelet aggregation Hematopoietic cell lineage Activation of Csk by cAMP-dependent Protein Kinase Inhibits BCR Signaling Pathway mCalpain and friends in Cell motility G13 Signaling Pathway Ras-Independent pathway in NK cell-mediated cytotoxicity Bi MAPK signaling pathway Prion Pathway Neurodegenerative Disorders TNFR1 Signaling Pathway SIG BCR Signaling Pathway receptors IL17 Signaling Pathway Acetylcholine Synthesis Prion disease Regulation of actin cytoskeleton Segmentation Clock Focal adhesion Calcium signaling pathway uCalpain and friends in Cell spread Table 2: Gene set enrichment analysis was applied to identify up-regulated and repressed processes after PPARα activation in wild-type mice. PPARα after processes and repressed 2: Gene set enrichment analysis was applied to identify up-regulated Table A s input the average expression values of 14,373 unique transcripts (genes) were used. The analyses were performed as described [39]. Presented are the top 40 significantly false discovery rate (FDR) was calculated to account for multiple hypothesis testing. Sources of the gene sets: 1BioCarta, 2GenMAPP, A repressed and up-regulated processes. top the at overrepresented is set gene this which to degree the is, that set; gene the for score enrichment the is ES set. gene the in genes of number the is N 5STKE. 4SA, 3KEGG, normalized been has it after set gene the for score enrichment the is, that score; enrichment normalized the is NES dataset. expression the in genes of list ranked the of bottom or to account for variations in gene set size. See supplemental text additional info. The functional outcomes of these transcriptome analyses are summarized in Figure 6. The figure underscores the role of PPARα as an important transcriptional regulator in small intestine, governing diverse processes ranging from apoptosis and cell cycle, to the immune response and numerous metabolic pathways. Many genes involved in fatty acid catabolism are known to be direct PPARα target genes [53]. However, the mechanisms by which PPARα activation results in down-regulation of numerous genes are less well understood. We therefore used Ingenuity Pathway Analyses (IPA) to search for biological interaction networks. As input all 1136 PPARα-dependently regulated genes were used. Of these genes, 588 were eligible for network analysis. For the remaining 548 genes molecular interaction information was lacking. IPA computed 83 networks, of which five scored equally best, as judged by a statistical likelihood approach [110].

Figure 6: PPARα regulates a variety of processes in small intestine. Summary of functional implications of PPARα activation as assessed by analyses of pre-deﬁned Nuclear receptors + transcription factors gene sets based on Gene • NR in lipid metabolism and toxicity Ontology, biochemical, • FXR+ LXR regulation of cholesterol metabolism • SREBP control of lipid synthesis metabolic or signal transduction routes. Apoptosis + cell cylce Amino acid metabolism • Granzyme A mediated • Valine, leucine, isoleucine • Caspasecascade degradation • G alpha i pathway • Tyrosine metabolism • Induction of apoptosis P • MAPK signaling pathway • Mitosis Lipid metabolism P • TAG synthesis • Sterol biosynthesis Host defense • Bile acid metabolism • Inflammatory response pathway A • Glycerolipid metabolism • Humoral immune response • Phospholipid metabolism • Antigen procession via MHCII 46 • Antigen presentation via MHCII R • B-cell receptor signaling Fatty acid metabolism • T-cell signal transduction α • FA activation • MMP cytokine connection • FA binding • Complement activation • FA transport (classical and lectin induced) • FA β-oxidation • FA ω-oxidation

Carbohydrate metabolism • Pentose phosphate pathway • Glycolysis/ gluconeogenesis • Hexose biosynthesis

The ﬁve networks with a score of 39 (p < 10-39) were combined to form a composite network representing the underlying biology of PPARα activation in small intestine (Figure 7). This complex comprised a network of 175 unique genes and their interactions. All genes were responsive to PPARα-activation, and every interaction between the genes was supported by published information. In addition, a right-tailed Fisher’s exact test identiﬁed twenty-three 2 Genome-wide analysis of PPARα activation in murine small intestine

canonical pathways significantly affected by PPARα activation (p<0.05) (data not shown). All of these were also identified with GSEA, and six canonical pathways linked to the composite network (Figure 7). Moreover, we identified six nodes that were central in connecting many of the changed genes. As expected, PPARα linked directly to most of the induced genes. The down-regulated genes of the merged network all linked to MYC, caspase 3 (CASP3), major histocompatibility class (MHC) 2 transactivator (MHC2TA), epidermal growth factor receptor (EGFR) or lymphocyte-specific protein tyrosine kinase (LCK), which themselves were also PPARα-dependently down-regulated. No direct interaction between PPARα and the other five central nodes could be identified. Since the maximum network size was limited to 35 genes, all genes presented in the composite network were regulated by PPARα. Yet, this does not imply that all genes connected to the central nodes were regulated. For example, IPA knowledge base linked PPARα and MYC to 109 and 459 genes, respectively (data not shown). Of these, 35 (32%) respectively 60 (13%) genes were significantly regulated (FC>1.5, FDR<0.05) in the small intestine. Taken together, the generation of biological interaction networks identified five genes that may play a central role in mediating the pleiotropic suppressive effects of PPARα activation.

47 PPARα dependent gene regulation occurs in villus cells To conﬁrm that PPARα dependent gene regulation occurred in the differentiated enterocytes, we isolated fractions enriched in villus cells from wild-type and PPARα-null mice treated with WY14643. The expression for selected PPARα target genes identiﬁed in the array analysis was then determined by qRT-PCR (Figure 8A-C). For comparison microarray data and qRT-PCR results of total tissue are presented as well. Expression of “classical” PPARα- target genes, cytochrome P450, family 4, subfamily a, polypeptide 10 (CYP4A10), enoyl-Coenzyme A, hydratase/3- hydroxyacyl Coenzyme A dehydrogenase (EHHADH), 2-4-dienoyl-Coenzyme A reductase 2, peroxisomal (DECR2) and angiopoietin-like 4 (ANGPTL4) [53, 94, 111, 112] was increased by WY14643 in total tissue and villus cells from wild type but not PPARα null mice (Figure 8A).

Figure 8 A: The effect of PPARα activation on expression of “classical” PPARα target genes in small intestine. For selected genes the outcome of PPARα activation was determined in wild-type and PPARα null mice. Independent sets of mice were analyzed. Left column: microarray data obtained from total tissue samples (n=3 mice per group); middle column: qRT-PCR data from total tissue samples (n=3 mice per group); right column: qRT-PCR data from fractions enriched in villus cells (n=3-4 mice per group, fractions f1-f3 were pooled). 48 (A) The effect of PPARα activation on expression of “classical” PPARα target genes in small intestine. [53, 94, 111, 112]: cytochrome P450, family 4, subfamily a, polypeptide 10 (CYP4A10), enoyl-Coenzyme A, hydratase/3-hydroxyacyl Co- enzyme A dehydrogenase (EHHADH), 2-4-dienoyl-Coenzyme A reductase 2, peroxisomal (DECR2), angiopoietin- like 4 (ANGPTL4). 2 Genome-wide analysis of PPARα activation in murine small intestine

Similar expression patterns were observed for four putative intestinal PPARα target genes aldo-keto reductase family 1, member B8 (AKR1B8), glutamate oxaloacetate transaminase 2, mitochondrial (GOT2), farnesoid X receptor; retinoid X receptor (FXR) and diacylglycerol O- acyltransferase 2 (DGAT2), which were specifically induced by WY14643 treatment in total tissue and isolated fractions (Figure 8B). Expression of PPARα itself was increased by WY14643 in total tissue and isolated fractions, indicating autoregulation of PPARα gene expression. Besides PPARα itself we also determined the expression levels of five other nodes identified by network analysis. Expression of CASP3, MHC2TA, EGFR, MYC and LCK was PPARα-dependently down-regulated (Figure 8C).

Figure 8 B: The effect of PPARα activation on expression of putative intestinal PPARα target genes identiﬁed by microarray analysis. For selected genes the outcome of PPARα activation was determined in wild-type and PPARα null mice. Independent sets of mice were analyzed. Left column: microarray data obtained from total tissue samples (n=3 mice per group); middle column: qRT-PCR data from total tissue samples (n=3 mice per group); right column: qRT-PCR data from fractions enriched in villus cells (n=3-4 mice per group, fractions f1-f3 were pooled). 49

(B) The effect of PPARα activation on expression of putative intestinal PPARα target genes: aldo-keto reductase family 1, member B8 (AKR1B), glutamate oxaloacetate transaminase 2, mitochondrial (GOT2), farnesoid X receptor; retinoid X receptor (FXR), diacylglycerol O-acyltransferase 2 (DGAT2). Figure 8 C: The effect of PPARα activation on expression of putative intestinal PPARα target genes identified by microarray analysis. For selected genes the outcome of PPARα activation was determined in wild-type and PPARα null mice. Independent sets of mice were analyzed. Left column: microarray data obtained from total tissue samples (n=3 mice per group); middle column: qRT-PCR data from total tissue samples (n=3 mice per group); right column: qRT-PCR data from fractions enriched in villus cells (n=3-4 mice per group, fractions f1-f3 were pooled). (C) The effect of PPARα activation on PPARα expression itself and five central nodes identified by network analysis: caspase 3 (CASP3), major histocompatibility class (MHC) 2 transactivator (MHC2TA), epidermal growth factor receptor (EGFR), myelocytomatosis oncogene (MYC) and lymphocyte protein tyrosine kinase (LCK). Micro- array data are presented 50 as fold change compared to wild-type mice fed the control diet; qRT-PCR data were standardized to cyclophillin (total tissue) or villin (cell fractions), wild-type control was then arbitrarily set to 1, data for total tissue samples are mean ± standard deviation. ** indicate a significant difference between the groups: for microarray data, gene satisfied the criterion of FDR < 5% (q-value < 0.05), and >1.5 fold changed in wild-type mice upon WY14643 treatment, but were not changed in treated PPARα-knockout mice; for total tissue qRT- PCR data p <0.01. 2 Genome-wide analysis of PPARα activation in murine small intestine

Thus, for all genes analyzed we observed great similarity in regulation in total tissue between microarray and qRT-PCR, as well as between total tissue and isolated villus cells. This again demonstrates the robustness of our microarray analysis.

PPARα activation by fenofibrate To study the specificity of PPARα activation, we compared the effects of WY14643 with the effects of fenofibrate another PPARα agonist. qRT-PCR was used to analyze expression of a set of “classical” and putative PPARα target genes as well as central nodes, which also were studied in isolated villus cells (Figure 9). The effects of fenofibrate were comparable to those of WY14643, as both agonists induced the expression of PPARα itself and (putative) PPARα target genes, and reduced the expression of the central nodes MHC2TA and CASP3. However, at the same concentration fenofibrate was less potent in activating or repressing genes compared to WY14643. Thus, we show that activation of PPARα by two different ligands does not result in qualitative differences in expression of a specific set of genes.

Figure 9: Activation of PPARα by two different ligands does not result in qualitative differences on expression of a set of genes. Wild-type (white bars) and PPARα-null (black bars) mice were treated with WY14643 (0.1% w/w) or fenofibrate (0.1% w/w) for five days (n=5 per group). For selected genes the outcome in expression was determined using qRT-PCR analysis. Messenger RNA 51 levels were standardized to cyclophilin; wild-type control was set to 1. Upper row “classical” PPARα target genes, middle row: putative intestinal PPARα target genes, lower row: PPARα and two other central nodes. Data are presented as mean ± standard deviation. a, b, c: different letters indicate a significant difference between the treatment groups (p<0.05), using ANOVA.

Morphometric changes of the villus of mice treated with WY14643 One of the outcomes of the transcriptome analyses was that PPARα activation suppressed cell proliferation and apoptosis, and enhanced cell differentiation. We hypothesized that morphologically this should result in elongated villi. We therefore prepared paraffin sections of control and treated intestines from wild-type and PPARα-null mice, and studied gross intestinal morphology and height of the villi (Figure 10, Table 3). Under basal conditions no differences were observed with respect to form, structure and morphometric parameters between wild-type and PPARα-null mice. In wild-type mice, treatment with WY14643 significantly increased the height of the villus, whereas no effect was observed in PPARα-null mice. Crypt depth was not affected by WY14643 treatment in both groups. These data indicate that PPARα activation specifically increases villus height, but does not alter crypt depth. The morphometric assessments are in line with the functional outcomes of the transcriptome analyses.

I III V VII

II IV VI VIII

WT-control WT-WY14643 PPARα-/--control PPARα-/--WY14643

Figure 10: Activation of PPARα increases villus height. Histological stainings were performed on jejunal sections from control and WY14643 treated wild-type and PPARα-null (PPARα-/-) mice. Two representative sections of different mice per treatment group are shown, (I, II) wild-type mice fed the control diet, (III, IV) wild-type mice fed the control diet supplemented with 0.1% WY14643 for five days, (V, VI) PPARα-null mice fed the control diet, 52 (VII, VII) PPARα-null mice fed the control diet supplemented with 0.1% WY14643 for five days. Original magni- fication 88x.

Table 3: Villus height, crypt depth and villus area upon PPARα activation in the jejunum.

Diet Wild-type PPARα-/- Villus height (µm) Control 331.9 ± 46.5 a 372.0 ± 32.3 a,b 0.1% WY14643 403.4 ± 43.4 b 341.7 ± 40.7 a Crypt depth (µm) Control 64.6 ± 5.9a 66.5 ±10.5a 0.1% WY14643 59.8 ± 4.8a 60.9 ± 6.3a Area (µm2) Control 17229.0 ± 3208.6 a 21149.1 ± 2851.1 a,b 0.1% WY14643 23031.4 ± 2991.4 b 17990.3 ± 1402.4 a Activation of PPARα significantly increases villus height and area. Histological stainings (HE) were performed on jejunal sections from control and treated wild-type and PPARα-null mice (for photos of stainings see Figure 10). Villus height, crypt depth and villus area were measured as explained in materials and methods; data are presented as mean ± standard deviation; n=5mice/group, N=20; a, b: different letters indicate a significant difference between the treatment groups (p<0.05), as identified using ANOVA. 2 Genome-wide analysis of PPARα activation in murine small intestine

53 Figure Figure 7: Generation of biological interaction networks identiﬁes central genes. Ingenuity pathway analysis was used to search interaction for networks biological (BINs). As input the dependently 1136 PPARα regulated genes were used, of which 588 were eligible for the analysis genes). (focus Five networks scored equally best, these were merged and are presented. Six major affected canonical are pathways, linked indicated as CP, to the merged network. Genes marked with a “*” are genes that are detected two or more times on Color the coding: array. Red: up-regulated gene, green: down-regulated gene. The intensity of the colors indicates the degree of up or down-regulation, respectively; a greater intensity represents a higher degree of regulation. Discussion

In this study we set out to determine the role of PPARα in the small intestine using genomics tools. We ﬁnd that PPARα is very well expressed in the small intestine in both mouse and human. PPARα expression is highest in villus cells and peaks in the proximal jejunum. Activation of PPARα results in altered expression of a large set of genes involved in a variety of pathways, including intestinal lipid handling, cell cycle, differentiation, apoptosis and host defense. These data suggest an important role for PPARα in the regulation of gene expression in the small intestine. Under control (fed) conditions we observed few changes in gene expression between wild-type and PPARα null mice. Only 21 genes were signiﬁcantly altered, most of which are involved in lipid metabolism. These observations are in accordance with numerous studies showing that the effect of PPARα deletion becomes mainly noticeable under conditions of metabolic stress, and does not a priori imply that the physiological role of PPARα is limited to lipid metabolism [5, 59, 113, 114].

Our finding that PPARα is expressed highest in villus cells from the jejunum is supported by crude tissue distribution studies performed in rats [88, 89]. Intestinal PPARα expression thus coincides with the main anatomical location where long-chain fatty acids are digested, taken up, and secreted into the body, suggesting an important regulatory function of PPARα in these processes. Indeed, activation of PPARα resulted in the specific induction of genes involved in fatty acid uptake, -binding, -transport and -catabolism, as well as genes involved in triacylglycerol-, and glycerolipid metabolism, in the small intestine of wild-type but not PPARα- null mice. Although PPARα is known to regulate fatty acid metabolism in other organs [53, 54], the link between PPARα and regulation of lipid handling in the intestine is new. Hence, at the level of the enterocyte, PPARα serves as a fatty acid sensor that is part of a feed-forward mechanism in which fatty acids stimulate their own catabolism, storage, transfer through the enterocyte, and secretion as triacylglycerols. PPARα thus tightly controls the intracellular levels of these potential toxic compounds. 54 Next to genes involved in fatty acid and triacylglycerol metabolism, genes coding for transcription factors and enzymes connected with steroid (sterol) and bile acid metabolism, including FXR, SHP, and SREBP1 are specifically induced. This demonstrates that cross-talk between PPARα and other lipid-regulated transcription factors also occurs in small intestine, in addition to liver [115, 116], and is functionally in line with the well-established roles of bile acids in absorption of dietary lipids [93]. Moreover, since it was recently shown that intestinal FXR target genes are also involved in enteroprotection and inhibition of bacterial overgrowth [117], our data suggest that activation of PPARα might also influence epithelial barrier function. 2 Genome-wide analysis of PPARα activation in murine small intestine

In addition to PPARα itself, network analyses identified five central genes that connected many of the repressed genes. These central genes (MYC, EGFR, CASP3, LCK and MHC2TA) themselves were also PPARα dependently down-regulated. The MYC gene encodes a multifunctional nuclear phosphoprotein and plays a role in cell cycle progression, apoptosis, cellular transformation and differentiation. The results of the microarray study show that MYC and related genes are repressed in a PPARα-dependent manner, including cyclin D1 (CCND1), CASP3, nuclear factor-κB (NFkB), signal transducer and activator of transcription 1 (STAT1) and EGFR via platelet-derived growth factor receptor β (PDGFRB). Inasmuch as MYC is known to repress major histocompatibility complex, class I-B (HLA-B), PDGFRB and N-myc downstream-regulated gene 1 (NDRG1) expression [118-120], the observed up-regulation of these genes by PPARα is likely mediated through repression of MYC. Our data for the first time connect PPARα with regulation of cell proliferation, differentiation and apoptosis in the small intestine. Our results indicate that an important consequence of intestinal PPARα activation is blocking cells in transition to the G1-S checkpoint of the cell cycle, resulting in reduced proliferation and increased differentiation of cells [121]. We speculate that the specific down-regulation in villus cells of CASP3, a key enzyme in the apoptotic cascade [122], points to inhibition of apoptosis, as has been reported for other cell types [123-126]. Since cell shedding is strongly associated with apoptosis, this in turn may result in reduced shedding of cells from the villus tips [127, 128]. Functional support of the gene expression data is provided by morphometric data showing a significant 22% increase in villus height, which is accompanied by a 34% increase in villus area. These findings were not observed in PPARα-null mice treated with WY14643. Combined, our data demonstrate that PPARα represses cell growth and apoptosis, and stimulates cell differentiation, resulting in an increased number of mature absorptive enterocytes. We believe that this may be an important adaptation mechanism of the small intestine aimed at adjusting lipid absorptive capacity to increased dietary fats (i.e. hydrolyzed TAGs).

Our study connects PPARα with the immune system of the small intestine. Activation of PPARα 55 suppresses complement activation, antigen presentation and B-cell receptor signaling. It is known that PPARα-null mice have abnormally prolonged hepatic responses to inflammatory stimuli [129]. In vascular cells the expression of interleukin-6, the vascular cell adhesion molecule (VCAM), and cyclooxygenase-2 in response to cytokine activation can be inhibited by PPARα ligands [130]. In these cells PPARα ligands may inhibit the functional expression of NF-κB, in part by augmenting the expression of inhibitor of NF-κB (IκBα) [131]. However, inhibitory effects of PPARα in the intestine as well as on the innate immune system have not been reported before. Although the lymphoid tissue of the gut is the primary system for host defense, it is known that intestinal epithelial cell expresses MHC2 molecules and can function as antigen-presenting cells, thus being capable of regulating mucosal T-cell responses [132-134]. Moreover, there is evidence that some of the complement proteins, such as complement C3, are synthesized in enterocytes [135, 136]. Network analyses showed that the repression of the positive “master” regulator MHC2TA by PPARα is likely responsible for the suppression of MHC2 gene transcription, whereas suppressed B-cell receptor signaling is linked to decreased expression of LCK [137]. Taken together, our data show that PPARα influences the immune and inflammatory response in the intestine, and support the possibility that enterocytes are involved in a local response to injury/inflammation at the epithelial surface. A repression of the inflammatory response in the intestine by PPARα might be therapeutically valuable for patients with inflammatory bowel disease. Although several studies suggest a link between PPARγ and inflammatory bowel disease [138-141], hardly anything is known about the effect of PPARα.

In summary, by using a combination of functional genomics experiments and current bioinformatics tools, we are the first to identify the pathways and processes under control of PPARα in the small intestine. Our data provide new insight into the role of PPARα in the small intestine and may be of particular importance for the development of fortified foods, and for prevention and therapies for treating obesity and inflammatory bowel diseases.

56 2 Genome-wide analysis of PPARα activation in murine small intestine

57 3 Organ-speciﬁc function of PPARα as revealed by gene expression proﬁling

Meike Bünger, Philip de Groot, Linda Sanderson, Larry Singh, Madeleen Bosma, Sridhar Hannenhalli, Sander Kersten, Michael Müller, Guido Hooiveld Abstract

Background The peroxisome proliferator-activated receptor alpha (PPARα) is a fatty acid- activated transcription factor expressed at high levels in small intestine and liver, two key organs in whole-body metabolism of (dietary) triglycerides (TGs). The importance of PPARα in TG metabolism is illustrated by the fact that agonists targeting PPARα, such as poly-unsaturated fatty acids (PUFAs) and ﬁbrates, are commonly used for the treatment of dyslipidemia. Whether PPARα functions differently in small intestine and liver is currently unknown. In this study, genome wide effects of PPARα activation in small intestine and liver were examined and compared.

Methods Wild-type and PPARα-null mice were fed a diet supplemented with the PPARα agonist WY14643 for 5 days. RNA was isolated from intestines and livers, and hybridized to Affymetrix mouse genome arrays. PPARα-dependently regulated genes were identiﬁed, and functionally analyzed based on Gene Ontology, metabolic pathways and signal transduction routes. A secretome analysis was performed to identify genes encoding secreted proteins. Finally, we examined the WY14643-induced interaction of PPARα with co-regulators, regulated transcription factors, and the occurance of cis regulatory motifs in promoters of regulated genes.

Results PPARα regulated 2 distinct, but overlapping sets of genes in intestine and liver. Based on the commonalities and differences we are able to assign its organ specific functions. In addition, PPARα regulated 326 genes encoding secreted proteins, and 249 genes encoding transcription factors. We also identified several organ-specific interactions with nuclear regulators as well as cis regulatory motifs that consisted of three transcription factors including PPARα.

Conclusion PPARα in small intestine and liver controls the expression of distinct sets of genes, demonstrating that these organs use different strategies to adapt to high levels of agonists. We identiﬁed a variety of PPARα-regulated secreted proteins, that may be important for intercellular signalling or can function as biomarkers for PPARα activation. Finally, we provide clues for the 60 transcriptional mechanisms involved in the observed organ-speciﬁcity.

3 Organ-speciﬁc function of PPARα as revealed by gene expression proﬁling

Introduction

The peroxisome proliferator-activated receptor alpha (PPARα) is a ligand-activated transcription factor with diverse functions and is activated by a variety of synthetic compounds, including drugs used for the treatment of dyslipidemia and type 2 diabetes [53, 54, 63]. High affinity natural ligands include eicosanoids, unsaturated as well as long-chain fatty acids, and their activated derivatives (acyl-CoA esters) [84-87, 142, 143]. In analogy with other nuclear receptors, when activated PPARα forms obligate heterodimers with the retinoid X receptor and stimulates gene expression by binding to peroxisome proliferator response elements (PPREs) located in the promoter regions of target genes [53]. For efficient transcriptional regulation by PPARα also coregulators are required. These are molecules that assist PPARα to positively or negatively influence the transcription of target genes, and thereby comprise an integral part of the transcriptional circuitry [144-146]. PPARα is also able to repress transcription by directly interacting with other transcription factors and interfere with their signaling pathways, a mechanism commonly referred to as transrepression [20, 53]. PPARα is expressed in a variety of tissues, including liver and small intestine [70, 88, 89]. In liver PPARα is critical for the coordinate transcriptional activation of genes involved in nutrient metabolism [53, 54] and it is suggested that PPARα is an important regulator of the hepatic acute phase response [147, 148]. Even though the small intestine expresses PPARα at high level and is frequently exposed to high levels of PPARα agonists via the diet, the role of PPARα in this organ was not investigated until recently. We showed that intestinal PPARα plays an important role, governing diverse processes ranging from numerous metabolic pathways and lipid handling to the control of apoptosis and cell cycle genes [70]. Thus, although PPARα activation and target gene regulation has been studied in a range of organs, until now nothing is known about its differential function in specific organs. Therefore, in the present study we identified the commonalities and differences of PPARα activation on gene expression in small intestine and liver, with special emphasis on functional implications and secreted proteins. As our data demonstrated that small intestine and liver responded differentially to PPARα activation, we also aimed at explaining the transcriptional 61 mechanism underlying these differences. The results contribute to the elucidation of potential transcriptional mechanisms by which PPARα activation can accomplish such organ-specificity. Materials and Methods

Experimental design and tissue handling Four month old male wild-type and PPARα-null mice were fed chow, or chow supplemented with WY14643 (0.1% w/w, Chemsyn, Lenexa, KA) for 5 days (n=6 mice per group, N=24). On the sixth day, mice were anaesthetized with a mixture of isoﬂuorane (1.5%), nitrous oxide (70%) and oxygen (30%). Liver and small intestine were excised and ﬂushed with ice-cold PBS, snap-frozen in liquid nitrogen and stored at -80 °C until subsequent analyses

RNA isolation and quality control The RNA isolation and quality control procedures have been earlier described [70].

Affymetrix GeneChip oligoarray hybridization and scanning For microarray analyses, we used RNA isolated from liver and full-length small intestine which subsequently was hybridized on an Affymetrix GeneChip Mouse Genome 430 2.0 arrays. This array detects 39,626 transcripts that represent approximately 16,000 known genes. For each experimental group, four biological replicates were hybridized, thus in total 32 arrays (16 arrays per organ) were performed. Total RNA (5 μg) was labelled using the Affymetrix One-cycle Target Labeling Assay kit (Affymetrix, Santa Clara, CA). The correspondingly labelled RNA samples were hybridized on Affymetrix Mouse Genome 430 2.0 plus arrays, washed, stained and scanned on an Affymetrix GeneChip 3000 7G scanner. Detailed protocols for the handling 62 of the arrays can be found in the Genechip Expression Analysis Technical Manual, section 2, chapter 2 (Affymetrix; P/N 701028, revision 5), and are also available upon request. Array data are available at the gene expression omnibus (GEO) [48], accession number GSE8345.

Statistical analyses and functional interpretation of microarray data Packages from the Bioconductor project [96], integrated in an in-house developed on-line management and analysis database for multiplatform microarray experiments (Gavai et al, submitted), were used for analysing the scanned Affymetrix arrays. Various advanced quality metrics, diagnostic plots, pseudo-images and classification methods were applied to ascertain only excellent quality arrays were used in the statistical analyses [149]. An extensive description of the applied criteria is available upon request. Probesets were redefined according to Dai et al. [150], because the genome information utilized by Affymetrix at the time of designing the arrays is not current anymore, which may result in an unreliable reconstruction of expression levels. In this study probes were reorganized based on the Entrez Gene database, build 36, version 2 (remapped CDF v9). Expression estimates were obtained by GC-robust multi-array (GCRMA) 3 Organ-specific function of PPARα as revealed by gene expression profiling

analysis, employing the empirical Bayes approach for background adjustment, followed by quantile normalization and median polish summarization [97]. Differentially expressed probesets were identified using linear models, applying moderated t-statistics that implement empirical Bayes regularisation of standard errors [151] P-values were corrected for multiple testing using a false discovery rate method [152]. Genes that satisfied the criteria of FDR < 1% (q-value < 0.01) were considered to be significantly regulated. Genes that were regulated in wild-type mice upon WY14643 treatment, but were not significantly changed in treated PPARα-knockout mice, were designated PPARα regulated. Functional analyses were performed as explained in detail [70].

Quantitative reverse-transcription polymerase chain reaction Quantitative RT-PCR analyses were performed as reported before [70].

In vitro PPARα-coregulator interaction assay WY14643-modulated interaction of coregulators with PPARα was assessed using PamChipH peptide microarrays (PamGene, ‘s-Hertogenbosch, The Netherlands), and were used according to the manufacturer’sinstructions. Upon binding a ligand, PPARα undergoes a conformational change, which leads to alteration of the interaction and association with coregulator molecules that can be measured [143].

Secreted protein analysis The secretome is defined as the population of gene products that are secreted from the cell [153]. With the goal of using the most comprehensive secretome list available, we compiled secreted protein data from three public databases. First, we included the list of secreted proteins listed in the LOCATE database [154]. LOCATE is a curated database of predicted protein properties derived from a computational conceptual translation of mouse transcripts, the subcellular locations of selected proteins as determined by a high-throughput immunofluorescence-based assay, and by the manual review of over 1700 peer-reviewed publications. Secondly, we included the content of 63 the eukaroyotic subcellular localization database (eSLDB) [155]. For each transcript this database contains information on experimental localization, homology based annotation and the predicted localization computed with machine learning based methods. Finally, results of ngLOC [156], an n-gram-based Bayesian classification method that predicts the localization of a protein sequence over ten distinct subcellular organelles, were included. A gene was assigned to encode for a secreted (extracellular) protein when it fulfilled this criterium in two out of the three databases.

Identiﬁcation of cis-regulatory modules Individual and combinations of putative transcription factor (TF) binding sites were identiﬁed in promoter regions of genes using the tool TREMOR [157]. In higher eukaryotes, TFs rarely operate by themselves, but rather bind to DNA in cooperation with other transcription factors [158-160]. The DNA footprint of this set of factors is called a cis-regulatory module (CRM), which consists of a set of TFBSs located in the promoter region of the gene being regulated. Promoter regions of regulated genes were retrieved from the Genomatix Promoter Database, release 4.5. Results

PPARα-dependent gene regulation in small intestine and liver In the current study we aimed to investigated the commonalities and differences of PPARα activation in small intestine and liver. To this end wild-type and PPARα-null mice were treated with the potent PPARα agonist WY14643 for 5days, followed by analyses of changes in global gene expression on Affymetrix Mouse Genome 430 2.0 arrays. Since we a priori assumed that differential regulation between the two organs occured, significantly regulated genes were selected based on a statistical significance only, thus no fold-change criterium was applied. All genes that satisfied the criterion of FDR < 1% (q-value < 0.01) in any comparison were considered to be significantly regulated. A complete list of regulated genes is available at http://www.bunger.nl/thesis in supplemental_data_1. Results on the number and overlap of significantly regulated genes in the two organs are summarized in Figure 1. We identified 1,626 and 3,884 PPARα-regulated genes in small intestine and liver, respectively. Of these, 711 genes were regulated in both organs, and this set of genes is defined as the commonly-regulated or overlaping gene set. The small intestine- and liver-specific PPARα-regulated genes sets comprised of 915 and 3173 genes, respectively. Of the intestine- specific genes, 572 genes were suppressed and 343 genes were induced after PPARα activation, in liver these numbers were 1707 and 1466, respectively (Figure 1, Table 1).

Small intestine Liver

PPAR -/- vs WT -/- vs α control control WT WY -14,643 vs WTcontrol PPAR α control WT control WT WY -14,643 vs WTcontrol (23) (1626) (110) (3886 21 98 2 1605 11 3786 0 1 0 0 0 1 0 0

PPAR -/- vs PPAR -/- -/- vs -/- α WY -14,643 α control PPAR α WY -14,643 PPAR α control (0) (2) 64

1626 915 711 3173 3884

Sm. intestine-specific Overlap Liver-specific

Figure 1: Identification of PPARα target genes in small intestine and liver. Affymetrix MOE430 2.0 arrays (n=4 per group) were hybridized with RNA isolated from small intestines and livers of control and WY14643-treated wild-type and PPARα-null mice. Groups were first compared for each organ separately, genes that satisfied the criteria of FDR<0.01 were designated as significantly changed genes. Numbers of changed genes are summarized in a Venn plot per organ. PPARα-/-c.: PPARα-null mice that received the control diet; PPARα-/- WY14643: PPARα null mice that received the control diet supplemented with 0.1% WY14643 for five days; WTc.: wild-type mice that received the control diet; WTWY14643: wild-type mice that received the control diet supplemented with 0.1% WY14643 for five days. 3 Organ-specific function of PPARα as revealed by gene expression profiling

Functional analyses of regulated processes in small intestine and liver To gain insight into the underlying biological phenomena differentially affected by PPARα activation, a scoring-based resampling method was applied to identify signiﬁcantly overrepresented Gene Ontology (GO) classes in intestine and liver [100]. The same signiﬁcance cut-off criterium (FDR<0.001) was used for both organs. We found 92 and 97 GO classes overrepresented in small intestine and liver, respectively (Figure 2).

Figure 2: Venn plot of overrepresented Gene Ontolo- gy classes of small intestine and liver. A scoring-based Overrepresented GO classes resampling method was used to identify signiﬁcantly overrepresented Gene Ontology (GO) classes upon PPARα activation in small intestine and liver. As input the more than twenty-two-thousand t-test p-values from 35 57 the probe set comparisons across the two diets and or- 40 gans in wild-type mice were used. The same signiﬁcance cut-off criterium (FDR<0.001) was used for both organs. Only classes for the concept “biological process” with a FDR <0.0001 are shown. For the analysis only classes containing 8 through 125 genes were taken into account. Intestine-specific Overlap Liver-specific We found 92 and 97 GO classes overrepresented in small intestine and liver, respectively (Figure 2).The analysis was performed using the tool ErmineJ [35].

The 57 classes that were overrepresented in both organs comprised 38 terms that were mainly descriptors for “lipid, carbohydrate or amino acid metabolism”. The other 19 classes could be grouped together as being involved in either “cell cycle regulation” or “immune response”. Intestine- specific overrepresented GO classes included terms that in addition to processes such as fatty acid, cholesterol and sterol biosynthesis and glucose and lipid catabolism described antigen processing and presentation, and B- and T-cell activation and regulation. Liver-specific overrepresented GO classes were more diverse, and described among others hemostasis, regulation of body fluids, protein amino acid glycosylation, glycerophospholipid metabolism, activation of plasma proteins during acute inflammatory response, cell redox homeostasis, acute-phase response, complement 65 activation, alternative pathway and hormone metabolism. A list of all altered GO descriptors can be found at http://www.bunger.nl/thesis in the supplemental_data_4. A parallel gene set enrichment analysis (GSEA) was used to focus on groups of genes that comprise specific biochemical, metabolic or signal transduction routes [39]. This well accepted method allowed us to identify up- or down-regulated processes (Table 2). However, due to overlap in the source databases, several functions (gene sets) are represented multiple times. Not surprisingly, gene sets that were induced in both organs corresponded to processes such as fatty acid metabolism (catabolic as well as anabolic, and related regulatory gene sets) and energy metabolism (electron transport chain, Krebs TCA cycle and oxidative phophorylation). On the other hand urea cycle, antigen processing and presentation as well as the IL4 pathway were repressed in both organs (Table 2). 9.1E-05 4.2E-04 1.2E-03 3.0E-02 1.8E-01 1.4E-03 6.3E-02 2.0E-02 3.6E-02 1.3E-01 3.5E-02 7.2E-04 6.3E-02 1.8E-01 3.5E-02 1.5E-02 1.5E-04 6.4E-04 6.3E-02 1.3E-01 1.3E-01 2.5E-01 FDR <9.1E-05 <9.1E-05 <9.1E-05 <9.1E-05 <9.1E-05 <9.1E-05 <9.1E-05 <9.1E-05 <9.1E-05 <9.1E-05 2.48 2.39 2.40 2.36 2.47 2.22 2.07 1.99 1.76 2.29 1.48 2.24 1.98 1.66 1.80 1.73 1.54 1.74 2.01 1.66 1.48 1.73 1.83 2.18 -2.05 -2.17 -1.73 -2.73 -1.64 -2.78 -1.65 -1.56 NES Liver 0.85 0.87 0.92 0.90 0.91 0.89 0.91 0.80 0.69 0.88 0.56 0.74 0.65 0.61 0.71 0.63 0.56 0.71 0.75 0.59 0.58 0.69 0.71 0.69 -0.80 -0.82 -0.55 -0.85 -0.64 -0.86 -0.63 -0.60 ES FDR 3.8E-05 4.9E-05 3.8E-04 6.0E-04 1.8E-03 3.2E-03 3.4E-03 4.8E-03 6.8E-03 3.1E-02 4.0E-02 5.5E-02 8.5E-02 1.5E-01 2.5E-01 2.3E-02 3.0E-02 6.9E-02 1.3E-01 2.0E-01 2.1E-01 2.1E-01 2.3E-01 <3.8E-05 <3.8E-05 <3.8E-05 <3.8E-05 <3.8E-05 <3.8E-05 <3.8E-05 <3.8E-05 <3.8E-05 intestine 2.72 2.63 2.45 2.42 2.41 2.59 2.41 2.30 2.29 2.15 2.23 2.09 2.07 2.00 1.96 1.96 1.92 1.89 1.74 1.69 1.63 1.56 1.44 1.30 -1.78 -1.75 -1.60 -1.49 -1.37 -1.35 -1.34 -1.32 NES Small 0.81 0.88 0.83 0.85 0.82 0.96 0.96 0.84 0.84 0.76 0.76 0.63 0.61 0.65 0.71 0.64 0.63 0.70 0.60 0.55 0.58 0.56 0.52 0.36 ES -0.66 -0.61 -0.46 -0.41 -0.50 -0.38 -0.47 -0.47 64 36 29 29 34 23 15 24 23 28 32 63 68 40 26 40 37 21 29 41 25 23 26 97 18 22 52 58 19 60 21 22 N 7 targets) α 2 2 3

66 3 2 3 3 2 all intestine and liver 3 7 3 7 2 in sm 3 2 3 2 3 3 3 3 3 2 3 3 5 3 3 2 Induced in small intestine and liver 2 Repressed 7 PPAR signaling pathway PPAR Fatty acid metabolism Fatty acid beta oxidation 1 Mitochondrial fatty acid oxidation Fatty acid beta oxidation meta Peroxisomal fatty acid oxidation Mitochondrial fatty acid betaoxidation Fatty acid synthesis Lipogenesis Nuclear receptors in lipid metabolism and toxicity Electron transport chain Adipocytokine signaling pathway Butanoate metabolism Propanoate metabolism Pyruvate metabolism Beta-alanine metabolism Glycerolipid metabolism Oxidative phosphorylation Urea cycle and metabolism of amino groups Urea cycle and metabolism of amino groups Antigen processing and presentation Complement and coagulation cascades Heparan sulfate biosynthesis Complement and coagulation cascades Nitrogen metabolism Interleukin 4 pathway Miscellaneous lipid metabolism (list biased towards putative PPAR leucine and isoleucine degradation Valine, cycle Krebs-TCA ligation Benzoate degradation via CoA synthesis Triacylglyceride cycle) Citrate cycle (TCA Table 2: Altered gene sets in small intestine and liver Altered 2: Table 3 Organ-speciﬁc function of PPARα as revealed by gene expression proﬁling - 3.0E-02 2.9E-02 3.9E-04 1.7E-01 1.3E-01 9.3E-03 4.2E-03 7.7E-04 2.5E-01 1.1E-01 1.1E-01 6.3E-02 7.9E-02 FDR 2.06 1.50 1.55 1.87 1.92 2.01 1.41 1.58 1.58 1.66 1.63 -1.82 -1.82 NES Liver 0.64 0.58 0.50 0.59 0.63 0.72 0.46 0.68 0.53 0.67 0.64 -0.59 -0.63 ES FDR 4.5E-05 3.8E-02 9.0E-03 1.6E-02 2.1E-02 6.8E-02 1.3E-01 1.3E-01 1.8E-01 2.2E-01 2.2E-01 2.5E-01 2.5E-01 intestine 2.16 1.70 -1.89 -1.83 -1.81 -1.60 -1.49 -1.47 -1.38 -1.32 -1.33 -1.29 -1.29 NES Small 0.67 0.60 ES -0.48 -0.61 -0.49 -0.42 -0.41 -0.45 -0.36 -0.51 -0.37 -0.46 -0.44 50 31 27 75 85 70 39 81 15 62 22 24 102 N 3

67 3 2 1 2 1 3 1 2 3 1 Induced in small intestine and repressed liver 3 2 Repressed in small intestine and upregulated liver Metabolism of xenobiotics by cytochrome p450 Androgen and estrogen metabolism Cell cycle TNFR1 signaling pathway Hematopoietic cell lineage Cell cycle Cell cycle-G1 to S control reactome ECM-receptor interaction P53 signaling pathway Alpha6-beta4-integrin netpath 1 Cyclins and cell cycle regulation Cell cycle G1 S check point DNA replication reactome DNA sented at the top or bottom of the ranked list of genes in the expression dataset. NES is the normalized enrichment score; that is, the enrichment score enrichment the is, that score; enrichment normalized the is NES dataset. expression the in genes of list ranked the of bottom or top the at sented Table 2 (continued): Altered gene sets in small intestine and liver Altered 2 (continued): Table wild- in activation PPARα after liver and intestine small in processes repressed and upregulated identify to applied was analysis enrichment set Gene type As mice. input the t-test values of the signiﬁcance analysis of being 16,297 and unique FDR<0.25 genes a were having used. The sets analyses gene were of performed intersection as the described is [39]. table this in Presented used. were genes 250 and 15 between size a having sets gene Only are signiﬁcantly repressed or upregulated in both organs. The false discovery rate (FDR) 7SK was calculated to account 6GEARRAY, TRANSDUCTION, for 5SIGNALING ALLIANCE, multiple 4SIGNALING 3KEGG, hypothesis testing. 2GENMAPP, 1BIOCARTA, : sets: gene the of Sources manual. N is the number of genes in the gene set. ES is the enrichment score for the gene set; that is, the degree to which this gene set is overrepre for the gene set after it has been normalized to account variations in size. See supplemental text additional info. Table 1: Summary of regulated genes after PPARα activation by WY14643 for 5 days in small intestine and liver Transcription Secreted proteins2 N factors1 PPARα dep. regulated. % of % of genes N N regulated genes regulated genes

Both up 376 10 2.7 17 4.5 Both down 201 13 6.5 18 9.0 Overlap SI-up, L-down 73 3 4.1 3 4.1 SI-down, L-up 61 0 0.0 4 6.6 Up 343 18 5.2 22 6.4 SI-speciﬁc Down 572 33 5.8 37 6.5 Up 1466 47 3.2 59 4.0 L-speciﬁc Down 1707 99 5.8 166 9.7 1 [170] 2 [154-156] and Materials and Methods, Secretome Analysis

Several gene sets were oppositely regulated in both organs, i.e. these were repressed (or induced) in small intestine and induced (or repressed) in liver. Those gene sets mapped to cell cycle regulation, DNA replication or genes that were shown to be altered via WNT, MYC or E2F1 signalling (Table 2, and data not shown). In contrast to liver, we identified a rather large number of suppressed gene sets in small intestine after PPARα activation. Most of these reflect processes related to the immune system (Table 3A). Specifically induced gene sets in the small intestine were among others bile acid synthesis, lipoprotein metabolism, ABC-transporters, tryptophan metabolism and adipogenesis. Gene sets that were specifically induced in liver are cell cycle G2/M checkpoint, cell communication, sphingolipid and glycerophospholipid metabolism as well as proteasome degradation (Table 3B). Taken together, our data clearly demonstrate that small intestine and liver respond differentially 68 to PPARα, both with respect to indivual genes and corresponding cellular processes. 3 Organ-specific function of PPARα as revealed by gene expression profiling FDR <4.1E-05 4.1E-05 3.1E-03 1.5E-02 2.8E-02 2.9E-02 3.3E-02 3.4E-02 3.8E-02 3.9E-02 5.3E-02 5.3E-02 5.8E-02 7.4E-02 8.8E-02 9.3E-02 9.6E-02 1.0E-01 1.0E-01 1.1E-01 1.1E-01 1.5E-01 1.5E-01 1.5E-01 1.5E-01 1.6E-01 1.7E-01 1.7E-01 2.0E-01 2.2E-01 NES 2.31 2.16 1.96 1.82 1.75 1.75 1.72 1.72 1.70 1.68 1.64 1.64 1.62 1.58 1.55 1.54 1.53 1.50 1.51 1.51 1.49 1.44 1.42 1.42 1.42 1.40 1.39 1.39 1.35 1.33 ES 0.79 0.72 0.65 0.68 0.59 0.56 0.61 0.65 0.56 0.63 0.61 0.43 0.52 0.62 0.49 0.60 0.56 0.50 0.61 0.45 0.48 0.48 0.57 0.55 0.57 0.51 0.45 0.42 0.54 0.36 N 31 33 35 21 31 49 25 19 37 21 22 126 44 16 46 19 25 40 15 66 47 36 15 17 16 22 46 63 16 113 1 3 3 3 3 3 3 7 3 2 3 3 3 2 3 3 3 3 3 7 2 3 1 3 3 3 2 2 3 2 Upregulated cellular processes Pantothenate and CoA biosynthesis Pantothenate and CoA Carbon fixation degradation Lysine metabolism Tyrosine metabolism Tryptophan Cholesterol synthesis esterification Nuclear receptors in lipid metabolism and toxicity Bile acid biosynthesis Glutathione metabolism Limonene and pinene degradation Lipoprotein metabolism Glycolysis/gluconeogenesis Pentose phosphate pathway ABC transporters - general Porphyrin and chlorophyll metabolism Adipogenesis Bisphenol a degradation 1- and 2-methylnaphthalene degradation Oxidative stress Nuclear receptors Ascorbate and aldarate metabolism Fructose and mannose metabolism Glycolysis and gluconeogenesis Akt signaling pathway Alkaloid biosynthesis II Nucleotide sugars metabolism Amino acid metabolism Arachidonic acid metabolism Amyotrophic lateral sclerosis (ALS) 1-tissue-endocrine and CNS FDR <8.0E-04 8.0E-04 1.0E-03 1.2E-03 1.7E-03 2.1E-03 5.9E-03 6.6E-03 7.8E-03 8.6E-03 8.7E-03 8.8E-03 9.2E-03 9.4E-03 9.9E-03 2.3E-02 2.3E-02 2.5E-02 2.6E-02 2.7E-02 2.7E-02 2.7E-02 2.9E-02 2.9E-02 2.9E-02 3.0E-02 3.0E-02 3.6E-02 3.7E-02 3.9E-02 Small intestine specific NES -2.11 -2.18 -2.12 -2.07 -2.02 -2.04 -1.96 -1.95 -1.92 -1.90 -1.90 -1.88 -1.88 -1.90 -1.88 -1.80 -1.78 -1.79 -1.79 -1.77 -1.76 -1.76 -1.75 -1.74 -1.74 -1.74 -1.73 -1.71 -1.71 -1.70 ES -0.54 -0.51 -0.59 -0.62 -0.52 -0.57 -0.60 -0.71 -0.64 -0.52 -0.57 -0.63 -0.73 -0.62 -0.45 -0.49 -0.59 -0.44 -0.48 -0.53 -0.58 -0.48 -0.45 -0.63 -0.66 -0.49 -0.62 -0.55 -0.54 -0.51

69 N 124 149 64 43 99 62 39 19 28 68 41 27 15 31 152 75 27 146 92 45 30 76 111 20 16 64 20 31 32 48 1 3 1 3 3 4 2 3 3 1 2 1 1 2 3 4 2 2 2 2 5 2 2 3 3 2 1 4 2 2 Repressed cellular processes T-cell signal transduction T-cell Inflammatory response pathway dependent signaling pathway in IL12 and STAT4 signaling pathway ( cd95 ) FAS disease Huntington’s netpath 6 KIT-receptor cell receptor signaling pathway T TNF and of FAS nef negative effector HIV-i B-cell receptor netpath 12 Phosphatidylinositol signaling system Leukocyte transendothelial migration B-cell receptor signaling pathway TH1 development Olfactory transduction IL-5 netpath 17 Sig chemotaxis The co-stimulatory signal during t-cell activation 2-tissues-blood and lymph Smooth muscle contraction TGF-beta signaling pathway TGF-beta-receptor netpath 7 IL-3 netpath 15 Sig BCR signaling pathway Sig PIP3 signaling in b lymphocytes Delta-Notch netpath 3 Cell adhesion molecules (CAMS) Caspase cascade in apoptosis Glycosphingolipid biosynthesis - ganglioseries Hypertrophy model Prostaglandin synthesis regulation T-cell-receptor netpath 11 T-cell-receptor Table 3A: Intestine-specifically altered gene sets upon PPARα activation gene sets upon PPARα 3A: Intestine-specifically altered Table FDR 1.7E-04 2.1E-02 2.3E-02 5.9E-02 1.1E-01 1.3E-01 1.3E-01 1.9E-01 2.1E-01 2.2E-01 2.19 1.80 1.78 1.67 1.57 1.53 1.53 1.46 1.45 1.44 NES ES 0.86 0.73 0.69 0.69 0.59 0.48 0.52 0.51 0.46 0.65 25 21 27 18 30 91 60 53 78 15 N 1 3 1 2 3 3 3 3 3 3 Upregulated cellular processes Proteasome Ribosome Ether lipid metabolism Cell cycle G2 M checkpoint Sphingolipid metabolism Cell communication Proteasome degradation Glycerophospholipid metabolism Pyrimidine metabolism Phosphoinositides and their downstream targets FDR 2.5E-04 4.1E-02 5.7E-02 1.1E-01 2.2E-01 Liver specific 0.0E+00 -2.54 -2.10 -1.78 -1.75 -1.67 -1.58 NES ES -0.67 -0.83 -0.66 -0.57 -0.67 -0.53

70 18 24 42 17 38 N 176 3 3 2 3 2 3 Repressed cellular processes 2-tissues-internal organs Blood clotting cascade Maturity onset diabetes of the young Glycine, serine and threonine metabolism Cysteine metabolism Linoleic acid metabolism Table 3B: Liver-specifically altered gene sets (biological processes only) upon PPARα activation only) upon PPARα gene sets (biological processes altered 3B: Liver-specifically Table Gene set enrichment analysis was applied to identify up-regulated and repressed processes in small intestine (A) or liver (B) activation after in PPARα wild-type As mice. input the t-test values of the significance analysis of 16,297 unique genes were used. The analyses were performed as described [39]. Presented in table (A) are 30 the Top gene sets having N a FDR<0.25 and being specific for the 6GEARRAY. TRANSDUCTION, small 5SIGNALING intestine. ALLIANCE, The false 4SIGNALING discovery 3KEGG, rate (FDR) was calculated 2GENMAPP, to account for multiple 1BIOCARTA, sets: gene the of Sources testing. hypothesis of bottom or top the at overrepresented is set gene this which to degree the is, that set; gene the for score enrichment the is ES set. gene the in genes of number the is normalized been has it after set gene the for score enrichment the is, that score; enrichment normalized the is NES dataset. expression the in genes of list ranked the to account for variations in gene set size. 3 Organ-specific function of PPARα as revealed by gene expression profiling

Secretome analysis: the systemic effects of PPARα regulation To identify potentially secreted proteins regulated by PPARα, we next performed a secretome analysis. Secreted proteins are critical to both short- and long-range intercellular signalling in multicellular organisms [153, 161]. Moreover, these proteins may be used as biomarkers for PPARα activation. Three databases that used curated data and different prediction algorithms were mined to identify potentially secreted proteins [154-156]. A gene was assigned to encode for a secreted protein when it fulfilled this criterium in two out of the three databases. An overview of the secretome analysis is given in supplemental Figure 3. PPARα regulated 326 genes encoding secreted proteins in small intestine and liver, of which 42 were regulated in both organs (organ-independent), 59 only in small intestine, and 225 genes encoded for liver-specific regulated extracellular proteins (Table 1). The relative contribution of secreted proteins compared to all PPARα regulated genes varied between 5.9% and 7.1%, depending on the organ-specificity (Table 1). Interestingly, in small intestine the percentages of induced and respressed genes encoding secreted protein were similar (6.4% vs 6.5%). However, for liver we observed a disbalance since many more genes enconding secreted proteins were repressed than induced (9.7% vs 4.0%). With respect to individual genes, our approach identified several genes, such as ANGPTL4, FGF21, FGF15 and IL18 that were recently shown to code for PPARα-dependently regulated secreted proteins, or being important inter- or intra-organ messengers [162-167]. This demonstrated the validity of our approach. Taken together, we identified a number of PPARα-regulated secreted proteins that are envisioned to have important intercellular functions.

Nuclear co-regulators

Figure 3: Coregulaor recruitment assay with WY14643 and microarray expression data of nuclear coregulators. The Nuclear Receptor PamChipH assay was used to measure the interaction between PPARα and immobilized peptides corresponding to specific coregulator-nuclear receptor binding regions. Measurements were performed in the presence of control (EtOH) or WY14643 (5 mM). (A) Shows both images. Arrows point to selected co-activators which except for PPARGC1A expression was confirmed by qRT-PCR in small intestine and liver wild-type mice (n=5). Images were taken after 100 msec exposure time. (B) Values are the measurements of light intensities of both images. (C) Microarray expression data for nuclear coregulators in small intestine and liver (values are mean, +/- SD, n=4). Identifying mechanisms for organ-specific PPARα-dependent gene regulation Our genome-wide screening revealed that PPARα in small intestine and liver regulated two distinct, but overlapping sets of genes. This distinction could be due to differences in expression of transcriptional coregulator proteins that are recruited upon binding of WY14643 to PPARα, or similarly, may be caused by tissue-specific employment of transcription factors. Coregulators are proteins which can repress (corepressors) or enhance (coactivators) nuclear receptor transcriptional activity by bridging transcription factors to the basic transcription machinery and by specifically modifying chromatin structure [14, 168, 169]. The Nuclear Receptor PamChip arrays was used to identify nuclear coregulators that were recruited or repulsed after activation of PPARα with WY14643 (Figure 3A, B). WY14643 promoted the interaction between PPARα and several coregulator peptides, and reduced the interaction with the corepressor NCOR1. We next determined the intestinal and hepatic mRNA expression levels of the coregulators (Figure 3C). Almost all coregulators were expressed at higher levels in intestine than liver, although the differences varied to a great extent, which was also found before [169]. Similar observations were made when comparing expression levels within each organ. For selected nuclear coactivators (CREBBP, NRIP1, MED1 and NCOA1) expression levels were measured by qRT-PCR analyses, which confirmed the results of the microarray data (Figure 4). Combined, the data presented in Figure 6 suggest that (part of) the organic-specific effects of PPARα activation may indeed be due to distinct expression levels of specific coregulators that are recruited by WY14643, such as JMJD1C, NCOA1, NCOR1, NRIP1 and PPARGC1. In parallel, we investigated whether the organ specific effects could be related to differences

Figure 4: qRT-PCR analysis of selected coactivators. qRT-PCR was used to determine the differential expression of s genes encoding coactivators in small intestine and (n=4 mice per group). The small intestine was set to 1. Values are means ± standard deviation (SD). Messenger RNA levels were standardized to cyclophilin. Data are presented as mean ± standard deviation. The small intestine was arbitrarily set to 1. 3 Organ-speciﬁc function of PPARα as revealed by gene expression proﬁling

Table 5: Organ-independently regulated transcription factors.

Small intestine Liver

WTWY/ WTWY/ Gene EGID Description WTc FDR WTc FDR (5d) (5d) PPARα 19013 peroxisome proliferator activated receptor alpha 6.49 1.4E-05 1.61 1.2E-04 Hlf 217082 hepatic leukemia factor 2.90 3.3E-03 1.48 9.3E-03 Nr1i2 (Pxr) 18171 nuclear receptor subfamily 1, group I, member 2 2.57 1.9E-04 2.51 8.4E-05 Yaf2 67057 YY1 associated factor 2 2.37 2.8E-06 2.39 2.7E-06 Nr2c1 (Tr2) 22025 nuclear receptor subfamily 2, group C, member 1 2.12 1.4E-03 2.12 1.7E-03 Nr1i3 (Car) 12355 nuclear receptor subfamily 1, group I, member 3 2.11 4.6E-03 1.94 6.3E-05 Hif1a 15251 hypoxia inducible factor 1, alpha subunit 1.94 4.4E-05 1.62 2.3E-04 Creb3l3 208677 cAMP responsive element binding protein 3-like 3 1.79 4.0E-05 1.56 4.7E-04 Zfp655 72611 zinc ﬁnger protein 655 1.51 3.8E-03 1.48 5.2E-03 Hdac3 15183 histone deacetylase 3 1.50 8.4E-04 1.37 1.5E-03 Arntl 11865 aryl hydrocarbon receptor nuclear translocator-like -7.84 9.4E-05 -9.27 9.3E-03 Bcl3 12051 B-cell leukemia/lymphoma 3 -2.54 2.2E-04 -1.87 5.1E-04 E2f5 13559 E2F transcription factor 5 -1.63 2.9E-03 -1.55 4.5E-03 Foxa2 15376 forkhead box A2 -1.46 9.1E-03 -2.32 2.6E-04 Ccnh 66671 cyclin H -1.43 7.1E-03 -1.48 4.2E-04 Pknox1 18771 Pbx/knotted 1 homeobox -1.41 7.3E-03 -2.13 7.0E-05 Pml 18854 promyelocytic leukemia -1.40 3.5E-03 -1.48 6.8E-03 Irf6 54139 interferon regulatory factor 6 -1.40 1.2E-03 -2.67 1.5E-03 Id2 15902 inhibitor of DNA binding 2 -1.38 9.8E-03 -3.56 9.8E-06 Irf1 16362 interferon regulatory factor 1 -1.38 8.6E-03 -1.89 3.3E-03 Tcf21 21412 transcription factor 21 -1.35 3.4E-03 -2.79 5.2E-05 Creb1 12912 cAMP responsive element binding protein 1 -1.29 5.2E-03 -1.37 7.6E-04 Arid1a 93760 AT rich interactive domain 1A (Swi1 like) -1.28 6.6E-03 -1.61 4.4E-04 Klf15 66277 Kruppel-like factor 15 4.17 2.5E-05 -1.41 8.9E-03 Nr0b2 (Shp) 23957 nuclear receptor subfamily 0, group B, member 2 2.94 2.0E-03 -2.50 7.2E-03 Cebpa 12606 CCAAT/enhancer binding protein (C/EBP), alpha 1.38 6.3E-03 -1.85 1.1E-03

in expression and regulation of transcription factors. From a curate list consisting of 1156 of transcription factors [170], we were able to link 1054 entries to the Affymetrix GeneChip Mouse Genome 430 2.0 array. Of the 1054 TF mRNAs probed, levels of 77 and 172 TFs were significantly 73 altered after PPARα activation in small intestine and liver, respectively, (Table 1). Twenty-six TFs were regulated in both organs (Table 4). Besides autoregulation of PPARα itself, four other nuclear receptors, namely SHP (NR0B2), PXR (NR1I2), CAR (NR1I3) and TR2 (NR2C1), were regulated in a PPARα dependent way in both organs. We identified 51 respectively 146 transcription factors that were specifically regulated in intestine and liver (Tables 5, 6). Regulated in intestine only were among others TCF23, FXR (NR1H4), ISX, SREBF1, REV-ERB (NR1D2) and HNF4g, all induced, whereas E2F2, PPARGC1A, IRF8, CIITA, MYB, RELB and SP1 were all reduced in intestine. Table 5: Small intestine-specifically regulated transcription factors. Small intestine-specific Fold change* Gene EGID Description WTWY/ FDR WTc (5d) Tcf23 69852 transcription factor 23 12.64 1.2E-05 Dbp 13170 D site albumin promoter binding protein 4.88 6.8E-03 Klf11 194655 Kruppel-like factor 11 4.62 1.4E-05 Tef 21685 thyrotroph embryonic factor 4.50 1.5E-03 Srebf1 20787 sterol regulatory element binding factor 1 3.44 6.0E-06 Nr1h4 (FXR) 20186 nuclear receptor subfamily 1, group H, member 4 2.08 4.5E-05 Gzf1 74533 GDNF-inducible zinc finger protein 1 2.06 5.4E-04 Nr1d2 (Rev-erb) 353187 nuclear receptor subfamily 1, group D, member 2 1.78 1.4E-03 Hnf4g 30942 hepatocyte nuclear factor 4, gamma 1.72 1.1E-03 Zfp160 224585 zinc finger protein 160 1.71 1.9E-03 Bin1 30948 bridging integrator 1 1.69 3.1E-03 Isx 71597 intestine specific homeobox 1.67 3.5E-04 Rorc 19885 RAR-related orphan receptor gamma 1.47 9.1E-03 Hic2 58180 hypermethylated in cancer 2 1.36 3.0E-03 Vdr 22337 vitamin D receptor 1.35 6.2E-03 Pou3f3 18993 POU domain, class 3, transcription factor 3 1.29 8.3E-03 Nr1h3 (LXR) 22259 nuclear receptor subfamily 1, group H, member 3 1.27 4.8E-03 Gata6 14465 GATA binding protein 6 1.25 7.0E-03 Atf3 11910 activating transcription factor 3 -3.09 2.9E-03 E2f2 242705 E2F transcription factor 2 -2.74 6.0E-03 Spib 272382 Spi-B transcription factor (Spi-1/PU.1 related) -2.32 5.9E-05 Ppargc1a 19017 peroxisome proliferative activated receptor, gamma, coactivator 1 alpha -1.97 4.2E-03 Irf8 15900 interferon regulatory factor 8 -1.91 1.5E-03 Fli1 14247 Friend leukemia integration 1 -1.81 8.3E-04 Zfp26 22688 zinc finger protein 26 -1.74 2.7E-04 Baz1a 217578 bromodomain adjacent to zinc finger domain 1A -1.69 1.1E-03 Ciita 12265 class II transactivator -1.63 2.9E-03 Relb 19698 avian reticuloendotheliosis viral (v-rel) oncogene related B -1.62 3.8E-04 Elf3 13710 E74-like factor 3 -1.62 8.4E-03 Mtf2 17765 metal response element binding transcription factor 2 -1.57 1.1E-03 Hes1 15205 hairy and enhancer of split 1 (Drosophila) -1.55 9.4E-03 74 Myb 17863 myeloblastosis oncogene -1.55 5.5E-03 Fubp1 51886 far upstream element (FUSE) binding protein 1 -1.51 2.8E-03 Tbx3 21386 T-box 3 -1.49 6.2E-03 Mll1 214162 myeloid/lymphoid or mixed-lineage leukemia 1 -1.47 5.3E-03 Snai2 20583 snail homolog 2 (Drosophila) -1.47 6.9E-03 Ezh1 14055 enhancer of zeste homolog 1 (Drosophila) -1.46 1.0E-02 Atoh1 11921 atonal homolog 1 (Drosophila) -1.42 7.8E-03 Bdp1 544971 B double prime 1, subunit of RNA polymerase III TIFIIIB -1.38 9.8E-04 Id4 15904 inhibitor of DNA binding 4 -1.38 1.7E-03 Sp1 20683 trans-acting transcription factor 1 -1.38 9.3E-03 Etv3 27049 ets variant gene 3 -1.37 9.1E-03 Ikzf1 22778 IKAROS family zinc finger 1 -1.37 3.0E-03 Atbf1 11906 AT motif binding factor 1 -1.34 6.2E-03 Crsp3 70208 cofactor required for Sp1 transcriptional activation, subunit 3 -1.33 8.6E-03 Trim27 19720 tripartite motif protein 27 -1.33 7.3E-03 Mkrn1 54484 makorin, ring finger protein, 1 -1.32 4.4E-03 Sox4 20677 SRY-box containing gene 4 -1.32 2.3E-03 Atf7ip 54343 activating transcription factor 7 interacting protein -1.32 6.0E-03 Bmi1 12151 Bmi1 polycomb ring finger oncogene -1.31 5.2E-03 Hod 74318 homeobox only domain -1.27 9.0E-03 *Values are “absolute” Fold changes, ratios <1 were divided by -1. 3 Organ-specific function of PPARα as revealed by gene expression profiling

Likewise PEG3, CBFA2T3H, NFE2L1, TRIM24, ZFP810, JUN, RARB, STAT5A and CEBPB were more than 2-fold induced in liver only, and KLF12, IRF2, MAFB, ELF1, STAT5B, STAT2, FOXA1 and ESR1 were PPARα-dependently expressed in liver. These data show that PPARα controls expression of many DNA-binding proteins, who in turn may initiate or regulate transcription of genes. In higher eukaryotes, TFs rarely operate by themselves, but rather bind to DNA in cooperation with other transcription factors [158-160]. The DNA footprint of this set of factors is called a cis- regulatory module (CRM). In addition to the identification of changed transcription factors upon PPARα activation, we identified transcription factor binding sites (TFBS) that were significantly enriched in the promoters of genes belonging to one of the three PPARα-dependently regulated subsets of genes (overlap, intestine-specific or liver-specific) and selected the CRMs that comprised PPAR. In the promoter regions of genes that were upregulated in both organs we identified 78 significantly enriched TFBS (FDR<0.01), including (as expected) a PPAR responsive element (PPARα:RXRα). In total 23 CRMs were found that comprised two other partners in addition to PPAR. Of these 23 CRMs for 9 the second partner was Pax4 (40%), for 5 R (22%) and for 4 E2F1 (17%), all combined with a third, varying transcriptional partner. Pax4 is a transcriptional repressor that plays a critical role in the differentiation and development of pancreatic islet beta cells. R and E2F1 both have an important role in controlling cell cycle. Promoters of the induced intestinal genes were significantly enriched for 51 TFBS (FDR<0.01), again including the PPRE (PPARα:RXRα). Fiftyseven CRMs could be identified that comprised two other TFBS next to PPARα. The AHR:ARNT was identified 15 times (30%) as second partner, and E2F1 13 times (25%). AHR:ARNT is a ligand-activated receptor hat activates the expression of multiple phase I and II xenobiotic chemical metabolizing enzyme. Finally, in the promoters of genes specifically induced in liver we identified 88 significantly enriched TFBS (FDR<0.01) including PPREs. Although many hits contained recognition sites for AHR:ARNT, ZID, or NF-1, only one PPAR- comprising CRM was enriched. In this case the transcriptional partner was CREB. Thus, the CRM analyses identified various combinations of transcription factors that were significantly 75 enriched in promoter regions of sets of regulates genes.

Table 6: Liver-specifically regulated transcription factors. Liver-specific Fold change* Gene EGID Description FDR WTWY/ WTc (5d) Peg3 18616 paternally expressed 3 22.51 1.0E-08 core-binding factor, runt domain, alpha subunit 2, translocated to, 3 Cbfa2t3h 12398 homolog (human) 17.41 3.5E-07 Myc 17869 myelocytomatosis oncogene 5.67 9.0E-05 Nfe2l1 18023 nuclear factor, erythroid derived 2,-like 1 4.96 3.3E-06 Trim24 21848 tripartite motif protein 24 4.68 6.5E-04 Zfp810 235050 zinc finger protein 810 2.65 2.4E-04 Jun 16476 Jun oncogene 2.55 3.0E-03 Rarb 218772 retinoic acid receptor, beta 2.50 1.6E-04 Stat5a 20850 signal transducer and activator of transcription 5A 2.31 7.3E-03 Gtf2h1 14884 general transcription factor II H, polypeptide 1 2.26 3.0E-06 Tfdp1 21781 transcription factor Dp 1 2.24 4.0E-03 Pknox2 208076 Pbx/knotted 1 homeobox 2 2.23 2.0E-03 Lmo4 16911 LIM domain only 4 2.14 3.8E-03 Olig1 50914 oligodendrocyte transcription factor 1 2.07 5.4E-04 Cebpb 12608 CCAAT/enhancer binding protein (C/EBP), beta 2.03 1.1E-04 Supt3h 109115 suppressor of Ty 3 homolog (S. cerevisiae) 2.00 4.2E-04 Ubp1 22221 upstream binding protein 1 1.90 4.5E-05 Csda 56449 cold shock domain protein A 1.85 8.3E-06 Mybbp1a 18432 MYB binding protein (P160) 1a 1.82 4.9E-04 Ezh2 14056 enhancer of zeste homolog 2 (Drosophila) 1.80 4.0E-03 Foxp2 114142 forkhead box P2 1.79 1.3E-04 Grhl1 195733 grainyhead-like 1 (Drosophila) 1.77 1.3E-03 Sp3 20687 trans-acting transcription factor 3 1.75 3.9E-04 Tfam 21780 transcription factor A, mitochondrial 1.72 3.3E-05 Zfp36l2 12193 zinc finger protein 36, C3H type-like 2 1.68 6.3E-04 Rnf24 51902 ring finger protein 24 1.62 5.5E-04 Rrn3 106298 RRN3 RNA polymerase I transcription factor homolog (yeast) 1.57 4.8E-04 Npat 244879 nuclear protein in the AT region 1.57 1.7E-03 Zfp423 94187 zinc finger protein 423 1.57 1.1E-03 Hmg20b 15353 high mobility group 20 B 1.56 2.1E-03 Creg1 433375 cellular repressor of E1A-stimulated genes 1 1.55 8.4E-05 Rb1 19645 retinoblastoma 1 1.55 2.4E-03 TAF6-like RNA polymerase II, p300/CBP-associated factor (PCAF)- 76 Taf6l 225895 associated factor 1.55 2.3E-03 Ar 11835 androgen receptor 1.54 6.3E-03 Brpf3 268936 bromodomain and PHD finger containing, 3 1.50 4.6E-03 Rqcd1 58184 rcd1 (required for cell differentiation) homolog 1 (S. pombe) 1.45 3.7E-03 Gtf2f2 68705 general transcription factor IIF, polypeptide 2 1.45 2.3E-03 Hdac5 15184 histone deacetylase 5 1.45 8.9E-03 Lztr1 66863 leucine-zipper-like transcriptional regulator, 1 1.41 7.2E-03 Gtf3c1 233863 general transcription factor III C 1 1.38 5.6E-03 Cutl1 13047 cut-like 1 (Drosophila) 1.38 3.9E-03 Gtf2i 14886 general transcription factor II I 1.32 3.0E-03 COP9 (constitutive photomorphogenic) homolog, subunit 5 Cops5 26754 (Arabidopsis thaliana) 1.30 1.3E-03 TAF4B RNA polymerase II, TATA box binding protein (TBP)-associated Taf4b 72504 factor 1.28 2.9E-03 Rxrb 20182 retinoid X receptor beta 1.25 3.2E-03 Phf10 72057 PHD finger protein 10 1.23 4.5E-03 Hmgn1 15312 high mobility group nucleosomal binding domain 1 1.20 9.9E-03 Hes6 55927 hairy and enhancer of split 6 (Drosophila) -5.54 7.4E-07 Nfil3 18030 nuclear factor, interleukin 3, regulated -5.28 2.2E-03 Irf5 27056 interferon regulatory factor 5 -4.92 1.7E-05 3 Organ-specific function of PPARα as revealed by gene expression profiling

Liver-specific (continued) Fold change* Gene EGID Description FDR WTWY/ WTc (5d) Crem 12916 cAMP responsive element modulator -3.98 8.0E-06 Nab2 17937 Ngfi-A binding protein 2 -2.92 1.6E-04 Klf12 16597 Kruppel-like factor 12 -2.82 2.1E-05 Tcea3 21401 transcription elongation factor A (SII), 3 -2.80 1.2E-05 Irf2 16363 interferon regulatory factor 2 -2.80 3.4E-04 v-maf musculoaponeurotic fibrosarcoma oncogene family, protein B Mafb 16658 (avian) -2.69 7.5E-03 Elf1 13709 E74-like factor 1 -2.58 1.7E-03 Stat5b 20851 signal transducer and activator of transcription 5B -2.57 1.7E-04 Prdm2 110593 PR domain containing 2, with ZNF domain -2.56 2.6E-05 Bhlhb2 20893 basic helix-loop-helix domain containing, class B2 -2.50 2.4E-03 Tle2 21886 transducin-like enhancer of split 2, homolog of Drosophila E(spl) -2.47 4.5E-04 Sox5 20678 SRY-box containing gene 5 -2.41 7.0E-04 Stat2 20847 signal transducer and activator of transcription 2 -2.34 2.3E-05 Nmi 64685 N-myc (and STAT) interactor -2.32 1.1E-05 Foxa1 15375 forkhead box A1 -2.28 2.0E-05 Esr1 13982 estrogen receptor 1 (alpha) -2.21 7.0E-05 Xbp1 22433 X-box binding protein 1 -2.21 1.1E-05 Atf4 11911 activating transcription factor 4 -2.15 3.7E-04 Nfix 18032 nuclear factor I/X -2.14 2.5E-06 Zfp36l1 12192 zinc finger protein 36, C3H type-like 1 -2.08 3.9E-04 Mlx 21428 MAX-like protein X -2.06 4.1E-05 BC063263 434178 cDNA sequence BC063263 -2.06 1.9E-04 Mlxipl 58805 MLX interacting protein-like -2.04 2.5E-04 Isgf3g 16391 interferon dependent positive acting transcription factor 3 gamma -1.93 1.0E-04 Sertad1 55942 SERTA domain containing 1 -1.93 5.3E-03 Foxa3 15377 forkhead box A3 -1.92 4.1E-04 Jarid1b 75605 jumonji, AT rich interactive domain 1B (Rbp2 like) -1.91 1.7E-03 Lsr 54135 lipolysis stimulated lipoprotein receptor -1.82 8.6E-05 Phf17 269424 PHD finger protein 17 -1.80 1.7E-03 Zbtb20 56490 zinc finger and BTB domain containing 20 -1.78 5.5E-05 Crebl1 12915 cAMP responsive element binding protein-like 1 -1.77 1.4E-04 Nr2f6 13864 nuclear receptor subfamily 2, group F, member 6 -1.70 3.1E-04 Stat3 20848 signal transducer and activator of transcription 3 -1.70 2.5E-04 Rbl2 19651 retinoblastoma-like 2 -1.69 3.2E-04 77 Zfpm1 22761 zinc finger protein, multitype 1 -1.67 7.3E-03 SWI/SNF related, matrix associated, actin dependent regulator of Smarcd2 83796 chromatin, subfamily d, member 2 -1.66 4.4E-05 Irf7 54123 interferon regulatory factor 7 -1.65 2.4E-03 Onecut1 15379 one cut domain, family member 1 -1.65 8.7E-03 Rbak 57782 RB-associated KRAB repressor -1.62 5.8E-04 Irf3 54131 interferon regulatory factor 3 -1.62 8.2E-05 Gata4 14463 GATA binding protein 4 -1.62 5.1E-03 Cnot4 53621 CCR4-NOT transcription complex, subunit 4 -1.61 5.1E-04 Znrd1 66136 zinc ribbon domain containing, 1 -1.61 1.7E-03 Tshz2 228911 teashirt zinc finger family member 2 -1.60 7.9E-03 Hdac11 232232 histone deacetylase 11 -1.60 9.1E-04 Rfx5 53970 regulatory factor X, 5 (influences HLA class II expression) -1.59 2.1E-03 Lhx2 16870 LIM homeobox protein 2 -1.58 6.4E-04 Pqbp1 54633 polyglutamine binding protein 1 -1.58 1.8E-04 Zfp148 22661 zinc finger protein 148 -1.58 1.5E-04 Ctnnb1 12387 catenin (cadherin associated protein), beta 1 -1.58 3.5E-04 Safb 224903 scaffold attachment factor B -1.57 3.5E-04 Nr5a2 26424 nuclear receptor subfamily 5, group A, member 2 -1.57 1.6E-03 Liver-specific Fold change* Gene EGID Description FDR WTWY/ WTc (5d) Epc1 13831 enhancer of polycomb homolog 1 (Drosophila) -1.55 2.5E-04 Bptf 207165 bromodomain PHD finger transcription factor -1.54 1.8E-04 Zfp281 226442 zinc finger protein 281 -1.54 8.4E-03 Stat6 20852 signal transducer and activator of transcription 6 -1.51 8.2E-03 Atf5 107503 activating transcription factor 5 -1.50 1.3E-03 Cnot2 72068 CCR4-NOT transcription complex, subunit 2 -1.49 8.3E-05 Tead3 21678 TEA domain family member 3 -1.47 6.9E-03 Zbtb48 100090 zinc finger and BTB domain containing 48 -1.45 4.3E-03 Zfp354a 21408 zinc finger protein 354A -1.44 1.4E-03 Tceb3 27224 transcription elongation factor B (SIII), polypeptide 3 -1.43 1.0E-03 Whsc2 24116 Wolf-Hirschhorn syndrome candidate 2 (human) -1.43 1.3E-03 Zbtb7b 22724 zinc finger and BTB domain containing 7B -1.43 7.0E-04 SWI/SNF related, matrix associated, actin dependent regulator of Smarce1 57376 chromatin, subfamily e, member 1 -1.43 9.3E-04 Sin3a 20466 transcriptional regulator, SIN3A (yeast) -1.42 2.2E-03 Tcf4 21413 transcription factor 4 -1.42 1.3E-03 Rbbp4 19646 retinoblastoma binding protein 4 -1.42 3.4E-03 TAF2 RNA polymerase II, TATA box binding protein (TBP)-associated Taf2 319944 factor -1.41 5.5E-03 Atf2 11909 activating transcription factor 2 -1.41 7.9E-04 Vps72 21427 vacuolar protein sorting 72 (yeast) -1.41 5.8E-03 Repin1 58887 replication initiator 1 -1.41 1.6E-03 Ahr 11622 aryl-hydrocarbon receptor -1.40 2.7E-03 Rara 19401 retinoic acid receptor, alpha -1.39 5.5E-03 Rnf2 19821 ring finger protein 2 -1.39 1.8E-03 TAF6 RNA polymerase II, TATA box binding protein (TBP)-associated Taf6 21343 factor -1.37 4.7E-03 Zeb2 24136 zinc finger E-box binding homeobox 2 -1.36 2.4E-03 Preb 50907 prolactin regulatory element binding -1.36 2.0E-03 Arid4b 94246 AT rich interactive domain 4B (Rbp1 like) -1.36 1.4E-03 Elf2 69257 E74-like factor 2 -1.34 1.7E-03 Arid2 77044 AT rich interactive domain 2 (Arid-rfx like) -1.34 2.6E-03 v-myc myelocytomatosis viral related oncogene, neuroblastoma 78 Mycn 18109 derived (avian) -1.32 4.0E-03 Son 20658 Son cell proliferation protein -1.32 9.2E-04 Arnt2 11864 aryl hydrocarbon receptor nuclear translocator 2 -1.32 6.3E-03 Mxi1 17859 Max interacting protein 1 -1.31 6.6E-03 Rela 19697 v-rel reticuloendotheliosis viral oncogene homolog A (avian) -1.31 6.3E-03 Tardbp 230908 TAR DNA binding protein -1.29 8.4E-03 Meis1 17268 myeloid ecotropic viral integration site 1 -1.28 4.5E-03 Pbx2 18515 pre B-cell leukemia transcription factor 2 -1.28 4.0E-03 Daxx 13163 Fas death domain-associated protein -1.28 4.8E-03 Brd4 57261 bromodomain containing 4 -1.28 2.1E-03 Safb2 224902 scaffold attachment factor B2 -1.26 7.3E-03 Brdt 114642 bromodomain, testis-specific -1.25 7.7E-03 Foxj2 60611 forkhead box J2 -1.25 8.1E-03 Rfxank 19727 regulatory factor X-associated ankyrin-containing protein -1.24 8.9E-03 Smad4 17128 MAD homolog 4 (Drosophila) -1.21 6.1E-03 *Values are “absolute” Fold changes, ratios <1 were divided by -1. 3 Organ-specific function of PPARα as revealed by gene expression profiling

Discussion

In this study we identified and compared the PPARα-dependently regulated genes in small intestine and liver, and based on the commonalities and differences we are able to assign its organ specific functions. Furthermore, we aimed at identifying transcriptional mechanisms by which organ-specific PPARα-dependent gene regulation might be achieved.

Our data demonstrate that small intestine and liver respond differentially to PPARα activation, both at level of indivual genes and corresponding cellular processes. With respect to each individual organ, our results are in agreement with the outcomes of independently performed experiments reported by our laboratory, although in these studies less comprehensive microarrays were used [70, 171, 172]. Metabolic processes that were altered in both intestine and liver relate to the best-studied function of PPARα, namely fatty acid metabolism. In addition to intestine and liver, PPARα also regulates metabolism of fatty in fatty acids in cardiomyocytes and skeletal muscle (reviewed in [11]), which thus likely deﬁnes the core-function of PPARα. However, although regulated in both organs, several processes were inversely regulated. For example, our data reveal the well-known induction in liver of certain genes involved in regulation of cell cycle and DNA repair, which upon chronic administration of PPARα agonists causes hepatocarcinoma in rodents [173]. In contrast, we did not observe this regulation in small intestine, which likely explains no such adverse effects have been reported for this organ. In the small intestine cell cycle is rather repressed because of the PPARα-dependent reduced expression of MYC and Caspase-3 [70].

Our secretome analysis identified numerous PPARα-regulated genes that potentially encode secreted proteins. For several of these proteins, such as ANGPTL4, FGF21, FGF15 and IL18, recent studies showed that they indeed function as important intercellular messengers [162-167]. This demonstrates the validity and relevance of our approach, but also highlights the necessity for further research on the biological role of the other, not well-studied secreted proteins, which 79 might reveal other mechansisms by which PPARα is able to signal to other organs. The use of organ-specific secreted PPARα-target proteins may allow the development of tissue- specific PPARα function tests with high therapeutic value. Because of its role in many metabolic and signalling routes these tests would allow to detect small deviations from a normal PPARα response due to an early pathological condition such as inflammation, steatosis or insulin resistance. Such early plasma biomarkers could be crucial in preventive strategies to sensitively detect in a less invasive way organ-dysfunction that are precursors of systemic metabolic diseases such as diabetes or CVDs. Finally, although the precise molecular mechanism(s) responsible for the organ specific responses to PPARα activation by WY14643 activation are still unclear, our data allows to speculate that these will include the differential expression, and therefore recruitment, of nuclear coregulators as well as transcription factors.

In conclusion, we provide here a comprehensive comparative microarray analysis of liver and small intestinal PPARα-dependent transcriptome. This analysis allows a better understanding of the biological importance of this fatty acid sensing nuclear receptor that comprises more broader expressed functions such as regulating lipid handling capacity or more organ- and cell-speciﬁc ones such as modulation of acute phase response (liver) or repression of T-cell speciﬁc functions (small intestine).

A series of future studies are required to investigate the different scientiﬁc issues in more detail and to allow more precisely the characterization of organ-speciﬁc functions with ultimately the whole body system biology of PPARα.

80 3 Organ-speciﬁc function of PPARα as revealed by gene expression proﬁling

81 4 PPARα-mediated effects of dietary lipids on intestinal barrier gene expression

Heleen de Vogel-van den Bosch, Meike Bünger, Philip de Groot, Hanneke Bosch-Vermeulen, Guido Hooiveld, Michael Müller

Published in BMC Genomics, 2008, May 19;9:231. PMID: 18489776 Abstract

Background The selective absorption of nutrients and other food constituents in the small intestine is mediated by a group of transport proteins and metabolic enzymes, often collectively called ‘intestinal barrier proteins’. An important receptor that mediates the effects of dietary lipids on gene expression is the peroxisome proliferator-activated receptor alpha (PPARα), which is abundantly expressed in enterocytes. In this study we examined the effects of acute nutritional activation of PPARα on expression of genes encoding intestinal barrier proteins. To this end we used triacylglycerols composed of identical fatty acids in combination with gene expression proﬁling in wild-type and PPARα-null mice. Treatment with the synthetic PPARα agonist WY14643 served as reference.

Results We identiﬁed 74 barrier genes that were PPARα-dependently regulated 6 hours after activation with WY14643. For eicosapentaenoic acid (EPA), docosahexaenoic acid (DHA) and oleic acid (OA) these numbers were 46, 41, and 19, respectively. The overlap between EPA-, DHA-, and WY14643-regulated genes was considerable, whereas OA treatment showed limited overlap. Functional implications inferred form our data suggested that nutrient-activated PPARα regulated transporters and phase I/II metabolic enzymes were involved in a) fatty acid oxidation, b) cholesterol, glucose, and amino acid transport and metabolism, c) intestinal motility, and d) oxidative stress defense.

Conclusion We identiﬁed intestinal barrier genes that were PPARα-dependently regulated after acute activation by fatty acids.This knowledge provides a better understanding of the impact dietary fat has on the barrier function of the gut, identiﬁes PPARα as an important factor controlling this key function, and underscores the importance of PPARα for nutrient-mediated gene regulation in intestine.

84 4 PPARα-mediated effects of dietary lipids on intestinal barrier gene expression

Background

The small intestine is the primary site for digestion and selective absorption of nutrients and other food constituents. Absorption of these molecules across the intestinal epithelium is mediated mainly by multiple transmembrane transporters that principally belong to two superfamilies, namely the solute carrier (SLC) and the ATP Binding Cassette (ABC) superfamily of transporters. SLC transporters located at the apical membrane of enterocytes are responsible for the selective uptake of macronutrients, such as di- and tripeptides, hexoses and fatty acids [174]. In contrast, ABC transporters are efflux transporters responsible for the active removal of substances, including nutrients such as cholesterol, regulating their intracellular concentrations [175]. Besides their presence in plasma membranes, SLC and ABC transporters are also located in intracellular organelles, such as mitochondria or peroxisomes, thus regulating intracellular and transcellular solute transport. In addition, it has become clear that the small intestine is an important metabolic active organ, to a great extend responsible for the first-pass metabolism of nutrients and xenobiotics [176, 177]. Numerous metabolic reactions occur in enterocytes, including those typically referred to as phase I and phase II metabolism. Phase I metabolism includes oxidative, peroxidative, and reductive metabolism of endogenous compounds and drugs, mediated by cytochrome P450 isoenzymes (CYPP450s) [178]. Phase II metabolism often succeeds phase I metabolism and yields mainly more hydrophilic metabolites, mostly by conjugation, thereby increasing the water solubility of lipophilic compounds. The most important phase II enzymes are sulfotransferases (SULTS) [179, 180], UDP-glucuronosyltransferases (UGTS) [181], glutathione S-transferases (GSTS) [182, 183], N-acetyltransferases (NATS) [184], and epoxide hydrolases (EPHS) [185]. Several ABC transporters are responsible for the excretion of metabolites resulting from phase I and phase II enzymatic transformations [175]. There is increasing interest in the molecular mechanisms underlying the beneficial or adverse effects of foods and food components. Nutrients impact gene expression mainly by activating or suppressing specific transcription factors [5, 11]. The most important group of transcription factors involved in mediating the effect of nutrients and their metabolites on gene transcription 85 is the superfamily of nuclear receptors, which consists of 48 members in the human genome [12]. This superfamily is subdivided into six families [13], of which the NR1 family is most relevant to nutrition. One important group of receptors that mediates the effects of dietary fatty acids and its derivatives on gene expression are the Peroxisome Proliferator Activated Receptors (PPARs, NR1C) [13, 17, 186]. Three PPAR isotypes, α (NR1C1), δ (also called β) (NR1C2), and γ (NR1C3) are distinguished and characterized by different biological roles. Transcriptional regulation by PPARs requires heterodimerization with the retinoid X receptor (RXR; NR2B) [13, 18, 19]. When activated by an agonist, the PPAR/RXR heterodimer stimulates transcription via binding to DNA response elements (PPREs) present in and around the promoter of target genes. Besides upregulating gene expression, PPARs are also able to repress transcription by directly interacting with other transcription factors and interfere with their signaling pathways, a mechanism commonly referred to as transrepression [20]. PPARα has been shown to be expressed at a high level in the small intestine [70]. Moreover, the average Western diet contains a high amount of triacylglycerols [187] that are hydrolyzed to monoacylglycerol and free fatty acids before entering the enterocyte [93]. As a result the small intestine is frequently exposed to high levels of natural PPARα agonists. However, currently little is known about the effects of PPARα activation by dietary fats on gene expression in small intestine. Although in several studies small intestinal gene expression was studied after high-fat feeding [188-190], the specific role of PPARα remains to be elucidated. Here, we take advantage of a unique experimental design using triglycerides composed of identical fatty acids in combination with gene expression profiling to examine the effects of individual dietary fatty acids on intestinal gene expression in mice. By conducting these experiments in wild-type and PPARα-null mice, and by limiting the exposure time to 6 hours, we were able to elucidate the specific, direct contribution of PPARα in regulating the expression of transport and phase I/II metabolism genes in small intestine.

86 4 PPARα-mediated effects of dietary lipids on intestinal barrier gene expression

Methods

Animals and materials Pure bred wild-type (129S1/SvImJ) and Pparα-null (129S4/SvJae) mice [94] were bred and housed as described [191]. All animal studies were approved by the Local Committee for Care and Use of Laboratory Animals. The synthetic triacylglycerols trieicosapentaenoin and tridocosahexaenoin were bought from Nu-Chek-Prep, Inc (Elysian, MN), whereas triolein was from Fluka (Zwijndrecht, the Netherlands). These are synthetic triacylglycerols with three identical acyl moieties, namely eicosapentaenoic acid (EPA), docosahexanoic acid (DHA) and oleic acid (OA), which are released as free fatty acids upon digestion in the small intestinal lumen. All three fatty acids have been reported to bind PPARα with varying afﬁnities in the micromolar range [85- 87, 142]. WY14643 was obtained from Chemsyn (Lenexa, KS).

Experimental design and tissue handling Four months-old male wild-type and PPARα-null mice were used in this study (n=4-5 per group). Two weeks before the start of the experiment all mice were put on a background diet, which was a modified AIN76A diet (Research diet services, Wijk bij Duurstede, The Netherlands). The AIN76A diet contains 5% w/w corn oil (~10 energy%) [192], which is a relatively low amount of fat. In the current study we replaced the corn oil by the same amount of olive oil (predominantly consisting of oleic acid), since Ren et al [193] demonstrated that an olive oil-rich diet did not regulate established PPARα target genes. The modified AIN76A diet was thus assumed to be a ‘poor PPARα-activating’ diet, and therefore we hypothesized the number of genes PPARα- dependently regulated by OA would be nominal. However, since the amount of OA in the diet was lower than the amount dosed by gavage (see below), some genes were expected to be regulated. At the day of the experiment mice were fasted for four hours. At 9 AM mice were dosed by oral gavage with 400 µl of the synthetic triacylglycerols triolein, trieicosapentaenoin, or tridocosahexaenoin, or 400 µl of a 0.1% WY14643 suspension in 0.5% carboxymethyl cellulose (Sigma-Aldrich, Zwijndrecht, the Netherlands). The volume of all doses (400μl) equalled the maximum recommended volume for gastric gavages for mice [194]. For the fatty acids these 87 doses corresponded to approximately 12.5g/kg body weight. To put this amount into perspective, data on food intake (not shown) revealed that the mice consumed approximately 4 gram of the modified AIN76A diet per day, which corresponds to approximately 200mg (6.7g/kg body weight) of fat. The amount of WY14643 the mice received (approximately 130mg/kg body weight) was based on previously published short-term study [195]. Six hours after the gavage the mice were anaesthetized with a mixture of isofluorane (1.5%), nitrous oxide (70%) and oxygen (30%). Small intestines were isolated and flushed with ice-cold phosphate-buffer saline and subsequent tissue handlings were performed on ice. Remaining fat and pancreatic tissue was carefully removed from the intestines. For RNA analyses of total tissue, we used full-length small intestine (microarray analyses), or sections obtained after dividing the small intestine into 10 equal parts (studying gene expression distribution along the proximal- distal axis). All small intestinal samples were snap-frozen in liquid nitrogen and stored at -80 °C until RNA isolation. RNA isolation, Affymetrix GeneChip hybridization and scanning, and qRT-PCR RNA isolation, Affymetrix GeneChip oligoarray hybridization and scanning, and quantitative real-time PCR were performed as described previously [191]. The sequences of primers used in qRT-PCR are available on request. For microarray analyses, RNA was isolated from the full-length small intestine. RNA was hybridized on an Affymetrix GeneChip Mouse Genome 430 2.0 array. This array detects 45,038 transcripts that represent 16,579 known genes. For each experimental group, four or five biological replicates were hybridized for wild-type and PPARα-null mice, and in total 35 arrays were used. Array data have been submitted to the Gene Expression Omnibus, accession number GSE9533.

Analyses of microarray data Microarrays were analyzed as described previously [191]. To compile a list of transport and phase I/II metabolism (barrier) genes represented on the array, annotation information from Affymetrix (release of July 2006) was queried for SLC transporters, ABC transporters, CYPP450s, the phase II metabolism enzymes glutathione S-transferases, sulfotransferases, epoxide hydrolases, aldo- keto reductases, N-acetyltransferases, and glucuronosyl transferases. Also glutathione reductase, glutathione synthetase, and glutathione peroxidases were included in this set. The final set consisted of 944 probesets, encoding for 529 unique genes. To study significantly expressed genes, only probesets with an expression estimate higher than 32 in either of the 8 experimental groups were selected for further analysis. This cut-off value was based qRT-PCR experiments, because regulation of genes with an expression estimate >32 on the array could all be confirmed by qRT-PCR [191]. The filtering was done after normalization and data analysis. Probesets that had a Bayesian comparison p-value <0.01 were considered to be significantly regulated; no cut- off value of the fold change was used. Of these, probesets that were changed in treated wild- type mice compared to treated PPARα-null mice, were designated PPARα-dependently regulated. qRT-PCR data confirming our array analysis is presented in Table 2 and additional data, Table 7. 88 Differences on the number of regulated genes between gene sets were tested for significance by a one-tailed binominal test. Interpretations of functional outcomes focused on groups of genes that are known to be functionally related (i.e. having a similar function or participating in the same pathway). 4 PPARα-mediated effects of dietary lipids on intestinal barrier gene expression

Results and discussion

Effect of acute PPARα activation with OA, EPA, DHA and WY14643 In this study we investigated the role of PPARα on the expression of genes encoding for transport proteins and phase I/II metabolic enzymes, collectively called barrier proteins, which are responsible for the selective absorption and metabolism of food components. Since we are specifically interested in the role PPARα plays in nutrient-mediated gene regulation, we used and compared in our studies 3 natural agonists normally found in the diet [oleic acid (OA, C18:1), eicosapentaenoic acid (EPA, C20:5) and docosahexaenoic acid (DHA, C22:6)]. As reference we used the synthetic agonist WY14643. We analyzed gene expression 6 hours after oral gavage, thus we primarily studied the direct effects of PPARα activation. This time point was chosen because in a pilot experiment we found that after an oral fat load plasma triacylglycerol levels peaked at 2-3 hours post loading and almost returned to basal levels after 6 hours (data not shown), indicating that at that point most of the fat bolus has passed through the enterocytes, and sufficient time remained for transcriptional events to occur. Expression of transport and phase I/II metabolism genes (barrier genes) was studied using microarrays. The Affymetrix GeneChip Mouse Genome 430 2.0 array comprised of 45,038 probesets, representing 16,579 unique genes. Annotation information from Affymetrix was queried to compile a list of transport and phase I/II metabolism genes present on the array (for details, see methods section). This set consisted of 944 probesets, encoding for 529 unique genes, and was used throughout all analyses. We identified 9,426 significantly expressed genes in small intestine, of which 264 were barrier genes. As expected, of all agonists used in this study, acute treatment with WY14643 provoked the most pronounced response, both with respect to the number of regulated genes and the magnitude of the fold changes. Treatment with WY14643 resulted in the PPARα-specific differential expression of 74 transport and phase I/II metabolism genes (Table 1), of which 32 were expressed at higher levels and 42 genes were reduced in wild-type mice compared to PPARα-null mice (for the full list of regulated genes please see additional data, Table 1).

Table 1: Number of PPARα-dependently regulated genes after treatment with different agonists. 89 OA EPA DHA WY14643 All Barrier All Barrier All Barrier All Barrier genes genes genes genes genes genes genes genes Number of PPARα dependent regulated 508 19 874 46 894 41 1218 74 genes

Percentage 5.4 7.2 9.3 17.4 9.5 15.5 12.9 28.0

Number of PPARα-dependently regulated genes and corresponding percentages for all genes and the barrier gene set after activation with OA, EPA, DHA, and WY14643. The percentages relate to the total number of expressed genes (9,426) and all barrier genes (264), respectively. On the other hand, treatment of wild-type and PPARα-null mice with OA identiﬁed only 19 PPARα-dependently regulated barrier genes (Table 1). Of these, 13 were induced and 6 repressed (additional data, Table 2). Treatment with EPA and DHA resulted in 46 and 41 PPARα-dependently regulated barrier genes, respectively (Table 1). Activation of PPARα by EPA increased the expression of 32 genes and suppressed 14 genes (additional data, Table 3), whereas for DHA these numbers were 22 and 19, respectively (additional data, Table 4).

Table 2: Confirmation of microarray results. WY14643 EPA DHA OA FC FC FC FC FC FC FC FC Gene symbol (MA) (qPCR) (MA) (qPCR) (MA) (qPCR) (MA) (qPCR) Fatp4 (Slc27a4) 1.7 (0.24) 2.2 (0.28)* 1.5 (0.14) 1.7 (0.30)* 1.4 (0.14) 1.5 (0.09)* 1.4 (0.16) 1.6 (0.41)* Abcd3 2.8 (0.38) 4.9 (0.75)* 1.7 (0.11) 2.0 (0.28)* 1.8 (0.08) 2.7 (0.50)* 1.4 (0.10) 2.0 (0.22)* Cyp2c65 2.6 (0.34) 3.7 (0.69)* 2.3 (0.24) 2.8 (0.51)* 2.5 (0.48) 3.9 (1.13)* 1.7 (0.23) 2.7 (0.94)* Cyp4F16 1.9 (0.36) 2.0 (0.44)* 1.5 (0.14) 1.3 (0.30)* 1.4 (0.17) 1.5 (0.25)* 1.9 (0.34) 2.5 (0.36)* Microarray results were confirmed with qRT-PCR. FC = Fold change, MA = microarray, qPCR (qRT-PCR) = quantitative real time PCR. For the qRT-PCR analysis: mRNA levels were standardized to cyclophilin; expression in the PPARα-null mice was arbitrarily set to 1. Significance was determined by a Bayesian t-test (array data) or student’s t-test (qRT-PCR data), *= p-value <0.05. Data are means ±standard error (n=4-5).

In Bünger et al [70] we reported that under basal (control) conditions only 21 genes were differentially expressed in small intestine of wild-type mice compared to PPARα-null mice. The currently investigated barrier gene set includes 2 of these genes, Slc25a20 and Cyp4a10, which were both expressed at lower levels in the null mice. We found that expression of Slc25a20 was only slightly elevated after acute treatment with WY14643, EPA, and DHA, which indicates that the regulation of Slc25a20 by PPARα may be of less relevance during (nutritional) activation of PPARα. In contrast, the fold induction of Cyp4a10 observed after acute treatment with WY14643, EPA and DHA was much larger than the basal difference, which implies that for this 90 gene activation of PPARα is of importance. When comparing the list of barrier genes that were PPARα-dependently regulated after acute treatment with WY14643 with that of a long-term (5 day) exposure experiment [70], we found an overlap of 74% (additional data, Table 5). This indicates that short-term regulation evoked with synthetic agonists is maintained for at least 5 days. Hirai et al [196] very recently reported their study in which they identiﬁed seven nutrient and drug transporters that were PPARα-dependently regulated in small intestine after 3 days exposure to two synthetic agonists. In concordance with their data, we found that all except 2 of these transporters were regulated after acute treatment with WY14643 as well. In addition, we observed that these carriers were also PPARα-dependently regulated by DHA and EPA. Like Hirai et al, we did not observe a PPARα-dependent regulation of Pept1 (Slc15a1), the ﬁrst intestinal nutrient transporter shown to be PPARα-dependently regulated during fasting [196, 197]. 4 PPARα-mediated effects of dietary lipids on intestinal barrier gene expression -oxidation [82] -oxidation [85] β β -oxidation [54] β -oxidation [68] β [96]

[93]

-oxidation [86] -oxidation [86, 87] -oxidation [88, 89] Function Carnitine cycle in mitochondrial Apical carnitine uptake for mitochondrial ω ω ω Glycogenolysis [90] Dicarboxylates uptake for Krebs cycle [91] Apical cholesterol uptake [92] Apical glucose uptake Glucose transport [94] Glucose + mannose transport [95] Apical glutamate uptake Apical neutral amino acids uptake [97] Basolateral cationic amino acids [98] Malate-aspartate shuttle: provides cytosolic aspartate [68] Serotonine uptake [72] Dopamine uptake [74] Apical sulphate uptake [100] Has acyl CoA properties to activate FA for subsequent peroxisomal properties to activate FA Has acyl CoA Basolateral cholesterol efflux [61, 63] Basolateral glucose efflux [94] Basolateral neutral amino acids efflux [98] Basolateral aromatic amino acids efflux [99] Apical long chain FA uptake [81] Apical long chain FA over the peroxisomal membrane for of VLFA Transport nc nc nc nc nc nc nc nc nc nc nc nc nc nc nc nc nc nc nc nc 1.4 1.4 1.9 2.6 -2.3 FC OA nc nc nc nc nc nc nc nc nc nc nc nc nc 1.4 2.4 4.2 1.8 1.4 -1.6 -2.0 -1.8 -2.0 -1.8 -1.3 FC 160.3 DHA nc nc nc nc nc nc nc nc 1.5 2.6 3.2 1.6 1.7 1.5 3.2 1.8 -2.2 -1.4 -1.7 -1.6 -1.6 -1.4 -1.2 -2.0 FC 120.2 EPA nc nc 1.7 6.8 7.7 2.2 2.8 1.7 1.9 1.5 -1.5 -2.1 -1.5 -1.4 -1.4 -1.5 -1.7 -1.4 -1.6 -1.7 -1.3 -1.4 -2.1 FC 12.1 WY

1447.2 91 Probeset ID 1424441_at 1423109_s_at 1421848_at 1416316_at 1416679_at 1424853_s_at 1416194_at 1417277_at 1417042_at 1418857_at 1438514_at 1450392_at 1455431_at 1449067_at 1426599_a_at 1439494_at 1448299_at 1428793_at 1417929_at 1436368_at 1447181_s_at 1428440_at 1417150_at 1417415_at 1430804_at LAT1 (SLC7A7) LAT1 + Gene symbol Fatty acid oxidation (SLC27A4) FATP4 OCTN2 (SLC22A5) ABCD3 CYP4A10 CYP4B1 CYP4F16 G6PT1 (SLC37A4) NADC1 (SLC13A2) Cholesterol ﬂux NPC1L1 ABCA1 Glucose transcport (SLC5A1) SGLT1 GLUT2 (SLC2A2) GLUT1 (SLC2A1) Amino acid metabolism EAAC1 (SLC1A1) Y ARALAR1 (SLC25A12) Intestinal Motility (SLC6A4) SERT NAS1 (SLC13A1) FATP2 (SLC27A2) FATP2 PAT1 (SLC36A1) PAT1 SGLT4 (SLC5A9) SGLT4 (SLC7A8) LAT2 (SLC6A3) DAT1 TAT1 (SLC16A10) TAT1 CACT (SLC25A20) CACT Table 3: Overview of regulated processes Table Function Pyruvate metabolism [68] ? ? Apical vitamin C uptake [101] Basolateral vitamin C uptake [101] Phase I metabolism Phase I metabolism Phase I metabolism ? Aldo-ketoreductase activity (Phase II) Aldo-ketoreductase activity (Phase II) Epoxide hydrolase activity (Phase II) Epoxide hydrolase activity (Phase II) Glutathione transferase activity (phase II) Glutathione transferase activity (phase II) Glutathione transferase activity (phase II) Glutathione transferase activity (phase II) Glutathione transferase activity (phase II) Glutathione transferase activity (phase II) Glutathione transferase activity (phase II) Glutathione transferase activity (phase II) Glutathione transferase activity (phase II) Glutathione transferase activity (phase II) Glutathione transferase activity (phase II) Apical heme secretion [102] nc nc nc nc nc nc nc nc nc nc nc nc nc nc nc nc nc nc nc nc nc nc nc nc 1.7 FC OA nc nc nc nc nc nc nc nc nc nc nc nc nc 1.8 2.5 5.1 1.9 1.3 1.8 1.5 1.5 2.6 1.8 1.2 1.3 FC DHA nc nc nc nc nc nc nc nc nc 1.5 1.5 1.7 3.2 2.3 1.5 3.2 1.6 1.4 1.2 1.3 1.7 2.4 1.8 1.3 1.3 FC EPA nc nc nc nc nc nc nc nc 1.6 5.5 1.4 2.6 1.7 1.7 1.5 1.3 2.0 1.2 1.4 1.9 -1.5 -1.7 -1.3 FC 10.7 13.4 92 WY Probeset ID 1416954_at 1420836_at 1453056_at 1421912_at 1445589_at 1417651_at 1429994_s_at 1419039_at 1448894_at 1422000_at 1418672_at 1422438_at 1448499_a_at 1421041_s_at 1423436_at 1416368_at 1452823_at 1448330_at 1427473_at 1424835_at 1416842_at 1422072_a_at 1417883_at 1415897_a_at 1422906_at Gene symbol Oxidative Stress DIC (SLC25A10) KMCP1 (SLC25A30) MCT13 (SLC16A13) SVCT1 (SLC23A1) SVCT2 (SLC23A2) CYP2C29 CYP2C65 CYP2D22 AKR1B8 AKR1C12 AKR1C13 EPHX1 EPHX2 GSTK1 GSTM1 GSTM3 GSTM4 GSTM5 GSTM6 GSTT2 MGST1 ABCG2 GSTA1 /// GSTA2 GSTA1 GSTA3 GSTA4 Table 3 (continued): Overview of regulated processes Table and results the in described are processes All intestine. small the in metabolism I/II phase and transport to related processes regulated dependently PPARα of Overview discussion section. FC= fold change, nc= not changed. 4 PPARα-mediated effects of dietary lipids on intestinal barrier gene expression

To characterize the importance of PPARα in controlling expression of small intestinal transporters and phase I/II metabolic enzymes, we compared the fraction of PPARα dependently regulated genes of the barrier gene set with that observed for all genes (Table 1). For treatment with OA the percentages were 7.2% and 5.4%, respectively, for the set of barrier genes and all expressed genes. This difference was not statistically signiﬁcant (p=0.11). However, the other two natural agonists showed a signiﬁcantly higher percentage of regulated genes for the barrier gene set than for all genes; for EPA this percentage was 9.3% for all genes, whereas 17.4% of the barrier genes were regulated (p<0.001). For DHA these percentages were 9.5% and 15.5% respectively (p<0.001). For WY14643 we observed an even larger difference between all genes and the barrier gene set; 12.9% respectively 28.0% of the genes were regulated in a PPARα-dependent manner (p<0.001). These results imply that PPARα plays an important role in regulating small intestinal gene expression of transporter and phase I/II metabolic enzymes.

Overlap between OA, EPA, DHA and WY14643 treatment We also determined the overlap of PPARα-dependently regulated genes between the different treatments. Most of the genes regulated upon treatment with OA were not regulated by DHA and EPA (Figure 1A) or WY14643 (data not shown). Only four genes, i.e. Slc27a4 (Fatp4), Cyp4F16, Cyp2c65, and Abcd3 were regulated upon all 4 treatments. For these genes additional qRT-PCR analyses were performed, which conﬁrmed the array results (Table 2). There was considerable overlap between the genes affected by EPA, DHA or WY14643 treatment (Figure 1B).

Figure 1. Overlap of A B PPARα-dependently regulated genes between the four agonists. The numbers in the Venn plots represent the numbers of PPARα-dependently regulated genes for each treatment. A) Overlap between OA, EPA and DHA, 93 B) Overlap between EPA, DHA, and WY14643.

These overlapping genes behaved the same in all treatments, i.e. they were either increased or suppressed in wild-type compared to PPARα-null mice upon all treatments. In Table 6 of the additional data the complete list of overlapping genes is presented. It is likely that OA treatment affected fewer genes, because the mice may be adapted to this fatty acid since they were fed a diet based on olive oil three weeks before gavage (for details, see methods section). In addition, it is generally accepted that polyunsaturated fatty acids activate PPARα better than monounsaturated fatty acids [85-87, 142], which is in line with our result that OA activated less genes PPARα- dependently than EPA and DHA. Although the overlap between WY14643-, EPA-, and DHA-regulated genes was high, we still observed differential gene activation between these treatments. The exact mechanism(s) underlying these differences are currently unclear, but we speculate this may be partially due to the differential recruitment of coactivators such as Src-1, Med1, Pgc1α, and p300 by the three agonists [14, 198-200]. Alternatively, hitherto unknown additional signaling routes not shared by the three agonists may exist.

Functional implications of acute PPARα activation in small intestine A summary of functional outcomes of PPARα activation by the agonists inferred from our data is presented in Table 3. Although in this study we only determined mRNA levels, it has been reported that for the majority of genes the mRNA levels reﬂect protein abundance very well [149, 201]. We therefore allow ourselves to speculate about the functional consequences of nutritional PPARα activation. Nevertheless, these implications should ultimately be evaluated in follow-up studies.

Role of PPARα in intestinal fatty acid oxidation It is well established that PPARα serves as a master regulator of fatty acid catabolism, which is also apparent from our data [53, 70]. Various transporters and phase I enzymes involved in fatty acid uptake and oxidation were PPARα-dependently regulated (Table 3). Although the extent varied somewhat, all 4 agonists regulated long chain fatty acid uptake, mitochondrial and peroxisomal β-oxidation, ω-oxidation, and the metabolism of energy-yielding substrates (glycogenolysis and Krebs cycle). For most genes this regulation was agonist-independent and is consistent with earlier findings [59, 202, 203]. It is known that enhanced fatty acid β-oxidation is correlated with reduced severity of inflammatory bowel disease [204]. Furthermore it has been shown that WY14643 treatment caused a reduction of colon injury in a murine DNBS experimental colitis model [205] and that WY14643 treatment might have an anti-inflammatory effect in the small intestine [70]. 94 It has been reported that the expression of Octn2 (Slc22a5), involved in apical carnitine uptake, is induced by WY14643 and clofibrate [206-208]. Here we showed that also EPA and DHA induced expression of Octn2. Recently it has been reported that two functionally relevant polymorphisms in the Octn2 (Slc22a5) gene are associated with increased risk for inflammatory bowel disease [209, 210], and that Octn2 expression is decreased in rats with induced inflammatory bowel disease [211]. Taken together, our data imply that nutritional activation of PPARα might be therapeutically valuable for patients with inflammatory bowel disease. 4 PPARα-mediated effects of dietary lipids on intestinal barrier gene expression

PPARα regulates intestinal cholesterol flux Expression of the apical cholesterol uptake transporter Npc1l1 was PPARα-dependently suppressed after treatment with WY14643, EPA, and DHA (Table 3), as also has been observed after fenofibrate treatment [212] and PPARδ activation [213]. It is known that treatment with WY14643 for 5 days induced expression of Abca1 [214]. Here we show that Abca1 is also acutely regulated after PPARα activation. Abca1, which promotes cholesterol efflux at the basolateral membrane to Apo-AI for HDL formation [215-218] was increased after treatment with WY14643 and EPA. Functionally these results suggest that less cholesterol is absorbed from the lumen and more cholesterol is transferred to Apo-A1, resulting in reduced intracellular cholesterol levels in enterocytes. Enterocytes likely compensate for this by increasing the activity of HMG-CoA reductase, as has been reported before [2, 3, 219].

PPARα regulates intestinal nutrient transport and metabolism Expression of the apical glucose uptake transporter Sglt1 (Slc5a1) and the basolateral glucose efflux transporter Glut2 (Slc2a2) was PPARα-dependently repressed after WY14643 treatment. Furthermore, WY14643, EPA, and DHA all reduced expression of the apical mannose and glucose uptake transporter Sglt4 (Slc5a9), suggesting that PPARα activation results in reduced glucose transport through the intestinal wall. In addition, several transporters involved in the amino acid metabolism were PPARα-dependently regulated (Table 3). Gene expression of small intestinal apical uptake as well as basolateral efflux amino acid transporters was PPARα-dependently suppressed. Furthermore, activation of PPARα reduced expression of Slc25a12 (Aralar1), which is involved in the malate-aspartate shuttle [220]. These effects are in line with data that showed PPARα-mediated downregulation of genes involved in hepatic amino acid metabolism [53, 56]. For liver it is suggested that amino acids are conserved for local synthetic processes, including protein and purine synthesis during for instance proliferation [221]. In the small intestine villus length is increased after WY14643 treatment [70], which implies that also in the small intestine amino acids are conserved for local anabolic processes. Taken together, our results suggest that PPARα activation leads to a diminished (neutral) amino acid flux through the enterocyte. 95 PPARα regulates intestinal motility Expression of the serotonin transporter Slc6a4 (Sert) was decreased after treatment with WY14643, EPA, and DHA (Table 3). Serotonin is a neurotransmitter secreted by enterochromaffin cells and is considered to play a key role in functioning of the gut, initiating peristaltic reflex pathways and facilitating propulsive activity [222]. Inactivation of serotonin is crucial to limit its activity, and this is mediated by Sert [223]. The observed reduced expression will result in a diminished activity of Sert, which in turn may increase intestinal motility [223]. Serotonin is detoxified by sulfation inside the enterocyte [224]. The apical sulfate import seems to be reduced as gene expression of the uptake transporter Slc13a1 (Nas1) was decreased. This might be a response to the decreased uptake of serotonin. We also showed that the dopamine transporter Dat1 was PPARα-dependently upregulated after EPA and OA treatment (Table 4). Dopamine increases contractile force of intestinal motility [225], thus more dopamine likely results in increased intestinal motility. Altogether, we believe it is likely that PPARα is involved in regulating intestinal motility. Our data suggest that in feeding conditions PPARα activation may result in speeding-up intestinal motility.

PPARα diminishes effects of oxidative stress Oxidative stress results from an imbalance between formation and degradation of pro-oxidants or decreased cellular antioxidant protection mechanisms and may result in increased cell damage and apoptosis [226]. Many genes included in our barrier gene set, such as CYPP450s, Gsts, and several Slc transporters, are involved in oxidative stress and were PPARα-dependently regulated. CYPP450s induce oxidative stress by oxidative, peroxidative, and reductive metabolism of endogenous compounds and drugs [178], whereas GSTS are involved in the defense against oxidative stress by catalyzing the conjugation of glutathione to a wide variety of endogenous and exogenous electrophilic compounds [227]. Various CYPP450 genes were PPARα-dependently upregulated (Table 4), which is in line with data obtained from liver [53]. However, since not 96 all CYPP450 genes are expressed in both organs, the regulated genes were not identical. Many GSTS were upregulated by activation with WY14643, EPA, and DHA (Table 3). In addition, various SLC transporters involved in oxidative stress defense were PPARα-dependently upregulated; DIC (SLC25A10), involved in the pyruvate-malate shuttle, citrate-pyruvate shuttle, and gluconeogenesis from pyruvate, is known to protect against oxidative stress [220]; SVCT2 (SLC23A2), a basolaterally-located uptake transporter for ascorbic acid [228]; MCT13 (SLC16A13), proposed to play an important role in communicating information on the redox state between cells [229]; ABCG2 (BCRP1), a secretion transporter of heme and porphyrins located in the apical membrane [229, 230]; and KMCP1 (SLC25A30), probably involved in protection from oxidative damage in situations of increased mitochondrial metabolism [231] 4 PPARα-mediated effects of dietary lipids on intestinal barrier gene expression

Taken together, we show that many barrier genes involved in defense against oxidative stress were PPARα-dependently upregulated. These data point towards an important role of PPARα in the defense against oxidative stress. In general oxidative stress results in increased cell damage and apoptosis [226] and our data might explain one of the mechanisms by which WY14643 suppresses many genes involved in apoptosis in the small intestine [70].

Longitudinal distribution of the transcriptional regulation during PPARα activation is not the same for PPARα-dependently regulated genes Finally we investigated the expression along the proximal-distal axis of PPARα and 4 genes that were PPARα-dependently regulated by all agonists (Figures 2 and 3). Expression was measured with qRT-PCR. For this analysis, the small intestine was divided in 10 equal parts; part 1 represents the proximal side, whereas part 10 represents the most distal end. Expression of PPARα was maximal in duodenum and jejunum, and then gradually declined in ileum, both under basal conditions and after acute activation with WY14643 (Figure 2). For all treatments only the expression pattern in treated wild-type and PPARα-null mice of the 4 genes is reported (Figure 3). FATP4, ABCD3, CYP2C65, and CYP4F16 are all involved in fatty acid metabolism; FATP4 mediates the apical uptake of long chain fatty acids [107], whereas ABCD3 is involved in the peroxisomal β-oxidation of long chain fatty acids [232]. The human homolog of CYP2C65 (CYP2C8) metabolizes arachidonic acid and generates epoxygenase products [233]. The rat homolog of CYP4F16 (CYP4F5) is involved in ω-oxidation of prostaglandins [234].

Figure 2. Expression of PPARα along the longitudinal axis of control and WY14643-treated wild-type mice. qRT-PCR was used to determine the relative expression levels of PPARα in sections isolated along the proximal-distal axis of the small intestine of wild-type mice that received the control diet (white, open bars), or were acutely treated (6hr) with WY14643 (black, closed 97 bars) (n=4 per group). Small intestines were divided into 10 equal parts; part 1 refers to the most proximal part (duodenum), part 10 refers to the most distal (terminal ileum). Messenger RNA levels were standardized to cyclophilin; part 1 of the non-treated mice was arbitrarily set to 1. Signiﬁcance of control versus treated wild-type mice was determined per segment using an unpaired student’s t-test. * p-value < 0.05. Data are presented as mean ± standard error.

In EPA-, DHA- and WY14643-treated mice we observed a similar expression pattern of Fatp4 (Figure 3A), which closely resembled that of PPARα under control and WY14643-activated conditions. In contrast, OA-treated wild-type mice did not show this pattern. In all treatments Figure 3: Expression of PPARα-dependently regulated genes along the longitudinal axis of treated PPARα-null and wild-type mice. qRT- PCR was used to determine relative expression levels of PPARα-dependently regulated genes in sections isolated along the proximal-distal axis of the small intestine from PPARα-null mice (white, open bars) and wild-type mice (black, closed bars) that were acutely treated (6hr) with the 4 agonists (n=4 per group). The small intestine was divided into 10 equal parts; part 1 refers to the most proximal part (duodenum), part 10 refers to the most distal (terminal ileum). Messenger RNA levels were standardized to cyclophilin; part 1 of the PPARα-null mice was arbitrarily set to 1. White bars represent the PPARα-null mice, black bars represent the wild- type mice. Significance of treated WT versus treated KO mice was determined per segment using student’s t-test. * p-value < 0.05. 98 no significant PPARα-dependent induction of FATP4 was observed in the distal part of the small intestine. ABCD3 was uniformly induced by all treatments (Figure 3B). Activation with WY14643 revealed a robust, equal induction in all segments in wild-type compared to PPARα-null mice, whereas these were less for the three natural agonists. CYP2C65 was predominantly expressed in the proximal part of the small intestine, and showed high similarity between agonists (Figure 3C). For each agonist we observed an induction of its expression which was equal along the complete longitudinal axis. CYP4F16 was uniformly expressed along the proximal-distal axis in treated PPARα-null mice (Figure 3D). However, treatment of wild-type mice with WY14643 and OA shifted the expression of CYP4F16 to more distal regions, whereas EPA and DHA treatment resulted in significant increased expression in more proximal segments. 4 PPARα-mediated effects of dietary lipids on intestinal barrier gene expression

Figure 3 (continued legend): Expression of PPARα-dependently regulated genes along the longitudinal axis of treated PPARα-null and wild-type mice. Data are presented as mean ± standard error. A) fatty acid transport protein 4 (FATP4). B) ATP-binding cassette, sub-family D, member 3 (ABCD3; ALD). C) cytochrome P450, family 2, subfamily c, polypeptide 65 (CYP2C65). D) cytochrome P450, family 4, subfamily f, polypeptide 16 (CYP4F16).

99 Taken together, the data in Figure 3 show that in general all agonists provoke a similar effect on expression of 4 PPARα-dependently regulated genes, and this induction also occurs in more distally-located cells. The latter demonstrates that despite its relatively low expression, PPARα is still able to regulate gene expression. Conclusion In the current study we have identiﬁed intestinal barrier genes that were PPARα-dependently regulated after acute activation by fatty acids. The functional outcomes inferred from our data suggest that nutritional-activated PPARα controls processes ranging from fatty acid oxidation and cholesterol-, glucose-, and amino acid-transport and metabolism to intestinal motility and oxidative stress. Altogether, we showed that PPARα has a great impact in controlling the barrier function of the gut, and this underscores the importance of PPARα for nutrient-mediated gene regulation in intestine.

100 4 PPARα-mediated effects of dietary lipids on intestinal barrier gene expression

101 5 PPARα regulates intestinal lipid absorption

Meike Bünger, Michael Müller, Guido Hooiveld

Submitted Abstract

Background The small intestine is the primary site for digestion and selective absorption of nutrients and other food constituents and is frequently exposed to high levels of PPARα agonists via the diet. Fatty acids affect gene expression by activating the peroxisome proliferator-activated receptor alpha (PPARα). We have previously shown that intestinal PPARα controls a variety of processes, including dietary lipid metabolism. Using a combination of gene expression and functional studies in wild-type and PPARα-null mice, we determined whether PPARα could regulate intestinal lipid absorption.

Results Genes involved in intestinal dietary lipid metabolism were specifically induced upon PPARα activation. Lipid absorption studies showed that wild-type mice absorbed more lipids than mice lacking PPARα, which was stimulated by the specific agonist WY14643 in wild-type, but not null mice. These findings were confirmed by intestinal lipid staining and by a nutritional intervention study. Inhibition of fatty acid oxidation in PPARα-activated wild-type mice resulted in a significantly higher accumulation of labeled lipids in both intestinal wall and circulation at all time points.

Conclusion Our results demonstrate that lipid absorption and chylomicron secretion is regulated by PPARα, and also suggest that oxidation of fatty acids is a quantitative important mechanism to control intracellular levels of fatty acids.

104 5 PPARα regulates intestinal lipid absorption

Introduction

An energy rich diet characterized by high intakes of dietary fat has been linked to the dramatic increase in obesity in Western societies in the last several decades. Obese individuals are at increased risk of developing the metabolic syndrome, which is defined by a clustering of major metabolic risk factors for cardiovascular diseases and diabetes [187, 253, 254]. Essentially, all of the dietary fat that humans ingest is in the form of triacylglycerol (TG); three fatty acids esterified to glycerol. The absorption of lipids from the lumen is generally highly efficient, and only about 4% of the ingested fat escapes into feces [93, 255]. Prior to absorption, TGs are hydrolyzed by gastric and pancreatic lipases to fatty acids (FA) and 2-monoacylglycerols (2MAGs) [93, 255], both of which are taken up by enterocytes. Since both products are potentially toxic, especially free FAs (FFAs) in high concentrations, they must be rapidly neutralized. It is generally thought that this may be accomplished through multiple mechanisms. The incoming FFAs are activated and may be sequestered by binding to e.g. fatty acid binding proteins (FABPs) [256], which are also involved in intracellular trafficking of FAs [256, 257]. Next to this, absorbed FFAs and 2MAG are rapidly resynthezised into TG, which as such is (temporarily) present in intracellular lipid droplets and is ultimately transported out of the cell as TG-rich lipoproptein particles, the chylomicrons [258]. Since the intestine expresses various enzymes involved in mitochondrial, peroxisomal and microsomal oxidation, FFAs may also be catabolized [189, 259]. The amount of TGs absorbed by the intestine is dependent on various physiological and nutritional factors, including the amount of and composition of TGs and phosphatidylcholine (PC) in the intestinal lumen [258, 260]. However, the molecular mechanisms underlaying these phenomena are still unclear. Nowadays fatty acids are no longer viewed as merely a source of energy. Through alterations in gene expression patterns they can also act as fine modulators of metabolism [186, 261, 262]. While numerous transcription factors have been shown to be activated by fatty acids in vitro [186, 261, 263, 264], recent data suggest that PPARα is dominant in mediating the effects of dietary fatty acids on hepatic gene transcription [143]. PPARα (NR1C1) is a member of the 105 superfamily of nuclear receptors and is closely related to the other PPAR isoforms β/δ (NR1C2) and γ (NR1C3), that are characterized by different biological roles [13, 17]. Transcriptional regulation by PPARs requires heterodimerization with the retinoid X receptor (RXR; NR2B) [13, 18, #962]. When activated by an agonist, the PPAR/RXR heterodimer stimulates transcription via binding to DNA response elements (PPREs) present in and around the promoter of target genes. In addition to being activated by fatty acids and their derivatives [85-87, 142], PPARs also serve as the molecular target for certain drugs used in the treatment of dyslipidemia and type 2 diabetes [63]. We previously demonstrated that PPARα is highly expressed in villus cells of the duodenum and jejunum [70], coinciding with the main anatomical location where fatty acids are digested, absorbed and transported into the body as chylomicrons. Moreover, our genome-wide analysis suggested that intestinal PPARα may control, among others, these processes [70]. In this study we therefore set out to determine whether PPARα could regulate intestinal lipid absorption. Using a combination of gene expression and functional studies in wild-type and PPARα-null mice, we show that PPARα-mediated induction of genes involved in intestinal lipid handling results in reduced intracellular lipid levels and increased formation of chylomicrons. Moreover, a long-term nutritional intervention study demonstrated when fed a high fat diet mice lacking PPARα excreted more fat in feces and gained less weight than wild-type mice. We conclude that PPARα regulates lipid absorption and tightly controls the intracellular levels of these potential toxic compounds.

106 5 PPARα regulates intestinal lipid absorption

Materials and Methods

Experimental design and tissue handling In all studies 4- to 5-month old male wild-type and PPARα-null mice were used.

Activation of small intestinal PPARα Mice were fed chow or chow supplemented with WY14643 (0.1% wt/wt; Chemsyn, Lenexa, KS) or fenofibrate (0.1% wt/wt; Sigma, St. Louis, MO) for 5 days (n = 5 mice per group, N=30). On the sixth day, mice were anaesthetized with a mixture of isoflurane (1.5%), nitrous oxide (70%), and oxygen (30%). Small intestines were excised and flushed with ice-cold PBS, and all subsequent tissue handling was performed on ice. Remaining fat and pancreatic tissue was carefully removed, and RNA was isolated from total tissue. Microarray hybridizations were performed on the full-length small intestine, whereas gene expression along the proximal-distal axis was studied in sections obtained after dividing the small intestine into 10 equal parts.

Long term low-fat (LFD) and high fat feeding (HFD) of wild-type and PPARα-null mice To investigate the effect of low- and high-fat diets on lipid absorption in mice, we used puriﬁed diets based on ‘Research Diets’ formulas D12450B/D12451, with adaptations regarding type of fat (palm oil in stead of lard) and carbohydrates to mimic the composition of the average human diet in Western societies. Thus, the high fat diet (HFD) mimics the ratio of saturated to monounsaturated to polyunsaturated fatty acids (40:40:20) in a human diet. The LFD and HFD provide 10% and 45% energy percent, respectively, in the form of TG. The complete composition 107 of the diets is given in supplemental_table_1. It should be noted that in these diets the energy density of all nutrients, except fat and starch, is equal. Mice were weighted every week to monitor weight gain, and received the LFD or HFD for in total 21 weeks. Food intake of all mice was measured during the whole experiment. After 17 weeks, 48h feces collection was performed for determination of lipid content.

Radiolabelled lipid absorption studies Two independent lipid absorption studies were performed: Food was removed five hours prior to the start of each of the experiments. This relatively short period was chosen since the PPARα- null display severe metabolic disturbances after a prolonged (over-night) fast [58, 114, 201]. Prior to the intragastric lipid load an intravenous injection of Triton WR1339 (500 mg/kg BW in saline) was given to block lipoprotein lipase-mediated lipolysis [265-267]. In the first study mice were fed chow, or chow supplemented with 0.1% WY14643 (Chemsyn, Lenexa, KA) for five days (n=12 mice per group, N=48). On the sixth day mice received an intragastric load consisting of 7μCi 3H-labeled triolein (glycerol tri [9,10(n)-3H] oleate, 21.0 Ci/mmol, Amersham, Buckinghamshire, UK) and 2μCi 14C-labeled palmitic acid in 400 mL olive oil. Two, three, four and five hours after the intragastric lipid load blood was collected by tail bleeding for the determination of 3H- and 14C-labelled lipids. At the five hour time point for the small intestine was removed, cut into three equal parts of approximately 11cm (denotated S1 [proximal], S2 [mid] and S3 [distal], respectively), washed in 15ml taurocholate (0.5 mmol/L) in phosphate- buffered saline (PBS) to remove excess luminal oil, and solubilized by the addition of solvable (PerkinElmer, Groningen, the Netherlands). Amounts of radiolabelled lipids were determined by liquid scintillation counting (Packard Instruments, Dowers Grove, IL), and were expressed as percentage of the original dose. The second lipid-loading study was performed as described above in wild-type mice only, except that 2h prior to the experiment mice received an oral dose of either PBS or etomoxir (25mg/kg BW in PBS) to block the activity of mitochondrial carnitine palmitoyltransferase 1 (CPT1; the rate-limiting enzyme for the transport of long-chain fatty acids into mitochondria) and carnitine octanoyltransferase (COT; faciliting the transport of medium- chain fatty acids through the peroxisomal membrane) to inhibit the oxidative breakdown of fatty acids [268].

Histology For histology analyses, a third study was performed in which mice were fed chow, or chow supplemented with WY14643 (0.1% w.w) for ﬁve days (n=10 mice per group, N=40). Five hours prior to an intragastric lipid load of 400µl olive oil chow was removed, and four hours post lipid load intestines were removed and divided into four equal parts. Each section was prepared using a “Swiss roll” technique [102] to evaluate the entire longitudinal section on one slide. Tissues were frozen in liquid nitrogen, sectioned at 5 µm, and stained with hematoxylin and eosin (HE), 108 and Oil-red O [269]. Sections were examined on a CKX41 microscope (Olympus, Zoeterwoude, the Netherlands).

RNA isolation and quality control The RNA isolation and quality control procedures have been earlier described [70].

Affymetrix GeneChip oligoarray hybridization, scanning and quality control Total RNA (5 μg) was labelled using the Affymetrix One-cycle Target Labeling Assay kit (Affymetrix, Santa Clara, CA). The correspondingly labelled RNA samples were hybridized on Affymetrix Mouse Genome 430 2.0 plus arrays, washed, stained and scanned on an Affymetrix GeneChip 3000 7G scanner. Detailed protocols for the handling of the arrays can be found in the Genechip Expression Analysis Technical Manual, section 2, chapter 2 (Affymetrix; P/N 701028, revision 5), and are also available upon request. 5 PPARα regulates intestinal lipid absorption

Packages from the Bioconductor project [96], integrated in an in-house developed on-line management and analysis database for multiplatform microarray experiments (Gavai et al, submitted), were used for analysing the scanned Affymetrix arrays. Various advanced quality metrics, diagnostic plots, pseudo-images and classiﬁcation methods were applied to ascertain only excellent quality arrays were used in the statistical analyses [149]. An extensive description of the applied criteria is available upon request. Array data have been submitted to the Gene Expression Omnibus [48], accession number GSExxx.

Statistical analyses of microarray data Probesets were redefined according to Dai et al. [150], because the genome information utilized by Affymetrix at the time of designing the arrays is not current anymore, which may result in an unreliable reconstruction of expression levels. In this study probes were reorganized based on the Entrez Gene database, build 36, version 2 (remapped CDF v9). Expression estimates were obtained by GC-robust multi-array (GCRMA) analysis, employing the empirical Bayes approach for background adjustment, followed by quantile normalization and median polish summarization [97]. Differentially expressed probesets were identified using linear models, applying moderated t-statistics that implement empirical Bayes regularisation of standard errors [151] P-values were corrected for multiple testing using a false discovery rate method [152]. Probesets that satisfied the criteria of FDR < 1% (q-value < 0.01) were considered to be significantly regulated.

109 Results

Activation of PPARα by two different agonists result in similar effects on the expression of genes involved in intestinal dietary lipid metabolism. To ascertain that genes involved in intestinal dietary lipid metabolism are PPARα dependently regulated irrespective of the agonist used, mice were treated with WY14643 and fenofibrate at equal concentrations for five days followed by analyses of changes in gene expression on Affymetrix Mouse Genome 430 2.0 plus arrays. An overview of selected genes, their function and cellular localization is presented in Figure 1. Outcomes of PPARα activation by the two agonists on expression of these genes in wild-type and PPARα-null mice are given in Table 1. It is evident that upon PPARα activation key genes involved in all steps of intestinal lipid absorption, thus ranging from intestinal fatty acid hydrolysis, uptake, intracellular binding, activation and resynthesis into triacylgycerols and packaging into chylomicrons, to lipid catabolism were significantly induced in wild-type mice only. For most genes, the effects of fenofibrate were comparable to those of WY14643. However, at the same concentration fenofibrate was less potent in activating genes compared to WY14643. We conclude that activation of PPARα by two different agonists result in similar differences on the expression of a whole array of genes involved in intestinal dietary lipid metabolism. As in our model WY14643 is a stronger activator of PPARα compared to fenofibrate, we chose to use WY14643 in the remaining experiments.

Figure 1: Cellular location of selected genes involved in intestinal lipid metabolism. Genes shown in this scheme are involved in intestinal lipid metabolism. For more information on their regulation upon PPARα activation see Table 1.

110 5 PPARα regulates intestinal lipid absorption

Activation of PPARα increases intestinal lipid uptake Accumulation of intragastrically administered 3H-triglycerides and 14C-palmitic acids in triton WR1339-treated wild-type and PPARα-null mice. Since an array of genes involved in intestinal dietary lipid metabolism was induced by PPARα, we next examined whether this regulation could be translated into functional changes such as increased absorption of dietary fat. Wild-type and PPARα-null mice were fed WY14643 for 5days, intragastrically dosed with 3H-triolein and 14C-palmitic acid, and lipoprotein lipase was blocked to ensure accumulation of radiolabelled lipids in blood. The use of these two labels allow the evaluation of effects of PPARα activation on intraluminal digestion of triglycerides independent of effects observed on lipid absorption. Figure 2 shows the accumulation of 3H-labelled (derived from 3H-oleic acid) and 14C-labelled (derived from 14C- palmitic acid) lipids in blood up to ﬁve hours post lipid load (expressed as %3H- and %14C- of the administred dose). We observed signiﬁcantly more lipid accumulation in wild-type compared

Figure 2: Time cours of 3H- and 14C-appearance in whole blood up to ﬁve hours after the intragastric lipid load of wild- type and PPARα-null mice on a control diet or upon PPARα activation. Appearance of either 3H (A) or 14C (B) in whole blood after intragastric load of 3H-triolein- and 14C-labeld olive oil in wild-type (WT) and PPARα-null (PPARα-/-) mice. Blood was collected at the indicated time points after the intragastric olive 111 oil load and intravenously injected triton WR1339, as outlined in the materials and methods section. Both genotypes were on chow or chow supplemented with 0.1%WY14643. The %3H (14C) of bolus in whole blood was calculated assuming a blood volume of 72ml/kg bodyweight. The statistical analysis was performed with the SPSS programme version 15, using two-way ANOVA with genotype and diet as ﬁxed factors. (For uptake rates see also table 2). Values are means ± standard deviations (n=12 mice per group, N=48). Table 1: Gene expression data of genes involved in intestinal lipid metabolism.

Absolute fold changes Wild-type PPARα-null Gene symbol EG ID WY14643/ FENO/ WY14643/ FENO/ Control Control Control Control

Intestinal lipases and phospholipases Mgll 23945 9.53** 1.92** 1.06 -1.03 Lipe 16890 3.68** 1.20 1.36 -1.32 Lipa 16889 1.97** 1.34** 1.10 1.10 Pla2g6 53357 2.20** 1.32** 1.14 1.10 Pnpla2 66853 2.01** 1.14 -1.00 -1.11 Pnpla8 67452 1.58** 1.07 1.01 1.07 Daglb 231871 -1.36** -1.13 1.14 1.03 Ces1 12623 1.95** 1.53 1.07 1.14 Ces3 104158 3.41** 1.31 -1.00 1.43 Fatty acid transport and binding Fat/Cd36 12491 3.55** 1.76** 1.16 1.02 Slc27a2 26458 3.27** 1.59** 1.24 1.14 Slc27a4 26569 1.39** 1.27* 1.13 -1.03 Acsl1 14081 10.14** 2.02** 1.11 -1.01 Acsl3 74205 2.80** 1.20 1.28 -1.05 Acsl5 433256 1.21** 1.07 1.17 1.02 Fabp2 14079 1.15** 1.10** 1.06 1.04 Fabp1 14080 1.39** 1.23** 1.09 1.02 Scarb1 20778 2.59** 1.26** 1.04 -1.11 Scarb2 12492 3.13** 1.49** 1.14 1.23 TAG (re)-synthesis Gpat1 14732 6.30** 1.30 1.11 1.03 Mogat2 233549 -1.27** -1.01 1.06 1.02 Dgat1 13350 1.07 1.08 1.05 1.01 Dgat2 67800 1.46* -1.02 1.23 1.25 Gpat3 231510 1.02 1.13 1.06 1.07 Agpat3 28169 1.46** -1.14 1.18 1.04 Chylomicron assembly and secretion Mttp 17777 1.37** 1.17** 1.10 1.09 Stx5a 56389 1.01 -1.02 -1.01 1.04 Vti1a 53611 1.04 -1.05 1.07 -1.04 112 Bet1 12068 1.84** 1.24** 1.27 1.06 Sar1a 20224 1.04 -1.01 -1.01 1.06 Sec23ip 207352 1.11 1.07 1.03 1.02 Sec24b 99683 -1.24** -1.07 -1.04 1.04 Sec24a 77371 1.15* -1.01 1.11 1.09 Sec13 110379 -1.05 -1.09 1.06 1.08 Sec31a 69162 -1.23** -1.07 1.02 1.02 Sec16a 227648 1.12 1.04 -1.01 -1.00 Sec16b 89867 1.01 1.02 1.09 1.10 Ykt6 56418 1.19* 1.03 1.06 -1.03 Aytl2 210992 1.02 1.02 -1.01 -1.07 Agpat7 99010 1.03 1.08 -1.06 -1.06 5 PPARα regulates intestinal lipid absorption

Table 1 (continued): Gene expression data of genes involved in intestinal lipid metabolism.

Absolute fold changes Gene symbol EG ID Wild-type PPARα-null WY14643/ FENO/ WY14643/ FENO/ Control Control Control Control

Apolipoproteins and related Apob 238055 -1.20 -1.50 1.44 -1.54 Apoa1 11806 -1.16** -1.08* 1.01 1.01 Apoa2 11807 3.59** 4.48** 2.86 -1.11 Apool 68117 1.27** -1.06 1.04 1.03 Apoa4 11808 -1.07* -1.00 1.06 1.02 Apoc2 11813 1.21** 1.06 1.15 -1.01 Apoc3 11814 -1.87** -1.11 1.03 -1.01 Apoe 11816 -1.45** -1.17 1.03 1.04 Apol3 75761 1.08 1.07 1.12 -1.10 Angptl4 57875 9.21** 3.17** 1.23 1.53 Vldlr 22359 6.78** -1.04 -1.11 -1.19 Mitochondrial, microsomal, peroxisomal fatty acid oxidation Acaa2 52538 2.36** 1.34** 1.17 -1.01 Acad10 71985 1.68** 1.01 1.19 1.28 Acad8 66948 1.17** -1.00 -1.01 -1.01 Acad9 229211 -1.03** -1.05 1.03 -1.08 Acadl 11363 2.02** 1.38** 1.12 1.15 Acadm 11364 1.72** 1.16 1.03 1.01 Acads 11409 3.36** 1.51* 1.25 -1.03 Acadsb 66885 -1.20** -1.13 -1.09 -1.05 Acadvl 11370 3.08** 1.57** 1.08 1.04 Acot10 64833 -1.90** -1.15 1.01 1.03 Acot2 171210 5.85** 1.65** 1.33 1.25 Acot9 56360 -1.46** -1.11 1.01 1.07 Aldh9a1 56752 1.51** 1.19** 1.11 1.12 Cpt1a 12894 2.42** 1.30* 1.12 1.21 Cpt2 12896 2.58** 1.46** 1.23 1.23 Crat 12908 3.99** 1.46** 1.15 1.04 Dci 13177 7.75** 2.52** 1.16 1.04 Decr1 67460 3.64** 1.43** 1.09 1.17 Hadha 97212 3.22** 1.60** 1.11 1.14 Hadhb 231086 1.77** 1.20* 1.13 1.09 113 Hibch 227095 1.30** -1.06 1.05 1.09 Slc22a5 20520 7.62** 2.32** 1.17 1.34 Slc25a20 57279 4.48** 1.82** 1.45 1.21 Aldh3a2 11671 4.97** 1.95** 1.22 1.19 Cyp4a10 13117 75.22** 30.27** 1.00 1.10 Abcd3 19299 2.63** 1.55** 1.16 1.14 Acaa1a 113868 3.78** 1.51 1.35 1.18 Acaa1b 235674 6.92** 3.08** 1.80** 1.03 Acot3 171281 149.46** 5.12** 1.07 -1.03 Acot4 171282 5.64** 1.81** 1.44 1.38 Acot5 217698 1.21* 1.07 -1.04 -1.01 Acot8 170789 3.71** 1.59** 1.04 1.05 Acox1 11430 2.90** 1.49** 1.18* 1.11 Acox2 93732 6.65** 1.73 2.12 1.22 Crot 74114 3.16** 1.45** 1.07 1.03 Decr2 26378 11.49** 2.24** 1.23 -1.41 Table 1 (continued): Gene expression data of genes involved in intestinal lipid metabolism.

Absolute fold changes Gene symbol EG ID Wild-type PPARα-null WY14643/ FENO/ WY14643/ FENO/ Control Control Control Control

Mitochondrial, microsomal, peroxisomal fatty acid oxidation (continued) Ech1 51798 3.15** 1.40** 1.08 1.04 Ehhadh 74147 7.25** 1.58** 1.04 1.01 Hsd17b4 15488 3.71** 1.63** 1.15 1.09 Peci 23986 6.15** 1.84** 1.11 1.08 Pecr 111175 1.94** 1.03 1.00 1.13 PPARα 19013 5.20** 2.40** 1.23 1.20 Ketone body synthesis Acat1 110446 1.00 -1.05 1.04 -1.08 Hmgcl 15356 1.79** 1.17* 1.03 -1.00 Hmgcs2 15360 10.94** 3.12** 1.62 1.34 Glycerophospholipid metabolism Scap 235623 2.13** -1.01 -1.04 -1.02 Fasn 14104 2.74** 1.40 1.31 -1.01 Chkb 12651 1.33** 1.08 -1.02 1.00 Pcyt1a 13026 -1.86** -1.21 -1.10 -1.13 Srebf1 20787 1.78** 1.17 1.23 -1.04 Chka 12660 -1.41** -1.02 1.12 1.08 Ptdss2 27388 2.26** 1.20 1.01 1.02 Pla2g6 53357 2.20** 1.32** 1.14 1.10 Gpd1 14555 1.69** 1.06 -1.00 -1.10 Cdipt 52858 -1.35** -1.06 1.02 1.07 Chkb 12651 1.33** 1.08 -1.02 1.00 Ppap2b 67916 -1.28** -1.19* 1.00 -1.02 Pcyt2 68671 1.26** 1.12 1.13 1.06 Wild-type and PPARα-null mice were treated with chow, or chow supplemented with 0.1%WY14643 or 0.1%fenofibrate (FENO) for 5days. RNA was isolated from small intestine of total length and whole tube and was hybridized to Affymetrix GeneChip arrays (MOE430 2.0). Fold changes having **, indicate genes that satisfied the significance critera of FDR<1% (q-value<0.01); fold changes with * have an FDR< 5%(q<0.05) for the respective comparison.

114 5 PPARα regulates intestinal lipid absorption

to PPARα-null mice. Moreover, at all time points except the last, treatment with WY14643 resulted in signiﬁcantly increased lipid accumulation in wild-type mice, but not in PPARα-null mice. Uptake rates of 3H- and 14C-labelled lipids in blood, as determined by the slope of the time course, were signiﬁcantly lower in PPARα-null mice for both labels (Table 2). However, WY14643 treatment did not affect this parameter in both mouse strains.

Table 2: Slope (uptake rates, speed) calculated for triolein (3H) and palmitic acid (14C) as % of the given bolus uptake/ hour. Group Slope 3H* Slope 14C* N Wild-type-Control 3.97 (1.05)a 4.10 (1.02)a 12 Wild-type-WY14643 4.03 (1.45)a 3.66 (1.30)a 12 PPARα-null -Control 2.88 (0.68)b 3.49 (0.83)b 12 PPARα-null -WY14643 2.55 (1.22)b 2.93 (1.05)b 12 * a, b, c: a different letter indicates a significant difference between the groups using two-way ANOVA with genotype and diet as fixed factors. Values are means ± standard deviations. The statistical analysis was performed with the SPSS programme version 15, using two-way ANOVA with genotype and diet as fixed factors. (For uptake rates see also table 2). Values are means ± standard deviations (n=12 mice per group, N=48).

Figure 3 shows the accumulation of both radioactive labels five hours after the lipid load in the intestinal lumen (Figure 3A, B) and intestinal wall along the proximal distal axis (Figure 3C, D). We observed no difference in luminal content for both labels between the different treatment groups (Figure 3A, B). However, the amount of radioactive lipids present in the intestinal wall was significantly reduced in treated wild-type mice compared to untreated mice. Treatment with WY14643 had no effect on tissue content in PPARα-null mice (Figure 3C, D). Detailed analyses of the tissue distribution of the radioactive lipids along the proximal-distal axis showed that for 3H-labelled lipids the effect was restricted to the first two sections (S1, S2), whereas for 14C-labelled lipids only the first section (S1) was significantly effected by PPARα activation (Figure 3C, D). 115

Table 3: Ratio of 14C to 3H after an intragastric load, in blood of wild-type and PPARα-null after 5days of WY14643 treatment. 14C/ 3H ratio in Blood* Group 2h 3h 4h 5h N Wild-type-Control 1.87 (0.37)a 1.43 (0.23)a 1.25 (0.16)a,b 1.17 (0.09)a,b 12 Wild-type-WY14643 1.60 (0.21)b 1.27 (0.19)a 1.14 (0.15)a 1.10 (0.14)a 12 PPARα-null -Control 1.97 (0.36)a 1.59 (0.24)b 1.42 (0.23)b,c 1.35 (0.21)b,c 12 PPARα-null -WY14643 2.08 (0.34)a 1.69 (0.45)b 1.56 (0.40)c 1.40 (0.36)c 12 * a, b, c: a different letter indicates a significant difference between the groups using two-way ANOVA with genotype and diet as fixed factors. Values are means ± standard deviations. The statistical analysis was performed with the SPSS programme version 15, using two-way ANOVA with genotype and diet as fixed factors. (For uptake rates see also table 2). Values are means ± standard deviations (n=12 mice per group, N=48). Similar results were obtained from an independent experiment in which a non-radiolabelled lipid load was given to wild-type and PPARα-null mice. An oil red O (ORO) staining was performed four hours post lipid load to visualize intestinal lipid content. Results of a typical staining are presented in Figure 4. Although an ORO staining is not suitable for determining quantitative differences in the lipid content of the sections, it is clear that in wild-type mice treated with WY14643 the lipid content in the intestinal wall was drastically reduced compared to the control diet (Figure 4). WY14643 treatment did not affect lipid content in PPARα-null mice. In Table 3 the ratio of 14C- to 3H-labelled lipids (14C/3H) in blood, measured at the different time points is shown.

116

Figure 3: 3H- and 14C-appearance in intestinal lumen and small intestinal sections five hours poast intragastric lipid load of wild-type and PPARα-null mice on a control diet or upon PPARα activation. Appearance of either 3H or 14C in intestinal lumen (A, B) or in intestinal sections (C, D) five hours after intragastric load of 3H-triolein- and 14C-labeld olive oil in wild-type (WT) and PPARα-null (PPARα-/-) mice. Both genotypes were on chow or chow supplemented with 0.1%WY14643. Small intestine has been divided into three equal sections, each section been washed and the luminal content has and collected and analyzed. For details see the materials and methods section. There was no difference of the luminal content between the various groups, whereas for the tissue sections significant differences have been observed either when PPARα was activated or absent from the model. The statistical analysis was performed with the SPSS programme version 15, using two-way ANOVA with genotype and diet as fixed factors. (For uptake rates see also table 2). Values are means ± standard deviations (n=12 mice per group, N=48, a, b, c: a different letter indicates a significant difference between the groups. 5 PPARα regulates intestinal lipid absorption

Figure 4: Typical Oil-red- O stainings four hours post lipid load of wild- type and PPARα-null mice under control conditions or treated with WY14643 for five days. Mice were again fed chow, or chow supplemented with 0.1% WY14643 for five. An intragastric lipid load of 400µl olive oil ws administered to all mice, four hours later mice were euthanized and the small intestine was divided into four equal parts. The images show the Oil-red O staining for the first part of the swiss rolled small intestine.

This ratio is an indicator for the relative “speed” of intraluminal hydrolysis of triolein, since the tritium labelled lipids can only be absorbed (and subsequently measured in blood) after hydrolysis, in contrast to 14C-labelled lipids that are derived from palmitic acid. Assuming the mechanism of intestinal uptake of oleic and palmitic acid is the same for both fatty acids, the 14C/3H ratio in time thus reﬂects the speed of lipid processing in small intestine. As expected, all groups showed a decrease in 14C/3H ratio with increasing time post lipid load, which reﬂects the increased availability of 3H-oleic acid due to luminal hydrolysis of 3H-triolein. At all time points a lower 14C/3H ratio was observed in blood of wild-type mice treated with WY14643 compared to all other groups, demonstrating these mice have an increased digestion and absorption of 3H- labelled lipid derived from triolein.

Table 4: Ratio of 14C to 3H after an intragastric load, in intestinal tissue of wild-type and PPARα-null after 117 5days of WY14643 treatment. 14C/3H ratio in intestinal tissue* Group S1 S2 S3 N Wild-type-Control 1.14 (0.12 ) a,b 0.60 (0.12) 0.46 (0.16) 12 Wild-type-WY14643 0.95 (0.11) a 0.70 (0.17) 0.53 (0.23) 12 PPARα-null -Control 1.25 (0.31) b 0.80 (0.38) 0.55 (0.30) 12 PPARα-null -WY14643 1.23 (0.44) b 0.66 (0.27) 0.60 (0.44) 12 * a, b, c: a different letter indicates a signiﬁcant difference between the groups using two-way ANOVA with genotype and diet as ﬁxed factors.

This is in concordance with our observations that PPARα activation in wild-type mice resulted in higher uptake of both 3H-oleic acid and 14C-palmitic acid derived lipids compared to the other groups (Figure 2). However, as already mentioned before, there was no difference in speed (expressed as uptake rates) when PPARα was activated (Table 2). Figure 5: Appearance of 3H and 14C in whole blood after an intragastric load of 3H-triolein- and 14C-labelled olive oil in wild-type mice after intravenous triton injection and pretreatment of etomoxir. Wild-type mice were treated with WY14643 for ﬁve days, pretreated with etomoxir and two hours later injected with triton WR1339, after which an intragastric olive oil load containing 3H-triolein and 14C-palmitic acid was given. Etomoxir is known to block the β-oxidation pathway thus prevents activated fatty acids from being burned. Whole blood was collected at the indicated time points, the luminal and intestinal tissue contents were collected 5hours post lipid load. Upon PPARα activation by WY14643 tissue and blood content both were signiﬁcantly induced compared to chow fed mice. Comparable results were obtained for the palmitic acid label 14C. Values are means ± standard deviations (n=3 mice per group, N=6). The statistical analysis was performed in SPSS version 15 using the student t-test statistics. 118

In Table 4 the 14C/3H ratio in the intestinal tissue is given. Along the proximal-distal-axis (S1-S3) this ratio decreased independent of treatment or genotype. Whereas PPARα activation in wild-type mice led to a lower 14C/3H ratio in the most proximal part, this difference did not reach statistical significance when compared to wild-type control fed mice (p=0.072) but it was significantly different from PPARα-null mice irrespective of the treatment. These data are in line with the observation that accumulation of 3H-labelled lipids in blood was higher in wild-type mice upon PPARα activation. PPARα-null mice displayed significantly higher 14C/3H ratios in blood throughout the whole experiment (Table 2) although absolute levels for both labels and treatments were significantly lower compared to wild-type mice (Figure 2). The amount of 3H-labelled lipids in the intestinal wall was also significantly 5 PPARα regulates intestinal lipid absorption

lower in PPARα-null mice compared to wild-type (Figure 3). These results show that PPARα-null mice better absorb free fatty acids compared to fatty acids derived from triglycerides, which suggests luminal hydrolysis of TG is reduced by the absence of PPARα. From the gene expression experiments we expected that WY14643 feeding should have resulted in a more pronounced effect on the appearance of 3H- and 14C-labelled lipids in blood of wild- type mice than we actually observed. It is well established that activation of PPARα results in increased peroxisomal and mitochondrial oxidation of fatty acids (Table 1 and [17, 53]). In the case absorbed fatty acids are preferentially oxidized instead of utilized for TG resynthesis, we hypothesized that this response could mask the effects of PPARα activation on chylomicron formation. In other words, inhibition of fatty acid oxidation should drastically enhance chylomicron formation. Wild-type mice were therefore pretreated with etomoxir, which potently inhibits mitochondrial and peroxisomal fatty acid oxidation [268], two hours before administration of the lipid load. Accumulation of radiolabelled lipids in the lumen, intestinal wall and blood is presented in Figure 5. Pretreatment with etomoxir did not signiﬁcantly affect the amount of radiolabelled lipids in the intestinal lumen of control and WY14643 treated mice. However, inhibition of fatty acid oxidation in PPARα-activated wild-type mice resulted in a signiﬁcantly higher accumulation of 3H-labeled lipids in both intestinal wall and circulation at all time points. Comparable results were observed for 14C-labelled lipids, except for tissue content (Figure 5). These data clearly demonstrate that lipid absorption and chylomicron secretion is regulated by PPARα, and also suggest that oxidation of fatty acids is a quantitative important mechanism to control intracellular levels of fatty acids.

Regional expression of genes involved in intestinal dietary lipid metabolism upon PPARα activation by WY14643. The effect of PPARα activation on the absorption of triolein-derived oleic acid and palmitic acid was most pronounced in part one and two (S1, S2) or only seen in the first part (S1), respectively. 119 We next determined whether this regional difference in lipid handling could be specifically linked to differential gene expression in the respective intestinal part. The longitudinal distribution of genes involved in dietary lipid metabolism was investigated in detail in wild-type mice (Figure 6A-D). For all genes we observed that the effect of PPARα activation translated on changes in expression along the complete proximal-distal axis. However, as expected, the majority of the genes were most profoundly regulated the first two-thirds of the small intestine (parts 1 to 6), Figure 6A, B: Longitudi- nal distribution of genes involved in dietary lipid metabolism in wild-type mice upon PPARα activation by WY14643. qRT-PCR was used to determine the differential expression of genes involved in dietary lipid metabolism after a 5day treatment with the PPARα ligand WY14643. The intestinal sections were isolated along the proximal-distal axis of the small intestine from adult 129Sv mice either on chow or chow supplemented with 0.1%WY14643 (n=4 mice per group). The small intestine was divided into 10 equal parts; part 1 refers to the most proximal part (duodenum), part 10 refers to the most distal (terminal ileum). The effect of PPARα activation in wild-type mice along the proximal distal axis (PDA) of (A) intestinal lipases and phospholipases: mono- glyceride lipase (MGLLl), lipase, lysosomal acid lipase A (LIPA), phospholipase A2, group VI (PLA2G6), patatin-like phospholipase domain containing 2 (PNPLA2), patatin-like phospholi- 120 pase domain containing 8 (PNPLA8); selected genes involved in (B) fatty acid transport and binding, CD36 antigen (CD36), solute carrier family 27 (fatty acid transporter), member 2 (SLC27A2), solute carrier family 27 (fatty acid transporter), member 4 (SLC27a4), acyl-CoA synthetase long-chain family member 1 (ACSL1), acyl-CoA synthetase long- chain family member 3 (ACSL3), acyl-CoA synthetase long-chain family member 5 (ACSL5);

WT-Control WT-WY14643 5 PPARα regulates intestinal lipid absorption

Figure 6C, D: Longitudi- nal distribution of genes involved in dietary lipid metabolism in wild-type mice upon PPARα activation by WY14643. The effect of PPARα activation in wild-type mice along the proximal distal axis (PDA) of (C) TAG (re-) synthesis, monoacylglycerol O-acyltransferase 2 (MOGAT2), diacylglycerol O-acyltransferase 1 (DGAT1), diacylglycerol O-acyltransferase 2 (DgAT2); (D) chylomicron formation, microsomal triglyceride transfer protein (MTTP), blocked early in transport 1 homolog (S. cerevisiae) (BET1), fatty acid binding protein 1, liver (FABP1). Messenger RNA levels were standardized to cyclophilin; part 10 of the control mice was arbitrarily set to 1. Data are presented as mean ± standard deviation. Comparable re- WT-Control WT-WY14643 sults were observed when 18S ribosomal RNA was used as reference.

121 which nicely followed the expression gradient of PPARα along the small intestine [70].

Effects of a low-fat versus a high-fat dietary intervention in wild-type and PPARα null mice. Thus far our results showed that activation of PPARα by a potent, synthetic agonist resulted in increased lipid absorption. Since it is known that chronic high-fat feeding activates hepatic PPARα and PPARα-signaling [59], we investigated whether such intervention also functionally affected intestinal lipid absorption in wild-type and PPARα-null mice. To this end mice were fed a low-fat (LFD) or a high fat (HFD) diet for 20 weeks. Results of this dietary intervention study are presented in Figure 7 and 8. PPARα mRNA levels were modestly

122

Figure 7: Bodyweight development, energy intake and fat content in feces of wild-type and PPARα-null mice on a HFD or LFD. Mice were fed a HFD or LFD. Left panel: upper diagram, bodyweight gain until week 17; lower diagram Bodyweight gain set out as difference between HFD and LFD for each genptype. In week 17, energy intake was measure and 48h feces were collected and the fat content measured as described in the materials and methods section. The statistical analysis was performed with the SPSS programme version 15 by using two-way ANOVA with genotype and diet as ﬁxed. Values are means ± standard deviations (N=40, n=5 (LFD), n=15 (HFD)). a, b: a different letter indicates a signiﬁcant difference between the groups. 5 PPARα regulates intestinal lipid absorption

but signiﬁcantly increased by high-fat feeding, as determined by qRT-PCR (Figure 8). To examine whether a HFD is also associated with increased PPARα activity in the small intestine, we measured mRNA expression of CYP4A10, HMGCS2, MOSC1, SCARB2 and L-FABP. The ﬁrst two genes are well established PPARα target genes that are sensitive to the presence and activation of PPARα [53], and the other three genes have been shown to be small intestinal target genes [70]. All genes can thus serve as markers for PPARα activity. Expression of CYP4A10 and HMGCS2 was induced by a HFD in wild-type mice, whereas expression was very low and not

Figure 8: qRT-PCR analysis of intestinal PPARα target genes after 20 weeks on a LFD or HFD in wild-type an PPARα- null mcie. qRT-PCR analysis of intestinal PPARα target genes after 20 weeks on a LFD or HFD in wild-type an PPARα-null mcie. N=4 or 5, values are means+/- SEM, relative to villin and LFD was set arbitrarily to 1.

123

inducible in PPARα-null mice (Figure 8C and D). Similar observations were made for the other three genes, indicating that feeding a HFD resulted in enhanced PPARα activity, although the effects were modest compared to WY14643 (Table 1). The feces of PPARα-null mice fed a HFD contained a signiﬁcantly higher amount of fat compared to the corresponding wild-type mice (Figure 7D), indicating that intestinal PPARα also controls dietary fat absorption under ‘normal’ nutritional conditions (i.e. not only after utilizing pharmacological-type of agonists). Energy intake did not differ between the experimental groups (Figure 7C), indication this reduced lipid absorption resulted in less weight gain in PPARα-null mice fed a HFD compared to wild-type mice (Figure 7A, B). Thus, this dietary intervention study demonstrated that a HFD indeed activates intestinal PPARα.

Discussion

In this study we set out to determine whether PPARα regulates intestinal lipid absorption and transfer into the body. We show that, irrespectively of the agonist used, the expression of many genes involved in all steps of intestinal dietary lipid metabolism is PPARα dependently regulated. These results are in agreement with the outcomes of independently performed experiments reported by our laboratory, in which a less comprehensive microarray was used [70]. Here we demonstrate that these changes in gene expression are functionally translated, because under control conditions wild-type mice accumulated more lipids than PPARα-null mice; treatment with WY14643 resulted in signiﬁcantly increased lipid accumulation in wild-type mice, but not in PPARα-null mice; and in a long-term dietary intervention study PPARα-null mice absorbed less fat than wild-type mice. Combined, our data clearly indicate that PPARα controls intestinal lipid absorption.

Roughly speaking, intestinal absorption and secretion of TGs are comprised of four steps [93, 255]. Luminal TGs are hydrolyzed to fatty acids and 2-monoacylglycerols (2MAGs), which are then taken up by enterocytes, rapidly resynthezised to TGs, and transported out of the cell in chylomicrons. We provide evidence that all steps, except lipid uptake in the enterocyte, are regulated by PPARα. Moreover, our data indicate that intestinal oxidation of post prandial-derived fatty acids is an important route to control intracellular levels of these potentially toxic compounds.

Our experimental setup allows discriminating between the effects of PPARα on lipids derived from free fatty acid (14C-palmitic acid) or TG-derived fatty acid (3H-oleic acid). Lack of PPARα leads to a reduction in TG-derived lipids in small intestinal tissue, whereas FFA-derived lipids were not affected (Figure 3C, D), suggesting that intraluminal hydrolysis of TGs is impaired in the absence 124 of PPARα. In accordance, expression data show that intestinal lipases and phospholipases, such as MGLL, LIPA, PLA2G6 and PNPLA2, are regulated by PPARα. Moreover, effects on tissue lipid contents were most clearly observed in the most proximal part (S1), coinciding with the main anatomical location where long-chain fatty acids are taken and the region PPARα is highest expressed [70, 93, 255]. Activation of PPARα in wild-type mice resulted in reduced levels of lipids derived from both sources in the proximal part, whereas PPARα-null mice were not affected. This is in agreement with various reports demonstrating that activation of PPARα reduces 5 PPARα regulates intestinal lipid absorption

triglyceride content in steatotic livers, which is believed to occur via increased oxidation of fatty acids [270-272].

Lack of PPARα resulted in less and slower accumulation of dietary lipids in the circulation, and the difference in accumulated amounts increased after PPARα activation (Figure 2 and 4). However, unexpectedly, the rate of appearance, expressed as percentage of dose per hour, was not altered after PPARα activation in wild-type mice. Thus, although after PPARα activation the expression of genes involved in TG resynthesis and chylomicron formation are induced, this does not necessarily result in enhanced secretion of chylomicrons. The absence of this functional effect is due to the preferential oxidation of fatty acids in the enterocytes upon PPARα activation, since inhibition by etomoxir of PPARα-induced mitochondrial and peroxisomal fatty acid oxidation [268] clearly reveals that PPARα induces chylomicron secretion. This is in contrast with hepatic VLDL production, which is reduced upon PPARα activation [172, 273, 274]. Our data also demonstrate that in addition to sequestration and TG resynthesis, oxidation of fatty acids is an important mechanism to control the levels of these potential highly toxic metabolites, which is also supported by published data [189, 259, 275]. Importantly, by comparing an obesity- resistant with a obesity-prone mouse strain, Kondo et al [189] showed that impaired intestinal fatty acid oxidation resulted in more dietary lipid entering the body, which was associated with increased diet-induced obesity. Studies reported in two other papers demonstrated that nutritional agonists of PPARα, i.e. polyunsaturated fatty acids, also PPARα-dependently induced the expression and activity of intestinal fatty acid oxidation [259, 275]. The signiﬁcance of a well-balanced intestinal fatty acid oxidation is further underlined by the reports that showed an enhanced intestinal oxidation is correlated with a reduced severity of inﬂammatory bowel disease [204, 276].

Taken together, we are the first to show that PPARα regulates intestinal lipid uptake. Our data provide new insight into the functional role of PPARα in the small intestine and may be of particular importance for the development of fortified foods, and for prevention and therapies for 125 treating obesity and inflammatory bowel diseases. 6 General discussion

PPARα was first described in 1990 as a receptor that is activated by peroxisome proliferators [51]. Shortly after this discovery, it was found in in vitro studies that PPARα could be activated by natural fatty acids and their derivatives [52, 85-87, 142]. Its tissue expression pattern indicated that PPARα was highly expressed in organs that carry out significant catabolism of fatty acids such as the liver, heart, small intestine and kidney [52-54]. It is therefore not surprising that the identification of PPARα target genes has concentrated mainly on cellular lipid metabolism in the context of the hepatocyte. Indeed, the first PPARα target gene identified in liver was acyl- coenzyme A oxidase [55] which is involved in peroxisomal fatty acid β-oxidation. This discovery was soon followed by the identification of many more PPARα target genes involved in several key functions of lipid metabolism [53]. Thus, while the function of PPARα in liver was well studied, little was known about PPARα and PPARα target genes in non-hepatic tissues. When we started with this research, this was especially the case with respect to the role of PPARα in the small intestine, which had only been addressed in few studies [90, 91]. Knowledge on the regulatory and physiological function of PPARα in the small intestine is of particular interest, since the average Western diet contains a high amount of triacylglycerols [92] that are hydrolyzed to monoacylglycerol and free fatty acids before entering the enterocyte [93]. Consequently the small intestine is frequently exposed to high levels of PPARα agonists. In this thesis we therefore aimed at characterizing the function of PPARα in the small intestine, with special emphasis on nutritional relevance, making use of functional genomics experiments and advanced bioinformatics tools, thus applying a so-called nutrigenomics approach.

As a ﬁrst step the expression of PPARα was measured in several human and mouse tissues. In chapter 2 we showed that PPARα was highly expressed in both human and murine small intestine. In addition, we demonstrated that the activation of PPARα by a potent PPARα agonist, called WY14643, resulted in differential expression of a large set of genes involved in a variety of pathways, including intestinal lipid handling, cell cycle, differentiation, apoptosis and host 128 defense. Next to genes involved in fatty acid and triacylglycerol metabolism, genes coding for transcription factors and enzymes connected with steroid (sterol) and bile acid metabolism, including FXR, SHP, and SREBP1 were speciﬁcally induced. This demonstrates that cross-talk between PPARα and other lipid-regulated transcription factors also occurs in small intestine, in addition to liver [115, 116], and is functionally in line with the well-established roles of bile acids for the absorption of dietary lipids [93, 255]. The results from our microarray experiments and the subsequent data analyses, suggested an important role for PPARα in the regulation of gene expression in the small intestine, of which some aspects were more thoroughly investigated. 6 General discussion

PPARα regulates intestinal lipid absorption Although PPARα was known to regulate fatty acid metabolism in other organs, such as liver [53, 54], the link between PPARα and the regulation of lipid handling in the intestine was new. A detailed analysis of PPARαexpression along the two small intestinal axis showed that PPARα was highest expressed in jejunal villus cells (chapter 2). These findings are in agreement with earlier observations performed by crude tissue distribution studies in rats [88, 89], and correlate with the main anatomical site fatty acids are taken up by the enterocyte [93, 255]. We not only showed that PPARα was expressed but also regulated in intestinal villus cells and used these conclusions to formulate new hypotheses on PPARα regulating intestinal lipid absorption. Thus, in chapter 5 we examined whether PPARα regulates the intestinal absorption of dietary triacylglcerol using a combination of gene expression and functional studies in wild-type and PPARα-null mice. We showed that PPARα-mediated induction of genes involved in intestinal lipid handling resulted in reduced intracellular lipid levels. Our results are in agreement with various reports demonstrating that activation of PPARα reduces triglyceride content in steatotic livers, which is believed to occur via increased oxidation of fatty acids [270-272]. By blocking fatty acid oxidation in the small intestine, we showed that oxidation of fatty acids is a quantitative important mechanism to control intracellular levels of fatty acids in this organ. Importantly, it has been reported that in addition to synthetic agonists, nutritional agonists of PPARα, i.e. polyunsaturated fatty acids, also PPARα-dependently induced the expression and activity of intestinal fatty acid oxidation [259, 275]. On the other hand, the observed increased formation of chylomicrons (chapter 5) upon PPARα activation contrast with reports suggesting hepatic VLDL production is reduced upon PPARα activation [172, 273, 274]. Although many studies have been conducted to examine the effect of PPARα activation on hepatic VLDL metabolism, it yet is not entirely known by which mechanism these effects are generated. Indeed, several PPARα target genes that have been identified the last couple of years, do just not fit the classical paradigm of hepatic PPARα function and it has become clear that the role of PPARα in TG-rich lipoprotein metabolism is much broader than previously envisioned [172, 277]. 129

PPARα activation by synthetic and dietary agonists and IBD In chapter 2 we suggested that PPARα influences the immune and inflammatory response of the small intestine, and support the possibility that enterocytes are involved in a local response to injury/inflammation at the epithelial surface. One could imagine that a repression of the inflammatory response in the intestine by PPARα might be therapeutically valuable for patients with inflammatory bowel disease (IBD). Although several studies suggest a link between PPARγ and inflammatory bowel disease [138-141], hardly anything was known about the effect of PPARα. By blocking the fatty acid oxidation we showed that oxidation of fatty acids is a quantitative important mechanism to control intestinal intracellular levels of fatty acids. We showed this effect in the small intestine was PPARα dependent (chapter 5). Our findings in chapter 2, 4 and 5 on the significance of a well-balanced intestinal fatty acid oxidation is further underlined by others showing an enhanced intestinal oxidation is correlated with a reduced severity of inflammatory bowel disease [204, 276]. Various studies, predominantly performed in rodents have shown that the administration of omega-3 fatty acids had positive effects on inflammatory bowel diseases [219, 278-282] In contrast, a recent study in which in patients were treated with either omega-3 free fatty acids or placebo for up to 58 weeks showed that treatment with omega-3 free fatty acids was not effective for the prevention of relapse in Crohn’s disease [283] In chapter 4 we observed differential gene activation in vivo between different long chain fatty acids, eicosapentaenoic acid (EPA, C20:5, n-3), docosahexaenoic acid (DHA, C22:6, n-3) and oleic acid (OA, C18:1, n-9) compared to WY14643. The exact mechanism(s) underlying these differences are still unclear. We speculated this may be partially due to the differential recruitment of coactivators such as SRC1, MED1, PGC1Α, and P300 by the agonists. However, a recent paper reported that no PPARα-coregulator interactions were stimulated specifically by C22:6, C20:5 or C18:1, and not by WY14643 [143]. Our second possible explanation that unknown additional signaling routes not shared by the three agonists may exist, is therefore more likely.

Organ-specific function of PPARα activation In chapter 3 we identified and compared the PPARα-dependently regulated genes in small intestine and liver, and based on the commonalities and differences we made first suggestions on PPARα’s organ specific functions. Such an approach has not been undertaken so far, which makes a comparison an connection with other results difficult. It shows however the innovative character of this approach we have performed in this chapter. We identify potentially secreted proteins regulated by PPARα. For several of these proteins, such as ANGPTL4, FGF21, FGF15 and IL18, recent studies showed that they indeed function as important PPARα-regulated intercellular messengers [162-167]. This demonstrates the validity and relevance of our approach and also highlights the necessity for further research on the biological role of the other, not well-studied 130 secreted proteins, which might reveal other mechanisms by which PPARα is able to signal to other organs. With this report we provide an extensive framework for further research on the (organ-specific) function of PPARα at different levels. The precise molecular mechanism(s) responsible for the organ specific responses to PPARα activation by WY14643 remain still unclear, however our data add a significant contribution and directs to further research in this research area. A series of future studies are required to investigate the different scientific issues in more detail and to allow more precisely the characterization of organ-specific functions with as ultimate goal to study the whole body system biology of PPARα. 6 General discussion

Conclusion and recommendations By maximally utilizing the unique possibilities offered in the post-genome era, the studies described in these thesis reported on the function of PPARα in small intestine. We conclude that intestinal PPARα plays an important role, is relevant for nutrition, and its effects are distinguishable from the hepatic PPARα response. These new insights provide a better understanding of normal intestinal physiology, and may be of particular importance for the development of fortiﬁed foods, and prevention and therapies for treating obesity and inﬂammatory bowel diseases.

The research described here shows that whole genome analyses combined with functional models and synthetic agonists provide extremely valuable tools to study and identify physiological functions of PPARα. We show that this strategy is also effective for specific dietary compounds, and believe such exemplar nutrigenomics approach is the only method that will improve our understanding of molecular mechanisms underlying the effects of nutrition on health. Mouse models have greatly added to our understanding of a myriad of human diseases, such as obesity, diabetes, cardiovascular disease, and cancer. In addition to our experiments, we envision an even more refined approach to study PPARα function will come from mice that have an organ- specific or cell-type specific ablation of PPARα. The recent launch of the International Knockout Mouse Consortium (IKMC) [284, 285] is expected to play a significant role in generating such mouse mutants, and it is envisioned that also the nutrigenomics research community will profit from this. Regarding the organ-organ comparison performed in this thesis, we believe that a shorter intervention period of e.g. 6 hours would possibly allow to discriminate between primary (acute) and secondary (chronic) effects of PPARα activation. As the cell composition of liver and small intestine is different, one could envision that part of the differences may be due to different cell- cell signaling (in liver e.g. hepatocyte to kupffer cell, hepatocyte-stellate cell, or hepatocyte- cholangiocyte signaling; in small intestine e.g. enterocyte-goblet cell, enterocyte-paneth cell, or enterocyte-enteroendocrine cel signalingl). This could be elucidated e.g. by using isolated hepatocytes after short term WY14643 intervention of mice, or villus cell isolation. Likewise, 131 cell isolation techniques such as laser capture microdissection is another technique valuable that still has to be applied to such nutrigenomics studies. Transcriptome analysis, or any ‘omics-technique’, generates large amounts of data, which in turn need to be translated into physiologically relevant information. In this post-genome era, a single group, let alone an individual researcher, is not able to efficiently cope with this. Only multidisciplinary research teams, that integrate expertise from different disciplines such as nutrition, physiology, statistics and bioinformatics, will be able to successfully do so. Good examples of such recently founded initiatives are the Nutrigenomics Consortia in the Netherlands, Australia & New Zealand, the USA, and Brazil. An even higher level of integration is being established by the European Nutrigenomics Organisation (NuGO), which is formed by over 20 universities, research institutes and companies all over Europe. Thus, in an ideal world, combining the knowledge of leading experts in the different research areas will ultimately lead to a win-win situation for each partner. A

132 Appendix

Supplemental data downloads can be found at http://www.bunger.nl/thesis

2 Genome-wide analysis of PPARα activation in murine small intestine Supplemental data 1: Detailed descriptions of the applied methods used Supplemental data 2: A complete list of regulated genes Supplemental data 3: Primer sequences are listed in table 1

3 Organ-specific function of PPARα as revealed by gene expression profiling Supplemental data 1: List of genes with classification in small intestine specific, liver specific and overlapping regulated genes Supplemental data 2: Overview of PPARα-dependant secretome analysis Supplemental data 3: Genes encoding secreted proteins regulated upon PPARα activation Supplemental data 4: Table of overrepresented Gene Ontology classes

4 PPARα-mediated effects of dietary lipids on intestinal barrier gene expression Additional table 1: PPARα-dependently barrier genes upon WY14643 treatment Additional table 2: PPARα-dependently regulated barrier genes upon OA treatment 133 Additional table 3: PPARα-dependently regulated barrier genes upon EPA treatment Additional table 4: PPARα-dependently regulated barrier genes upon DHA treatment Additional table 5: PARα-dependently regulated barrier genes after acute (6hr) and long-term (5 day) treatment with WY14643 Additional table 6: Overlap of PARα-dependently regulated barrier genes after acute treatment with WY14643, EPA, and DHA Additional table 7: Additional conﬁrmatory qRT-PCR data

5 PPARα regulates intestinal lipid absorption Supplemental table 1: Composition of the low-fat and high fat diet R

134 References

1. J.M. Mariadason, C. Nicholas, K.E. L’Italien, M. Zhuang, H.J.M. Smartt, B.G. Heerdt, W. Yang, G.A. Corner, A.J. Wilson, L. Klampfer, D. Arango and L.H. Augenlicht, Gastroenterology 128 (2005) 1081- 1088. 2. A. Stegmann, M. Hansen, Y. Wang, J.B. Larsen, L.R. Lund, L. Ritie, J.K. Nicholson, B. Quistorff, P. Simon-Assmann, J.T. Troelsen and J. Olsen, Physiol Genomics 27 (2006) 141-55. 3. C.R. Erwin, M.D. Jarboe, M.A. Sartor, M. Medvedovic, K.F. Stringer, B.W. Warner and M.D. Bates, Gastroenterology 130 (2006) 1324-32. 4. N. Fogg-Johnson and A. Merolli, Nutraceuticals World (2000) 86-95. 5. M. Müller and S. Kersten, Nat Rev Genet 4 (2003) 315-22. 6. J.M. Ordovas and V. Mooser, Curr Opin Lipidol 15 (2004) 101-8. 7. J. Kaput and R.L. Rodriguez, Physiol Genomics 16 (2004) 166-77. 8. L. Afman and M. Müller, J Am Diet Assoc 106 (2006) 569-76. 9. C.B. Clish, E. Davidov, M. Oresic, T.N. Plasterer, G. Lavine, T. Londo, M. Meys, P. Snell, W. Stochaj, A. Adourian, X. Zhang, N. Morel, E. Neumann, E. Verheij, J.T. Vogels, L.M. Havekes, N. Afeyan, F. Regnier, J. van der Greef and S. Naylor, Omics 8 (2004) 3-13. 10. M. Oresic, C.B. Clish, E.J. Davidov, E. Verheij, J. Vogels, L.M. Havekes, E. Neumann, A. Adourian, S. Naylor, J. van der Greef and T. Plasterer, Appl Bioinformatics 3 (2004) 205-17. 11. B. Desvergne, L. Michalik and W. Wahli, Physiol Rev 86 (2006) 465-514. 12. Z. Zhang, P.E. Burch, A.J. Cooney, R.B. Lanz, F.A. Pereira, J. Wu, R.A. Gibbs, G. Weinstock and D.A. Wheeler, Genome Res 14 (2004) 580-90. 13. P. Germain, B. Staels, C. Dacquet, M. Spedding and V. Laudet, Pharmacol Rev 58 (2006) 685-704. 14. K.W. Nettles and G.L. Greene, Annu Rev Physiol 67 (2005) 309-33. 15. D.L. Bain, A.F. Heneghan, K.D. Connaghan-Jones and M.T. Miura, Annu Rev Physiol 69 (2007) 201-20. 16. H. Sampath and J.M. Ntambi, Nutr Rev 62 (2004) 333-9. 17. B. Desvergne and W. Wahli, Endocr Rev 20 (1999) 649-688. 135 18. L. Michalik, J. Auwerx, J.P. Berger, V.K. Chatterjee, C.K. Glass, F.J. Gonzalez, P.A. Grimaldi, T. Kadowaki, M.A. Lazar, S. O’Rahilly, C.N. Palmer, J. Plutzky, J.K. Reddy, B.M. Spiegelman, B. Staels and W. Wahli, Pharmacol Rev 58 (2006) 726-41. 19. J.N. Feige, L. Gelman, C. Tudor, Y. Engelborghs, W. Wahli and B. Desvergne, J Biol Chem 280 (2005) 17880-90. 20. M. Ricote and C.K. Glass, Biochim Biophys Acta (2007). 21. A.C. Syvanen, Nat Genet 37 Suppl (2005) S5-10. 22. J.L. Freeman, G.H. Perry, L. Feuk, R. Redon, S.A. McCarroll, D.M. Altshuler, H. Aburatani, K.W. Jones, C. Tyler-Smith, M.E. Hurles, N.P. Carter, S.W. Scherer and C. Lee, Genome Res 16 (2006) 949-61. 23. J.B. Fan, M.S. Chee and K.L. Gunderson, Nat Rev Genet 7 (2006) 632-44. 24. J.M. Stuart, E. Segal, D. Koller and S.K. Kim, Science 302 (2003) 249-55. 25. E. Segal, N. Friedman, N. Kaminski, A. Regev and D. Koller, Nat Genet 37 Suppl (2005) S38-45. 26. J. Quackenbush, N Engl J Med 354 (2006) 2463-72. 27. M.F. Shannon and S. Rao, Science 296 (2002) 666-9. 28. N.V. Taverner, J.C. Smith and F.C. Wardle, Genome Biol 5 (2004) 210. 29. E.M. Southern, J Mol Biol 98 (1975) 503-17. 30. J.C. Alwine, D.J. Kemp and G.R. Stark, Proc Natl Acad Sci U S A 74 (1977) 5350-4. 31. F.C. Kafatos, C.W. Jones and A. Efstratiadis, Nucleic Acids Res 7 (1979) 1541-52. 32. G. Hardiman, Pharmacogenomics 5 (2004) 487-502. 33. F.E. Ahmed, Expert Rev Mol Diagn 6 (2006) 535-50. 34. P.K. Wolber, P.J. Collins, A.B. Lucas, A. De Witte and K.W. Shannon, Methods Enzymol 410 (2006) 28-57. 35. D.D. Dalma-Weiszhausz, J. Warrington, E.Y. Tanimoto and C.G. Miyada, Methods Enzymol 410 (2006) 3-28. 36. M.B. Eisen and P.O. Brown, Methods Enzymol 303 (1999) 179-205. 37. D.B. Allison, X. Cui, G.P. Page and M. Sabripour, Nat Rev Genet 7 (2006) 55-65. 38. M. Ashburner, C.A. Ball, J.A. Blake, D. Botstein, H. Butler, J.M. Cherry, A.P. Davis, K. Dolinski, S.S. Dwight, J.T. Eppig, M.A. Harris, D.P. Hill, L. Issel-Tarver, A. Kasarskis, S. Lewis, J.C. Matese, J.E. Richardson, M. Ringwald, G.M. Rubin and G. Sherlock, Nat Genet 25 (2000) 25-9. 39. A. Subramanian, P. Tamayo, V.K. Mootha, S. Mukherjee, B.L. Ebert, M.A. Gillette, A. Paulovich, S.L. Pomeroy, T.R. Golub, E.S. Lander and J.P. Mesirov, Proc Natl Acad Sci U S A 102 (2005) 15545-50. 40. P.K. Tan, T.J. Downey, E.L. Spitznagel, Jr., P. Xu, D. Fu, D.S. Dimitrov, R.A. Lempicki, B.M. Raaka and M.C. Cam, Nucleic Acids Res 31 (2003) 5676-84. 41. J. Zhang, R.P. Finney, R.J. Clifford, L.K. Derr and K.H. Buetow, Genomics 85 (2005) 297-308. 42. J.E. Larkin, B.C. Frank, H. Gavras, R. Sultana and J. Quackenbush, Nat Methods 2 (2005) 337-44. 43. R.A. Irizarry, D. Warren, F. Spencer, I.F. Kim, S. Biswal, B.C. Frank, E. Gabrielson, J.G. Garcia, J. Geoghegan, G. Germino, C. Grifﬁn, S.C. Hilmer, E. Hoffman, A.E. Jedlicka, E. Kawasaki, F. Martinez-Murillo, L. Morsberger, H. Lee, D. Petersen, J. Quackenbush, A. Scott, M. Wilson, Y. Yang, S.Q. Ye and W. Yu, Nat Methods 2 (2005) 345-50. 44. L. Shi, L.H. Reid, W.D. Jones, R. Shippy, J.A. Warrington, S.C. Baker, P.J. Collins, F. de Longueville, E.S. Kawasaki, K.Y. Lee, Y. Luo, Y.A. Sun, J.M. Willey, R.A. Setterquist, G.M. Fischer, W. Tong, Y.P. Dragan, D.J. Dix, F.W. Frueh, F.M. Goodsaid, D. Herman, R.V. Jensen, C.D. Johnson, E.K. Lobenhofer, R.K. Puri, U. Schrf, J. Thierry-Mieg, C. Wang, M. Wilson, P.K. Wolber, L. Zhang, W. Slikker, Jr., L. Shi and L.H. Reid, Nat Biotechnol 24 (2006) 1151-61. 45. O. Larsson, K. Wennmalm and R. Sandberg, Omics 10 (2006) 381-97. 46. A. Brazma, P. Hingamp, J. Quackenbush, G. Sherlock, P. Spellman, C. Stoeckert, J. Aach, W. Ansorge, C.A. Ball, H.C. Causton, T. Gaasterland, P. Glenisson, F.C. Holstege, I.F. Kim, V. Markowitz, J.C. Matese, H. Parkinson, A. Robinson, U. Sarkans, S. Schulze-Kremer, J. Stewart, R. Taylor, J. Vilo and M. Vingron, Nat Genet 29 (2001) 365-71. 47. C.A. Ball, G. Sherlock, H. Parkinson, P. Rocca-Sera, C. Brooksbank, H.C. Causton, D. Cavalieri, T. Gaasterland, P. Hingamp, F. Holstege, M. Ringwald, P. Spellman, C.J. Stoeckert, Jr., J.E. Stewart, R. Taylor, A. Brazma and J. Quackenbush, Science 298 (2002) 539. 48. T. Barrett, D.B. Troup, S.E. Wilhite, P. Ledoux, D. Rudnev, C. Evangelista, I.F. Kim, A. Soboleva, M. Tomashevsky and R. Edgar, Nucleic Acids Res 35 (2007) D760-5. 49. H. Parkinson, M. Kapushesky, M. Shojatalab, N. Abeygunawardena, R. Coulson, A. Farne, E. Holloway, N. Kolesnykov, P. Lilja, M. Lukk, R. Mani, T. Rayner, A. Sharma, E. William, U. Sarkans and A. Brazma, Nucleic Acids Res 35 (2007) D747-50. 136 50. O. Larsson and R. Sandberg, Nat Biotechnol 24 (2006) 1322-3. 51. I. Issemann and S. Green, Nature 347 (1990) 645-50. 52. W. Wahli, O. Braissant and B. Desvergne, Chem Biol 2 (1995) 261-6. 53. S. Mandard, M. Müller and S. Kersten, Cell Mol Life Sci 61 (2004) 393-416. 54. P. Lefebvre, G. Chinetti, J.C. Fruchart and B. Staels, J Clin Invest 116 (2006) 571-80. 55. C. Dreyer, G. Krey, H. Keller, F. Givel, G. Helftenbein and W. Wahli, Cell 68 (1992) 879. 56. S. Kersten, S. Mandard, P. Escher, F.J. Gonzalez, S. Tafuri, B. Desvergne and W. Wahli, FASEB J. 15 (2001) 1971-1978. 57. D. Patsouris, S. Mandard, P.J. Voshol, P. Escher, N.S. Tan, L.M. Havekes, W. Koenig, W. Marz, S. Tafuri, W. Wahli, M. Müller and S. Kersten, J Clin Invest 114 (2004) 94-103. 58. S. Kersten, J. Seydoux, J.M. Peters, F.J. Gonzalez, B. Desvergne and W. Wahli, J. Clin. Invest. 103 (1999) 1489-1498. 59. D. Patsouris, J.K. Reddy, M. Müller and S. Kersten, Endocrinology 147 (2006) 1508-1516. 60. M. Cherkaoui-Malki, K. Meyer, W.Q. Cao, N. Latruffe, A.V. Yeldandi, M.S. Rao, C.A. Bradﬁeld and J.K. Reddy, Gene Expr 9 (2001) 291-304. 61. H.K. Hamadeh, P.R. Bushel, S. Jayadev, K. Martin, O. DiSorbo, S. Sieber, L. Bennett, R. Tennant, R. Stoll, J.C. Barrett, K. Blanchard, R.S. Paules and C.A. Afshari, Toxicol Sci 67 (2002) 219-31. 62. L. Guo, H. Fang, J. Collins, X.H. Fan, S. Dial, A. Wong, K. Mehta, E. Blann, L. Shi, W. Tong and Y.P. Dragan, BMC Bioinformatics 7 Suppl 2 (2006) S18. References

63. B. Staels and J.C. Fruchart, Diabetes 54 (2005) 2460-70. 64. K.S. Frederiksen, E.M. Wulff, P. Sauerberg, J.P. Mogensen, L. Jeppesen and J. Fleckner, J. Lipid Res. 45 (2004) 592-601. 65. L. Richert, C. Lamboley, C. Viollon-Abadie, P. Grass, N. Hartmann, S. Laurent, B. Heyd, G. Mantion, S.D. Chibout and F. Staedtler, Toxicol Appl Pharmacol 191 (2003) 130-46. 66. J.E. Klaunig, M.A. Babich, K.P. Baetcke, J.C. Cook, J.C. Corton, R.M. David, J.G. DeLuca, D.Y. Lai, R.H. McKee, J.M. Peters, R.A. Roberts and P.A. Fenner-Crisp, Crit Rev Toxicol 33 (2003) 655-780. 67. C. Cheung, T.E. Akiyama, J.M. Ward, C.J. Nicol, L. Feigenbaum, C. Vinson and F.J. Gonzalez, Cancer Res 64 (2004) 3849-54. 68. K. Morimura, C. Cheung, J.M. Ward, J.K. Reddy and F.J. Gonzalez, Carcinogenesis 27 (2006) 1074-80. 69. N.F. Cariello, E.H. Romach, H.M. Colton, H. Ni, L. Yoon, J.G. Falls, W. Casey, D. Creech, S.P. Anderson, G.R. Benavides, D.J. Hoivik, R. Brown and R.T. Miller, Toxicol Sci 88 (2005) 250-64. 70. M. Bünger, H.M. van den Bosch, J. van der Meijde, S. Kersten, G.J.E.J. Hooiveld and M. Müller, Physiol Genomics 30 (2007) 192-204. 71. J.G. Granneman, P. Li, Z. Zhu and Y. Lu, Am J Physiol Endocrinol Metab 289 (2005) E608-16. 72. B.N. Finck, C. Bernal-Mizrachi, D.H. Han, T. Coleman, N. Sambandam, L.L. LaRiviere, J.O. Holloszy, C.F. Semenkovich and D.P. Kelly, Cell Metabolism 1 (2005) 133. 73. A.T. De Souza, P.D. Cornwell, X. Dai, M.J. Caguyong and R.G. Ulrich, Toxicol Sci 92 (2006) 578-86. 74. S. Hasmall, G. Orphanides, N. James, W. Pennie, K. Hedley, A. Soames, I. Kimber and R. Roberts, Toxicol Sci 68 (2002) 304-13. 75. K. Yamazaki, J. Kuromitsu and I. Tanaka, Biochem Biophys Res Commun 290 (2002) 1114-22. 76. F. Yadetie, A. Laegreid, I. Bakke, W. Kusnierczyk, J. Komorowski, H.L. Waldum and A.K. Sandvik, Physiol Genomics 15 (2003) 9-19. 77. E.S. Tien, J.P. Gray, J.M. Peters and J.P. Vanden Heuvel, Cancer Res 63 (2003) 5767-80. 78. E.S. Tien, J.W. Davis and J.P. Vanden Heuvel, J Biol Chem 279 (2004) 24053-63. 79. Y. Yang, S.J. Abel, R. Ciurlionis and J.F. Waring, Pharmacogenomics 7 (2006) 177-86. 80. A.T. De Souza, X. Dai, A.G. Spencer, T. Reppen, A. Menzie, P.L. Roesch, Y. He, M.J. Caguyong, S. Bloomer, H. Herweijer, J.A. Wolff, J.E. Hagstrom, D.L. Lewis, P.S. Linsley and R.G. Ulrich, Nucl. Acids Res. (2006) gkl609. 81. K. Motojima and T. Hirai, Febs J 273 (2006) 292-300. 82. S. Narravula and S.P. Colgan, J Immunol 166 (2001) 7543-7548. 83. P. Li, Z. Zhu, Y. Lu and J.G. Granneman, Am J Physiol Endocrinol Metab 289 (2005) E617-26. 84. H.A. Hostetler, A.B. Kier and F. Schroeder, Biochemistry 45 (2006) 7669-81. 85. B.M. Forman, J. Chen and R.M. Evans, PNAS 94 (1997) 4312-4317. 86. S.A. Kliewer, S.S. Sundseth, S.A. Jones, P.J. Brown, G.B. Wisely, C.S. Koble, P. Devchand, W. Wahli, T.M. Willson, J.M. Lenhard and J.M. Lehmann, PNAS 94 (1997) 4318-4323. 87. G. Krey, O. Braissant, F. L’Horset, E. Kalkhoven, M. Perroud, M.G. Parker and W. Wahli, Mol Endocrinol 11 (1997) 779-91. 88. P. Escher, O. Braissant, S. Basu-Modak, L. Michalik, W. Wahli and B. Desvergne, Endocrinology 142 (2001) 4195-4202. 137 89. O. Braissant, F. Foufelle, C. Scotto, M. Dauca and W. Wahli, Endocrinology 137 (1996) 354-66. 90. K. Motojima, P. Passilly, J.M. Peters, F.J. Gonzalez and N. Latruffe, J. Biol. Chem. 273 (1998) 16710-16714. 91. K. Motojima, Eur J Biochem 271 (2004) 4141-6. 92. C.C. Bastie, T. Hajri, V.A. Drover, P.A. Grimaldi and N.A. Abumrad, Diabetes 53 (2004) 2209-2216. 93. C.T. Phan and P. Tso, Front Biosci 6 (2001) D299-319. 94. S. Lee, T. Pineau, J. Drago, E. Lee, J. Owens, D. Kroetz, P. Fernandez-Salguero, H. Westphal and F. Gonzalez, Mol. Cell. Biol. 15 (1995) 3012-3022. 95. N. Flint, F.L. Cove and G.S. Evans, Biochem J 280 (1991) 331-4. 96. R.C. Gentleman, V.J. Carey, D.M. Bates, B. Bolstad, M. Dettling, S. Dudoit, B. Ellis, L. Gautier, Y. Ge, J. Gentry, K. Hornik, T. Hothorn, W. Huber, S. Iacus, R. Irizarry, F. Leisch, C. Li, M. Maechler, A.J. Rossini, G. Sawitzki, C. Smith, G. Smyth, L. Tierney, J.Y. Yang and J. Zhang, Genome Biol 5 (2004) R80. 97. Z. Wu, R.A. Irizarry, R. Gentleman, F. Martinez-Murillo and F. Spencer, Journal of the American Statistical Association 99 (2004) 909-917. 98. G.K. Smyth. in (R. Gentleman, V.C., S. Dudoit, R. Irizarry, W. Huber ed.) Bioinformatics and Computational Biology Solutions using R and Bioconductor, Springer, New York 2005, pp. 397-420. 99. J.D. Storey and R. Tibshirani, Proc Natl Acad Sci U S A 100 (2003) 9440-9445. 100. H.K. Lee, W. Braynen, K. Keshav and P. Pavlidis, BMC Bioinformatics 6 (2005) 269. 101. X. Wang and B. Seed, Nucl. Acids Res. 31 (2003) e154. 102. C. Moolenbeek and E.J. Ruitenberg, Lab Anim 15 (1981) 57-9. 103. R.P. Ferraris, S.A. Villenas and J. Diamond, Am J Physiol Gastrointest Liver Physiol 262 (1992) G1047-1059. 104. S. Robine, C. Huet, R. Moll, C. Sahuquillo-Merino, E. Coudrier, A. Zweibaum and D. Louvard, PNAS 82 (1985) 8488-8492. 105. T.S. Stappenbeck, J.C. Mills and J.I. Gordon, PNAS 100 (2003) 1004-1009. 106. M. Chen, Y. Yang, E. Braunstein, K.E. Georgeson and C.M. Harmon, Am J Physiol Endocrinol Metab 281 (2001) E916-923.

107. A. Stahl, D.J. Hirsch, R.E. Gimeno, S. Punreddy, P. Ge, N. Watson, S. Patel, M. Kotler, A. Raimondi, L.A. Tartaglia and H.F. Lodish, Mol Cell 4 (1999) 299-308. 108. S. Narisawa, L. Huang, A. Iwasaki, H. Hasegawa, D.H. Alpers and J.L. Millan, Mol. Cell. Biol. 23 (2003) 7525-7530. 109. M.H. Wong, P. Oelkers, A.L. Craddock and P.A. Dawson, J Biol Chem 269 (1994) 1340-7. 110. S.E. Calvano, W. Xiao, D.R. Richards, R.M. Felciano, H.V. Baker, R.J. Cho, R.O. Chen, B.H. Brownstein, J.P. Cobb, S.K. Tschoeke, C. Miller-Graziano, L.L. Moldawer, M.N. Mindrinos, R.W. Davis, R.G. Tompkins and S.F. Lowry, Nature 437 (2005) 1032-7. 111. Y. Jia, C. Qi, Z. Zhang, T. Hashimoto, M.S. Rao, S. Huyghe, Y. Suzuki, P.P. Van Veldhoven, M. Baes and J.K. Reddy, J. Biol. Chem. 278 (2003) 47232-47239. 112. E.F. Johnson, M.-H. Hsu, U. Savas and K.J. Griffin, Toxicology 181-182 (2002) 203. 113. F. Djouadi, C.J. Weinheimer, J.E. Saffitz, C. Pitchford, J. Bastin, F.J. Gonzalez and D.P. Kelly, J Clin Invest 102 (1998) 1083-91. 114. T.C. Leone, C.J. Weinheimer and D.P. Kelly, Proc Natl Acad Sci U S A 96 (1999) 7473-8. 115. T. Claudel, B. Staels and F. Kuipers, Arterioscler Thromb Vasc Biol 25 (2005) 2020-30. 116. S.M. Houten, M. Watanabe and J. Auwerx, Embo J 25 (2006) 1419-25. 117. T. Inagaki, A. Moschetta, Y.-K. Lee, L. Peng, G. Zhao, M. Downes, R.T. Yu, J.M. Shelton, J.A. Richardson, J.J. Repa, D.J. Mangelsdorf and S.A. Kliewer, PNAS 103 (2006) 3920-3925. 118. M. Griffioen, W.T. Steegenga, I.J. Ouwerkerk, L.T. Peltenburg, A.G. Jochemsen and P.I. Schrier, Mol Immunol 35 (1998) 829-35. 119. J.D. Watson, S.K. Oster, M. Shago, F. Khosravi and L.Z. Penn, J Biol Chem 277 (2002) 36921-30. 120. S.K. Oster, C.S. Ho, E.L. Soucie and L.Z. Penn, Adv Cancer Res 84 (2002) 81-154. 121. M. van de Wetering, E. Sancho, C. Verweij, W. de Lau, I. Oving, A. Hurlstone, K. van der Horn, E. Batlle, D. Coudreuse, A.P. Haramis, M. Tjon-Pon-Fong, P. Moerer, M. van den Born, G. Soete, S. Pals, M. Eilers, R. Medema and H. Clevers, Cell 111 (2002) 241-50. 122. W.C. Earnshaw, L.M. Martins and S.H. Kaufmann, Annu Rev Biochem 68 (1999) 383-424. 123. R.A. Roberts, N.H. James, N.J. Woodyatt, N. Macdonald and J.D. Tugwood, Carcinogenesis 19 (1998) 43-8. 138 124. S.C. Hasmall, N.H. James, N. Macdonald, F.J. Gonzalez, J.M. Peters and R.A. Roberts, Mutat Res 448 (2000) 193-200. 125. T. Nakajima, Y. Kamijo, N. Tanaka, E. Sugiyama, E. Tanaka, K. Kiyosawa, Y. Fukushima, J.M. Peters, F.J. Gonzalez and T. Aoyama, Hepatology 40 (2004) 972-80. 126. C.H. Yeh, T.P. Chen, C.H. Lee, Y.C. Wu, Y.M. Lin and P.J. Lin, Shock 26 (2006) 262-70. 127. F. Radtke and H. Clevers, Science 307 (2005) 1904-1909. 128. T.F. Bullen, S. Forrest, F. Campbell, A.R. Dodson, M.J. Hershman, D.M. Pritchard, J.R. Turner, M.H. Montrose and A.J. Watson, Lab Invest 86 (2006) 1052-63. 129. P.R. Devchand, H. Keller, J.M. Peters, M. Vazquez, F.J. Gonzalez and W. Wahli, Nature 384 (1996) 39-43. 130. P. Delerive, K. De Bosscher, S. Besnard, W. Vanden Berghe, J.M. Peters, F.J. Gonzalez, J.C. Fruchart, A. Tedgui, G. Haegeman and B. Staels, J Biol Chem 274 (1999) 32048-54. 131. P. Delerive, P. Gervois, J.C. Fruchart and B. Staels, J Biol Chem 275 (2000) 36703-7. 132. R.M. Hershberg, D.H. Cho, A. Youakim, M.B. Bradley, J.S. Lee, P.E. Framson and G.T. Nepom, J Clin Invest 102 (1998) 792-803. 133. S. Strobel and A.M. Mowat, Immunol Today 19 (1998) 173-81. 134. R.M. Hershberg and L.F. Mayer, Immunol Today 21 (2000) 123-8. References

135. O. Ahrenstedt, L. Knutson, B. Nilsson, K. Nilsson-Ekdahl, B. Odlind and R. Hallgren, N Engl J Med 322 (1990) 1345-9. 136. A. Andoh, Y. Fujiyama, T. Bamba, S. Hosoda and W.R. Brown. in (Mestecky, J., Russell, M.W., Jackson, S., Michaelek, S.M., Maskalova-Hogenova, H. and Strezc, J., eds.) Advances in Mucosal Immunology, Plenum, New York 1995, pp. 211-215. 137. V. Steimle, C.A. Siegrist, A. Mottet, B. Lisowska-Grospierre and B. Mach, Science 265 (1994) 106-109. 138. D. Kelly, J.I. Campbell, T.P. King, G. Grant, E.A. Jansson, A.G.P. Coutts, S. Pettersson and S. Conway, Nat Immunol 5 (2004) 104. 139. M. Sasaki, P. Jordan, T. Welbourne, A. Minagar, T. Joh, M. Itoh, J.W. Elrod and J.S. Alexander, BMC Physiol 5 (2005) 3. 140. J. Bassaganya-Riera and R. Hontecillas, Clin Nutr 25 (2006) 454-465. 141. J. Bassaganya-Riera, K. Reynolds, S. Martino-Catt, Y. Cui, L. Hennighausen, F. Gonzalez, J. Rohrer, A.U. Benninghoff and R. Hontecillas, Gastroenterology 127 (2004) 777-91. 142. H.E. Xu, M.H. Lambert, V.G. Montana, D.J. Parks, S.G. Blanchard, P.J. Brown, D.D. Sternbach, J.M. Lehmann, G.B. Wisely, T.M. Willson, S.A. Kliewer and M.V. Milburn, Mol Cell 3 (1999) 397-403. 143. L.M. Sanderson, P.J. de Groot, G.J. Hooiveld, A. Koppen, E. Kalkhoven, M. Müller and S. Kersten, PLoS ONE 3 (2008) e1681. 144. N.J. McKenna, R.B. Lanz and B.W. O’Malley, Endocr Rev 20 (1999) 321-44. 145. J.N. Feige and J. Auwerx, Trends Cell Biol 17 (2007) 292-301. 146. D.M. Lonard, R.B. Lanz and B.W. O’Malley, Endocr Rev 28 (2007) 575-87. 147. R. Stienstra, S. Mandard, D. Patsouris, C. Maass, S. Kersten and M. Müller, Endocrinology 148 (2007) 2753-2763. 148. R. Stienstra, S. Mandard, N.S. Tan, W. Wahli, C. Trautwein, T.A. Richardson, E. Lichtenauer-Kaligis, S. Kersten and M. Müller, Journal of Hepatology 46 (2007) 869-877. 149. S. Heber and B. Sick, Omics 10 (2006) 358-68. 150. M. Dai, P. Wang, A.D. Boyd, G. Kostov, B. Athey, E.G. Jones, W.E. Bunney, R.M. Myers, T.P. Speed, H. Akil, S.J. Watson and F. Meng, Nucleic Acids Res 33 (2005) e175. 151. G.K. Smyth, Statistical Applications in Genetics and Molecular Biology 3 (2004) http://www.bepress.com/sagmb/vol3/iss1/art3. 152. J.D. Storey and R. Tibshirani, Proc Natl Acad Sci U S A 100 (2003) 9440-5. 153. D. Greenbaum, N.M. Luscombe, R. Jansen, J. Qian and M. Gerstein, Genome Res 11 (2001) 1463-8. 154. J.L. Fink, R.N. Aturaliya, M.J. Davis, F. Zhang, K. Hanson, M.S. Teasdale, C. Kai, J. Kawai, P. Carninci, Y. Hayashizaki and R.D. Teasdale, Nucleic Acids Res 34 (2006) D213-7. 155. A. Pierleoni, P.L. Martelli, P. Fariselli and R. Casadio, Nucl. Acids Res. (2006) gkl775. 156. B.R. King and C. Guda, Genome Biol 8 (2007) R68. 157. L.N. Singh, L.S. Wang and S. Hannenhalli, Nucleic Acids Res 35 (2007) 7360-71. 158. E.H. Davidson, Genomic regulatory systems: development and evolution, Academic Press, San Diego, 2001. 159. M. Levine and E.H. Davidson, Proc Natl Acad Sci U S A 102 (2005) 4936-42. 160. M. Blanchette, A.R. Bataille, X. Chen, C. Poitras, J. Laganiere, C. Lefebvre, G. Deblois, V. Giguere, 139 V. Ferretti, D. Bergeron, B. Coulombe and F. Robert, Genome Res 16 (2006) 656-68. 161. S.M. Grimmond, K.C. Miranda, Z. Yuan, M.J. Davis, D.A. Hume, K. Yagi, N. Tominaga, H. Bono, Y. Hayashizaki, Y. Okazaki and R.D. Teasdale, Genome Res 13 (2003) 1350-9. 162. S. Kersten, S. Mandard, N.S. Tan, P. Escher, D. Metzger, P. Chambon, F.J. Gonzalez, B. Desvergne and W. Wahli, J. Biol. Chem. 275 (2000) 28488-28493. 163. M.L. Reitman, Cell Metabolism 5 (2007) 405-407. 164. M.K. Badman, P. Pissios, A.R. Kennedy, G. Koukos, J.S. Flier and E. Maratos-Flier, Cell Metabolism 5 (2007) 426-437. 165. T. Inagaki, M. Choi, A. Moschetta, L. Peng, C.L. Cummins, J.G. McDonald, G. Luo, S.A. Jones, B. Goodwin, J.A. Richardson, R.D. Gerard, J.J. Repa, D.J. Mangelsdorf and S.A. Kliewer, Cell Metab 2 (2005) 217-25. 166. S. Chikano, K. Sawada, T. Shimoyama, S.I. Kashiwamura, A. Sugihara, K. Sekikawa, N. Terada, K. Nakanishi and H. Okamura, Gut 47 (2000) 779-86. 167. P. Garside, Gut 48 (2001) 6-7. 168. J.N. Feige, L. Gelman, L. Michalik, B. Desvergne and W. Wahli, Prog Lipid Res 45 (2006) 120-59. 169. S. Yu and J.K. Reddy, Biochim Biophys Acta 1771 (2007) 936-51. 170. J.C. Roach, K.D. Smith, K.L. Strobe, S.M. Nissen, C.D. Haudenschild, D. Zhou, T.J. Vasicek, G.A. Held, G.A. Stolovitzky, L.E. Hood and A. Aderem, Proc Natl Acad Sci U S A 104 (2007) 16245-50. 171. M. Rakhshandehroo, L.M. Sanderson, M. Matilainen, R. Stienstra, C. Carlberg, P.J. de Groot, M. and S. Kersten, PPAR Res 2007 (2007) 26839. 172. S. Kersten, PPAR Res 2008 (2008) 132960. 173. F.J. Gonzalez and Y.M. Shah, Toxicology 246 (2008) 2-8. 174. M.A. Hediger, M.F. Romero, J.B. Peng, A. Rolfs, H. Takanaga and E.A. Bruford, Pﬂugers Arch 447 (2004) 465-8. 175. P. Borst and R.O. Elferink, Annu Rev Biochem 71 (2002) 537-92. . 176. V.J. Wacher, L. Salphati and L.Z. Benet, Adv Drug Deliv Rev 46 (2001) 89-102. 177. L.S. Kaminsky and Q.Y. Zhang, Drug Metab Dispos 31 (2003) 1520-5. 178. P.B. Danielson, Curr Drug Metab 3 (2002) 561-97. 179. R.M. Weinshilboum, D.M. Otterness, I.A. Aksoy, T.C. Wood, C. Her and R.B. Raftogianis, FASEB J. 11 (1997) 3-14. 180. E. Banoglu, Curr Drug Metab 1 (2000) 1-30. 181. R.H. Tukey and C.P. Strassburg, Annual Review of Pharmacology and Toxicology 40 (2000) 581-616. 182. J.A. Moscow and K.H. Dixon, Cytotechnology 12 (1993) 155-70. 183. K.D. Tew and Z.e. Ronai, Drug Resistance Updates 2 (1999) 143. 184. K.P. Vatsis, W.W. Weber, D.A. Bell, J.M. Dupret, D.A. Evans, D.M. Grant, D.W. Hein, H.J. Lin, U.A. Meyer, M.V. Relling and et al., Pharmacogenetics 5 (1995) 1-17. 185. M. Arand, A. Cronin, M. Adamska and F. Oesch, Methods Enzymol 400 (2005) 569-88. 186. H. Sampath and J.M. Ntambi, Annu Rev Nutr 25 (2005) 317-40. 187. E.T. Kennedy, S.A. Bowman and R. Powell, J Am Coll Nutr 18 (1999) 207-212. 188. J. Fu, S. Gaetani, F. Oveisi, J. Lo Verme, A. Serrano, F. Rodriguez De Fonseca, A. Rosengarth, H. Luecke, B. Di Giacomo, G. Tarzia and D. Piomelli, Nature 425 (2003) 90-3. 189. H. Kondo, Y. Minegishi, Y. Komine, T. Mori, I. Matsumoto, K. Abe, I. Tokimitsu, T. Hase and T. Murase, Am J Physiol Endocrinol Metab 291 (2006) E1092-1099. 190. T. Murase, A. Nagasawa, J. Suzuki, T. Wakisaka, T. Hase and I. Tokimitsu, J. Nutr. 132 (2002) 3018-3022. 191. H.M. van den Bosch, M. Bünger, P.J. de Groot, J. van der Meijde, G.J.E.J. Hooiveld and M. Müller, BMC Genomics 8 (2007) 267. 192. J.G. Bieri, J Nutr 110 (1980) 1726. 193. B. Ren, A.P. Thelen, J.M. Peters, F.J. Gonzalez and D.B. Jump, J Biol Chem 272 (1997) 26827-32. 194. K.H. Diehl, R. Hull, D. Morton, R. Pﬁster, Y. Rabemampianina, D. Smith, J.M. Vidal and C. van de Vorstenbosch, J Appl Toxicol 21 (2001) 15-23. 195. E. Compe, P. Drane, C. Laurent, K. Diderich, C. Braun, J.H. Hoeijmakers and J.M. Egly, Mol Cell Biol 25 (2005) 6065-76. 196. T. Hirai, Y. Fukui and K. Motojima, Biol Pharm Bull 30 (2007) 2185-90. 197. J. Shimakura, T. Terada, H. Saito, T. Katsura and K. Inui, Am J Physiol Gastrointest Liver Physiol 291 (2006) G851-6. 198. J.N. Feige, L. Gelman, D. Rossi, V. Zoete, R. Metivier, C. Tudor, S.I. Anghel, A. Grosdidier, C. Lathion, Y. Engelborghs, O. Michielin, W. Wahli and B. Desvergne, J. Biol. Chem. 282 (2007) 19152-19166. 199. S. Westin, R. Kurokawa, R.T. Nolte, G.B. Wisely, E.M. McInerney, D.W. Rose, M.V. Milburn, 140 M.G. Rosenfeld and C.K. Glass, Nature 395 (1998) 199-202. 200. B. Desvergne, I.J. A, P.R. Devchand and W. Wahli, J Steroid Biochem Mol Biol 65 (1998) 65-74. 201. T. Hashimoto, W.S. Cook, C. Qi, A.V. Yeldandi, J.K. Reddy and M.S. Rao, J Biol Chem 275 (2000) 28918-28. 202. S. Fourcade, S. Savary, S. Albet, D. Gauthe, C. Gondcaille, T. Pineau, J. Bellenger, M. Bentejac, A. Holzinger, J. Berger and M. Bugaut, Eur J Biochem 268 (2001) 3490-500. 203. S. Albet, C. Causeret, M. Bentejac, J.L. Mandel, P. Aubourg and B. Maurice, FEBS Lett 405 (1997) 394-7. 204. W.E. Roediger and S. Nance, Br J Exp Pathol 67 (1986) 773-82. 205. S. Cuzzocrea, R. Di Paola, E. Mazzon, T. Genovese, C. Muia, T. Centorrino and A.P. Caputi, Lab Invest 84 (2004) 1643-54. 206. R. Ringseis, S. Posel, F. Hirche and K. Eder, Pharmacol Res 56 (2007) 175-83. 207. N. van Vlies, S. Ferdinandusse, M. Turkenburg, R.J. Wanders and F.M. Vaz, Biochim Biophys Acta 1767 (2007) 1134-42. 208. Y. Kato, M. Sugiura, T. Sugiura, T. Wakayama, Y. Kubo, D. Kobayashi, Y. Sai, I. Tamai, S. Iseki and A. Tsuji, Mol Pharmacol 70 (2006) 829-837. 209. B. Newman, X. Gu, R. Wintle, D. Cescon, M. Yazdanpanah, X. Liu, V. Peltekova, M. Van Oene, C.I. Amos and K.A. Siminovitch, Gastroenterology 128 (2005) 260-9. 210. S. Waller, M. Tremelling, F. Bredin, L. Godfrey, J. Howson and M. Parkes, Gut 55 (2006) 809-814. References

211. G. D’Argenio, M. Calvani, A. Casamassimi, O. Petillo, S. Margarucci, M. Rienzo, I. Peluso, R. Calvani, A. Ciccodicola, N. Caporaso and G. Peluso, FASEB J. 20 (2006) 2544-2546. 212. M.A. Valasek, S.L. Clarke and J.J. Repa, J Lipid Res (2007). 213. J.N. van der Veen, J.K. Kruit, R. Havinga, J.F.W. Baller, G. Chimini, S. Lestavel, B. Staels, P.H.E. Groot, A.K. Groen and F. Kuipers, J. Lipid Res. 46 (2005) 526-534. 214. B.L. Knight, D.D. Patel, S.M. Humphreys, D. Wiggins and G.F. Gibbons, J Lipid Res 44 (2003) 2049-58. 215. G. Schmitz and T. Langmann, Curr Opin Lipidol 12 (2001) 129-40. 216. G.A. Francis, R.H. Knopp and J.F. Oram, J Clin Invest 96 (1995) 78-87. 217. J.D. Mulligan, M.T. Flowers, A. Tebon, J.J. Bitgood, C. Wellington, M.R. Hayden and A.D. Attie, J. Biol. Chem. 278 (2003) 13356-13366. 218. S. Murthy, E. Born, S.N. Mathur and F.J. Field, J. Lipid Res. 43 (2002) 1054-1064. 219. N. Mahmud and D.G. Weir, Eur J Gastroenterol Hepatol 13 (2001) 93-5. 220. F. Palmieri, Pflugers Arch 447 (2004) 689-709. . 221. K. Sheikh, G. Camejo, B. Lanne, T. Halvarsson, M.R. Landergren and N.D. Oakes, Am J Physiol Endocrinol Metab 292 (2007) E1157-1165. 222. M. Lordal, H. Wallen, P. Hjemdahl, O. Beck and P.M. Hellstrom, Clin Sci (Lond) 94 (1998) 663-70. 223. F. Martel, Pharmacol Res 54 (2006) 73-6. 224. M.C. Liu, Y. Sakakibara and C.C. Liu, Biochem Biophys Res Commun 254 (1999) 65-9. 225. J.K.L. Walker, R.R. Gainetdinov, A.W. Mangel, M.G. Caron and M.A. Shetzline, Am J Physiol Gastrointest Liver Physiol 279 (2000) G311-318. 226. H.C. Lee and Y.H. Wei, J Biomed Sci 7 (2000) 2-15. 227. D.M. Townsend, K.D. Tew and H. Tapiero, Biomed Pharmacother 57 (2003) 145-55. 228. S.A. Tariq, Mol Biotechnol 37 (2007) 62-5. 229. A.P. Halestrap and D. Meredith, Pflugers Arch 447 (2004) 619-28. 230. J.W. Jonker, M. Buitelaar, E. Wagenaar, M.A. Van Der Valk, G.L. Scheffer, R.J. Scheper, T. Plosch, F. Kuipers, R.P. Elferink, H. Rosing, J.H. Beijnen and A.H. Schinkel, Proc Natl Acad Sci U S A 99 (2002) 15649-54. 231. A. Haguenauer, S. Raimbault, S. Masscheleyn, M. del Mar Gonzalez-Barroso, F. Criscuolo, J. Plamondon, B. Miroux, D. Ricquier, D. Richard, F. Bouillaud and C. Pecqueur, J. Biol. Chem. 280 (2005) 22036-22043. 232. R.J. Wanders, W.F. Visser, C.W. van Roermund, S. Kemp and H.R. Waterham, Pflugers Arch 453 (2007) 719-734. 233. A.B. Rifkind, C. Lee, T.K. Chang and D.J. Waxman, Arch Biochem Biophys 320 (1995) 380-9. 234. H. Kawashima, E. Kusunose, C.M. Thompson and H.W. Strobel, Arch Biochem Biophys 347 (1997) 148-54. 235. S.J. Steinberg, S.J. Wang, D.G. Kim, S.J. Mihalik and P.A. Watkins, Biochemical and Biophysical Research Communications 257 (1999) 615-621. 236. R.T. Okita and J.R. Okita, Curr Drug Metab 2 (2001) 265-81. 237. M.B. Fisher, Y.M. Zheng and A.E. Rettie, Biochem Biophys Res Commun 248 (1998) 352-5. 238. S.D. Harmon, X. Fang, T.L. Kaduce, S. Hu, V. Raj Gopal, J.R. Falck and A.A. Spector, Prostaglandins Leukot Essent Fatty Acids 75 (2006) 169-77. 239. V. Le Quere, E. Plee-Gautier, P. Potin, S. Madec and J.P. Salaun, J Lipid Res 45 (2004) 1446-58. 240. L. Bartoloni and S.E. Antonarakis, Pflugers Arch 447 (2004) 780-3. 141 241. A.M. Pajor, Annu Rev Physiol 61 (1999) 663-82. 242. S.W. Altmann, H.R. Davis, Jr., L.-j. Zhu, X. Yao, L.M. Hoos, G. Tetzloff, S.P.N. Iyer, M. Maguire, A. Golovko, M. Zeng, L. Wang, N. Murgolo and M.P. Graziano, Science 303 (2004) 1201-1204. 243. E.M. Wright and E. Turk, Pflugers Arch 447 (2004) 510-8. . 244. M. Uldry and B. Thorens, Pflugers Arch 26 (2004) 26. 245. S. Tazawa, T. Yamato, H. Fujikura, M. Hiratochi, F. Itoh, M. Tomae, Y. Takemura, H. Maruyama, T. Sugiyama, A. Wakamatsu, T. Isogai and M. Isaji, Life Sci 76 (2005) 1039-50. 246. Y. Kanai and M.A. Hediger, Nature 360 (1992) 467-71. 247. M. Boll, H. Daniel and B. Gasnier, Pflugers Arch 447 (2004) 776-9. 248. F. Verrey, E.I. Closs, C.A. Wagner, M. Palacin, H. Endou and Y. Kanai, Pflugers Arch 447 (2004) 532-42. . 249. T. Ramadan, S.M. Camargo, V. Summa, P. Hunziker, S. Chesnov, K.M. Pos and F. Verrey, J Cell Physiol 206 (2006) 771-9. 250. D. Markovich and H. Murer, Pflugers Arch 447 (2004) 594-602. . 251. J.C. Boyer, C.E. Campbell, W.J. Sigurdson and S.M. Kuo, Biochem Biophys Res Commun 334 (2005) 150-6. 252. P. Krishnamurthy, D.D. Ross, T. Nakanishi, K. Bailey-Dell, S. Zhou, K.E. Mercer, B. Sarkadi, B.P. Sorrentino and J.D. Schuetz, J. Biol. Chem. 279 (2004) 24218-24225. 253. G.M. Reaven, Physiol Rev 75 (1995) 473-86. 254. D.E. Moller and K.D. Kaufman, Annu Rev Med 56 (2005) 45-62. 255. M.C. Carey, D.M. Small and C.M. Bliss, Annu Rev Physiol 45 (1983) 651-77. 256. J. Storch and B. Corsico, Annu Rev Nutr (2008). 257. I. Neeli, S.A. Siddiqi, S. Siddiqi, J. Mahan, W.S. Lagakos, B. Binas, T. Gheyi, J. Storch and C.M. Mansbach, II, J. Biol. Chem. 282 (2007) 17974-17984. 258. C.M. Mansbach, II and F. Gorelick, Am J Physiol Gastrointest Liver Physiol 293 (2007) G645-650. 259. T. Mori, H. Kondo, T. Hase, I. Tokimitsu and T. Murase, J Nutr 137 (2007) 2629-34. 260. H. Mu and C.-E. Hoy, Progress in Lipid Research 43 (2004) 105-133. 261. E. Duplus and C. Forest, Biochemical Pharmacology 64 (2002) 893. 262. D.B. Jump and S.D. Clarke, Annual Review of Nutrition 19 (1999) 63-90. 263. J.P. Pegorier, C. Le May and J. Girard, J Nutr 134 (2004) 2444S-2449S. 264. D.B. Jump, D. Botolin, Y. Wang, J. Xu, B. Christian and O. Demeure, J Nutr 135 (2005) 2503-6. 265. S. Otway and D.S. Robinson, J Physiol 190 (1967) 309-19. 266. S. Otway and D.S. Robinson, J Physiol 190 (1967) 321-32. 267. J. Borensztajn, M.S. Rone and T.J. Kotlar, Biochem J 156 (1976) 539-43. 268. M. Morillas, J. Clotet, B. Rubi, D. Serra, J. Arino, F.G. Hegardt and G. Asins, Biochem J 351 Pt 2 (2000) 495-502. 269. F.B. Johnson. in (Prophet, E.B., Mills, B., Arrington, J.B. and Sobin, L.H., eds.) Laboratory methods in histotechnology, American Registry of Pathology, Washington, D.C. 1992, pp. 177. 270. Y. Harano, K. Yasui, T. Toyama, T. Nakajima, H. Mitsuyoshi, M. Mimani, T. Hirasawa, Y. Itoh and T. Okanoue, Liver Int 26 (2006) 613-20. 271. T. Nagasawa, Y. Inada, S. Nakano, T. Tamura, T. Takahashi, K. Maruyama, Y. Yamazaki, J. Kuroda and N. Shibata, Eur J Pharmacol 536 (2006) 182-91. 272. G. Svegliati-Baroni, C. Candelaresi, S. Saccomanno, G. Ferretti, T. Bachetti, M. Marzioni, S. De Minicis, L. Nobili, R. Salzano, A. Omenetti, D. Pacetti, S. Sigmund, A. Benedetti and A. Casini, Am J Pathol 169 (2006) 846-60. 273. D. Linden, M. Alsterholm, H. Wennbo and J. Oscarsson, J Lipid Res 42 (2001) 1831-40. 274. K. Tordjman, C. Bernal-Mizrachi, L. Zemany, S. Weng, C. Feng, F. Zhang, T.C. Leone, T. Coleman, D.P. Kelly and C.F. Semenkovich, J Clin Invest 107 (2001) 1025-34. 275. H.M. de Vogel-van den Bosch, M. Bünger, P.J. de Groot, H. Bosch-Vermeulen, G.J. Hooiveld and M. Müller, BMC Genomics 9 (2008) 231. 276. J.W. Lee, P.J. Bajwa, M.J. Carson, D.R. Jeske, Y. Cong, C.O. Elson, C. Lytle and D.S. Straus, Gastroenterology 133 (2007) 108-23. 277. C. Duval, M. Müller and S. Kersten, Biochim Biophys Acta 1771 (2007) 961-71. 278. J.M. Campbell, G.C. Fahey, Jr., C.A. Lichtensteiger, S.J. Demichele and K.A. Garleb, J Nutr 127 (1997) 137-45. 279. D. Camuesco, J. Galvez, A. Nieto, M. Comalada, M.E. Rodriguez-Cabezas, A. Concha, J. Xaus and A. Zarzuelo, J Nutr 135 (2005) 687-94. 142 280. D. Camuesco, M. Comalada, A. Concha, A. Nieto, S. Sierra, J. Xaus, A. Zarzuelo and J. Galvez, Clin Nutr 25 (2006) 466-76. 281. H. Yuceyar, O. Ozutemiz, A. Huseyinov, M. Saruc, M. Alkanat, S. Bor, I. Coker and Y. Batur, Prostaglandins Leukot Essent Fatty Acids 61 (1999) 339-45. 282. K.D. Cashman and F. Shanahan, Eur J Gastroenterol Hepatol 15 (2003) 607-13. 283. B.G. Feagan, W.J. Sandborn, U. Mittmann, S. Bar-Meir, G. D’Haens, M. Bradette, A. Cohen, C. Dallaire, T.P. Ponich, J.W. McDonald, X. Hebuterne, P. Pare, P. Klvana, Y. Niv, S. Ardizzone, O. Alexeeva, A. Rostom, G. Kiudelis, J. Spleiss, D. Gilgen, M.K. Vandervoort, C.J. Wong, G.Y. Zou, A. Donner and P. Rutgeerts, Jama 299 (2008) 1690-7. 284. F.S. Collins, J. Rossant and W. Wurst, Cell 128 (2007) 9-13. 285. F.S. Collins, R.H. Finnell, J. Rossant and W. Wurst, Cell 129 (2007) 235-235. References

143 S

144 Summary

In the last decade it has become well accepted that various nutrients, such as fatty acids, not only serve as source of energy, but also can act as potent regulators of gene transcription. Moreover, there is increasing interest in the molecular mechanisms underlying the beneficial or adverse effects of foods and food components. Driven by the continuing and accelerating discoveries in omics technology, unique possibilities have emerged to investigate the genome-wide effects of nutrients at the molecular level. This research field of gene-nutrient interactions is called nutritional genomics or nutrigenomics, and encompasses the fields of biotechnology, genomics, molecular medicine and human nutrition. In chapter 1 these developments are summarized, the importance of the nuclear receptor peroxisome proliferator-activated receptor alpha (PPARα) in mediating effects of dietary fatty acids on gene expression is discussed, and studies that used a range of models and molecular tools combined with microarrays to study gene regulation by PPARα are reviewed to exemplify a successful nutrigenomics research strategy.

The primary function of the small intestine is the digestion of food and absorption of nutrients. Another important function is to prevent the translocation of bacteria and foreign antigens to extra-intestinal sites. In order to accomplish these roles the small intestine needs to function as a selective barrier. Enterocytes, one of the four cell types in the small intestinal epithelium, 145 are polarized cells with a basal nucleus and an apical brush border having tiny hairy structures called microvilli, and represent the dominant cell population of small intestinal villi. Enterocytes are responsible for the selective barrier function. Surprisingly, although few studies reported that PPARα is expressed in the small intestine, no attempts were made to characterize its exact location and function in this organ. Since the small intestine is frequently exposed to high levels of PPARα agonists we envisioned an important functional role of this transcription factor in this organ. In chapter 2 we therefore investigated in detail the expression of PPARα along the proximal-distal and the crypt-villus axis of the small intestine, and characterized the function of PPARα using functional genomics experiments and bioinformatics tools. We found that PPARα was expressed at high levels in both human and murine small intestine. Moreover, we demonstrated that PPARα was predominantly expressed in differentiated enterocytes (top villus cells), and its mRNA expression was highest in the proximal jejunum, which co-localized with other genes involved in fatty acid metabolism.

Utilizing genome-wide expression profiling in wild-type and PPARα-null mice in combination with a diet supplemented with the potent and specific agonist WY14643, we were the first to provide a comprehensive overview of processes controlled by PPARα in small intestine (chapter 2). We found that in addition to genes involved in fatty acid and triacylglycerol metabolism, transcription factors and enzymes connected to sterol and bile acid metabolism were specifically induced. In contrast, genes involved in cell cycle and differentiation, apoptosis, and host defense were repressed by PPARα activation. These results are functionally in line with the well-established roles of bile acids in absorption of dietary lipids, and were also corroborated by morphometric assessments. We conclude that PPARα plays an important role in the regulation of intestinal function by governing diverse processes ranging from numerous metabolic pathways to the control of apoptosis and cell cycle.

In chapter 3 we determined the commonalities and differences of PPARα activation in small intestine and liver. We identified intestinal and hepatic PPARα target genes that clustered into three sets of genes: those regulated in both organs, and those exclusively regulated in either small intestine or liver. At the level of corresponding cellular processes we found that in both tissues metabolism of lipids and energy-rich intermediates, organelle organization and biogenesis, and humoral immune responses were regulated. In contrast, specific for small intestine were antigen presentation via MHCII, B-lymphocyte proliferation, various cell signaling pathways, and bile acid metabolism. Liver specific processes were coagulation and wound healing, non-lymphocyte mediated immune response, non-lipid nutrient metabolism, and control of cell growth. Whereas for example c-myc was induced in the liver, which results in proliferation of hepatocytes, PPARα repressed its expression in the small intestine. To identify potentially secreted proteins regulated 146 by PPARα, we next performed a secretome analyses. Secreted proteins, including agonists and receptors, are critical to both short- and long-range intercellular signaling in multi-cellular organisms. PPARα regulated 326 genes encoding for secreted proteins in small intestine and liver, several of which were specific for small intestine or liver. The validity of our approach was demonstrated by the fact that several recently identified PPARα-regulated secreted proteins were picked-up, such as ANGPTL4, FGF15 and FGF21. Finally, we investigated the molecular basis of the observed commonalities and differences of PPARα activation. We investigated whether the tissue specificity could be related to nuclear co-regulators or transcription factors that are differentially expressed between intestine and liver. Of the studied co-regulators NR0B2 (SHP) and TRIP4 were higher expressed in liver compared to small intestine. We identified several combinations of transcription factors, so called cis-regulatory modules, that were enriched in the respective sets of regulated genes. Summary

Combined, our data provides a comprehensive overview of the differential effects of PPARα in small intestine and liver, we identiﬁed potentially secreted proteins that among others may be used as markers for intestinal or hepatic PPARα activity, and we provide clues that may underlay this tissue speciﬁcity.

It is generally accepted that PPARα is activated by fatty acids and their metabolites. As an extension of the experiments described in chapters 2 and 3, and to unambiguously demonstrate the nutritional relevance of intestinal PPARα, in chapter 4 we studied the effects of acute nutritional activation of PPARα focusing on expression of genes encoding intestinal barrier proteins. These enzymes are responsible for the selective absorption and metabolism of nutrients and other food constituents. We show that many barrier genes are PPARα-dependently regulated by dietary lipids, and functional implications inferred from our data suggested that nutrient- activated PPARα regulated barrier genes involved in fatty acid oxidation; cholesterol, glucose, and amino acid transport and metabolism; intestinal motility; and oxidative stress defence. This knowledge provides a better understanding of the impact dietary fat has on the barrier function of the gut, identiﬁes PPARα as an important factor controlling this key function, and underscores the importance of PPARα for nutrient-mediated gene regulation in the small intestine.

In chapter 5 the effect of PPARα activation on dietary lipid absorption was investigated. We show that PPARα dependent gene regulation of genes involved in intestinal dietary lipid metabolism is functionally translated into increased lipid uptake and transfer into the body. In line with these data, a dietary intervention study revealed that PPARα-null mice had an increased faecal fat secretion. Our data also demonstrate that in addition to sequestration and TG resynthesis, oxidation of fatty acids is an important mechanism to control the levels of these potential highly toxic metabolites. Taken together, we are the ﬁrst to show that PPARα regulates intestinal lipid uptake.

147 S

148 Samenvatting

Het doel van mijn onderzoeksproject dat ik bij de Voeding, Metabolisme & Genomics groep van de afdeling Humane Voeding, Wageningen Universiteit heb uitgevoerd, was het bestuderen van effecten die vetzuren (komend uit de voeding) hebben op expressie van genen in de dunne darm en het vertalen van deze effecten naar veranderde fysiologisch processen. Het is bekend dat de transcriptiefactor PPARα verschillen in concentraties van vetzuren kan omzetten in een veranderde genexpressie. Op deze manier kan een orgaan, of organisme, zich dynamisch aanpassen aan veranderde omstandigheden. In onze studies hebben we gebruik gemaakt van een diermodel waarin deze transcriptie factor niet meer aanwezig is. Een dergelijk muis model wordt ook wel een knockout muis genoemd, in ons geval hebben we dus gewerkt met een PPARα knockout muis. Door bestuderen van effecten van vetzuren in de darm van ‘normale’ muizen, en deze vervolgens te vergelijken met die in de PPARα knockout muizen, komen we allereerst te weten wat vetzuren in de darm doen, maar wordt ook onweerlegbaar de rol van PPARα hierin vastgesteld. Om een beter inzicht te krijgen over mijn onderzoek geef ik in het volgende stukje eerst wat achtergrond informatie om vervolgens de resultaten van de uitgevoerde studies samen te vatten en conclusies te trekken.

Achtergrond informatie 149 Alle genetische informatie van levende organismen ligt opgeslagen in het DNA. Het DNA bevindt zich binnen de cel in de celkern in de vorm van chromosomen. Op een chromosoom bevinden zich duizenden genen. Genen bestaan uit DNA-sequenties die gebruikt worden om boodschapper (‘messenger’) RNA (mRNA) te maken. mRNA wordt afgeschreven van het DNA (transcriptie). Na de transcriptie verlaten deze mRNA-molekulen de celkern en verplaatsen zich naar andere delen van de cel. Daar worden de RNA-moleculen vertaald in aminozuren (translatie), die in een lange keten een eiwit vormen. Eiwitten vervullen binnen en buiten de cel een zeer grote verscheidenheid aan biologische functies.

Genexpressie wordt gemeten door het bepalen van de hoeveelheid mRNA moleculen die van een gen zijn afgeschreven, en deze voorspelt het gehalte van het corresponderende eiwit in deze cel. Door de genexpressie te bepalen van een behandeling ten opzichte van de controle conditie, kan er bepaald worden welke genen door de behandeling verschillend (differentieel) tot expressie komen. Door gebruik te maken van zogenaamde microarrays kunnen wij in een experiment de expressie meten van allen genen die in een weefsel tot expressie komen. Dit wordt ook wel een transcriptomics analyse genoemd. Dit is een vrij unieke en nieuwe methode en op deze manier kan men in een keer het verschil in expressie van 15,000 genen bepalen.

PPARα is een transcriptie factor. Een transcriptie factor is een eiwit dat het afschrijven van genen controleert. Genen die door een bepaalde transcriptie factor worden veranderd hebben een DNA- bindingsplaats voor deze transcriptie factor in hun promoter gebied. Sommige transcriptiefactoren, zoals PPARα, worden door agonisten geactiveerd. Deze agonisten kunnen natuurlijk voorkomende componenten uit de voeding zijn zoals b.v. vetzuren, maar ook synthetische componenten zoals medicijnen kunnen transcriptie factoren activeren. Nadat een agonist een transcriptiefactor heeft gebonden en geactiveerd, gaat deze op de bindingsplaats van de genen zitten die hij reguleert en gaat deze afschrijven. Van deze genen word dus meer mRNA aangemaakt en vervolgens meer eiwit geproduceerd.

De belangrijkste functie van de dunne darm is het verteren van voeding en het opnemen van nutriënten uit de voeding. Ook heeft de darm een belangrijke functie in het afweersysteem doordat het een barrière vormt en zo de translocatie van bacterie en andere lichaamsvreemde antigenen naar het lichaam voorkomt. De dunne darm bestaat uit verschillende soorten cellen, waarvan de enterocyt de meest voorkomende celsoort is. Enterocyten zijn verantwoordelijk voor de opname van voedingsstoffen door de dunne darm.

Onderzoeksvraagstelling PPARα is een transcriptie factor die onder andere in de lever, de dunne darm en het hart aanwezig is. De afgelopen jaren is er veel onderzoek verricht naar de functie van PPARα in de lever. Er is 150 echter nog weinig bekend over de rol van PPARα in de dunne darm. Een belangrijke functie van de dunne darm is het opnemen van vet en andere nutriënten uit het dieet. Daardoor wordt de dunne darm veelvuldig blootgesteld aan zowel verzadigde als onverzadigde vetten. Vetzuren uit het dieet zijn activatoren voor PPARα. Dit betekent dat deze voedingsstoffen PPARα kunnen activeren en zo een effect op gen expressie kunnen hebben. Het doel van mijn onderzoek was de rol van PPARα in de darm te bestuderen, met name door het meten van verschillen in genexpressie, en vervolgens te bepalen welke fysiologische processen hierbij betrokken en veranderd zijn. Dit onderzoeksgebied waarin de gen-nutriënt interactie wordt bekeken heet nutrigenomics en maakt gebruik van technieken uit biotechnologie, genomics, biomedische wetenschappen, bioinformatica en humane voeding. Samenvatting

Resultaten In hoofdstuk 1 wordt een overzicht gegeven over de algemene functie van het maag-darmstel en wordt het concept van het nutrigenomics onderzoek uitgelegd. Verder worden studies uit de literatuur die gebruik maken van microarrays om de PPARα afhankelijke gen-regulatie te meten beschreven en hun resultaten bediscussieerd. Hieruit wordt geconcludeerd dat de microarray analyse tot nu toe alleen nog maar gebruikt was om een globaal inzicht te verkrijgen welke genen door PPPARα veranderd worden om dan vervolgens de regulatie van slechts één (of enkele genen) in detail te bestuderen. Dit lijkt vooral te komen omdat er tot voorkort weinig methoden bekend waren voor de post-microarray data analyse. Verder kan geconcludeerd worden dat er over de rol van PPARα in de dunne darm nog weinig bekend is.

In het hoofdstuk 2 bestuderen we de genoom-breede genregulatie door PPARα in de dunne darm. We laten eerst zien dat PPARα het hoogst tot expressie komt in villus cellen op de plek waar het vet in het lichaam wordt opgenomen. Vervolgens is een experimentele opzet gebruikt waarin PPARα met een synthetische agonist, genaamd WY14643, geactiveerd wordt. Op een genoom- breede schaal is er gekeken wat er met de genexpressie gebeurt. Omdat er naast het normale (wild-type) muizen ook muizen zijn gebruikt die geen functionele PPARα (PPARα-knockout) hebben, was het mogelijk de speciﬁeke effecten van PPARα na activering te bepalen (in de PPARα-null muizen gebeurt na activering op gen niveau helemaal niets). De verschillen in gen expressie na PPARα activatie hebben wij kunnen relateren aan processen die na PPARα activatie zijn veranderd. In de dunne darm reguleert PPARα het vetzuur en triacylglyceride metabolisme en induceert andere transcriptie factoren en de expressie van enzymen die betrokken zijn bij b.v. het sterol en galzuur metabolisme. Wij zien ook dat PPARα in de dunne darm gerelateerd is aan het afweersystem. Genen betrokken bij de cel cyclus en celdood komen lager tot expressie als PPARα geactiveerd is. Om aan te tonen dat deze effecten ook functioneel relevant zijn, laten wij zien dat door PPARα activatie de villi in wild-type muizen signiﬁcant langer zijn vergeleken met de controle conditie. Dit is gerelateerd aan het onderdrukken van celdood en het bevorderen van celdifferentiatie richting een enterocyt fenotype. Uit deze studie concluderen wij dat PPARα een 151 belangrijke rol heeft in de regulatie van gen expressie in de dunne darm.

Omdat PPARα zowel in de dunne darm als ook in de lever hoog tot expressie komt, waren wij geïnteresseerd in de overeenkomsten en verschillen in gen expressie na PPARα activering in deze twee organen. Dit wordt in hoofdstuk 3 bestudeerd. Er zijn genen gevonden die zowel in de lever als de darm door PPARα veranderd worden, andere worden alleen in de darm of juist alleen in de lever veranderd.. Zowel in de darm als lever reguleert PPARα genen die betrokken zijn bij het metabolisme van vet en energierijke tussenproducten en de organellen die hiervoor verantwoordelijk zijn. Op het niveau van veranderde processen vonden wij dat de darmspecifiek veranderde genen vooral betrokken zijn bij antigen presentatie via MHCII, lymfocyt proliferatie, verschillende cell signaling pathways en galzuur metabolisme. Leverspecifieke genen waren onder andere betrokken bij processen zoals bloed klontering, controle van de cel groei, non-lymfocyten gemediteerd immune response, non-vet nutriënt metabolisme, en de humorale immune response. Op zoek naar markereiwitten is er een secretome analyse uitgevoerd. Van de PPARα afhankelijk veranderde genen waren er 326 genen die coderen voor potentieel extracellulaire eiwitten. Sommigen genen zijn specifiek veranderd en kunnen allen door de darm of de lever uitgescheiden worden. Onder deze 326 genen vonden wij b.v. ANGPTL4, FGF15 and FGF21. Deze zijn recent geïdentificeerd en beschreven in literatuur als PPARα veranderde genen. Uiteindelijk hebben wij geprobeerd te achterhalen wat de moleculaire (transcriptionele) basis voor de gemeten verschillen zou kunnen zijn. We laten zien dat de verschillen in PPARα afhankelijke gen regulatie gerelateerd kan worden aan weefsel specifieke regulatie van transcriptie factoren of de expressie van weefsel specifieke nucleaire co-regulatoren die door WY14643 activatering gerekruteerd worden.

Omdat PPARα door vetzuren en hun metabolieten uit voeding geactiveerd kan worden is er in hoofdstuk 4 ervoor gekozen om te bestuderen of deze activering niet alleen in de reageerbuis gebeurt, maar mogelijk ook met een voedingsinterventie bereikt kan worden. In dit hoofdstuk is acute (6 uur) PPARα activering onderzocht. Hiervoor is PPARα activatie door sterke PPARα agonist WY14643 vergeleken met drie (meervoudig) onverzadigde vetzuren (oliezuur, EPA en DHA). In deze studie hebben we ons gericht op genen die coderen voor transport en barrière eiwitten en betrokken zijn bij de fase I/II metabolisme genen omdat deze groep van genen een belangrijke rol speelt in de fysiologie van de dunne darm. Van de bestudeerde genen uit deze dataset waren 7%, 16%, 17% en 28% PPARα afhankelijk veranderd door respectievelijk oliezuur, DHA, EPA en WY14643. Deze PPARα afhankelijk veranderden genen waren vooral betrokken bij 1) de vetzuur verbranding, 2) cholesterol, glucose en aminozuur metabolisme, 3) de beweegelijkheid (motiliteit) van de dunne darm en 4) oxidatieve stress.

152 In hoofdstuk 5 wordt door het uivoeren van functionele studies gekeken of de eerder gemeten PPARα afhankelijke gen regulatie vertaald kan worden naar fysiologisch meetbare parameters, zoals vet opname. In dit hoofdstuk laten wij zien dat differentiële gen expressie van genen betrokken bij vet metabolisme een verhoogde opname en transport van vet naar het lichaam ter gevolg hebben. Als PPARα geactiveerd wordt, is de vet opname hoger dan wanneer PPARα niet aanwezig of niet geactiveerd is. Bovendien laten wij zien dat na PPARα activering de verbranding van vrije vetzuren een mogelijk rol speelt bij het onschadelijk maken van deze potentiële toxische metabolieten. Samenvatting

Conclusies De conclusie van mijn onderzoek en de verschillende studies is dat PPARα in de dunne darm een belangrijke transcriptionele regulator is. Verschillende processen staan onder de controle van PPARα in dit orgaan.

Voor vet opname zijn functioneele studies uitgevoerd die laten zien dat het verschill dat wij meten op gen-expressie niveau vertaald wordt in een fysiologisch meetbare verschill in het opgenomen vet na PPARα activering. PPARα afhankelijke gen-expressie in de darm is daroom functioneel.

PPARα activering in de darm en dr lever resulteert in een verschillend en een overlappend groep van genen.

Een signiﬁcant gedeelte van de fase I/II metabolisme genen zijn na een korte behandeling met DHA en EPA PPARα afhankelijk veranderd.

Met behulp van microarray technologie (transcriptomics) en nieuwe bioinformatics data analyse programma’s hebben wij de rol van PPARα in de darm kunnen bestuderen.

153 A

154 Acknowledgements

...und fertig-Schluss-aus-und-Punkt! Het is Af! Finished!

(K) ein ganz normales Dankwort!

Dear all, while I am sitting here and enjoying the thought of being ready with this part of my research career, I wonder if I should write these very last words in German, Dutch or English. I am not sure, really. So I just start.

Michael, ich erinnere mich noch gut als ich zum ersten Mal bei dir ins Büro kam. Ich wollte mehr Informationen über ein laufendes IOP Gut Health Projekt, dass die Thematik „Fettsäure abhängige Genexpression“ beinhaltete. Ein ideales Projekt für eine Oecotrophologin! Ich war begeistert. Ich stand vor deiner Tür und Marie (lies: Maari) erwähnte beiläuﬁg, „du weißt doch sicher dass Professor Müller auch ein Deutscher ist, oder?“ Mhh, die Striche über dem ‘ü’ von Müller waren mir nicht entgangen - sie fehlten. Ein Deutscher Professor, kann ja spannend werden. Ich konnte gerade Deutsch, Englisch und Niederländisch ein wenig voneinander ‘unterscheiden’, aber dann muss mir keiner in einer Niederländischen Umgebung ﬂiessend Deutsch reden. Unser 155 Gespräch war sehr entspannt, aber spannend. Für das IOP Gut Health Projekt warst du noch auf der Suche nach einem, frei übersetzt „Schaf mit 5 Beinen“, ein Übermensch sozusagen, ambitiös wie immer. Nach unserem kurzen Gespräch, dessen Inhalt ich nicht mehr richtig weiss, stand für mich fest: Hier will ich arbeiten, in diesem neuen Forschungsgebiet Nutrigenomics. Und so geschah es dann auch und begonn ich wenig später bei Nutrition Metabolism & Genomics welches ein Teil der Humanernährung in Wageningen ausmacht.Vielen Dank für dein Vertrauen und die fortwährenden innovativen Ideen. Ich bin stolz Teil deines Forschungsteams zu sein und zu der ersten Generation von Nutrigenomics Forschern zu gehören. Look at the bright side of life. Mit Michael hat man Guido. Beste Guido, ergens in November 2001 kwam ik voor het eerst op de zesde verdieping je kantoor binnen lopen. Toen was alles nog heel erg kleinschalig en je hebt me even rondgeleid. Dat was dan ook zo gedaan. Kein zerstreuerter, abwesender nicht zu begreifender, sondern ein zutiefst bodenständiger Forscher mit vielen neuen Ideen. Afkomstig uit Groningen waar jij bent gepromoveerd, was je zelf net begonnen als postdoc en je weg aan het zoeken in Wageningen. Niet dat je Wageningen van vroeger niet kende. De Microarrays en alles erop en eraan is echt je passie geworden, daar werk je gedreven aan. Bedankt voor veel goede en nieuwe ideeën die je tijdens mijn AIO tijd hebt opgedaan (lees: dit betekende wel altijd weer veel werk voor mij, maar goed als een artikel er beter van wordt). Met veel energie heb je mij bij de laatste loodjes geholpen. Bekijk het resultaat en ben er ook trots op ;o). Beste Sander. Zonder jouw hulp en ideen was het PPARα werk best wel onmogelijk. Bedankt voor de steun tijdens ontalligen experimenten, schrijven en data analyzes. Je positieve insteek heeft mij erg goed gedaan. Vielen Dank!

Die lieve Heleen(e). Het klikte meteen, toen ik je zag zitten op de zesde verdieping van het Biotechnion (in je gele trui, naast jou de oranje backpack, in dezelfde kleur als je proefschrift). Je lachte vrolijk naar mij zoals altijd en was lekker aan het werk. Jij hebt mij de eerste dingen in het lab laten zien. Met name werken in de celkweek, RNA isolatie en qRT-PCR heb je mij uitgelegd en we hebben samen heerlijk onderzoek gedaan. Wat je al eerder hebt gezegd het is te veel om hier allemaal te noemen, maar ik wil je vanharte bedankt voor het er zijn. Als vriendin, maar ook als collega mis ik je nogal in Wageningen, maar Maastricht is eigenlijk ook om de hoek, toch? ;o) Ik verheug me op de dag van mijn promotie, samen met jou en Carla naast mij, kan er gewoon niets mis gaan!

Kerensa, door jou ben ik überhaupt op het idee gekomen om een promotie onderzoek in te gaan. Met je enthousiaste manier van onderzoek doen en je drang naar meer weten, heb je mij op 15B weten te inspireren. We hebben het er ondertussen zo vaak over gehad, en nu staat het hier zwart op wit en ga ik 12 September promoveren. Je goed gekeurde project heb je verdiend en ik ben erg trots op jou en blij dat ik jou heb ontmoet. Ik ben benieuwd hoe het alles zo verder gaat, bedankt 156 voor je vriendschap, energie en eerlijkheid.

Allerliefste Carla, mijn tweede paranimf. Heerlijk dingen doen, klimmen, naar de ﬁlm of de fads (hier was ik niet vaak aanwezig ;o)) of samen lunchen in de stad. Jij en Mark, jullie hebben altijd wel een idee of plan voor een gezellig uitje. Veel dank ervoor! Nu staan we samen in de Aula, ik kijk er naar uit. Beste Jocelijn, mijn nieuwe kamergenoot sinds Heleens vertrek. Heleen viel natuurlijk niet te vervangen, toch hebben wij een rustig en erg gezellige kamer 0068. Bedankt voor veel verhelderende gesprekken. Oh en nog eens sorry dat je plantje jouw vakantie niet heeft overleeft ;S. En poefff….toen verdween je naar Spanje voor 1.5 jaar of was je gewoon weggetoverd? Jolanda, wij hebben veel experimenten samen gedaan. Bedankt voor veel inzet in het labwerk en allen gesprekken bij de the en telefoontjes vanuit Spanje! Lieve Wilma, aan jou helder en gestructureerd manier van aanpak heb ik veel gehad in tijden dat het niet zo goed met mij ging. Bedankt voor je steun als ik die nodig had. Er zijn zeker niet veel worden nodig. Wanneer Acknowledgements

mag ik achterop de motor met je meerijden? Lieve Lydia, onderzoeker naar mijn smaak. Ik kan nog veel van jou leren. Mosbach 2004 was ons eerste congres waar wij samen naartoe gingen. Sindsdien zijn er velen gevolgd, en ik verheug me nu al op de NWO dagen 2008. Lekker de nacht doorkletsen over van alles en nog wat, maar vooral ook over onderzoek doen. Bedankt voor je rationele en vooral positieve benadering van dingen. We begrijpen elkaar, toch? Lieve Mechteld en Jenny, jullie moet men hebben als het gaat om goede microarrays te draaien. Bedankt voor jullie inzet en energie! Thank you Shohreh! The workaholic technician as she says herself. Always lovely and honestly concerned about everybody. Die Liebe Seele der Schaepmanstraat. Will we ever enjoy one of those charma days? Veel dank ook aan “Michael’s darmenclub” (Guido, Jolanda en Heleen, Roelof, Nicole, Hanneke en Els) voor alle discussies. Op naar nog veel meer vruchtbare discussies en baanbrekende ideeën. Dear Eileen, the Irish girl of CLA’s. Thanks for having been part of the NMG group. I absolutely enjoyed Ireland and the house of your parents. It has been a while, a lot has been happened, so when will we meet again? ;o)

Natuurlijk ook bijzonder bedankt aan de rest van de huidige NMGers en ook de voormalige NMGers voor een gezellige sfeertje, lunchpauzes in de zon (als ik erbij was), erg grappige NMG-uitjes (die veel te schaars zijn), Sinterklaas, kerst, afscheids- of vrijgezellen feestje(s), lab- opruim-dagen, enorme hoeveelheden taart (bei jeder Gelegenheit gibt’s Torte oder Kuchen!) die in de zomer in ijsjes transformeerden, snoep die er meestal niet meer lag toen ik kwam of ging, ik weet het niet meer….maar snel was het door Shohreh aangevuld. Hopelijk heb ik niet al te veel mensen vergeten, vergeeft mij (bitte). Volgens Prof. dr. Witkamp, of te wel Renger hoort het bij/ of na een zwangerschap erbij dingen te vergeten en volgens velen anderen “moeders” gaat het zelfs nooit meer weg. Het mag wel beter worden. Maar kortom: Allen NMGers bedankt voor een goede werksfeer!

Ook wil ik de Humane Voeding AIO’s bedanken voor de twee gezellige AIO reizen naar 157 Australië en Engeland, Ierland en Schotland. Voor mij was Australië mijn favoriete, aangezien ik deze samen met Andrea, Annet, Brian, Kristel en Maaike heb voorbereid. Maandag avond was ExCie avond. Bedankt voor de vergaderingen, etentjes in de mensa en het heerlijk ontspannen Heide-weekend met alle activiteiten. Wel wil ik nog zeggen: Annet-te-retette, last but zeker, zeker not least, kom op meid, zet hem op!! Dione, ook al zit je lekker thuis in de zon in je tuin, je bent altijd online (of niet soms?). Bedankt voor je snelle hulp met mijn website, trouwens bij deze www.bunger.nl is online. ;o).

Beste Teun, bedankt voor hulp met de radioactiviteitmetingen, de cappuccino’s en bizarre opmerkingen die ik soms niet begreep. Het was gezellig. Geniet nu maar van je vrije tijd. Betty en jou zie ik binnenkort wel eens verschijnen, anders steht und spricht unsere Kleine bald ;o). In ons onderzoeksgebied is onderzoek zonder het CKP onmogelijk en zo heeft het CKP zich onmisbaar gemaakt in mijn onderzoeksproject. Mijn dank gaat in het bijzondere uit naar Rene, Bert, Wilma en Judith. Met hun heb ik de meesten experimenten gedaan en met veel plezier heb ik altijd uitgekeken naar de dagen in de “kelder” (ik weet het is niet echt een kelder) om er vervolgens weer een jaar of langer met de data zoet te zijn. Bedankt voor jullie inzet! Graag wil ik ook de mensen van het IOP ‘Genomics’ Gut Health project bedanken en dan met name Rachelle en de AIO’s Bart, Eline, Kaatje, Milka en Rolf. Ik hoop wij lopen elkaar snel weer eens tegen het lijf. Eline dat zal bij jou in Australië wat minder toevallig kunnen gebeuren, super gaaf! De verschillen van onderzoeksgebieden makten het ons soms niet makkelijk elkaar een steun te zijn, toch ben ik blij jullie ontmoet te hebben. And yes, I really enjoyed discussing with you Milka, we should have done that more often! See you in AMS! Mijn student Maria van Noort bedankt voor het verrichten van labwerk, en het werken met kaders. Ondertussen verhuist en getrouwd, …hoe de tijd vergaat, ik ben blij dat ik je mocht begeleiden Maria. Zonder functionerende PC ben je toch ook nergens. Beste Harro bedankt voor je al je hulp bij kleine of grote PC probleempjes. Rood is een moeie kleur haar! Susan, mijn allerliefste ex-medebewonster. Nu kan het bijna weer, lekker voor een cocktailtje naar Arnhem en/of dansen in het Dollars, wat dan ook, tijd met jou is altijd gezellig.

Und zum Schluss: die Deutschen Freunde: ein Deutsches soziales Leben ist doch auch kein unwesentlicher Beitrag zum Erfolg! Ein deja-vu der Schulzeit, ja ja damals… früher… Meine liebe Eva-Maria (unser Evchen), liebe Melanie (die aus Gellenbeck) und die liebe Tanja (von der Nordstrasse, eeh Asternstrasse): Ich weiss nicht ob ihr euch noch erinnern könnt, aber Herr Triphaus ermahnte mich damals auf dem Gymi nach fast jeder Deutsch Klausur: Meike oder sagte er Frau Bünger?: “ …in der Kürze liegt die Würze… und diese Rechtschreibfehler (ein schüttelnder Kopf mit Blick nach unten) beim nächsten Mal gibt es dafür einen Punkt Abzug.” Und den gab es! Mich hat dann vor allem der erste Satzt oft und bis zum heutigen Tag vervolgt (für 158 den zweiten gibt es ja spelling control, wenn man die auf seinem PC installiert), die Anwendung und Umsetzung vom soeben genannt ﬁel und fällt mir leider immernoch schwer. Das Dankwort ist ein gutes Beispiel. Und auch meine Kollegen können davon sicherlich ein Liedchen singen. Ich bekam dann auch meine Manuskripte, recht “kontrolliert” (lies: röter) zurück. Reden konnte ich umso besser. Obwohl man mich des öfteren fälschlicher Weise des Schwatzens beschuldigte (Tanja erinnert sich sicherlich, stille Wasser sind tief kann ich nur sagen). Zum Thema Schuhe: Schuhe mache ich immernoch so richtig gerne zu, früher vor allem auch bei anderen und mit viel Hingabe. Melanie, nochmals gern geschehen “da-nich-für”. ;o). Weitere Knaller auf Parties, 1.Maifestlichkeiten, Schützenfesten, Zeltlagern, Urlauben, Karneval-parties oder Hochzeiten und all den anderen Blödsinn den wir früher so ausgefressen haben, erspare ich vor allem den restlichen Lesern, wenn die es überhaupt soweit geschafft haben. Die Frage für euch bleibt: “Und was haben wir nun an diesem Buch und der Doktorarbeit beigetragen?” Die Antwort ist ganz einfach: Ihr seid einfach immer für mich da! EINEN GANKZ LIEBEN DANK! Acknowledgements

Das gilt im übrigen auch für eure wunderbaren Männer, Johannes, Thorsten und Titus. Für uns hoffe ich, dass wir uns weiterhin und genauso regelmässig sehen wie die letzten +/-20 Jahre. Tanja ich sage nur: Aus der Ruhe kommt die Kraft. ;o) Herr Rehm lässt grüssen… Lieber Jens, liebe Alex, vielen Dank für die vielen Telefonate, solange Peter zahlt (http://www.peterzahlt.de) dürfte das kein Ende nehmen mit den Gesprächen. Ich freue mich schon drauf! Wenke nach 100 Jahren bist du nun endlich in Wageningen gewesen. War doch gar nicht so schwer und so weit ist das doch gar nicht von Holzbunge (Rendsburg). Ich fand’s echt super Klasse - wir brauchen wenig Worte - verstehen uns. Ein Urlaub zu zweit wäre auch mal wieder schön, oder?!

Allerliefste familie Heck, en dat zijn Jan, Annette, Martijn (PhD), Jeroen, Remco, Kathelijn en natuurlijk hoort Tjitske er ook bij. Ik voel me thuis tussen jullie en wil jullie bedanken voor allen discussies lekkere etentjes met Kerst, een ontspannen sfeer en de steun die jullie mij geven. Ik ben thuis. Apropos thuis: Liebe Mama, lieber Papa. So nun ist es dann soweit, eure Tochter bekommt den Doktortitel. Ein langer Weg von Kiel bis Wageningen über Bennekom hat hier dann ein Ende und eröffnet mir neue Türen. Wir werden sehen wo ich in ein paar Jahren bin. Ohne euch wäre ich in jedem Fall nicht hier. Vielen Dank dass ihr mir studieren ermöglicht habt. Eure immerwährende Kraft und Energie Jeroen und mir beizustehen und fast auf Abruf für uns da zu sein, um uns zu helfen bewundere ich. Einen ganz lieben und herzlichen Dank. Ich bin eine äusserst gelungene Kombination von euch beiden geworden, nicht wahr? Geniesst den Tag! Eine zweite nicht weniger gute Kombi bist du Thorsten. Mit deiner Familie bestehend aus Petra, Marvin und Marie-Lena machst du mir ein Heimkommen immer angenehm und ist immer was los auf der Kiewitsheide 27a.

Mein allerliebster Schatz der Jeroen, nu weet ik al helemaal niet meer welke taal ik zal gaan spreken. Wij hebben elkaar bijna 10 jaar gelden in de Rijnsteeg leren kennen. De ersten jaren 159 moesten wij nog van Wageningen naar Kiel of andersom op-en-neer reizen om elkaar te zien. Toch is dat vaak gelukt en heb ik van die tijd met jou in Kiel in mijn klein appartement genoten. New Zealand was onbeschrijfelijk mooi en een bijzonder ervaring. Samen reizen, ja dat is gaaf en doen wij snel weer. 2007 was voor ons in vele inzichten een heel bijzonder jaar. Uiteindelijk komt het toch allemaal weer goed! Was het echt maar een jaar? Wij meisteren echt elke situatie, dus problemen komt maar op. Ik ben trots op ons tween en onze dochter Liena. We zien wel wat ons de toekomst brengt. Bedankt voor je relativeringsvermogen en je manier mij te laten zijn zoals ik nu eenmal ben. Wij vullen elkaar heel goed aan! En nu ga jij je boekje afmaken! Een heel dikke kus!

Wir sehen uns alle am 12. September! See you all the 12th of September! Meike C

160 Curriculum Vitae

Meike Bünger was born October 31st, 1975 in Georgsmarienhuette (Germany). She completed secondary school at her hometown in 1995 (Gymnasium Oesede, Georgsmarienhuette, Germany) and started to study Oecotrophology at the Christian-Albrechts-University (CAU) of Kiel. During her studies she was an assistant to a scientist at the Faculty of Agricultural- and Nutritional science and performed a study abroad, in the Netherlands at the Wageningen Univertity, termination with the “European Master” of Science. In 2001 she graduated at the CAU in Kiel as Master of Science (Diplom) in Oecotrophology. In 2002 she started her PhD study which was part of the IOP project ‘An integrated genomics approach towards gut health’ and entitled “Probing the role of PPARa in the small intestine: A functional nutrigenomcis approach” at the Division of Human Nutrition of the Wageningen University. She worked at the chair of “Nutrition Metabolism and Genomics”. Her project was supervised by Prof. dr. M. Muller and dr. G.J.E.J. Hooiveld. Since March 2007 she is working as postdoc for the nutrigenomics consortium (NGC), the Netherlands.

161 L

162 List of Publications

Bünger M, van den Bosch HM, van der Meijde J, Kersten S, Hooiveld GJ, Müller M. Genome- wide analysis of PPARalpha activation in murine small intestine. Physiol Genomics. 2007 Jul 18;30(2):192-204. PMID: 17426115

Bünger M, Hooiveld GJ, Kersten S, Müller M. Exploration of PPAR functions by microarray technology--a paradigm for nutrigenomics. Biochim Biophys Acta. 2007 Aug;1771(8):1046-64. PMID: 17632033 van den Bosch HM, Bünger M, de Groot PJ, van der Meijde J, Hooiveld GJ, Müller M., Gene expression of transporters and phase I/II metabolic enzymes in murine small intestine during fasting. BMC Genomics. 2007 Aug 7;8:267. PMID: 17683626 de Vogel-van den Bosch HM, Bünger M, de Groot PJ, Bosch-Vermeulen H, Hooiveld GJ, Müller M. PPARalpha-mediated effects of dietary lipids on intestinal barrier gene expression. BMC Genomics. 2008 May 19;9:231. PMID: 18489776

Bünger M, de Groot PJ, Sanderson L , Singh L, Bosma M, Hannenhalli S, Kersten S, Müller M, 163 Hooiveld GJ. Organ-speciﬁc function of PPARα as revealed by gene expression proﬁling.

Bünger M, Müller M, Hooiveld GJ. PPARα regulates intestinal lipid absorption. E

164 Education

Overview of completed training activities

Courses VLAG PhD week 2002 BMT course 2002, Wageningen University BIT course 2003, Wageningen University Scientiﬁc writing 2005, Wageningen University Laboratory animal science 2003, Universiteit Utrecht Ecophysiology of the GI-tract 2003, VLAG CASCADE Summer course school on Nuclear Hormone Receptors 2004, Lyon, France FASEB SRC, Molecular Biology of Intestinal Lipid Transport and Metabolism 2006, USA Personal efﬁcacy 2007

Meetings Ernst Klenk symposium 2003, University of Cologne, Germany 3rd Nutrigenomics Masterclass in Wageningen, The Netherlands 1st Proteomics Masterclass in Maastricht, The Netherlands 165 Mosbach Colloquium 2004, Mosbach, Germany Digestive Disease Week, Chicago 2005, USA Digestive Disease Week, Los Angeles 2006, USA NWO Nutrition, Papendal 2002-2008, The Netherlands Annual NVGE days 2003-2006, Veldhoven, The Netherlands Annual Netherlands Gut meeting (“Darmendag”) 2005-2006, The Netherlands The European Nutrigenomics Organisation (NuGO) week 2004, Wageningen, The Netherlands NuGO week 2005, Tuscany, Italie NuGO week 2007, Oslo, Norway PhD-tour 2004, Australia PhD-tour 2006, England, Ireland, Scotland Financial support by TI Food and Nutrition, Promege Benelux, 166 Affymetrix and the Wageningen University for printing of this thesis is gratefully acknowledged.

The research described in this thesis was supported by the Dutch Ministry of Economic Affairs through the innovation-oriented research program on genomics IOP-IGE01016 (2002-2006) and the Nutrigenomics Consortium (NGC) of the TI Food and Nutrition.

Printing of this thesis was ﬁnancially supported by the Section Experimental Gastroenterology (SEG) of the Dutch Society of Gastroenterology (NVGE).