Intestinal Cytoplasmic Lipid Droplets, Associated Proteins, and the Regulation of Dietary Fat Absorption Theresa M

Purdue University Purdue e-Pubs

Open Access Dissertations Theses and Dissertations

8-2016 Intestinal cytoplasmic lipid droplets, associated proteins, and the regulation of dietary fat absorption Theresa M. D'Aquila Purdue University

Follow this and additional works at: https://docs.lib.purdue.edu/open_access_dissertations Part of the Biochemistry Commons, Cell Biology Commons, Molecular Biology Commons, Nutrition Commons, and the Physiology Commons

Recommended Citation D'Aquila, Theresa M., "Intestinal cytoplasmic lipid droplets, associated proteins, and the regulation of dietary fat absorption" (2016). Open Access Dissertations. 747. https://docs.lib.purdue.edu/open_access_dissertations/747

This document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries. Please contact [email protected] for additional information. Graduate School Form

30 Updated ¡ ¢¡£ ¢¡¤ ¥

PURDUE UNIVERSITY GRADUATE SCHOOL Thesis/Dissertation Acceptance

This is to certify that the thesis/dissertation prepared

By Theresa Marie D'Aquila

Entitled INTESTINAL CYTOPLASMIC LIPID DROPLETS, ASSOCIATED PROTEINS, AND THE ABSORPTION OF DIETARY FAT.

For the degree of Doctor of Philosophy

Is approved by the final examining committee:

Kimberly K. Buhman Chair Jon A. Story

Kee- Hong Kim

Nana A. Gletsu-Miller

To the best of my knowledge and as understood by the student in the Thesis/Dissertation Agreement, Publication Delay, and Certification Disclaimer (Graduate School Form 32), this thesis/dissertation adheres to the provisions of Purdue University’s “Policy of Integrity in Research” and the use of copyright material.

Approved by Major Professor(s): Kimberly K. Buhman

Approved by: Connie M. Weaver 7/20/2016 Head of the Departmental Graduate Program Date

INTESTINAL CYTOPLASMIC LIPID DROPLETS, ASSOCIATED PROTEINS, AND

THE REGULATION OF DIETARY FAT ABSORPTION

A Dissertation

Submitted to the Faculty

Purdue University

Theresa M D’Aquila

In Partial Fulfillment of the

Requirements for the Degree

Doctor of Philosophy

August 2016

Purdue University

West Lafayette, Indiana

For my parents, brother, and my adventure buddy

iii

ACKNOWLEDGEMENTS

This dissertation would not have been possible without the help, guidance, and

encouragement of advisors, friends, and my family.

First and foremost, I would like to express my sincerest gratitude to my advisor

Dr. Kim Buhman. Her guidance and support helped me develop into the scientist that I

am today. Thank you for always being there ready to listen, give advice, and bullshit

science with me. I would also like to thank my academic advisory committee, Dr. Jon

Story, Dr. Kee Hong Kim, and Dr. Nana Gletsu-Miller. Your knowledge, advice, and encouragement have been invaluable through this process.

I would also like to thank the current and former lab members of the Buhman lab including, Amy Hung, Alicia Carreiro, Lenna Peterson and Dr. Aki Uchida. I’m grateful for your help, advice, and some good laughs.

Lastly, I would like to thank my climbing buddies, particularly Bryce Heckaman.

Nothing like beers around a campfire after a hard day of climbing to keep everything in perspective.

TABLE OF CONTENTS

Page LIST OF TABLES ...... x LIST OF FIGURES ...... xi LIST OF ABBREVIATIONS ...... xiii ABSTRACT ...... xv CHAPTER 1. INTRODUCTION ...... 1 1.1 Dietary fat, health, and disease ...... 1 1.1.1 Dietary fat structure and composition ...... 1 1.1.2 Dietary fat, fat soluble vitamins, and essential fatty acids ...... 2 1.1.3 Dietary fat and risk for disease ...... 2 1.2 Intestinal metabolism of dietary fat ...... 4 1.2.1 Digestion and uptake of dietary fat ...... 4 1.2.2 Fatty acid uptake and metabolism ...... 5 1.2.3 Fatty acid re-esterification ...... 7 1.2.4 Triacylglycerol synthesis ...... 10 1.2.5 Chylomicron synthesis and secretion ...... 10 1.2.6 Cytoplasmic lipid droplets and triacylglycerol storage ...... 12 1.3 Cytoplasmic lipid droplet metabolism ...... 13 1.3.1 General knowledge of CLD metabolism ...... 13 1.3.1.1 Physiological conditions promoting CLDs ...... 13 1.3.1.2 CLD composition ...... 13 1.3.1.2.1 Neutral and phospholipid composition ...... 14 1.3.1.2.2 General CLD associated proteins ...... 14 1.3.1.3 CLD connection with organelles ...... 16

Page 1.3.1.4 Roles of CLDs ...... 16 1.3.1.5 CLD synthesis ...... 18 1.3.1.6 CLD catabolism ...... 20 1.3.1.6.1 Adipose Triglyceride Lipase ...... 20 1.3.1.6.2 Comparative Gene Identification- 58 ...... 21 1.3.1.6.3 Hormone Sensitive Lipase ...... 21 1.3.1.6.4 Monoacylglycerol Lipase ...... 22 1.3.1.6.5 Putative methods of CLD catabolism ...... 22 1.3.2 CLD protein in enterocytes ...... 24 1.3.2.1 Physiological conditions promoting CLDs in enterocytes ...... 24 1.3.2.2 CLD composition ...... 25 1.3.2.2.1 Neutral and phospholipid composition ...... 25 1.3.2.2.2 Perilipins...... 26 1.3.2.3 CLDs may serve multiple roles in the enterocyte ...... 27 1.3.2.4 CLD synthesis ...... 28 1.3.2.5 CLD catabolism ...... 29 1.3.2.5.1 Adipose Triglyceride Lipase ...... 29 1.3.2.5.2 Comparative Gene Identification-58 ...... 30 1.3.2.5.3 Hormone Sensitive Lipase ...... 30 1.3.2.5.4 Monoacylglycerol Lipase ...... 31 1.3.2.5.5 Putative methods of CLD catabolism ...... 32 1.3.2.5.6 Autophagy ...... 33 1.4 Methods for identifying CLD associated proteins ...... 33 1.4.1 Introduction to identifying CLD associated proteins ...... 34 1.4.2 Isolation of CLD associated proteins ...... 34 1.4.3 Proteomics methods ...... 35 1.4.3.1 Targeted approach ...... 35 1.4.3.2 Global approach ...... 35 1.4.4 Confirmation of CLD associated proteins by imaging ...... 36

Page 1.5 Recent discoveries and new mysteries of CLD associated proteins ...... 37 1.5.1 General observation of newly identified CLD associated proteins ...... 37 1.5.2 Newly identified CLD associated proteins in enterocytes ...... 37 1.6 Effects of obesity on dietary fat absorption ...... 38 1.6.1 Whole animal/ physiological consequences ...... 38 1.6.2 Enterocyte specific consequences ...... 39 1.7 Conclusion ...... 40 1.8 References ...... 43 CHAPTER 2. CHARACTERIZATION OF THE PROTEOME OF CYTOPLASMIC LIPID DROPLETS IN MOUSE ENTEROCYTES AFTER A DIETARY FAT CHALLENGE ...... 57 2.1 Abstract ...... 57 2.2 Introduction ...... 58 2.3 Materials and methods ...... 60 2.3.1 Ethics statement ...... 60 2.3.2 Mice ...... 60 2.3.3 Mouse procedure for lipid droplet isolation and proteomic analysis ...... 60 2.3.4 Enterocyte Isolation ...... 61 2.3.5 CLD isolation ...... 61 2.3.6 TAG and protein analysis ...... 62 2.3.7 In solution digestion and LC MS/MS ...... 62 2.3.8 Data analysis and bioinformatics ...... 63 2.3.9 Immunofluorescence Imaging ...... 65 2.3.10 Transmission electron microscopy ...... 66 2.3.10.1 Cardiac perfusion fixation and TEM imaging of jejunum...... 66 2.3.10.2 TEM imaging of isolated CLDs ...... 66 2.3.10.3 Immunoelectron microscopy ...... 66 2.4 Results ...... 67

vii

Page 2.4.1 Lipid accumulates in CLDs in enterocytes in response to a dietary fat challenge...... 67 2.4.2 The CLD fraction is enriched with Plin3 and has a high TAG to protein ratio...... 67 2.4.3 Proteins were identified in the CLD fraction through LC MS/MS ...... 68 2.4.4 Some identified proteins in the CLD fraction are associated with known lipid metabolism related functions ...... 69 2.4.5 Immunofluorescence imaging of proteins identified in the CLD fraction supports their CLD localization ...... 70 2.5 Discussion ...... 72 2.6 Acknowledgements ...... 77 2.7 References ...... 87 CHAPTER 3. DIET INDUCED OBESITY ALTERS CYTOPLASMIC LIPID DROPLET MORPHOLOGY AND ASSOCIATED PROTEINS IN RESPONSE TO AN ACUTE DIETARY FAT CHALLENGE ...... 92 3.1 Abstract ...... 92 3.2 Introduction ...... 93 3.3 Materials and methods ...... 95 3.3.1 Mice ...... 95 3.3.2 Generation of DIO mouse model ...... 95 3.3.3 Mouse procedure for lipid droplet isolation and proteomics analysis ...... 96 3.3.4 Enterocyte isolation ...... 96 3.3.5 CLD isolation ...... 96 3.3.6 In solution digestion and LC MS/MS ...... 97 3.3.7 Protein identification ...... 98 3.3.8 Data analysis and bioinformatics ...... 99 3.3.9 Immunofluorescence imaging ...... 99 3.3.10 Cardiac perfusion, fixation, and transmission electron microscopy ...... 100 3.3.11 Cytoplasmic lipid droplet size analysis ...... 101

viii

Page 3.4 Results ...... 101

3.4.1 Lipid accumulates in CLDs in enterocytes in response to an acute dietary fat

¡¢££¤¥¦ ¤§¢¨©¤ ¥¤© ¡¨ ¢ ¤¢¥¢£¢© ¥ ¨ ¢ ¥ ££ ¤ £§ ¥

lean and DIO mice ...... 101 3.4.2 CLDs in enterocyte of DIO mice are larger and have a different size distribution compared to lean mice in response to an acute dietary fat challenge .. 102 3.4.3 Proteins were identified associated with isolated CLDs from lean and DIO mice in response to an acute dietary fat challenge ...... 102 3.4.4 Proteins identified associated with CLDs from lean and DIO mice are associated with a variety of biological processes ...... 103 3.4.5 Proteins associated with lipid related processes are found at differentially present levels and have been previously characterized associated with CLD in enterocytes proteomic or targeted protein analysis ...... 103 3.4.6 Plin3 and Plin2 localize to CLDs from lean and DIO mice in response to an acute dietary fat challenge...... 104 3.4.7 Plin2 and Plin3 localize to different CLDs in different regions of the cell in DIO mice...... 105 3.5 Discussion ...... 105 3.6 Acknowledgements ...... 112 3.7 References ...... 122 CHAPTER 4. LONG CHAIN FATTY ACYL COA SYNTHETASE 5 ABLATION DECREASES TRIGLYCERIDE STORAGE IN ENTEROCYTES AND SECRETION IN RESPONSE TO AN ACUTE DIETARY FAT CHALLENGE COMPARED TO WILD TYPE MICE ...... 125 4.1 Abstract ...... 125 4.2 Introduction ...... 126 4.3 Materials and methods ...... 128 4.3.1 Diets and mice ...... 128 4.3.2 Tissue imaging ...... 128

Page 4.3.3 TAG secretion ...... 129 4.3.4 High fat feeding and food intake analysis ...... 130 4.3.5 Fecal lipid analysis ...... 130 4.4 Results ...... 130 4.4.1 Acsl5-deficient mice accumulate less neutral lipids in CLDs in response to an acute dietary fat challenge compared to WT mice...... 130 4.4.2 Decreased TAG secretion in Acsl5-deficient mice compared to WT mice. 131 4.4.3 Acsl5-deficeint mice have increased food intake without differences in body weight or fecal lipid content compared to WT mice when fed a high fat diet...... 131 4.4.4 Nuclear localization of Acsl5 in enterocytes in response to an acute dietary fat challenge in WT mice...... 132 4.5 Discussion ...... 132 4.6 Acknowledgments ...... 136 4.7 References ...... 142 CHAPTER 5. SUMMARY AND FUTURE DIRECTIONS ...... 144 5.1 Summary ...... 144 5.2 Future directions ...... 146 5.2.1 CLD morphology and associated proteins along the length of small intestine ...... 146 5.2.2 Changes in CLD morphology and associated proteins over time in response to an acute dietary fat challenge...... 147 5.2.3 The role of Acsl5 in the nucleus ...... 148 5.3 References ...... 150 VITA ...... 151 PUBLICATIONS ...... 153

LIST OF TABLES

Table ...... Page Table 1-1 Proteins identified by proteomic analysis associated with CLDs from enterocytes ...... 41

Table 2-1 GO Terms associated with protein classification term...... 78

Table 2-2 A sub group of proteins are associated with known lipid related functions. .... 79

Table 3-1 Label free quantitation of proteins associated with lipid related Gene Ontology

terms...... 113

LIST OF FIGURES

Figure ...... Page Figure 2-1 Lipid accumulates in CLDs in enterocytes in response to a dietary fat challenge...... 81

Figure 2-2 Isolated CLD fraction is enriched with CLD marker, Plin3, and has a high

TAG to protein ratio...... 82

Figure 2-3 Proteins identified by LC MS/MS and grouped by biological process and

molecular function...... 83

Figure 2-4 Lipid related proteins identified cluster based on predicted physical and

functional interactions...... 84

Figure 2-5 Immunofluorescence imaging of Plin3 and ApoA-IV demonstrates localization on or around CLDs...... 85

Figure 2-6 Immunofluoresence and immunoelectron imaging of Acsl5 demonstrates

localization on or around CLDs...... 86

Figure 3-1 Lipids accumulate in CLDs in enterocytes from lean and DIO mice following

an acute dietary fat challenge...... 115

Figure 3-2 CLDs from enterocytes of DIO mice have a larger average diameter and have

a larger maximum diameter compared to lean mice following an acute dietary fat

challenge...... 116

xii

Figure Page Figure 3-3 Number of proteins identified associated with isolated CLDs from enterocytes

of lean and DIO mice following an acute dietary fat challenge...... 117

Figure 3-4 STRING analysis of proteins associated with lipid related Gene Ontology

terms highlight proteins clusters ...... 118

Figure 3-5 Plin3 and Plin2 localize to CLDs from lean and DIO mice in response to an

acute dietary fat challenge...... 119

Figure 3-6 Plin3 localizes to CLDs on the apical (AP) side of the nuclei compared to

Plin2 which localizes to CLDs on the AP and basolateral (BL) side of the nuclei in DIO

mice in response to an acute dietary fat challenge...... 120

Figure 3-7 Proposed model ...... 121

Figure 4-1 Acsl5-deficient mice accumulate less neutral lipids in CLDs in response to an

acute dietary fat challenge compared to WT mice...... 138

Figure 4-2 Decreased TAG secretion in Acsl5-deficient mice compared to WT mice. . 139

Figure 4-3 Acsl5-deficeint mice have increased food intake without differences in body

weight or fecal lipid content compared to WT mice when fed a high fat diet ...... 140

Figure 4-4 Nuclear localization of Acsl5 in enterocytes in response to an acute dietary fat challenge in WT mice ...... 141

xiii

LIST OF ABBREVIATIONS

Abbreviation Term 3BHS1 3-Beta-hydroxysteroid dehydrogenase AADAC Arylacetamide deacetylase ACS Acyl-CoA synthetase ACSL Acyl-CoA synthetase long chain AGPAT Acylglycerophosphate acyltransferase APO Apolipoprotein ATGL Adipocyte triglyceride lipase CARS Coherent anti-stokes ramen scattering spectroscopy CCT Phosphocholine cytidylyltransferase CD36 Cluster of differentiation 36 CE Cholesteryl ester CES Carboxylesterase CGI-58 Comparative gene identification-58 CIDE Cell death inducing DFF45-like effector CLD Cytoplasmic lipid droplet CM Chylomicron COPII Coat protein complex II DAG Diacylglycerol DGAT Acyl-CoA:diacylglycerol acyltransferase

DHB2 ¡¢-Hydroxysteroid dehydrogenase type 2 DIO Diet-induced obesity ER Endoplasmic reticulum FA Fatty acid

xiv

Abbreviation Term FABP Fatty acid binding protein FAO Fatty acid oxidation FATP4 Fatty acid transport protein 4 FFA Free fatty acid G3P Glycerol-3-phosphate

GPAT Glycerol phosphate acyltransferase

¡¢£¤ Hepatocyte nuclear factor 4 alpha HSL Hormone sensitive lipase LAL Lysosomal acid lipase LCFA Long chain fatty acid LLDs Lumenal lipid droplets LPCAT Lysophosphatidylcholine acyltransferase MAG Monoacylglycerol MGAT or Mogat Acyl-CoA:monoacylglycerol acyltransferase MGL Monoacylglycerol lipase MTP or Mttp Microsomal triglyceride transfer protein PC Phosphatidylcholine PCTV Pre-chylomicron transport vesicle PL Phospholipid

PLIN Perilipin

¥¥¦§ ¤ Peroxisome proliferator-activated receptor alpha SAR1B GTPase Secretion associated Ras related 1B GTPase SNARE Soluble NSF attachment receptor TAG Triacylglycerol TEM Transmission electron microscopy

ABSTRACT

D’Aquila, Theresa M. Ph.D., Purdue University, August 2016. Intestinal Cytoplasmic Lipid Droplets, Associated Proteins, and the Regulation of Dietary Fat Absorption. Major Professor: Kimberly Buhman.

Dietary fat provides essential nutrients, contributes to energy balance, and

regulates blood lipid concentrations. These functions are important to health, but can also

become dysregulated and contribute to diseases such as obesity, diabetes, and

cardiovascular disease. The small intestine absorbs dietary fat through an efficient multi

step process of digestion, uptake, metabolism, and secretion or storage. When dietary fat is taken up by the absorptive cells of the small intestine, enterocytes, it can be secreted into circulation where it contributes to blood lipid levels or temporarily stored in cytoplasmic lipid droplets (CLDs). The objective of this dissertation is to investigate the

presence and metabolism of CLDs in the intestine.

Dietary fat absorption by the small intestine is a multi-step process that regulates the uptake and delivery of essential nutrients and energy. One step of this process is the temporary storage of dietary fat in cytoplasmic lipid droplets (CLDs). The storage and mobilization of dietary fat is thought to be regulated by proteins that associate with the

CLD; however, mechanistic details of this process are currently unknown. In this study we analyzed the proteome of CLDs isolated from enterocytes harvested from the small intestine of mice following a dietary fat challenge. In this analysis we identified 181

xvi

proteins associated with the CLD fraction, of which 37 are associated with known lipid

related metabolic pathways. We confirmed the localization of several of these proteins on or around the CLD through confocal and electron microscopy, including perilipin 3, apolipoprotein A-IV, and acyl-CoA synthetase long-chain family member 5. The identification of the enterocyte CLD proteome provides new insight into potential regulators of CLD metabolism and the process of dietary fat absorption.

We also investigated the effect of diet induced obesity on CLD morphology and associated proteins. It is currently unclear what effect diet induced obesity (DIO) has on the balance between dietary fat storage and secretion. Specifically, there is limited knowledge of how DIO affects the level and diversity of proteins that associate with

CLDs and regulate CLD dynamics. In the current study, we characterize CLDs from lean

and DIO mice through histological and proteomic analyses. We demonstrate that DIO

mice have larger intestinal CLDs compared to lean mice in response to dietary fat.

Additionally we identified 375 proteins associated with CLDs in lean and DIO mice. We identify a subgroup of lipid related proteins which are either increased or unique to the

DIO CLD proteome. These proteins are involved in steroid synthesis pathways, TAG

synthesis, and lipolysis. This analysis expands upon our knowledge of the effect of DIO

obesity on the process of dietary fat absorption in the small intestine.

Lastly, we investigated the role of one of the newly identified CLD associated

proteins on dietary fat absorption. Little is known of the role of Acsl5 in the metabolism

of dietary fat in enterocytes. To understand the role of Acsl5 in dietary fat absorption, we

challenged WT and Acsl5-deficient mice with acute and chronic dietary fat challenges.

Despite no evidence of quantitative fat malabsorption in the absence of Acls5, we found

xvii

significantly reduced TAG storage and secretion by enterocytes compared to WT mice.

Interestingly, Acsl5-deficient consumed more food, but did not gain more weight

compared to WT mice during two weeks of HFD feeding. Furthermore, we found that

Acsl5 localizes to the nucleus after an acute dietary fat challenge. Together this data

suggests that although Acsl5 is not essential for quantitative dietary fat absorption, it

plays a role in intestinal metabolism of dietary fat.

The results presented in this dissertation provide new knowledge of regulators of

CLD metabolism. The long term objectives of these studies are to identify mechanistic

details of the process of intestinal dietary fat absorption. Understanding the process of dietary fat absorption will provide new treatments for the prevention of diseases associated with dyslipidemia.

CHAPTER 1. INTRODUCTION

1.1 Dietary fat, health, and disease

1.1.1 Dietary fat structure and composition

Dietary fat represents a diverse group of chemical compounds with different

structures, functions, health benefits, and risk for disease. Dietary fat is a broad

classification for lipid compounds which are consumed[1]. Dietary fat can be found from

both animal and plant sources and is a significant source of energy. In the western diet,

dietary fat accounts for approximately 35% of calories consumed. The predominant form

of dietary fat consumed is the form of long chain triacylglycerol (TAG), which is

composed of 3 fatty acids esterified to a glycerol backbone. Dietary fat also contains

smaller amounts of cholesteryl esters (CE) and phospholipids (PL). Fatty acids are

present within TAG, CE, or PL and consist of a chain of carbons of various chain lengths

and degree of saturation. Fatty acid species include saturated fatty acids,

monounsaturated fatty acids, polyunsaturated fatty acids (including omega 6 and 3 fatty

acids) and trans-fatty acids.

Portions of this section have been previously published in:

T. D'Aquila, Y.H. Hung, A. Carreiro, K.K. Buhman, Recent discoveries on absorption of dietary fat: Presence, synthesis, and metabolism of cytoplasmic lipid droplets within enterocytes, Biochimica et biophysica acta, 1861 (2016) 730-747.

1.1.2 Dietary fat, fat soluble vitamins, and essential fatty acids

Dietary fat plays a critical role in health by promoting the absorption of essential nutrients of fat soluble vitamins and essential fatty acids[2]. Vitamins A (in the form of retinol esters), vitamin D, vitamin E, and vitamin K, are fat soluble vitamins whose absorption is dependent on the presence of fat in the diet. These vitamins enter the circulation as components of chylomicrons [2] [3]. Malabsorption of these vitamins may lead to deficiency and ultimately xerophthalmia, rickets, neuromuscular disorders, hemorrhaging. Essential fatty acids are also absorbed by the intestine and secreted into circulation [4]. Essential fatty acid deficiency, while rare, can result in neurological and cognitive development problems. The small intestine plays an essential role in the absorption of fat soluble vitamins, essential fatty acids, and promotion of health.

1.1.3 Dietary fat and risk for disease

Dietary fat plays a critical role in the etiology of cardiovascular disease [5]. The association of diet composition and risk for coronary heart disease has been extensively studied for the last 50 years. Despite intense investigation, the link between dietary fat and coronary heart disease is a contested subject. In 1977, public health officials recommended to limit intake of dietary fat to less than 30% of daily calories in an effort to reduce the risk for coronary artery disease[6]. To assess the impact of these recommendations, The Women's Health Initiative Randomized Controlled Dietary

Modification Trial, tested low fat (20% kcal from fat)/high carbohydrate intervention diet against a usual diet (>32% kcal from fat) group. This study demonstrated no effect of

reducing dietary fat on risk for coronary artery disease and cardiovascular disease [7].

This lack luster result shifted the attention from total dietary fat to dietary fatty acid

composition.

Diets high in saturated fat have been implicated in an increased risk for coronary

artery disease. Dietary intervention trials which reduced saturated fat or replaced

saturated fat with other fatty acid types reduced cardiomorbidity and cardiovascular

events, but no had effect on cardiovascular mortality [8]. Based on numerous intervention trials and meta-analyses, the National Heart Foundation of Australia recommended reducing saturated and trans-fat intake and increasing omega 3, omega 6, and monounsaturated fatty acid intake, to reduce the risk of cardiovascular disease [9].

The United States has recently changed dietary guidelines to help Americans eat a healthier diet. The 2015 dietary guidelines suggest limiting the intake of calories from saturated fat to less than 10% of total calories. Additionally, the dietary guideline advisory committee suggests replacing saturated fat with polyunsaturated fat to reduce the risk for coronary artery disease [10]. More specifically, current guidelines suggest reducing trans fats and saturated fats to 1 and 7% respectively and replacing these fats with polyunsaturated and monounsaturated fatty acids [11]. Understanding the molecular mechanisms by which the small intestine absorbs dietary fat is critical for the prevention and treatment of diseases associated with dyslipidemia.

1.2 Intestinal metabolism of dietary fat

1.2.1 Digestion and uptake of dietary fat

The small intestine absorbs dietary fat through an efficient multi step process of digestion, uptake, metabolism, and secretion or storage. Dietary fat absorption is an extremely efficient process (greater than 98%) even when large amounts of dietary fat are consumed [12]. Disruption of the process of dietary fat digestion, uptake, metabolism, and secretion results in fat malabsorption as indicated by steatorrhea or presence of fat in the feces.

Dietary fats are digested and emulsified in the lumen of the small intestine and ultimately taken up by enterocytes through the apical brush border membrane. Crude emulsion of dietary fat enters the small intestine from the stomach and is combined with bile acids from the gall bladder/liver and pancreatic juices. In the lumen of the small intestine, dietary fat is emulsified into mixed micelles with bile acids and phospholipids and undergo hydrolysis by pancreatic lipase [13]. The digestion of dietary TAG results in the production of free fatty acids (FFA) and sn-2 monoacylglycerol (MAG) [14] that are emulsified with bile generating mixed micelles and presented to the brush border membrane of enterocytes.

1.2.2 Fatty acid uptake and metabolism

FFA and MAG are taken up by enterocytes however the mechanisms through

which this occurs is not completely clear. MAG passively diffuses across the plasma

membrane of enterocytes [15]. FFA on the other hand may be taken up by the enterocytes

by passive diffusion or by protein mediated uptake [16]. FFA may diffuse passively

across the plasma membrane via a flip flop mechanism. Since long chain fatty acid

(LCFA) uptake by diffusion is a saturable process [17], a protein dependent mechanism

has been proposed to promote cellular uptake of FFA. FFA may be facilitated across the plasma membrane by CD36, by enterocytes in the proximal section of the small intestine.

Isolated enterocytes from the proximal section of CD36 deficient mice have decreased rate (50%) of fatty acid uptake [18]. In particular, CD36 has a high affinity and promotes absorption of very long chain fatty acids [19]. FFA may be facilitated across the plasma membrane by fatty acid transport protein 4 (FATP4). FATP4 is present on the apical side of the enterocyte and when overexpressed, increases the uptake of LCFA [20].

Interestingly, both CD36 and FATP4 deficient mice have no fat malabsorption. It has been suggested that CD36 and FATP4 may play important however non-essential roles in regulating dietary fat absorption. Although the mechanism through which fatty acids are taken up is unclear, the intestine is capable of responding to high lipid load as demonstrated by low fecal lipid content in healthy humans.

FFA may be metabolically trapped in enterocytes via fatty acid binding proteins and directed toward specific metabolic fates. The upper region of the small intestine expresses two fatty acid binding proteins (FABP), FABP1 or liver FABP and FABP2 or

6 intestinal FABP. Both FABP1 and FABP2 are highly expressed in the small intestine and have a high affinity for LCFA [21]. It is hypothesized that FABP1 and FABP2 bind to

FFA in enterocytes and limit the toxic effects of FFA [22]. FABP1 has high binding affinity for fatty acids but also other lipid species such as lysophospholipids, MAG, fatty acyl-coA, and prostaglandins [23]. FABP1 deficient mice have increased deposition of lumenally derived MAG into PL, MAG, and diacylglycerol (DAG) and with a decreased deposition in TAG. This data indicates FABP1 directs dietary MAG toward TAG synthesis. FABP1 deficient mice have a decrease deposition of lumenally derived fatty acids into mucosal oxidation indicating FABP1 directs fatty acids toward oxidation [24].

FABP2 on the other hand has a high binding affinity for fatty acids but has not been shown to have non fatty acid ligands. FABP2 deficient mice have increased deposition of lumenally derived fatty acids into PL and decreased into TAG indicating FABP2 directs fatty acids into TAG synthesis and away from PL synthesis. Interestingly, both FABP1 and FABP2 deficient mice do not have fat malabsorption as indicated by steatorrhea on chow or high fat diets. While the roles of these two proteins in the intestine is not completely clear, FABP1 and FABP2 may play a role in fatty acids transport and targeting of fatty acids towards different metabolic fates [25].

FFA may be metabolically trapped in enterocytes by conversion to fatty acyl-CoA and directed toward specific metabolic fates. The thioesterification of LCFA to a long chain fatty acyl-CoA is catalyzed by long chain fatty acyl-CoA synthetases (ACSLs) and is an obligatory step in the cellular lipid metabolism. There are at least 25 different acyl-

CoA synthetases with different tissue distribution patterns [26]. The primary ACSLs found in the intestine are Fatp4 and Acsl5. Acyl-CoA synthetases can promote cellular

7 fatty acid uptake via vectorial acylation [27]. Vectorial acylation has not been investigated in the small intestine, however FATP4 inhibition in isolated enterocytes by antisense oligonucleotides reduces the uptake of LCFA [20]. The role of Acsl5 in fatty acid uptake in enterocytes is unknown. In addition to driving fatty acid uptake, ACSLs have been proposed to partition fatty acids toward specific metabolic fates [28]. ACSL partitioning of fatty acids has been characterized in other cell types but has not been fully investigated in enterocytes. Acsl5 directs fatty acyl-coA toward TAG synthesis in a cell model of hepatocytes [29]. The role of Acsl5 in intestinal dietary fat absorption has not been thoroughly investigated. Initial report of Acsl5 deletion in mice shows a 60% decrease in jejunal ACSL activity but no difference in TAG secretion in response to an acute dietary fat challenge [30]. Intestinal deletion of FATP4 does not affect TAG secretion or fecal lipid content in response to an acute dietary fat challenge, however iFATP4 deficient mice have increased fecal lipid TAG content when placed on a Western diet when compared to wild type mice [31]. The investigators report no changes in expression of other FATPs in iFATP4 deficient mice, however Acsl5 expression was not measured and potentially compensates for loss of iFATP4. These results indicate highly expressed ACSLs in the intestine are dispensable for dietary fat absorption.

1.2.3 Fatty acid re-esterification

Fatty acids are re-esterified generating a TAG molecule through the glycerol-3- phosphate (G3P) and acyl-coA: monoacylglycerol acyltransferase (MGAT) pathways.

The first step in the G3P pathway is the acylation of G3P, which is catalyzed by glycerol-

3-phosphate acyltransferase (GPAT) enzymes and yields lysophosphatidic acid [32]. In

the small intestine, two isoforms of GPAT are expressed: GPAT3 (high expression,

jejunum) and GPAT4 (moderate expression) [32-34]. Gpat3-deficient mice have

decreased TAG secretion and increased TAG storage in enterocytes following a dietary

fat challenge [33]. The lysophophatidic acid formed during this first reaction is then

acylated by acylglycerolphosphate acyltransferase (AGPAT) enzymes, also known as lysophosphatidic acid acyltransferases, to form phosphatidic acid. In the small intestine,

AGPAT2, AGPAT4 or AGPAT8 may contribute to the formation of phosphatidic acid.

Lipin enzymes, or phosphatidic acid phosphatases (PAPs), then remove the phosphate

group to form DAG. Lipin-3 mRNA is present in the small intestine of mice and humans

[35]. The presence of players in the G3P pathway and results found in Gpat3-deficient

mice highlight that this pathway may play a more important role in intestinal TAG

metabolism than previously recognized.

Alternatively, free fatty acids may be re-esterified by the MGAT pathway. MGAT

activity accounts for 70-80% of TAG generated for secretion [36]. Multiple MGAT

enzymes have been identified and are differentially expressed in tissues and species. The

first step of the MGAT pathway for TAG synthesis is the acylation of MAG by fatty

acyl-CoA to produce DAG, which is catalyzed by MGAT activity (reviewed in [37]).

MGAT activity is highest in the proximal small intestine and progressively decreases in distal regions [38]. Three MGAT isoforms have been identified, but only MGAT2 and

MGAT3 are present in the small intestine [39]. MGAT2 is abundant in the intestine of both mice and humans, with both mRNA and protein levels reflecting MGAT activity along the length of the small intestine [38]. MGAT3, on the other hand, is expressed in

humans, but not mice, and is highest in the ileum [40]. The role of MGAT2 in dietary fat absorption has been investigated in mouse models. Mgat2 protein and mRNA levels increase in response to a high fat diet, suggesting this enzyme contributes to the increase in MGAT activity that occurs during dietary fat absorption [38]. Two different Mogat2- deficient mouse models were generated to understand the function of Mgat2 in intestinal

TAG synthesis and dietary fat absorption [41, 42]. Although these mice still synthesize

TAG in enterocytes and secrete CM-sized lipoproteins, they exhibit a reduced postprandial triglyceridemic response and a decreased TAG secretion rate in response to an oral fat load compared to wild-type mice [41, 42]. In addition, they remain efficient at absorbing dietary fat when fed a high fat diet. These results suggest that although other enzymes must be present to catalyze this activity, the fate of TAG synthesized in enterocytes may be different in Mogat2-deficient mice than wild-type mice. In fact, while less TAG is secreted in the Mogat2-deficient mouse models, more TAG is found in CLDs of jejunal enterocytes in Mogat2-deficient mice fed a high fat diet than in wild-type mice

[41]. Similar effects have been observed in intestine-specific Mogat2-deficient mice [43] and in mice treated with an MGAT2 inhibitor [44]. In addition, restoring Mogat2 expression specifically in the intestine of Mogat2-deficient mice normalized the TAG absorption rate [45]. Together these results suggest that MGAT2 is not critical for absorption, but important for balancing intestinal TAG storage and secretion. Whether

DAG is generated via G3P or MGAT pathway, the final step in TAG synthesis is catalyzed by Acyl- CoA: diacylglycerol acyltransferase (DGAT).

1.2.4 Triacylglycerol synthesis

Acyl Co-A: diacylglycerol acyltransferase (DGAT) catalyzes the final step in

TAG synthesis. Two proteins with DGAT activity have been identified and investigated

in intestinal lipid metabolism. DGAT enzymes catalyze the acylation of DAG, resulting

in the formation of TAG [37, 39]. This reaction is the final, committed step of TAG

synthesis, and is the point at which the MGAT and G3P pathways converge. Evidence of

DGAT activity on both the luminal (latent activity) and cytosolic (overt activity) sides of

the ER in rat liver microsomes suggests that distinct pools of TAG destined for storage

and secretion may be present in lipoprotein-producing cells [46]. Total DGAT activity is

highest in the proximal intestine and decreases distally along the length of the intestine

[47]. The two known DGAT enzymes, DGAT1 and DGAT2, are both integral ER

membrane proteins and are both present in the small intestine. DGAT2 has also been

shown to localize to CLDs and associate with mitochondria in other cell types [48, 49].

Although both of these enzymes catalyze the same reaction, they are members of two

different gene families, and differ from one another in terms of structure, expression

pattern, and biochemical properties. This suggests that DGAT1 and DGAT2 may play

unique roles in synthesizing TAG for storage versus secretion in enterocytes.

1.2.5 Chylomicron synthesis and secretion

TAG can be packaged onto chylomicrons (CM) for secretion from enterocytes.

CMs are the major TAG rich lipoprotein produced by enterocytes in response to a meal containing dietary fat (reviewed in [2, 50]). CMs are made by a two-step pathway within

11 the ER and are then secreted via pre-CM transport vesicles (PCTVs) through the Golgi and secretory pathway on the basolateral side of the enterocyte to lymph [51]. First,

TAGs and CEs are synthesized by enzymes that localize to the ER membrane. These neutral lipids are packaged onto apolipoprotein (APO) B48 with the help of microsomal triglyceride transfer protein (MTP or Mttp) [52]. APOB48 serves as a lipid acceptor and structural protein of CMs. MTP serves as a lipid transfer protein and chaperone for maintaining APOB48 structure. The neutral lipids, TAGs and CE, are surrounded by a PL monolayer containing free cholesterol and proteins. This initial structure is referred to as the primordial CM [53]. The second step in the synthesis of pre-CMs is the expansion of the CM core by TAG [54]. The mechanisms underlying the addition of TAG is unclear but it is known that MTP is required, and apoB-free ER lumenal lipid droplets (LLDs) are the most likely source of TAG for core expansion [55, 56] . One possible mechanism for this expansion is the fusion of LLDs with primordial CMs. Alternatively, it has been suggested that hydrolysis and re-esterification of LLDs may provide lipids for the second step of core expansion during CM assembly [55].

Pre-CMs exit the ER and are transported to the Golgi for further processing.

These particles are transported within PCTVs containing coat protein complex II (COPII).

COPII is made up of five core proteins: secretion associated, Ras related 1B (Sar1B)

GTPase, Sec23, Sec24, Sec13, and Sec31 [56, 57]. The budding of PCTVs from the surface of the ER is mediated by liver fatty acid binding protein [58, 59] and protein kinase C isoform zeta [60]. PCTVs are larger than protein transport vesicles (250 nm versus 55–70 nm, respectively) and also contain the unique vesicle-associated membrane protein 7 [61]. After budding, the PCTV fuses with the cis-Golgi [62]. Within the Golgi,

pre-CMs are further modified by addition of APOAI [63] and glycosylation of APOB48

[64]. Mature CMs are then destined for secretion where they are released from the

basolateral side of the enterocyte by exocytosis to the lamina propria as imaged by

transmission electron microscopy [65, 66]. The contractile movement of lacteals mediates

the process of CM drainage to mesenteric lymphatic vessels [67]. CMs are then

transported via the thoracic duct that empties into circulation at the left subclavian vein.

In addition to the nonexchangeable APOB48, several exchangeable apolipoproteins can

be found on the surface of CMs, including APOAI, APOAII, APOAIV, APOC and

APOE [68].

The importance of the role in dietary fat absorption in health is highlighted by

human congenital disorders where proteins chylomicron synthesis and secretion are

defective or deficient [69]. Deficiencies in ApoB, Mttp, and Sar1b result in

hypobetalipoproteinemia, abetalipoproteinemia, and chylomicron retention disease

respectively. These diseases are characterized by fat malabsorption presenting as

steatorrhea, low blood lipid (TAG and CHOL) levels, deficiencies in fat soluble vitamins

and essential fatty acids and ultimately failure to thrive.

1.2.6 Cytoplasmic lipid droplets and triacylglycerol storage

Alternatively to secretion in CM, dietary TAG can be stored in CLDs within

enterocytes. The partitioning of TAG into CM synthesis has been well characterized and less is known regarding the synthesis and catabolism of CLDs in enterocytes. The

balance between TAG storage and secretion regulates the appearance of TAG in the

13 blood. The molecular mechanisms of CLD metabolism in enterocytes are relatively unknown. Identification of mediators of intestinal CLD metabolism has the potential to significantly impact the dietary fat absorption and ultimately health and disease.

1.3 Cytoplasmic lipid droplet metabolism

1.3.1 General knowledge of CLD metabolism

1.3.1.1 Physiological conditions promoting CLDs

Cytoplasmic lipid droplets are the main site of lipid storage in eukaryotic cells.

CLDs are efficient stores of excess energy and are generated when nutrients are present in excess. Almost every cell type is capable of generating CLDs, however the size, number, composition, and metabolism of CLDs varies based on cell type and physiological conditions. While CLD characteristics differ, there are some common features. This section will highlight general characteristics of CLD biology and metabolism and the following sections will focus on CLDs specifically in the small intestine.

1.3.1.2 CLD composition

CLDs are composed of a neutral lipid core, surrounded by a phospholipid monolayer and a coat of associated proteins. Additionally, CLDs may be physically connected or closely associated with other organelles which contribute to the metabolism

and function of CLDs. The composition of CLD components vary based on cell type and

physiological condition. For example, CLDs in adipocytes are long term energy stores

while CLDs in hepatocytes and enterocytes are more transient stores for excess fatty

acids. Additionally, the presence, composition, and metabolism of CLDs may affect cell,

tissue, and organism, physiology and function.

1.3.1.2.1 Neutral and phospholipid composition

CLDs are composed of a neutral lipid core with a phospholipid monolayer coat.

The neutral lipid core is predominantly TAG and sterol esters however additional hydrophobic compounds have been detected in isolated CLDs albeit at much lower levels.

These lipid species include retinyl esters, wax esters, and ether lipids [70, 71]. The surface of the CLD is comprised of phospholipids, mainly phsophstidylcholine,

phosphatidylethanolamine, and phosphatidynlinositol [70]. This amphipathic interface

allows for the interaction with proteins and other organelles [72].

1.3.1.2.2 General CLD associated proteins

Various proteins associate with the CLD and regulate CLD metabolism. Perilipin

(Plin) was the first protein identified associated with the CLD in 1991 [73]. Since the

initial discovery of Plin, other members of the perilipin family have been identified

across species with similar sequence and ability to bind to CLDs [74]. The members of

the perilipin family include, Plin1 (Perilipin), Plin2 (Adipocyte differentiation related

protein or ADRP), Plin3 (47 kDa mannose 6-phosphate receptor-binding protein or TIP-

47), Plin4 (Adipocyte protein S3-12, or S3-12), and Plin5 (Lipid droplet-associated protein PAT-1 or OxPAT). It is still unclear what role the different Plin family members play in regulating CLD metabolism. While members of the Plin family are often characterized as regulators of CLD catabolism, new research suggest a potential role in glucose homeostasis [75, 76]. A large number of studies have been performed to identify additional proteins associated with CLDs from a variety of cell types under various physiological conditions. Fueled by technological advances and association of CLDs with various metabolic disorders, the field of CLD biology has advance dramatically in the last

10 years.

A current gap in the field revolves around the targeting proteins to CLDs. Proteins may embed directly into the CLD by a hydrophobic hairpin motif or associate with the surface of the CLD by amphipathic helices or hydrophobic domains [77]. Additionally the ability of a protein to bind to CLDs may change during CLD growth and metabolism

[78]. During lipolysis, CLD surface area decreases resulting in “protein crowding” and ultimately the removal of proteins which are more loosely bound to the CLD. The removal of proteins during CLD shrinkage has the potential to alter CLD composition and metabolism. Understanding which proteins associate with CLDs during CLD synthesis, growth, and catabolism remains an active of research [72].

1.3.1.3 CLD connection with organelles

CLDs are highly connected organelles that physically interact with ER, other

CLDs, mitochondria, and lysosomes [79]. CLDs are thought to emerge from the ER and

may maintain a physical connection or close contact with the ER [80]. The bridge

between ER and CLDs has been suggested to facilitate the transport of proteins and lipids

between the two organelles [49] [81]. This is also a proposed mechanism of CLD growth.

Additionally, CLDs are connected to other CLDs through CIDE proteins. Both CIDEC

and CIDEA localize to CLD contact sites [82, 83]. CLDs are also connected or closely

associated with mitochondria. Fatty acids mobilized from CLD stores can channel to

mitochondria for oxidation [84], although the proteins which mediate the interaction

between the two organelles remain unknown. CLDs are connected to autophagosomes.

Electron microscopy images demonstrate that autophagosomes surround the CLD during

starvation conditions in hepatocytes to mobilize stored TAG [85]. The process of lipid

mobilization through the process of lipophagy will be described later. Recognition of

CLDs as highly connected organelles is transforming the concept of CLDs from mere

lipid storage sites to central players in whole cell metabolism.

1.3.1.4 Roles of CLDs

CLDs store excess energy and help protect cells from free fatty acid induced

cytotoxicity. Tissues specialized in energy storage, such as adipose tissue, can store

incredible amounts of energy in the form of TAG and stored in CLDs. A non-obese

person can store over 500,000 kJ of energy as TAG in CLDs of adipocytes, enough to run

about 30 marathons [72]. Additionally, CLDs protect cells against toxic effects of free

fatty acids. Excess non esterified free fatty acids or cholesterol can trigger cell membrane damage and trigger inflammatory responses. When esterified to TAG or CE and stored in

CLDs, excess fatty acids are buffered against the cellular environment and serve as a

source of lipid for cellular processes.

CLDs serve as a source of fatty acids for cell function. Fatty acids mobilized from

CLD stores can be used to provide energy for the cell through oxidation. Fatty acids

mobilized from CLDs can be used to generate complex lipids. Fatty acids can be used for

synthesis of PL, CE, or TAG. CLDs may also serve as a source of fatty acids or other

lipid species functioning as signaling molecules. Adipose triglyceride lipase (ATGL), a

cytosolic lipase that localizes to the CLD, catalyzes the hydrolysis of a FFA from stored

TAG in the lipid droplet. The resulting FFA potentially acts as a ligand for peroxisome

proliferator- activated receptor alpha (PPARa). Liver specific knockdown of ATGL

decreases the expression of PPARa target genes and the effect is reversed when PPARa

agonists are administered [86].

More recently, CLDs have been hypothesized to serve as protein storage sites.

The most well characterized example of CLDs serving as protein depot is during

embryogenesis in Drosphilia. Histones localize to CLDs until needed for cell replication

[87]. Alternatively, CLDs may serve as a site for targeting proteins for degradation. In

particular ApoB has been shown to localize to CLDs and aggregate when proteasomal

degradation is inhibited [88]. Lastly, proteins may transiently localize to CLDs to be

transferred to other organelles [89].

1.3.1.5 CLD synthesis

CLDs are generated and expand through multiple mechanisms. There are several

proposed models of CLD biogenesis, including ER budding, bicellar excision, and vesicle

formation (reviewed in [90]). The model with the most current support is ER budding. In the ER budding model, the accumulation of neutral lipids within the ER membrane bilayer eventually leads to the budding of a CLD, with the cytosolic ER membrane leaflet forming its PL monolayer. This budding process may be driven solely by lipid accumulation, or mediated by proteins that bind to the ER membrane. A model of CLD synthesis with less support, the bicellar model, involves neutral lipid accumulation in the

ER membrane followed by excision of this lipid along with both surrounding membrane leaflets. The issue with this model is that it would result in a transient hole in the ER membrane. Another model of CLD synthesis with less support is the formation of CLDs within bilayer vesicles. This mechanism would be mediated by vesicle formation machinery of the secretory pathway, and the lipid accumulation could occur either within the ER or after the vesicle buds. The precise mechanism of CLD synthesis in the small intestine has not been directly investigated. Once CLDs are formed, they are able to continue to accumulate TAG and increase in size. One mechanism of CLD growth is through TAG synthesis locally at the CLD surface [49, 90]. PL synthesis is important for

CLD monolayer expansion in this growth mechanism. A second mechanism for CLD growth is through fusion of CLDs [90, 91]. Although there is not much known about

CLD growth in enterocytes specifically, it may occur through one or both of these mechanisms.

TAG synthesis is known to occur at the ER membrane, but more recently this process has been observed at the CLD surface and shown to contribute locally to the

growth of CLDs (reviewed in [90]). For this to occur, TAG synthesis machinery needs to

localize to the CLD surface. One isoform of each of the enzymes involved in the G3P

pathway of TAG synthesis (GPAT4, AGPAT3, PAP, and DGAT2) has been shown to localize to CLDs in either Drosophila S2 cells, 3T3-L1 adipocytes, or COS7 cells [48]

[49] and [92]. In addition, specific isoforms of enzymes that activate FAs to form acyl-

CoAs for incorporation into TAG (ACSL1, ACSL3, and ACSL4) have been identified on

CLDs in several cell types (reviewed in [93]). In enterocytes, TAG synthesis enzymes that have been identified on CLDs are ACSL3, ACSL4, ACSL5, GPAT3 and MGAT2 in mice and humans (reviewed in [33] and [94]). Of these, ACSL3 [95], ACSL5 [96], and

GPAT3 [33] have been validated by confocal or electron microscopy to be present on

CLDs in enterocytes.

During CLD growth, additional PLs are required for expansion of the PL monolayer surrounding CLDs to prevent their coalescence [97]. Phosphatidylcholine (PC) is the most abundant PL in these structures, and plays an important role in CLD growth

[97, 98]. Three pathways mediate PC synthesis and are present in most cell types. These pathways include the de novo, or Kennedy pathway, the phosphatidylethanolamine methyl transferase (PEMT) pathway (liver cells only), and the Lands Cycle.

1.3.1.6 CLD catabolism

Lipolysis mediates the biochemical hydrolysis of neutral lipids and CLD catabolism. The biochemical reaction of lipolysis is catalyzed by enzymes that hydrolyze ester bonds of lipids including TAGs, CEs, and PLs. TAGs are hydrolyzed to glycerol and FAs, CEs are hydrolyzed to free cholesterol and FAs and PLs are hydrolyzed to lysophospholipids and FAs. In a cell, lipolysis often refers to the breakdown of CLDs; however, lipolysis can take place wherever lipids containing ester bonds are present.

Through lipolysis, the neutral lipids in the core of CLDs are able to be mobilized to a variety of metabolic fates.

There are two identified pathways that mediate lipolysis of CLDs in cells.

Cytoplasmic TAG lipolysis is the most well-known pathway and has been defined primarily by studies in adipocytes (reviewed in [99]). Lipophagy is a newly-identified pathway and refers to the selective degradation of CLDs through autophagy (reviewed in

[100]). The enzymes and co-activator involved in cytoplasmic TAG lipolysis and lipophagy are present and associate with CLDs in multiple cell types.

1.3.1.6.1 Adipose Triglyceride Lipase

ATGL catalyzes the first step of cytoplasmic lipolysis by hydrolyzing an ester bond in TAG to generate DAG and FA [101]. TAG storage in CLDs inversely correlates with ATGL activity in enterocytes. In enterocytes of C. elegans, reducing ATGL activity via RNA interference leads to increased TAG storage in CLDs, whereas increasing

ATGL activity by overexpression decreases TAG storage in CLDs [102]. In addition,

whole body Atgl-deficient mice have a severe defect in lipolysis and abnormal

accumulation of TAG in CLDs in many tissues (including enterocytes), and they develop

obesity [103].

1.3.1.6.2 Comparative Gene Identification- 58

CGI-58 is a cofactor that regulates ATGL activity and is present in intestine of humans, mice and C. elegans. CGI-58 binds to ATGL and is required for efficient ATGL

activity; however, it has no measurable hydrolase activity [104]. Mutations in CGI-58 in

humans cause Chanarin–Dorfman Syndrome [105]. Chanarin–Dorfman Syndrome is a

rare genetic disease resulting in ichthyosis and abnormal lipid accumulation in multiple

tissues including the intestine [106]. CGI-58 also plays an important role in mouse

physiology, including intestinal lipid metabolism. Whole body Cgi-58-deficient mice die

shortly after birth due to a severe skin barrier defect [107].

1.3.1.6.3 Hormone Sensitive Lipase

HSL has broad substrate specificity; but is thought to primarily catalyze the

second step of TAG hydrolysis, the hydrolysis of an ester bond in DAG to generate MAG

and FA. HSL also possesses hydrolytic activity towards TAGs, MAGs, CEs and retinoid

esters, but lacks phospholipase activity [reviewed in [108]). Whole body Hsl-deficient

mice accumulate DAGs in adipose and non-adipose tissues but do not develop obesity

[109, 110]. The link between HSL deficiency and human disease is not clear; however, a

recent clinical study identified a frameshift mutation in the LIPE gene (encodes HSL)

that reduces adipose tissue lipolysis in both basal and stimulated states [111].

1.3.1.6.4 Monoacylglycerol Lipase

MGL catalyzes the final step of cytoplasmic lipolysis by hydrolyzing MAG to

glycerol and FA. MGL localizes not only to CLDs, but also cell membranes and the

cytoplasm [112]. Beyond hydrolyzing dietary MAG species, MGL also hydrolyzes the

endogenous cannabinoid 2-arachidonylglycerol and is known to regulate cannabinoid

signaling in the central nervous system (reviewed in [113]). Therefore, effects of altered

MGL expression may be due to either TAG hydrolysis or cannabinoid signaling, and are challenging to differentiate. Mouse models deficient in MGL have a decreased rate of lipolysis and accumulation of MAG [114].

1.3.1.6.5 Putative methods of CLD catabolism

Several lines of evidence suggest the presence of additional lipases particularly in non-adipose tissue (Reviewed in [115]). In the liver, ATGL accounts for less than 50% of

TAG hydrolysis [116] and ATGL deficient mice develop hepatosteatosis [103]. Residual

TAG activity in ATGL deficient mice and normal VLDL production suggest the presence of additional TAG hydrolases in the liver. While putative lipases have been briefly studied in the liver, very little is known about putative lipases in other cell types.

1.3.1.6.5.1 Arylacetamide deacetylase

In addition to classical lipases and mediators of autophagy, several putative lipases have recently been identified. Arylacetamide deacetylase (Aadac) is an enzyme with mRNA expression in hepatocytes, intestinal mucosa, pancreas, and adrenal glands

[117]. Aadac has sequence homology with hormone sensitive lipase. In the intestine Aada mRNA levels increased with fenofibrate administration. In the liver Aadac is characterized as a putative microsomal triglyceride lipase [118]. HuH7.5 cells, a human hepatoma cell line, expressing a short hairpin RNA targeting Aadac has impaired TAG lipolysis and VLDL secretion. This data indicates that Aadac may play a role in TAG lipolysis in non--adipose cell types.

1.3.1.6.5.2 Carboxylesterase

Members of the carboxylesterase family are potential lipases. Carboxylesterase 3

(Ces1d, or triglyceride hydrolyze) has triglyceride hydrolase activity [119]. Liver specific deletion of Ces3 results in reduced plasma TAG without hepatosteatosis [120]. Other members of the carboxylesterase family have not been investigated for their role in TAG lipolysis.

1.3.1.6.5.3 Autophagy

Autophagy is the catabolic process that breaks down cellular components sequestered within double-membrane organelles called autophagosomes by transporting them to lysosomes. Emerging studies identify CLDs as a selective substrate for

autophagy which are delivered by autophagosomes to lysosomes, where lysosomal acid lipase (LAL) hydrolyzes ester bonds in lipids. This process is referred to as lipophagy

(reviewed in [100]). LAL is expressed ubiquitously and is responsible for hydrolyzing

ester bonds in TAGs and CEs within the lysosome, generating free cholesterol, glycerol

and FAs. Mutations in LAL in humans result in two diseases: Wolman disease (complete

loss of LAL) and CE storage disease (CESD; partial loss of LAL) [121, 122]. Patients

with Wolman disease usually have failure to thrive during infancy due to severe

gastrointestinal issues including vomiting, diarrhea and malabsorption. Massive storage

of TAGs and CEs are observed in the liver and small intestine of these patients[123].

Whole-body Lal-deficient mice have similar features of Wolman disease. These mice can

survive into adulthood, but have a shorter lifespan than wild-type mice [124].

1.3.2 CLD protein in enterocytes

1.3.2.1 Physiological conditions promoting CLDs in enterocytes

CLDs are present in the small intestine when chylomicron synthesis is inhibited

and in response to a high fat meal. CLDs are present in the intestine when there are

deficiencies in proteins involved in chylomicron synthesis including ApoB, Mttp, and

Sar1b result in hypobetalipoproteinemia, abetalipoproteinemia, and chylomicron

retention disease. These diseases are characterized by accumulation of TAG in

enterocytes which results in steatorrhea due to enterocyte turnover. These diseases are

also characterized by low blood lipid (TAG and CHOL) levels, deficiencies in fat soluble

vitamins and essential fatty acids and ultimately failure to thrive. Additionally, CLDs are

present in the intestine when chylomicron synthesis and release is inhibited by chemical inhibitors. A commonly used non-ionic hydrophobic surfactant known as Pluronic L-81

reversibly inhibit the transport of intestinal chylomicrons [125]. Pluronic L-81 treated

mice accumulate TAG in enterocytes in response to a meal, however it is not clear if

TAG accumulates in CLD or in other cellular organelles such as the ER or the Golgi.

Lastly, CLDs are transiently present in the intestine after a high fat meal. TAG levels in

enterocytes increase and decrease over time [126].

1.3.2.2 CLD composition

1.3.2.2.1 Neutral and phospholipid composition

CLDs in enterocytes contain neutral lipids, phospholipids, and highly

hydrophobic lipids. Isolated CLDs contain primarily TAG but also cholesterol ester, diacylglycerol ether, and phospholipid, when Caco-2 cells were exposed to mixed micelles containing 2 mM sodiumtaurocholate, 0.6 mM oleic acid, 0.2 mM lysophosphatidylcholine, 0.05 mM cholesterol and 0.2 mM 1-O-octadecyl-rac-glycerol

[127]. In a separate experiment CLDs were isolated from the duodenum after a cholesterol (2%) and sunflower oil challenge. The lipid content of these isolated CLDs were analyzed by lipidomics and found to contain triglyceride, cholesterol ester, free cholesterol, phosphatidylcholine, sphingomyelin, ceramides, free fatty acids and mono and di-acylglycerols. Not surprisingly, the lipid species found in isolated CLDs correspond to substrates or products of the enzymes which associate with the CLD [128].

1.3.2.2.2 Perilipins

Members of the perilipin family associate with CLDs in enterocytes. The first proteins identified associated with CLDs in enterocytes were members of the PLIN family, perilipin 2 (Plin2, also adipocyte related protein (ADRP) or adipophilin)) and

Plin3 (also Tip47) [129]. These proteins were investigated because the PLIN family of proteins is well established to associate with CLDs and regulate lipid storage and hydrolysis in many cell types [130]. Plin2 and Plin3 are the only members of the family identified in intestine and are differentially expressed and localize to CLDs in enterocytes

based on specific dietary conditions. Plin3, an exchangeable CLD associated protein,

¡¢£ ¤¥¦§¨¡© § £ ¨ ¦£ ¡ ¤¦ ¡¤ ¡ § ¨ ¡¢£ ¤¥¦§ ¡ ¤ ¨ ¦ ¢¨ ¡ £§

and is absent from CLDs during fasting, as well as in chronic high-fat fed mice in the fed state, as determined by immunofluorescence microscopy [129].

The role of Plin2 in dietary fat absorption was further investigated in a Plin2- deficient mouse model [131]. Plin2-deficient mice have increased fecal TAG content and decreased CLDs in jejunum and ileum when fed either low-fat or high-fat diets compared to wild-type mice. Although the effects of Plin2-deficiency on CM synthesis and secretion were not investigated, these results indicate that Plin2 plays a role in dietary fat absorption, and Plin3 cannot fully compensate for the loss of Plin2 in this process.

1.3.2.3 CLDs may serve multiple roles in the enterocyte

CLDs may protect enterocytes from free fatty acid toxicity[132].CLDs may help

regulate the rate and amount of dietary fat absorption after a meal by serving as a sink

and store of dietary TAG. After a meal, dietary fat can go toward chylomicron synthesis

or toward storage in CLDs. Additionally, CLDs may regulate the length of the post

prandial period by controlling the mobilization of stored TAG for chylomicron synthesis.

CLDs may regulate the rate of steroid synthesis. Several proteins associated with steroid

metabolism have been identified and confirmed on the CLD in enterocytes [94] including

¡£¤¥¦¡§¨©¤¥ ££© ¡£¤¥ © §© ¢ 3-¢ - -hydroxysteroid dehydrogenase 2

(DHB2), and Oestradiol 17-¢ -dehydrogenase 11 (HSD17B11). While the small intestine

is not typically classified as a steroidogenic organ, the presence of steroid hormone

metabolism proteins associated with the CLD suggests local production or metabolism of

steroids.

The FAs or lipids released from enterocyte CLDs have the potential to serve as

signaling molecules that regulate intestinal and/or systemic physiology. These FAs or

lipids may serve as ligands for transcription factors, or substrates for synthesizing

specialized, bioactive lipids that regulate diverse cellular functions. In the small intestine,

! ""#!

¨ ¥ ¨¥¤ § § § £ FAs and lipids act as ligands for trans ¤

(see details in Section 5.3). In addition, arachidonic acid is the major FA precursor for

synthesizing eicosanoids (such as prostaglandins, leukotrienes, and other related

molecules) and endocannabinoids (anandamide and 2-arachidonoylglycerol). Eicosanoids

and arachidonic acid metabolites in the intestine have been shown to mediate

inflammatory processes and play a role in inflammatory bowel disease [133, 134].

Furthermore, endocannabinoids have been shown to activate the endocannabinoid system

by binding to their receptors in the intestine and other cell types, resulting in protective effects against inflammation [135] and [136], regulation of food intake [137], and regulation of energy homeostasis [138]. Therefore, the dynamic CLD storage in

enterocytes may be important in the regulation of lipid-mediated signaling pathways.

¡¢£¤¥¦ §¨©¦ § ¢ ¨ ¥¦ ¤ ¤¦ ©§ ¤© §¤¨£ -oxidation to generate

ATP. Although glutamine and glutamate are the major fuel sources in enterocytes [139-

141] and intestinal FAO activity is low in fed and fasted states [142, 143], emerging studies highlight that regulation of intestinal FAO activity has the capacity to alter enterocyte TAG storage and secretion. Altering FAO in the intestine may influence systemic FA availability, which could alter postprandial blood TAG levels and/or energy balance.

1.3.2.4 CLD synthesis

The method of CLD biogenesis in enterocytes is thought to be the result of TAG accumulation within the ER leaflet. The mechanism through which TAG is partitioned to chylomicron synthesis or CLD synthesis is unknown however may be influenced by

DGAT1/2 and MGAT2 as reviewed in [37].

1.3.2.5 CLD catabolism

In enterocytes, the size and number of CLDs increase and then decrease over time

after dietary fat is consumed [126]. This depletion of CLDs over time suggests active

lipolysis during dietary fat absorption. In addition, an in vitro experiment showed that the

cytoplasm exhibits the highest lipolytic activity compared to other fractions in Caco-2

cells [144], indicating that CLD catabolism is an important function of intestinal lipid

metabolism. CLD catabolism may regulate the quantity and rate at which dietary fat is

absorbed.

1.3.2.5.1 Adipose Triglyceride Lipase

Intestine-specific Atgl-deficient (iAtgl-deficient) mice were generated to

investigate the role of ATGL in intestinal TAG metabolism and dietary fat absorption

[145]. iAtgl-deficient mice have reduced TAG hydrolase activity in enterocytes from the

proximal small intestine and increased TAG storage in CLDs compared with wild-type

mice fed either chow or a high fat diet. However, iAtgl-deficient mice have similar TAG

secretion from enterocytes and no fat malabsorption compared to wild-type mice. iAtgl-

¡¢ £¤¥ ¦§¨ © £ © ¨ © ¢ § ¢ ¡ ¦¤¥¦ § ¤ ¡¨ ©¨ ¤¡ defici

regulating fatty acid oxidation (FAO), oxidative stress response, and cholesterol

£¤¥ ¦ ¥ § ¤¡ ¢ § ¢ absorption in enterocytes compared to wild-¢

genes in iAtgl-deficient mice may contribute to their intracellular TAG accumulation, as

well as their delayed cholesterol absorption. Overall, these results suggest that intestinal

ATGL directs FAs released from TAG stored in CLDs towards oxidation mediated by

¡¢£ activation, and is not directly involved in CM synthesis for TAG secretion into

circulation.

1.3.2.5.2 Comparative Gene Identification-58

Intestine-specific Cgi-58-deficient (iCgi-58-deficient) mice have a significant

reduction in intestinal TAG hydrolase activity, which results in massive TAG storage in

CLDs in enterocytes compared to wild-type mice when fed chow or a high fat diet [146].

Unlike iAtgl-deficient mice, the TAG accumulation in enterocytes of iCgi-58-deficient

mice leads to steatorrhea, indicating fat malabsorption. In addition, iCgi-58-deficient

mice have lower postprandial blood TAG levels and reduced intestinal FAO activity

compared to wild-type mice. These results suggest that CGI-58 is involved in mobilizing

FAs stored in CLDs that are used for both secretion and oxidation. Recently, lipid droplet protein 1, a homolog of CGI-58, was identified in enterocytes of C. elegans and found to activate lipolysis by binding to ATGL during fasting [102].

1.3.2.5.3 Hormone Sensitive Lipase

HSL is expressed along the length of the intestine in mice and primarily regulates intestinal cholesterol metabolism. Whole body Hsl-deficient mice have mild reductions in

DAG hydrolase activity in the intestine but totally abolished CE hydrolase activity [147].

Mice with intestine-specific HSL deficiency (iHSL-deficient mice) were generated to

31 study the role of HSL in intestinal lipid metabolism [148]. iHSL-deficiency has little effect on TAG metabolism in the intestine. iHSL-deficient mice have similar TAG secretion into circulation and intracellular lipid species (including TAG, DAG, FA and

PL) in enterocytes compared to wild-type mice. However, HSL-deficiency has a significant effect on intestinal cholesterol metabolism [148]. iHSL-deficient mice have higher levels of CEs, enhanced cholesterol uptake and reduced cholesterol biosynthesis in enterocytes compared to wild-type mice. These results suggest that in the intestine HSL is not required for TAG metabolism, but plays an important role in CE hydrolysis, which is required for normal intestinal cholesterol metabolism.

1.3.2.5.4 Monoacylglycerol Lipase

Several mouse models with altered Mgl expression have been generated to study the role of this enzyme in intestinal lipid metabolism. Whole body Mgl-deficient mice have significantly less MAG hydrolase activity and massive accumulation of MAG species in multiple tissues, including small intestine [149, 150]. Whole body Mgl- deficient mice are either leaner compared to wild-type mice fed either a low or high fat diet [149], or of similar body weight with improved glucose tolerance and insulin sensitivity [150]. In addition, these mice display blunted intestinal TAG secretion without fat malabsorption, as determined by fecal fat content [149]. They also have reduced mRNA levels of genes involved in the G3P pathway and increased levels of genes involved in the MGAT pathway in enterocytes. Overall, the alterations in lipid metabolism in the absence of Mgl may contribute, in part, to the lean and/or improved

glucose and insulin sensitivity phenotype observed in these models. Mice with intestine-

specific overexpression of MGL (iMgl overexpression mice) have higher intestinal MGL activity, resulting in decreases in MAG species in the intestine compared to wild-type mice [151]. iMgl overexpression mice are hyperphagic and have reduced energy expenditure, which contributes to increased susceptibility to obesity when fed a high fat diet. This phenotype is probably due to altered endocannabinoid signaling in the intestine since iMgl overexpression mice have little changes in intestinal lipid metabolism.

Intestinal TAG synthesis, metabolism of MAG into different lipid species, and intestinal

FAO activity are similar in iMgl overexpression mice compared to wild-type mice.

However, iMgl overexpression mice have significantly greater TAG accumulation in the intestine compared to wild-type mice when fed a high fat diet, but do not have fat malabsorption, as determined by fecal fat content. Whether this intestinal TAG accumulation has effects on dietary fat absorption or contributes to the obese phenotype is unclear.

1.3.2.5.5 Putative methods of CLD catabolism

1.3.2.5.5.1 Carboxylesterase

Members of the carboxylesterase family are putative or potential lipases.

Carboxylesterase 2 is a (Ces2) is a highly expressed esterase in the liver and the intestine

[152]. Recombinant mouse Ces2 demonstrates TAG hydrolase activity and liver lysates

overexpressing Ces2 have increased TAG hydrolysis [153]. Members of the Ces2 family

have been identified on intestinal CLDs [96, 127] and [95] however the role of Ces2 in the intestinal lipid metabolism has not been investigated.

1.3.2.5.6 Autophagy

CLDs are catabolized in the enterocyte through autophagy. Lipophagy is stimulated in Caco-2 cells in response to treatment with lipid micelles, as determined by the co-localization of TAG and components of autophagy machinery using confocal microscopy [154]. Although little is known about mechanistic activation and regulation of lipophagy in enterocytes, inhibition of autophagy initiators/mediators by miRNA and drug inhibition of LAL result in an increase in TAG storage in CLDs in Caco-2 cells.

Together these results highlight that lipophagy is active and regulates intestinal CLD metabolism during dietary fat absorption.

1.4 Methods for identifying CLD associated proteins

Methods for identifying CLD associated proteins by either targeted or global proteomic approaches vary from study to study based on cell type of interest and endpoint, however have some consistent key features. Briefly, cells are isolated and lysed,

CLDs are isolated via centrifugation, and proteins are identified by Western Blot or tandem mass spectrometry analysis. Identifying proteins associated with CLDs help identify regulators of CLD metabolism and expand the model of the role of CLDs in cell biology and physiology.

1.4.1 Introduction to identifying CLD associated proteins

There are multiple methods for identifying proteins which associate with CLDs.

In general, these methods fall into one of two categories, targeted and global proteomic

approaches. Both of these approaches depend on isolation of CLDs without

contamination of associated organelles, associated membranes, or non-CLD lipid

containing particles.

1.4.2 Isolation of CLD associated proteins

To identify CLD associated proteins, cells are harvested under conditions where

CLDs are present. The cells are lysed using various lysis methods based on cell type as

described in [155]. The method of cell lysis and homogenization is dependent on the

CLD size and cell type. A French Press or nitrogen bomb can be used to disrupt cells.

Excess force or mechanical shearing may disrupt very large CLDs as found in adipocytes.

The low density of CLDs allows the organelles to float in an aqueous layer after centrifugation. The speed at which cells are centrifuged is dependent on CLD size. Larger

CLDs may be isolated at much slower speeds than smaller CLDs. In the case of lipoprotein producing tissues such as enterocytes and hepatocytes, a slower centrifugation step may allow for the separation of CLDs from lipoprotein sized contaminating particles.

To identify bona fide CLD associated proteins, some investigators wash the isolated CLD fraction to remove contaminating membranes. This washing process has been shown to result in the removal of both contaminating proteins as well as reduce the level of bona fide proteins [156]. Once isolated, the CLD fraction is delipidated and proteins are

precipitated and prepared for mass spectrometry analysis [Reviewed in[157]] Once a

protein is identified associated with the CLD fraction by mass spectrometry analysis, the

localization of the protein to the CLD should be confirmed using imaging techniques.

1.4.3 Proteomics methods

Proteomic methods for identifying CLD associated proteins fall into two general categories, targeted and global proteomics.

1.4.3.1 Targeted approach

CLD associated proteins can be identified through a targeted proteomic approach.

This approach involves the analysis of isolated CLD associated proteins by Western Blot

analysis. This approach is cost effective, relatively simple, and does not require special

equipment or analysis software. The downside to a targeted approach is the analysis is

limited by previous knowledge of proteins which are thought to play a role in CLD

metabolism or localize to CLDs.

1.4.3.2 Global approach

CLD associated proteins can be identified through a global proteomic approach.

This approach involves the analysis of CLD associated proteins by mass spectrometry

analysis. This approach is more complex and requires specialized software and analysis

equipment. This approach is more sensitive than traditional Western Blot analysis,

particularly with modern high resolution mass spectrometers. Within the category of

global proteomics there are two sub classifications, labeled and label free proteomics.

Several labeling methods exist [158] which revolves around using stable isotopes to label

amino acids to quantitatively compare levels of proteins in two or more treatment groups.

Alternatively, quantitative levels of CLD associate proteins can be determined using label

free proteomic approach [159]. This approach utilizes normalization software to

determine the relative levels of a protein in a sample.

1.4.4 Confirmation of CLD associated proteins by imaging

Recent advances in imaging techniques have expanding upon the ability to

visualize CLDs under different physiological conditions and resolutions. Commonly used

microscopy techniques to visualize lipids are fluorphore labeled lipid (ex. BODIPY

labeling of fatty acids), Coherent anti-stokes ramen scattering spectroscopy (CARS), and

transmission electron microscopy. [Reviewed in [160]]. Each technique has benefits and

limitations which should be taken into consideration when interpreting results. A number

of techniques have been employed to visualize the localization of proteins to CLDs including conjugating proteins to fluorescent tags and conjugating proteins to fluorescently or gold labeled antibodies for confocal or TEM analysis respectively. High resolution microscopy is required to determine if a protein localizes to CLDs or surrounding structures.

1.5 Recent discoveries and new mysteries of CLD associated proteins

1.5.1 General observation of newly identified CLD associated proteins

A significant body of work has been generated regarding the identification of new

CLD associated protein. Newly identified CLD associated proteins are associated with a wide range of biological processes and metabolic pathways. These proteins include various mitochondrial, ER, chaperone, membrane trafficking, and carbohydrate metabolism related proteins. Non-lipid related proteins have been identified in CLD proteomic profiles from multiple species and cell types [96, 156, 161-167]. The identification of the roles of these non-lipid related proteins on the CLD is currently an understudied and underexplored area of research within the field of CLD biology.

1.5.2 Newly identified CLD associated proteins in enterocytes

The proteomes of isolated CLDs derived from Caco-2 cells and mouse enterocytes have been analyzed using a tandem mass spectrometry approach [95, 96, 127] and have been previously reviewed [94]. These studies identified both known and novel

CLD associated proteins. Proteins were categorized based on their gene ontology into lipid related and non-lipid related proteins. Within the lipid related category, proteins fall into various sub groupings including: modifiers of FAs, PL synthesis enzymes, lipid transporters, and steroid synthesis enzymes (Table 1-1). Several of these proteins were

confirmed on the CLD through imaging techniques, including 3-beta-hydroxysteroid ¥¦¨ dehydrogenase (3BHS1) [95, 127] , ACSL3 [95], ACSL5 [96], APOA- ¡ ¢£¤¥ ¦§¨ © -

38 hydroxysteroid dehydrogenase type 2 (DHB2) [95], and LPCAT2 [95, 127]. It is currently unclear what function many of these proteins serve on the CLD; however, the presence of these proteins can be used to generate several intriguing hypotheses.

1.6 Effects of obesity on dietary fat absorption

The effect of obesity on intestinal lipid metabolism is currently unclear, but may be dysregulated by obesity and associated diseases (recently reviewed [2]). Much of the research on obesity and blood fat levels have been conducted in the fasted state, however after work done by Zilversmit et al [168] which correlated increased post prandial response with an increased risk for atherosclerosis, more attention has been paid to blood fat in response to a meal. It is often to distinguish between the effect of obesity, and its comorbidity insulin resistance, on post prandial lipidemia.\

1.6.1 Whole animal/ physiological consequences

Obesity is associated with dyslipidemia in the post prandial state which has been suggested as a risk factor for coronary artery disease [169]. The small intestine contributes to blood lipid levels in the post prandial state by secreting chylomicrons. The level of fat in the blood after a meal is a balance between the secretion of lipoproteins and clearance by peripheral tissues. The effect of obesity on this balance is relatively unknown.

Obese subjects with high levels of visceral adipose tissue had a prolonged or

higher postprandial triglyceride response compared to lean subjects [170-173]. The authors conclude that the postprandial response was due to in part by the contributions of liver lipoproteins [170].

1.6.2 Enterocyte specific consequences

Obesity affects lipid metabolism in enterocytes. Duodenal explants from insulin resistant obese patients had a decreased triglyceride rich lipoprotein production compared to insulin sensitive obese patients [174]. DIO mice had similar TAG content in intestinal villi in response to a dietary fat challenge as determined by CARS microscopy compared to lean mice. Additionally, DIO mice have decreased rate of TAG secretion in response to a dietary fat challenge [175, 176].

Chronic high-fat diet consumption in mice results in DIO and alters intestinal

CLD metabolism. DIO mice have a reduced TAG secretion rate in response to dietary fat compared to lean mice [4] and [197]. In the fed state, DIO mice have abundant TAG storage in CLDs within enterocytes, but after a six hour fast this TAG storage is significantly reduced. In DIO mice given an acute dietary fat challenge, mRNA levels of genes involved in lipolysis (Atgl and Hsl) are significantly reduced compared to before the challenge [4]. This is a response to dietary fat that does not occur in lean mice. These results suggest that DIO induces adaptations in CLD metabolism resulting in decreased mobilization of enterocyte TAG stores and decreased TAG secretion.

1.7 Conclusion

Identifying regulators of intestinal CLD metabolism can provide potential targets for the prevention and treatment for chronic diseases associated with dyslipidemia including obesity, Type II Diabetes, and cardiovascular disease (Pirillo 2014) (Patsch

1992). The small intestine is the source of chylomicrons in the post prandial state. The balance of dietary TAG storage and secretion may regulate post prandial TAG levels. The proteins that associate with CLDs regulate CLD metabolism and TAG mobilization.

Identifying proteins that associate with CLDs and regulate CLD metabolism are potential targets for regulating chylomicron lipidation and post prandial hypertriglyceridemia. A few of the outstanding questions regarding the role of CLDs, and associated proteins in the regulation of dietary fat absorption that this thesis intends to address are:

1) What proteins associated with CLDs in enterocytes in response to a dietary fat challenge? (Chapter 2).

2) What effect does obesity have on CLDs and associated proteins? (Chapter 3)

3) What role does Acsl5 play in dietary fat absorption? (Chapter 4)

Table 1-1 Proteins identified by proteomic analysis associated with CLDs from enterocytes

.Gene symbol Protein name Biological pathway Reference CYB5R3 NADH–cytochrome b5 reductase 3 Cholesterol metabolism 95, 96, 127 DHCR7 7-Dehydrocholesterol reductase Cholesterol metabolism 127 NAD(P)H steroid dehydrogenase-like NSDHL protein Cholesterol metabolism 95, 127 ACSL3 Long chain acyl-CoA synthetase 3 Fatty acid metabolism 95, 127 ACSL4 Long chain acyl-CoA synthetase 4 Fatty acid metabolism 127 ACSL5 Long chain acyl-CoA synthetase 5 Fatty acid metabolism 95, 96 ECHS1 Enoyl-CoA hydratase, mitochondrial Fatty acid oxidation 127 Hadh Hydroxyacyl-coenzyme A dehydrogenase Fatty acid oxidation 96 Ugt1a7c UDP-glucuronosyltransferase 1–7C Hormone metabolism 96 Bpnt1 3(2), 5-bisphosphate nucleotidase 1 Hydrolase 96 Ces2a Pyrethroid hydrolase Hydrolase 96 Ces2c Acylcarnitine hydrolase Hydrolase 96 Ces2e Pyrethroid hydrolase Hydrolase 96 EPHX1 Epoxide hydrolase 1 Hydrolase 127 Ephx2 Bifunctional epoxide hydrolase 2 Hydrolase 96 LDAH Lipid droplet-associated hydrolase Hydrolase 95, 127 Lipid droplet CGI-58 Comparative gene identification-58 metabolism 127 Lipid droplet FAF2 FAS-associated factor 2 metabolism 95, 127 Lipid droplet Lipe Hormone-sensitive lipase metabolism 96, 128 Lipid droplet PLIN2 Perilipin 2/ADRP metabolism 95, 127, 128, 129 Lipid droplet PLIN3 Perilipin 3/TIP47 metabolism 95, 96, 127, 129 Patatin-like phospholipase domain- Lipid droplet PNPLA2 containing protein 2 metabolism 95, 127, 128 Beta-1, 4 N-acetylgalactosaminyltransferase B4galnt2 2 Lipid glycosylation 96 Scp2 Non-specific lipid-transfer protein Lipid metabolism 96 Mogat2 Monoacylglycerol: acylCoA transferase 2 Lipid synthesis 128 Clint Clathrin interactor 1 Lipid transport 96 Diazepam binding inhibitor, Acyl-CoA- Dbi binding protein Lipid transport 96 Fabp1 Fatty acid-binding protein, liver Lipid transport 96 Fabp2 Fatty acid-binding protein, intestinal Lipid transport 96 ANXA2 Annexin A2 Lipoprotein metabolism 96, 127 Apoa1 Apolipoprotein A-I Lipoprotein metabolism 96

Table 1-1 Continued

APOA4 Apolipoprotein A-IV Lipoprotein metabolism 96, 127 Apob Apolipoprotein B Lipoprotein metabolism 96 APOE Apolipoprotein E Lipoprotein metabolism 127 MGLL Monoglyceride lipase Lipoprotein metabolism 95, 127 MTTP Microsomal triglyceride transfer protein Lipoprotein metabolism 95, 96, 127, 128 P4HB Protein disulphide-isomerase Lipoprotein metabolism 95, 96, 127 ANXA4 Annexin A4 Membrane metabolism 96, 127 ELMOD2 ELMO domain-containing protein 2 Miscellaneous 127 Dehydrogenase/reductase SDR family Oxidation-reduction DHRS1 member 1 process 95, 96, 127 Phospholipid HSPA8 Heat-shock cognate 71 kDa protein metabolism 96, 127 Phospholipid LPCAT2 Lysophosphatidylcholine acyltransferase 2 metabolism 95, 127 Phospholipid PCYT1A Choline-phosphate cytidylyltransferase A metabolism 127 Phospholipid Prdx6 Peroxiredoxin-6 metabolism 96 Regulation of lipid Atp5a1 ATP synthase subunit alpha, mitochondrial metabolism 96 Regulation of lipid Atp5b ATP synthase subunit beta, mitochondrial metabolism 96, 127 Regulation of lipid Cat Catalase metabolism 96 Regulation of lipid Golm1 Golgi membrane protein 1 metabolism 96 Adh1 Alcohol dehydrogenase 1 Retinal metabolism 96 Aldh1a1 Retinal dehydrogenase 1 Retinal metabolism 96 DHRS3 Short-chain dehydrogenase/reductase 3 Retinal metabolism 95, 127 Rbp2 Retinol-binding protein 2 Retinal metabolism 96 RDH10 Retinol dehydrogenase 10 Retinal metabolism 95, 127 Sphingolipid SGPL1 Sphingosine-1-phosphate lyase 1 metabolism 127 Cyb5a Cytochrome b5 Steroid metabolism 96 Cyp2b10 Cytochrome P450 2B10 Steroid metabolism 96

HSD17B11 Oestradiol 17-¢ -dehydrogenase 11 Steroid metabolism 95, 96, 127 HSD17B2 Estradiol 17-beta-dehydrogenase 2 Steroid metabolism 95 HSD17B7 3-Oxo-steroid reductase Steroid metabolism 95, 127

HSD3B1 3-¢ -Hydroxysteroid dehydrogenase Steroid metabolism 95, 127 Sulfotransferase family cytosolic 1B Sult1b1 member 1 Steroid metabolism 96 LSS Lanosterol synthase Sterol metabolism 95, 127 SQLE Squalene monooxygenase Sterol metabolism 95

1.8 References

[1] M.C. Carey, D.M. Small, C.M. Bliss, Lipid digestion and absorption, Annual review of physiology, 45 (1983) 651-677. [2] M.M. Hussain, Intestinal lipid absorption and lipoprotein formation, Current opinion in lipidology, 25 (2014) 200-206. [3] P.M. Brannon, J.C. Fleet, Vitamin D, Advances in nutrition (Bethesda, Md.), 2 (2011) 365-367. [4] D.M. Minich, R.J. Vonk, H.J. Verkade, Intestinal absorption of essential fatty acids under physiological and essential fatty acid-deficient conditions, Journal of lipid research, 38 (1997) 1709-1721. [5] A.H. Lichtenstein, E. Kennedy, P. Barrier, D. Danford, N.D. Ernst, S.M. Grundy, G.A. Leveille, L. Van Horn, C.L. Williams, S.L. Booth, Dietary fat consumption and health, Nutrition reviews, 56 (1998) S3-19; discussion S19-28. [6] U.G.P. Office, Dietary goals for the United States, in: U.S.S.C.o.N.a.H. Needs (Ed.), Washington, DC, 1977. [7] B.V. Howard, L. Van Horn, J. Hsia, J.E. Manson, M.L. Stefanick, S. Wassertheil- Smoller, L.H. Kuller, A.Z. LaCroix, R.D. Langer, N.L. Lasser, C.E. Lewis, M.C. Limacher, K.L. Margolis, W.J. Mysiw, J.K. Ockene, L.M. Parker, M.G. Perri, L. Phillips, R.L. Prentice, J. Robbins, J.E. Rossouw, G.E. Sarto, I.J. Schatz, L.G. Snetselaar, V.J. Stevens, L.F. Tinker, M. Trevisan, M.Z. Vitolins, G.L. Anderson, A.R. Assaf, T. Bassford, S.A. Beresford, H.R. Black, R.L. Brunner, R.G. Brzyski, B. Caan, R.T. Chlebowski, M. Gass, I. Granek, P. Greenland, J. Hays, D. Heber, G. Heiss, S.L. Hendrix, F.A. Hubbell, K.C. Johnson, J.M. Kotchen, Low-fat dietary pattern and risk of cardiovascular disease: the Women's Health Initiative Randomized Controlled Dietary Modification Trial, JAMA : the journal of the American Medical Association, 295 (2006) 655-666. [8] L. Hooper, C.D. Summerbell, R. Thompson, D. Sills, F.G. Roberts, H.J. Moore, G.D. Smith, Reduced or modified dietary fat for preventing cardiovascular disease, Sao Paulo medical journal = Revista paulista de medicina, 134 (2016) 182-183. [9] W.S. Shrapnel, G.D. Calvert, P.J. Nestel, A.S. Truswell, Diet and coronary heart disease. The National Heart Foundation of Australia, The Medical journal of Australia, 156 Suppl (1992) S9-16.[10] B.E. Millen, S. Abrams, L. Adams-Campbell, C.A. Anderson, J.T. Brenna, W.W. Campbell, S. Clinton, F. Hu, M. Nelson, M.L. Neuhouser, R. Perez-Escamilla, A.M. Siega-Riz, M. Story, A.H. Lichtenstein, The 2015 Dietary Guidelines Advisory Committee Scientific Report: Development and Major Conclusions, Advances in nutrition (Bethesda, Md.), 7 (2016) 438-444.[11] S. Hammad, S. Pu, P.J. Jones, Current Evidence Supporting the Link Between Dietary Fatty Acids and Cardiovascular Disease, Lipids, 51 (2016) 507-517. [12] H. Kasper, Faecal fat excretion, diarrhea, and subjective complaints with highly dosed oral fat intake, Digestion, 3 (1970) 321-330.[13] C.T. Phan, P. Tso, Intestinal lipid absorption and transport, Frontiers in bioscience : a journal and virtual library, 6 (2001) D299-319. [14] H. Mu, C.E. Hoy, The digestion of dietary triacylglycerols, Progress in lipid research, 43 (2004) 105-133.

[15] G. Schulthess, G. Lipka, S. Compassi, D. Boffelli, F.E. Weber, F. Paltauf, H. Hauser, Absorption of monoacylglycerols by small intestinal brush border membrane, Biochemistry, 33 (1994) 4500-4508. [16] N.A. Abumrad, N.O. Davidson, Role of the gut in lipid homeostasis, Physiological reviews, 92 (2012) 1061-1085. [17] P.J. Trotter, S.Y. Ho, J. Storch, Fatty acid uptake by Caco-2 human intestinal cells, Journal of lipid research, 37 (1996) 336-346. [18] F. Nassir, B. Wilson, X. Han, R.W. Gross, N.A. Abumrad, CD36 is important for fatty acid and cholesterol uptake by the proximal but not distal intestine, The Journal of biological chemistry, 282 (2007) 19493-19501. [19] V.A. Drover, D.V. Nguyen, C.C. Bastie, Y.F. Darlington, N.A. Abumrad, J.E. Pessin, E. London, D. Sahoo, M.C. Phillips, CD36 mediates both cellular uptake of very long chain fatty acids and their intestinal absorption in mice, The Journal of biological chemistry, 283 (2008) 13108-13115. [20] A. Stahl, D.J. Hirsch, R.E. Gimeno, S. Punreddy, P. Ge, N. Watson, S. Patel, M. Kotler, A. Raimondi, L.A. Tartaglia, H.F. Lodish, Identification of the major intestinal fatty acid transport protein, Molecular cell, 4 (1999) 299-308. [21] A.M. Gajda, J. Storch, Enterocyte fatty acid-binding proteins (FABPs): different functions of liver and intestinal FABPs in the intestine, Prostaglandins, leukotrienes, and essential fatty acids, 93 (2015) 9-16. [22] B.P. Atshaves, G.G. Martin, H.A. Hostetler, A.L. McIntosh, A.B. Kier, F. Schroeder, Liver fatty acid-binding protein and obesity, The Journal of nutritional biochemistry, 21 (2010) 1015-1032. [23] J. Storch, L. McDermott, Structural and functional analysis of fatty acid-binding proteins, Journal of lipid research, 50 Suppl (2009) S126-131. [24] W.S. Lagakos, A.M. Gajda, L. Agellon, B. Binas, V. Choi, B. Mandap, T. Russnak, Y.X. Zhou, J. Storch, Different functions of intestinal and liver-type fatty acid- binding proteins in intestine and in whole body energy homeostasis, American journal of physiology. Gastrointestinal and liver physiology, 300 (2011) G803-814. [25] J. Storch, A.E. Thumser, Tissue-specific functions in the fatty acid-binding protein family, The Journal of biological chemistry, 285 (2010) 32679-32683. [26] D.G. Mashek, L.O. Li, R.A. Coleman, Long-chain acyl-CoA synthetases and fatty acid channeling, Future lipidology, 2 (2007) 465-476. [27] I. Niot, H. Poirier, T.T. Tran, P. Besnard, Intestinal absorption of long-chain fatty acids: evidence and uncertainties, Progress in lipid research, 48 (2009) 101-115. [28] D.E. Cooper, P.A. Young, E.L. Klett, R.A. Coleman, Physiological Consequences of Compartmentalized Acyl-CoA Metabolism, The Journal of biological chemistry, 290 (2015) 20023-20031. [29] S.Y. Bu, D.G. Mashek, Hepatic long-chain acyl-CoA synthetase 5 mediates fatty acid channeling between anabolic and catabolic pathways, Journal of lipid research, 51 (2010) 3270-3280. [30] N. Meller, M.E. Morgan, W.P. Wong, J.B. Altemus, E. Sehayek, Targeting of Acyl- CoA synthetase 5 decreases jejunal fatty acid activation with no effect on dietary long-chain fatty acid absorption, Lipids in health and disease, 12 (2013) 88.

[31] J. Shim, C.L. Moulson, E.P. Newberry, M.H. Lin, Y. Xie, S.M. Kennedy, J.H. Miner, N.O. Davidson, Fatty acid transport protein 4 is dispensable for intestinal lipid absorption in mice, Journal of lipid research, 50 (2009) 491-500. [32] K. Takeuchi, K. Reue, Biochemistry, physiology, and genetics of GPAT, AGPAT, and lipin enzymes in triglyceride synthesis, American journal of physiology. Endocrinology and metabolism, 296 (2009) E1195-1209. [33] I. Khatun, R.W. Clark, N.B. Vera, K. Kou, D.M. Erion, T. Coskran, W.F. Bobrowski, C. Okerberg, B. Goodwin, Characterization of a Novel Intestinal Glycerol-3- phosphate Acyltransferase Pathway and Its Role in Lipid Homeostasis, The Journal of biological chemistry, 291 (2016) 2602-2615. [34] A. Yamashita, Y. Hayashi, N. Matsumoto, Y. Nemoto-Sasaki, S. Oka, T. Tanikawa, T. Sugiura, Glycerophosphate/Acylglycerophosphate acyltransferases, Biology, 3 (2014) 801-830. [35] J. Donkor, M. Sariahmetoglu, J. Dewald, D.N. Brindley, K. Reue, Three mammalian lipins act as phosphatidate phosphatases with distinct tissue expression patterns, The Journal of biological chemistry, 282 (2007) 3450-3457. [36] F. Kern, Jr., B. Borgstrom, Quantitative study of the pathways of triglyceride synthesis by hamster intestinal mucosa, Biochimica et biophysica acta, 98 (1965) 520-531. [37] C.L. Yen, D.W. Nelson, M.I. Yen, Intestinal triacylglycerol synthesis in fat absorption and systemic energy metabolism, Journal of lipid research, 56 (2015) 489- 501. [38] J. Cao, E. Hawkins, J. Brozinick, X. Liu, H. Zhang, P. Burn, Y. Shi, A predominant role of acyl-CoA:monoacylglycerol acyltransferase-2 in dietary fat absorption implicated by tissue distribution, subcellular localization, and up-regulation by high fat diet, The Journal of biological chemistry, 279 (2004) 18878-18886. [39] Y. Shi, D. Cheng, Beyond triglyceride synthesis: the dynamic functional roles of MGAT and DGAT enzymes in energy metabolism, American journal of physiology. Endocrinology and metabolism, 297 (2009) E10-18. [40] D. Cheng, T.C. Nelson, J. Chen, S.G. Walker, J. Wardwell-Swanson, R. Meegalla, R. Taub, J.T. Billheimer, M. Ramaker, J.N. Feder, Identification of acyl coenzyme A:monoacylglycerol acyltransferase 3, an intestinal specific enzyme implicated in dietary fat absorption, The Journal of biological chemistry, 278 (2003) 13611-13614. [41] C.L. Yen, M.L. Cheong, C. Grueter, P. Zhou, J. Moriwaki, J.S. Wong, B. Hubbard, S. Marmor, R.V. Farese, Jr., Deficiency of the intestinal enzyme acyl CoA:monoacylglycerol acyltransferase-2 protects mice from metabolic disorders induced by high-fat feeding, Nature medicine, 15 (2009) 442-446. [42] T. Tsuchida, S. Fukuda, H. Aoyama, N. Taniuchi, T. Ishihara, N. Ohashi, H. Sato, K. Wakimoto, M. Shiotani, A. Oku, MGAT2 deficiency ameliorates high-fat diet- induced obesity and insulin resistance by inhibiting intestinal fat absorption in mice, Lipids in health and disease, 11 (2012) 75. [43] D.W. Nelson, Y. Gao, M.I. Yen, C.L. Yen, Intestine-specific deletion of acyl- CoA:monoacylglycerol acyltransferase (MGAT) 2 protects mice from diet-induced obesity and glucose intolerance, The Journal of biological chemistry, 289 (2014) 17338-17349.

[44] C. Okuma, T. Ohta, H. Tadaki, H. Hamada, T. Oda, H. Taniuchi, K. Yamanaka, Y. Ishii, Y. Ohe, S. Yata, J. Nishiu, Y. Aratsu, S. Oshida, S. Kume, M. Kakutani, JTP- 103237, a novel monoacylglycerol acyltransferase inhibitor, modulates fat absorption and prevents diet-induced obesity, European journal of pharmacology, 758 (2015) 72- 81. [45] Y. Gao, D.W. Nelson, T. Banh, M.I. Yen, C.L. Yen, Intestine-specific expression of MOGAT2 partially restores metabolic efficiency in Mogat2-deficient mice, Journal of lipid research, 54 (2013) 1644-1652. [46] M.R. Owen, C.C. Corstorphine, V.A. Zammit, Overt and latent activities of diacylglycerol acytransferase in rat liver microsomes: possible roles in very-low- density lipoprotein triacylglycerol secretion, The Biochemical journal, 323 ( Pt 1) (1997) 17-21. [47] M.R. Grigor, R.M. Bell, Separate monoacylglycerol and diacylglycerol acyltransferases function in intestinal triacylglycerol synthesis, Biochimica et biophysica acta, 712 (1982) 464-472. [48] S.J. Stone, M.C. Levin, P. Zhou, J. Han, T.C. Walther, R.V. Farese, Jr., The endoplasmic reticulum enzyme DGAT2 is found in mitochondria-associated membranes and has a mitochondrial targeting signal that promotes its association with mitochondria, The Journal of biological chemistry, 284 (2009) 5352-5361. [49] F. Wilfling, H. Wang, J.T. Haas, N. Krahmer, T.J. Gould, A. Uchida, J.X. Cheng, M. Graham, R. Christiano, F. Frohlich, X. Liu, K.K. Buhman, R.A. Coleman, J. Bewersdorf, R.V. Farese, Jr., T.C. Walther, Triacylglycerol synthesis enzymes mediate lipid droplet growth by relocalizing from the ER to lipid droplets, Developmental cell, 24 (2013) 384-399. [50] C.M. Mansbach, S.A. Siddiqi, The biogenesis of chylomicrons, Annual review of physiology, 72 (2010) 315-333. [51] I.J. Cartwright, J.A. Higgins, Direct evidence for a two-step assembly of ApoB48- containing lipoproteins in the lumen of the smooth endoplasmic reticulum of rabbit enterocytes, The Journal of biological chemistry, 276 (2001) 48048-48057. [52] M.M. Hussain, P. Rava, M. Walsh, M. Rana, J. Iqbal, Multiple functions of microsomal triglyceride transfer protein, Nutr Metab (Lond), 9 (2012) 14. [53] M.M. Hussain, A proposed model for the assembly of chylomicrons, Atherosclerosis, 148 (2000) 1-15. [54] J. Luchoomun, M.M. Hussain, Assembly and secretion of chylomicrons by differentiated Caco-2 cells. Nascent triglycerides and preformed phospholipids are preferentially used for lipoprotein assembly, The Journal of biological chemistry, 274 (1999) 19565-19572. [55] R. Lehner, J. Lian, A.D. Quiroga, Lumenal lipid metabolism: implications for lipoprotein assembly, Arteriosclerosis, thrombosis, and vascular biology, 32 (2012) 1087-1093. [56] C.M. Mansbach, 2nd, F. Gorelick, Development and physiological regulation of intestinal lipid absorption. II. Dietary lipid absorption, complex lipid synthesis, and the intracellular packaging and secretion of chylomicrons, American journal of physiology. Gastrointestinal and liver physiology, 293 (2007) G645-650.

[57] B. Jones, E.L. Jones, S.A. Bonney, H.N. Patel, A.R. Mensenkamp, S. Eichenbaum- Voline, M. Rudling, U. Myrdal, G. Annesi, S. Naik, N. Meadows, A. Quattrone, S.A. Islam, R.P. Naoumova, B. Angelin, R. Infante, E. Levy, C.C. Roy, P.S. Freemont, J. Scott, C.C. Shoulders, Mutations in a Sar1 GTPase of COPII vesicles are associated with lipid absorption disorders, Nature genetics, 34 (2003) 29-31. [58] I. Neeli, S.A. Siddiqi, S. Siddiqi, J. Mahan, W.S. Lagakos, B. Binas, T. Gheyi, J. Storch, C.M. Mansbach, 2nd, Liver fatty acid-binding protein initiates budding of pre-chylomicron transport vesicles from intestinal endoplasmic reticulum, The Journal of biological chemistry, 282 (2007) 17974-17984. [59] S. Siddiqi, C.M. Mansbach, 2nd, Phosphorylation of Sar1b protein releases liver fatty acid-binding protein from multiprotein complex in intestinal cytosol enabling it to bind to endoplasmic reticulum (ER) and bud the pre-chylomicron transport vesicle, The Journal of biological chemistry, 287 (2012) 10178-10188. [60] S.A. Siddiqi, C.M. Mansbach, 2nd, PKC zeta-mediated phosphorylation controls budding of the pre-chylomicron transport vesicle, Journal of cell science, 121 (2008) 2327-2338. [61] S.A. Siddiqi, J. Mahan, S. Siddiqi, F.S. Gorelick, C.M. Mansbach, 2nd, Vesicle- associated membrane protein 7 is expressed in intestinal ER, Journal of cell science, 119 (2006) 943-950. [62] S.A. Siddiqi, F.S. Gorelick, J.T. Mahan, C.M. Mansbach, 2nd, COPII proteins are required for Golgi fusion but not for endoplasmic reticulum budding of the pre- chylomicron transport vesicle, Journal of cell science, 116 (2003) 415-427. [63] S.A. Siddiqi, S. Siddiqi, J. Mahan, K. Peggs, F.S. Gorelick, C.M. Mansbach, 2nd, The identification of a novel endoplasmic reticulum to Golgi SNARE complex used by the prechylomicron transport vesicle, The Journal of biological chemistry, 281 (2006) 20974-20982. [64] N. Berriot-Varoqueaux, A.H. Dannoura, A. Moreau, N. Verthier, A. Sassolas, G. Cadiot, A. Lachaux, A. Munck, J. Schmitz, L.P. Aggerbeck, M.E. Samson-Bouma, Apolipoprotein B48 glycosylation in abetalipoproteinemia and Anderson's disease, Gastroenterology, 121 (2001) 1101-1108. [65] S.M. Sabesin, S. Frase, Electron microscopic studies of the assembly, intracellular transport, and secretion of chylomicrons by rat intestine, Journal of lipid research, 18 (1977) 496-511. [66] N.J. Christensen, C.E. Rubin, M.C. Cheung, J.J. Albers, Ultrastructural immunolocalization of apolipoprotein B within human jejunal absorptive cells, Journal of lipid research, 24 (1983) 1229-1242. [67] K. Choe, J.Y. Jang, I. Park, Y. Kim, S. Ahn, D.Y. Park, Y.K. Hong, K. Alitalo, G.Y. Koh, P. Kim, Intravital imaging of intestinal lacteals unveils lipid drainage through contractility, The Journal of clinical investigation, 125 (2015) 4042-4052. [68] K.R. Feingold, C. Grunfeld, Introduction to Lipids and Lipoproteins, in: L.J. De Groot, P. Beck-Peccoz, G. Chrousos, K. Dungan, A. Grossman, J.M. Hershman, C. Koch, R. McLachlan, M. New, R. Rebar, F. Singer, A. Vinik, M.O. Weickert (Eds.) Endotext, MDText.com, Inc., South Dartmouth MA, 2000. [69] E. Levy, Insights from human congenital disorders of intestinal lipid metabolism, Journal of lipid research, 56 (2015) 945-962.

[70] R. Bartz, W.H. Li, B. Venables, J.K. Zehmer, M.R. Roth, R. Welti, R.G. Anderson, P. Liu, K.D. Chapman, Lipidomics reveals that adiposomes store ether lipids and mediate phospholipid traffic, Journal of lipid research, 48 (2007) 837-847. [71] W.S. Blaner, S.M. O'Byrne, N. Wongsiriroj, J. Kluwe, D.M. D'Ambrosio, H. Jiang, R.F. Schwabe, E.M. Hillman, R. Piantedosi, J. Libien, Hepatic stellate cell lipid droplets: a specialized lipid droplet for retinoid storage, Biochimica et biophysica acta, 1791 (2009) 467-473. [72] T.C. Walther, R.V. Farese, Jr., Lipid droplets and cellular lipid metabolism, Annual review of biochemistry, 81 (2012) 687-714. [73] A.S. Greenberg, J.J. Egan, S.A. Wek, N.B. Garty, E.J. Blanchette-Mackie, C. Londos, Perilipin, a major hormonally regulated adipocyte-specific phosphoprotein associated with the periphery of lipid storage droplets, The Journal of biological chemistry, 266 (1991) 11341-11346. [74] P.E. Bickel, J.T. Tansey, M.A. Welte, PAT proteins, an ancient family of lipid droplet proteins that regulate cellular lipid stores, Biochimica et biophysica acta, 1791 (2009) 419-440. [75] K.A. Cho, P.B. Kang, PLIN2 inhibits insulin-induced glucose uptake in myoblasts through the activation of the NLRP3 inflammasome, International journal of molecular medicine, 36 (2015) 839-844. [76] S. Senthivinayagam, A.L. McIntosh, K.C. Moon, B.P. Atshaves, Plin2 inhibits cellular glucose uptake through interactions with SNAP23, a SNARE complex protein, PloS one, 8 (2013) e73696. [77] N. Kory, R.V. Farese, Jr., T.C. Walther, Targeting Fat: Mechanisms of Protein Localization to Lipid Droplets, Trends in cell biology, 26 (2016) 535-546. [78] N. Kory, A.R. Thiam, R.V. Farese, Jr., T.C. Walther, Protein Crowding Is a Determinant of Lipid Droplet Protein Composition, Developmental cell, 34 (2015) 351-363. [79] A.D. Barbosa, D.B. Savage, S. Siniossoglou, Lipid droplet-organelle interactions: emerging roles in lipid metabolism, Current opinion in cell biology, 35 (2015) 91-97. [80] H. Robenek, O. Hofnagel, I. Buers, M.J. Robenek, D. Troyer, N.J. Severs, Adipophilin-enriched domains in the ER membrane are sites of lipid droplet biogenesis, Journal of cell science, 119 (2006) 4215-4224. [81] F. Wilfling, A.R. Thiam, M.J. Olarte, J. Wang, R. Beck, T.J. Gould, E.S. Allgeyer, F. Pincet, J. Bewersdorf, R.V. Farese, Jr., T.C. Walther, Arf1/COPI machinery acts directly on lipid droplets and enables their connection to the ER for protein targeting, eLife, 3 (2014) e01607. [82] W. Xu, L. Wu, M. Yu, F.J. Chen, M. Arshad, X. Xia, H. Ren, J. Yu, L. Xu, D. Xu, J.Z. Li, P. Li, L. Zhou, Differential Roles of Cell Death-inducing DNA Fragmentation Factor-alpha-like Effector (CIDE) Proteins in Promoting Lipid Droplet Fusion and Growth in Subpopulations of Hepatocytes, The Journal of biological chemistry, 291 (2016) 4282-4293. [83] L. Xu, L. Zhou, P. Li, CIDE proteins and lipid metabolism, Arteriosclerosis, thrombosis, and vascular biology, 32 (2012) 1094-1098.

[84] A.S. Rambold, S. Cohen, J. Lippincott-Schwartz, Fatty acid trafficking in starved cells: regulation by lipid droplet lipolysis, autophagy, and mitochondrial fusion dynamics, Developmental cell, 32 (2015) 678-692. [85] R. Singh, S. Kaushik, Y. Wang, Y. Xiang, I. Novak, M. Komatsu, K. Tanaka, A.M. Cuervo, M.J. Czaja, Autophagy regulates lipid metabolism, Nature, 458 (2009) 1131- 1135. [86] K.T. Ong, M.T. Mashek, S.Y. Bu, A.S. Greenberg, D.G. Mashek, Adipose triglyceride lipase is a major hepatic lipase that regulates triacylglycerol turnover and fatty acid signaling and partitioning, Hepatology (Baltimore, Md.), 53 (2011) 116-126. [87] S. Cermelli, Y. Guo, S.P. Gross, M.A. Welte, The lipid-droplet proteome reveals that droplets are a protein-storage depot, Current biology : CB, 16 (2006) 1783-1795. [88] Y. Ohsaki, J. Cheng, A. Fujita, T. Tokumoto, T. Fujimoto, Cytoplasmic lipid droplets are sites of convergence of proteasomal and autophagic degradation of apolipoprotein B, Molecular biology of the cell, 17 (2006) 2674-2683. [89] M.A. Welte, Proteins under new management: lipid droplets deliver, Trends in cell biology, 17 (2007) 363-369. [90] F. Wilfling, J.T. Haas, T.C. Walther, R.V. Farese, Jr., Lipid droplet biogenesis, Current opinion in cell biology, 29 (2014) 39-45. [91] P. Bostrom, L. Andersson, M. Rutberg, J. Perman, U. Lidberg, B.R. Johansson, J. Fernandez-Rodriguez, J. Ericson, T. Nilsson, J. Boren, S.O. Olofsson, SNARE proteins mediate fusion between cytosolic lipid droplets and are implicated in insulin sensitivity, Nature cell biology, 9 (2007) 1286-1293. [92] L. Kuerschner, C. Moessinger, C. Thiele, Imaging of lipid biosynthesis: how a neutral lipid enters lipid droplets, Traffic (Copenhagen, Denmark), 9 (2008) 338-352. [93] A. Pol, S.P. Gross, R.G. Parton, Review: biogenesis of the multifunctional lipid droplet: lipids, proteins, and sites, The Journal of cell biology, 204 (2014) 635-646. [94] F. Beilstein, V. Carriere, A. Leturque, S. Demignot, Characteristics and functions of lipid droplets and associated proteins in enterocytes, Experimental cell research, 340 (2016) 172-179. [95] F. Beilstein, J. Bouchoux, M. Rousset, S. Demignot, Proteomic analysis of lipid droplets from Caco-2/TC7 enterocytes identifies novel modulators of lipid secretion, PloS one, 8 (2013) e53017. [96] T. D'Aquila, D. Sirohi, J.M. Grabowski, V.E. Hedrick, L.N. Paul, A.S. Greenberg, R.J. Kuhn, K.K. Buhman, Characterization of the proteome of cytoplasmic lipid droplets in mouse enterocytes after a dietary fat challenge, PloS one, 10 (2015) e0126823. [97] N. Krahmer, Y. Guo, F. Wilfling, M. Hilger, S. Lingrell, K. Heger, H.W. Newman, M. Schmidt-Supprian, D.E. Vance, M. Mann, R.V. Farese, Jr., T.C. Walther, Phosphatidylcholine synthesis for lipid droplet expansion is mediated by localized activation of CTP:phosphocholine cytidylyltransferase, Cell metabolism, 14 (2011) 504-515. [98] A. Penno, G. Hackenbroich, C. Thiele, Phospholipids and lipid droplets, Biochimica et biophysica acta, 1831 (2013) 589-594. [99] S.G. Young, R. Zechner, Biochemistry and pathophysiology of intravascular and intracellular lipolysis, Genes & development, 27 (2013) 459-484.

[100] R. Singh, A.M. Cuervo, Lipophagy: connecting autophagy and lipid metabolism, International journal of cell biology, 2012 (2012) 282041. [101] R. Zechner, P.C. Kienesberger, G. Haemmerle, R. Zimmermann, A. Lass, Adipose triglyceride lipase and the lipolytic catabolism of cellular fat stores, Journal of lipid research, 50 (2009) 3-21. [102] J.H. Lee, J. Kong, J.Y. Jang, J.S. Han, Y. Ji, J. Lee, J.B. Kim, Lipid droplet protein LID-1 mediates ATGL-1-dependent lipolysis during fasting in Caenorhabditis elegans, Molecular and cellular biology, 34 (2014) 4165-4176. [103] G. Haemmerle, A. Lass, R. Zimmermann, G. Gorkiewicz, C. Meyer, J. Rozman, G. Heldmaier, R. Maier, C. Theussl, S. Eder, D. Kratky, E.F. Wagner, M. Klingenspor, G. Hoefler, R. Zechner, Defective lipolysis and altered energy metabolism in mice lacking adipose triglyceride lipase, Science (New York, N.Y.), 312 (2006) 734-737. [104] A. Lass, R. Zimmermann, G. Haemmerle, M. Riederer, G. Schoiswohl, M. Schweiger, P. Kienesberger, J.G. Strauss, G. Gorkiewicz, R. Zechner, Adipose triglyceride lipase-mediated lipolysis of cellular fat stores is activated by CGI-58 and defective in Chanarin-Dorfman Syndrome, Cell metabolism, 3 (2006) 309-319. [105] C. Lefevre, F. Jobard, F. Caux, B. Bouadjar, A. Karaduman, R. Heilig, H. Lakhdar, A. Wollenberg, J.L. Verret, J. Weissenbach, M. Ozguc, M. Lathrop, J.F. Prud'homme, J. Fischer, Mutations in CGI-58, the gene encoding a new protein of the esterase/lipase/thioesterase subfamily, in Chanarin-Dorfman syndrome, American journal of human genetics, 69 (2001) 1002-1012. [106] I. Chanarin, A. Patel, G. Slavin, E.J. Wills, T.M. Andrews, G. Stewart, Neutral- lipid storage disease: a new disorder of lipid metabolism, British medical journal, 1 (1975) 553-555. [107] F.P. Radner, I.E. Streith, G. Schoiswohl, M. Schweiger, M. Kumari, T.O. Eichmann, G. Rechberger, H.C. Koefeler, S. Eder, S. Schauer, H.C. Theussl, K. Preiss-Landl, A. Lass, R. Zimmermann, G. Hoefler, R. Zechner, G. Haemmerle, Growth retardation, impaired triacylglycerol catabolism, hepatic steatosis, and lethal skin barrier defect in mice lacking comparative gene identification-58 (CGI-58), The Journal of biological chemistry, 285 (2010) 7300-7311. [108] C. Holm, T. Osterlund, H. Laurell, J.A. Contreras, Molecular mechanisms regulating hormone-sensitive lipase and lipolysis, Annu Rev Nutr, 20 (2000) 365- 393. [109] J. Osuga, S. Ishibashi, T. Oka, H. Yagyu, R. Tozawa, A. Fujimoto, F. Shionoiri, N. Yahagi, F.B. Kraemer, O. Tsutsumi, N. Yamada, Targeted disruption of hormone- sensitive lipase results in male sterility and adipocyte hypertrophy, but not in obesity, Proceedings of the National Academy of Sciences of the United States of America, 97 (2000) 787-792. [110] G. Haemmerle, R. Zimmermann, M. Hayn, C. Theussl, G. Waeg, E. Wagner, W. Sattler, T.M. Magin, E.F. Wagner, R. Zechner, Hormone-sensitive lipase deficiency in mice causes diglyceride accumulation in adipose tissue, muscle, and testis, The Journal of biological chemistry, 277 (2002) 4806-4815.

[111] J.S. Albert, L.M. Yerges-Armstrong, R.B. Horenstein, T.I. Pollin, U.T. Sreenivasan, S. Chai, W.S. Blaner, S. Snitker, J.R. O'Connell, D.W. Gong, R.J. Breyer, 3rd, A.S. Ryan, J.C. McLenithan, A.R. Shuldiner, C. Sztalryd, C.M. Damcott, Null mutation in hormone-sensitive lipase gene and risk of type 2 diabetes, The New England journal of medicine, 370 (2014) 2307-2315. [112] T. Sakurada, A. Noma, Subcellular localization and some properties of monoacylglycerol lipase in rat adipocytes, Journal of biochemistry, 90 (1981) 1413- 1419. [113] T.P. Dinh, D. Carpenter, F.M. Leslie, T.F. Freund, I. Katona, S.L. Sensi, S. Kathuria, D. Piomelli, Brain monoglyceride lipase participating in endocannabinoid inactivation, Proceedings of the National Academy of Sciences of the United States of America, 99 (2002) 10819-10824. [114] G. Fredrikson, H. Tornqvist, P. Belfrage, Hormone-sensitive lipase and monoacylglycerol lipase are both required for complete degradation of adipocyte triacylglycerol, Biochimica et biophysica acta, 876 (1986) 288-293. [115] R. Zechner, R. Zimmermann, T.O. Eichmann, S.D. Kohlwein, G. Haemmerle, A. Lass, F. Madeo, FAT SIGNALS--lipases and lipolysis in lipid metabolism and signaling, Cell metabolism, 15 (2012) 279-291. [116] B.N. Reid, G.P. Ables, O.A. Otlivanchik, G. Schoiswohl, R. Zechner, W.S. Blaner, I.J. Goldberg, R.F. Schwabe, S.C. Chua, Jr., L.S. Huang, Hepatic overexpression of hormone-sensitive lipase and adipose triglyceride lipase promotes fatty acid oxidation, stimulates direct release of free fatty acids, and ameliorates steatosis, The Journal of biological chemistry, 283 (2008) 13087-13099. [117] J.I. Trickett, D.D. Patel, B.L. Knight, E.D. Saggerson, G.F. Gibbons, R.J. Pease, Characterization of the rodent genes for arylacetamide deacetylase, a putative microsomal lipase, and evidence for transcriptional regulation, The Journal of biological chemistry, 276 (2001) 39522-39532. [118] M. Nourbakhsh, D.N. Douglas, C.H. Pu, J.T. Lewis, T. Kawahara, L.F. Lisboa, E. Wei, S. Asthana, A.D. Quiroga, L.M. Law, C. Chen, W.R. Addison, R. Nelson, M. Houghton, R. Lehner, N.M. Kneteman, Arylacetamide deacetylase: a novel host factor with important roles in the lipolysis of cellular triacylglycerol stores, VLDL assembly and HCV production, Journal of hepatology, 59 (2013) 336-343. [119] K.G. Soni, R. Lehner, P. Metalnikov, P. O'Donnell, M. Semache, W. Gao, K. Ashman, A.V. Pshezhetsky, G.A. Mitchell, Carboxylesterase 3 (EC 3.1.1.1) is a major adipocyte lipase, The Journal of biological chemistry, 279 (2004) 40683- 40689. [120] J. Lian, E. Wei, S.P. Wang, A.D. Quiroga, L. Li, A. Di Pardo, J. van der Veen, S. Sipione, G.A. Mitchell, R. Lehner, Liver specific inactivation of carboxylesterase 3/triacylglycerol hydrolase decreases blood lipids without causing severe steatosis in mice, Hepatology (Baltimore, Md.), 56 (2012) 2154-2162. [121] J.L. Goldstein, S.E. Dana, J.R. Faust, A.L. Beaudet, M.S. Brown, Role of lysosomal acid lipase in the metabolism of plasma low density lipoprotein. Observations in cultured fibroblasts from a patient with cholesteryl ester storage disease, The Journal of biological chemistry, 250 (1975) 8487-8495.

[122] R.A. Anderson, G.M. Bryson, J.S. Parks, Lysosomal acid lipase mutations that determine phenotype in Wolman and cholesterol ester storage disease, Molecular genetics and metabolism, 68 (1999) 333-345. [123] A. Tanaka, [Acid lipase deficiency: Wolman disease and cholesteryl ester storage disease], Nihon rinsho. Japanese journal of clinical medicine, 53 (1995) 3004-3008. [124] H. Du, M. Heur, M. Duanmu, G.A. Grabowski, D.Y. Hui, D.P. Witte, J. Mishra, Lysosomal acid lipase-deficient mice: depletion of white and brown fat, severe hepatosplenomegaly, and shortened life span, Journal of lipid research, 42 (2001) 489-500. [125] P. Tso, S.R. Gollamudi, Pluronic L-81: a potent inhibitor of the transport of intestinal chylomicrons, The American journal of physiology, 247 (1984) G32-36. [126] J. Zhu, B. Lee, K.K. Buhman, J.X. Cheng, A dynamic, cytoplasmic triacylglycerol pool in enterocytes revealed by ex vivo and in vivo coherent anti-Stokes Raman scattering imaging, Journal of lipid research, 50 (2009) 1080-1089. [127] J. Bouchoux, F. Beilstein, T. Pauquai, I.C. Guerrera, D. Chateau, N. Ly, M. Alqub, C. Klein, J. Chambaz, M. Rousset, J.M. Lacorte, E. Morel, S. Demignot, The proteome of cytosolic lipid droplets isolated from differentiated Caco-2/TC7 enterocytes reveals cell-specific characteristics, Biology of the cell / under the auspices of the European Cell Biology Organization, 103 (2011) 499-517. [128] A. Seyer, M. Cantiello, J. Bertrand-Michel, V. Roques, M. Nauze, V. Bezirard, X. Collet, D. Touboul, A. Brunelle, C. Comera, Lipidomic and spatio-temporal imaging of fat by mass spectrometry in mice duodenum during lipid digestion, PloS one, 8 (2013) e58224. [129] B. Lee, J. Zhu, N.E. Wolins, J.X. Cheng, K.K. Buhman, Differential association of adipophilin and TIP47 proteins with cytoplasmic lipid droplets in mouse enterocytes during dietary fat absorption, Biochimica et biophysica acta, 1791 (2009) 1173-1180. [130] C. Sztalryd, A.R. Kimmel, Perilipins: lipid droplet coat proteins adapted for tissue- specific energy storage and utilization, and lipid cytoprotection, Biochimie, 96 (2014) 96-101. [131] D.N. Frank, E.S. Bales, J. Monks, M.J. Jackman, P.S. MacLean, D. Ir, C.E. Robertson, D.J. Orlicky, J.L. McManaman, Perilipin-2 Modulates Lipid Absorption and Microbiome Responses in the Mouse Intestine, PloS one, 10 (2015) e0131944. [132] L.L. Listenberger, X. Han, S.E. Lewis, S. Cases, R.V. Farese, Jr., D.S. Ory, J.E. Schaffer, Triglyceride accumulation protects against fatty acid-induced lipotoxicity, Proceedings of the National Academy of Sciences of the United States of America, 100 (2003) 3077-3082. [133] D.J. Fretland, S.W. Djuric, T.S. Gaginella, Eicosanoids and inflammatory bowel disease: regulation and prospects for therapy, Prostaglandins, leukotrienes, and essential fatty acids, 41 (1990) 215-233. [134] W.F. Stenson, Role of eicosanoids as mediators of inflammation in inflammatory bowel disease, Scandinavian journal of gastroenterology. Supplement, 172 (1990) 13-18.

[135] J. Fichna, M. Bawa, G.A. Thakur, R. Tichkule, A. Makriyannis, D.M. McCafferty, K.A. Sharkey, M. Storr, Cannabinoids alleviate experimentally induced intestinal inflammation by acting at central and peripheral receptors, PloS one, 9 (2014) e109115. [136] A.A. Izzo, F. Fezza, R. Capasso, T. Bisogno, L. Pinto, T. Iuvone, G. Esposito, N. Mascolo, V. Di Marzo, F. Capasso, Cannabinoid CB1-receptor mediated regulation of gastrointestinal motility in mice in a model of intestinal inflammation, British journal of pharmacology, 134 (2001) 563-570. [137] N.V. DiPatrizio, G. Astarita, G. Schwartz, X. Li, D. Piomelli, Endocannabinoid signal in the gut controls dietary fat intake, Proceedings of the National Academy of Sciences of the United States of America, 108 (2011) 12904-12908. [138] A.A. Izzo, F. Piscitelli, R. Capasso, G. Aviello, B. Romano, F. Borrelli, S. Petrosino, V. Di Marzo, Peripheral endocannabinoid dysregulation in obesity: relation to intestinal motility and energy processing induced by food deprivation and re-feeding, British journal of pharmacology, 158 (2009) 451-461. [139] M. Watford, P. Lund, H.A. Krebs, Isolation and metabolic characteristics of rat and chicken enterocytes, The Biochemical journal, 178 (1979) 589-596. [140] H.G. Windmueller, A.E. Spaeth, Respiratory fuels and nitrogen metabolism in vivo in small intestine of fed rats. Quantitative importance of glutamine, glutamate, and aspartate, The Journal of biological chemistry, 255 (1980) 107-112. [141] H.G. Windmueller, A.E. Spaeth, Uptake and metabolism of plasma glutamine by the small intestine, The Journal of biological chemistry, 249 (1974) 5070-5079. [142] H.G. Windmueller, A.E. Spaeth, Identification of ketone bodies and glutamine as the major respiratory fuels in vivo for postabsorptive rat small intestine, The Journal of biological chemistry, 253 (1978) 69-76. [143] P.H. Duee, B. Darcy-Vrillon, F. Blachier, M.T. Morel, Fuel selection in intestinal cells, The Proceedings of the Nutrition Society, 54 (1995) 83-94. [144] J.H. Spalinger, E.G. Seidman, D. Menard, E. Levy, Endogenous lipase activity in Caco-2 cells, Biochimica et biophysica acta, 1393 (1998) 119-127. [145] S. Obrowsky, P.G. Chandak, J.V. Patankar, S. Povoden, S. Schlager, E.E. Kershaw, J.G. Bogner-Strauss, G. Hoefler, S. Levak-Frank, D. Kratky, Adipose triglyceride lipase is a TG hydrolase of the small intestine and regulates intestinal PPARalpha signaling, Journal of lipid research, 54 (2013) 425-435. [146] P. Xie, F. Guo, Y. Ma, H. Zhu, F. Wang, B. Xue, H. Shi, J. Yang, L. Yu, Intestinal Cgi-58 deficiency reduces postprandial lipid absorption, PloS one, 9 (2014) e91652. [147] J. Grober, S. Lucas, M. Sorhede-Winzell, I. Zaghini, A. Mairal, J.A. Contreras, P. Besnard, C. Holm, D. Langin, Hormone-sensitive lipase is a cholesterol esterase of the intestinal mucosa, The Journal of biological chemistry, 278 (2003) 6510-6515. [148] S. Obrowsky, P.G. Chandak, J.V. Patankar, T. Pfeifer, S. Povoden, R. Schreiber, G. Haemmerle, S. Levak-Frank, D. Kratky, Cholesteryl ester accumulation and accelerated cholesterol absorption in intestine-specific hormone sensitive lipase-null mice, Biochimica et biophysica acta, 1821 (2012) 1406-1414.

[149] J.D. Douglass, Y.X. Zhou, A. Wu, J.A. Zadrogra, A.M. Gajda, A.I. Lackey, W. Lang, K.M. Chevalier, S.W. Sutton, S.P. Zhang, C.M. Flores, M.A. Connelly, J. Storch, Global deletion of MGL in mice delays lipid absorption and alters energy homeostasis and diet-induced obesity, Journal of lipid research, 56 (2015) 1153-1171. [150] U. Taschler, F.P. Radner, C. Heier, R. Schreiber, M. Schweiger, G. Schoiswohl, K. Preiss-Landl, D. Jaeger, B. Reiter, H.C. Koefeler, J. Wojciechowski, C. Theussl, J.M. Penninger, A. Lass, G. Haemmerle, R. Zechner, R. Zimmermann, Monoglyceride lipase deficiency in mice impairs lipolysis and attenuates diet-induced insulin resistance, The Journal of biological chemistry, 286 (2011) 17467-17477. [151] S.H. Chon, J.D. Douglass, Y.X. Zhou, N. Malik, J.L. Dixon, A. Brinker, L. Quadro, J. Storch, Over-expression of monoacylglycerol lipase (MGL) in small intestine alters endocannabinoid levels and whole body energy balance, resulting in obesity, PloS one, 7 (2012) e43962. [152] R.D. Jones, A.M. Taylor, E.Y. Tong, J.J. Repa, Carboxylesterases are uniquely expressed among tissues and regulated by nuclear hormone receptors in the mouse, Drug metabolism and disposition: the biological fate of chemicals, 41 (2013) 40-49. [153] Y. Li, M. Zalzala, K. Jadhav, Y. Xu, T. Kasumov, L. Yin, Y. Zhang, Carboxylesterase 2 prevents liver steatosis by modulating lipolysis, endoplasmic reticulum stress, and lipogenesis and is regulated by hepatocyte nuclear factor 4 alpha in mice, Hepatology (Baltimore, Md.), 63 (2016) 1860-1874. [154] S.A. Khaldoun, M.A. Emond-Boisjoly, D. Chateau, V. Carriere, M. Lacasa, M. Rousset, S. Demignot, E. Morel, Autophagosomes contribute to intracellular lipid distribution in enterocytes, Molecular biology of the cell, 25 (2014) 118-132. [155] Y. Ding, S. Zhang, L. Yang, H. Na, P. Zhang, H. Zhang, Y. Wang, Y. Chen, J. Yu, C. Huo, S. Xu, M. Garaiova, Y. Cong, P. Liu, Isolating lipid droplets from multiple species, Nature protocols, 8 (2013) 43-51. [156] A.E. Crunk, J. Monks, A. Murakami, M. Jackman, P.S. Maclean, M. Ladinsky, E.S. Bales, S. Cain, D.J. Orlicky, J.L. McManaman, Dynamic regulation of hepatic lipid droplet properties by diet, PloS one, 8 (2013) e67631. [157] H. Na, P. Zhang, Y. Ding, L. Yang, Y. Wang, H. Zhang, Z. Xie, F. Yang, S. Cichello, P. Liu, Proteomic studies of isolated lipid droplets from bacteria, C. elegans, and mammals, Methods in cell biology, 116 (2013) 1-14. [158] O. Chahrour, D. Cobice, J. Malone, Stable isotope labelling methods in mass spectrometry-based quantitative proteomics, Journal of pharmaceutical and biomedical analysis, 113 (2015) 2-20. [159] J.W. Wong, G. Cagney, An overview of label-free quantitation methods in proteomics by mass spectrometry, Methods in molecular biology (Clifton, N.J.), 604 (2010) 273-283. [160] S. Daemen, M.A. van Zandvoort, S.H. Parekh, M.K. Hesselink, Microscopy tools for the investigation of intracellular lipid storage and dynamics, Molecular metabolism, 5 (2016) 153-163. [161] Y. Ding, Y. Wu, R. Zeng, K. Liao, Proteomic profiling of lipid droplet-associated proteins in primary adipocytes of normal and obese mouse, Acta biochimica et biophysica Sinica, 44 (2012) 394-406.

[162] S. Larsson, S. Resjo, M.F. Gomez, P. James, C. Holm, Characterization of the lipid droplet proteome of a clonal insulin-producing beta-cell line (INS-1 832/13), Journal of proteome research, 11 (2012) 1264-1273. [163] P. Liu, Y. Ying, Y. Zhao, D.I. Mundy, M. Zhu, R.G. Anderson, Chinese hamster ovary K2 cell lipid droplets appear to be metabolic organelles involved in membrane traffic, The Journal of biological chemistry, 279 (2004) 3787-3792. [164] S. Sato, M. Fukasawa, Y. Yamakawa, T. Natsume, T. Suzuki, I. Shoji, H. Aizaki, T. Miyamura, M. Nishijima, Proteomic profiling of lipid droplet proteins in hepatoma cell lines expressing hepatitis C virus core protein, Journal of biochemistry, 139 (2006) 921-930. [165] L. Yang, Y. Ding, Y. Chen, S. Zhang, C. Huo, Y. Wang, J. Yu, P. Zhang, H. Na, H. Zhang, Y. Ma, P. Liu, The proteomics of lipid droplets: structure, dynamics, and functions of the organelle conserved from bacteria to humans, Journal of lipid research, 53 (2012) 1245-1253. [166] M. Beller, D. Riedel, L. Jansch, G. Dieterich, J. Wehland, H. Jackle, R.P. Kuhnlein, Characterization of the Drosophila lipid droplet subproteome, Molecular & cellular proteomics : MCP, 5 (2006) 1082-1094. [167] D.L. Brasaemle, G. Dolios, L. Shapiro, R. Wang, Proteomic analysis of proteins associated with lipid droplets of basal and lipolytically stimulated 3T3-L1 adipocytes, The Journal of biological chemistry, 279 (2004) 46835-46842. [168] D.B. Zilversmit, Atherogenesis: a postprandial phenomenon, Circulation, 60 (1979) 473-485. [169] K.N. Frayn, Insulin resistance, impaired postprandial lipid metabolism and abdominal obesity. A deadly triad, Medical principles and practice : international journal of the Kuwait University, Health Science Centre, 11 Suppl 2 (2002) 31-40. [170] G.F. Lewis, N.M. O'Meara, P.A. Soltys, J.D. Blackman, P.H. Iverius, A.F. Druetzler, G.S. Getz, K.S. Polonsky, Postprandial lipoprotein metabolism in normal and obese subjects: comparison after the vitamin A fat-loading test, The Journal of clinical endocrinology and metabolism, 71 (1990) 1041-1050. [171] C. Couillard, N. Bergeron, D. Prud'homme, J. Bergeron, A. Tremblay, C. Bouchard, P. Mauriege, J.P. Despres, Postprandial triglyceride response in visceral obesity in men, Diabetes, 47 (1998) 953-960. [172] L. Wideman, L.A. Kaminsky, M.H. Whaley, Postprandial lipemia in obese men with abdominal fat patterning, The Journal of sports medicine and physical fitness, 36 (1996) 204-210. [173] Y. Nabeno-Kaeriyama, Y. Fukuchi, S. Hayashi, T. Kimura, A. Tanaka, M. Naito, Delayed postprandial metabolism of triglyceride-rich lipoproteins in obese young men compared to lean young men, Clinica chimica acta; international journal of clinical chemistry, 411 (2010) 1694-1699. [174] A. Veilleux, E. Grenier, P. Marceau, A.C. Carpentier, D. Richard, E. Levy, Intestinal lipid handling: evidence and implication of insulin signaling abnormalities in human obese subjects, Arteriosclerosis, thrombosis, and vascular biology, 34 (2014) 644-653.

[175] A. Uchida, M.C. Whitsitt, T. Eustaquio, M.N. Slipchenko, J.F. Leary, J.X. Cheng, K.K. Buhman, Reduced triglyceride secretion in response to an acute dietary fat challenge in obese compared to lean mice, Frontiers in physiology, 3 (2012) 26. [176] J.D. Douglass, N. Malik, S.H. Chon, K. Wells, Y.X. Zhou, A.S. Choi, L.B. Joseph, J. Storch, Intestinal mucosal triacylglycerol accumulation secondary to decreased lipid secretion in obese and high fat fed mice, Frontiers in physiology, 3 (2012) 25.

CHAPTER 2. CHARACTERIZATION OF THE PROTEOME OF CYTOPLASMIC LIPID DROPLETS IN MOUSE ENTEROCYTES AFTER A DIETARY FAT CHALLENGE

2.1 Abstract

Dietary fat absorption by the small intestine is a multistep process that regulates the uptake and delivery of essential nutrients and energy. One step of this process is the

temporary storage of dietary fat in cytoplasmic lipid droplets (CLDs). The storage and

mobilization of dietary fat is thought to be regulated by proteins that associate with the

CLD; however, mechanistic details of this process are currently unknown. In this study

we analyzed the proteome of CLDs isolated from enterocytes harvested from the small

intestine of mice following a dietary fat challenge. In this analysis we identified 181

proteins associated with the CLD fraction, of which 37 are associated with known lipid

related metabolic pathways. We confirmed the localization of several of these proteins on

or around the CLD through confocal and electron microscopy, including perilipin 3,

apolipoprotein A-IV, and acyl-CoA synthetase long-chain family member 5. The

identification of the enterocyte CLD proteome provides new insight into potential

regulators of CLD metabolism and the process of dietary fat absorption.

This work was published in PLOSone

T. D'Aquila, D. Sirohi, J.M. Grabowski, V.E. Hedrick, L.N. Paul, A.S. Greenberg, R.J. Kuhn, K.K. Buhman, Characterization of the proteome of cytoplasmic lipid droplets in mouse enterocytes after a dietary fat challenge, PloS one, 10 (2015) e0126823.

2.2 Introduction

Dietary fat is the most energy dense macronutrient consumed and is required for the absorption of essential fatty acids and other lipophilic nutrients including fat soluble vitamins. However, when present in excess, dietary fat increases the risk for chronic diseases such as cardiovascular disease and obesity [1-4]. Therefore, understanding the regulators of dietary fat absorption and metabolism is important for both the promotion of health and prevention of disease.

Dietary fat absorption by the small intestine is a multistep process.

Triacylglycerol (TAG) is hydrolyzed by pancreatic lipase in the intestinal lumen producing monoacylglycerol and free fatty acids. These digestive products are taken up by the absorptive cells of the intestine, enterocytes, where they are rapidly resynthesized to TAG. The TAG is then packaged in the core of a chylomicron for systemic delivery of nutrients throughout the body [3,5,6]. Alternatively, when fatty acids are present in excess, the newly synthesized TAG may be incorporated into cytoplasmic lipid droplets

(CLDs) within enterocytes. The size and number of CLDs within enterocytes increases and then decreases after consumption of dietary fat [7]; however, the factors that regulate

CLD synthesis and catabolism within enterocytes are relatively unknown. The partitioning of TAG into chylomicrons or CLDs is important for determining the amount and rate of fatty acids delivered systemically. Therefore, identification of factors that regulate enterocyte CLD metabolism is important for understanding the overall process of dietary fat absorption.

Proteins that associate with CLDs in various cell types have been shown to

regulate the synthesis and catabolism of CLDs [8-11]. Recently, our laboratory identified two CLD associated proteins, perilipin 2 (Plin2) and perilipin 3 (Plin3), on CLDs within enterocytes after a dietary fat challenge [12]. These well-established CLD associated proteins are thought to regulate lipolysis. In addition, mice deficient in mediators of lipolysis, including adipose triglyceride lipase [13] and abhydrolase domain containing 5

[14], have altered catabolism of CLDs in enterocytes. These results strongly suggest CLD associated proteins regulate the synthesis and catabolism of CLDs in enterocytes. The identification of additional CLD associated proteins within enterocytes has the potential to establish novel mediators of dietary fat absorption.

The objective of this study was to identify CLD associated proteins in enterocytes after a dietary fat challenge. To fulfill this objective, we isolated CLDs from the small intestine of mice two hours after an oil bolus and identified proteins involved in the

regulation of lipid trafficking and metabolism using a global proteomic approach. We

further validated the presence of selected proteins on or around CLDs by confocal and

immunoelectron microscopy.

2.3 Materials and methods

2.3.1 Ethics statement

We conducted the study in strict accordance with the recommendations in the

Guide for the Care and Use of Laboratory Animals of the National Institute of Health.

The protocol was approved by the Purdue Animal Care and Use Committee (PACUC#

1111000154). All effort was made to minimize pain and suffering of mice and the number of mice used.

2.3.2 Mice

C57BL/6 male mice from an in-house breeding colony were used for this study.

The mice were maintained on a chow diet (PicoLab 5053, Lab Diets, Richmond, IN,

USA) that consisted of 62.1% of calories from carbohydrate (starch), 24.7% from protein, and 13.2% from fat. The mice were housed in a temperature and humidity controlled facility with a 12 hour light/dark cycle (6AM/6PM) with ad libitum access to food and water.

2.3.3 Mouse procedure for lipid droplet isolation and proteomic analysis

Four, 5 month old male C57BL/6 mice were fasted for 4 hours at the beginning of the light cycle. An oral gavage of 200 µl olive oil was administered and two hours after

the oil bolus, the mice were euthanized via CO2 asphyxiation. The small intestine was

excised and divided into five equal length segments in relation to the stomach and labeled

sections 1-5. Segments 2 and 3, representing the jejunum, were used for analysis.

2.3.4 Enterocyte Isolation

Enterocytes were isolated from sections 2 and 3 of the small intestine as

previously described [12,15]. Briefly, the intestinal sections were washed in tissue buffer

(Hank's Balanced Salt Solution with 25 mM HEPES and 1% fetal calf serum) and then

placed in isolation buffer (Calcium and magnesium free Hank's Balanced Salt Solution

with 1.5mM EDTA). The intestine segments were incubated for 15 minutes at 37° C with

rotation. The sample was mixed briefly (vortexed) and the supernatant containing

enterocytes was removed and saved. The process was repeated and the supernatants

containing isolated enterocytes were combined.

2.3.5 CLD isolation

CLDs were isolated from enterocytes using a previously established sucrose

gradient ultracentrifugation protocol [16]. Enterocytes were lysed in ice cold sucrose lysis

buffer (175mM sucrose, 10 mM HEPES and 1 mM EDTA pH 7.4). Cells were disrupted

by passing through a 27 gauge, 1 inch needle, eight times. The resulting two mLs of cell lysate were carefully layered with six mLs of sucrose-free lysis buffer and centrifuged at

£ 20,000 x g at 4 £ C for two hours. After centrifugation, the sample was frozen at -80 C.

The frozen sample was sliced into seven sequential fractions which were approximately 1

cm in length.

2.3.6 TAG and protein analysis

TAG concentration of each fraction was determined using an L Type TG M assay

(Wako Diagnostics, Richmond VA, USA). Protein concentration of each fraction was determined by a BCA assay (Thermo Scientific, Rockford IL, USA). Additionally, a

Western Blot analysis was performed on each fraction. Briefly, samples from the

isolated fractions were delipidated using 10% sodium dodecyl sulfate, separated by

polyacrylamide gel electrophoresis, and transferred to a PVDF membrane. The

membranes were probed overnight at 4 £ C with antibodies for Plin3 (a gift from Dr. Perry

Bickel at the University of Texas Southwestern, Dallas TX, USA), glyceraldehyde 3-

phosphate dehydrogenase (Gapdh) (Abcam, Cambridge MA, USA), or calnexin (Cnx)

(Santa Cruz Biotechnology, Dallas TX, USA). These proteins were used as markers for

the CLD, cytosolic, or membrane fractions, respectively. The blots were then probed with

a LI-COR IR DYE 800CW secondary antibody and imaged using Odyssey CLx Infrared

imaging system (LI-COR Biosciences, Lincoln NE, USA).

2.3.7 In solution digestion and LC MS/MS

In preparation for proteomic analysis, the isolated CLD fractions were delipidated using 2:1 chloroform methanol and proteins precipitated using ice cold acetone. The

63 protein pellet was denatured using 8M urea and 10 mM DTT for 1.5 hours at 37°C. The sample was digested for 12 hours using trypsin (Sigma-Aldrich, St. Louis MO, USA) using a ratio of 1 µg trypsin to 50 µg protein. The reaction was quenched using trifluoroacetic acid. Tryptic peptides were separated on a nanoLC system (1100 Series

LC, Agilent Technologies, Santa Clara, CA). The peptides were loaded on the Agilent

300SB-C18 enrichment column for concentration and the enrichment column was switched into the nano-flow path after five minutes. Peptides were separated with a C18 reversed phase ZORBAX 300SB-C18 column. The column was connected to the emission tip and coupled to the nano-electrospray ionization (ESI) source of the high resolution hybrid ion trap mass spectrometer LTQ-Orbitrap LX (Thermo Fisher Scientific,

Waltham MA, USA). The LTQ-orbitrap mass spectrometer was operated in the data- dependent positive acquisition mode in which each full MS scan (30.000 resolving power) was followed by six MS/MS scans where the six most abundant molecular ions were selected and fragmented by collision induced dissociation (CID) using a normalized collision energy of 35%..

2.3.8 Data analysis and bioinformatics

The peak list files containing MS and MS/MS data were analyzed using

MaxQuant version 1.4.08[17]. For protein identification, the MS/MS data was searched against the Uniprot protein database which includes the Swissprot (manually annotated and reviewed) and TrEMBL (automatically annotated and not reviewed) databases [18].

The database was searched using the MASCOT search engine utilizing Andromeda as the

peptide search algorithm that is incorporated in the MaxQuant platform [19]. The search

was conducted using the following settings: trypsin cleavage with a maximum of two

missed cleavages, fixed modification of iodoethanol addition to cysteine, variable modification of oxidation of methionine. The MS mass tolerance was set at 4.5 ppm with a maximum number of five modifications. The false discovery rate was set at 0.01 for proteins and peptides and was run against a decoy revert database. Peptides required a minimum length of seven amino acids. The MS/MS tolerance was set at 0.1 Da for protein identification. The minimum score for modified and unmodified peptides was set at forty. At least two peptides were required for protein identification.

The bioinformatics and statistical package Perseus 1.4.1.3 was used to analyze the

MaxQuant output. Contaminants identified by Perseus, such as keratin, were removed from the analysis. For label free quantitation (LFQ), the intensity for each protein was transformed log2(x) and samples who had a peptide intensity below the level of detection were assigned a value of 14. We limited the analysis to proteins which were identified in at least 3 out of the 4 biological replicates.

Proteins identified were clustered by their Gene Ontology (GO) Term based on biological process or molecular function using the Database for Annotation, Visualization and Integrated Discovery (DAVID) v 6.7 [20,21]. Visualization of protein interactions

was accomplished using STRING version 9.1[22]. We used the confidence view with a

score of 0.4, indicating a medium confidence level. We did not limit the number of

interactions in the analysis.

2.3.9 Immunofluorescence Imaging

Four, C57BL/6 male mice were fasted for 4 hours at the beginning of the light

cycle and then administered a 200 µl olive oil bolus. A small (5mm) section of the

jejunum was harvested from the mice two hours after the dietary fat challenge and was

frozen in optimal cutting temperature embedding media. The tissue was stored at -80°C

until processed by immunofluorescence microscopy. The tissue was sliced into 10 µm

tissue sections, fixed in 2% paraformaldehyde, and permeabilized with 0.1% saponin.

The tissue was probed with previously validated antibodies for Plin3 [23] (a gift from Dr.

Perry Bickel at the University of Texas Southwestern, Dallas TX, USA), apolipoprotein

A-IV (ApoA-IV) [24] (a gift from Dr. Patrick Tso from the University of Cincinnati,

Reading OH, USA), and acyl-CoA synthetase long-chain family member 5 (Acsl5)

(personal communication with A. Greenberg) (a gift from Dr. Andrew Greenberg from

Tufts University, Boston MA, USA). The sections were also stained for neutral lipids

§¨© ¡¢£¤¥¦¤ -difluoro-1,3,5,7,8-pentamethyl-4-bora-3a,4a-diaza-s-indacene

(BODIPY) (Life technologies, Grand Island NY, USA), a secondary AlexaFluor antibody

(Life technologies), and for nuclei using 300 nM DAPI (Life technologies). The sections were imaged using a Nikon A1R confocal microscope (Nikon Instruments Inc., Melville

NY, USA). Image processing, co-localization analysis, and multidimensional imaging were conducted using NIS- Elements C acquisition and analysis software.

2.3.10 Transmission electron microscopy

2.3.10.1 Cardiac perfusion fixation and TEM imaging of jejunum

Two, C57BL/6 mice were fasted for 4 hours at the beginning of the light cycle.

An oral gavage of 200 µl olive oil was administered and two hours after the oil bolus, the mice were anesthetized using inhaled isoflurane and perfused with 1.5% glutaraldehyde in 0.1M sodium cacodylate via cardiac infusion. After fixation, a sample of the jejunum section of the small intestine was isolated, stained with osmium tetroxide, dehydrated, and embedded in resin. Ultrathin sections were stained with lead citrate and uranyl acetate and examined using a Tecnai T20 Transmission Electron Microscope (FEI,

Hillsboro OR, USA).

2.3.10.2 TEM imaging of isolated CLDs

A sample from the isolated CLD fraction was negatively stained using 2% w/v aqueous uranyl acetate. The isolated CLDs were imaged using a Tecnai T20

Transmission Electron Microscope (FEI).

2.3.10.3 Immunoelectron microscopy

Two, C57BL/6 mice were fasted for 4 hours at the beginning of the light cycle.

An oral gavage of 200 µl olive oil was administered and two hours after the oil bolus, the mice were anesthetized using inhaled isoflurane and perfused with 4% paraformaldehyde and 0.05% glutaraldehyde in 0.1M phosphate buffer via cardiac infusion. After fixation, a

sample of the jejunum section of the small intestine was isolated, dehydrated, and

embedded in resin. Ultrathin sections were probed with an antibody for Acsl5 (a gift from

Dr. Andy Greenberg), gold conjugated secondary antibody (Ted Pella Inc. Redding, CA), and stained with uranyl acetate. The samples were examined using Tecnai T20

Transmission Electron Microscope (FEI).

2.4 Results

2.4.1 Lipid accumulates in CLDs in enterocytes in response to a dietary fat challenge.

Neutral lipids accumulate in distinct regions of the enterocyte following an acute

olive oil bolus. Enterocytes from the jejunum section of the small intestine, two hours

after a dietary fat challenge, were stained with osmium tetroxide and examined by

transmission electron microscopy (Figure 1). The various fates of neutral lipids within

enterocytes following a dietary fat challenge are identified in this image. Neutral lipids

accumulate in large CLDs which are indicated by white asterisks. Also visible is the

Golgi apparatus (white arrows) which contains smaller chylomicron sized particles.

Additionally, the white areas between enterocytes contain secreted chylomicrons (black

cross). These results highlight the complex, multistep process of dietary fat absorption.

2.4.2 The CLD fraction is enriched with Plin3 and has a high TAG to protein ratio.

Enterocyte CLDs were isolated via density gradient ultracentrifugation using a

previously established protocol [16]. After centrifugation, the samples were frozen and

sliced into seven fractions, where fraction 1 contains CLDs (Figure 2a). To confirm the successful isolation of CLDs, the isolated fractions were subjected to Western blot analysis. Antibodies for Plin3, Gapdh, and Cnx, were used as markers of the CLD, cytosolic, and membrane fractions respectively. The CLD marker Plin3 was found in fractions 1, 4, 5, 6, and 7 (Figure 2c), representing the various fractions Plin3 has been previously described to locate [25]. Plin3 is found primarily in the cytosol when lipids are absent and translocate to the nascent CLD in the early stages of CLD formation. Gapdh was used as a cytosolic marker and was found primarily in the cytosolic fractions

(fractions 2-6). In addition, Gapdh was notably absent in fraction 1. Cnx was used as a marker for membranes and found primarily in fraction 7. To validate the presence of

CLDs, the TAG to protein ratio of each fraction was examined (Figure 2b). Fraction 1 contained a high concentration of TAG compared to protein, indicating the presence of

CLDs. Fraction 1 was also examined using negative staining electron microscopy which showed an abundance of round, CLD sized particles (Figure 2d). This data suggests that

CLDs were successfully isolated from mouse enterocytes.

2.4.3 Proteins were identified in the CLD fraction through LC MS/MS

To identify proteins that associate with the CLD fraction, we utilized a shotgun proteomic approach. Through high resolution tandem mass spectroscopy, we identified

181 proteins in at least 3 out of the 4 biological replicates (Data not shown). Proteins were identified by comparing the MS/MS spectra against the Uniprot database. The

Uniprot database combines FASTA files from the Swissprot and TrEMBL databases.

Using Maxquant’s LFQ [26], the relative abundance of each protein within each sample

was identified. It should be noted that the LFQ intensity is the result of multiple factors,

including the abundance of the protein within the sample and the ability of the peptides to

be ionized and detected. To identify the function of the proteins, we used the functional

annotation tool in DAVID. The Gene Ontology (GO) terms of molecular function and

biological process associated with each protein was identified and then GO terms were

clustered into broader classifications as described in Table 1. The percentage of proteins

categorized within broader classifications is found in Figure 3. On several occasions, the

GO terms associated with a protein would fall under multiple broader classifications.

Under these circumstances, a manual analysis of the protein’s function was conducted

using the Uniprot and NCBI databases. Based on the quantity and quality of reported

protein function, the protein was assigned into the most appropriate GO term and broader

classification.

2.4.4 Some identified proteins in the CLD fraction are associated with known lipid

metabolism related functions

Using DAVID’s functional analysis tool, 19% of the proteins identified in the

CLD fraction were found to be associated with a biological process or molecular function related to lipid metabolism. The relative abundance of these proteins was determined by

Maxquant’s LFQ (Table 2). Approximately half of these proteins have been identified in other CLD proteomic studies from diverse cell types (Table 2). We specifically highlighted proteins which have been previously identified associated with CLDs in an

intestinal cell model through proteomic analysis. A map of predicted interactions was generated using a STRING analysis (Figure 4) [22]. The resulting network shows several protein clusters including proteins responsible for lipoprotein synthesis and modifiers of fatty acids.

2.4.5 Immunofluorescence imaging of proteins identified in the CLD fraction supports

their CLD localization

To confirm the localization of ApoA-IV and Acsl5 on or around CLDs, we probed intestinal sections with antibodies for these proteins from four, male C57BL/6 mice, two hours after a 200 µl olive oil bolus. To visualize CLDs, we stained neutral lipids with BODIPY (Life technologies) and used immunofluorescence staining of a well-established marker of enterocyte CLDs, Plin3 [12]. Due to the process of fixation, neutral lipids are often depleted from CLDs; therefore, where applicable, both Plin3 and neutral lipid staining was used to enhance our ability to detect CLDs. The first protein investigated was ApoA-IV, a protein known regulate intestinal lipid absorption and chylomicron synthesis [27,28]. To confirm the localization of ApoA-IV to the area on or around the CLD, the signals from Plin3 (Figure 5A) and ApoA-IV (Figure 5B) were merged and overlaid with signals from nuclei (blue) and neutral lipid (orange) (Figure

5C). A region of interest was highlighted (white capped line) in Figure 5C. The intensity profile of all fluorescent signals along this line was generated (Figure 5D). The intensity signals from Plin3 and ApoA-IV overlap and form two peaks which flank the signal from neutral lipids. Additionally, the signal from ApoA-IV and Plin3 antibodies colocalize with a Pearson’s coefficient of r= 0.59 and a Mander’s overlap of r= 0.96 which indicates

a high degree of correlation. This indicates that ApoA-IV localizes to the area on or

around the CLD. A three dimensional volume view was generated from Z-series images

(Figure 5E) and illustrates the localization of Plin3 and ApoA-IV to the area on or around

the CLD forming a protein coat. The intensity profile, colocalization analysis, and

volume view was generated using NIS- Elements C software.

The proteomic analysis of isolated CLDs also identified Acsl5, an enzyme that

activates fatty acids by the addition of coenzyme A [29]. In this experiment, we were

unable to use Plin3 as a marker for CLD due to the fact the antibodies for Acsl5 and Plin3 were generated in the same host species, so CLDs are identified using only BODIPY.

Intestinal sections were probed with an antibody for Acsl5 (green) (Figure 6a), stained for

neutral lipids using BODIPY (orange) and nuclei using DAPI (blue) (Figure 6B). Acsl5

forms distinct ring like structures on or around CLDs as indicated by the white arrows. A

3D volume view was generated from Z series images (Figure 6C) showing the

localization of Acsl5 to the area on or around CLDs. Additionally, to confirm the

localization of Acsl5 to CLDs in enterocytes following a 200 µl olive oil bolus, we

performed immunogold labeling of ultra-thin sections of the jejunum section of the small

intestine. An electromicrograph (Figure 6D) shows an enterocyte with large white CLDs

present. Areas of interest are highlighted by colored arrows which correspond with

figures 6 E-H. Acsl5 conjugated gold particles are found on the CLD surface which are

indicated by the white arrows (Figure 6E-G). In figure 6E, Acsl5 can be seen localized to

the CLD as opposed to the endoplasmic reticulum membrane which can been seen on the

72 left hand side of the CLD. Additionally, Acsl5 can be found localized to the mitochondria

(Figure 6H) which has been previously reported in Caco-2 cells[30,31] and liver

McArdle-RH7777 cells[32].

2.5 Discussion

To increase our understanding of the process of dietary fat absorption, we investigated a key step of this process, the storage of dietary fat in CLDs. Proteins associated with CLDs are likely involved in both the storage and trafficking of fatty acids as well as other cellular processes. Therefore, we analyzed the proteome of CLDs from enterocytes isolated from the jejunum section of the small intestine following an oil bolus using high resolution tandem mass spectroscopy. We identified 181 proteins associated with the CLD fraction (Data not shown) that are associated with a wide range of biological and molecular functions (Figure 3). Several of the identified proteins associated with lipid metabolism have been previously identified in other CLDs proteomic analyses from many cell types, while other identified proteins have not been previously reported (Table 2). Furthermore, we confirmed the localization of select proteins through immunohistochemistry and immunoelectron microscopy (Figures 5 and

6).

Of the 181 proteins identified in the CLD fraction of mouse enterocytes, thirty seven proteins are associated with known lipid related processes. Within this subgroup, twenty three proteins (62%) have been identified in CLD proteomic analyses in various cell types and fourteen proteins (38%) have not been previously reported in CLD proteomic analyses from any cell type (Table 2). Interestingly, three of the CLD proteins,

ApoA-IV, Acsl5, and microsomal triglyceride transfer protein (Mttp), have been

previously identified in a human cell model of enterocytes, Caco-2 cells, but not other

cell types [33,34]. Therefore, these proteins may be unique to the CLD of enterocytes and serve a function on CLDs in dietary fat absorption not previously recognized. We

confirmed the localization of ApoA-IV and Acsl5 on or around CLDs through

immunohistochemistry and immunoelectron microscopy analysis of mouse enterocytes

after a high fat challenge.

ApoA-IV may play important roles in dietary fat absorption due to its localization

on lipoproteins and CLDs [33,35]. Both lipoproteins and CLDs function as lipid storage

vesicles and are surrounded by a phospholipid monolayer and a diverse array of proteins

[3,36]. Many of these proteins regulate synthesis and catabolism of these similar

structures. Therefore, the identification of common proteins on lipoproteins and CLDs is

not surprising. We observed ApoA-IV forming a distinct ring like structure around CLDs

and co-localizing with Plin3, an established CLD protein, through immunofluorescence

imaging. Consistent with this result, ApoA-IV was also identified on CLDs in Caco-2

cells via immunoelectron microscopy [33]. Although the precise function of ApoA-IV

on lipoproteins is not clear, expression of ApoA-IV regulates lipoprotein synthesis and

satiety. When over-expressed, ApoA-IV increases TAG secretion from the liver and

intestine, by increasing the amount of lipid packaged on apolipoprotein B (ApoB)

containing lipoprotein particles [37-40]. Furthermore, ApoA- IV regulates feeding

behavior by trafficking to the brain and serving as a satiety signal [41]. Current

knowledge of ApoA-IV’s role in lipoprotein metabolism can be applied to the formation

of hypotheses regarding the role of ApoA-IV on CLDs. ApoA-IV may regulate CLD

synthesis and/or regulate satiety through sequestration of the protein on the CLD. These

hypotheses remain to be tested.

Although the identification of lipoprotein associated proteins in the CLD fraction

is not surprising, it is important to consider the potential for contamination of the CLD

fraction by lipoproteins. To minimize contamination of the CLD fraction by smaller, lipid

containing particles the samples were centrifuged at a relatively low speed (20,000 x g).

Smaller lipid containing particles such as lumenal lipid droplets, pre-chylomicron

transport vesicles, and chylomicrons, are isolated at much faster centrifugation speeds,

106,000 x g [42], 100,000 x g [43], and 197,000 x g [44], respectively. Despite efforts to

minimize lipoprotein contamination, we still identified several proteins which are

established for association with lipoproteins including ApoA-IV, ApoB, and Mttp. This

observation is consistent with other CLD proteomic analyses from a range of cell types

that have also identified lipoprotein associated proteins [33,45], including cell types not known to produce lipoproteins [46,47]. Furthermore, ApoA-IV and ApoB were demonstrated to localize to CLDs through immunoelectron microscopy in Caco-2 cells

[33] and hepatocytes [48] respectively. Together, these results support that a protein may be associated with both lipoproteins and CLDs.

Acsl5 may play a role in localized TAG synthesis on CLDs in enterocytes as part of the process of dietary fat absorption. It was recently demonstrated in lipid challenged

Drosophilia S2 cells that TAGs synthesis enzymes localize to CLDs including glycerol-3- phosphate acyltransferase 4, 1-acyl-sn-glycerol-3-phsophate acyltransferase gamma, and diacylglycerol acyltransferase 2 [11]. In the intestine, two TAG synthesis enzymes that catalyze the committed step in TAG biosynthesis, diacylglycerol acyltransferase 1 and

diacylglycerol acyltransferase 2 were identified as associated with CLDs by western blotting [49]. Additionally, Acsl enzymes activate fatty acids and channel them towards specific metabolic fates within cells [50]. Interestingly, other Acsl family members including Acsl1 [45,46], Acsl3 [33,34,46], and Acsl4 [33,46] have been identified on

CLDs in various cell types. In particular, Acsl3 over-expression in Cos-1 cells results in generation and expansion of CLDs in response to fatty acids [51]. Finally, Acsl5 directs exogenous fatty acids toward TAG synthesis in hepatocytes [32,52]. Therefore, one hypothesis for the function of Acsl5 on CLDs in enterocytes is the activation of fatty

acids for TAG synthesis; however, this hypothesis remains to be tested.

A limitation to a proteomic approach for identifying CLD associated proteins is

the challenge to distinguish protein contamination from other cellular organelles from

bona fide CLD proteins in the isolated CLD fraction. It is possible that proteins from

other organelles associate with the CLD fraction nonspecifically during cell lysis because

of the normal tight association of CLDs with other organelles [53]. Methods described in

the literature for isolation of CLDs vary in how cells are lysed and the type and numbers

of washes used. Currently there is not one commonly accepted method for CLD isolation

to alleviate this problem. CLD isolation protocols involving multiple washing steps

remove proteins known to associate with other organelles, as well as decrease the levels

of well-established CLD proteins [45]. In the current study, we used a single step

isolation protocol to preserve potentially loosely bound CLD proteins and reduce the risk

of missing key players in CLD metabolism. However, we only included proteins

identified in 3 of 4 biological replicates to rule out random contamination. Due to these

limitations, each protein identified needs to be validated by other methods such as

imaging to confirm the localization to CLDs in enterocytes before making firm

conclusions. Despite these limitations, the identification of Acsl5 on CLDs, ER, and

mitochondria highlights that a single protein may have multiple associations in the cell.

The protein list generated here is intended to be used as an informed, hypothesis

generating tool to prioritize future research, not as a list of established proteins on CLDs

in enterocytes.

The proteins identified in the current study build upon an existing body literature

of CLD biology, from which interesting commonalities are being observed. A key feature,

of all CLD proteomic analyses from multiple cell types, is the identification of non-lipid

related proteins in the CLD fraction. These proteins fall into a wide range of biological

and molecular functions such as amino acid metabolism, carbohydrate metabolism, and

membrane trafficking (Figure 3) [33,34,45-47,54-58]. This consistent observation has not

been fully investigated and remains an intriguing paradox of CLD biology. Several

hypotheses have been proposed to address this issue including the role of CLDs as

sequestration sites [59,60] or in the regulation of membrane trafficking [61]. The

identification of non-lipid related protein in the CLD fraction has expanded our

perception of the CLD from a simple lipid storage site to a complex and multi-functional

organelle [62].

We successfully isolated CLDs from the jejunum of mice after a dietary fat

challenge and identified the proteome of the CLDs. Similar to previously reported CLD

proteomic analyses, the proteins identified in the CLD fraction are associated with a wide

range of biological processes and molecular functions. Of interest, ApoA-IV and Acsl5

were identified in the proteomic analysis of the CLD fraction and the presence was

77 confirmed through immunofluorescence and immunoelectron microscopy. The role of many of these lipid related proteins on the CLD is currently unknown, however the proteomic analysis provides a foundation to further investigate the role of these proteins and advance the model of dietary fat absorption and lipid trafficking in the small intestine.

2.6 Acknowledgements

We would like to acknowledge the help and support of Yu-Han Hung and the

Bindley Bioscience Center, in particular Ernesto S. Nakayasu, Aaron Taylor, and members of the multi-scale imaging center for their help and expertise.

This project was supported by the Indiana Clinical and Translational Sciences

Institute, by Grant # UL1TR001108 from the National Institutes of Health, National

Center for Advancing Translational Sciences, Clinical and Translational Sciences Award

(KKB), Grant # 7-13-IN-05 by the American Diabetes Association, Innovation Award

(KKB), Grant # R01 DK098606 from the National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) (KKB, ASG), Grant # P30

DK046200 National Institutes of Health, NIDDK (ASG), Grant # U01 ES020958

National Institutes of Health, National Institute of Environmental Health Science (ASG), and the Purdue Research Foundation, Graduate Fellowship Award (KKB, TD). This material is based upon work supported by the U.S. Department of Agriculture, under agreement No. 58-1950-4-003 (ASG).

Table 2-1 GO Terms associated with protein classification term. GO terms of biological process and molecular function were grouped under broader classifications. Proteins were assigned into a broader classifications based on the associated GO terms.

Classification GO_id GO Term Number of proteins Translation GO:0006412 Translation 22 GO:0006417 regulation of translation 5 Carbohydrate GO:0044275 cellular carbohydrate catabolic process 8 GO:0016052 carbohydrate catabolic process 11 GO:0044262 cellular carbohydrate metabolic process 11 GO:0030246 carbohydrate binding 9 GO:0016051 carbohydrate biosynthetic process 6 Cytoskeletal GO:0030029 actin filament-based process 12 GO:0051015 actin filament binding 5 GO:0008092 cytoskeletal protein binding 11 Lipid GO:0006629 lipid metabolic process 21 GO:0044255 cellular lipid metabolic process 18 GO:0008289 lipid binding 17 GO:0019216 regulation of lipid metabolic process 10 GO:0016042 lipid catabolic process 7 GO:0005811 lipid particle 6 GO:0016788 hydrolase activity, acting on ester bonds 10 Protein GO:0008104 protein localization 18 localization GO:0032880 regulation of protein localization 6 and transport GO:0015031 protein transport 15 Protein GO:0019538 protein metabolic process 47 folding and GO:0071822 protein complex subunit organization 23 metabolism GO:0042803 protein homodimerization activity 19 GO:0006461 protein complex assembly 18 GO:0032403 protein complex binding 17 GO:0051246 regulation of protein metabolic process 16 GO:0022613 ribonucleoprotein complex biogenesis 12 RNA and GO:0051252 regulation of RNA metabolic process 17 transcription GO:0006403 RNA localization 4 GO:2001141 regulation of RNA biosynthetic process 15 GO:0006396 RNA processing 14 Redox GO:0055114 oxidation-reduction process 27 GO:0045454 cell redox homeostasis 4 GO:0051920 peroxiredoxin activity 3

Table 2-2 A sub group of proteins are associated with known lipid related functions. Thirty seven proteins associated with known lipid metabolism pathways were identified, of which twenty three proteins have been previously identified in other CLD proteomic analyses. Relative levels of the proteins were determined by LFQ and the average is reported (n=4 mice).

Identified in Identified in Avg Uniprot Protein name Gene name other CLD intestine LFQ IDs studies model Long-chain-fatty-acid--CoA 21.801 Acsl5 Q8JZR0 [34] ligase 5 20.926 Alcohol dehydrogenase 1 Adh1 Q3UKA4 [45,54,55,58] 20.006 Retinal dehydrogenase 1 Aldh1a1 P24549 [45] [46,47,55,58, 21.008 Annexin A2 Anxa2 P07356 [33] 63] 21.993 Annexin A4 Anxa4 Q7TMN7 [47] [33] 23.367 Apolipoprotein A-I Apoa1 Q3V2G1 [45-47] 24.529 Apolipoprotein A-IV Apoa4 Q9DBN0 [33] 18.634 Apolipoprotein B Apob E9Q414 ATP synthase subunit alpha, 19.522 Atp5a1 Q03265 [45,47] mitochondrial ATP synthase subunit beta, 22.053 Atp5b P56480 [45,47] mitochondrial Beta-1,4 N- 18.834 acetylgalactosaminyltransferase B4galnt2 Q09199

2 3(2),5-bisphosphate 19.963 Bpnt1 Q9Z0S1 nucleotidase 1 19.926 Catalase Cat Q8C6E3 [45,64] 18.289 Carboxylesterase 2a Ces2a Q8QZR3 20.492 Carboxylesterase 2c Ces2c Q91WG0 23.281 Carboxylesterase 2e Ces2e Q8BK48 16.681 Clathrin interactor 1 Clint1 Q5SUH7 22.123 Cytochrome b5 Cyb5a P56395 [45,47]

Table 2-2: Continued

22.653 NADH-cytochrome b5 reductase 3 Cyb5r3 Q9DCN2 [45] [33,34] 20.515 Cytochrome P450 2B10 Cyp2b10 Q9WUD0 21.318 Acyl-CoA-binding protein Dbi Q548W7 [47] 19.61 Bifunctional epoxide hydrolase 2 Ephx2 Q3UQ71 [45] 23.106 Fatty acid-binding protein, liver Fabp1 Q3V2F7 [45] 22.526 Fatty acid-binding protein, intestinal Fabp2 Q53YP5 18.818 Golgi membrane protein 1 Golm1 Q91XA2 Hydroxyacyl-coenzyme A 18.524 Hadh Q61425 [46] dehydrogenase 21.474 Estradiol 17-beta-dehydrogenase 11 Hsd17b11 Q9EQ06 [45,46] [33,34] [46,47,56 18.527 Hormone-sensitive lipase Lipe P54310 [33,34] ,57] Microsomal triglyceride transfer 23.765 Mttp O08601 [33,34] protein 20.812 Nucleoside diphosphate kinase Nme2 Q01768 [46,56,58 25.467 Perilipin-3 Plin3 Q9DBG5 [33,34] ,64-66] 19.272 Peroxiredoxin-6 Prdx6 Q6GT24 [45] 18.859 Retinol-binding protein 2 Rbp2 Q08652 20.936 Non-specific lipid-transfer protein Scp2 P32020 [45,46] Sulfotransferase family cytosolic 1B 19.562 Sult1b1 Q9QWG7 member 1 21.097 UDP-glucuronosyltransferase 1-7C Ugt1a7c Q6ZQM8 Transitional endoplasmic reticulum 18.177 Vcp Q8BNF8 [45-47] ATPase

2.7 References

[1] J.E. Lambert, E.J. Parks, Postprandial metabolism of meal triglyceride in humans, Biochimica et biophysica acta, 1821 (2012) 721-726. [2] S. Warnakula, J. Hsieh, K. Adeli, M.M. Hussain, P. Tso, S.D. Proctor, New insights into how the intestine can regulate lipid homeostasis and impact vascular disease: frontiers for new pharmaceutical therapies to lower cardiovascular disease risk, The Canadian journal of cardiology, 27 (2011) 183-191. [3] S. Demignot, F. Beilstein, E. Morel, Triglyceride-rich lipoproteins and cytosolic lipid droplets in enterocytes: key players in intestinal physiology and metabolic disorders, Biochimie, 96 (2014) 48-55. [4] H. Duez, M. Pavlic, G.F. Lewis, Mechanism of intestinal lipoprotein overproduction in insulin resistant humans, Atherosclerosis. Supplements, 9 (2008) 33-38. [5] N.A. Abumrad, N.O. Davidson, Role of the gut in lipid homeostasis, Physiological reviews, 92 (2012) 1061-1085. [6] M.M. Hussain, Intestinal lipid absorption and lipoprotein formation, Current opinion in lipidology, 25 (2014) 200-206. [7] J. Zhu, B. Lee, K.K. Buhman, J.X. Cheng, A dynamic, cytoplasmic triacylglycerol pool in enterocytes revealed by ex vivo and in vivo coherent anti-Stokes Raman scattering imaging, Journal of lipid research, 50 (2009) 1080-1089. [8] A. Lass, R. Zimmermann, M. Oberer, R. Zechner, Lipolysis - a highly regulated multi-enzyme complex mediates the catabolism of cellular fat stores, Progress in lipid research, 50 (2011) 14-27. [9] P.J. McFie, Y. Jin, S.L. Banman, E. Beauchamp, L.G. Berthiaume, S.J. Stone, Characterization of the interaction of diacylglycerol acyltransferase-2 with the endoplasmic reticulum and lipid droplets, Biochimica et biophysica acta, 1841 (2014) 1318-1328. [10] P.J. McFie, S.L. Banman, S. Kary, S.J. Stone, Murine diacylglycerol acyltransferase- 2 (DGAT2) can catalyze triacylglycerol synthesis and promote lipid droplet formation independent of its localization to the endoplasmic reticulum, The Journal of biological chemistry, 286 (2011) 28235-28246[11] F. Wilfling, H. Wang, J.T. Haas, N. Krahmer, T.J. Gould, A. Uchida, J.X. Cheng, M. Graham, R. Christiano, F. Frohlich, X. Liu, K.K. Buhman, R.A. Coleman, J. Bewersdorf, R.V. Farese, Jr., T.C. Walther, Triacylglycerol synthesis enzymes mediate lipid droplet growth by relocalizing from the ER to lipid droplets, Developmental cell, 24 (2013) 384-399. [12] B. Lee, J. Zhu, N.E. Wolins, J.X. Cheng, K.K. Buhman, Differential association of adipophilin and TIP47 proteins with cytoplasmic lipid droplets in mouse enterocytes during dietary fat absorption, Biochimica et biophysica acta, 1791 (2009) 1173-1180.

[13] S. Obrowsky, P.G. Chandak, J.V. Patankar, S. Povoden, S. Schlager, E.E. Kershaw, J.G. Bogner-Strauss, G. Hoefler, S. Levak-Frank, D. Kratky, Adipose triglyceride lipase is a TG hydrolase of the small intestine and regulates intestinal PPARalpha signaling, Journal of lipid research, 54 (2013) 425-435. [14] P. Xie, F. Guo, Y. Ma, H. Zhu, F. Wang, B. Xue, H. Shi, J. Yang, L. Yu, Intestinal Cgi-58 deficiency reduces postprandial lipid absorption, PloS one, 9 (2014) e91652. [15] Y. Xie, F. Nassir, J. Luo, K. Buhman, N.O. Davidson, Intestinal lipoprotein assembly in apobec-1-/- mice reveals subtle alterations in triglyceride secretion coupled with a shift to larger lipoproteins, American journal of physiology. Gastrointestinal and liver physiology, 285 (2003) G735-746. [16] D.L. Brasaemle, N.E. Wolins, Isolation of lipid droplets from cells by density gradient centrifugation, Current protocols in cell biology / editorial board, Juan S. Bonifacino ... [et al.], Chapter 3 (2006) Unit 3 15. [17] J. Cox, M. Mann, MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification, Nature biotechnology, 26 (2008) 1367-1372. [18] C. UniProt, Activities at the Universal Protein Resource (UniProt), Nucleic acids research, 42 (2014) D191-198. [19] J. Cox, N. Neuhauser, A. Michalski, R.A. Scheltema, J.V. Olsen, M. Mann, Andromeda: a peptide search engine integrated into the MaxQuant environment, Journal of proteome research, 10 (2011) 1794-1805. [20] W. Huang da, B.T. Sherman, R.A. Lempicki, Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources, Nature protocols, 4 (2009) 44-57. [21] W. Huang da, B.T. Sherman, R.A. Lempicki, Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists, Nucleic acids research, 37 (2009) 1-13. [22] L.J. Jensen, M. Kuhn, M. Stark, S. Chaffron, C. Creevey, J. Muller, T. Doerks, P. Julien, A. Roth, M. Simonovic, P. Bork, C. von Mering, STRING 8--a global view on proteins and their functional interactions in 630 organisms, Nucleic acids research, 37 (2009) D412-416. [23] N.E. Wolins, B.K. Quaynor, J.R. Skinner, M.J. Schoenfish, A. Tzekov, P.E. Bickel, S3-12, Adipophilin, and TIP47 package lipid in adipocytes, The Journal of biological chemistry, 280 (2005) 19146-19155. [24] T. Vowinkel, M. Mori, C.F. Krieglstein, J. Russell, F. Saijo, S. Bharwani, R.H. Turnage, W.S. Davidson, P. Tso, D.N. Granger, T.J. Kalogeris, Apolipoprotein A-IV inhibits experimental colitis, The Journal of clinical investigation, 114 (2004) 260-269. [25] N.E. Wolins, D.L. Brasaemle, P.E. Bickel, A proposed model of fat packaging by exchangeable lipid droplet proteins, FEBS letters, 580 (2006) 5484-5491. [26] J. Cox, M.Y. Hein, C.A. Luber, I. Paron, N. Nagaraj, M. Mann, Accurate proteome- wide label-free quantification by delayed normalization and maximal peptide ratio extraction, termed MaxLFQ, Molecular & cellular proteomics : MCP, 13 (2014) 2513- 2526.

[27] A.B. Kohan, F. Wang, X. Li, S. Bradshaw, Q. Yang, J.L. Caldwell, T.M. Bullock, P. Tso, Apolipoprotein A-IV regulates chylomicron metabolism-mechanism and function, American journal of physiology. Gastrointestinal and liver physiology, 302 (2012) G628- 636. [28] T. Simon, V.R. Cook, A. Rao, R.B. Weinberg, Impact of murine intestinal apolipoprotein A-IV expression on regional lipid absorption, gene expression, and growth, Journal of lipid research, 52 (2011) 1984-1994. [29] E. Soupene, F.A. Kuypers, Mammalian long-chain acyl-CoA synthetases, Experimental biology and medicine (Maywood, N.J.), 233 (2008) 507-521. [30] C. Klaus, U. Schneider, C. Hedberg, A.K. Schutz, J. Bernhagen, H. Waldmann, N. Gassler, E. Kaemmerer, Modulating effects of acyl-CoA synthetase 5-derived mitochondrial Wnt2B palmitoylation on intestinal Wnt activity, World journal of gastroenterology : WJG, 20 (2014) 14855-14864. [31] E. Kaemmerer, A. Peuscher, A. Reinartz, C. Liedtke, R. Weiskirchen, J. Kopitz, N. Gassler, Human intestinal acyl-CoA synthetase 5 is sensitive to the inhibitor triacsin C, World journal of gastroenterology : WJG, 17 (2011) 4883-4889. [32] D.G. Mashek, M.A. McKenzie, C.G. Van Horn, R.A. Coleman, Rat long chain acyl- CoA synthetase 5 increases fatty acid uptake and partitioning to cellular triacylglycerol in McArdle-RH7777 cells, The Journal of biological chemistry, 281 (2006) 945-950. [33] J. Bouchoux, F. Beilstein, T. Pauquai, I.C. Guerrera, D. Chateau, N. Ly, M. Alqub, C. Klein, J. Chambaz, M. Rousset, J.M. Lacorte, E. Morel, S. Demignot, The proteome of cytosolic lipid droplets isolated from differentiated Caco-2/TC7 enterocytes reveals cell- specific characteristics, Biology of the cell / under the auspices of the European Cell Biology Organization, 103 (2011) 499-517. [34] F. Beilstein, J. Bouchoux, M. Rousset, S. Demignot, Proteomic analysis of lipid droplets from Caco-2/TC7 enterocytes identifies novel modulators of lipid secretion, PloS one, 8 (2013) e53017. [35] F. Wang, A.B. Kohan, C.M. Lo, M. Liu, P. Howles, P. Tso, Apolipoprotein A-IV: a protein intimately involved in metabolism, Journal of lipid research, (2015). [36] T.C. Walther, R.V. Farese, Jr., Lipid droplets and cellular lipid metabolism, Annual review of biochemistry, 81 (2012) 687-714. [37] R.B. Weinberg, J.W. Gallagher, M.A. Fabritius, G.S. Shelness, ApoA-IV modulates the secretory trafficking of apoB and the size of triglyceride-rich lipoproteins, Journal of lipid research, 53 (2012) 736-743. [38] M.A. VerHague, D. Cheng, R.B. Weinberg, G.S. Shelness, Apolipoprotein A-IV expression in mouse liver enhances triglyceride secretion and reduces hepatic lipid content by promoting very low density lipoprotein particle expansion, Arteriosclerosis, thrombosis, and vascular biology, 33 (2013) 2501-2508. [39] S. Lu, Y. Yao, X. Cheng, S. Mitchell, S. Leng, S. Meng, J.W. Gallagher, G.S. Shelness, G.S. Morris, J. Mahan, S. Frase, C.M. Mansbach, R.B. Weinberg, D.D. Black, Overexpression of apolipoprotein A-IV enhances lipid secretion in IPEC-1 cells by increasing chylomicron size, The Journal of biological chemistry, 281 (2006) 3473-3483. [40] S. Lu, Y. Yao, S. Meng, X. Cheng, D.D. Black, Overexpression of apolipoprotein A- IV enhances lipid transport in newborn swine intestinal epithelial cells, The Journal of biological chemistry, 277 (2002) 31929-31937.

[41] F. Wang, K.J. Pearson, W.S. Davidson, P. Tso, Specific sequences in N termini of apolipoprotein A-IV modulate its anorectic effect, Physiology & behavior, 120 (2013) 136-142. [42] R. Tan, W. Wang, S. Wang, Z. Wang, L. Sun, W. He, R. Fan, Y. Zhou, X. Xu, W. Hong, T. Wang, Small GTPase Rab40c associates with lipid droplets and modulates the biogenesis of lipid droplets, PloS one, 8 (2013). [43] D.M. Wong, J.P. Webb, P.M. Malinowski, J. Macri, K. Adeli, Proteomic profiling of the prechylomicron transport vesicle involved in the assembly and secretion of apoB-48- containing chylomicrons in the intestinal enterocytes, Proteomics, 9 (2009) 3698-3711. [44] J. Luchoomun, M.M. Hussain, Assembly and secretion of chylomicrons by differentiated Caco-2 cells. Nascent triglycerides and preformed phospholipids are preferentially used for lipoprotein assembly, The Journal of biological chemistry, 274 (1999) 19565-19572. [45] A.E. Crunk, J. Monks, A. Murakami, M. Jackman, P.S. Maclean, M. Ladinsky, E.S. Bales, S. Cain, D.J. Orlicky, J.L. McManaman, Dynamic regulation of hepatic lipid droplet properties by diet, PloS one, 8 (2013) e67631. [46] H. Zhang, Y. Wang, J. Li, J. Yu, J. Pu, L. Li, S. Zhang, G. Peng, F. Yang, P. Liu, Proteome of skeletal muscle lipid droplet reveals association with mitochondria and apolipoprotein a-I, Journal of proteome research, 10 (2011) 4757-4768. [47] Y. Ding, Y. Wu, R. Zeng, K. Liao, Proteomic profiling of lipid droplet-associated proteins in primary adipocytes of normal and obese mouse, Acta biochimica et biophysica Sinica, 44 (2012) 394-406. [48] Y. Ohsaki, J. Cheng, A. Fujita, T. Tokumoto, T. Fujimoto, Cytoplasmic lipid droplets are sites of convergence of proteasomal and autophagic degradation of apolipoprotein B, Molecular biology of the cell, 17 (2006) 2674-2683. [49] A. Seyer, M. Cantiello, J. Bertrand-Michel, V. Roques, M. Nauze, V. Bezirard, X. Collet, D. Touboul, A. Brunelle, C. Comera, Lipidomic and spatio-temporal imaging of fat by mass spectrometry in mice duodenum during lipid digestion, PloS one, 8 (2013) e58224. [50] D.G. Mashek, L.O. Li, R.A. Coleman, Long-chain acyl-CoA synthetases and fatty acid channeling, Future lipidology, 2 (2007) 465-476. [51] A. Kassan, A. Herms, A. Fernandez-Vidal, M. Bosch, N.L. Schieber, B.J. Reddy, A. Fajardo, M. Gelabert-Baldrich, F. Tebar, C. Enrich, S.P. Gross, R.G. Parton, A. Pol, Acyl-CoA synthetase 3 promotes lipid droplet biogenesis in ER microdomains, The Journal of cell biology, 203 (2013) 985-1001. [52] S.Y. Bu, D.G. Mashek, Hepatic long-chain acyl-CoA synthetase 5 mediates fatty acid channeling between anabolic and catabolic pathways, Journal of lipid research, 51 (2010) 3270-3280. [53] J.M. Goodman, Demonstrated and inferred metabolism associated with cytosolic lipid droplets, Journal of lipid research, 50 (2009) 2148-2156. [54] P. Liu, Y. Ying, Y. Zhao, D.I. Mundy, M. Zhu, R.G. Anderson, Chinese hamster ovary K2 cell lipid droplets appear to be metabolic organelles involved in membrane traffic, The Journal of biological chemistry, 279 (2004) 3787-3792.

[55] R. Bartz, J.K. Zehmer, M. Zhu, Y. Chen, G. Serrero, Y. Zhao, P. Liu, Dynamic activity of lipid droplets: protein phosphorylation and GTP-mediated protein translocation, Journal of proteome research, 6 (2007) 3256-3265. [56] D.L. Brasaemle, G. Dolios, L. Shapiro, R. Wang, Proteomic analysis of proteins associated with lipid droplets of basal and lipolytically stimulated 3T3-L1 adipocytes, The Journal of biological chemistry, 279 (2004) 46835-46842. [57] S. Larsson, S. Resjo, M.F. Gomez, P. James, C. Holm, Characterization of the lipid droplet proteome of a clonal insulin-producing beta-cell line (INS-1 832/13), Journal of proteome research, 11 (2012) 1264-1273. [58] S. Turro, M. Ingelmo-Torres, J.M. Estanyol, F. Tebar, M.A. Fernandez, C.V. Albor, K. Gaus, T. Grewal, C. Enrich, A. Pol, Identification and characterization of associated with lipid droplet protein 1: A novel membrane-associated protein that resides on hepatic lipid droplets, Traffic (Copenhagen, Denmark), 7 (2006) 1254-1269. [59] Z. Li, K. Thiel, P.J. Thul, M. Beller, R.P. Kuhnlein, M.A. Welte, Lipid droplets control the maternal histone supply of Drosophila embryos, Current biology : CB, 22 (2012) 2104-2113. [60] Z. Li, M.R. Johnson, Z. Ke, L. Chen, M.A. Welte, Drosophila lipid droplets buffer the H2Av supply to protect early embryonic development, Current biology : CB, 24 (2014) 1485-1491. [61] R.S. Kiss, T. Nilsson, Rab proteins implicated in lipid storage and mobilization, Journal of biomedical research, 28 (2014) 169-177. [62] E.L. Arrese, F.Z. Saudale, J.L. Soulages, Lipid Droplets as Signaling Platforms Linking Metabolic and Cellular Functions, Lipid insights, 7 (2014) 7-16. [63] Y. Fujimoto, H. Itabe, J. Sakai, M. Makita, J. Noda, M. Mori, Y. Higashi, S. Kojima, T. Takano, Identification of major proteins in the lipid droplet-enriched fraction isolated from the human hepatocyte cell line HuH7, Biochimica et biophysica acta, 1644 (2004) 47-59. [64] T. Orban, G. Palczewska, K. Palczewski, Retinyl ester storage particles (retinosomes) from the retinal pigmented epithelium resemble lipid droplets in other tissues, The Journal of biological chemistry, 286 (2011) 17248-17258. [65] E. Umlauf, E. Csaszar, M. Moertelmaier, G.J. Schuetz, R.G. Parton, R. Prohaska, Association of stomatin with lipid bodies, The Journal of biological chemistry, 279 (2004) 23699-23709. [66] S. Sato, M. Fukasawa, Y. Yamakawa, T. Natsume, T. Suzuki, I. Shoji, H. Aizaki, T. Miyamura, M. Nishijima, Proteomic profiling of lipid droplet proteins in hepatoma cell lines expressing hepatitis C virus core protein, Journal of biochemistry, 139 (2006) 921- 930.

CHAPTER 3. DIET INDUCED OBESITY ALTERS CYTOPLASMIC LIPID DROPLET MORPHOLOGY AND ASSOCIATED PROTEINS IN RESPONSE TO AN ACUTE DIETARY FAT CHALLENGE

3.1 Abstract

Dietary fat absorption by the small intestine is an efficient, multistep process that

regulates the uptake and delivery of essential nutrients and energy. Fatty acids taken up

by enterocytes, the absorptive cells of the small intestine, are resynthesized into

triacylglycerol (TAG) and either secreted in chylomicrons or temporarily stored in

cytoplasmic lipid droplets (CLDs). Proteins that associate with CLDs are thought to

regulate the dynamics of TAG storage and mobilization. It is currently unclear what

affect diet induced obesity (DIO) has on the balance between dietary fat storage and

secretion. Specifically, there is limited knowledge of how DIO affects the level and diversity of proteins that associate with CLDs and regulate CLD dynamics. In the current study, we characterize CLDs from lean and DIO mice through histological and proteomic analyses. We demonstrate that DIO mice have larger intestinal CLDs compared to lean mice in response to dietary fat. Additionally we identified 375 proteins associated with

CLDs in lean and DIO mice. We identify a subgroup of lipid related proteins which are either increased or unique to the DIO CLD proteome. These proteins are involved in

93 steroid synthesis pathways, TAG synthesis, and lipolysis. This analysis expands upon our knowledge of the effect of DIO obesity on the process of dietary fat absorption in the small intestine.

3.2 Introduction

Enterocytes, the absorptive cells of the small intestine, are responsible for the uptake, repackaging, and secretion of dietary fat. In addition, enterocytes are capable of temporarily storing dietary fat in CLDs in response to a dietary fat challenge (Reviewed in [1, 2]) Dietary fat in the form of TAG is digested in the lumen of the small intestine producing FFA and MAG which are incorporated into mixed micelles. Mixed micelles interact with the brush border membrane of enterocytes and FFA and MAG are taken up by enterocytes. In the enterocyte FFA and MAG are resynthesized into TAG by MGAT and DGAT at the ER. The resulting TAG can be packaged onto a chylomicron particle for secretion or temporarily stored in CLDs.

CLDs are composed of a neutral lipid core, a phospholipid monolayer, and associated proteins [3]. Proteins which associate with CLDs are thought to regulate the storage and mobilization of TAG. Proteins associated with CLDs isolated from enterocytes after a lipid challenge have been identified by targeted [4] and global proteomic [5-7] approaches. These studies have identified and confirmed members of the perilipin family (Plin2 and Plin3) associated with CLDs. Interestingly, these studies have also identified proteins associated with a wide variety of biological pathways including lipid catabolic and anabolic pathways.

Obesity alters the process of dietary fat absorption. The effect of obesity on

intestinal lipid metabolism is currently unclear, but may be dysregulated by obesity and

associated diseases (recently reviewed [8]). Recent studies in high fat diet fed (non obese),

ob/ob mice, and DIO mice have demonstrated decreased TAG secretion rates in response

to an acute dietary fat challenge, when lipoprotein lipase has been inhibited with

tyloxapol [9, 10]. Inhibition of lipoprotein lipase blocks the clearance of lipoproteins

resulting in a dramatic increase in blood TAG levels. DIO alters intestinal mRNA levels

of proteins involved in chylomicron assembly, TAG synthesis, and fatty acid trafficking

[10].

Little is known on the effect of obesity on CLD morphology and associated

proteins in enterocytes. DIO alters CLD associated proteins in adipocytes and

hepatocytes [11, 12].Little is known on the effect of obesity on the CLD proteome in

enterocytes. Previous studies have shown alterations in mRNA levels of proteins known

to associate with CLDs including lipases [10] and members of the perilipin family[13].

Plin2 and 3 localize to CLDs in response to a dietary fat challenge, however under

different physiological conditions. Plin3 localizes to CLDs in response to an acute dietary

¡¢£¤¡ ¥ ¥¦§¨¦©¦ §¦ © ¡ ¤ ¡ ¢¦ ¡ © § ¢ ¡¢ § ¡ ¥ ¥ ¦ ¥ ¡ ¥ ¨¡¡¨¦

Plin2 localizes to CLDs in response to chronic high fat feeding in a DIO mouse model

[13]. A comprehensive analysis of proteins associated with CLDs from lean and DIO

mice in response to an acute dietary fat challenge has not been conducted

` To determine the effects of DIO on enterocyte CLD morphology and associated proteins in response to consumption of an acute dietary fat challenge, we investigated

CLDs within enterocytes of lean and DIO mice given an oil bolus orally by transmission

electron microscopy, immunofluorescence microscopy, and proteomic analysis. Based on

previous observations [9, 10, 13] of altered TAG secretion in DIO mice in response to a

dietary fat challenge, we hypothesized that DIO mice have altered CLD morphology and

CLD associated proteins.

3.3 Materials and methods

3.3.1 Mice

C57BL/6 male mice from an in-house breeding colony were used for this study.

The mice were maintained on a chow diet (PicoLab 5053, Lab Diets, Richmond, IN,

USA) that consisted of 62.1% of calories from carbohydrate (starch), 24.7% from protein,

and 13.2% from fat from weaning to 5 weeks of age. The mice were housed in a

temperature and humidity controlled facility with a 12 hour light/dark cycle (6AM/6PM)

with ad libitum access to food and water.

3.3.2 Generation of DIO mouse model

C57BL/6 male mice were placed on a either high fat diet or low fat diet. The high fat diet (D12492, Research Diet, New Brunswick, NJ, USA) consists of 20% of calories from carbohydrates, 20% from protein, and 60% from fat. The low fat, D12492 matched diet (D12450J, Research Diet, New Brunswick, NJ, USA) consists of 70% of calories from carbohydrates, 20% from protein, and 10% from fat. The mice were maintained on the diet for 12 weeks and body weights were monitored weekly.

3.3.3 Mouse procedure for lipid droplet isolation and proteomics analysis

Five mice from the low fat diet group and 5 mice from the high fat diet group

¡¢£¤¥ ¦§ ¨ ©¦ § ¢ ¡ £ £©¤ ¤ ¦ £©¤ ©£ ¤ ¦§¡ ¡¡ ¤ ¦ ¦ ¤ were oil was administered and two hours after the oil bolus, the mice were euthanized via CO2 asphyxiation. The small intestine was excised and divided into three equal length segments in relation to the stomach. The middle segment, representing the jejunum, was used for the analysis.

3.3.4 Enterocyte isolation

Enterocytes were isolated from the middle section of the small intestine as previously described [13, 14]. Briefly, the intestinal sections were washed in tissue buffer

(Hank's Balanced Salt Solution with 25 mM HEPES and 1% fetal calf serum) and then placed in isolation buffer (Calcium and magnesium free Hank's Balanced Salt Solution with 1.5mM EDTA). The intestine segments were incubated for 15 minutes at 37°C with rotation. The sample was mixed briefly (vortexed) and the supernatant containing enterocytes was removed and saved. The process was repeated and the supernatants containing isolated enterocytes were combined.

3.3.5 CLD isolation

CLDs were isolated from enterocytes using a previously established sucrose gradient ultracentrifugation protocol [7]. Enterocytes were lysed in ice cold sucrose lysis

buffer (175mM sucrose, 10 mM HEPES and 1 mM EDTA pH 7.4). Cells were disrupted

by passing through a 27 gauge, 1 inch needle, eight times. The resulting two mLs of cell lysate were carefully layered with six mLs of sucrose-free lysis buffer and centrifuged at

20,000 x g at 4°C for two hours. After centrifugation, the sample was frozen at -80°C.

The frozen sample was sliced into seven sequential fractions which were approximately 1

cm in length.

3.3.6 In solution digestion and LC MS/MS

In preparation for proteomic analysis, the isolated CLD fractions were delipidated using 2:1 chloroform methanol and proteins precipitated using ice cold acetone. The protein pellet was denatured using 8M urea and 10 mM DTT for 1.5 hours at 37°C. The

sample was digested for 12 hours using trypsin (Sigma-Aldrich, St. Louis MO, USA)

¡ ¢ £¤ ¥ ¦ ¥§¢ ¨ ¨ © ¤ § ¦ ¡ ¢ £ § ¨ ¤ ¦¨§ ¢£ ¦¥§¢¨ £ ¥¡ £ ¡¢£¤

trifluoroacetic acid. Tryptic peptides were separated on a nanoLC system (1100 Series

LC, Agilent Technologies, Santa Clara, CA). The peptides were loaded on the Agilent

300SB-C18 enrichment column for concentration and the enrichment column was

switched into the nano-flow path after five minutes. Peptides were separated with a C18

reversed phase ZORBAX 300SB-C18 column. The column was connected to the

emission tip and coupled to the nano-electrospray ionization source of the high resolution hybrid ion trap mass spectrometer LTQ-Orbitrap LX (Thermo Fisher Scientific, Waltham

MA, USA). The LTQ-orbitrap mass spectrometer was operated in the data-dependent

positive acquisition mode in which each full MS scan (30.000 resolving power) was

followed by six MS/MS scans where the six most abundant molecular ions were selected

and fragmented by collision induced dissociation (CID) using a normalized collision

energy of 35%..

3.3.7 Protein identification

The peak list files containing MS and MS/MS data were analyzed using

MaxQuant version 1.4.08 [15, 16]. For protein identification, the MS/MS data was

searched against the Uniprot protein database[17]. The database was searched using the

MASCOT search engine utilizing Andromeda as the peptide search algorithm that is

incorporated in the MaxQuant platform [18]. The search was conducted using the following settings: trypsin cleavage with a maximum of two missed cleavages, fixed modification of iodoethanol addition to cysteine, variable modification of oxidation of methionine and acetylation of the N-terminal. The MS mass tolerance was set at 4.5 ppm with a maximum number of five modifications. The false discovery rate was set at 0.01

for proteins and peptides and was run against a decoy revert database. Peptides required a

minimum length of seven amino acids. The MS/MS tolerance was set at 0.1 Da for

protein identification. The minimum score for modified and unmodified peptides was set

at forty. At least two peptides were required for protein identification. The bioinformatics

and statistical package Perseus 1.4.1.3 was used to analyze the MaxQuant output.

Contaminants identified by Perseus, such as keratin, were removed from the analysis. For

label free quantitation (LFQ), the intensity for each protein was transformed log2(x). A

protein was considered identified in a treatment group if it was identified in at least 3 out

of the 5 biological replicates.

3.3.8 Data analysis and bioinformatics

The Gene Ontology (GO) Terms for biological process associated with identified proteins were determined using Gene Ontology Enrichment Analysis and Visualization

Tool (GOrilla) [19, 20]. Visualization of enriched GO terms was accomplished using

GOrilla comparing the isolated CLD proteomic profile to the entire mouse proteome

obtained from the Uniprot protein database. The p value threshold was set at p< .01.

Lipid related GO terms were identified and proteins associated with the lipid related GO terms were compiled and are here on referred to as “Lipid related proteins”. Visualization of protein interactions was accomplished using STRING version 10.0 [21]. We used the confidence view with a score of 0.4, indicating a medium confidence level.

3.3.9 Immunofluorescence imaging

Four mice from low fat diet treatment group and the high fat treatment group were

¡¢£ ¤¥ ¦§¨©¦ §¢¡££©¤ ¤ ¦ £©¤ ©£ ¤ ¡¥£©¤¡¥ ¢£¤§¤¥ ¡ ¦ ¤

oil bolus. A small (5mm) section of the jejunum was harvested from the mice two hours

after the dietary fat challenge and was frozen in optimal cutting temperature embedding

media in 2-methyl butane cooled with dry ice. The tissue was stored at -80°C until

§¦ ¤¢¢ ¤¥ ¦§ ¦ ¦§¤¢ ¤¤ §¦¢ ¦ ©¤ £ ¢¢ ¤ ¡¢¢ ¤¥ £ ¦ £ ¢¢ ¤

100

sections, fixed in 2% paraformaldehyde, permeabilized with 0.1% saponin, and blocked

with 3% bovine serum albumin in PBS. The tissue was probed with previously validated

antibodies for Plin3 [22] (a gift from Dr. Perry Bickel at the University of Texas

Southwestern, Dallas TX, USA) and Plin2 [22]. The sections were also stained for neutral

¡¢¡£¤ ¥¤ ¡ ¦§ ¨ © § -difluoro-1,3,5,7,8-pentamethyl-4-bora-3a,4a-diaza-s-indacene

(BODIPY) (Life technologies, Grand Island NY, USA), a secondary AlexaFluor antibody

(Life technologies), and for nuclei using 300 nM DAPI (Life technologies). The sections

were imaged using a Nikon A1R confocal microscope (Nikon Instruments Inc., Melville

NY, USA). Image processing was conducted using NIS- Elements C acquisition and

analysis software.

3.3.10 Cardiac perfusion, fixation, and transmission electron microscopy

Three mice from low fat diet treatment group and high fat treatment group, were

¤ £ ¥ ¤ §¡¦¦¡¦§ ¡§ ¦ § § © ¡ ¡

was administered and two hours after the oil bolus, the mice were anesthetized using

inhaled isoflurane and perfused with 1.5% glutaraldehyde in 0.1M sodium cacodylate via

cardiac infusion. After fixation, a sample of the jejunum section of the small intestine was

isolated, stained with osmium tetroxide, dehydrated, and embedded in resin. Ultrathin

sections were stained with lead citrate and uranyl acetate and examined using a Tecnai

T20 Transmission Electron Microscope (FEI, Hillsboro OR, USA).

101

3.3.11 Cytoplasmic lipid droplet size analysis

Thirty cells were examined from (n=3 mice) from each treatment group. The cells

analyzed were located in the middle section of the villi and at least 3 villi were measured

per mouse. CLD diameter was measured using ImageJ. Image stitching was performed

using Autostitch.. Data is expressed as mean ± SEM. Statistical significance was

determined using a Type 3 mixed model for fixed effects using SAS v 9.4; P<0.05 was

considered significant. Lipid droplet size distribution was analyzed by two-sample

Kolmogorov–Smirnov test; P<0.05 was considered significant.

3.4 Results

3.4.1 Lipid accumulates in CLDs in enterocytes in response to an acute dietary fat

¡¢££¤¥¦¤ § ¢ ¨ ©¤ ¥¤© ¡ ¨ ¢ ¤¢¥ ¢£¢© ¥ ¨¢ ¥ ££ ¤ £ § ¥

lean and DIO mice

Neutral lipids accumulate in CLDs in enterocytes of the jejunum section of the small intestine in lean and DIO mice in response to an acute dietary fat challenge (Fig 1).

The representative micrographs highlight the size disparity of CLDs that accumulate in enterocytes of lean (Figure 1A) compared to DIO mice (Figure 1B).

102

3.4.2 CLDs in enterocyte of DIO mice are larger and have a different size distribution

compared to lean mice in response to an acute dietary fat challenge

The average CLD diameter was measured from 30 cells per mouse (n=3 mice) from each treatment group. No difference was observed in the number of CLD per cell in lean and obese mice in response to an acute dietary fat challenge (Figure 2a). DIO mice

(black bars) have significantly greater CLD diameter compared to lean mice in response to an acute dietary fat challenge (Figure 2b). CLDs in enterocytes from DIO mice have a statistically significant size distribution compared to lean mice in response to an acute dietary fat challenge as determined by two-sample Kolmogorov–Smirnov test (Figure 2c).

3.4.3 Proteins were identified associated with isolated CLDs from lean and DIO mice in

response to an acute dietary fat challenge

To identify proteins which may contribute to the size distribution disparity between the lean and DIO mice in response to an acute dietary fat challenge, we identified proteins associated with isolated CLDs by LC MS/MS. A protein was considered identified in a treatment group if it was identified in at least 3 out of 5 biological replicates in at least one of the treatment groups. We identified a total of 375 proteins (Data not shown). Of the 375 identified proteins, a portion was found unique to a treatment group, some found at similar levels and others were found at differentiated levels (Figure 3).

103

3.4.4 Proteins identified associated with CLDs from lean and DIO mice are associated

with a variety of biological processes

The gene ontologies associated with the 375 identified proteins were compared to the complete mouse proteome and visualized using GOrilla (Data not shown). GOrilla highlights biological processes which are enriched within the CLD associated protein dataset. Biological processes highlighted in red indicate the most significant enrichment.

The proteins identified in the current analysis are associated with a wide variety of biological processes (Data not shown). Biological process pathways that have the greatest enrichment include oxidation-reduction process (GO:0055114), cellular lipid catabolic process (GO:0044242), and organonitrogen compound metabolic process (GO:1901564).

To identify proteins which may contribute to CLD size and lipid accumulation, we limited our analysis to the 81 (out of 375) proteins that are associated with lipid related processes as determined using GOrilla (Data not shown).

3.4.5 Proteins associated with lipid related processes are found at differentially present

levels and have been previously characterized associated with CLD in enterocytes

proteomic or targeted protein analysis

Of the 81 lipid related proteins identified in the analysis, 25 are found either unique or at differentiated present levels in lean and DIO samples (Table 1)Targeted

STRING analysis of lipid related proteins identify interacting protein clusters associated with various lipid pathways.

104

To identify lipid functional pathways enriched in the CLD proteome, we

conducted a STRING analysis of the 81 lipid related proteins (Figure 4). The STRING

analysis identified protein clusters related to lipoprotein metabolism, TAG storage and

mobilization, and fatty acid catabolism.The 25 proteins which are found unique or at

higher levels in either lean or DIO samples are indicated by green and red boxes

respectively and their relative levels are found in Table 1. Proteins found at higher levels

in DIO samples are associated mainly with fatty acid catabolism, lipid synthesis, and

lipid catabolism.

3.4.6 Plin3 and Plin2 localize to CLDs from lean and DIO mice in response to an acute

dietary fat challenge.

To confirm the presence of CLD associated proteins identified by CLD proteomic analysis, we used immunofluorescence microscopy of 2 proteins which were identified,

Plin2 and Plin3 (Figure 5). Plin3 (A-B and E-F) is found at relatively equal abundance associated with CLDs from lean (A-B) and DIO (E-F) mice. Plin2 (C-D and G-H) is found at higher levels and forming a more distinct ring like structure around CLDs from

DIO mice (G-H) compared to lean mice (C-D). These results validate that Plin2 and 3 associate with CLDs in lean and DIO mice and Plin2 is present at differentiated levels.

An interesting observation was made regarding to the localization of Plin2 and Plin3, particularly in DIO samples.

105

3.4.7 Plin2 and Plin3 localize to different CLDs in different regions of the cell in DIO

mice.

Plin3 (A, C, and E) localizes to CLDs primarily in the apical region (above the

nucleus) of the cell. Plin2 (B, D, and F) localizes to CLDs both in the apical and

basolateral side of the cell.

3.5 Discussion

We investigated the effects of DIO on enterocyte CLD morphology and

associated proteins in response to an acute dietary fat challenge. DIO mice have larger

CLDs with a greater maximal size compared to lean (Figures 1 and 2). 375 proteins were identified on CLDs from lean and DIO mice in response to an acute dietary fat challenge

(Data not shown). A portion of the proteins were found unique to one of the treatment

groups, a portion found at similar levels between the treatment groups, and some found at

differentially expressed levels (Figure 3). The identified proteins represent a diverse

range of proteins functions (Data not shown). Eighty one of the identified proteins are

associated with lipid related pathways (Figure 4). Of the 81 lipid related proteins, 25

proteins are either unique to CLDs from DIO mice or found at differentially present

levels in lean and DIO mice in response to an acute dietary fat challenge based on label free quantitation levels (Table 1). Plin3 is found associated with CLDs from lean and

DIO mice at similar levels while Plin2 is found at increased levels on CLDs from DIO

106

mice compared to lean in response to an acute dietary fat challenge (Figure 5). Plin2 and

Plin3 are found on different pools of CLDs in DIO mice in response to an acute dietary

fat challenge (Figure 6).

DIO alters CLD size which may contribute to the observed decreased TAG

secretion rates in DIO compared to lean mice in response to an acute dietary fat challenge.

High fat diet fed (non obese), ob/ob mice, and DIO mice have decreased TAG secretion

rates after an olive oil bolus, when lipoprotein lipase has been inhibited with tyloxapol [9,

10]. Interestingly, while there is a decrease in TAG secretion, analysis of intestinal

mucosa by either biochemical analysis or Coherent Anti-Scatter Ramen Spectroscopy[10]

demonstrated no difference in TAG content between lean and DIO in response to an

acute dietary fat challenge. The current study demonstrates an increased CLD size in DIO

and lean mice in response to an acute dietary fat challenge by transmission electron

microscopy. The difference between the current study and previous analysis of intestinal

mucosa is the use of transmission electron microscopy to identify the subcellular localization of TAG in CLDs as compared to the analysis of the total enterocyte TAG content. Biochemical and CARS analysis cannot distinguish between subcellular

localization of TAG in secretory or storage pathways in enterocytes. We hypothesize that

the decreased TAG secretion rate and increased TAG storage in CLDs in DIO mice compared to lean in response to an acute dietary fat challenge may be an adaptation of

DIO mice to chronic consumption of a high fat diet. Adaptation of enterocytes to chronic high fat diet feeding may allow the intestine to maintain efficient dietary fat uptake and absorption.

107

DIO alters CLD associated proteins in response to an acute dietary fat challenge.

Previous observations identified Plin3 localized to the CLD of lean mice in response to an acute dietary fat challenge while Plin2 localized to the CLD of DIO mice under chronic high fat feeding [13]. This study is the first report of CLD associated proteins in enterocytes in DIO mice in response to an acute dietary fat challenge. Surprisingly, we identified Plin2 on CLDs from lean mice in response to an acute dietary fat challenge. In our previous report, we did not identify Plin2 associated with CLDs in enterocytes of lean mice in response to an acute dietary fat challenge [13]. The previous results are consistent with a lipid maturation hypothesis where Plin2 is thought of associating with more mature CLDs and Plin3 associates with nascent CLDs [22]. While it is unclear why Plin2 was identified on CLDs from lean mice in response to an acute dietary fat challenge, there are differences in the methodology in the current study compared to previous data generated in our laboratory. The current study uses a more sensitive proteomic technique to identify CLD associated proteins. Additionally, fresh frozen tissue used in the current study allows for imaging of a greater number of enterocytes compared to isolated enterocytes used in the previous report [13]. As seen in Figure 5, the Plin2 antibody generates a weaker signal around the CLDs in lean mice compared to DIO mice in response to a dietary fat challenge. It is possible that Plin2 was undetected in previous reports because it was present at levels that were below the level of detection.

Lipid related proteins which were found identified either unique or at higher abundance associated with CLD from DIO mice compared to lean in response to an acute dietary fat challenge are similar to those found in models of hepatic steatosis induced by high fat feeding. To further our understanding of the effect of DIO on CLD associated

108

proteins we compared the findings of the current study to a previous report of liver CLD associated protein induced by high fat feeding [12]. The liver is an appropriate model system to compare the small intestine to in regards to lipid metabolism because they share several common features such as the ability to generate both CLDs and lipoproteins.

Although both liver and intestine are capable of generating both triglyceride rich particles, the liver is often thought as a storage site for TAG while the small intestine is thought of as a through fare for TAG. A model of diet induced hepatic steatosis was used for comparison. Hepatic steatosis was induced by chronic high fat feeding (60 kcal% Bio-

Serv, F3282 for 9 weeks) [12]. Several lipid related proteins which were found identified either unique or at higher abundance associated with CLD from DIO mice compared to lean in response to an acute dietary fat challenge which have also been previously reported in the models of hepatic steatosis include ApoC2, Faah, Abcd3, and Acadvl [12].

It is unclear if these proteins are present due to the effect of obesity on a CLD accumulation in response to an acute dietary fat challenge or if they are residual proteins present on CLDs induced by high fat feeding from the previous meal of high fat diet. The concept of proteomic differences between “old” CLD which are residual from the chronic consumption of high fat diet and newly formed CLDs generated in response to a meal is an interesting one, but this study does not address this issue.

Different CLD pools have different protein composition in response to an acute dietary fat challenge. An unexpected observation made in this study was that Plin2 and

Plin3 localize to different CLD pools in enterocytes of DIO mice in response to an acute dietary fat challenge (Figure 6). There are several lines of evidence suggesting that there are sub populations of CLDs in various cell types with different proteins associated with

109

them. While it has not been previously observed in the intestine, the perilipin family

members associate with CLDs at different times during CLD maturation in adipocytes

[22]. Additionally, members of the Cell death-inducing DNA fragmentation factor- ¡-like

effector (CIDE) family of proteins localize to different CLD subpopulations in hepatocytes[23]. Lastly, TAG synthesis enzymes localize to different CLD sub

populations [24]. In particular AGPAT3 and GPAT4 colocalize on a subpopulation of

CLDs. More recently, it has been hypothesized that two groups of lipid droplets exist in

the cell. These CLDs are described as initial lipid droplets and expanding lipid droplets

which have their own distinct protein compositions[25]. The current study provides

supporting evidence for the possibility of different CLD subpopulations which have

distinct sub-proteomes and unique metabolism. This study highlights the power of

tandem mass spectroscopy to identify novel CLD associated proteins and compare the

CLD proteome under different conditions, however it cannot distinguish between

subpopulations of CLDs. This limitations highlights the need for microscopy techniques

to confirm the presence of proteins on or around CLDs and if it the protein localizes to distinct CLD sub population.

Proteins that associate with CLDs in lean and DIO mice in response to an acute dietary fat challenge are associated with a wide range of lipid related pathways (Figure 4 and Table 1). Proteins associated with lipid anabolic and catabolic pathways localize to

CLDs. TAG synthesis enzymes localize to CLDs in Cos7 cells[24] including GPAT4,

AGPAT3, ACSL1 and 3, and DGAT2. In the current study we identify Mogat2, Acsl5, and Dgat1 associated with CLDs from DIO mice (Table 1). Although we did not measure localized TAG synthesis, the presence of these TAG synthesis proteins on the CLD in

110 response to an acute dietary fat challenge indicates the potential for localized TAG synthesis. Enzymes that regulate CLD lipolysis have been well characterized in other cell types including hepatocytes [26] and adipocytes [27] but enzymes that conduct lipolysis in the intestine are under investigated. Several proteins which have been identified as lipases in other cell types have also been identified in this analysis. Lipe (Hormone sensitive lipase) was identified in CLDs from lean and DIO mice in response to an acute dietary fat challenge. Members of the carboxylesterase family have been identified on

CLDs from lean and DIO mice. Some Carboxylesterase family members have been characterized as lipases, however the isoform Ces1f, which was identified on the CLD from DIO mice has not been characterized [28](Table 1), Arylacetamide deacetylase

(Aadac) was identified on CLDs from lean and DIO mice but at higher levels on CLDs from DIO mice and has been characterized as a lipase in the HuH7.5 cells[29].

Additionally, members of the short-chain dehydrogenase/reductase family associate with

CLD and may regulate TAG storage. Members of the short-chain dehydrogenase/reductase family including Dhrs1, HSD17B11, and HSD17B13 have been identified on the CLD from lean and DIO mice. The role of these proteins on the CLD is currently unknown, but evidence suggests that at least 1 family member, HSD17B13 is associated with human subjects with NAFLD and over expression results in increased accumulation of CLDs in mice[30].

The results from this study help generate a hypothetical model to generate hypotheses for further investigation (Figure 7) to explain the effect of DIO on intestinal lipid metabolism. DIO alters CLD morphology and associated proteins in response to an acute dietary fat challenge. Proteins found associated with the CLD are associated with

111 both anabolic and catabolic pathways. Proteins found associated with the CLD may regulate the balance between storage of dietary TAG and the mobilization of TAG to be utilized for chylomicron synthesis. The identification of proteins associated with TAG lipolysis and lipid synthesis associated with CLDs from DIO mice may indicate that mobilized FFA are being utilized for lipid synthesis directly on the CLD instead of being utilized for chylomicron synthesis. This is consistent with the previously reported decrease in TAG secretion observed in DIO mice in response to an acute dietary fat challenge. The identified CLD associated proteins provide potential targets for further investigation into the role of these proteins in dietary fat absorption

We demonstrated that DIO alters CLD morphology and associated proteins in response to an acute dietary fat challenge. CLDs from DIO mice are larger with a different size distribution compared to lean mice. CLDs from DIO mice have some proteins in common and some unique compared to lean mice. Proteins found at higher levels or unique to CLDs from DIO mice are associated with lipid catabolic and anabolic pathways. Proteins identified in this analysis help expand upon the model of the effect of

DIO on dietary fat absorption.

112

3.6 Acknowledgements

We would like to thank member of the Bindley Bioscience Center for their help and expertise.This project was supported by the Indiana Clinical and Translational

Sciences Institute, by Grant # UL1TR001108 from the National Institutes of Health,

National Center for Advancing Translational Sciences, Clinical and Translational

Sciences Award (KKB), Grant # 7-13-IN-05 by the American Diabetes Association,

Innovation Award (KKB),

113

Table 3-1 Label free quantitation of proteins associated with lipid related Gene Ontology terms. 375 identified proteins, 81 proteins are associated with lipid related Gene Ontology terms as identified by GOrilla (Data not shown). Of the 81 lipid related proteins, 25 were found at differentially present levels associated with CLDs from lean or DIO mice.

DIO T Majority Lean LFQ test protein Gene LFQ level P Intestinal IDs Protein names names levels s value CLD Pathway Fatty acid-binding Q3V2F7 protein, liver Fabp1 23.72 25.10 0.029 [7] Lipid binding ATP-binding cassette Lipid transport, fatty Q9DBZ6 sub-family D member 3 Abcd3 22.20 acid oxidation NADPH--cytochrome Oxidoreductase activity, P37040 P450 reductase Por 21.99 23.92 0.017 [5] FAO Long-chain specific acyl-CoA dehydrogenase, Mitochondrial beta P51174 mitochondrial Acadl 21.52 oxidation Very long-chain specific acyl-CoA dehydrogenase, Mitochondrial beta P50544 mitochondrial Acadvl 22.56 oxidation Hydroxyacyl-coenzyme A dehydrogenase, Mitochondrial beta Q61425 mitochondrial Hadh 22.80 [7] oxidation Fatty-acid amide Q3UUF0 hydrolase 1 Faah 20.65 Fatty acid catabolism Q91WU0 Carboxylesterase 1f Ces1f 22.64 Triglyceride lipase Arylacetamide Q99PG0 deacetylase Aadac 21.66 23.41 0.011 Triglyceride lipase 1-acylglycerol-3- phosphate O- Q922Z5 acyltransferase ABHD5 Abhd5 21.53 [5, 6] Lipid homeostasis Core histone macro- Regulation of lipid Q9QZQ8 H2A.1 H2afy 21.56 metabolism Alcohol dehydrogenase Q3UKA4 1 Adh1 20.21 [7] Lipid metabolic process Fatty aldehyde Long-chain-aldehyde Q8C3C9 dehydrogenase Aldh3a2 22.16 23.47 0.049 dehydrogenase activity CMP-N- acetylneuraminate-beta- galactosamide-alpha- Q91Y74 2,3-sialyltransferase 4 St3gal4 23.34 24.37 0.013 [7] Ganglioside metabolism Hydroxymethylglutaryl- CoA synthase, Cholesterol/steroid Q8N7N8 mitochondrial Hmgcs2 21.33 biosynthesis

114

Table 3-1 Continued

G5E850 Cytochrome b5 Cyb5 25.41 24.44 0.039 [7] Sterol biosynthesis

Estradiol 17-beta- Q9EQ06 dehydrogenase 11 Hsd17b11 23.46 25.38 0.027 [5, 6, 7] Steroid synthesis Choline-phosphate P49586 cytidylyltransferase A Pcyt1a 22.63 21.52 0.021 [5] Phospholipid synthesis 2-acylglycerol O- Q80W94 acyltransferase 2 Mogat2 19.78 [4] Triglyceride synthesis Triosephosphate H7BXC3 isomerase Tpi1 22.10 [5, 7] G3P metabolic process Bifunctional ATP- dependent dihydroxyacetone Q8VC30 kinase/FAD-AMP lyase Dak 22.79 21.85 0.005 [7] G3P metabolic process

Glycerol-3-phosphate dehydrogenase E0CXN5 [NAD(+)], cytoplasmic Gpd1 21.01 [7] G3P metabolic process

17-beta-hydroxysteroid Oxidation-reduction Q8VCR2 dehydrogenase 13 Hsd17b13 21.88 process

Q3UJG0 Apolipoprotein C-II Apoc2 19.12 Lipoprotein metabolism [4, 5, 6, Q3U711 Perilipin-2 Plin2 23.92 25.99 0.026 13] Lipid storage

122

3.7 References

[1] S. Demignot, F. Beilstein, E. Morel, Triglyceride-rich lipoproteins and cytosolic lipid droplets in enterocytes: key players in intestinal physiology and metabolic disorders, Biochimie, 96 (2014) 48-55. [2] T. D'Aquila, Y.H. Hung, A. Carreiro, K.K. Buhman, Recent discoveries on absorption of dietary fat: Presence, synthesis, and metabolism of cytoplasmic lipid droplets within enterocytes, Biochimica et biophysica acta, 1861 (2016) 730-747. [3] T.C. Walther, R.V. Farese, Jr., Lipid droplets and cellular lipid metabolism, Annual review of biochemistry, 81 (2012) 687-714. [4] A. Seyer, M. Cantiello, J. Bertrand-Michel, V. Roques, M. Nauze, V. Bezirard, X. Collet, D. Touboul, A. Brunelle, C. Comera, Lipidomic and spatio-temporal imaging of fat by mass spectrometry in mice duodenum during lipid digestion, PloS one, 8 (2013) e58224. [5] J. Bouchoux, F. Beilstein, T. Pauquai, I.C. Guerrera, D. Chateau, N. Ly, M. Alqub, C. Klein, J. Chambaz, M. Rousset, J.M. Lacorte, E. Morel, S. Demignot, The proteome of cytosolic lipid droplets isolated from differentiated Caco-2/TC7 enterocytes reveals cell- specific characteristics, Biology of the cell / under the auspices of the European Cell Biology Organization, 103 (2011) 499-517. [6] F. Beilstein, J. Bouchoux, M. Rousset, S. Demignot, Proteomic analysis of lipid droplets from Caco-2/TC7 enterocytes identifies novel modulators of lipid secretion, PloS one, 8 (2013) e53017. [7] T. D'Aquila, D. Sirohi, J.M. Grabowski, V.E. Hedrick, L.N. Paul, A.S. Greenberg, R.J. Kuhn, K.K. Buhman, Characterization of the proteome of cytoplasmic lipid droplets in mouse enterocytes after a dietary fat challenge, PloS one, 10 (2015) e0126823. [8] M.M. Hussain, Intestinal lipid absorption and lipoprotein formation, Current opinion in lipidology, 25 (2014) 200-206. [9] J.D. Douglass, N. Malik, S.H. Chon, K. Wells, Y.X. Zhou, A.S. Choi, L.B. Joseph, J. Storch, Intestinal mucosal triacylglycerol accumulation secondary to decreased lipid secretion in obese and high fat fed mice, Frontiers in physiology, 3 (2012) 25. [10] A. Uchida, M.C. Whitsitt, T. Eustaquio, M.N. Slipchenko, J.F. Leary, J.X. Cheng, K.K. Buhman, Reduced triglyceride secretion in response to an acute dietary fat challenge in obese compared to lean mice, Frontiers in physiology, 3 (2012) 26. [11] Y. Ding, Y. Wu, R. Zeng, K. Liao, Proteomic profiling of lipid droplet-associated proteins in primary adipocytes of normal and obese mouse, Acta biochimica et biophysica Sinica, 44 (2012) 394-406.[12] S.A. Khan, E.E. Wollaston-Hayden, T.W. Markowski, L. Higgins, D.G. Mashek, Quantitative analysis of the murine lipid droplet- associated proteome during diet-induced hepatic steatosis, Journal of lipid research, 56 (2015) 2260-2272. [13] B. Lee, J. Zhu, N.E. Wolins, J.X. Cheng, K.K. Buhman, Differential association of adipophilin and TIP47 proteins with cytoplasmic lipid droplets in mouse enterocytes during dietary fat absorption, Biochimica et biophysica acta, 1791 (2009) 1173-1180.

123

[14] Y. Xie, F. Nassir, J. Luo, K. Buhman, N.O. Davidson, Intestinal lipoprotein assembly in apobec-1-/- mice reveals subtle alterations in triglyceride secretion coupled with a shift to larger lipoproteins, American journal of physiology. Gastrointestinal and liver physiology, 285 (2003) G735-746. [15] J. Cox, M.Y. Hein, C.A. Luber, I. Paron, N. Nagaraj, M. Mann, Accurate proteome- wide label-free quantification by delayed normalization and maximal peptide ratio extraction, termed MaxLFQ, Molecular & cellular proteomics : MCP, 13 (2014) 2513- 2526. [16] J. Cox, M. Mann, MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification, Nature biotechnology, 26 (2008) 1367-1372. [17] C. UniProt, Activities at the Universal Protein Resource (UniProt), Nucleic acids research, 42 (2014) D191-198. [18] J. Cox, N. Neuhauser, A. Michalski, R.A. Scheltema, J.V. Olsen, M. Mann, Andromeda: a peptide search engine integrated into the MaxQuant environment, Journal of proteome research, 10 (2011) 1794-1805. [19] E. Eden, R. Navon, I. Steinfeld, D. Lipson, Z. Yakhini, GOrilla: a tool for discovery and visualization of enriched GO terms in ranked gene lists, BMC bioinformatics, 10 (2009) 48. [20] E. Eden, D. Lipson, S. Yogev, Z. Yakhini, Discovering motifs in ranked lists of DNA sequences, PLoS computational biology, 3 (2007) e39. [21] D. Szklarczyk, A. Franceschini, S. Wyder, K. Forslund, D. Heller, J. Huerta-Cepas, M. Simonovic, A. Roth, A. Santos, K.P. Tsafou, M. Kuhn, P. Bork, L.J. Jensen, C. von Mering, STRING v10: protein-protein interaction networks, integrated over the tree of life, Nucleic acids research, 43 (2015) D447-452. [22] N.E. Wolins, B.K. Quaynor, J.R. Skinner, M.J. Schoenfish, A. Tzekov, P.E. Bickel, S3-12, Adipophilin, and TIP47 package lipid in adipocytes, The Journal of biological chemistry, 280 (2005) 19146-19155. [23] W. Xu, L. Wu, M. Yu, F.J. Chen, M. Arshad, X. Xia, H. Ren, J. Yu, L. Xu, D. Xu, J.Z. Li, P. Li, L. Zhou, Differential Roles of Cell Death-inducing DNA Fragmentation Factor-alpha-like Effector (CIDE) Proteins in Promoting Lipid Droplet Fusion and Growth in Subpopulations of Hepatocytes, The Journal of biological chemistry, 291 (2016) 4282-4293. [24] F. Wilfling, H. Wang, J.T. Haas, N. Krahmer, T.J. Gould, A. Uchida, J.X. Cheng, M. Graham, R. Christiano, F. Frohlich, X. Liu, K.K. Buhman, R.A. Coleman, J. Bewersdorf, R.V. Farese, Jr., T.C. Walther, Triacylglycerol synthesis enzymes mediate lipid droplet growth by relocalizing from the ER to lipid droplets, Developmental cell, 24 (2013) 384- 399. [25] N. Kory, R.V. Farese, Jr., T.C. Walther, Targeting Fat: Mechanisms of Protein Localization to Lipid Droplets, Trends in cell biology, (2016). [26] A.D. Quiroga, R. Lehner, Liver triacylglycerol lipases, Biochimica et biophysica acta, 1821 (2012) 762-769. [27] S. D'Andrea, Lipid droplet mobilization: The different ways to loosen the purse strings, Biochimie, 120 (2016) 17-27.

124

[28] H. Okazaki, M. Igarashi, M. Nishi, M. Tajima, M. Sekiya, S. Okazaki, N. Yahagi, K. Ohashi, K. Tsukamoto, M. Amemiya-Kudo, T. Matsuzaka, H. Shimano, N. Yamada, J. Aoki, R. Morikawa, Y. Takanezawa, H. Arai, R. Nagai, T. Kadowaki, J. Osuga, S. Ishibashi, Identification of a novel member of the carboxylesterase family that hydrolyzes triacylglycerol: a potential role in adipocyte lipolysis, Diabetes, 55 (2006) 2091-2097. [29] M. Nourbakhsh, D.N. Douglas, C.H. Pu, J.T. Lewis, T. Kawahara, L.F. Lisboa, E. Wei, S. Asthana, A.D. Quiroga, L.M. Law, C. Chen, W.R. Addison, R. Nelson, M. Houghton, R. Lehner, N.M. Kneteman, Arylacetamide deacetylase: a novel host factor with important roles in the lipolysis of cellular triacylglycerol stores, VLDL assembly and HCV production, Journal of hepatology, 59 (2013) 336-343. [30] W. Su, Y. Wang, X. Jia, W. Wu, L. Li, X. Tian, S. Li, C. Wang, H. Xu, J. Cao, Q. Han, S. Xu, Y. Chen, Y. Zhong, X. Zhang, P. Liu, J.A. Gustafsson, Y. Guan, Comparative proteomic study reveals 17beta-HSD13 as a pathogenic protein in nonalcoholic fatty liver disease, Proc Natl Acad Sci U S A, 111 (2014) 11437-11442.

125

CHAPTER 4. LONG CHAIN FATTY ACYL COA SYNTHETASE 5 ABLATION DECREASES TRIGLYCERIDE STORAGE IN ENTEROCYTES AND SECRETION IN RESPONSE TO AN ACUTE DIETARY FAT CHALLENGE COMPARED TO WILD TYPE MICE

4.1 Abstract

Acyl-CoA synthetases (ACSs) activate fatty acids generating fatty acyl-CoA

which is further metabolized by the cell. One member of the ACS family, long chain fatty

acyl Co-A synthetase 5 (Acsl5), is highly expressed in the intestine. Little is known of the

role of Acsl5 in the metabolism of dietary fat in enterocytes. To understand the role of

Acsl5 in dietary fat absorption, we challenged WT and Acsl5-deficient mice with acute and chronic dietary fat challenges. Despite no evidence of quantitative fat malabsorption in the absence of Acls5, we found significantly reduced TAG storage and secretion by enterocytes compared to WT mice. Interestingly, Acsl5-deficient consumed more food, but did not gain more weight compared to WT mice during two weeks of HFD feeding.

Furthermore, we found that Acsl5 localizes to the nucleus after an acute dietary fat challenge. Together this data suggests that although Acsl5 is not essential for quantitative dietary fat absorption, it plays a role in intestinal metabolism of dietary fat.

126

4.2 Introduction

ACSs initiate the first step of fatty acid metabolism in the cell by converting fatty

acid to fatty acyl-CoA [1].There are 25 known members of the ACS family with different

tissue distribution patterns and fatty acid substrate preferences. Long chain fatty acyl

CoA synthetases (Acsls) prefer long chain fatty acids with chain lengths of 13-21 carbons.

Acsl5 is highly expressed in the small intestine, liver, and brown adipose tissue [2]. Acsl5 has multiple subcellular localizations including mitochondria [3-6] and ER [2, 5]. More recently, Acsl5 was also identified [7, 8] and confirmed [8] to be associated with CLDs in enterocytes. Multiple areas of sub cellular localization of Acsl5 suggest Acsl5 may play multiple roles in cellular lipid metabolism.

ACSs can have multiple functions in cells including fatty acid uptake and partitioning of fatty acids into different metabolic fates [9]. Over-expression of Acsl5 in

McArdle-RH7777 cells, a rat hepatoma cell line, increased fatty acid uptake, and preferentially directed exogenous fatty acids toward TAG synthesis [3]. In a separate study, knocking down Acsl5 by siRNA in rat primary hepatocytes decreased fatty acid channeling toward anabolic pathways [10]. In this model, knocking down Acsl5 decreased incorporation of fatty acids into TAG, phospholipids, and cholesterol esters, but did not affect fatty acid uptake. The decrease in incorporation of fatty acids into

complex lipids was accompanied by decreased CLD formation and lipoprotein secretion.

Lastly, Acsl5 knockdown cells have increased fatty acid oxidation. These results suggest

Acsl5 directs fatty acids toward anabolic pathways.

127

To investigate the role of intestinal Acsl5 in dietary fat absorption, Meller et al.

generated an Acsl5-deficient mouse model[11]. Acsl5-deficient mice had a 60% decrease

in total ACS activity in the jejunum compared to WT mice. Despite the decrease in

ACSL activity, Acsl5-deficient mice had similar TAG secretion in response to an acute

dietary fat challenge compared to WT mice. Additionally, Acsl5-deficient mice had

similar growth curves when fed a Western Diet (17% protein, 49% carbohydrates, 21% fat) compared to WT mice. This data suggests that Acsl5 does not play a role in dietary fat absorption by the small intestine.

A new Acsl5-deficient mouse model has been generated with an improved

metabolic phenotype compared to WT mice. Whole body deletion of Acsl5 resulted in a

¡ ¡ reduction of total Acsl activity by ¡80% in jejunal mucosa, 50% in liver, and 37% in

brown adipose tissue lysates [12]. Overall Acsl5-deficient mice had decreased fat mass,

increased metabolic rate, increased respiratory quotient, reduced serum TAG, and

improved insulin sensitivity. This is the mouse model that was used for this study.

To determine the role of Acsl5 in dietary fat absorption we challenged Acsl5=deficient

and WT mice with acute and chronic dietary fat challenges. We hypothesized that Acsl5

directs fatty acids toward TAG synthesis and that the absence of Acsl5 would result in a

decrease in enterocyte TAG storage in CLDs and a decreased secretion rate in response to

an acute dietary fat challenge. To test this hypothesis we analyzed TAG storage in the

jejunum section of Acsl5-deficient mice compared to WT mice in response to an acute

dietary fat challenge. Additionally, we measured TAG secretion in response to an acute

dietary fat challenge in Acsl5-deficient and WT mice. Lastly, to determine if Acsl5

128

deficiency results in fat malabsorption as indicated by steatorrhea, we fed Acsl5-deficient

and WT mice a high fat diet and measured fecal lipid content.

4.3 Materials and methods

4.3.1 Diets and mice

All procedures were approved by the Purdue Animal Care and Use Committee. A

conditional, Acsl5-deficient mouse line was generated by Dr Andrew Greenberg at Tuft’s

University and backcrossed to C57Bl6J line as previously reported (Bowman 2016). All

mice were maintained on a 12- hour light/dark cycle (6AM/6PM) with ad libitum access

to food and water unless indicated. Mice were fed a chow diet or where indicated, a high

fat diet. This chow diet (PicoLab 5043, Lab Diets, Richmond, IN) has 62.1% of calories

from carbohydrate (starch), 24.7 % from protein, and13.2% from fat. The high fat diet

(D12492, Research Diets Inc, New Brunswick, NJ) has 20% calories from carbohydrate

(35% sucrose, 65% from starch), 20% from protein, and 60% from fat (lard). Mice were

euthanized via CO2 asphyxiation.

4.3.2 Tissue imaging

Mice were fasted for 4 hours at the beginning of the light cycle. Mice were then

¡¢£¤£¥¦§¨§¡ © olive oil bolus via oral gavage. Mice were euthanized 2 hours after the oil bolus by CO2 asphyxiation. The small intestine was isolated and flushed with ice cold PBS and divided into 5 equal sections. Sections 2 and 3, representing the jejunum,

129

were analyzed. A small 1 cm2 section of the intestine was taken and cut longitudinally

exposing the lumen. The section was embedded in optimal cutting temperature media

(OCT) and frozen in 2-Methylbutane cooled by dry ice. Embedded tissue was stored at -

20 £C until sectioned.

plus charged microscopy slides (Fisher scientific, Hanover Park, IL, USA). The tissue

was fixed in 2% paraformaldehyde, and permeabilized with 0.1% saponin. The tissue was

probed with a previously validated antibody Acsl5 (a gift from Dr. Andrew Greenberg

from Tufts University, Boston MA, USA). The sections were also stained for neutral

¤¤ § §¤¥ -difluoro-1,3,5,7,8-pentamethyl-4-bora-3a,4a-diaza-s-indacene

(BODIPY) (Life technologies, Grand Island NY, USA), a secondary AlexaFluor antibody

(Life technologies), and for nuclei using 300 nM DAPI (Life technologies). The sections

were imaged using a Nikon A1R confocal microscope (Nikon Instruments Inc., Melville

NY, USA). Image processing was conducted using NIS- Elements C acquisition and

analysis software.

4.3.3 TAG secretion

To assess TAG secretion in response to an acute dietary fat challenge, mice were

fasted for 4 h, starting at the beginning of the light cycle. Mice were then injected with

tyloxapol (500 mg/kg) IP to block lipases required for clearance of TAG from blood.

Thirty minutes after tyloxapol injection, blood was collected via submandibular bleed for

analysis of plasma TAG concentrations (Time 0). Mice were then immediately gavaged

130

gavage for plasma TAG analysis. Plasma TAG concentrations were determined by Wako

L-Type TG M kit (Wako Chemicals USA).

4.3.4 High fat feeding and food intake analysis

Mice were fed a high fat diet (60% kcal) for 2 weeks and body weights

determined weekly. To determine food intake, mice were housed individually in

metabolic cages and food was weighed daily for the last 4 days of high fat feeding.

4.3.5 Fecal lipid analysis

To assess fecal lipid content, feces were collected from mice for the last 4 days of

the 2 week high fat diet feeding. Feces were dried, weighed, ground using mortar and

pestle, and 0.5g was used for lipid analysis. Lipids were extracted using the Folch

extraction (2:1 chloroform: methanol) procedure. Percent fecal lipid content was

determined: (mg extracted lipid/mg dried feces)*100.

4.4 Results

4.4.1 Acsl5-deficient mice accumulate less neutral lipids in CLDs in response to an

acute dietary fat challenge compared to WT mice.

To assess the effect of Acsl5 on enterocyte neutral lipid storage in response to an

¢£ ¤ ¥¦ ¤£ §¨ © £¡ ¤ ¤ ¦¤ ¦ ¢ ¥¦ ¦£¤§¤¥£ ¥ ¡ ¡ - deficient mice. Two hours after the acute dietary fat challenge, the small intestine was

131

isolated, sectioned and stained for Acsl5, neutral lipids, and nuclei (Figure 1). Acsl5 is present in enterocytes of WT mice (Fig 1A) but absent in Acsl5-deficient mice (Fig 1D).

WT mice had greater accumulation of neutral lipids (orange) (Fig 1B) compared to

Acsl5-deficient mice (E) in response to an acute dietary fat challenge.

4.4.2 Decreased TAG secretion in Acsl5-deficient mice compared to WT mice.

administered to WT and Acsl5-deficient mice pretreated with a lipase inhibitor, that

prevents clearance of TAG from blood, Tyloxapol. Blood was collected at T=0, 2, and 4

hours after the olive oil bolus. TAG secretion at 4 hours after an acute dietary fat

challenge was significantly reduced in Acsl5-deficient compared to WT mice (Fig 2).

4.4.3 Acsl5-deficeint mice have increased food intake without differences in body

weight or fecal lipid content compared to WT mice when fed a high fat diet.

To assess the effect of Acsl5 deficiency on weight gain, food intake, and

quantitative dietary fat absorption, WT and Acsl5-deficient mice were fed a HFD (60%

kcal) for two weeks body weight, food intake, and fecal lipid content were determined.

WT and Acsl5 had similar body weight curves during this time period (Fig 3a). Despite

similar body weight gain, Acsl5 deficient mice consumed significantly greater amounts

of food compared to WT mice during the last 3 days of HFD feeding (Fig 3b) with

similar excretion of lipid in feces (Fig 3c).

132

4.4.4 Nuclear localization of Acsl5 in enterocytes in response to an acute dietary fat

challenge in WT mice.

To assess Acsl5 subcellular location in enterocytes in response to an acute dietary

Samples from the jejunum section of the small intestine were isolated before the oil bolus

(after a 4 hour fast) and two hours after the oil bolus. The small intestine was isolated,

sectioned, and probed for Acsl5 (green) and counter stained for lipid (Bodipy, orange)

and nuclei (DAPI, blue) (Figure 4). Before the oil bolus, Acsl5 is found dispersed in the

cell (Figure 4A,B). Two hours after the oil bolus, Acsl5 distinctly localizes to nuclei

(Figure 4C-F).

4.5 Discussion

To understand the role of Acsl5 in dietary fat absorption, we challenged WT and

Acsl5-deficient mice with acute and chronic dietary fat challenges. Despite no evidence

of quantitative fat malabsorption in the absence of Acls5, we found significantly reduced

TAG storage and secretion by enterocytes compared to WT mice. Interestingly, Acsl5-

deficient consumed more food, but did not gain more weight compared to WT mice

during two weeks of HFD feeding. Furthermore, we found that Acsl5 localizes to the

nucleus after an acute dietary fat challenge. Together this data suggests that although

Acsl5 is not essential for quantitative dietary fat absorption, it plays a role in intestinal

metabolism of dietary fat.

133

Members of the ACSL family have been identified associated with CLDs and

contribute to CLD metabolism [13]. Recent proteomic analyses of isolated CLDs from

enterocytes identified [7, 8] and confirmed [8] Acsl5 localization to CLDs after a lipid

challenge. We found less TAG storage in CLDs in Acsl5-deficient mice in response to an

acute dietary fat challenge compared to WT mice. These results are similar to those

observed in primary rat hepatocytes where Acsl5 knockdown resulted in a decrease in

TAG storage [10]. Taken together, these results suggest that Acsl5 works in concert with

TAG synthesis machinery on the CLD to generate TAG for storage.

Blood TAG levels in the postprandial state are a balance between TAG secretion

from the intestine and clearance by peripheral tissues. We found that Acsl5-deficient

mice have reduced TAG secretion in response to an acute dietary fat challenge compared

to WT mice. These results are in contrast to previous reports from a different mouse model that is also deficient in Acsl5 where TAG secretion rates were found to be similar to WT mice [11]. There are several important differences between the two studies. The

differing genetic backgrounds might explain some of the observed phenotypic differences.

The Bowman Acsl5-deficient mice were on a pure C57BL6J background while the

Meller Acsl5-deficient mice were on a mixed C57BL6J/129sJ background. The Bowman

Acsl5-deficient mice had reduced total ACSL activity in liver extracts consistent with

previous published studies showing that siRNA knockdown of ACSL5 in isolated

hepatocytes significantly reduced total ACSL activity [10]. Meller et al. did not report the

expression of all of the other ACSL isoforms in their mouse livers, thus, another ACSL

isoform may have compensated for any reductions in ACSL activity. The two studies also

differ in experimental methodology. The first issue regarding the TAG secretion assay is

134

the amount of dietary fat which was used to challenge the mice. The Meller mouse model

live oil used in the

stress the intestine to delineate differences between Acsl5 deficient and WT mice. The

second issue regarding the TAG secretion assay is the time period of assessment. The

Meller mouse model collected blood samples until 90 minutes. In the current study, we

collected samples 4 hours after the dietary fat challenge and it was only at the 4 hour time

point that TAG secretion differs between Acsl5 deficient and WT mice. These data

suggests that Acsl5 regulates dietary fat secretion in the small intestine, however under specific physiological conditions.

Deficiencies in dietary fat uptake and secretion are often accompanied by fat

malabsorption. Since Acsl5 deficient mice have a decrease in TAG storage and TAG

secretion compared to WT mice, we analyzed fecal lipid content in mice fed a high fat

diet to assess steatorrhea. Surprisingly, Acsl5 deficient mice do not have an increase in

fecal lipid content compared to WT mice when fed a high fat diet. One hypothesis to

explain this observation is Acsl5 deficient mice take up fatty acids in different regions of

the small intestine compared to WT mice. Acsl5 deficient mice may have lower rates of

fatty acid uptake in enterocytes in the upper region of the intestine; instead take up

dietary fatty acids in smaller quantities in different regions. By diffusing dietary fat

uptake along the length of the intestine, Acsl5 deficient mice can have decreased TAG

secretion rates and TAG storage without fat malabsorption. In this study we only assessed

TAG storage in the jejunum section of the small intestine and only assessed TAG

135

secretion and uptake on a whole body physiological level. Future studies to assess fatty

acid uptake and TAG storage and secretion at a cellular level in specific regions of the

small intestine are important to clarify this issue.

We also investigated weight gain and food intake of high fat diet in Acsl5

deficient mice and WT mice. Acsl5-deficient and WT mice consumed similar quantities

of food when fed a chow diet [12]. When Acsl5-deficient mice were fed a HFD however,

Acsl5-deficient mice consumed significantly more food compared to WT mice. Increased food intake may be attributed to a number of factors and the factors which influence an

increase in food intake on a HFD in Acsl5 deficient compared to WT mice have not been

studied. One hypothesis for increased food intake in Acsl5 deficient mice compared to

WT mice is an increased circulating level of FGF21[12]. Overexpression of FGF21

increases food intake when mice are fed chow or high fat diets [15]. Acsl5-deficient mice

may consume more high fat diet compared to WT mice due to their increased levels of

serum FGF21. Despite the increase in food intake, Acsl5 deficient mice had similar

weight gain on high fat diet compared to WT mice when fed for a 2 week period. The

increase in caloric intake may be offset by the increased energy expenditure in Acsl5-

deficient mice compared to WT mice [12].

A surprising observation made during imaging of Acsl5 in response to an acute dietary fat challenge was the dramatic change in subcellular localization of Acsl5. In the

before an oil bolus, Acsl5 is found dispersed in the cell, however when WT mice are

given an acute dietary fat challenge, Acsl5 distinctly localizes to the nucleus of

enterocytes. Acsl5 is known to localize to mitochondria and CLDs in enterocytes [4, 8]

however nuclear localization pattern of Acsl5 has not been previously observed in

136

enterocytes. The presence of long chain fatty acyl-CoA and ACSLs has been detected in

the nucleus of rat liver [16-20]. Acyl-CoAs are thought to be present within the nucleus

and serve as ligands for transcription factors that may regulate expression of proteins

involved in TAG storage and secretion. Acyl-CoAs may be transported into the nucleus

or based on this observation generated within the nucleus. This unexpected discovery

represents a very interesting area for future research.

Acsl5 plays a role in intestinal lipid metabolism in response to dietary fat. Despite

no evidence of quantitative fat malabsorption in the absence of Acls5, we found

significantly reduced TAG storage and secretion by enterocytes compared to WT mice.

Interestingly, Acsl5-deficient consumed more food, but did not gain more weight compared to WT mice during two weeks of HFD feeding. Furthermore, we found that

Acsl5 localizes to the nucleus after an acute dietary fat challenge. Together this data suggests that although Acsl5 is not essential for quantitative dietary fat absorption, it plays a role in intestinal metabolism of dietary fat.

4.6 Acknowledgments

We would like to thank Dr. Andrew Greenberg and members of his lab for help and guidance. In particular we would like to thank, Dr. Tom Bowman and John Griffen for their help.

This work was supported by the Purdue Research Foundation, American Diabetes

Association (7-13-IN-05), the Indiana Clinical and Translational Sciences Institute funded, in part by Grant Number Grant # UL1TR001108 from the National Institutes of

137

Health, National Center for Advancing Translational Sciences, Clinical and Translational

Sciences Award. NIEHS (UO1-ES-020958, RO3-ES-0227), NIDDK (R01 DK098606-

02), NIDDK-Boston Nutrition Obesity Research Center (P30-DK-46200), T32

DK062032-24, and the U.S. Department of Agriculture, Agricultural Research Service, under agreement no. 58-1950-7-70

139

2500

2000

1500 * 1000 WT Acsl5 -/- Plasma TG (mg/dl) 500

0 024 Time post oil bolus (Hrs)

Figure 4-2 Decreased TAG secretion in Acsl5-deficient mice compared to WT mice.

Postprandial TAG secretion rate was measured in WT and Acsl5 -/- mice. Mice were injected with 500 mg/kg Tyloxapol into the interperitoneal cavity to inhibit lipoprotein

lipase activity. Thirty minutes after Tyloxapol injection, the mice were bled (T=0) and

¡ ¢£¤ ¥¦¦ § ¨ ©¨¡¢£ ©¡ ¨ ¢¡ © ¨ ¢ £ ¨ £ £ ¥ ¤ © post oil bolus. Data represented as mean +/- SEM. Asterisk denotes significant difference at the time point (Student T test, p<0.05). n= 3-5 male mice.

140

B A 50 4.0 40 * * * 3.0 30

20 2.0 1.0 Body Weight (g) Body Weight 10 Food intake (g)

0 0.0 0 5 10 15 1234 Days of HFD feeding Day

C 3

2 WT

Acsl5 -/- 1

Fecal lipid (%) Fecal lipid (%)

Figure 4-3 Acsl5-deficeint mice have increased food intake without differences in body weight or fecal lipid content compared to WT mice when fed a high fat diet No different in body weight between Acsl5-/- and WT mice on a high fat diet for 2 weeks. Body weights of Acsl5 -/- and WT mice fed a high fat diet (60% kcal) were monitored for 2 weeks (A). No difference in body weight was determined using a Student t test (p<.05) Data represented as mean +/- SEM. n= 4-5 mice. Increased food intake of Acsl5 - /- mice compared to WT mice when fed a high fat diet (B).Food intake was monitored during the last 4 days of a two week period of high fat diet (60% kcal) feeding. Acsl5 -/- mice had increased food intake compared to WT mice as determined by a Student t test (* p<.05). Data represented as mean +/- SEM n= 4-5 mice. No difference in fecal lipid content in Acsl5 -/- compared to WT mice after 2 weeks of high fat feeding (C). Acsl5 -/- and WT mice were fed a high fat diet (60% kcal) for 2 weeks. Fecal samples were collected during the last 4 days of feeding. Lipid content was determined by Foch extraction of ground feces. Data represented as mean +/- SEM. No difference in fecal lipid content levels was determined by Student T Test p <.05 n=4-5 male mice.

142

4.7 References

[1] T.J. Grevengoed, E.L. Klett, R.A. Coleman, Acyl-CoA metabolism and partitioning, Annu Rev Nutr, 34 (2014) 1-30. [2] D.G. Mashek, L.O. Li, R.A. Coleman, Rat long-chain acyl-CoA synthetase mRNA, protein, and activity vary in tissue distribution and in response to diet, Journal of lipid research, 47 (2006) 2004-2010. [3] D.G. Mashek, M.A. McKenzie, C.G. Van Horn, R.A. Coleman, Rat long chain acyl- CoA synthetase 5 increases fatty acid uptake and partitioning to cellular triacylglycerol in McArdle-RH7777 cells, The Journal of biological chemistry, 281 (2006) 945-950. [4] C. Klaus, E. Kaemmerer, A. Reinartz, U. Schneider, P. Plum, M.K. Jeon, J. Hose, F. Hartmann, M. Schnolzer, N. Wagner, J. Kopitz, N. Gassler, TP53 status regulates ACSL5-induced expression of mitochondrial mortalin in enterocytes and colorectal adenocarcinomas, Cell and tissue research, 357 (2014) 267-278. [5] N. Gassler, W. Roth, B. Funke, A. Schneider, F. Herzog, J.J. Tischendorf, K. Grund, R. Penzel, I.G. Bravo, J. Mariadason, V. Ehemann, J. Sykora, T.L. Haas, H. Walczak, T. Ganten, H. Zentgraf, P. Erb, A. Alonso, F. Autschbach, P. Schirmacher, R. Knuchel, J. Kopitz, Regulation of enterocyte apoptosis by acyl-CoA synthetase 5 splicing, Gastroenterology, 133 (2007) 587-598. [6] E. Kaemmerer, A. Peuscher, A. Reinartz, C. Liedtke, R. Weiskirchen, J. Kopitz, N. Gassler, Human intestinal acyl-CoA synthetase 5 is sensitive to the inhibitor triacsin C, World journal of gastroenterology : WJG, 17 (2011) 4883-4889. [7] F. Beilstein, J. Bouchoux, M. Rousset, S. Demignot, Proteomic analysis of lipid droplets from Caco-2/TC7 enterocytes identifies novel modulators of lipid secretion, PloS one, 8 (2013) e53017. [8] T. D'Aquila, D. Sirohi, J.M. Grabowski, V.E. Hedrick, L.N. Paul, A.S. Greenberg, R.J. Kuhn, K.K. Buhman, Characterization of the proteome of cytoplasmic lipid droplets in mouse enterocytes after a dietary fat challenge, PloS one, 10 (2015) e0126823. [9] D.E. Cooper, P.A. Young, E.L. Klett, R.A. Coleman, Physiological Consequences of Compartmentalized Acyl-CoA Metabolism, The Journal of biological chemistry, 290 (2015) 20023-20031. [10] S.Y. Bu, D.G. Mashek, Hepatic long-chain acyl-CoA synthetase 5 mediates fatty acid channeling between anabolic and catabolic pathways, Journal of lipid research, 51 (2010) 3270-3280 [11] N. Meller, M.E. Morgan, W.P. Wong, J.B. Altemus, E. Sehayek, Targeting of Acyl- CoA synthetase 5 decreases jejunal fatty acid activation with no effect on dietary long- chain fatty acid absorption, Lipids in health and disease, 12 (2013) 88. [12] T.A. Bowman, K.R. O'Keeffe, T. D'Aquila, Q.W. Yan, J.D. Griffin, E.A. Killion, D.M. Salter, D.G. Mashek, K.K. Buhman, A.S. Greenberg, Acyl CoA synthetase 5 (ACSL5) ablation in mice increases energy expenditure and insulin sensitivity and delays fat absorption, Molecular metabolism, 5 (2016) 210-220.

143

[13] Y. Fujimoto, H. Itabe, T. Kinoshita, K.J. Homma, J. Onoduka, M. Mori, S. Yamaguchi, M. Makita, Y. Higashi, A. Yamashita, T. Takano, Involvement of ACSL in local synthesis of neutral lipids in cytoplasmic lipid droplets in human hepatocyte HuH7, Journal of lipid research, 48 (2007) 1280-1292. [14] J. Zhu, B. Lee, K.K. Buhman, J.X. Cheng, A dynamic, cytoplasmic triacylglycerol pool in enterocytes revealed by ex vivo and in vivo coherent anti-Stokes Raman scattering imaging, Journal of lipid research, 50 (2009) 1080-1089. [15] L. Straub, C. Wolfrum, FGF21, energy expenditure and weight loss - How much brown fat do you need?, Molecular metabolism, 4 (2015) 605-609. [16] L. Amigo, M.C. McElroy, M.N. Morales, M. Bronfman, Subcellular distribution and characteristics of ciprofibroyl-CoA synthetase in rat liver. Its possible identity with long- chain acyl-CoA synthetase, The Biochemical journal, 284 ( Pt 1) (1992) 283-287. [17] M. Elholm, A. Garras, S. Neve, D. Tornehave, T.B. Lund, J. Skorve, T. Flatmark, K. Kristiansen, R.K. Berge, Long-chain acyl-CoA esters and acyl-CoA binding protein are present in the nucleus of rat liver cells, Journal of lipid research, 41 (2000) 538-545. [18] A. Ves Losada, R.R. Brenner, Incorporation of delta 5 desaturase substrate (dihomogammalinolenic acid, 20:3 n-6) and product (arachidonic acid 20:4 n-6) into rat liver cell nuclei, Prostaglandins, leukotrienes, and essential fatty acids, 59 (1998) 39-47. [19] A. Ves-Losada, R.R. Brenner, Long-chain fatty Acyl-CoA synthetase enzymatic activity in rat liver cell nuclei, Molecular and cellular biochemistry, 159 (1996) 1-6. [20] A. Ves-Losada, R.R. Brenner, Fatty acid delta 5 desaturation in rat liver cell nuclei, Molecular and cellular biochemistry, 142 (1995) 163-170.

144

CHAPTER 5. SUMMARY AND FUTURE DIRECTIONS

5.1 Summary

Dietary fat provides essential nutrients, contributes to energy balance, and regulates blood lipid concentrations. These functions are important to health, but can also become dysregulated and contribute to diseases such as obesity, diabetes, cardiovascular disease, and cancer. Within enterocytes, the digestive products of dietary fat are resynthesized into triacylglycerol, which is either secreted on chylomicrons or stored within cytoplasmic lipid droplets (CLDs). CLDs were originally thought to be inert stores of neutral lipids, but are now recognized as dynamic organelles that function in multiple cellular processes in addition to lipid metabolism. The objective of this dissertation was

to investigate CLDs and associated proteins and their role in dietary fat absorption.

To identify regulators of CLD metabolism we analyzed the proteomic profile of isolated CLDs from enterocytes in response to an acute dietary fat challenge. We identified 181 proteins associated with the isolated CLD fraction using high resolution

tandem mass spectrometry. Interestingly, the identified proteins are associated with a

wide range of biological and molecular functions. Several of the identified proteins have

been previously identified in other CLDs proteomic analyses from many cell types, while

other identified proteins have not been previously reported. Furthermore, we confirmed

the localization of Plin3, ApoA-IV, and Acsl5 to CLDs by immunohistochemistry and

145 immunoelectron microscopy. Additionally, we investigated the effect of DIO on intestinal CLD morphology and associated proteins. We demonstrated that DIO mice have larger CLDs compared to lean mice in response to an acute dietary fat challenge.

CLDs isolated from DIO mice have some proteins in common and some unique compared to CLDs isolated from lean mice in response to an acute dietary fat challenge.

Proteins found at higher levels or unique to CLDs isolated from DIO mice are associated with both lipid anabolic and catabolic pathways. The newly identified CLD associated proteins expand upon the model of protein regulators of CLD metabolism in enterocytes.

Lastly, we investigated the role of Acsl5 in dietary fat absorption. To characterize the role of Acsl5 in the small intestine we challenged Acsl5-deficient mice and WT mice with acute and chronic dietary fat challenges. We demonstrate Acsl5-deficient mice have decreased TAG secretion and decreased neutral lipid storage in response to an acute dietary fat challenge. Additionally, we show Acsl5-deficeint mice have no defect in quantitative dietary fat absorption when placed on a high fat diet for two weeks.

Additionally, Acsl5-deficient mice have increased food intake without associated weight gain compared to WT mice when fed a high fat diet. Lastly, we observed a dramatic shift to nuclear localization of Acsl5 in response to an acute dietary fat challenge. Taken together, these results indicate that Acsl5 plays an important role in dietary fat metabolism.

The results presented in this dissertation identify new regulators of CLD metabolism and dietary fat absorption in the small intestine. These results provide the groundwork for further investigation into mechanistic details of dietary fat absorption.

146

5.2 Future directions

The chapters in this dissertation highlight how much is unknown regarding the role of CLDs in dietary fat absorption. Several areas of unanswered questions include 1) identifying differences in CLD morphology and associated proteins in different regions of the intestine, 2) identifying changes in CLD morphology and associated proteins over time, and 3) what role Acsl5 plays in the nucleus of enterocytes.

5.2.1 CLD morphology and associated proteins along the length of small intestine

Enterocytes along the length of the intestine are capable of absorbing dietary fat however the mechanistic details of dietary fat absorption along the length of the intestine is relatively unknown. When a small amount of dietary fat is administered, the majority is taken up in the proximal section of the intestine. When dietary fat load increases, the lower region of the small intestine absorbs dietary fat [1]. Most dietary TAG is taken up and stored in CLDs in the proximal section of the small intestine, however TAG storage occurs along the length of the intestine with smaller quantities found in the distal regions

[2]. Little is known about CLD morphology and associated proteins in the duodenum and ileum. Regulators of CLD metabolism may differ in different regions of the intestine and alter that region’s contribution to post prandial lipedema. Identifying CLD associated proteins along the length of the intestine is important for identifying regulators of dietary fat absorption, particularly when a large amount of dietary fat is consumed.

147

5.2.2 Changes in CLD morphology and associated proteins over time in response to an

acute dietary fat challenge.

CLDs are dynamic stores of TAG which increase and decrease over time in

response to an acute dietary fat challenge [2]. This dynamic nature is most likely

regulated by proteins which associate with the CLD. CLD protein composition changes

over time in other cell types [3, 4]. We hypothesize CLDs associated proteins in

enterocytes change over time as CLDs grow and shrink. To test this hypothesis, we

would administer an acute dietary fat challenge to mice and isolate CLDs from enterocytes at 1, 2, 4, and 6 hours after the bolus. Proteins associated with the CLDs at

these time points will be identified by high resolution tandem mass spectroscopy.

We would predict protein composition to change based on protein function and

protein structure. We would predict proteins identified at earlier time points would

include more lipid synthesis proteins compared to later time points. Lipid synthesis

enzymes would allow nascent CLDs to grow and store TAG and other neutral lipid

species. Conversely we would expect more lipid catabolism proteins to be present on

CLDs at later time points compared to earlier time points to mobilize stored lipid.

Additionally, we would predict protein composition to change over time based on

protein structure. Proteins identified at later time points may have structural features

which would allow the protein to directly bind to CLDs. Proteins loosely associated with

CLDs are predicted to be displaced as stored TAG is catabolized and the CLDs shrink [5].

Identifying how CLD associated proteins change over time can identify important

regulators of CLD metabolism and TAG secretion rate.

148

5.2.3 The role of Acsl5 in the nucleus

Localization of Acsl5 to the nucleus of enterocytes in response to dietary fat challenge may regulate dietary fat absorption in the small intestine. Long chain fatty acyl-

CoA has been identified in the nucleus in other cell types. Isolated nuclei from rat livers contain C16:1-CoA and C18:3-CoA [6, 7]. Additionally, long chain fatty acyl-CoA synthestase activity has been detected in isolated rat liver nuclei for a variety of long chain fatty acids [8, 9]. Lastly, long chain fatty acyl-CoAs act as ligands for hepatic nuclear factor 4 alpha (Hnf4a) [10].

Hnf4a is expressed in the enterocyte and are activated by fatty acids derived from the luminal side of the enterocytes [11]. Hnf4a controls the expression of lipid metabolism genes including genes coding for apolipoproteins and microsomal triglyceride transfer protein (Mttp). Mice with an intestine specific inducible Hnf4a knockout have decreased triglyceride tolerance and decreased triglyceride secretion compared to Hnf4a loxp/loxp counterpart [12]. Hnf4a knockout mice have decreased

Apolipoprotein B and Mttp protein expression, Mttp activity and decreased TAG accumulation in enterocytes in the post oil bolus state [13]. The similarities between

Hnf4a knockout mice and Acsl5-deficient mice suggest the two proteins may share similar biological pathways or are mechanistically connected.

We hypothesize that Acsl5 localizes to the nucleus in response to a dietary fat absorption to provide long chain fatty acyl-CoA to serve as ligands for Hnf4a. The nuclear localization of Acsl5 indirectly regulates intestinal dietary fat uptake, metabolism, and secretion. To test this hypothesis we will confirm the localization and activity of

149

Acsl5 in isolated nuclei from enterocytes. We will confirm the presence of Acsl5 in isolated nuclei by Western Blotting analysis and acyl-coa synthetase activity using a variety of long chain fatty acids in an activity assay. Additionally, we will look at expression of Acsl5 downstream target genes in the post bolus state in WT and Acsl5- deficeint mice. Finally, we will determine the binding of Hnf4a to downstream targets using an electrophoretic mobility shift assay in WT and Acsl5-deficient mice. The results of these experiments will indirectly demonstrate the regulation of Hnf4a by Acsl5.

150

5.3 References

[1] C.C. Booth, A.E. Read, E. Jones, Studies on the site of fat absorption: 1. The sites of absorption of increasing doses of I-labelled triolein in the rat, Gut, 2 (1961) 23-31. [2] J. Zhu, B. Lee, K.K. Buhman, J.X. Cheng, A dynamic, cytoplasmic triacylglycerol pool in enterocytes revealed by ex vivo and in vivo coherent anti-Stokes Raman scattering imaging, Journal of lipid research, 50 (2009) 1080-1089. [3] N.E. Wolins, D.L. Brasaemle, P.E. Bickel, A proposed model of fat packaging by exchangeable lipid droplet proteins, FEBS letters, 580 (2006) 5484-5491. [4] N.E. Wolins, B.K. Quaynor, J.R. Skinner, M.J. Schoenfish, A. Tzekov, P.E. Bickel, S3-12, Adipophilin, and TIP47 package lipid in adipocytes, The Journal of biological chemistry, 280 (2005) 19146-19155. [5] N. Kory, A.R. Thiam, R.V. Farese, Jr., T.C. Walther, Protein Crowding Is a Determinant of Lipid Droplet Protein Composition, Developmental cell, 34 (2015) 351- 363. [6] L. Amigo, M.C. McElroy, M.N. Morales, M. Bronfman, Subcellular distribution and characteristics of ciprofibroyl-CoA synthetase in rat liver. Its possible identity with long- chain acyl-CoA synthetase, The Biochemical journal, 284 ( Pt 1) (1992) 283-287. [7] M. Elholm, A. Garras, S. Neve, D. Tornehave, T.B. Lund, J. Skorve, T. Flatmark, K. Kristiansen, R.K. Berge, Long-chain acyl-CoA esters and acyl-CoA binding protein are present in the nucleus of rat liver cells, Journal of lipid research, 41 (2000) 538-545. [8] A. Ves-Losada, R.R. Brenner, Fatty acid delta 5 desaturation in rat liver cell nuclei, Molecular and cellular biochemistry, 142 (1995) 163-170. [9] A. Ves-Losada, R.R. Brenner, Long-chain fatty Acyl-CoA synthetase enzymatic activity in rat liver cell nuclei, Molecular and cellular biochemistry, 159 (1996) 1-6. [10] R. Hertz, J. Magenheim, I. Berman, J. Bar-Tana, Fatty acyl-CoA thioesters are ligands of hepatic nuclear factor-4alpha, Nature, 392 (1998) 512-516. [11] V. Carriere, R. Vidal, K. Lazou, M. Lacasa, F. Delers, A. Ribeiro, M. Rousset, J. Chambaz, J.M. Lacorte, HNF-4-dependent induction of apolipoprotein A-IV gene transcription by an apical supply of lipid micelles in intestinal cells, The Journal of biological chemistry, 280 (2005) 5406-5413. [12] V. Frochot, M. Alqub, A.L. Cattin, V. Carriere, A. Houllier, F. Baraille, L. Barbot, S. Saint-Just, A. Ribeiro, M. Lacasa, P. Cardot, J. Chambaz, M. Rousset, J.M. Lacorte, The transcription factor HNF-4alpha: a key factor of the intestinal uptake of fatty acids in mouse, American journal of physiology. Gastrointestinal and liver physiology, 302 (2012) G1253-1263. [13] V. Marcil, E. Seidman, D. Sinnett, F. Boudreau, F.P. Gendron, J.F. Beaulieu, D. Menard, L.P. Precourt, D. Amre, E. Levy, Modification in oxidative stress, inflammation, and lipoprotein assembly in response to hepatocyte nuclear factor 4alpha knockdown in intestinal epithelial cells, The Journal of biological chemistry, 285 (2010) 40448-40460.

VITA

151

VITA

Theresa D’Aquila PhD Candidate, Interdepartmental Nutrition Program Department of Nutrition Science Purdue University EDUCATION

Purdue University, West Lafayette Indiana Doctorate of Philosophy, August 2016 Interdepartmental Nutrition Program Dissertation: Intestinal Cytoplasmic Lipid Droplets, Associated Proteins, and the Regulation of Dietary Fat Absorption Advisor: Dr. Kimberly Buhman

College of the Holy Cross, Worchester, Massachusetts Bachelor of Arts, Degree in Biology, May 2008 Advisor: Dr. Mary Lee Ledbetter

RESEARCH EXPERIENCE

Graduate Research Assistant, Department of Nutrition Science, Purdue University, August 2011- August 2016 Investigating lipid metabolism in the small intestine and the regulatory role of proteins associated with cytoplasmic lipid droplets and triglyceride storage in dietary fat absorption.

Associate Scientist, Cardiovascular, metabolic, and endocrine diseases unit, Pfizer Inc., June 2008- June 2011 Investigation of small molecule treatments for the pharmacological treatment of Type II Diabetes Mellitus

TEACHING EXPERIENCE

Purdue University • Department of Nutrition Science • Fall 2015 and Spring 2016 NUTR 303: Essentials of Nutrition

152

AWARDS AND HONORS

2016 Most Outstanding Interdisciplinary Award Honorable Mention, Office of Interdisciplinary Graduate Programs Purdue University Purdue Graduate Student Government Travel Grant Purdue University Compton Travel Grant Purdue University Three Minute Thesis (2nd place) Purdue University Sigma Xi Poster Symposium 1st place poster winner Purdue University 2015 Three Minute Thesis (2nd place) Purdue University Applied Management Principles Certificate Purdue University Interdepartmental Nutrition Program Poster Session honorable mention Purdue University Purdue Graduate Student Government Travel Grant Purdue University Compton Travel Grant Purdue University 2014 Purdue Research Fellowship Purdue University Most Outstanding Interdisciplinary Award Honorable Mention, Office of Interdisciplinary Graduate Programs Purdue University 2013 Purdue Research Fellowship Purdue University Office of Interdisciplinary Graduate Programs Spring Reception certificate of excellence Purdue University Sigma Xi Poster Symposium 2nd place poster winner Purdue University 2012 Purdue Research Fellowship Purdue University Chronic Diseases Poster Symposium 2nd place poster winner Purdue University Obesity and Cancer Research Symposium poster winner Purdue University 2011 Lynn Fellowship Purdue University

PUBLICATIONS

153

PUBLICATIONS

D’Aquila, T., Hung Y., Carreiro, A., and Buhman, K.K. (2016) Recent discoveries on absorption of dietary fat: Presence, synthesis, and metabolism of cytoplasmic lipid droplets within enterocytes. Biochim Biophys Acta – Molecular and Cell Biology of Lipids

Bowman TA, O’Keeffe K, D’Aquila T, Yan QW, Griffen JD, Killon EA, Salter DM, Mashek DG, Buhman KK, Greenberg AS. Acyl CoA Synthetase 5 (ACSL5) Ablation in Mice Increases Energy Expenditure and Insulin Sensitivity and Delays Fat Absorption. Molecular Metabolism (2016) Volume 5 , Issue 3 , 210 - 220.

D'Aquila T, Sirohi D, Grabowski J.M, Hedrick V.E., Paul L.N., Greenberg A.S., Kuhn R.J., Buhman K.K., Characterization of the proteome of cytoplasmic lipid droplets in mouse enterocytes after a dietary fat challenge, PloS one, 10 (2015) e0126823

Erion D.M., Lapworth A., Amor P.A., Bai G., Vera N.B., Clark R.W., Yan Q., Zhu Y., Ross T.T., Purkal J., Gorgoglione M., Zhang G., Bonato V., Baker L., Barucci N., D'Aquila T., Robertson A., Aiello R.J., Yan J., Trimmer J., Rolph T.P., Pfefferkorn J.A,, The hepatoselective glucokinase activator PF-04991532 ameliorates hyperglycemia without causing hepatic steatosis in diabetic rats, PloS one, 9 (2014) e97139.

Pfefferkorn J.A, Tu M., Filipski K.J., Guzman-Perez A., Bian J., Aspnes G.E., Sammons M.F., Song W., Li J.C., Jones C.S., Patel L., Rasmusson T., Zeng D., Karki K., Hamilton M., Hank R., Atkinson K., Litchfield J., Aiello R., Baker L., Barucci N., Bourassa P., Bourbonais F., D'Aquila T., Derksen D.R., MacDougall M., Robertson A., The design and synthesis of indazole and pyrazolopyridine based glucokinase activators for the treatment of type 2 diabetes mellitus, Bioorganic & medicinal chemistry letters, 22 (2012) 7100-7105

154

Pfefferkorn J.A., Guzman-Perez A., Litchfield J., Aiello R., Treadway J.L., Pettersen J., Minich M.L., Filipski K.J., Jones C.S., Tu M., Aspnes G., Risley H., Bian J., Stevens B.D., Bourassa P., D'Aquila T., Baker L., Barucci N., Robertson A.S., Bourbonais F., Derksen D.R., Macdougall M., Cabrera O., Chen J., Lapworth A.L., Landro J.A., Zavadoski W.J., Atkinson K., Haddish-Berhane N., Tan B., Yao L., Kosa R.E., Varma M.V., Feng B., Duignan D.B., El-Kattan A., Murdande S., Liu S., Ammirati M., Knafels J., Dasilva-Jardine P., Sweet L., Liras S., Rolph T.P., Discovery of (S)-6-(3-cyclopentyl- 2-(4-(trifluoromethyl)-1H-imidazol-1-yl)propanamido)nicotini c acid as a hepatoselective glucokinase activator clinical candidate for treating type 2 diabetes mellitus, J Med Chem, 55 (2012) 1318-1333.

Pfefferkorn J, Guzman-Perez A, Oates P, Litchfield J, Aspnes G, Basak A, Benbow J, Berliner M, Bian J, Choi C, Freeman-Cook K, Corbett J, Didiuk M, Dunetz J, Filipski K, Hungerford W, Jones C, Karki K, Ling A, Li J, Patel L, Perreault C, Risley H, Saenz J, Song W, Tu M, Aiello R, Atkinson K, Barucci N, Beebe D, Bourassa P, Bourbounais F, Brodeur A, Burbey R, Chen J, D'Aquila T, Derksen D, Haddish-Berhane N, Huang C, Landro J, Lapworth A, MacDougall M, Perregaux D, Pettersen J, Robertson A, Tan B, Treadway J, Liu S, Qiu X, Knafels J, Ammirati M, Song X, DaSilva-Jardine P, Liras S, Sweet L and Rolph T. (2011) Designing glucokinase activators with reduced hypoglycemia risk: discovery of N,N-dimethyl-5-(2-methyl-6-((5-methylpyrazin-2-yl)- carbamoyl)benzofuran-4-yloxy)pyrimidine-2-carboxamide as a clinical candidate for the treatment of type 2 diabetes mellitus. Med. Chem. Commun., 2011, 2, 828-839.