Dietary Sphingomyelin for the Prevention of Diet-Induced Metabolic Dysfunction in Mice" (2018)

Home , Sphingomyelin, Sphingomyelin synthase

University of Connecticut OpenCommons@UConn

Doctoral Dissertations University of Connecticut Graduate School

5-10-2018 Dietary Sphingomyelin for the Prevention of Diet- induced Metabolic Dysfunction in Mice Gregory Harrison Norris University of Connecticut - Storrs, [email protected]

Follow this and additional works at: https://opencommons.uconn.edu/dissertations

Recommended Citation Norris, Gregory Harrison, "Dietary Sphingomyelin for the Prevention of Diet-induced Metabolic Dysfunction in Mice" (2018). Doctoral Dissertations. 1801. https://opencommons.uconn.edu/dissertations/1801 Dietary Sphingomyelin for the Prevention of Diet-induced Metabolic Dysfunction in Mice

Gregory H. Norris

University of Connecticut, 2018

Metabolic dysfunction directly links of obesity and metabolic disorders. Although poor diet and lack of exercise are known factors for the development of metabolic disorders, there is a clear need to identify dietary components that may be effective for preventing metabolic dysfunction. Therefore, this dissertation aimed to define the effects of dietary sphingomyelin (SM) on diet-induced metabolic dysfunction. In Aim 1, we explored the short-term effects of dietary milk and egg SM on lipid homeostasis in C57BL/6J mice fed a high-fat diet. Milk SM reduced both hepatic lipid accumulation and serum cholesterol, which correlates well with the existing literature. Conversely, egg SM increased both hepatic and serum lipids. Having demonstrated differences between two sources of SM, the study progressed to Aim 2, where we investigated the effects of both egg and milk SM on preventing metabolic dysfunction in C57BL/6J mice on an obesogenic diet. Egg SM reduced hypercholesterolemia and fasting hyperglycemia. Both SM treatments reduced hepatic steatosis and adipose inflammation, however, egg SM was more effective than milk SM. When designing Aim 3, we focused on defining possible anti- inflammatory effects of milk SM. We observed that feeding milk SM reduced circulating cytokine concentrations in mice fed an obesogenic diet. We also tested the effects of milk SM and its hydrolytic products, ceramide and sphingosine, on LPS-stimulation of RAW264.7 macrophages.

We observed all three compounds to be effective inhibitors of the LPS-induced inflammatory response. Additionally, SM hydrolysis was required for SM effectiveness. In a normal diet, the consumption of dietary SM found in foods will be accompanied by other dietary phospholipids.

Therefore, we also tested the effects of milk phospholipids supplied by beta-serum, a dairy Gregory H. Norris

University of Connecticut, 2018 byproduct, on diet-induced metabolic dysfunction and inflammation in hypercholesterolemic

LDL-receptor knockout mice. We found that the highest dose of milk phospholipids markedly reduced cholesterol in both serum and liver, with more modest effects on reducing inflammation markers in liver and adipose tissues. Overall, this work demonstrates a clear potential for dietary

SM to be useful for the prevention of obesity-associated metabolic dysfunction. More research is warranted to translate this work into human clinical trials. Dietary Sphingomyelin for the Prevention of Diet-induced Metabolic Dysfunction in Mice

Gregory H. Norris

B.S., University of Connecticut 2010

A Dissertation

Submitted in Partial Fulfillment of the

Requirements for the Degree of

Doctor of Philosophy

at the

University of Connecticut

2018

Gregory H. Norris

2018

APPROVAL PAGE

Doctor of Philosophy Dissertation

Dietary Sphingomyelin for the Prevention of Diet-induced Metabolic Dysfunction in Mice

Presented by

Gregory H. Norris, B.S.

Major Advisor ______

Christopher N. Blesso, Ph.D.

Associate Advisor ______

Cameron Faustman, Ph.D.

Associate Advisor ______

Sung I. Koo, Ph.D.

Associate Advisor ______

Ji-Young Lee, Ph.D.

Associate Advisor ______

Alison Kohan, Ph.D.

University of Connecticut

2018

iii

This dissertation is dedicated in memory of my sister, Kimberly L. Norris (1989-2013), who consistently presented me with challenges throughout my life and provided me with the motivation to finish this degree. May we all remember to be perfectly imperfect.

Acknowledgments

First and foremost, I would like to thank my major advisor, Dr. Blesso. You presented me with an opportunity to transform my career, from a lack luster undergraduate to a Ph.D. I will always appreciate the chance you took with me. I hope you feel as though it has paid dividends. Throughout these four years I have always had funding to carry out the research. It was with your guidance that I have been able to navigate these past 4 years and been as successful as I have been. Throughout my time in your lab, you kept a constant work flow coupled with more difficult challenges to facilitate my intellectual growth. This has given me the practice and adversity it takes to feel as though I have become a full-fledged scientist. I truly feel prepared to take the next step in my path to becoming a professor myself.

As for the rest of my committee:

Dr. Faustman, first I would like to thank you for your precious time. As the interim Dean of CANHR, I know every moment is valuable for you. You have embodied how engaging one can be while still remaining technically sound. Your food chemistry class was easily my favorite elective and I will always refer to your advice to “tell a story” when writing and presenting science.

Dr. Koo, I also am appreciative of your time as our department head. Throughout my time in this department you have contributed funds for pilot studies, travel, the annual holiday party, and American Society of Nutrition memberships. It is this work that keeps our department feeling lively and like a second home. Aside from that, I appreciate all of your critiques through my exams and in my writings. You have an expertise in the field of dietary sphingomyelin and it is well- appreciated to learn from someone so knowledgeable in the field of nutrition.

Dr. Lee, if it was not for your NUSC 4236 I am not sure I would have applied to nutrition doctoral programs. The class married my interests of biochemistry and health in a way I did not imagine in my undergraduate career. It was no more than a year later that I found myself neck deep in “Inflammation, Diets, and Chronic Disease”, which to this day is the hardest, yet most rewarding class I have ever taken. It is with full confidence that I can say that class prepared me for the rest of my graduate course work and jumpstarted my ability to fully understand research articles.

Dr. Kohan, it was your positive attitude that raised my spirits again after feeling like I fumbled my way through my oral defense. You have never been too busy for a random visit to your office for advice on job hunting. Additionally, you were the first one to broaden my vision of what a Ph.D. means. During lunch at my first CANHR forum you told me how foreign languages use to be a requirement for a Ph.D. because it demonstrates your capacity to learn. From that, I have made a conscious effort to learn more than only molecular nutrition. I believe this had made me a more well-rounded person and a better scientist.

The rest of the faculty members in our department have also done an exceptional job of nurturing my curiosity about topics outside of molecular nutrition. Numerous professors have either raised questions that gave me new perspective or taught me some of the basics of research, writing, and presenting, all vital skills for my future.

Special thanks go to Kathryn Parlin and Camilla Crossgrove for managing the financial side of graduate school and lab work for me.

In my time in the Blesso lab, I have had the pleasure of working alongside my fellow graduate students Nicholas Farrell, Courtney Millar, Quinn Duclos, and Christina Jiang. I want to thank the four of you for your support and the sense of comradery in the lab. We have also seen a plethora of undergraduate students in our lab. Caitlin Porter, Christina Jiang, Julia Ryan, Julia Reedy, Nicholas Santangelo, Alexander Cruz, Chelsea Garcia, Jessica Seltz, and Julia Colliton, it was a pleasure to watch you develop in your undergraduate efforts within the Blesso lab, I appreciate every single tube you guy have labeled.

As for my fellow UConn Nutrition graduate students, past and present, there are far too many of you to name in this section. I have had the honor to be among the ranks of some of the most intelligent, motivated, and generally good-spirited people during my time here. Every single one of you has made this department feel like a home to me and I wish you all nothing but the best.

Olivia DeWald, thank you for the love and support throughout the past year and a half. You have been my biggest cheerleader, all the while overachieving as an Au.D. student. I will never cease to be amazed by you. Thank you for pushing me to be my best and to find a near perfect Postdoc in Chicago.

To my family, I would like to thank you for your continued love and support over the past 26 years. I owe my resiliency and work ethic to you. At some point in my life, each one of you have embodied these values and shown me exactly what is possible when you simply refuse to concede.

vii

List of Abbreviations

3-Hydroxy-3-methyl-glutaryl-coenzyme A Reductase HMGCR Acetyl-CoA Carboxylase ACC Acyl-Coenzyme A Dehydrogenase 1 ACAD1 Acyl-Coenzyme A Oxidase 1 ACOX1 Adiponectin (gene) Adipoq Alanine Transaminase ALT Alkaline Sphingomyelinase Alk-SMase Apolipoprotein Apo Aspartate Transaminase AST ATP-Binding Cassette Transporter ABC Bicinchoninic Acid BCA Body Mass Index BMI Carbohydrate Response Element Binding Protein ChREBP Cardiovascular Disease CVD Carnitine Palmitoyltransferase 1 α/β CPT1α/β C-C Motif Chemokine Ligand 2 CCL2 C-C Motif Chemokine Ligand 4 CCL4 CCAAT/enhancer-binding protein α C/EBPα Cell Counting Kit-8 CCK-8 Center for Disease Control CDC Ceramide-Rich Fraction Cer-fr Cholesteryl Esters CE C-Jun N-terminal Kinase JNK Cluster of Differentiation 36 CD36 Crown-like Structures CLS Dextran Sodium Sulfate DSS Diacylglycerol DAG Diacylglycerol Acyltransferase 1/2 DGAT1/2 Dihydrosphingomyelin DHSM Dithiothreitol DTT Endoplasmic Reticulum ER Fatty Acid Binding Protein 4 FABP4 Fatty Acid Synthase FAS Fluorescein Isothiocyanate FITC Forkhead Transcription Factor 1 FOXO1 Free Cholesterol FC Ganglioside GS Glucose Transporter 4 GLUT4

viii

Glucose Transporter 4 (gene) SCL2A4 Glucosylceramide GC Glyceraldehyde 3-phosphate Dehydrogenase GAPDH High Molecular Weight HMW High-Density Lipoprotein Cholesterol HDL-C High-Fat diet HFD Homeostatic Model Assessment of Insulin Resistance HOMA-IR Human Embryonic Kidney HEK Inducible Degrader of LDLr IDOL Insulin Resistance IR Insulin-induced Gene Product INSIG Interferon-inducible T-cell Alpha Chemoattractant I-TAC Interluekin 1β IL-1β Interluekin 6 IL-6 Intraperitoneal IP Limulus Amoebocyte Lysate LAL Lipid-concentrated Butter Serum LC-BS Lipopolysaccharide LPS Lipoprotein Lipase LPL Liver X Receptor LXR Low-Density Lipoprotein Cholesterol LDL-C Low-Density Lipoprotein Receptor LDLr LPS-Binding Protein LBP Macrophage Inflammatory Protein 2 MIP-2 Mechanistic Target of Rapamycin mTOR Microbial Analysis, Resources and Service MARS Milk Fat Globular Membrane MFGM Milk Polar Lipids MPL Mitogen-activated Protein Kinase MAPK Monocyte Chemoattractant Protein 1 MCP-1 Myocyte Enhancer Factor 2 MEF2 Niemann-Pick C1-Like 1 NPC1L1 Non-alcoholic Fatty Liver Disease NAFLD Non-alcoholic Steatohepatitis NASH Nuclear Factor κB NFκB Peroxisome Proliferator-activated Receptors PPARs Phenylmethylsulfonyl Fluoride PMSF Phosphatidylcholine PC Phospholipid PL Phospholipid-Rich Dairy Milk Extract PLRDME Phytosphingosine PS

Polymorphonuclear Leukocytes PMN Polyunsaturated Fatty Acids PUFAs Proprotein Convertase Subtilisin/Kexin Type 9 PCSK9 Quantitative Real-time Polymerase Chain Reaction qRT-PCR Reactive Oxygen Species ROS Regulated on Activation, Normal T Cell Expressed and Secreted RANTES Ribosomal Protein, large, P0 36B4 Serine Palmitoyltransferase SPT Short Chain Fatty Acids SCFAs Soluble Intercellular Adhesion Molecule sICAM Sphingolipids SL Sphingomyelin SM SREBP Cleavage-Activating Protein SCAP Stearoyl-CoA 1 SCD1 Sterol Response Element Binding Protein SREBP Toll-like Receptor 4 TLR4 Total Cholesterol TC Tricyclic Anti-depressant TCA Triglyceride TG Tumor Necrosis Factor α TNF-α Type II Diabetes Mellitus T2DM Unfolded Protein Response UPR Very Low-Density Lipoprotein VLDL Wild-type WT

Table of Contents Chapter 1. Introduction ...... 1 1.1. Introduction ...... 2 1.2. Obesity and Metabolic Dysfunction ...... 2 1.2.1. Inhibition of Lipid Absorption Can Improve Metabolic Dysfunction ...... 4 1.3. Dietary Sphingolipids as a Part of the Western Diet...... 4 1.4. Dietary Sphingomyelin for the Prevention of Metabolic Disease ...... 6 1.5 Objective and Central Hypothesis...... 7 1.6 References ...... 10 Chapter 2. Literature Review ...... 14 2.1. Introduction ...... 16 2.2. Pathogenesis of Metabolic Dysfunction and Insulin Resistance ...... 17 2.2.1. Adipose Tissue Dysfunction ...... 17 2.2.2. The Role of the Gut in Metabolic Dysfunction ...... 18 2.2.3. Systemic Consequences of Metabolic Dysfunction ...... 19 2.3. Molecular Mechanisms Governing Lipid Metabolism ...... 26 2.3.1. Insulin ...... 26 2.3.2. Peroxisome Proliferator-Activated Receptors ...... 27 2.3.3. Liver X Receptor ...... 28 2.3.4. Sterol Response Element Binding Proteins...... 29 2.4. Dietary Sphingomyelin for the Prevention of Metabolic Dysfunction ...... 30 2.4.1. Structure of Sphingomyelin ...... 30 2.4.2. Sphingolipids in the Typical Diet ...... 31 2.4.3. Dietary Sphingomyelin and Lipid Absorption ...... 33 2.4.4. Dietary Sphingolipids and Other Markers of Metabolic Dysfunction ...... 37 2.5. Other potential bioactive roles of dietary SM ...... 44 2.5.1. Anti-inflammatory Effects of Dietary Sphingolipids In Vitro ...... 44 2.5.2. Animal Models of Acute Inflammation ...... 47 2.5.3. Dietary Sphingolipids and Inflammation: Clinical Trials ...... 48 2.5.4. Dietary Sphingolipids Attenuate Animal Models of Chronic Inflammation ...... 49 2.4.5. Dietary Sphingolipids Interactions with the Gut Microbiota ...... 50

2.5. References ...... 53 Chapter 3. Dietary Milk Sphingomyelin Improves Lipid Metabolism and Alters Gut Microbiota in High Fat Diet-fed C57BL/6J Mice ...... 69 3.1. Abstract ...... 71 3.2. Introduction ...... 71 3.3. Materials and Methods ...... 74 3.3.1 Animals and Diets ...... 74 3.3.2. Serum Biochemical Analysis ...... 76 3.3.3. Hepatic Lipid Extraction and Analysis ...... 76 3.3.4. RNA Isolation, cDNA Synthesis and qRT-PCR ...... 76 3.3.5. Gut Permeability Analysis ...... 77 3.3.6. Gut Microbiota Analysis ...... 77 3.3.7. Statistical Analysis ...... 78 3.4. Results ...... 78 3.4.1. Body and Tissue Weights ...... 78 3.4.2. MSM Reduced Cholesterol, While ESM Increased Serum Lipids ...... 79 3.4.3. MSM Alters Intestinal mRNA Expression of Cholesterol Transporters ...... 81 3.4.4. MSM Reduced, While ESM Increased Hepatic Lipids...... 81 3.4.5. SM Increases Skeletal Muscle GLUT4 mRNA Expression with No Effects on Fatty Acid Oxidation-related Gene Expression ...... 82 3.4.6. MSM Reduce Serum LPS but Does Not Change Gut Permeability to FITC-dextran ...... 82 3.4.7. MSM Alters Distal Gut Microbiota Phylogenetic Abundance and Gram-negative Bacteria ... 84 3.5. Discussion ...... 87 3.6. References ...... 94 Chapter 4. Dietary Sphingomyelin Attenuates Hepatic Steatosis and Adipose Tissue Inflammation in High-fat-diet-induced Obese C57BL/6J Mice ...... 98 4.1. Abstract ...... 100 4.2. Introduction ...... 100 4.3. Materials and Methods ...... 102 4.3.1 Animals and Diets ...... 102 4.3.2. Serum Biochemical Analysis ...... 104 4.3.3. Tissue Lipid Extraction and Analysis ...... 105

xii

4.3.4. Tissue Histology ...... 105 4.3.5. RNA Isolation, cDNA Synthesis, and Real Time qRT-PCR ...... 105 4.3.6. SDS-PAGE and Immunoblotting ...... 106 4.3.7. Statistical Analysis ...... 107 4.4. Results ...... 107 4.4.1. Egg SM Reduced HFD-induced Weight Gain and Increases in Adiposity ...... 107 4.4.2. Egg SM Reduced Serum Total Cholesterol and Fasting Glucose ...... 107 4.4.3. Dietary SM Attenuated HFD-induced Hepatic Steatosis and Reduced PPARγ-related mRNA Expression ...... 109 4.4.4. Dietary Egg SM, and to a Lesser Extent Milk SM, Reduces Epididymal Adipose Tissue Inflammation ...... 113 4.4.5. Dietary Egg SM Reduces Skeletal Muscle Lipid Accumulation Induced by HFD ...... 113 4.5. Discussion ...... 115 4.6. References ...... 123 Chapter 5. Dietary Milk Sphingomyelin Reduces Systemic Inflammation in Diet-Induced Obese Mice and Inhibits LPS Activity in Macrophages ...... 127 5.1. Abstract ...... 129 5.2. Introduction ...... 129 5.3. Materials and Methods ...... 132 5.3.1. Animals and Diets ...... 132 5.3.2. Serum Biochemical Analysis ...... 133 5.3.3. Gut Microbiota Analysis ...... 133 5.3.4. Cell Culture ...... 134 5.3.5. Effects of Sphingolipids on LPS Stimulation of RAW264.7 Macrophages ...... 134 5.3.6. RNA isolation, cDNA Synthesis, and qRT-PCR ...... 135 5.2.7. Statistical Analysis ...... 136 5.4. Results ...... 136 5.4.1. Dietary milk SM reduces systemic inflammation and tends to lower circulating LPS ...... 136 5.4.2. Gut microbiota composition is mostly unaffected by 0.1% (w/w) dietary milk SM ...... 139 5.4.3. Milk SM inhibits LPS stimulation of macrophages ...... 139 5.4.4. Ceramides and sphingosine, but not dihydroceramides, inhibit LPS stimulation of macrophages ...... 141 5.5. Discussion ...... 146

xiii

5.6. References ...... 152 Chapter 6. Dietary Milk Phospholipids Improve Dyslipidemia and Hepatic Steatosis in LDLr-/- Mice Fed a Western-type Diet ...... 156 6.1. Abstract ...... 157 6.2. Introduction ...... 157 6.3. Materials and Methods ...... 159 6.3.1 Animals and Diets ...... 159 6.3.2. Serum Biochemical Analysis ...... 160 6.3.3. Hepatic Lipid Extraction and Analysis ...... 161 6.3.4. Tissue Histology ...... 162 3.3.5. RNA Isolation, cDNA Synthesis and qRT-PCR ...... 162 3.3.6. Statistical Analysis ...... 163 6.4. Results ...... 163 6.4.1. Dietary Phospholipids Do Not Alter Body Weight and Food Intake ...... 163 6.4.2. MPL Treatment Attenuated Hypercholesterolemia and Reduced Circulating NEFAs ...... 163 6.4.3. Hepatic Cholesterol Accumulation and Inflammation were reduced by MPLs ...... 164 6.4.4. Consuming MPLs Reduces Ccl2 mRNA Expression in the Epididymal Adipose ...... 166 6.5. Discussion ...... 166 6.6. Acknowledgments ...... 172 6.7. References ...... 173 Chapter 7. Summary, Conclusions, and Future Directions ...... 176 7.1. Summary and Conclusions...... 177 7.2. Future Directions ...... 181 7.2.1. Understanding the Impact of Food Matrix on Dietary SM ...... 181 7.2.2 Further Elucidating the Molecular Mechanism of SM’s Anti-inflammatory Effects ...... 184 7.2.3. Interactions between Dietary SM and Gut Microbiota ...... 186 7.2.4. Dietary SM and Prevention of Chronic Disease ...... 187 7.2.5. Clinical Trials to Measure the Efficacy of SM to Inhibit Cholesterol Absorption in Humans189 7.2.6. Does Dietary SM Correlate with Disease Outcomes? ...... 189 7.3. References ...... 191 Appendix A. Primer List ...... 194

xiv

Index of Figures and Tables Figure 2.1. Schematic summary of metabolic dysfunction ...... 25 Figure 2.2. Structure of common bases, ceramide, and SM ...... 31 Table 2.1. Common dietary sphingolipids ...... 33 Figure 2.3. Compositional differences between egg- and milk-derived SM ...... 34 Table 2.2. Clinical trials evaluating hypolipidemic effects of dietary sphingolipids ...... 38-39 Table 2.3. Dietary sphingolipids and hepatic steatosis ...... 44-45 Table 3.1. Diet composition (Aim 1) ...... 75 Table 3.2. Body and tissue weights of C57BL/6J mice after 4 weeks of high-fat diets ...... 79 Table 3.3. Serum markers of C57BL/6J mice after 4 weeks of high-fat diets ...... 80 Figure 3.1. Serum cholesterol is correlated with serum SM but not serum PC ...... 80 Figure 3.2. MSM significantly altered gene expression related with intestinal cholesterol flux ...... 81 Figure 3.3. MSM reduces hepatic triglycerides and alters cholesterol-responsive gene expression ...... 83 Figure 3.4. MSM increases skeletal muscle GLUT4 and decreases IL-6 mRNA ...... 85 Figure 3.5. MSM reduces serum LPS but does not influence gut permeability ...... 86 Figure 3.6. MSM group distal gut microbiota has significantly reduced Gram-negative bacteria ...... 89 Table 4.1. Diet composition (Aim 2) ...... 104 Figure 4.1. Egg SM reduces body weight and epididymal fat with no change in food intake...... 108 Table 4.2. Serum markers of C57BL/6J mice after 10 weeks of diets ...... 109 Figure 4.2. Dietary SM attenuates hepatic steatosis and reduces hepatic PPARγ-related gene expression ...... 110 Figure 4.3. Hepatic PPARγ-related lipogenic gene expression positively correlates with hepatic TGs ... 111 Figure 4.4. Dietary SM does not alter total PPARγ protein in liver. Hepatic PPARγ protein expression was visualized by immunoblotting ...... 112 Figure 4.5. Dietary SM attenuates inflammation in adipose tissue ...... 114 Figure 4.6 Egg SM attenuates skeletal muscle triglyceride accumulation with HFD ...... 116 Figure 5.1. Dietary milk SM reduces serum inflammation markers in diet-induced obese mice ...... 137 Figure 5.2. Intestinal and mesenteric adipose tissue mRNA expression of diet-induced obese mice ..... 138 Figure 5.3. Gut microbiota composition of diet-induced obese mice ...... 140 Figure 5.4. Milk SM inhibits LPS-activation of macrophages ...... 142 Figure 5.5. Long-chain ceramides inhibit LPS-activation of macrophages ...... 143 Figure 5.6. Sphingosine inhibits LPS-activation of macrophages ...... 144 Figure 5.7. Ceramides and sphingosine do not affect cell viability or apoptosis ...... 145 Table 6.1 Dietary SM content by source ...... 158

Table 6.2. Diet composition by component ...... 161 Table 6.3. Diet composition by macronutrient ...... 162 Table 6.4. Body and tissue weights after 14 weeks of diets ...... 163 Table 6.5 Serum values of LDLr-/- after 14 weeks of diets ...... 164 Figure 6.1. Dietary milk phospholipids reduce hepatic cholesterol accumulation and inflammation ...... 165 Figure 6.2. Milk phospholipids mildly reduce adipose inflammation ...... 167 Figure 7.1. Schematic of dietary sphingomyelin’s effect on metabolic dysfunction ...... 181

xvi

Chapter 1. Introduction

1.1. Introduction

Obesity is a public health crisis. More than 1 out of every 3 Americans are obese and over two- thirds are considered overweight [1]. Nutritional scientists have set forth to address this crisis through multiple methods. Epidemiological and community-based nutrition researchers examine populations’ eating habits, behaviors, and how those are interrelated with the public’s obesity rates.

Clinical researchers are interested in dietary interventions and how they may alter a patient’s metabolic phenotype. Meanwhile, molecular nutrition researchers focus on preclinical trials, finding new potential nutraceuticals to help improve obesity-related outcomes. Many molecular researchers have focused on phytochemicals, plant-based chemicals, as potential nutraceuticals.

This is in part because the epidemiological studies indicate positive health outcomes associated with an increase intake of plant-based foods [2]. This may have suggested animal-based foods to be unhealthy. On the contrary, animal-based foods have many potentially beneficial components including: high-quality protein [3], micronutrients [4], and carotenoids [5]. In fact, the field of zoochemicals, animal-based chemicals, is continuing to produce potential nutraceuticals. One such nutraceutical is sphingomyelin (SM), a sphingolipid commonly found in dairy, meat, and eggs [6].

1.2. Obesity and Metabolic Dysfunction

Obesity raises the risk of all-cause mortality by 21-79% [7, 8]. To combat this, the US spends $200 billion annually, in response to obesity-related health issues [9]. Cardiovascular disease (CVD), type II diabetes (T2D), non-alcoholic fatty liver disease (NAFLD), and cancer all have strong ties to obesity [10]. Thus, curbing obesity would be an effective measure to reduce the prevalence of these life-threatening diseases and the economic burden. Both epidemiological and clinical trials have reported calorie restriction combined with exercise to be an effective means to reduce obesity for most individuals [11-13]. However, according to the Centers for Disease Control (CDC), only

25% of Americans meet weekly goals for both strength and aerobic exercise [14]. Meanwhile, according to the Food and Agriculture Organization, the average American is consuming over

3,000 kilocalories per day [15]. Therefore, despite the knowledge of preventing obesity through calorie restriction and exercise, obesity is projected to increase to 42% of the adult US population by the year 2030 [16].

The increased risk of health complications is not due to merely obesity. It is the underlying metabolic consequences that come from overtaxing the adipose tissues’ energy buffering capacity.

This results in metabolic dysfunction, which manifests itself as dyslipidemia, insulin resistance, and endotoxemia or elevated lipopolysaccharide (LPS) activity [10]. These symptoms of metabolic dysfunction are the defining difference between metabolically-healthy and unhealthy individuals.

Compared to the typically obese population, the metabolically-healthy population has a reduced relative risk of all-cause mortality [8]. Therefore, improving metabolic health could improve the health and economic burdens on the United States. Rodents fed calorically-dense Western-type diets are a common preclinical model of obesity. Western-type diet is an umbrella term encompassing several diets rich in saturated fat, simple sugars, and sometimes added cholesterol.

Rodents fed Western-type diets have increased weight gain, dyslipidemia, insulin resistance, inflammation, and steatotic livers, which is similar to humans [17, 18].

Briefly, the pathogenesis of metabolic syndrome and insulin resistance (IR) stems from overwhelming the lipid buffering capacity of the adipose tissue [19]. The increase in lipids originates both from the diet and an increase in de novo lipogenesis. As adipose tissue enlarges, the adipocytes become stressed, insulin resistant, and secrete inflammatory cytokines and non- esterified fatty acids (NEFA) into circulation [20]. These excessive lipids will accumulate ectopically in the liver and skeletal muscle causing IR in these tissues [21]. This will impair the

3 handling of lipids and glucose, consequentially elevating the risk of the chronic metabolic-related diseases such as CVD, T2D, NAFLD, and cancer.

1.2.1. Inhibition of Lipid Absorption Can Improve Metabolic Dysfunction

As the field of molecular nutrition has grown, the search for potential therapeutic targets to slow down obesity has increased. The ability to prevent metabolic dysfunction is the underlying goal.

Reducing intestinal lipid absorption has been a mechanistic target that has shown some success.

First, inhibiting cholesterol absorption through use of the drug ezetimibe has shown benefits.

Ezetimibe inhibits the Niemann-Pick C1-Like 1 (NPC1L1) protein, which is the major importer of luminal free cholesterol into the enterocyte [22]. Luminal cholesterol is sourced from both dietary and biliary cholesterol. When inhibiting cholesterol absorption, the net result is a lowering of circulating cholesterol [23], and this can reduce hepatic and adipose inflammation associated with a Western-type diet [24-26]. Additionally, ezetimibe has been shown to reduce visceral fat mass, consequently improving metabolic dysfunction [27]. Another commonly prescribed drug is orlistat. Orlistat inhibits intestinal lipases which slows the absorption of lipids by inhibiting their breakdown. This is an effective treatment to reduce body weight [28]. However, pharmacological agents can be costly. Additionally, ezetimibe and orlistat are not without their side effects [29, 30].

Therefore, the inhibition of lipid absorption is a viable target for improving metabolic dysfunction but, there is a need for novel therapies.

1.3. Dietary Sphingolipids as a Part of the Western Diet

On average, 2-8 g/day of a typical American’s diet are phospholipids [31]. Within those 2-8 g, sphingolipids constitute 200-400 mg/day [32]. Animal-derived sphingolipids are by far the most abundant and the most common of them is SM [33]. Therefore, the majority of dietary sphingolipids are SM. The average American adult consumes 200-400 mg of cholesterol per day

[34]. Despite the relative abundances of these lipids, dietary cholesterol has dominated the literature (>9,000 papers available on PubMed as of January 2018) in comparison to dietary sphingolipids (<100 papers). This discrepancy exists in spite of the potential bioactive effects of sphingolipids [32]. Therefore, there is a clear lack of research into the effects of dietary sphingolipids.

Once dietary SM reaches the small intestine its hydrolysis is initiated by bile salt-dependent alkaline sphingomyelinase (alk-SMase), located on the brush border membrane in mammals and additionally found in human bile [35]. Sphingomyelinase activity is low in the duodenum and peaks in the mid-jejunum [36]. The delay in activity is due in part to the inhibition of alk-SMase by common luminal lipids [37]. Hydrolysis of SM produces a phosphocholine and ceramide.

Subsequently, the ceramide will be digested by neutral ceramidase to produce sphingosine and a fatty acid [35]. The sphingosine backbone will be readily absorbed by the enterocyte through a passive diffusion mechanism [38]. Once absorbed, most of the sphingosine will enter the sphingosine-1-phosphate lyase pathway to recover a fatty acid while, some can be used to resynthesize complex sphingolipids [35].

Different food sources of SM contain structurally distinct SM profiles. Egg-derived SM contains predominantly C16:0 N-linked acyl chains and d18:1 as the sphingoid backbone [39].

Comparatively, milk-derived SM contains N-linked acyl chains ranging from C16:0-C24:0 and sphingoid backbones ranging from 16-19 carbons with and without a double bond [40]. The longer saturated chains could increase hydrophobic interactions with cholesterol [41]. The SM species with a saturated backbone in milk-derived SM are classified as dihydrosphingomyelin.

Dihydrosphingomyelin has been shown to slow cholesterol’s desorption from a model membrane to cyclodextrin more strongly than sphingomyelin [42]. Dihydroceramides have been shown to be

5 inert in terms of apoptosis when compared to ceramide [43]. However, new insights into the potential biological role of dihydroceramide metabolism are surfacing. While exogenous dihydroceramides seem biologically inert [44], a build-up of dihydroceramide has been shown to induce autophagy and coincides with both oxidative stress and hypoxia [45]. However, due to the interconnected nature of sphingolipid metabolism, it is hard to discern whether these effects are associated with increases in dihydroceramide or depletion of ceramide. Regardless, there is increasing data suggesting that subtle differences between ceramide species can have distinct biological effects [46]. Therefore, structurally-distinct dietary SM could have differing biological effects.

1.4. Dietary Sphingomyelin for the Prevention of Metabolic Disease

Exogenous SM and its metabolites have been shown in cell culture models to modulate cholesterol absorption [47-49], in part by decreasing the release of cholesterol from the micelles [50].

Furthermore, in animal models, reductions in both triglyceride and cholesterol absorption were observed [41, 50-52]. In lymph duct-cannulated rats, milk-derived SM was more potent than egg- derived SM when inhibiting cholesterol and triglycerides during duodenal infusion [41]. This suggests that SM from different sources will have differential effects on lipid absorption. Since then, studies conducted in rodents have observed reductions in serum and hepatic cholesterol during chronic feeding of SM on a high-fat diet [52-54]. In humans, intervention studies with purified dietary SM are scarce, but the consumption of dietary milk-derived SM for 14 days has been shown to raise high-density lipoprotein cholesterol (HDL-C) in healthy adult men and women

[55]. There are consistently observed reductions in serum and hepatic lipids with feeding of SM.

However, comparing distinct SMs could elucidate further understanding of the mechanism.

Dietary SM has also been investigated for its role in chemically-induced colitis and inflammation. Two separate research groups found that dietary SM mixed into the diet reduced inflammation induced by dextran sodium sulfate (DSS) treatment of mice [56, 57]. However, when mice were gavaged with similar quantities of egg-derived SM, both DSS-induced and IL-10-/-- induced forms of colitis were exacerbated [58]. This could be due to differences in the timing of the dose, as the studies inhibiting inflammation gave SM over the course of the day as opposed to a single dose. This observation is consistent with large doses of exogenous SM and ceramide having been shown to be pro-inflammatory in cell culture [59], while lower doses of exogenous sphingolipids are anti-inflammatory [60]. This suggests a delicate balance in sphingolipids regulates their role in inflammation.

1.5 Objective and Central Hypothesis

The main objective of this work is to further clarify the role of dietary SM as a therapeutic food component in preventing diet-induced metabolic dysfunction and inflammation utilizing both animal and cell models. We hypothesize that dietary SM will be beneficial in attenuating the progression of dyslipidemia, metabolic dysfunction, and inflammation in mice fed Western-type diets. These effects will likely be related to the ability of dietary SM to inhibit the absorption of other dietary, biliary, and possibly microbial lipids. Metabolic dysfunction is also associated with systemic inflammation. Therefore, we expect to see a reduction in systemic inflammation with dietary SM. Dietary SM has shown potential in alleviating chemically-induced colitis through a poorly understood mechanism. Logically, there is potential for SM to reduce systemic inflammation in a manner independent from its effects on lipid homeostasis. Thus, we will test an in vitro model of inflammation to examine potential anti-inflammatory effects of SM. Purified SM

7 is currently prohibitively expensive and therefore we will test a less expensive alternative, beta- serum, a dairy byproduct that is rich in phospholipids, like SM, for reducing hypercholesterolemia.

Question 1: Are there differences in the effects of structurally-distinct dietary SM derived from egg and milk on metabolic dysfunction when fed a short-term Western-type diet?

Hypothesis 1: We hypothesize that both SM sources will improve Western-type diet-induced metabolic dysfunction and endotoxemia, but milk-derived SM will be more potent compared to egg-derived SM. This difference will be due to milk-derived SM’s more potent inhibitory effect on lipid absorption compared to egg-derived SM.

Question 2: Will either milk or egg SM slow the progression of diet-induced obesity and the onset of metabolic dysfunction in mice fed a Western-type diet?

Hypothesis 2: Both SM treatment groups will reduce the metabolic dysfunction phenotype, due to SM’s effects on both lipid absorption and inflammation compared to a high-fat control.

However, milk-derived SM will be more effective than egg-derived SM because of the stronger inhibitory effect on lipid absorption.

Question 3: Will exogenous SM inhibit the LPS-stimulation of macrophages, and will this effect be dependent on SM hydrolysis?

Hypothesis 3: We hypothesize exogenous SM treatment will inhibit LPS-stimulation of

RAW264.7 macrophages. Due to the fact that SM metabolites have been shown to have inhibitory properties, blocking hydrolysis of SM will reduce SM’s effectiveness for inhibiting LPS- stimulation. Consequently, the addition of the hydrolytic products of SM will also inhibit LPS-

8 stimulation. Furthermore, dihydroceramide and ceramide with have differential effects on inflammatory response, as they have previously shown different bioactivities.

Question 4: Does phospholipid supplementation confer hypolipidemic benefits and improve metabolic dysfunction in genetically hypercholesterolemic LDLr-/- mice?

Hypothesis 4: Phospholipid supplementation will reduce dyslipidemia and insulin resistance in

Western-type diet-fed LDLr-/- mice. This will result in an improved hepatic phenotype and decreased adipose inflammation.

1.6 References

1. Ogden CL, Carroll MD, Fryar CD, Flegal KM: Prevalence of Obesity Among Adults and Youth: United States, 2011-2014. NCHS Data Brief 2015:1-8. 2. Rolls BJ, Ello-Martin JA, Tohill BC: What can intervention studies tell us about the relationship between fruit and vegetable consumption and weight management? Nutr Rev 2004, 62:1-17. 3. Tome D: Criteria and markers for protein quality assessment - a review. Br J Nutr 2012, 108 Suppl 2:S222-229. 4. De Smet S, Vossen E: Meat: The balance between nutrition and health. A review. Meat Sci 2016, 120:145-156. 5. Nimalaratne C, Wu J, Schieber A: Egg Yolk Carotenoids: Composition, Analysis, and Effects of Processing on Their Stability. In Carotenoid Cleavage Products. Volume 1134: American Chemical Society; 2013: 219-225: ACS Symposium Series]. 6. Weihrauch JL, Son Y-S: Phospholipid content of foods. Journal of the American Oil Chemists' Society 1983, 60:1971-1978. 7. Flegal KM, Kit BK, Orpana H, Graubard BI: Association of All-Cause Mortality With Overweight and Obesity Using Standard Body Mass Index Categories: A Systematic Review and Meta-analysis. JAMA 2013, 309:71-82. 8. Hamer M, Stamatakis E: Metabolically Healthy Obesity and Risk of All-Cause and Cardiovascular Disease Mortality. The Journal of Clinical Endocrinology & Metabolism 2012, 97:2482-2488. 9. Cawley J, Meyerhoefer C: The medical care costs of obesity: an instrumental variables approach. J Health Econ 2012, 31:219-230. 10. Monteiro R, Azevedo I: Chronic Inflammation in Obesity and the Metabolic Syndrome. Mediators of Inflammation 2010, 2010:289645. 11. Lee YJ, Lee A, Yoo HJ, Kim M, Kim M, Jee SH, Shin DY, Lee JH: Effect of weight loss on circulating fatty acid profiles in overweight subjects with high visceral fat area: a 12-week randomized controlled trial. Nutr J 2018, 17:28. 12. Gay JL, Monsma EV, Hein KD: Weight Management Behaviors Among Mexican American Youth: Cross-Sectional Variation by Timing of Growth and Maturation. Am J Health Promot 2018, 32:392-399. 13. Arent SM, Walker AJ, Pellegrino JK, Sanders DJ, McFadden BA, Ziegenfuss TN, Lopez HL: The Combined Effects of Exercise, Diet, and a Multi-Ingredient Dietary Supplement on Body Composition and Adipokine Changes in Overweight Adults. J Am Coll Nutr 2018, 37:111-120. 14. Early release of selected estimates based on data from the 2016 National Health Interview Survey, data tables for figures 7.1, 7.4. 15. NASA/Food and Agriculture Organization of the United Nations, 2013. Food Balance Sheet. 16. Finkelstein EA, Khavjou OA, Thompson H, Trogdon JG, Pan L, Sherry B, Dietz W: Obesity and severe obesity forecasts through 2030. Am J Prev Med 2012, 42:563-570. 17. Hariri N, Thibault L: High-fat diet-induced obesity in animal models. Nutrition Research Reviews 2010, 23:270-299. 18. Mensink RP, Zock PL, Kester AD, Katan MB: Effects of dietary fatty acids and carbohydrates on the ratio of serum total to HDL cholesterol and on serum lipids and

10 apolipoproteins: a meta-analysis of 60 controlled trials. The American Journal of Clinical Nutrition 2003, 77:1146-1155. 19. Blüher M: Adipose Tissue Dysfunction in Obesity. Exp Clin Endocrinol Diabetes 2009, 117:241-250. 20. Chan P-C, Hsieh P-S: The Role of Adipocyte Hypertrophy and Hypoxia in the Development of Obesity-Associated Adipose Tissue Inflammation and Insulin Resistance. In Adiposity - Omics and Molecular Understanding. Edited by Gordeladze JO. Rijeka: InTech; 2017: Ch. 07 21. Guebre-Egziabher F, Alix PM, Koppe L, Pelletier CC, Kalbacher E, Fouque D, Soulage CO: Ectopic lipid accumulation: A potential cause for metabolic disturbances and a contributor to the alteration of kidney function. Biochimie 2013, 95:1971-1979. 22. Phan BAP, Dayspring TD, Toth PP: Ezetimibe therapy: mechanism of action and clinical update. Vascular Health and Risk Management 2012, 8:415-427. 23. Lipka LJ: Ezetimibe: a first-in-class, novel cholesterol absorption inhibitor. Cardiovasc Drug Rev 2003, 21:293-312. 24. Subramanian S, Han CY, Chiba T, McMillen TS, Wang SA, Haw A, 3rd, Kirk EA, O'Brien KD, Chait A: Dietary cholesterol worsens adipose tissue macrophage accumulation and atherosclerosis in obese LDL receptor-deficient mice. Arterioscler Thromb Vasc Biol 2008, 28:685-691. 25. Umemoto T, Subramanian S, Ding Y, Goodspeed L, Wang S, Han CY, Teresa AS, Kim J, O'Brien KD, Chait A: Inhibition of intestinal cholesterol absorption decreases atherosclerosis but not adipose tissue inflammation. J Lipid Res 2012, 53:2380-2389. 26. Subramanian S, Goodspeed L, Wang S, Kim J, Zeng L, Ioannou GN, Haigh WG, Yeh MM, Kowdley KV, O'Brien KD, et al: Dietary cholesterol exacerbates hepatic steatosis and inflammation in obese LDL receptor-deficient mice. J Lipid Res 2011, 52:1626-1635. 27. Takase H, Dohi Y, Okado T, Hashimoto T, Goto Y, Kimura G: Effects of ezetimibe on visceral fat in the metabolic syndrome: a randomised controlled study. European Journal of Clinical Investigation 2012, 42:1287-1294. 28. Jain SS, Ramanand SJ, Ramanand JB, Akat PB, Patwardhan MH, Joshi SR: Evaluation of efficacy and safety of orlistat in obese patients. Indian Journal of Endocrinology and Metabolism 2011, 15:99-104. 29. Stolk MF, Becx MC, Kuypers KC, Seldenrijk CA: Severe hepatic side effects of ezetimibe. Clin Gastroenterol Hepatol 2006, 4:908-911. 30. Garcia SB, Barros LT, Turatti A, Martinello F, Modiano P, Ribeiro-Silva A, Vespucio MV, Uyemura SA: The anti-obesity agent Orlistat is associated to increase in colonic preneoplastic markers in rats treated with a chemical carcinogen. Cancer Lett 2006, 240:221-224. 31. Cohn JS, Kamili A, Wat E, Chung RWS, Tandy S: Dietary Phospholipids and Intestinal Cholesterol Absorption. Nutrients 2010, 2:116-127. 32. Vesper H, Schmelz EM, Nikolova-Karakashian MN, Dillehay DL, Lynch DV, Merrill AH, Jr.: Sphingolipids in food and the emerging importance of sphingolipids to nutrition. J Nutr 1999, 129:1239-1250. 33. Canela N, Herrero P, Marine S, Nadal P, Ras MR, Rodriguez MA, Arola L: Analytical methods in sphingolipidomics: Quantitative and profiling approaches in food analysis. J Chromatogr A 2016, 1428:16-38.

34. Services UDoHaH: Scientific Report of the 2015 Dietary Guidelines Advisory Committee. (Agriculture UDo ed. pp. 13. Washington DC: US Department of Health and Human Services; 2016:13. 35. Duan R-D: Alkaline sphingomyelinase: An old enzyme with novel implications. Biochimica et Biophysica Acta (BBA) - Molecular and Cell Biology of Lipids 2006, 1761:281-291. 36. Nyberg L, Nilsson Å, Lundgren P, Duan R-D: Localization and capacity of sphingomyelin digestion in the rat intestinal tract. The Journal of Nutritional Biochemistry 1997, 8:112-118. 37. Liu J-J, Nilsson Å, Duan R-D: Effects of phospholipids on sphingomyelin hydrolysis induced by intestinal alkaline sphingomyelinase: an in vitro study. The Journal of Nutritional Biochemistry 2000, 11:192-197. 38. Garmy N, Taieb N, Yahi N, Fantini J: Apical uptake and transepithelial transport of sphingosine monomers through intact human intestinal epithelial cells: physicochemical and molecular modeling studies. Arch Biochem Biophys 2005, 440:91-100. 39. Byrdwell WC, Perry RH: Liquid chromatography with dual parallel mass spectrometry and (31)P nuclear magnetic resonance spectroscopy for analysis of sphingomyelin and dihydrosphingomyelin. I. Bovine brain and chicken egg yolk. J Chromatogr A 2006, 1133:149-171. 40. Byrdwell WC, Perry RH: Liquid chromatography with dual parallel mass spectrometry and 31P nuclear magnetic resonance spectroscopy for analysis of sphingomyelin and dihydrosphingomyelin. II. Bovine milk sphingolipids. J Chromatogr A 2007, 1146:164-185. 41. Noh SK, Koo SI: Milk sphingomyelin is more effective than egg sphingomyelin in inhibiting intestinal absorption of cholesterol and fat in rats. J Nutr 2004, 134:2611-2616. 42. Kuikka M, Ramstedt B, Ohvo-Rekila H, Tuuf J, Slotte JP: Membrane properties of D- erythro-N-acyl sphingomyelins and their corresponding dihydro species. Biophys J 2001, 80:2327-2337. 43. Bielawska A, Crane HM, Liotta D, Obeid LM, Hannun YA: Selectivity of ceramide- mediated biology. Lack of activity of erythro-dihydroceramide. J Biol Chem 1993, 268:26226- 26232. 44. Bielawska A, Crane HM, Liotta D, Obeid LM, Hannun YA: Selectivity of ceramide- mediated biology. Lack of activity of erythro-dihydroceramide. Journal of Biological Chemistry 1993, 268:26226-26232. 45. Siddique MM, Li Y, Chaurasia B, Kaddai VA, Summers SA: Dihydroceramides: From Bit Players to Lead Actors. Journal of Biological Chemistry 2015, 290:15371-15379. 46. Hannun YA, Obeid LM: Many Ceramides. Journal of Biological Chemistry 2011, 286:27855-27862. 47. Nyberg L, Duan R-D, Nilsson Å: A mutual inhibitory effect on absorption of sphingomyelin and cholesterol. Journal of Nutritional Biochemistry 2000, 11:244-249. 48. Feng D, Ohlsson L, Ling W, Nilsson Å, Duan R-D: Generating Ceramide from Sphingomyelin by Alkaline Sphingomyelinase in the Gut Enhances Sphingomyelin-Induced Inhibition of Cholesterol Uptake in Caco-2 Cells. Digestive Diseases and Sciences 2010, 55:3377-3383. 49. Garmy N, Taieb N, Yahi N, Fantini J: Interaction of cholesterol with sphingosine: physicochemical characterization and impact on intestinal absorption. J Lipid Res 2005, 46:36-45.

50. Eckhardt ER, Wang DQ, Donovan JM, Carey MC: Dietary sphingomyelin suppresses intestinal cholesterol absorption by decreasing thermodynamic activity of cholesterol monomers. Gastroenterology 2002, 122:948-956. 51. Noh SK, Koo SI: Egg sphingomyelin lowers the lymphatic absorption of cholesterol and alpha-tocopherol in rats. J Nutr 2003, 133:3571-3576. 52. Chung RWS, Kamili A, Tandy S, Weir JM, Gaire R, Wong G, Meikle PJ, Cohn JS, Rye K-A: Dietary Sphingomyelin Lowers Hepatic Lipid Levels and Inhibits Intestinal Cholesterol Absorption in High-Fat-Fed Mice. PLOS ONE 2013, 8:e55949. 53. Duivenvoorden I, Voshol PJ, Rensen PC, van Duyvenvoorde W, Romijn JA, Emeis JJ, Havekes LM, Nieuwenhuizen WF: Dietary sphingolipids lower plasma cholesterol and triacylglycerol and prevent liver steatosis in APOE*3Leiden mice. The American Journal of Clinical Nutrition 2006, 84:312-321. 54. Yunoki K, Renaguli M, Kinoshita M, Matsuyama H, Mawatari S, Fujino T, Kodama Y, Sugiyama M, Ohnishi M: Dietary sphingolipids ameliorate disorders of lipid metabolism in Zucker fatty rats. J Agric Food Chem 2010, 58:7030-7035. 55. Ramprasath VR, Jones PJ, Buckley DD, Woollett LA, Heubi JE: Effect of dietary sphingomyelin on absorption and fractional synthetic rate of cholesterol and serum lipid profile in humans. Lipids Health Dis 2013, 12:125. 56. Mazzei JC, Zhou H, Brayfield BP, Hontecillas R, Bassaganya-Riera J, Schmelz EM: Suppression of intestinal inflammation and inflammation-driven colon cancer in mice by dietary sphingomyelin: importance of peroxisome proliferator-activated receptor gamma expression. J Nutr Biochem 2011, 22:1160-1171. 57. Furuya H, Ohkawara S, Nagashima K, Asanuma N, Hino T: Dietary sphingomyelin alleviates experimental inflammatory bowel disease in mice. Int J Vitam Nutr Res 2008, 78:41- 49. 58. Fischbeck A, Leucht K, Frey-Wagner I, Bentz S, Pesch T, Kellermeier S, Krebs M, Fried M, Rogler G, Hausmann M, Humpf HU: Sphingomyelin induces cathepsin D-mediated apoptosis in intestinal epithelial cells and increases inflammation in DSS colitis. Gut 2011, 60:55-65. 59. Sakamoto H, Yoshida T, Sanaki T, Shigaki S, Morita H, Oyama M, Mitsui M, Tanaka Y, Nakano T, Mitsutake S, et al: Possible roles of long-chain sphingomyelines and sphingomyelin synthase 2 in mouse macrophage inflammatory response. Biochem Biophys Res Commun 2017, 482:202-207. 60. Jozefowski S, Czerkies M, Lukasik A, Bielawska A, Bielawski J, Kwiatkowska K, Sobota A: Ceramide and ceramide 1-phosphate are negative regulators of TNF-alpha production induced by lipopolysaccharide. J Immunol 2010, 185:6960-6973.

Chapter 2. Literature Review

Portions of this chapter are reprinted from “Norris GH, Blesso CN: Dietary sphingolipids: potential for management of dyslipidemia and nonalcoholic fatty liver disease. Nutrition Reviews 2017, 75:274-285.” As the author of this Oxford article, Gregory Norris retains the right to include this article in a dissertation, permission from the publisher is not require.

Portions of this chapter are reprinted from “Norris G, Blesso C: Dietary and Endogenous Sphingolipid Metabolism in Chronic Inflammation. Nutrients 2017, 9:1180.” As the author of this MPDI article, Gregory Norris retains the right to include this article in a dissertation, permission from the publisher is not required.

2.1. Introduction

In 2014, 36.5% of the US adult population was considered obese with a body mass index (BMI)

≥30 [1]. Obesity is estimated to cost the US $200 billion annually due to complications associated with increased body weight [2]. Obesity increases the relative risk of all-cause mortality by 21-

79% [3, 4] due to metabolic dysfunction. Metabolic dysfunction is characterized by dyslipidemia, insulin resistance (IR), and endotoxemia [5]. These symptoms are driving factors in the development of type II diabetes, cardiovascular disease (CVD), and non-alcoholic fatty liver disease (NAFLD) [5]. This is conceptually supported by metabolically healthy obese individuals, that have a reduced relative risk of all-cause mortality compared to metabolically unhealthy patients [4]. The metabolically healthy obese population lacks some of the critical symptoms of metabolic dysfunction. Although proper diet and exercise can reduce BMI and lessen metabolic dysfunction [6], the number of obese US adults is expected to increase to 42% of the total adult population by the year 2030 [7].

One cause of increased BMI is the overconsumption of nutrients. Chronic consumption of a high-fat diet (HFD) may provide the greatest risk of over-nutrition, due to high energy density in concert with a lesser impact on satiety and satiation per calorie compared to other macronutrients

[8]. Several epidemiological studies show when dietary fat increases, so does the prevalence of obesity [9-11]. High-fat diets are used in rodent models to induce characteristics similar to human obesity, such as increased weight gain, dyslipidemia, IR, inflammation, and steatotic livers [12,

13]. Thus, HFD-fed rodents are commonly used in preclinical trials to study obesity. The strong association of these comorbidities with metabolic disease warrants the identification of dietary interventions to prevent or slow their progression. The overall goal of these interventions is to reduce the incidence of obesity-related disease.

2.2. Pathogenesis of Metabolic Dysfunction and Insulin Resistance

2.2.1. Adipose Tissue Dysfunction

Historically, obesity’s role in metabolic dysfunction was attributed to adipose tissue inflammation and IR. More recently, the gut microbiota and intestinal inflammation are being recognized as potential initiators of metabolic dysfunction that may compound the effects of adipose tissue dysfunction [14]. Gut microbiota and intestinal inflammation will be discussed later in this review.

When adults gain fat mass, they enlarge existing adipocytes via hypertrophy to deal with the increased fat load, to a much greater extent than creating more adipocytes via hyperplasia [15].

When adipocytes approach their maximum size/capacity, which depends heavily on genetic factors and fat depot, they can begin to malfunction; setting off the cascade of events associated with metabolic dysfunction. This malfunction originates from many related pathways. For example, as the size of adipocytes increase, adipocyte stress may increase in part due to hypoxia. Larger adipocytes are typically >100 μm in diameter [16]. Some larger adipocytes may not receive a full supply of oxygen from blood vessels due to oxygen’s limited diffusion distance in tissues (100-

200 μm) [17]. Diet-induced obese mice have been found to have <50% of the oxygenation of adipose tissue of lean mice [18]. Hypoxia will induce several inflammatory pathways and inhibit insulin signaling [18, 19]. This inflammation will contribute to the IR and lipolysis seen in dysfunctional adipose tissue.

Endoplasmic reticulum (ER) stress is another insult to hypertrophic adipocytes. High-fat fed and genetically obese ob/ob mice both show increased adipose tissue ER stress compared to lean controls [20, 21]. ER stress is characterized by the unfolded protein response (UPR). The

UPR occurs when the demand for protein synthesis produces more protein than cellular chaperone folding proteins can manage or when there is damage to the ER, such as disruption of the ER

17 membrane. One inducer of ER stress is high cellular concentration of long-chain free fatty acids.

Saturated fatty acids seem to be more cytotoxic and may disrupt the fluidity of the ER membrane

[22]. This has been shown by using chemical chaperones to increase protein folding to alleviate

ER stress in mice [23]. ER stress will induce apoptosis [24], and in the adipose tissue this will release fatty acids into the interstitial space and circulation.

Systemic oxidative stress is another contributor to adipose tissue dysfunction.

Unsurprisingly, obesity and systemic oxidative stress correlate well [25]. Oxidative stress is increased by a myriad of causes, including hyperleptinemia [26], hyperglycemia [27], and mitochondrial dysfunction [28]. Several of these have direct links to obesity and adipose hypertrophy. Serum leptin concentrations are positively correlated with adipose tissue mass [29].

Hyperleptinemia increases oxidative stress and adipose dysfunction in mice and differentiated mouse pre-adipocytes (3T3-L1) [30] Hyperglycemia can drive metabolic production of reactive oxygen species (ROS) from oxidative phosphorylation in the mitochondria [31]. Insulin resistance reduces glucose transporter 4 (GLUT4)-dependent glucose uptake and subsequently increases hyperglycemia [32, 33]. Hyperlipidemia is associated with ectopic lipid accumulation. Ectopic lipid accumulation will progress to lipotoxicity contributing to IR in non-adipose tissues [34]. It has also been shown that excessive storage of lipids leads to an increase in lipid peroxidation within the adipose tissue itself [35]. Oxidative stress, in general, will lead to inflammation signaling and vice versa, which will drive IR in adipose tissue.

2.2.2. The Role of the Gut in Metabolic Dysfunction

The gut microbiota may contribute to metabolic dysfunction by extracting more energy from undigested food which contributes to weight gain, and by activating the immune system. Germ- free mice are resistant to diet-induced metabolic dysfunction and obesity [36, 37]. However, it is

18 still unclear if gut microbiota from obese mice promote weight gain more than from lean mice, although they do differentially affect IR [38]. High-fat diet has been shown to induce ileal inflammation compared to low-fat diet in conventionally raised mice in as little as 2 weeks. This effect was absent in germ-free mice [39]. Indeed, consumption of HFD can result in inflammation that acutely increases gut permeability [40]. This would lead to a “leaky gut” which allows for more bacterial byproducts to translocate into the intestinal tissue. Chronic consumption of HFD can disrupt the balance of the gut microbiota [41], potentially reducing commensal bacteria, allowing for more pathogens to grow. Gut-derived bacterial byproducts can also contribute to adipose tissue inflammation. The most well studied of these byproducts is lipopolysaccharide

(LPS). Lipopolysaccharide activates Toll-like receptor 4 (TLR4), and in adipocytes increases tumor necrosis factor α (TNF-α) production [42]. Lipopolysaccharide is also taken up by the enterocyte likely via TLR4 and localizes to the Golgi, the site of chylomicron maturation [43].

Large amounts of dietary fat have been shown to increase the concentration of LPS-rich chylomicrons in the lymph of mice in acute experimental conditions [44]. This effect was facilitated by chylomicron production, as pluronic L-81, which inhibits chylomicron secretion, reduced the appearance of LPS in the lymph [44]. Additionally, Gram-negative bacteria are taken up by enterocytes in a manner similar to phagocytosis, through a TLR4-dependent mechanism

[45]. Although the mechanisms have not been fully elucidated, HFD can increase intestinal inflammation and LPS translocation; both of which would increase systemic inflammation and contribute to metabolic dysfunction.

2.2.3. Systemic Consequences of Metabolic Dysfunction

Dysfunction of the adipose has major consequences for other metabolic tissues, such as the liver and skeletal muscle. Local inflammation of the adipose tissue will contribute to the circulating

19 pool of cytokines, including TNF-α, C-C motif chemokine ligand 2/4 (CCL2 and CCL4), interleukin 6 (IL-6), IL-1β, and interferon-γ (IFN-γ) [46]. Adipose tissue inflammation and IR will also increase the rate of lipolysis, or release of FAs [47]. These circulating fatty acids can be stored in other tissues if they are not oxidized and their accumulation contributes to systemic IR. This typically can occur with diacylglycerol (DAG) buildup, which is a well-known activator of protein kinase C [48]. Protein kinase C can phosphorylate the insulin receptor, subsequently disrupting its signaling [48]. Insulin resistance in the adipose tissue will also increase lipolysis further increasing the amount of FA efflux.

Adiponectin is an adipokine, known for its anti-inflammatory effects. Serum adiponectin is reduced in obese patients compared to healthy controls [49]; furthermore, adiponectin is inversely associated with non-alcoholic steatohepatitis (NASH), the inflammatory progression of

NAFLD [50, 51]. Mice pretreated with adiponectin were resistant to chemically-induced acute liver failure, which was dependent on the protective effects of IL-10 [52]. High molecular weight

(HMW) adiponectin is thought to be the more bioactive form and is also found to be decreased in obesity [53]. Adiponectin can upregulate RAW 264.7 macrophage IL-10 production, by initially increasing nuclear factor kappa-B (NFκB)-dependent TNF-α production [54]. This mechanism can dampen LPS sensitivity in macrophages [54]. Treating primary human macrophages with adiponectin inhibited LPS- and TNF-α-induced inflammation independent of IL-10 [55].

Adiponectin also inhibits TNF-α and IL-6 production by LPS-stimulated human macrophages

[56]. Thus, adiponectin, which is reduced in obesity, has anti-inflammatory effects which may be useful in preventing metabolic dysfunction.

2.2.3.1 Effects of Metabolic Dysfunction on the Liver

The liver functions as a hub for lipoprotein transport, bile production, and detoxification. Adipose dysfunction and IR manifests itself as hepatic lipid accumulation which precedes NAFLD and

NASH. One of the oldest hypotheses surrounding NAFLD/NASH is the “two-hit hypothesis” [57].

The “first hit” is hepatic triglyceride (TG) accumulation. The “first hit” is present in 60-90% of obese individuals [58]. During the fed and fasted states, about 60-90% of plasma NEFAs are from the adipose tissue in patients with both NAFLD and IR [59]. Insulin signaling inhibits lipolysis in a healthy individual [60, 61]. Additionally, when insulin sensitivity is impaired, lipoprotein lipase

(LPL) activity in insulin-sensitive tissue is reduced [32, 33]. Delayed metabolism of chylomicrons by the systemic tissues can increase the hepatic uptake of diet-derived TGs [62]. Lastly, if circulating free NEFAs and glucose exceed the metabolic needs of the body, they can accumulate and drive ectopic lipid droplet formation [63]. The excess glucose will be converted to lipid through de novo lipogenesis. Although the classical de novo lipogenesis pathways through sterol response element binding protein (SREBP) are insulin responsive [64]. The mechanistic target of rapamycin (mTOR) is required for insulin-SREBP-1c signaling [65]. However, SREBP-1c expression is increased despite IR, this has been hypothesized as a result of mTOR’s responsiveness to nutrients [66]. Namely, the glucose signaling pathways including carbohydrate response element binding protein (ChREBP) also contributes to de novo synthesis of lipids independent of SREBP [67]. These accumulating lipid droplets and plasma free NEFAs exacerbate hepatic IR through diacylglycerol activating protein kinase ε/θ [68] and saturated fatty acids activating TLR4 [69], worsening the metabolic conditions of the liver. Although simple hepatic steatosis is thought to be relatively benign, it sensitizes the liver to the “second hit.”

The “second hit” exploits the heightened sensitivity of the liver to oxidative damage.

Oxidative stress is typically the initial insult stemming from the production of ROS in response to metabolic dysfunction or inflammation. Healthy hepatocytes will produce metabolic ROS and reactive nitrogen species that will be sufficiently neutralized/quenched with cellular antioxidant defense systems [70]. However, in mitochondrial dysfunction the normal metabolic reactions are perturbed or overwhelmed, leading to an increase in ROS-mediated cellular damage. It has been shown that diet-induced mitochondrial dysfunction precedes NAFLD in fatty rats [71]. Takamura et al. suggested obesity contributes to the leakage of electrons from electron transport chain, which increases metabolic ROS production [72]. Reactive oxygen species drive lipid peroxidation, especially in polyunsaturated fatty acids (PUFAs). Lipid peroxidation also occurs enzymatically by lipoxygenases and cyclooxygenases [73]. Oxidized PUFAs will fragment, producing several oxidized products which create a more pro-oxidative environment [74]. This can contribute to the

UPR causing hepatocyte death and liver damage [75]. Lipid peroxides can induce cell death by either disruption of the membrane or by oxidizing protein and DNA [76]. DNA damage can induce apoptosis through activation of the caspase mechanism [77].

Although the two-hit hypothesis appears to explain the increased prevalence of NASH in obese populations (37%) [78] compared to the general population (13%) [79], only an estimated

3-20% of steatosis cases progress to NASH [80]. Therefore, there must be other underlying factors that drive the progression of steatosis to NASH. Some of these factors are likely associated with metabolic dysfunction associated with obesity. Supporting this viewpoint, there is a population of lean NASH patients, whom are seemingly opposite of the metabolically healthy obese population, in that despite their healthy BMI, they have central obesity and metabolic dysfunction. This population presents with hallmarks of metabolic dysfunction, such as increased visceral adiposity,

22 dyslipidemia, and IR [81]. Additionally, adipocyte size is an independent predictor of hepatic steatosis from obesity [82], suggesting obesity may drive the “first hit” through dysregulation of adipocyte metabolism more so than obesity itself.

Lipotoxicity is the accumulation of toxic lipid species, such as DAG and free cholesterol, both of which are a likely contributor to NAFLD progression. Overexpression of DAG acyltransferase 1 and 2 (DGAT1/2) in mice fed HFD, which led to a decrease hepatic DAG, demonstrated that hepatic TG accumulation alone was insufficient to induce inflammation and IR in this model [83]. Triglyceride production may actually be preventative in mouse models of

NASH because TGs are biologically inert compared to DAG. Deleting stearoyl-CoA desaturase 1

(Scd1) protected mice from hepatic steatosis, by inhibiting TG synthesis. However, these mice had exacerbated apoptosis and inflammation in liver when they were fed a methionine choline- deficient diet [84]. A high-sucrose, low-fat diet fed to Scd1-/- mice doubled hepatic free cholesterol and increased serum hepatic enzymes compared to wild-type (WT) mice [85]. High dietary cholesterol is a common tool to induce NASH in animal models [86]. Genetically hypercholesteremic mice fed Western-type diets (high-fat, high-cholesterol) are also commonly used as animal models of NASH [87]. These models may develop NASH, in part, via elevations in hepatic free cholesterol, which is known to activate inflammasomes [88] and induce ER stress

[89]. There are several other factors that may contribute to NASH progression, including decreased adiponectin, increased leptin, and genetic predisposition.

2.2.3.2. Effects on Skeletal Muscle

In addition to playing an essential role in mobility, skeletal muscle also functions as both an endocrine organ and a major site for macronutrient disposal [90]. Adipose tissue dysfunction

23 drastically affects skeletal muscle metabolism due to its insulin-sensitive nature, albeit typically to a lesser extent than in the liver. The causes of skeletal muscle dysfunction are similar to liver dysfunction in that they share the accumulation of ectopic lipid [91] and mitochondrial dysfunction

[92]. The accumulation of lipids in the muscle is associated with IR in this tissue [93, 94]. Insulin resistance within the muscle will decrease glucose uptake, contributing to hyperglycemia.

Mitochondrial dysfunction is associated with increased ROS production [95]. This can lead to damage of the mitochondria that, if not repaired, could lead to decreased mitochondrial density through autophagy [96]. The most well-studied myokine is IL-6. While IL-6 is classically thought to be proinflammatory, IL-6 can improve skeletal muscle glucose uptake and improve whole body insulin sensitivity [97]. Obesity’s effects on IL-6 production in the skeletal muscle remain controversial, as some studies report increases [98] and others decreases [99]; although, this could be due to differences in the study populations used for tissue biopsies. Decreased myocyte IL-6 was found in diabetics, while increases were found in healthy individuals in response to conditioned medium intended to mimic obese conditions. Skeletal muscle can also secrete pro- inflammatory cytokines such as TNF-α and CCL2 [99, 100] which, similar to adipose tissue and liver, will contribute to IR. TNF-α and CCL2 are increased in the skeletal muscle with the occurrence of obesity and diabetes [99, 100]. Ultimately, obesity disrupts the skeletal muscle’s capacity to catabolize nutrients, leaving more for the non-adipose tissues to handle, and increasing

IR. The inflammatory cytokines secreted by skeletal muscle may also contribute to systemic IR.

Figure 2.1 functions as a summary of the systemic interactions that are important for metabolic dysfunction.

Figure 2.1. Schematic summary of metabolic dysfunction

2.3. Molecular Mechanisms Governing Lipid Metabolism

2.3.1. Insulin

Insulin is known for its role in modulating blood glucose levels. However, it also controls several factors crucial for lipid synthesis and storage. As reviewed above, insulin alters skeletal muscle and adipose tissue metabolism. Several tissues are insulin sensitive, such as liver, skeletal muscle, and adipose tissue [101]. Insulin signaling in the liver has major implications for both carbohydrate and lipid homeostasis. In terms of carbohydrate metabolism, insulin induces GLUT4 translocation to the plasma membranes of skeletal muscle and adipose tissue. This allows for maximal glucose uptake in these tissues [102]. Insulin will also inhibit the production of glucose in the liver, derived from both gluconeogenesis and glycogenolysis. This is substantial because the liver is the major site of gluconeogenesis in the body [103]. Insulin signaling will inhibit gluconeogenesis through forkhead transcription factor 1 (FoxO1) [104], which is a major regulator of gluconeogenic gene expression. Hepatic insulin signaling also stimulates glycogen synthesis [105]. In terms of lipid metabolism, insulin stimulates many of the de novo lipogenesis genes including fatty acid synthase

(Fas) and acetyl-CoA carboxylase (Acc) [106]. Fatty acid synthesis in healthy individuals usually accounts for about 5% of FAs in the serum but can climb to 24% in the post-absorptive state [107].

In a healthy liver, insulin affects lipid metabolism in two ways: it will 1) upregulate de novo lipogenesis and TG synthesis; and 2) inhibit very low-density lipoprotein (VLDL) secretion [108].

This will increase the temporary storage of hepatic TGs that will be released after the post-prandial state.

Factors outside of local insulin signaling will have effects on local insulin responsive pathways. For example, hyperglycemia and hyperinsulinemia can suppress gluconeogenesis and glycogenolysis independent of hepatic insulin signaling [109]. Glucagon is known to upregulate

26 both hepatic gluconeogenesis and glycogenolysis [110]. Insulin signaling in pancreatic α-cells will suppress glucagon production [111]. Another consideration is insulin’s effects on gluconeogenic substrate availability. Triglyceride hydrolysis from adipose tissue and skeletal muscle proteolysis can supply glycerol and amino acids, respectively. Insulin will inhibit both of these actions, and thus limits the availability of these substrates critical for gluconeogenesis [112]. It is important to note that when examining mice with a liver-specific insulin receptor knockout, insulin was unable to suppress gluconeogenesis during a euglycemic clamp [113]. Thus, direct liver insulin signaling must remain intact to suppress gluconeogenesis. This highlights the importance of considering a whole-body system when examining insulin’s effects.

2.3.2. Peroxisome Proliferator-Activated Receptors

Regulation of lipogenesis is controlled through multiple transcription factors. Peroxisome proliferator-activated receptors (PPARs) are major factors in the regulation of lipid metabolism.

To date, three isoforms have been identified: PPARγ, PPARα, and PPARβ/δ [114]. All of these isoforms are ligand-binding nuclear receptors that respond to various natural and synthetic molecules. Of these isoforms, PPARγ and PPARα are the most well-studied and control opposing mechanisms in lipid metabolism. PPARα is a controller of genes involved in lipolysis and fatty acid oxidation. PPARα is most strongly expressed in the heart, kidney, small intestine, liver, and skeletal muscle [115]. PPARα regulates the expression of carnitine palmitoyltransferase 1 α/β

(Cpt1α/β), acyl-coenzyme A oxidase 1 (Acox1), and acyl-coenzyme A dehydrogenase (Acad1), all of which are key enzymes in fatty acid oxidation [116].

PPARγ controls genes encoding crucial enzymes in TG synthesis and lipid storage.

Adipose tissue expresses the highest levels of PPARγ; however, PPARγ can be upregulated in the liver during hepatic steatosis. In the adipose tissue, PPARγ is vital for adipocyte development and

27 alongside CCAAT/enhancer-binding protein α (C/EBPα) controls adipocyte differentiation [117].

Knocking down hepatic PPARγ has been shown to reduce de novo lipogenesis and TG synthesis

[118]. In obese patients, PPARγ mRNA is significantly upregulated [119]. In obese mice, the mRNA of a splice variant, PPARγ2, is strongly upregulated in response to HFD, suggesting it is more nutritionally-responsive than PPARγ1 [120]. Some PPARγ target genes include: Scd1, cluster of differentiation 36 (Cd36), fatty acid binding protein 4 (Fabp4) [121], lipoprotein lipase

(Lpl) [122], and glucose transporter 4 (Scl2a4) [123]. Thus, PPARγ plays a vital role in TG accumulation within tissues. It may be hypothesized that PPARγ is responsible for hepatic steatosis because its positive association with steatosis and lipogenic genes [124]. However, PPARγ agonists such as thiazolidinediones decrease the accumulation of hepatic lipids in type II diabetes patients likely through improving adipose tissue insulin sensitivity [125-127]. This suggests that while hepatic PPARγ is nutritionally-responsive and contributes to hepatic steatosis, other transcription factors and adipose insulin resistance could be important contributors to hepatic steatosis associated with obesity.

2.3.3. Liver X Receptor

Liver X receptor (LXR) is a major nuclear receptor controlling lipid metabolism [128]. The natural ligands for LXR are the oxysterols, which are produced when cellular cholesterol levels are high

[129]. LXR is partially responsible for maintaining cellular cholesterol balance. Two isoforms of this transcription factor exist, LXRα/β. LXRα is more prominent in major metabolic tissues such as the intestines, liver, and adipose [130], while LXRβ is ubiquitously expressed and at its highest in the brain [130]. LXRα target genes include inducible degrader of the low-density lipoprotein receptor (Idol) [131], proprotein convertase subtilisin/kexin type 9 (Pcsk9), Niemann-pick C1 like-

1 (Npc1l1), and the ATP-binding cassette transporter sub family (Abca1, Abcg1, Abcg5, Abcg8).

Pcsk9 [132] and Idol [133] will both impair the recycling and induce the degradation of low- density lipoprotein receptor (LDLr), decreasing its presence on the cellular membrane, and ultimately reducing LDL uptake. Npc1l1 is required for cholesterol absorption in the intestine and will be increased when intestinal cholesterol levels are lowered [134]. Abcg5 functions in a heterodimer with Abcg8 and will efflux free cholesterol back into the intestinal lumen [135].

Overall, LXRα functions to reduce cellular cholesterol levels.

2.3.4. Sterol Response Element Binding Proteins

Three major SREBP isoforms exist, -1a, -1c, and -2. Classically, nascent SREBP is bound to

SREBP cleavage-activating protein (SCAP) and insulin-induced gene product (INSIG) [136].

When cellular cholesterol levels are high, SREBP is retained within the endoplasmic reticulum and effectively inactive [137]. When the ER membrane is depleted of cholesterol, SCAP and

INSIG dissociate, releasing SREBP from the ER and it is shuttled to the Golgi apparatus where it is further matured by enzymatic cleavage [137]. Insulin can stimulate SREBP-1c protein expression through LXR activity [138, 139]. However, LXR does not seem to strongly induce

SREBP-1c maturation [140]. Insulin also downregulates the expression of INSIG posttranslationally, which has been shown to increase SREBP-1c maturation [136]. SREBP-1c contributes to de novo fatty acid synthesis by controlling the gene expression of several related enzymes. Some of its target genes are Fas, Scd1, and Acc [141]. SREBP-2 stimulates cholesterol biosynthesis upon activation. Most notably, SREBP-2 positively regulates the transcription of 3- hydroxy-3-methyl-glutaryl-coenzyme A reductase (Hmgcr) [142] and Ldlr [143]. The rate- limiting enzyme in cholesterol biosynthesis is Hmgcr. Additionally, Ldlr is an apolipoprotein (apo)

B-100 and apoE-recognizing receptor that is responsible for uptake of LDL, a cholesterol-rich

29 lipoprotein, into the liver. Thus, the SREBPs help increase de novo lipogenesis when cellular cholesterol is depleted.

The network of transcriptional regulation for lipid metabolism in the liver, adipose, and skeletal muscle is diverse and extensive. The focus of this review was to orient the reader to some of the key transcription factors and target genes used in the following aims to evaluate the metabolic phenotype of mice. Although many pharmaceuticals have targeted the numerous pathways mentioned above, they are not without negative consequences. One such pathway is at the level of intestinal lipid absorption. Two drugs commonly used to inhibit lipid absorption are orlistat and ezetimibe. Ezetimibe inhibits intestinal cholesterol absorption leading to lower plasma low-density lipoprotein cholesterol (LDL-C) and CVD risk [144]. In obese patients with IR, therapies targeting cholesterol absorption (e.g., ezetimibe) are effective in treating dyslipidemia

[145, 146], reducing visceral fat mass [146], and systemic inflammation [145]. Orlistat inhibits lipid absorption by disrupting lipase activity. This would inhibit fatty acid uptake by not allowing

TGs to be hydrolyzed. Orlistat has been found to be an effective weight loss drug and improves the metabolic outcomes of patients [147]. However, as with most drugs, undesirable side effects may occur, including liver damage (e.g., ezetimibe, orlistat) [148, 149], gastrointestinal distress

(e.g., orlistat) [150], and potentially cancer-promoting effects (e.g., orlistat) [151]. Therefore, a need exists for alternative or complementary therapies targeting lipid absorption.

2.4. Dietary Sphingomyelin for the Prevention of Metabolic Dysfunction

2.4.1. Structure of Sphingomyelin

Sphingomyelin (SM) is a dietary lipid found in eggs, dairy, and meat. Sphingomyelin is considered a sphingolipid. Despite their immense structural diversity, all sphingolipids are grouped together by their amino alcohol backbone [152]. Although >60 sphingoid backbones exist, the most

30 common in mammalian tissues is sphingosine ((2S,3R,4E)-2-aminooctadec-4-ene-1,3-diol) [153].

Some of the most common variations include sphinganine, also known as dihydrosphingosine

((2S,3R-2-aminooctadecane-1,3-diol), which lacks the 4C double bond and phytosphingosine

((2S,3S,4R)-2-aminooctadecane-1,3,4-triol), which replaces the 4C double bond of sphingosine with an additional hydroxyl group [153, 154]. Shorthand nomenclature for sphingoid backbones are similar to fatty acids, in that they follow the form C:D, where C is the number of carbons and

D is the number of double bonds. The distinction between fatty acids and sphingoid bases is the inclusion of a “d” or “t” to indicate the number of hydroxyl groups. For instance, sphingosine can be written as d18:1, sphinganine is d18:0, while phytosphingosine is t18:0 [155]. Ceramide is generated with the addition of an N-acylated fatty amide. In general, SM is a ceramide with phosphodiester bound choline. For a comparison of the structures of sphingoid backbones, ceramide, and SM refer to Figure 2.2.

Figure 2.2. Structure of common bases, ceramide, and SM

2.4.2. Sphingolipids in the Typical Diet

On average, a Western diet contains 200-400 mg/day of sphingolipids [156]. Most dietary sphingolipids are supplied by animal products, such as meat, eggs, and dairy. Sphingomyelin is

31 the predominant sphingolipid in mammalian cells, and therefore, most dietary sphingolipids are

SM. Table 2.1 contains a list of some of the most common dietary sphingolipids.

It is inaccurate to think of dietary SM as a single molecular species. In fact, SM is an umbrella term for a family of related lipid molecules containing a sphingoid backbone, fatty amide, and phosphorylcholine group. For comparison, both milk-derived SM and egg-derived SM vary greatly from each other. In terms of the N-acyl linked fatty amide, milk-derived SM species have a more varied fatty amide composition (16:0-24:0) compared to egg-derived SM species, which contain primarily palmitate. Similarly, milk-derived SM species have sphingoid backbones ranging from d16:0-d19:0 in contrast to egg-derived SM species, which contain almost exclusively sphingosine [157, 158]. Figure 2.3 illustrates the comparison of milk and egg SM structures from a commercial supplier (Avanti Polar Lipids). This difference in sphingoid backbones means milk

SM contains higher levels of dihydroSM [158]. DihydroSM binds more tightly to cholesterol than

SM in an in vitro model membrane [159]. Sphingomyelin is a potential bioactive component of the milk fat globular membrane (MFGM) and milk polar lipid (MPL) fractions, both of which have shown beneficial effects [160]. Studies containing these provide insight into possible whole-food sources of dietary sphingolipids, but the bioactive effects of other phospholipids cannot be ruled out [160, 161].

Sphingomyelin hydrolysis begins in the intestinal tract by the action of alkaline SMase

(alk-SMase) [168]. Ceramide will then be hydrolyzed further to sphingosine and a FA by neutral ceramidase [169]. Sphingosine itself is readily absorbed [170]. Sphingomyelin digestion is delayed by other dietary lipids through the inhibition of alk-SMase activity [171]. Alk-SMase activity peaks around the middle of the intestines after most

32 most other lipids have been absorbed. Due to the delayed activation of alk-SMase, around 20% of intact dietary SM reaches the ileum in humans [172]. Once absorbed, sphingosine has two metabolic fates. It can be resynthesized into ceramide or, through a two-step process, be converted to hexadecane and ethanolamine after conversion to sphingosine-1-phosphate. The majority of intestinal sphingosine follows the latter path [169].

Table 2.1. Common dietary sphingolipids Sphingolipid Dietary Sources Content (mg/100g) Ref Bovine Milk, Whole 9 Beef 44-69 Egg 82 Sphingomyelin [162] Cottage Cheese 139 Mackerel 224 Chicken Liver 291 Rice Bran 5.6 [163] Ceramide Soybean 11.5 [164] Corn 11.5 [163] Rice Bran 11.5 [164]

Soybean 20 [165] Cerebroside 310 [166] Amaranth Grain 39 [165] Bovine Milk, Whole 0-11 Mackerel 6.48 Ganglioside [167] Chicken Egg Yolk 15.9 Chicken Liver 29 200-400 mg Total Sphingolipids All Sources [156] consumed/day

2.4.3. Dietary Sphingomyelin and Lipid Absorption

Dietary SM dose-dependently reduces intestinal absorption of cholesterol, TG, and long-chain FAs in rodents [173-177]. In vitro experiments demonstrated that the products of SM digestion, ceramide and sphingosine, inhibit both cholesterol and FA uptake [178-180]. The inhibitory effects of SM are seemingly dependent on the source. When comparing egg SM to milk SM, milk SM

33 inhibited the intestinal absorption of fat and cholesterol more so than egg SM in dual-cannulated rats [176]. This is likely due to the structural differences as mentioned above (see Figure 2.3).

Figure 2.3. Compositional differences between egg- and milk-derived SM

A) Fatty Amide Composition of Egg- and Milk-derived SM B) Sphingoid Backbone of Egg- and Milk- Derived SM

2.4.3.1 Animal Studies and Serum Lipids

Due to the scarcity of studies that fed purified SM and examined serum lipids, we will review the studies feeding any purified sphingolipid or sphingolipid-rich extracts. In several rodent feeding studies, dietary sphingolipids or sphingolipid-rich fractions have shown mixed effects on serum lipids. Most studies reported reductions [177, 181-185], however, others observed no effect [185-

188], and some found increases [181, 189]. Duivenvoorden et al. fed hyperlipidemic

APOE*3Leiden mice a Western-type diet (15% cocoa butter, 0.25% cholesterol by weight) supplemented with various sphingolipids, ranging from SM to sphingosine [177]. At 0.2% and

0.4% (w/w of the diet) all dietary sphingolipids caused more than a 20% reduction of plasma cholesterol and TG [177]. Zucker-fatty rats fed chicken-derived SM or corn-derived

34 glycoceramides showed reduced plasma cholesterol after 4 weeks compared to control [183]. In contrast, the addition of egg SM to a milkfat-containing high-cholesterol diet fed to C57BL/6J mice had no effect on plasma lipids [186]. Apolipoprotein E-/- mice fed a HFD with and without egg SM (1.2% wt:wt) demonstrated no differences in plasma lipids and no effect on atherosclerosis development [188]. However, low-fat fed apoE-/- mice demonstrated a decrease in en face aortic plaques with 1.2% (wt:wt) egg SM supplementation [188].

Mixed fractions of polar lipids containing SM also show equivocal results. When feeding

SM as part of a milk-derived polar lipid extract to mice fed a HFD (21% w:w butter fat, 0.15% cholesterol), serum cholesterol and hepatic lipids were significantly reduced [184]. Similarly, MPL fractions, fed to genetically obese AAy mice, reduced plasma cholesterol after only 4 weeks [182].

Conversely, a milk phospholipid concentrate (1.2% phospholipid w/w of diet, containing ~ 0.25% wt:wt of the diet SM) added to a HFD (20% fat by weight) had no effect on plasma lipids over 8 weeks [187]. Plasma cholesterol, plasma SM, and aortic plaques were increased by a modified

AIN-76A diet with 1% (w/w) ox brain sphingolipids (28% SM) fed to female LDLr-/- mice for 12 weeks [189]. In the context of HFD or obesity, dietary sphingolipids generally reduce plasma cholesterol or hepatic lipids, which will be discussed below. Only in genetically hypercholesterolemic mice on a chow diet were elevations in plasma cholesterol reported.

Therefore, there is sufficient evidence to suggest dietary SM has hypolipidemic effects when chronically fed to rodents.

2.4.3.2 Clinical Trials and Serum Lipid Outcomes

Unlike rodents, human bile contains alk-SMase, so the relative efficiency of SM digestion is likely increased [172]. There is currently one study that supplemented humans with purified SM [190].

Ramprasath et al. found that 1 g/day of purified milk SM supplementation over 2 weeks of prepared meals raised high-density lipoprotein cholesterol (HDL-C) but did not affect non-HDL-

C or cholesterol absorption in healthy adults [190]. There are studies examining the effects of other sphingolipids or sphingolipid-rich extracts on clinical lipid profiles. These may provide insight into potential effects of feeding sphingolipids to humans. A double-blind, placebo-controlled crossover trial supplying 1 g/day of phytosphingosine for 4 weeks to adult men with metabolic syndrome reported a significant reduction in plasma LDL-C [191]. Rosqvist et al. supplemented overweight adults with 40 g of milkfat/day as whipping cream in a single-blind, randomized, controlled, isocaloric parallel study for 8 weeks. Consuming 40 g of milkfat/day as whipping cream significantly lowered plasma LDL-C, non-HDL-C, and apoB:apoA-I ratio compared to the same amount of milkfat as butter oil [192]. Compared to butter oil, whipping cream is rich in milkfat globular membranes, the polar lipid rich fraction in milk [192]. Two clinical trials supplementing buttermilk for 4 weeks found different results. Ohlsson et al. observed no effect of

4-week supplementation of a buttermilk drink enriched in MFGM, supplying 700 mg milk SM/day when measuring plasma lipids, in a placebo-controlled parallel study with healthy adults. There was a trend for a reduction in plasma LDL-C (p = 0.056) in women who drank the treatment drink

[193]. This study was relatively small with 18 participants, and the parallel design does not help the statistical power. In contrast, 4-week ingestion of 45 g/day of buttermilk resulted in significant reductions in both plasma cholesterol and TG compared to a calorically-matched placebo [194].

This study was a double-blind randomized cross-over study with healthy adults and indirectly measured cholesterol absorption through measuring serum β-sitosterol concentration. Serum β- sitosterol concentration was correlated with the reductions in plasma LDL-C, suggesting a reduction in cholesterol absorption was responsible [194]. This study was much more robust with

38 participants, and the crossover design improved statistical power. In aggregate, there is some clinical data supporting the hypolipidemic effects found in animals. Due to the scarcity of studies and the sample sizes used (see Table 2.2), there is a clear need for further clinical trials to confirm the effects. It is worth noting that rodent studies typically use larger doses of sphingolipids, even when correcting for body surface area, therefore the smaller effect sizes in clinical studies are expected.

2.4.4. Dietary Sphingolipids and Other Markers of Metabolic Dysfunction

As reviewed above, obesity can lead to IR and hepatic lipid accumulation. Although no studies at this point have examined dietary SM for its effects on IR, it is important to recognize that phytosphingosine (t18:0) has shown potential to reduce IR in both rodents [195] and humans [191].

Feeding a HFD supplemented with 0.2% (w/w) phytosphingosine to mice for 24 weeks reduced the expression of a macrophage marker (F4/80) in adipose tissue and improved the response to a glucose tolerance test [195]. A clinical trial examining the daily intake of phytosphingosine (1 g/day) reported a significant reduction in fasting glucose (-5%) along with an improved index of insulin sensitivity (+10%) in men with MetS and with only minor gastrointestinal distress [191].

A substantial amount of research that examines dietary sphingolipids for the prevention of hepatic lipid accumulation, the hepatic manifestation of IR, show an inhibitory effect towards steatosis (see Table 2.3). Supplementing Western-type (21% butter fat, 0.15% cholesterol by weight) diet-fed mice with egg SM at 0.6% and 1.2% (w/w) for 4 weeks reduced hepatic total lipid content by 33% and 40%, respectively [186]. These mice also demonstrated reduced in hepatic total cholesterol content. Meanwhile, mRNA expression of genes involved in fatty acid synthesis increased in the 0.6% egg SM-fed mice [186]. Male Zucker fatty rats fed an AIN-76A diet with

Table 2.2. Clinical trials evaluating hypolipidemic effects of dietary sphingolipids Author Year Population Study Design Treatment and Duration Results

4 weeks of 2 drinks/day totaling 975 ↔ Plasma lipids between groups Ohlsson Healthy adults mg SL (700 mg SM, 180 mg GC, 95 2009 Parallel, placebo trend for ↓ plasma LDL-C (p = 0.056) in [193] (n = 48) mg GS) of test drink or 119 mg total SL women compared to placebo in isocaloric placebo drink

Men with ↓ Serum cholesterol (-5.6%), Snel Double-blind, 4 weeks of phytosphingosine (1 g/day) 2010 MetS ↓ LDL-C (-8.8%), ↔ TG, ↔ HDL-C [191] placebo, crossover or placebo in capsule form (n = 12) compared to placebo

↓ Plasma cholesterol (-3.1%), 4 weeks of buttermilk powder (187.5 ↓ TG (-10.7%), Conway Healthy adults Double-blind, mg total PL, 23.6 mg SM per day) or trended for ↓ LDL-C (p = 0.057) compared 2013 [194] (n = 34) placebo, crossover placebo (34.6 mg total PL) matched for to placebo, calories and minerals ↓ LDL-C (-5.6%) in participants with highest (top 50%) baseline LDL-C

2 weeks of prepared diets (30% kcal ↑ HDL-C (+13%), Ramprasath Healthy adults Controlled diets, 2013 from fat, 240 mg cholesterol/day) with ↔ total cholesterol, [190] (n = 10) crossover or without 1 g/day added milk SM ↔ non-HDL-C, ↔ TG with milk SM

8 weeks of intervention meals with either 40 g milkfat/day as whipping ↓ Plasma cholesterol, ↓ LDL-C, Overweight Rosqvist Single-blind, cream (19.8 mg total PL) or control diet ↓ non-HDL-C, 2015 adults [192] parallel with butter oil (1.3 mg total PL), ↓ apoB:apoA-I ratio with whipping cream (n = 57) matched for calories, macronutrients, vs. control group and calcium

1ApoA-I, apolipoprotein A-I; ApoB, apolipoprotein B; GC, glucosylceramide; GS, gangliosides; MetS, metabolic syndrome; PL, phospholipid; SL, sphingolipids; SM, sphingomyelin; TG, triglyceride; ↑, increase; ↓ decrease; ↔, no change.

39 and without either 0.5% (w/w) SM prepared from chicken skin, or 0.5% (w/w) glucosylceramide from corn had significantly lower hepatic lipid content on both treatment diets compared to control

[183]. Additionally, these rats had reduced fasting plasma insulin as well as lower hepatic Scd1 and Acc mRNA in the treatment groups. Conversely, hepatic Pparα mRNA expression was increased in the SM and glucosylceramide-fed rats, suggesting greater fatty acid oxidation [183].

Chronic supplementation of mice fed a Western-type diet (21% butter fat, 0.15% cholesterol wt:wt) with milk phospholipid extracts (0.25%-0.35% SM w/w of diet) significantly lowered both hepatic cholesterol and TG [184, 196]. Milk-derived ceramides (0.35% w/w of diet) given to genetically obese KK-Ay mice for 4 weeks also reduced hepatic cholesterol and TG [182].

Mechanistically, mice fed a milk SM-enriched HFD (0.35% SM w/w, 28% fat by weight) tended

(p = 0.058) to have lower fractional synthesis of hepatic TG and palmitic acid incorporation (p =

0.062). These mice had a significant reduction in the ratio of hepatic palmitoleic acid (C16:1n7) to palmitic acid (C16:0), an index of SCD1 activity [197]. In addition, in vitro studies that applied very-long chain ceramides (C22:0- & C24:0-ceramides) to primary mouse hepatocytes and Huh7 hepatoma cells reduced the expression of both Pparg and Cd36 [198, 199]. This reduction in lipogenic genes may explain the effect found in vivo [181, 182, 185]. Duivenvoorden et al. found when female APOE*3Leiden mice were fed a Western-type diet (15% cocoa butter, 0.25% cholesterol by weight) supplemented with 1% phytosphingosine (w/w of the diet) for 5 weeks, their hepatic TG and cholesterol content, as well as plasma alanine transaminase and serum amyloid A were significantly reduced. This demonstrates a less steatotic liver that is potentially less inflamed. Interestingly, hepatic mRNA expression indicated lipid biosynthesis was upregulated compared to control, suggesting some compensatory mechanism [177].

Structurally distinct ceramides seemingly have different biological functions [200]. Thus, the distribution of ceramide species could be important for normal tissue function. Notably, the ceramide profile of rodent livers is associated with hepatic function. As an example, primary hepatocytes from mice lacking ceramide synthase 2, an enzyme required for C22- and C24- ceramide synthesis, have reduced insulin sensitivity [198, 201, 202]. C16-ceramide synthesis increases in these mice [202], C16:0-ceramide is also associated with deleterious effects on hepatocyte function [203, 204]. Therefore, a balance may be required between different length ceramides. Diet-induced obesity causes an increased hepatic C16:0-ceramide concentration and ceramide synthase 5 and 6 expression, the major enzymes responsible for C16:0-ceramide synthesis [204, 205]. Thus, a balance between the C22-/C24-ceramides and C16-ceramide may be important for proper liver function. Consequently, the effects of dietary sphingolipids on hepatic sphingolipid profiles could be an important endpoint. Unfortunately, this is largely unknown. One study reported no change in total ceramide in the livers of mice supplemented with 0.6% egg SM

(wt:wt), although C16:0-ceramide was reduced and C22- and C24- ceramides were unaffected

[186]. This was reported with an overall improved hepatic phenotype.

In summary, dietary SM has shown promise to reduce the occurrence of dysfunctional lipid metabolism in mice. Although there are few clinical trials in the field, they show either beneficial effects of dietary sphingolipids or may have suffered from a lack of statistical power. Therefore, more research is needed to fully determine the efficacy of dietary SM for improvements in lipid metabolism, especially in humans. Additionally, absorption studies suggest a functional difference between SM derived from egg and milk due to their structural differences. This coupled with a lack of studies comparing the two, leaves an obvious gap in the literature to directly compare the hypolipidemic effects of dietary SM from different sources.

Table 2.3. Dietary sphingolipids and hepatic steatosis Author Year Animal Model Diet and Duration Hepatic Lipids Results

Female ApoE*3 5 weeks of Western-type diet (15% cocoa Duivenvoorden PS: ↓ TG (-56%), ↓ TC (-61%), ↓ FC 2006 Leiden mice (n = butter; 0.25% cholesterol) supplemented with [177] (-11%) 6) 0.1% (w/w) phytosphingosine (PS)

8 weeks of either LFD or Western-type diet PLRDME with LFD: ↔ total lipid, Male C57BL/6 (21% milkfat; 0.15% cholesterol) with and ↔ TG, ↔TC, ↔PL PLRDME with Wat 2009 mice without 1.2% PL (w/w) from phospholipid-rich Western-type diet: ↓ total lipid (- [184] (n = 10) dairy milk extract (PLRDME) (0.26% of diet by 33%), ↓ TG (-44%), ↓ TC (-48%), ↓ weight as SM) PL (-16%)

8 weeks of Western-type diet (21% milkfat; PLRDME (0.26% SM w/w): ↓ total 0.15% cholesterol) with and without 1.2% PL Male C57BL/6 lipid (-41%), ↓ TG (-47%), ↓ TC (- Kamili (w/w) from phospholipid-rich dairy milk extract 2010 mice 39%)PC-700 (0.34% SM w/w): ↓ [196] (PLRDME) (0.26% SM w/w) or 0.2% (w/w) (n = 10) total lipid (-45%), ↓ TG (-63%), ↓ milk phospholipid concentrate (PC-700) (0.34% TC (-57%) SM w/w)

45 days of AIN-93 diet supplemented with 0.5% Yunoki Male Zucker rats SM: ↓ total lipid (-24%), ↔ TC GC: 2010 (w/w) of chicken skin SM or 0.5% (w/w) corn [183] (n = 6) ↓ total lipid (-18%), ↔ TC GC

4 weeks of AIN-93G diet with1.7% (w/w) of lipid-concentrated butter serum (LC-BS) (0.13% Female KK-Ay SM, 0.18% ceramides by weight in diet) or LC-BS: ↓ TG (-27%), ↔ TCCer-fr: Watanabe 2011 mice 0.5% (w/w) ceramide-rich fraction (Cer-fr) ↓ TG (-38%), ↓TC (-47%) SM-fr: ↔ [182] (n = 7) (0.34% ceramide by weight in diet) or 0.5% TG, ↔ TC (w/w) SM-rich fraction (SM-fr) (0.25% SM by weight in diet)

0.3% SM: ↓ total lipid (-20%), ↓ TC Male C57BL/6 4 weeks on a Western-type diet (21% milkfat; (- 36%), ↓ TG (-31%)0.6% SM: ↓ Chung 2013 mice 0.15% cholesterol) with 0.3, 0.6, 1.2% (w/w) total lipid (-33%), ↓ TC (- 48%), ↓ [186] (n = 10) egg SM TG (-46%)1.2% SM: ↓ total lipid (- 40%), ↓ TC (- 67%), ↓ TG (-56%)

Cer-fr, ceramide-rich fraction; FC, free cholesterol; GC, glucosylceramides; HFD, high-fat diet; LC-BS, lipid-concentrated butter serum; LFD, low-fat diet; PL, phospholipids; PLRDME, phospholipid-rich dairy milk extract; PS, phytosphingosine; SL, sphingolipids; SM, sphingomyelin; SM-fr, sphingomyelin-rich fraction; TC, total cholesterol; TG, triglycerides; ↑, increase; ↓, decrease; ↔, no change.

2.5. Other potential bioactive roles of dietary SM

In addition to the reported hypolipidemic and insulin-sensitizing effects, dietary SM may have other bioactivities. Thus, it may confer additional benefits compared to the hypolipidemic drugs orlistat and ezetimibe. Two of these effects include modulation of the gut microbiota and attenuation of inflammatory signalling.

2.5.1. Anti-inflammatory Effects of Dietary Sphingolipids In Vitro

Several cell studies utilizing exogenous sphingolipids have shown these lipids to be anti- inflammatory. J774 immortalized cells and thioglycollate-elicited peritoneal macrophages had lowered LPS-induced TNF-α and macrophage inflammatory protein 2 (MIP-2) secretion when pretreated with either 10 µM C8-ceramide or sphingosine treatment [206]. Aiming to elucidate the mechanism, this study proceeded to explore the potential for the synthesis of sphingosine-1- phosphate, a sphingolipid metabolite, to be anti-inflammatory. Supporting this hypothesis, sphingosine-1-phosphate was found to be a PPARγ ligand [207]. However, sphingosine-1- phosphate synthesis was found to be dispensable, as a sphingosine kinase inhibitor had no effect on ceramide and sphingosine’s effects on LPS-induced TNF-α production [206]. Exogenous sphingosine given to RAW 264.7 macrophages transfected with a PPARγ luciferase plasmid acted as a modest agonist of PPARγ [208]. PPARγ binding can repress NF-κB activity [209], giving a putative mechanism for the inhibition of inflammatory gene expression. Bovine brain SM was shown to inhibit LPS-stimulation of granulocytes [210]. C8-ceramide (10 µg/mL) inhibited LPS- stimulated secretion of IL-10 and IL-5 from bone marrow-derived mast cells, although ceramide did not alter TNF-α, IL-13, or IL-6 [211]. Additionally, thioglycollate-elicited peritoneal macrophages treated with C8-ceramide had a reduced production of IL-6, IL-12p40, and TNF-α in response to LPS. The difference between macrophage and mast cell responses were attributed

44 to variations in PI3K-Akt and mitogen-activated protein kinase (MAPK) phosphorylation between the two cell types [211]. Thus, exogenous sphingolipids may not affect all cells similarly. Treating

RAW 264.7 macrophages with 10-50 μM of C2-ceramide inhibited both the activation of NF-κB and adapter protein-1 by LPS, while reducing the production of nitric oxide and prostaglandin

[212].

Other complex sphingolipids and sphingolipid-rich dairy fractions have shown anti- inflammatory effects. Gangliosides administered ex vivo to infant bowel epithelial cells, taken from viable infants, inhibited LPS- and hypoxia-induced inflammation, measured by nitrite release, IL-8, IL-6, IL-1β, eicosanoid, and hydrogen peroxide production [213]. Most studies used

LPS as the inflammatory stimuli, suggesting there may be a lipid-lipid interaction with the sphingolipids as the potential mechanism. Interestingly, sphingolipid-rich dairy fractions inhibited monosodium urate-induced IL-8 mRNA and protein expression in THP-1 monocytes. Although, in this study the dairy fraction had no effect on cytokine production in response to LPS [214].

Some in vivo observations support sphingolipids for inhibiting LPS activity. For example, sphingolipid-containing lipoproteins neutralize LPS activity by forming phospholipid:LPS complexes [210, 215]. After intraperitoneal (IP) injection of LPS, the liver increases sphingolipid production, enriching circulating lipoproteins [216], possibly in an effort to detoxify LPS.

Sphingosine is known to potently inhibit protein kinase C [217], an enzyme that drives inflammatory signalling in macrophages [218].

In summary, these studies illustrate mechanistic evidence for exogenous sphingolipids and its derivatives to have inhibitory effects on inflammation. Due to the dynamic nature of sphingolipid metabolism, at this time it is impossible to pinpoint which metabolite is responsible.

If the catabolic metabolites are the causal factor, the effects of glycosphingolipids could be applied to SM.

In contrast to the above studies, some experiments demonstrate pro-inflammatory effects of exogenous sphingolipids. Treating thioglycollate-elicited peritoneal macrophages with 50-100

µM milk-derived SM, C16-SM, or C24-SM elevated iNOS mRNA [219]. Human embryonic kidney (HEK) cells transfected with TLR4 have an increase in TNF-α production in response to exogenous 1.5 µM C8- and C2-ceramide treatment, however, adding soluble CD14 inhibited the effect of C2-ceramide [220]. It should be noted that HEK cells lack CD14, while both membrane- anchored and soluble CD14 are expressed across many cell types to varying degrees [221]. Thus, disparity in CD14 expression may cause cell type-specific effects, as cells that express less CD14 may respond differently to sphingolipids. Interestingly, longer-chain ceramides may actually compete with LPS and block TLR4 signalling, as they bind CD14 in monocytes and appear to reduce TLR4-CD14 interactions [222]. C2-ceramide (50 μM) induces c-Jun N-terminal kinase

(JNK) and MAPK in macrophages [223]. However, C2-ceramide failed to activate NF-κB.

Furthermore, C2-ceramide had no effect on LPS-stimulated translocation of NF-κB. In another study, RAW 264.7 macrophages incubated with 100 μM C2-ceramide stimulated JNK activity, despite no effect on MAPK and NF-κB [224]. Although some studies report pro-inflammatory effects, the studies with long-chain sphingolipids use much higher concentrations. C2 and C8- ceramides have questionable biological relevance since they are not naturally produced and are typically only used because they are cell permeable, unlike their long chain counterparts.

In summary, results from several studies supporting an anti-inflammatory role of exogenous long-chain sphingolipids. However, the evidence indicating pro-inflammatory effects

46 of sphingolipids in the absence of LPS when given in high doses warrants more research to uncover the precise molecular mechanisms of exogenous sphingolipids and inflammation.

2.5.2. Animal Models of Acute Inflammation

The anti-inflammatory effects of dietary sphingolipids found in cell studies have also been demonstrated in rodents. After chronically feeding young male Sprague Dawley rats a HFD containing dairy-derived ganglioside (0.02% wt:wt of the diet), elevations in serum IL-1β and

TNF-α were attenuated six hours after IP injection of LPS compared to the HFD control group

[225]. The intestinal mucosa of these rats expressed less platelet activating factor, leukotriene B4, prostaglandin E2, IL-1β, and TNF-α, alongside elevated ganglioside content, suggesting that dietary ganglioside protected against both systemic and intestinal LPS-induced inflammation

[225]. Supporting these findings, Sprague Dawley rats fed the same ganglioside-supplemented diet had heightened plasma and intestinal IL-10 when treated with LPS, while iNOS expression as well as nitrate and nitrite production were blunted with ganglioside treatment [226]. These ganglioside- rich diets also impaired the LPS-induced loss of occludin protein from intestinal mucosa suggesting improved gut barrier function [226]. Male BALB/c mice on a chow diet fed a MFGM- supplemented (12.5% wt:wt) for 5 weeks were resistant to the effects of an IP LPS-challenge [227].

The MFGM mice had reduced gut permeability and reductions in serum interferon γ (IFNγ), IL-6,

IL-10, IL-17, IL-12p70, monocyte chemoattractant protein 1 (MCP-1), and TNF-α after 24 hours

[227]. MFGM also improved mouse survival at 48 hours compared to control [227]. Upon IP injection of monosodium urate, mice fed a diet supplemented with phospholipid- and sphingolipid- rich dairy fractions (0.8-7.0% wt/wt of the diet) had reduced cell migration into the peritoneal space compared to control [214]. The dietary phospholipids seemingly reduced the recruitment of

47 immune cells. Overall, animal studies examining the effects of sphingolipid and polar lipid mixtures on acute inflammation showed a reduced inflammatory response. There are no studies examining the effects of purified SM on acute inflammation. However, again if the effects are due to the catabolic products as suggested by the in vitro studies, these findings could be applicable to dietary SM.

2.5.3. Dietary Sphingolipids and Inflammation: Clinical Trials

Like the hypolipidemic effects, there are limited clinical trials exploring the potential benefits of dietary sphingolipids in inflammation. These studies examined post-prandial inflammation due to ethical restrictions on inducing sepsis in humans. Additionally, increased post-prandial inflammation is linked with increased risk for cardiometabolic disease [228]. Thus, treatments that can attenuate this inflammation could be preventative in cardiometabolic diseases. A milk fat globular membrane fraction supplementation given to overweight adults during saturated fat-rich meals increased serum IL-10 and decreased soluble intercellular adhesion molecule (sICAM) compared to isocaloric control meals [229]. IL-10 and sICAM are inversely related, as IL-10 is anti-inflammatory and sICAM is associated with inflammation [230]. A pilot study comparing isocaloric smoothies containing palm oil, made with and without MFGM, found modest reductions in sICAM when MFGM was fed to obese adults [231]. This study did not match fatty acids between the meals making the detection of differences in post-prandial inflammation difficult. The limited number of human studies, along with the use of complex mixtures of phospholipids, makes it difficult to pinpoint the effects of sphingolipids on post-prandial inflammation in humans. Thus, more research is needed with both polar lipid extracts and purified sphingolipids to elucidate the effects in humans.

2.5.4. Dietary Sphingolipids Attenuate Animal Models of Chronic Inflammation

2.5.4.1. Colitis Models

Dietary SM (0.1% wt/wt, unspecified source) reduced colonic myeloperoxidase and histological damage induced by dextran sodium sulfate (DSS) [232]. Additionally, feeding milk SM (0.1% wt:wt) to DSS-treated mice moderately attenuated colitis in a PPARγ-dependent manner. These mice had immunostimulatory gene expression including Ccl11/19/20 and Cxcl9/11 [208].

However, some anti-inflammatory genes were also upregulated and therefore, may have reduced the potential histological damage. BALB/c mice fed a maize glucosylceramide-supplemented

(0.1% wt/wt) diet were resistant to 1,2-dimethylhydrazine induced colonic expression of TNF-α, regulated on activation, normal T cell expressed and secreted (RANTES), macrophage colony stimulating factor, and interferon-inducible T-cell alpha chemoattractant (I-TAC) [233].

Corroborating this, DSS-treated BALB/c mice fed glucosylceramide (0.1% wt:wt) also had reduced colonic expression IL-1α/β, sICAM1, IL-16, interferon gamma-induced protein 10, monokine induced by gamma interferon, and tissue inhibitor metallopeptidase 1 [234]. Overall, there are studies suggesting an inhibitory role for dietary sphingolipids in chemically-induced colitis. Conversely, 4 mg/day of egg SM administered via gavage exacerbated colitis and apoptosis in both DSS and IL-10-/- colitis mouse models [235]. This study found increased ceramide content in the intestinal epithelium. Although this study found pro-inflammatory effects, the administration method could account for the differences. The gavage method administered the 4 mg in one dose, whereas the chronic feeding studies dispersed the dose throughout the day. In line with what has been observed with in vitro studies, single large doses may have pro-inflammatory effects in vivo as well.

Other potential roles of sphingolipids in the intestine are related to the gut barrier. Rat pups fed milk SM have shown improved markers of gut maturation [236], suggesting dietary SM found in breast milk [237] may be important for new-born gut development and health. Furthermore, egg

SM protects intestinal epithelial cells from the cytotoxic detergent effects of bile salts [238]. This could have implications in HFD as deoxycholic acid plays a crucial role in HFD-induced gut barrier dysfunction [239].

2.5.4.2. Animal Models of High-Fat Diet-Induced Inflammation

As reviewed above, inflammation is crucial in the development of metabolic dysfunction. Lecomte et al. supplemented mice fed a HFD (40% kcal from fat) with 2% (wt/wt) MPL, supplying about

0.3% milk SM. In general, MPL protected mice from the deleterious effects of HFD [240]. For instance, MPL increased the intestinal expression of Mucin2, suggesting increased intestinal integrity. The adipose tissue of MPL-fed mice had reduced CD68 mRNA, a marker of adipose macrophage infiltration [240]. The effects on inflammation may be due in part to MPL ability to reduce intestinal leakiness [227], thus limiting endotoxin translocation. The complexity of phospholipid extracts makes it difficult to discern the causal agent in these studies. Not only have dietary phospholipids been reported to have their own beneficial health effects [160], they can also increase milk SM-derived ceramide appearance in the lymph [241], and thus potentially increase

SM absorption. Further studies are needed to link dietary sphingolipids to diet-induced inflammation, however, there is potential for anti-inflammatory effects as shown by MPL feeding.

2.4.5. Dietary Sphingolipids Interactions with the Gut Microbiota

The gut microbiota spans the length of the intestines but is orders of magnitude more

50 concentrated in the colon [242]. It is important to note that significant quantities of both dietary

SM and its hydrolytic products can reach the colon in humans [172] where they may affect the bulk of the gut microbiota. Sprong et al. reported sphingosine has in vitro bactericidal properties against a variety of enteropathogens including bacteria from different phyla, as well as both Gram- negative and -positive bacteria [243]. Examining the translatability of this is difficult, only a few studies report gut microbiota or microbiota-related outcomes (e.g. serum LPS) after sphingolipid feeding. Milk fat globular membrane consumption attenuated gastrointestinal complications associated with the administration of a diarrheagenic strain of E. coli in humans [244]. Infant formulas containing gangliosides significantly altered the fecal microbiota by decreasing E. coli and increasing Bifidobacterium [245]. This study suggested the effects may be related to the oligosaccharide moieties acting as prebiotics. However, the broad spectrum bactericidal effects of sphingosine [243] and interactions of specific ceramide moieties with bacteria may also play a role

[246]. In contrast, Reis et al. did not observe any significant differences in cecal microbiota of mice in response to various sphingolipid-rich diets for 5 weeks [197]. Another consideration for interactions between dietary SM and gut microbiota is the production of trimethylamine by the gut microbiota when in the presence of excess choline. Trimethylamine is a precursor to trimethylamine N-oxide, which has been tentatively related to CVD disease [247]. However, recently Rye and colleagues demonstrated that only at excessively large doses of dietary egg SM did serum trimethylamine N-oxide modestly increase [188]. Furthermore, there was a reduction in aortic lesion area when mice were fed egg SM on a low-fat diet but, there was no difference between the control and SM group when broad-spectrum antibiotics were used to deplete the gut microbiota. This suggests there is an important, potentially protective interaction between SM and

51 gut microbiota on atherosclerosis [188]. At this moment, more research is needed to fully understand the interaction of dietary SM and gut microbiota.

2.5. References

1. Ogden CL, Carroll MD, Fryar CD, Flegal KM: Prevalence of Obesity Among Adults and Youth: United States, 2011-2014. NCHS Data Brief 2015:1-8. 2. Cawley J, Meyerhoefer C: The medical care costs of obesity: an instrumental variables approach. J Health Econ 2012, 31:219-230. 3. Flegal KM, Kit BK, Orpana H, Graubard BI: Association of All-Cause Mortality With Overweight and Obesity Using Standard Body Mass Index Categories: A Systematic Review and Meta-analysis. JAMA 2013, 309:71-82. 4. Hamer M, Stamatakis E: Metabolically Healthy Obesity and Risk of All-Cause and Cardiovascular Disease Mortality. The Journal of Clinical Endocrinology & Metabolism 2012, 97:2482-2488. 5. Monteiro R, Azevedo I: Chronic Inflammation in Obesity and the Metabolic Syndrome. Mediators of Inflammation 2010, 2010:289645. 6. Gepner Y, Shelef I, Schwarzfuchs D, Zelicha H, Tene L, Yaskolka Meir A, Tsaban G, Cohen N, Bril N, Rein M, et al: Effect of Distinct Lifestyle Interventions on Mobilization of Fat Storage Pools: The CENTRAL MRI Randomized Controlled Trial. Circulation 2017. 7. Finkelstein EA, Khavjou OA, Thompson H, Trogdon JG, Pan L, Sherry B, Dietz W: Obesity and severe obesity forecasts through 2030. Am J Prev Med 2012, 42:563-570. 8. Bellissimo N, Akhavan T: Effect of Macronutrient Composition on Short-Term Food Intake and Weight Loss. Advances in Nutrition: An International Review Journal 2015, 6:302S- 308S. 9. Curb JD, Marcus EB: Body fat and obesity in Japanese Americans. The American Journal of Clinical Nutrition 1991, 53:1552S-1555S. 10. Sonne-Holm S, Sørensen TIA: Post-War course of the prevalence of extreme overweight among Danish young men. Journal of Chronic Diseases 1977, 30:351-358. 11. Price RA, Charles MA, Pettitt DJ, Knowler WC: Obesity in Pima Indians: Large increases among post-World War II birth cohorts. American Journal of Physical Anthropology 1993, 92:473-479. 12. Hariri N, Thibault L: High-fat diet-induced obesity in animal models. Nutrition Research Reviews 2010, 23:270-299. 13. Mensink RP, Zock PL, Kester AD, Katan MB: Effects of dietary fatty acids and carbohydrates on the ratio of serum total to HDL cholesterol and on serum lipids and apolipoproteins: a meta-analysis of 60 controlled trials. The American Journal of Clinical Nutrition 2003, 77:1146-1155. 14. Zhang P, Meng X, Li D, Calderone R, Mao D, Sui B: Commensal Homeostasis of Gut Microbiota-Host for the Impact of Obesity. Front Physiol 2017, 8:1122. 15. Spalding KL, Arner E, Westermark PO, Bernard S, Buchholz BA, Bergmann O, Blomqvist L, Hoffstedt J, Näslund E, Britton T, et al: Dynamics of fat cell turnover in humans. Nature 2008, 453:783. 16. Skurk T, Alberti-Huber C, Herder C, Hauner H: Relationship between Adipocyte Size and Adipokine Expression and Secretion. The Journal of Clinical Endocrinology & Metabolism 2007, 92:1023-1033. 17. Brahimi-Horn MC, Pouyssegur J: Oxygen, a source of life and stress. FEBS Lett 2007, 581:3582-3591.

18. Ye J, Gao Z, Yin J, He Q: Hypoxia is a potential risk factor for chronic inflammation and adiponectin reduction in adipose tissue of ob/ob and dietary obese mice. American Journal of Physiology-Endocrinology and Metabolism 2007, 293:E1118-E1128. 19. Ban J-J, Ruthenborg RJ, Cho KW, Kim J-w: Regulation of obesity and insulin resistance by hypoxia-inducible factors. Hypoxia 2014, 2:171-183. 20. Kawasaki N, Asada R, Saito A, Kanemoto S, Imaizumi K: Obesity-induced endoplasmic reticulum stress causes chronic inflammation in adipose tissue. Scientific Reports 2012, 2:799. 21. Ozcan U, Cao Q, Yilmaz E, Lee AH, Iwakoshi NN, Ozdelen E, Tuncman G, Gorgun C, Glimcher LH, Hotamisligil GS: Endoplasmic reticulum stress links obesity, insulin action, and type 2 diabetes. Science 2004, 306:457-461. 22. Nivala AM, Reese L, Frye M, Gentile CL, Pagliassotti MJ: Fatty acid-mediated endoplasmic reticulum stress in vivo: differential response to the infusion of Soybean and Lard Oil in rats. Metabolism 2013, 62:753-760. 23. Ozcan U, Yilmaz E, Ozcan L, Furuhashi M, Vaillancourt E, Smith RO, Gorgun CZ, Hotamisligil GS: Chemical chaperones reduce ER stress and restore glucose homeostasis in a mouse model of type 2 diabetes. Science 2006, 313:1137-1140. 24. Sano R, Reed JC: ER stress-induced cell death mechanisms. Biochimica et Biophysica Acta (BBA) - Molecular Cell Research 2013, 1833:3460-3470. 25. Furukawa S, Fujita T, Shimabukuro M, Iwaki M, Yamada Y, Nakajima Y, Nakayama O, Makishima M, Matsuda M, Shimomura I: Increased oxidative stress in obesity and its impact on metabolic syndrome. Journal of Clinical Investigation 2004, 114:1752-1761. 26. Pandey G, Shihabudeen MS, David HP, Thirumurugan E, Thirumurugan K: Association between hyperleptinemia and oxidative stress in obese diabetic subjects. Journal of Diabetes and Metabolic Disorders 2015, 14:24. 27. Fiorentino TV, Prioletta A, Zuo P, Folli F: Hyperglycemia-induced oxidative stress and its role in diabetes mellitus related cardiovascular diseases. Curr Pharm Des 2013, 19:5695- 5703. 28. de Mello AH, Costa AB, Engel JDG, Rezin GT: Mitochondrial dysfunction in obesity. Life Sci 2018, 192:26-32. 29. Al Maskari MY, Alnaqdy AA: Correlation between Serum Leptin Levels, Body Mass Index and Obesity in Omanis. Sultan Qaboos University Medical Journal 2006, 6:27-31. 30. Gupta A, Beg M, Kumar D, Shankar K, Varshney S, Rajan S, Srivastava A, Singh K, Sonkar S, Mahdi AA, et al: Chronic hyper-leptinemia induces insulin signaling disruption in adipocytes: Implications of NOS2. Free Radical Biology and Medicine 2017, 112:93-108. 31. Sada K, Nishikawa T, Kukidome D, Yoshinaga T, Kajihara N, Sonoda K, Senokuchi T, Motoshima H, Matsumura T, Araki E: Hyperglycemia Induces Cellular Hypoxia through Production of Mitochondrial ROS Followed by Suppression of Aquaporin-1. PLOS ONE 2016, 11:e0158619. 32. Ong JM, Kern PA: Effect of feeding and obesity on lipoprotein lipase activity, immunoreactive protein, and messenger RNA levels in human adipose tissue. Journal of Clinical Investigation 1989, 84:305-311. 33. Pollare T, Vessby B, Lithell H: Lipoprotein lipase activity in skeletal muscle is related to insulin sensitivity. Arterioscler Thromb 1991, 11:1192-1203. 34. Zambo V, Simon-Szabo L, Szelenyi P, Kereszturi E, Banhegyi G, Csala M: Lipotoxicity in the liver. World J Hepatol 2013, 5:550-557.

35. Furukawa S, Fujita T, Shimabukuro M, Iwaki M, Yamada Y, Nakajima Y, Nakayama O, Makishima M, Matsuda M, Shimomura I: Increased oxidative stress in obesity and its impact on metabolic syndrome. J Clin Invest 2004, 114:1752-1761. 36. Rabot S, Membrez M, Bruneau A, Gérard P, Harach T, Moser M, Raymond F, Mansourian R, Chou CJ: Germ-free C57BL/6J mice are resistant to high-fat-diet-induced insulin resistance and have altered cholesterol metabolism. The FASEB Journal 2010, 24:4948-4959. 37. Bäckhed F, Ding H, Wang T, Hooper LV, Koh GY, Nagy A, Semenkovich CF, Gordon JI: The gut microbiota as an environmental factor that regulates fat storage. Proceedings of the National Academy of Sciences of the United States of America 2004, 101:15718-15723. 38. Rabot S, Membrez M, Blancher F, Berger B, Moine D, Krause L, Bibiloni R, Bruneau A, Gérard P, Siddharth J, et al: High fat diet drives obesity regardless the composition of gut microbiota in mice. Scientific Reports 2016, 6:32484. 39. Ding S, Chi MM, Scull BP, Rigby R, Schwerbrock NMJ, Magness S, Jobin C, Lund PK: High-Fat Diet: Bacteria Interactions Promote Intestinal Inflammation Which Precedes and Correlates with Obesity and Insulin Resistance in Mouse. PLOS ONE 2010, 5:e12191. 40. Cani PD, Bibiloni R, Knauf C, Waget A, Neyrinck AM, Delzenne NM, Burcelin R: Changes in Gut Microbiota Control Metabolic Endotoxemia-Induced Inflammation in High- Fat Diet–Induced Obesity and Diabetes in Mice. Diabetes 2008, 57:1470. 41. Murphy EA, Velazquez KT, Herbert KM: Influence of High-Fat-Diet on Gut Microbiota: A Driving Force for Chronic Disease Risk. Current opinion in clinical nutrition and metabolic care 2015, 18:515-520. 42. Bes-Houtmann S, Roche R, Hoareau L, Gonthier MP, Festy F, Caillens H, Gasque P, Lefebvre d'Hellencourt C, Cesari M: Presence of functional TLR2 and TLR4 on human adipocytes. Histochem Cell Biol 2007, 127:131-137. 43. Hornef MW, Frisan T, Vandewalle A, Normark S, Richter-Dahlfors A: Toll-like Receptor 4 Resides in the Golgi Apparatus and Colocalizes with Internalized Lipopolysaccharide in Intestinal Epithelial Cells. The Journal of Experimental Medicine 2002, 195:559. 44. Ghoshal S, Witta J, Zhong J, de Villiers W, Eckhardt E: Chylomicrons promote intestinal absorption of lipopolysaccharides. J Lipid Res 2009, 50:90-97. 45. Neal MD, Leaphart C, Levy R, Prince J, Billiar TR, Watkins S, Li J, Cetin S, Ford H, Schreiber A, Hackam DJ: Enterocyte TLR4 Mediates Phagocytosis and Translocation of Bacteria Across the Intestinal Barrier. The Journal of Immunology 2006, 176:3070. 46. Makki K, Froguel P, Wolowczuk I: Adipose Tissue in Obesity-Related Inflammation and Insulin Resistance: Cells, Cytokines, and Chemokines. ISRN Inflammation 2013, 2013:139239. 47. Rittig N, Bach E, Thomsen HH, Pedersen SB, Nielsen TS, Jørgensen JO, Jessen N, Møller N: Regulation of Lipolysis and Adipose Tissue Signaling during Acute Endotoxin-Induced Inflammation: A Human Randomized Crossover Trial. PLoS ONE 2016, 11:e0162167. 48. Erion DM, Shulman GI: Diacylglycerol-mediated insulin resistance. Nature medicine 2010, 16:400-402. 49. Gavrila A, Chan JL, Yiannakouris N, Kontogianni M, Miller LC, Orlova C, Mantzoros CS: Serum Adiponectin Levels Are Inversely Associated with Overall and Central Fat Distribution but Are Not Directly Regulated by Acute Fasting or Leptin Administration in Humans: Cross-Sectional and Interventional Studies. The Journal of Clinical Endocrinology & Metabolism 2003, 88:4823-4831.

50. Musso G, Gambino R, Durazzo M, Biroli G, Carello M, Fagà E, Pacini G, De Michieli F, Rabbione L, Premoli A, et al: Adipokines in NASH: Postprandial lipid metabolism as a link between adiponectin and liver disease. Hepatology 2005, 42:1175-1183. 51. Choi ES, Kim MK, Song MK, Kim JM, Kim ES, Chung WJ, Park KS, Cho KB, Hwang JS, Jang BK: Association between Serum Irisin Levels and Non-Alcoholic Fatty Liver Disease in Health Screen Examinees. PLoS ONE 2014, 9:e110680. 52. Wolf AM, Wolf D, Avila MA, Moschen AR, Berasain C, Enrich B, Rumpold H, Tilg H: Up-regulation of the anti-inflammatory adipokine adiponectin in acute liver failure in mice. J Hepatol 2006, 44:537-543. 53. Eglit T, Ringmets I, Lember M: Obesity, High-Molecular-Weight (HMW) Adiponectin, and Metabolic Risk Factors: Prevalence and Gender-Specific Associations in Estonia. PLOS ONE 2013, 8:e73273. 54. Park P-h, McMullen MR, Huang H, Thakur V, Nagy LE: Short-term Treatment of RAW264.7 Macrophages with Adiponectin Increases Tumor Necrosis Factor-α (TNF-α) Expression via ERK1/2 Activation and Egr-1 Expression: ROLE OF TNF-α IN ADIPONECTIN-STIMULATED INTERLEUKIN-10 PRODUCTION. The Journal of biological chemistry 2007, 282:21695-21703. 55. Folco EJ, Rocha VZ, López-Ilasaca M, Libby P: Adiponectin Inhibits Pro-inflammatory Signaling in Human Macrophages Independent of Interleukin-10. The Journal of Biological Chemistry 2009, 284:25569-25575. 56. Saijo S, Nagata K, Nakano Y, Tobe T, Kobayashi Y: Inhibition by adiponectin of IL-8 production by human macrophages upon coculturing with late apoptotic cells. Biochem Biophys Res Commun 2005, 334:1180-1183. 57. Day CP, James OF: Steatohepatitis: a tale of two "hits"? Gastroenterology 1998, 114:842-845. 58. Machado M, Marques-Vidal P, Cortez-Pinto H: Hepatic histology in obese patients undergoing bariatric surgery. J Hepatol 2006, 45:600-606. 59. Donnelly KL, Smith CI, Schwarzenberg SJ, Jessurun J, Boldt MD, Parks EJ: Sources of fatty acids stored in liver and secreted via lipoproteins in patients with nonalcoholic fatty liver disease. J Clin Invest 2005, 115:1343-1351. 60. Froesch ER: The physiology and pharmacology of adipose tissue lipolysis: Its inhibition and implications for the treatment of diabetes. Diabetologia 1967, 3:475-487. 61. Lewis GF, Carpentier A, Adeli K, Giacca A: Disordered fat storage and mobilization in the pathogenesis of insulin resistance and type 2 diabetes. Endocr Rev 2002, 23:201-229. 62. Field PA, Gibbons GF: Regulation of hepatic chylomicron remnant uptake: effects of obesity and insulin resistance. Biochem Soc Trans 1997, 25:S678. 63. Savage DB, Petersen KF, Shulman GI: Disordered Lipid Metabolism and the Pathogenesis of Insulin Resistance. Physiological reviews 2007, 87:507-520. 64. Azzout-Marniche D, Bécard D, Guichard C, Foretz M, Ferré P, Foufelle F: Insulin effects on sterol regulatory-element-binding protein-1c (SREBP-1c) transcriptional activity in rat hepatocytes. Biochemical Journal 2000, 350:389-393. 65. Li S, Brown MS, Goldstein JL: Bifurcation of insulin signaling pathway in rat liver: mTORC1 required for stimulation of lipogenesis, but not inhibition of gluconeogenesis. Proceedings of the National Academy of Sciences 2010, 107:3441-3446.

66. Laplante M, Sabatini DM: mTORC1 activates SREBP-1c and uncouples lipogenesis from gluconeogenesis. Proceedings of the National Academy of Sciences of the United States of America 2010, 107:3281-3282. 67. Towle HC: Glucose as a regulator of eukaryotic gene transcription. Trends Endocrinol Metab 2005, 16:489-494. 68. Samuel VT, Liu ZX, Qu X, Elder BD, Bilz S, Befroy D, Romanelli AJ, Shulman GI: Mechanism of hepatic insulin resistance in non-alcoholic fatty liver disease. J Biol Chem 2004, 279:32345-32353. 69. Shi H, Kokoeva MV, Inouye K, Tzameli I, Yin H, Flier JS: TLR4 links innate immunity and fatty acid-induced insulin resistance. J Clin Invest 2006, 116:3015-3025. 70. Sheu S-S, Nauduri D, Anders MW: Targeting antioxidants to mitochondria: A new therapeutic direction. Biochimica et Biophysica Acta (BBA) - Molecular Basis of Disease 2006, 1762:256-265. 71. Rector RS, Thyfault JP, Uptergrove GM, Morris EM, Naples SP, Borengasser SJ, Mikus CR, Laye MJ, Laughlin MH, Booth FW, Ibdah JA: Mitochondrial dysfunction precedes insulin resistance and hepatic steatosis and contributes to the natural history of non-alcoholic fatty liver disease in an obese rodent model. J Hepatol 2010, 52:727-736. 72. Takamura T, Misu H, Matsuzawa-Nagata N, Sakurai M, Ota T, Shimizu A, Kurita S, Takeshita Y, Ando H, Honda M, Kaneko S: Obesity upregulates genes involved in oxidative phosphorylation in livers of diabetic patients. Obesity (Silver Spring) 2008, 16:2601-2609. 73. Sumida Y, Niki E, Naito Y, Yoshikawa T: Involvement of free radicals and oxidative stress in NAFLD/NASH. Free Radic Res 2013, 47:869-880. 74. Liu W, Yin H, Akazawa YO, Yoshida Y, Niki E, Porter NA: Ex vivo oxidation in tissue and plasma assays of hydroxyoctadecadienoates: Z,E/E,E stereoisomer ratios. Chem Res Toxicol 2010, 23:986-995. 75. Rao J, Zhang C, Wang P, Lu L, Qian X, Qin J, Pan X, Li G, Wang X, Zhang F: C/EBP homologous protein (CHOP) contributes to hepatocyte death via the promotion of ERO1alpha signalling in acute liver failure. Biochem J 2015, 466:369-378. 76. Gaschler MM, Stockwell BR: Lipid peroxidation in cell death. Biochemical and Biophysical Research Communications 2017, 482:419-425. 77. Cichoż-Lach H, Michalak A: Oxidative stress as a crucial factor in liver diseases. World Journal of Gastroenterology : WJG 2014, 20:8082-8091. 78. Lopez-Velazquez JA, Silva-Vidal KV, Ponciano-Rodriguez G, Chavez-Tapia NC, Arrese M, Uribe M, Mendez-Sanchez N: The prevalence of nonalcoholic fatty liver disease in the Americas. Ann Hepatol 2014, 13:166-178. 79. Spengler EK, Loomba R: Recommendations for Diagnosis, Referral for Liver Biopsy, and Treatment of Nonalcoholic Fatty Liver Disease and Nonalcoholic Steatohepatitis. Mayo Clin Proc 2015, 90:1233-1246. 80. Vernon G, Baranova A, Younossi ZM: Systematic review: the epidemiology and natural history of non-alcoholic fatty liver disease and non-alcoholic steatohepatitis in adults. Aliment Pharmacol Ther 2011, 34:274-285. 81. Das K, Chowdhury A: Lean NASH: distinctiveness and clinical implication. Hepatology International 2013, 7:806-813. 82. Petaja EM, Sevastianova K, Hakkarainen A, Orho-Melander M, Lundbom N, Yki-Jarvinen H: Adipocyte size is associated with NAFLD independent of obesity, fat distribution, and PNPLA3 genotype. Obesity (Silver Spring) 2013, 21:1174-1179.

83. Monetti M, Levin MC, Watt MJ, Sajan MP, Marmor S, Hubbard BK, Stevens RD, Bain JR, Newgard CB, Farese RV, Sr., et al: Dissociation of hepatic steatosis and insulin resistance in mice overexpressing DGAT in the liver. Cell Metab 2007, 6:69-78. 84. Li ZZ, Berk M, McIntyre TM, Feldstein AE: Hepatic Lipid Partitioning and Liver Damage in Nonalcoholic Fatty Liver Disease: ROLE OF STEAROYL-CoA DESATURASE. The Journal of Biological Chemistry 2009, 284:5637-5644. 85. Flowers MT, Groen AK, Oler AT, Keller MP, Choi Y, Schueler KL, Richards OC, Lan H, Miyazaki M, Kuipers F, et al: Cholestasis and hypercholesterolemia in SCD1-deficient mice fed a low-fat, high-carbohydrate diet. J Lipid Res 2006, 47:2668-2680. 86. Savard C, Tartaglione EV, Kuver R, Haigh WG, Farrell GC, Subramanian S, Chait A, Yeh MM, Quinn LS, Ioannou GN: Synergistic interaction of dietary cholesterol and dietary fat in inducing experimental steatohepatitis. Hepatology 2013, 57:81-92. 87. Subramanian S, Goodspeed L, Wang S, Kim J, Zeng L, Ioannou GN, Haigh WG, Yeh MM, Kowdley KV, O'Brien KD, et al: Dietary cholesterol exacerbates hepatic steatosis and inflammation in obese LDL receptor-deficient mice. J Lipid Res 2011, 52:1626-1635. 88. Bashiri A, Nesan D, Tavallaee G, Sue-Chue-Lam I, Chien K, Maguire GF, Naples M, Zhang J, Magomedova L, Adeli K, et al: Cellular cholesterol accumulation modulates high fat high sucrose (HFHS) diet-induced ER stress and hepatic inflammasome activation in the development of non-alcoholic steatohepatitis. Biochim Biophys Acta 2016, 1861:594-605. 89. Hager L, Li L, Pun H, Liu L, Hossain MA, Maguire GF, Naples M, Baker C, Magomedova L, Tam J, et al: Lecithin:cholesterol acyltransferase deficiency protects against cholesterol- induced hepatic endoplasmic reticulum stress in mice. J Biol Chem 2012, 287:20755-20768. 90. Wu H, Ballantyne CM: Skeletal muscle inflammation and insulin resistance in obesity. The Journal of Clinical Investigation 2017, 127:43-54. 91. Perseghin G, Scifo P, De Cobelli F, Pagliato E, Battezzati A, Arcelloni C, Vanzulli A, Testolin G, Pozza G, Del Maschio A, Luzi L: Intramyocellular triglyceride content is a determinant of in vivo insulin resistance in humans: a 1H-13C nuclear magnetic resonance spectroscopy assessment in offspring of type 2 diabetic parents. Diabetes 1999, 48:1600-1606. 92. Heo JW, No MH, Park DH, Kang JH, Seo DY, Han J, Neufer PD, Kwak HB: Effects of exercise on obesity-induced mitochondrial dysfunction in skeletal muscle. Korean J Physiol Pharmacol 2017, 21:567-577. 93. Kelley DE, Goodpaster BH: Skeletal Muscle Triglyceride. Diabetes Care 2001, 24:933. 94. Goodpaster BH, Leland Thaete F, Simoneau J-A, Kelley DE: Subcutaneous Abdominal Fat and Thigh Muscle Composition Predict Insulin Sensitivity Independently of Visceral Fat. Diabetes 1997, 46:1579. 95. Mootha VK, Lindgren CM, Eriksson KF, Subramanian A, Sihag S, Lehar J, Puigserver P, Carlsson E, Ridderstrale M, Laurila E, et al: PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nat Genet 2003, 34:267- 273. 96. Twig G, Elorza A, Molina AJ, Mohamed H, Wikstrom JD, Walzer G, Stiles L, Haigh SE, Katz S, Las G, et al: Fission and selective fusion govern mitochondrial segregation and elimination by autophagy. Embo j 2008, 27:433-446. 97. Carey AL, Steinberg GR, Macaulay SL, Thomas WG, Holmes AG, Ramm G, Prelovsek O, Hohnen-Behrens C, Watt MJ, James DE, et al: Interleukin-6 increases insulin-stimulated glucose disposal in humans and glucose uptake and fatty acid oxidation in vitro via AMP- activated protein kinase. Diabetes 2006, 55:2688-2697.

98. Varma V, Yao-Borengasser A, Rasouli N, Nolen GT, Phanavanh B, Starks T, Gurley C, Simpson P, McGehee RE, Kern PA, Peterson CA: Muscle inflammatory response and insulin resistance: synergistic interaction between macrophages and fatty acids leads to impaired insulin action. American Journal of Physiology - Endocrinology and Metabolism 2009, 296:E1300-E1310. 99. Green CJ, Pedersen M, Pedersen BK, Scheele C: Elevated NF-kappaB activation is conserved in human myocytes cultured from obese type 2 diabetic patients and attenuated by AMP-activated protein kinase. Diabetes 2011, 60:2810-2819. 100. Ciaraldi TP, Ryan AJ, Mudaliar SR, Henry RR: Altered Myokine Secretion Is an Intrinsic Property of Skeletal Muscle in Type 2 Diabetes. PLoS ONE 2016, 11:e0158209. 101. Wilcox G: Insulin and Insulin Resistance. Clinical Biochemist Reviews 2005, 26:19-39. 102. Furtado LM, Somwar R, Sweeney G, Niu W, Klip A: Activation of the glucose transporter GLUT4 by insulin. Biochem Cell Biol 2002, 80:569-578. 103. Rui L: Energy Metabolism in the Liver. Comprehensive Physiology 2014, 4:177-197. 104. Taniguchi CM, Kondo T, Sajan M, Luo J, Bronson R, Asano T, Farese R, Cantley LC, Kahn CR: Divergent regulation of hepatic glucose and lipid metabolism by phosphoinositide 3-kinase via Akt and PKClambda/zeta. Cell Metab 2006, 3:343-353. 105. Cohen P, Nimmo HG, Proud CG: How does insulin stimulate glycogen synthesis? Biochem Soc Symp 1978:69-95. 106. Amemiya-Kudo M, Shimano H, Hasty AH, Yahagi N, Yoshikawa T, Matsuzaka T, Okazaki H, Tamura Y, Iizuka Y, Ohashi K, et al: Transcriptional activities of nuclear SREBP- 1a, -1c, and -2 to different target promoters of lipogenic and cholesterogenic genes. Journal of Lipid Research 2002, 43:1220-1235. 107. Timlin MT, Parks EJ: Temporal pattern of de novo lipogenesis in the postprandial state in healthy men. The American Journal of Clinical Nutrition 2005, 81:35-42. 108. Lewis GF, Steiner G: Acute effects of insulin in the control of VLDL production in humans. Implications for the insulin-resistant state. Diabetes Care 1996, 19:390-393. 109. Petersen KF, Laurent D, Rothman DL, Cline GW, Shulman GI: Mechanism by which glucose and insulin inhibit net hepatic glycogenolysis in humans. Journal of Clinical Investigation 1998, 101:1203-1209. 110. Ramnanan CJ, Edgerton DS, Kraft G, Cherrington AD: Physiologic action of glucagon on liver glucose metabolism. Diabetes Obes Metab 2011, 13 Suppl 1:118-125. 111. Kawamori D, Kurpad AJ, Hu J, Liew CW, Shih JL, Ford EL, Herrera PL, Polonsky KS, McGuinness OP, Kulkarni RN: Insulin signaling in alpha cells modulates glucagon secretion in vivo. Cell Metab 2009, 9:350-361. 112. Jeremy M Berg, John L Tymoczko, Stryer L: Glucose Can Be Synthesized from Noncarbohydrate Precursors. In Biochemistry 5th Edition. New York: WH Freeman 2002 113. Fisher SJ, Kahn CR: Insulin signaling is required for insulin's direct and indirect action on hepatic glucose production. J Clin Invest 2003, 111:463-468. 114. Tyagi S, Gupta P, Saini AS, Kaushal C, Sharma S: The peroxisome proliferator- activated receptor: A family of nuclear receptors role in various diseases. Journal of Advanced Pharmaceutical Technology & Research 2011, 2:236-240. 115. Bookout AL, Jeong Y, Downes M, Yu RT, Evans RM, Mangelsdorf DJ: Anatomical Profiling of Nuclear Receptor Expression Reveals a Hierarchical Transcriptional Network. Cell 2006, 126:789-799.

116. Rakhshandehroo M, Hooiveld G, Müller M, Kersten S: Comparative Analysis of Gene Regulation by the Transcription Factor PPARα between Mouse and Human. PLOS ONE 2009, 4:e6796. 117. Siersbaek R, Nielsen R, Mandrup S: PPARgamma in adipocyte differentiation and metabolism--novel insights from genome-wide studies. FEBS Lett 2010, 584:3242-3249. 118. Zhang Y-L, Hernandez-Ono A, Siri P, Weisberg S, Conlon D, Graham MJ, Crooke RM, Huang L-S, Ginsberg HN: Aberrant Hepatic Expression of PPARγ2 Stimulates Hepatic Lipogenesis in a Mouse Model of Obesity, Insulin Resistance, Dyslipidemia, and Hepatic Steatosis. Journal of Biological Chemistry 2006, 281:37603-37615. 119. Pettinelli P, Videla LA: Up-regulation of PPAR-gamma mRNA expression in the liver of obese patients: an additional reinforcing lipogenic mechanism to SREBP-1c induction. J Clin Endocrinol Metab 2011, 96:1424-1430. 120. Vidal-Puig A, Jimenez-Liñan M, Lowell BB, Hamann A, Flier JS, Moller DE, Hu E, Spiegelman B: Regulation of PPAR γ gene expression by nutrition and obesity in rodents. Journal of Clinical Investigation 1996, 97:2553-2561. 121. Rogue A, Spire C, Brun M, Claude N, Guillouzo A: Gene Expression Changes Induced by PPAR Gamma Agonists in Animal and Human Liver. PPAR Research 2010, 2010:16. 122. Laplante M, Sell H, MacNaul KL, Richard D, Berger JP, Deshaies Y: PPAR-gamma activation mediates adipose depot-specific effects on gene expression and lipoprotein lipase activity: mechanisms for modulation of postprandial lipemia and differential adipose accretion. Diabetes 2003, 52:291-299. 123. Wu Z, Xie Y, Morrison RF, Bucher NL, Farmer SR: PPARgamma induces the insulin- dependent glucose transporter GLUT4 in the absence of C/EBPalpha during the conversion of 3T3 fibroblasts into adipocytes. Journal of Clinical Investigation 1998, 101:22-32. 124. Inoue M, Ohtake T, Motomura W, Takahashi N, Hosoki Y, Miyoshi S, Suzuki Y, Saito H, Kohgo Y, Okumura T: Increased expression of PPARgamma in high fat diet-induced liver steatosis in mice. Biochem Biophys Res Commun 2005, 336:215-222. 125. Carey DG, Cowin GJ, Galloway GJ, Jones NP, Richards JC, Biswas N, Doddrell DM: Effect of rosiglitazone on insulin sensitivity and body composition in type 2 diabetic patients [corrected]. Obes Res 2002, 10:1008-1015. 126. Sutinen J, Hakkinen AM, Westerbacka J, Seppala-Lindroos A, Vehkavaara S, Halavaara J, Jarvinen A, Ristola M, Yki-Jarvinen H: Rosiglitazone in the treatment of HAART-associated lipodystrophy--a randomized double-blind placebo-controlled study. Antivir Ther 2003, 8:199-207. 127. Teranishi T, Ohara T, Maeda K, Zenibayashi M, Kouyama K, Hirota Y, Kawamitsu H, Fujii M, Sugimura K, Kasuga M: Effects of pioglitazone and metformin on intracellular lipid content in liver and skeletal muscle of individuals with type 2 diabetes mellitus. Metabolism 2007, 56:1418-1424. 128. Janowski BA, Grogan MJ, Jones SA, Wisely GB, Kliewer SA, Corey EJ, Mangelsdorf DJ: Structural requirements of ligands for the oxysterol liver X receptors LXRalpha and LXRbeta. Proc Natl Acad Sci U S A 1999, 96:266-271. 129. Olkkonen VM, Béaslas O, Nissilä E: Oxysterols and Their Cellular Effectors. Biomolecules 2012, 2:76-103. 130. Zhao C, Dahlman-Wright K: Liver X receptor in cholesterol metabolism. J Endocrinol 2010, 204:233-240.

131. Zelcer N, Hong C, Boyadjian R, Tontonoz P: LXR regulates cholesterol uptake through Idol-dependent ubiquitination of the LDL receptor. Science 2009, 325:100-104. 132. Lagace TA: PCSK9 and LDLR degradation: regulatory mechanisms in circulation and in cells. Current Opinion in Lipidology 2014, 25:387-393. 133. Zhang CP, Tian Y, Zhang M, Tuo QH, Chen JX, Liao DF: IDOL, inducible degrader of low-density lipoprotein receptor, serves as a potential therapeutic target for dyslipidemia. Med Hypotheses 2016, 86:138-142. 134. Jia L, Betters JL, Yu L: Niemann-Pick C1-Like 1 (NPC1L1) Protein in Intestinal and Hepatic Cholesterol Transport. Annual review of physiology 2011, 73:239-259. 135. Wang J, Mitsche MA, Lutjohann D, Cohen JC, Xie XS, Hobbs HH: Relative roles of ABCG5/ABCG8 in liver and intestine. J Lipid Res 2015, 56:319-330. 136. Yellaturu CR, Deng X, Park EA, Raghow R, Elam MB: Insulin Enhances the Biogenesis of Nuclear Sterol Regulatory Element-binding Protein (SREBP)-1c by Posttranscriptional Down-regulation of Insig-2A and Its Dissociation from SREBP Cleavage-activating Protein (SCAP)·SREBP-1c Complex. The Journal of Biological Chemistry 2009, 284:31726-31734. 137. Goldstein JL, DeBose-Boyd RA, Brown MS: Protein sensors for membrane sterols. Cell 2006, 124:35-46. 138. Okazaki H, Goldstein JL, Brown MS, Liang G: LXR-SREBP-1c-Phospholipid Transfer Protein Axis Controls Very Low Density Lipoprotein (VLDL) Particle Size. Journal of Biological Chemistry 2010, 285:6801-6810. 139. Tian J, Goldstein JL, Brown MS: Insulin induction of SREBP-1c in rodent liver requires LXRα-C/EBPβ complex. Proceedings of the National Academy of Sciences of the United States of America 2016, 113:8182-8187. 140. Hegarty BD, Bobard A, Hainault I, Ferré P, Bossard P, Foufelle F: Distinct roles of insulin and liver X receptor in the induction and cleavage of sterol regulatory elementbinding protein-1c. Proceedings of the National Academy of Sciences of the United States of America 2005, 102:791-796. 141. !!! INVALID CITATION !!! {Amemiya-Kudo, 2002 #29}. 142. Tao R, Xiong X, DePinho RA, Deng C-X, Dong XC: Hepatic SREBP-2 and cholesterol biosynthesis are regulated by FoxO3 and Sirt6. Journal of Lipid Research 2013, 54:2745-2753. 143. Horton JD, Shimomura I, Brown MS, Hammer RE, Goldstein JL, Shimano H: Activation of cholesterol synthesis in preference to fatty acid synthesis in liver and adipose tissue of transgenic mice overproducing sterol regulatory element-binding protein-2. Journal of Clinical Investigation 1998, 101:2331-2339. 144. Lipka LJ: Ezetimibe: a first-in-class, novel cholesterol absorption inhibitor. Cardiovasc Drug Rev 2003, 21:293-312. 145. Chan DC, Watts GF, Gan SK, Ooi EM, Barrett PH: Effect of ezetimibe on hepatic fat, inflammatory markers, and apolipoprotein B-100 kinetics in insulin-resistant obese subjects on a weight loss diet. Diabetes Care 2010, 33:1134-1139. 146. Takase H, Dohi Y, Okado T, Hashimoto T, Goto Y, Kimura G: Effects of ezetimibe on visceral fat in the metabolic syndrome: a randomised controlled study. Eur J Clin Invest 2012, 42:1287-1294. 147. Jain SS, Ramanand SJ, Ramanand JB, Akat PB, Patwardhan MH, Joshi SR: Evaluation of efficacy and safety of orlistat in obese patients. Indian Journal of Endocrinology and Metabolism 2011, 15:99-104.

148. Stolk MF, Becx MC, Kuypers KC, Seldenrijk CA: Severe hepatic side effects of ezetimibe. Clin Gastroenterol Hepatol 2006, 4:908-911. 149. Montero JL, Muntane J, Fraga E, Delgado M, Costan G, Serrano M, Padillo J, de la Mata M, Mino G: Orlistat associated subacute hepatic failure. J Hepatol 2001, 34:173. 150. Filippatos TD, Derdemezis CS, Gazi IF, Nakou ES, Mikhailidis DP, Elisaf MS: Orlistat- associated adverse effects and drug interactions: a critical review. Drug Saf 2008, 31:53-65. 151. Garcia SB, Barros LT, Turatti A, Martinello F, Modiano P, Ribeiro-Silva A, Vespucio MV, Uyemura SA: The anti-obesity agent Orlistat is associated to increase in colonic preneoplastic markers in rats treated with a chemical carcinogen. Cancer Lett 2006, 240:221-224. 152. Sud M, Fahy E, Cotter D, Brown A, Dennis E, Glass C, Murphy R, Raetz C, Russell D, Subramaniam S: LMSD: LIPID MAPS structure database. Nucleic Acids Research 2006, 35:D527-532. 153. Karlsson KA: On the chemistry and occurrence of sphingolipid long-chain bases. Chemistry and Physics of Lipids 1970, 5:6-43. 154. Karlsson KA: Sphingolipid long chain bases. Lipids 1970, 5:878-891. 155. Pruett ST, Bushnev A, Hagedorn K, Adiga M, Haynes CA, Sullards MC, Liotta DC, Merrill AH, Jr.: Biodiversity of sphingoid bases ("sphingosines") and related amino alcohols. J Lipid Res 2008, 49:1621-1639. 156. Vesper H, Schmelz EM, Nikolova-Karakashian MN, Dillehay DL, Lynch DV, Merrill AH, Jr.: Sphingolipids in food and the emerging importance of sphingolipids to nutrition. J Nutr 1999, 129:1239-1250. 157. Byrdwell WC, Perry RH: Liquid chromatography with dual parallel mass spectrometry and 31P nuclear magnetic resonance spectroscopy for analysis of sphingomyelin and dihydrosphingomyelin. II. Bovine milk sphingolipids. J Chromatogr A 2007, 1146:164-185. 158. Byrdwell WC, Perry RH: Liquid chromatography with dual parallel mass spectrometry and (31)P nuclear magnetic resonance spectroscopy for analysis of sphingomyelin and dihydrosphingomyelin. I. Bovine brain and chicken egg yolk. J Chromatogr A 2006, 1133:149-171. 159. Epand RM: Cholesterol in Bilayers of Sphingomyelin or Dihydrosphingomyelin at Concentrations Found in Ocular Lens Membranes. Biophysical Journal 2003, 84:3102-3110. 160. Blesso CN: Egg phospholipids and cardiovascular health. Nutrients 2015, 7:2731-2747. 161. Kullenberg D, Taylor LA, Schneider M, Massing U: Health effects of dietary phospholipids. Lipids Health Dis 2012, 11:3. 162. Weihrauch JL, Son Y-S: Phospholipid content of foods. Journal of the American Oil Chemists' Society 1983, 60:1971-1978. 163. Fujino Y, Ohnishi M, Ito S: Molecular Species of Ceramide and Mono-, Di-, Tri-, and Tetraglycosylceramide in Bran and Endosperm of Rice Grains. Agricultural and Biological Chemistry 1985, 49:2753-2762. 164. Wang L, Wang T, Fehr WR: HPLC quantification of sphingolipids in soybeans with modified palmitate content. J Agric Food Chem 2006, 54:7422-7428. 165. Takakuwa N, Saito K, Ohnishi M, Oda Y: Determination of glucosylceramide contents in crop tissues and by-products from their processing. Bioresour Technol 2005, 96:1089-1092. 166. Sugawara T, Miyazawa T: Separation and determination of glycolipids from edible plant sources by high-performance liquid chromatography and evaporative light-scattering detection. Lipids 1999, 34:1231-1237.

167. Khor GL, Shyam S, Misra S, Fong B, Chong MHZ, Sulaiman N, Lee YL, Cannan R, Rowan A: Correlation between dietary intake and serum ganglioside concentrations: a cross- sectional study among Malaysian toddlers. BMC Nutrition 2016, 2:74. 168. Leese HJ, Semenza G: On the Identity between the Small Intestinal Enzymes Phlorizin Hydrolase and Glycosylceramidase. Journal of Biological Chemistry 1973, 248:8170-8173. 169. Nilsson A, Duan RD: Absorption and lipoprotein transport of sphingomyelin. J Lipid Res 2006, 47:154-171. 170. Garmy N, Taieb N, Yahi N, Fantini J: Apical uptake and transepithelial transport of sphingosine monomers through intact human intestinal epithelial cells: physicochemical and molecular modeling studies. Arch Biochem Biophys 2005, 440:91-100. 171. Liu JJ, Nilsson A, Duan RD: Effects of phospholipids on sphingomyelin hydrolysis induced by intestinal alkaline sphingomyelinase: an in vitro study. J Nutr Biochem 2000, 11:192-197. 172. Ohlsson L, Hertervig E, Jonsson BA, Duan RD, Nyberg L, Svernlov R, Nilsson A: Sphingolipids in human ileostomy content after meals containing milk sphingomyelin. Am J Clin Nutr 2010, 91:672-678. 173. Nyberg L, Duan RD, Nilsson A: A mutual inhibitory effect on absorption of sphingomyelin and cholesterol. J Nutr Biochem 2000, 11:244-249. 174. Eckhardt ER, Wang DQ, Donovan JM, Carey MC: Dietary sphingomyelin suppresses intestinal cholesterol absorption by decreasing thermodynamic activity of cholesterol monomers. Gastroenterology 2002, 122:948-956. 175. Noh SK, Koo SI: Egg sphingomyelin lowers the lymphatic absorption of cholesterol and alpha-tocopherol in rats. J Nutr 2003, 133:3571-3576. 176. Noh SK, Koo SI: Milk sphingomyelin is more effective than egg sphingomyelin in inhibiting intestinal absorption of cholesterol and fat in rats. J Nutr 2004, 134:2611-2616. 177. Duivenvoorden I, Voshol PJ, Rensen PC, van Duyvenvoorde W, Romijn JA, Emeis JJ, Havekes LM, Nieuwenhuizen WF: Dietary sphingolipids lower plasma cholesterol and triacylglycerol and prevent liver steatosis in APOE*3Leiden mice. Am J Clin Nutr 2006, 84:312-321. 178. Narita T, Naganuma T, Sase Y, Kihara A: Long-chain bases of sphingolipids are transported into cells via the acyl-CoA synthetases. Sci Rep 2016, 6:25469. 179. Garmy N, Taieb N, Yahi N, Fantini J: Interaction of cholesterol with sphingosine: physicochemical characterization and impact on intestinal absorption. J Lipid Res 2005, 46:36-45. 180. Feng D, Ohlsson L, Ling W, Nilsson A, Duan RD: Generating ceramide from sphingomyelin by alkaline sphingomyelinase in the gut enhances sphingomyelin-induced inhibition of cholesterol uptake in Caco-2 cells. Dig Dis Sci 2010, 55:3377-3383. 181. Norris GH, Jiang C, Ryan J, Porter CM, Blesso CN: Milk sphingomyelin improves lipid metabolism and alters gut microbiota in high fat diet-fed mice. J Nutr Biochem 2016, 30:93- 101. 182. Watanabe S, Takahashi T, Tanaka L, Miyashita K: The effect of milk polar lipids separated from butter serum on the lipid levels in the liver and the plasma of obese-model mouse (KK-Ay). J Funct Foods 2011, 3:313-320. 183. Yunoki K, Renaguli M, Kinoshita M, Matsuyama H, Mawatari S, Fujino T, Kodama Y, Sugiyama M, Ohnishi M: Dietary sphingolipids ameliorate disorders of lipid metabolism in Zucker fatty rats. J Agric Food Chem 2010, 58:7030-7035.

184. Wat E, Tandy S, Kapera E, Kamili A, Chung RW, Brown A, Rowney M, Cohn JS: Dietary phospholipid-rich dairy milk extract reduces hepatomegaly, hepatic steatosis and hyperlipidemia in mice fed a high-fat diet. Atherosclerosis 2009, 205:144-150. 185. Norris GH, Porter CM, Jiang C, Millar CL, Blesso CN: Dietary sphingomyelin attenuates hepatic steatosis and adipose tissue inflammation in high fat diet-induced obese mice. J Nutr Biochem 2016. 186. Chung RW, Kamili A, Tandy S, Weir JM, Gaire R, Wong G, Meikle PJ, Cohn JS, Rye KA: Dietary sphingomyelin lowers hepatic lipid levels and inhibits intestinal cholesterol absorption in high-fat-fed mice. PLoS One 2013, 8:e55949. 187. Lecomte M, Couedelo L, Meugnier E, Plaisancie P, Letisse M, Benoit B, Gabert L, Penhoat A, Durand A, Pineau G, et al: Dietary emulsifiers from milk and soybean differently impact adiposity and inflammation in association with modulation of colonic goblet cells in high-fat fed mice. Mol Nutr Food Res 2016, 60:609-620. 188. Chung RWS, Wang Z, Bursill CA, Wu BJ, Barter PJ, Rye K-A: Effect of long-term dietary sphingomyelin supplementation on atherosclerosis in mice. PLOS ONE 2017, 12:e0189523. 189. Li Z, Basterr MJ, Hailemariam TK, Hojjati MR, Lu S, Liu J, Liu R, Zhou H, Jiang XC: The effect of dietary sphingolipids on plasma sphingomyelin metabolism and atherosclerosis. Biochim Biophys Acta 2005, 1735:130-134. 190. Ramprasath VR, Jones PJ, Buckley DD, Woollett LA, Heubi JE: Effect of dietary sphingomyelin on absorption and fractional synthetic rate of cholesterol and serum lipid profile in humans. Lipids Health Dis 2013, 12:125. 191. Snel M, Sleddering MA, Pijl H, Nieuwenhuizen WF, Frolich M, Havekes LM, Romijn JA, Jazet IM: The effect of dietary phytosphingosine on cholesterol levels and insulin sensitivity in subjects with the metabolic syndrome. Eur J Clin Nutr 2010, 64:419-423. 192. Rosqvist F, Smedman A, Lindmark-Mansson H, Paulsson M, Petrus P, Straniero S, Rudling M, Dahlman I, Riserus U: Potential role of milk fat globule membrane in modulating plasma lipoproteins, gene expression, and cholesterol metabolism in humans: a randomized study. Am J Clin Nutr 2015, 102:20-30. 193. Ohlsson L, Burling H, Nilsson A: Long term effects on human plasma lipoproteins of a formulation enriched in butter milk polar lipid. Lipids Health Dis 2009, 8:44. 194. Conway V, Couture P, Richard C, Gauthier SF, Pouliot Y, Lamarche B: Impact of buttermilk consumption on plasma lipids and surrogate markers of cholesterol homeostasis in men and women. Nutr Metab Cardiovasc Dis 2013, 23:1255-1262. 195. Murakami I, Mitsutake S, Kobayashi N, Matsuda J, Suzuki A, Shigyo T, Igarashi Y: Improved high-fat diet-induced glucose intolerance by an oral administration of phytosphingosine. Biosci Biotechnol Biochem 2013, 77:194-197. 196. Kamili A, Wat E, Chung RW, Tandy S, Weir JM, Meikle PJ, Cohn JS: Hepatic accumulation of intestinal cholesterol is decreased and fecal cholesterol excretion is increased in mice fed a high-fat diet supplemented with milk phospholipids. Nutr Metab (Lond) 2010, 7:90. 197. Reis MG, Roy NC, Bermingham EN, Ryan L, Bibiloni R, Young W, Krause L, Berger B, North M, Stelwagen K, Reis MM: Impact of dietary dairy polar lipids on lipid metabolism of mice fed a high-fat diet. J Agric Food Chem 2013, 61:2729-2738.

198. Park WJ, Park JW, Merrill AH, Storch J, Pewzner-Jung Y, Futerman AH: Hepatic fatty acid uptake is regulated by the sphingolipid acyl chain length. Biochim Biophys Acta 2014, 1841:1754-1766. 199. Li Y, Dong J, Ding T, Kuo MS, Cao G, Jiang XC, Li Z: Sphingomyelin synthase 2 activity and liver steatosis: an effect of ceramide-mediated peroxisome proliferator-activated receptor gamma2 suppression. Arterioscler Thromb Vasc Biol 2013, 33:1513-1520. 200. Hannun YA, Obeid LM: Many Ceramides. Journal of Biological Chemistry 2011, 286:27855-27862. 201. Park JW, Park WJ, Kuperman Y, Boura-Halfon S, Pewzner-Jung Y, Futerman AH: Ablation of very long acyl chain sphingolipids causes hepatic insulin resistance in mice due to altered detergent-resistant membranes. Hepatology 2013, 57:525-532. 202. Raichur S, Wang ST, Chan PW, Li Y, Ching J, Chaurasia B, Dogra S, Ohman MK, Takeda K, Sugii S, et al: CerS2 haploinsufficiency inhibits beta-oxidation and confers susceptibility to diet-induced steatohepatitis and insulin resistance. Cell Metab 2014, 20:687-695. 203. Hla T, Kolesnick R: C16:0-ceramide signals insulin resistance. Cell Metab 2014, 20:703-705. 204. Turpin SM, Nicholls HT, Willmes DM, Mourier A, Brodesser S, Wunderlich CM, Mauer J, Xu E, Hammerschmidt P, Bronneke HS, et al: Obesity-induced CerS6-dependent C16:0 ceramide production promotes weight gain and glucose intolerance. Cell Metab 2014, 20:678- 686. 205. Cinar R, Godlewski G, Liu J, Tam J, Jourdan T, Mukhopadhyay B, Harvey-White J, Kunos G: Hepatic cannabinoid-1 receptors mediate diet-induced insulin resistance by increasing de novo synthesis of long-chain ceramides. Hepatology 2014, 59:143-153. 206. Jozefowski S, Czerkies M, Lukasik A, Bielawska A, Bielawski J, Kwiatkowska K, Sobota A: Ceramide and ceramide 1-phosphate are negative regulators of TNF-alpha production induced by lipopolysaccharide. J Immunol 2010, 185:6960-6973. 207. Parham KA, Zebol JR, Tooley KL, Sun WY, Moldenhauer LM, Cockshell MP, Gliddon BL, Moretti PA, Tigyi G, Pitson SM, Bonder CS: Sphingosine 1-phosphate is a ligand for peroxisome proliferator-activated receptor-gamma that regulates neoangiogenesis. Faseb j 2015, 29:3638-3653. 208. Mazzei JC, Zhou H, Brayfield BP, Hontecillas R, Bassaganya-Riera J, Schmelz EM: Suppression of intestinal inflammation and inflammation-driven colon cancer in mice by dietary sphingomyelin: importance of peroxisome proliferator-activated receptor gamma expression. J Nutr Biochem 2011, 22:1160-1171. 209. Martin H: Role of PPAR-gamma in inflammation. Prospects for therapeutic intervention by food components. Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis 2010, 690:57-63. 210. Wurfel MM, Wright SD: Lipopolysaccharide-binding protein and soluble CD14 transfer lipopolysaccharide to phospholipid bilayers: preferential interaction with particular classes of lipid. J Immunol 1997, 158:3925-3934. 211. Chiba N, Masuda A, Yoshikai Y, Matsuguchi T: Ceramide inhibits LPS-induced production of IL-5, IL-10, and IL-13 from mast cells. Journal of Cellular Physiology 2007, 213:126-136. 212. Hsu YW, Chi KH, Huang WC, Lin WW: Ceramide inhibits lipopolysaccharide- mediated nitric oxide synthase and cyclooxygenase-2 induction in macrophages: effects on protein kinases and transcription factors. J Immunol 2001, 166:5388-5397.

213. Schnabl KL, Larsen B, Van Aerde JE, Lees G, Evans M, Belosevic M, Field C, Thomson AB, Clandinin MT: Gangliosides protect bowel in an infant model of necrotizing enterocolitis by suppressing proinflammatory signals. J Pediatr Gastroenterol Nutr 2009, 49:382-392. 214. Dalbeth N, Gracey E, Pool B, Callon K, McQueen FM, Cornish J, MacGibbon A, Palmano K: Identification of dairy fractions with anti-inflammatory properties in models of acute gout. Ann Rheum Dis 2010, 69:766-769. 215. Rauchhaus M, Coats AJ, Anker SD: The endotoxin-lipoprotein hypothesis. Lancet 2000, 356:930-933. 216. Memon RA, Holleran WM, Moser AH, Seki T, Uchida Y, Fuller J, Shigenaga JK, Grunfeld C, Feingold KR: Endotoxin and cytokines increase hepatic sphingolipid biosynthesis and produce lipoproteins enriched in ceramides and sphingomyelin. Arterioscler Thromb Vasc Biol 1998, 18:1257-1265. 217. Hannun YA, Bell RM: Regulation of protein kinase C by sphingosine and lysosphingolipids. Clin Chim Acta 1989, 185:333-345. 218. Foey AD, Brennan FM: Conventional protein kinase C and atypical protein kinase Czeta differentially regulate macrophage production of tumour necrosis factor-alpha and interleukin-10. Immunology 2004, 112:44-53. 219. Sakamoto H, Yoshida T, Sanaki T, Shigaki S, Morita H, Oyama M, Mitsui M, Tanaka Y, Nakano T, Mitsutake S, et al: Possible roles of long-chain sphingomyelines and sphingomyelin synthase 2 in mouse macrophage inflammatory response. Biochem Biophys Res Commun 2016. 220. Fischer H, Ellstrom P, Ekstrom K, Gustafsson L, Gustafsson M, Svanborg C: Ceramide as a TLR4 agonist; a putative signalling intermediate between sphingolipid receptors for microbial ligands and TLR4. Cell Microbiol 2007, 9:1239-1251. 221. Jersmann HPA: Time to abandon dogma: CD14 is expressed by non-myeloid lineage cells. Immunol Cell Biol 2005, 83:462-467. 222. Pfeiffer A, Bottcher A, Orso E, Kapinsky M, Nagy P, Bodnar A, Spreitzer I, Liebisch G, Drobnik W, Gempel K, et al: Lipopolysaccharide and ceramide docking to CD14 provokes ligand-specific receptor clustering in rafts. Eur J Immunol 2001, 31:3153-3164. 223. Medvedev AE, Blanco JCG, Qureshi N, Vogel SN: Limited Role of Ceramide in Lipopolysaccharide-mediated Mitogen-activated Protein Kinase Activation, Transcription Factor Induction, and Cytokine Release. Journal of Biological Chemistry 1999, 274:9342-9350. 224. MacKichan ML, DeFranco AL: Role of Ceramide in Lipopolysaccharide (LPS)- induced Signaling: LPS INCREASES CERAMIDE RATHER THAN ACTING AS A STRUCTURAL HOMOLOG. Journal of Biological Chemistry 1999, 274:1767-1775. 225. Park EJ, Suh M, Thomson B, Ma DW, Ramanujam K, Thomson AB, Clandinin MT: Dietary ganglioside inhibits acute inflammatory signals in intestinal mucosa and blood induced by systemic inflammation of Escherichia coli lipopolysaccharide. Shock 2007, 28:112-117. 226. Park EJ, Thomson AB, Clandinin MT: Protection of intestinal occludin tight junction protein by dietary gangliosides in lipopolysaccharide-induced acute inflammation. J Pediatr Gastroenterol Nutr 2010, 50:321-328. 227. Snow DR, Ward RE, Olsen A, Jimenez-Flores R, Hintze KJ: Membrane-rich milk fat diet provides protection against gastrointestinal leakiness in mice treated with lipopolysaccharide. J Dairy Sci 2011, 94:2201-2212.

228. Klop B, Proctor SD, Mamo JC, Botham KM, Castro Cabezas M: Understanding Postprandial Inflammation and Its Relationship to Lifestyle Behaviour and Metabolic Diseases. International Journal of Vascular Medicine 2012, 2012:11. 229. Demmer E, Van Loan MD, Rivera N, Rogers TS, Gertz ER, German JB, Smilowitz JT, Zivkovic AM: Addition of a dairy fraction rich in milk fat globule membrane to a high- saturated fat meal reduces the postprandial insulinaemic and inflammatory response in overweight and obese adults. J Nutr Sci 2016, 5:e14. 230. Witkowska AM: Soluble ICAM-1: a marker of vascular inflammation and lifestyle. Cytokine 2005, 31:127-134. 231. Rogers TS, Demmer E, Rivera N, Gertz ER, German JB, Smilowitz JT, Zivkovic AM, Van Loan MD: The role of a dairy fraction rich in milk fat globule membrane in the suppression of postprandial inflammatory markers and bone turnover in obese and overweight adults: an exploratory study. Nutr Metab (Lond) 2017, 14:36. 232. Furuya, Ohkawara, Nagashima, Asanuma, Hino: Dietary Sphingomyelin Alleviates Experimental Inflammatory Bowel Disease in Mice. International Journal for Vitamin and Nutrition Research 2008, 78:41-48. 233. Yamashita S, Sakurai R, Hishiki K, Aida K, Kinoshita M: Effects of Dietary Plant-origin Glucosylceramide on Colon Cytokine Contents in DMH-treated Mice. J Oleo Sci 2017, 66:157-160. 234. Arai K, Mizobuchi Y, Tokuji Y, Aida K, Yamashita S, Ohnishi M, Kinoshita M: Effects of Dietary Plant-Origin Glucosylceramide on Bowel Inflammation in DSS-Treated Mice. J Oleo Sci 2015, 64:737-742. 235. Fischbeck A, Leucht K, Frey-Wagner I, Bentz S, Pesch T, Kellermeier S, Krebs M, Fried M, Rogler G, Hausmann M, Humpf HU: Sphingomyelin induces cathepsin D-mediated apoptosis in intestinal epithelial cells and increases inflammation in DSS colitis. Gut 2011, 60:55-65. 236. Motouri M, Matsuyama H, Yamamura J, Tanaka M, Aoe S, Iwanaga T, Kawakami H: Milk sphingomyelin accelerates enzymatic and morphological maturation of the intestine in artificially reared rats. J Pediatr Gastroenterol Nutr 2003, 36:241-247. 237. Blaas N, Schuurmann C, Bartke N, Stahl B, Humpf HU: Structural profiling and quantification of sphingomyelin in human breast milk by HPLC-MS/MS. J Agric Food Chem 2011, 59:6018-6024. 238. Moschetta A, vanBerge-Henegouwen GP, Portincasa P, Palasciano G, Groen AK, van Erpecum KJ: Sphingomyelin exhibits greatly enhanced protection compared with egg yolk phosphatidylcholine against detergent bile salts. J Lipid Res 2000, 41:916-924. 239. Stenman LK, Holma R, Eggert A, Korpela R: A novel mechanism for gut barrier dysfunction by dietary fat: epithelial disruption by hydrophobic bile acids. Am J Physiol Gastrointest Liver Physiol 2013, 304:G227-234. 240. Lecomte M, Couedelo L, Meugnier E, Plaisancie P, Letisse M, Benoit B, Gabert L, Penhoat A, Durand A, Pineau G, et al: Dietary emulsifiers from milk and soybean differently impact adiposity and inflammation in association with modulation of colonic goblet cells in high-fat fed mice. Mol Nutr Food Res 2015. 241. Lecomte M, Bourlieu C, Meugnier E, Penhoat A, Cheillan D, Pineau G, Loizon E, Trauchessec M, Claude M, Menard O, et al: Milk Polar Lipids Affect In Vitro Digestive Lipolysis and Postprandial Lipid Metabolism in Mice. J Nutr 2015, 145:1770-1777.

242. Mu C, Yang Y, Su Y, Zoetendal EG, Zhu W: Differences in Microbiota Membership along the Gastrointestinal Tract of Piglets and Their Differential Alterations Following an Early-Life Antibiotic Intervention. Frontiers in Microbiology 2017, 8:797. 243. Sprong RC, Hulstein MF, Van der Meer R: Bactericidal activities of milk lipids. Antimicrob Agents Chemother 2001, 45:1298-1301. 244. Ten Bruggencate SJ, Frederiksen PD, Pedersen SM, Floris-Vollenbroek EG, Lucas-van de Bos E, van Hoffen E, Wejse PL: Dietary Milk-Fat-Globule Membrane Affects Resistance to Diarrheagenic Escherichia coli in Healthy Adults in a Randomized, Placebo-Controlled, Double-Blind Study. J Nutr 2016, 146:249-255. 245. Rueda R, Sabatel JL, Maldonado J, Molina-Font JA, Gil A: Addition of gangliosides to an adapted milk formula modifies levels of fecal Escherichia coli in preterm newborn infants. J Pediatr 1998, 133:90-94. 246. Martin MJ, Martin-Sosa S, Alonso JM, Hueso P: Enterotoxigenic Escherichia coli strains bind bovine milk gangliosides in a ceramide-dependent process. Lipids 2003, 38:761- 768. 247. Wang Z, Klipfell E, Bennett BJ, Koeth R, Levison BS, Dugar B, Feldstein AE, Britt EB, Fu X, Chung YM, et al: Gut flora metabolism of phosphatidylcholine promotes cardiovascular disease. Nature 2011, 472:57-63.

Chapter 3. Dietary Milk Sphingomyelin Improves Lipid Metabolism and Alters Gut Microbiota in High Fat Diet-fed C57BL/6J Mice

This chapter is reprinted from “Norris GH, Jiang C, Ryan J, Porter CM, Blesso CN: Milk sphingomyelin improves lipid metabolism and alters gut microbiota in high fat diet-fed mice. J Nutr Biochem 2016, 30:93-101.” As the author of this Elsevier article, Gregory Norris retains the right to include this article in a dissertation, permission from the publisher is not required.

3.1. Abstract

High dietary fat intake can cause elevated serum and hepatic lipids, as well as contribute to gut dysbiosis, intestinal barrier dysfunction and increased circulating lipopolysaccharide (LPS).

Dietary milk sphingomyelin (SM) has been shown to inhibit lipid absorption in rodents. We evaluated the effects of milk SM on lipid metabolism and LPS levels in C57BL/6J mice fed a high- fat diet for 4 weeks and compared it with egg SM. Mice were fed a high-fat diet (45% kcal from fat) (CTL, n = 10) or the same diet modified to contain 0.25% (wt/wt) milk SM (MSM, n = 10) or

0.25% (wt/wt) egg SM (ESM, n = 10). After 4 weeks, MSM had gained significantly less weight and had reduced serum cholesterol compared to CTL. ESM had increases in serum cholesterol, triglycerides, phospholipids and SM compared to CTL. MSM significantly decreased, while ESM increased, hepatic triglycerides. This may have been related to induction of hepatic stearoyl-CoA desaturase-1 mRNA observed in ESM. MSM displayed intestinal and hepatic gene expression changes consistent with cholesterol depletion. MSM had significantly lower serum LPS compared to CTL, which may have been due to altered distal gut microbiota. Fecal Gram-negative bacteria were significantly lower, while fecal Bifidobacterium were higher, in MSM. These results suggest that milk SM is more effective than egg SM at combating the detrimental effects of a high-fat diet in mice. Additionally, distal gut microbiota is altered with milk SM and this may have contributed to the lower serum LPS observed.

3.2. Introduction

High dietary fat is thought to contribute to obesity incidence due to its high energy density and therefore higher risk of overconsumption [1]. Compared with low-fat diets, feeding mice high-fat diets or “Western”-type diets (40–60% kcal as fat) ad libitum results in development of obesity and diabetes, characterized by increased adiposity, hepatic steatosis, systemic inflammation and

71 insulin resistance [2]. In addition, high-fat diets contribute to elevated serum LPS concentrations

(also known as endotoxin), which may precede the associated increases in systemic inflammation, insulin resistance and hepatic steatosis [3,4]. Dietary fat may increase appearance of gut-derived

LPS in circulation by enhancing transcellular transport via chylomicrons [5,6] and/or paracellular leakage across the gut barrier [7]. High-fat-diet feeding of mice has also been shown to markedly alter the gut microbiota [8,9]. Gut dysbiosis has been linked with endotoxin-mediated hepatic steatosis in mouse models [10]. Detrimental effects of high-fat diet in mice appear to be influenced by gut microbes, since administration of broad-spectrum antibiotics attenuated the elevation in plasma LPS and improved metabolic parameters [8]. Along with excessive dietary fat, higher cholesterol absorption is linked with deleterious health effects. Cholesterol absorption is widely recognized to influence fasting plasma lipids and risk for cardiovascular disease [11]. Besides the well-known effects on plasma lipids, dietary cholesterol exacerbates inflammation in adipose tissue [12] and liver [13,14] in mouse models of obesity. Therefore, an important strategy to manage complications of obesity would be to reduce lipid absorption as a means to improve complications such as dyslipidemia, hepatic steatosis and low-grade inflammation.

Dietary sphingomyelin (SM) and other sphingolipids have been shown to dose- dependently reduce the absorption of cholesterol, triglycerides and fatty acids in rodents [15–18].

SM consists of a ceramide (a fatty acid linked to a long-chain sphingoid base through an amide linkage) with a phosphorylcholine head group. Digestion of SM in the intestine is slow and incomplete but yields ceramide and sphingosine that can be absorbed into intestinal mucosal cells [19]. Products of SM digestion, such as ceramide and sphingosine, also appear to reduce lipid absorption [18,20]. Sphingosine has been shown to exhibit antimicrobial properties against pathogenic bacteria, such as Escherichia coli and Salmonella enteritidis [21], suggesting that

72 dietary sphingolipids can alter the gut microbiota, along with lipid absorption. The estimated dietary intakes of sphingolipids is 300–400 mg/day in Western diets, most of which comes from

SM found in dairy products, meats and eggs [22]. SM is abundant in animal cell membranes but is not found in plants or fungi. Egg SM and milk SM differ in their fatty acid and sphingoid base composition and, therefore, may have different biological effects on lipid metabolism in vivo. In rats, milk SM has been shown to be a more potent inhibitor of cholesterol and fatty acid absorption when compared to egg SM [16]. Although milk SM has been shown repeatedly to inhibit lipid absorption in rodents [16,17,23], the effects of chronic milk SM intake on lipid metabolism in vivo have not been studied extensively. While milk SM and digestion products can impact intestinal luminal interactions and absorption of cholesterol, the sphingolipids themselves may demonstrate bioactivity and affect lipid metabolism after absorption. Furthermore, whether milk

SM displays a more potent hypolipidemic effect than egg SM is not known. A comparison of dietary milk SM and egg SM on lipid metabolism has not yet been examined in vivo.

Therefore, we investigated the effects of dietary milk SM on lipid metabolism in C57BL/6J mice fed a high-fat diet and compared it with egg SM feeding. We hypothesized that milk SM would improve lipid metabolism, while also being more potent than egg SM at preventing the elevations in serum and hepatic lipids that occur in mice fed a high-fat diet. We also hypothesized that the addition of dietary milk SM would result in lower circulating LPS compared with a high- fat control diet, by protecting the gut barrier from the damaging effects of absorbed dietary lipids and altering the gut microbiota.

3.3. Materials and Methods

3.3.1 Animals and Diets

Male C57BL/6J mice (8 weeks old, n = 30) were obtained from the Jackson Laboratory (Bar

Harbor, ME) and allowed to acclimate to the animal facility for 2 weeks before being fed 1 of 3 anhydrous milk-fat-based, high-fat diets for 4 weeks: high-fat diet control (CTL; 45% kcal as fat; n = 10), high-fat diet with 0.25% of milk SM added by weight (MSM; 45% kcal as fat; n = 10) and high-fat diet with 0.25% of egg SM added by weight (ESM; 45% kcal as fat; n = 10). Detailed composition of the diets is listed in Table 3.1. The ingredients for experimental diets were obtained from Dyets, Inc. (Bethlehem, PA). Milk SM (bovine) and egg SM (chicken egg) were both obtained at > 99% purity in powdered form from Avanti Polar Lipids, Inc. (Alabaster, AL) and were substituted into test diets for an equal amount of anhydrous milk fat. The fatty acid distribution of the SM obtained varied between milk SM (34% C23:0, 21% C24:0, 20% C22:0,

16% C16:0, 3% C24:1 and 6% unknown) and egg SM (86% C16:0, 6% C18:0, 3% C22:0, 3%

C24:1 and 2% unknown). The sphingoid base composition has also been reported to be different between egg SM and milk SM, with egg SM containing almost entirely sphingosine (d18:1) bases, whereas milk SM has a more varied distribution (d16:0 to d19:0) [24]. The anhydrous milk fat used in high-fat diets generally contains phospholipid concentrations of ~ 0.2–1% (wt/wt) [25,26],

~ 20–25% is SM [25]. Therefore, the overall contribution of SM to the control high-fat diet is minimal (0.008–0.0525%, wt/wt) compared to what is found in the diets with added SM (0.258–

0.3025%, wt/wt). The amount of dietary SM provided by the 0.25% SM diets is equivalent to

22 mg/kg of body weight in a human or roughly 1.5 g/day to a 70-kg human, based on body surface normalization [27].

Table 3.1. Diet composition

g/kg of diet

CTL MSM ESM Casein 195 195 195 L-Cystine 3 3 3 Sucrose 340 340 340 Corn Starch 56.86 56.86 56.86 Maltodextrin 60 60 60 Anhydrous Milk Fat 210 207.5 207.5 Soybean Oil 20 20 20 Cellulose 50 50 50 Mineral Mix, AIN-93G-MX 43 43 43 Vitamin Mix, AIN-93-VX 19 19 19 Choline Bitartrate 3 3 3 TBHQ 0.04 0.04 0.04 Milk Sphingomyelin 0 2.5 0 Egg Sphingomyelin 0 0 2.5

Food intake was assessed biweekly and body weight was assessed weekly. Fresh food was provided on a weekly basis. After 4 weeks on experimental diets, mice were fasted for 6–8 h prior to blood collection by cardiac puncture following euthanasia. Blood was allowed to clot at room temperature for 30 min before serum was isolated by centrifugation (10,000gfor 10 min at 4°C) and then stored at − 80°C. Tissues were perfused with saline before being harvested, snap-frozen in liquid nitrogen and stored at − 80°C. The liver, skeletal muscle (quadriceps femoris), epididymal adipose and small intestines were collected from animals. The entire small intestine from the pylorus to the cecum was collected, cleaned and divided into four segments (SI-I, SI-II, SI-III and

SI-IV) before being snap-frozen. During the last 3 days prior to sacrifice, fecal samples were

75 collected directly from the mouse into sterile tubes and stored frozen at − 20°C until analysis. All mice were housed in a temperature-controlled room and maintained in a 12-h light/12-h dark cycle at the University of Connecticut-Storrs vivarium. The Animal Care and Use Committee of the

University of Connecticut-Storrs approved all procedures used in the current study.

3.3.2. Serum Biochemical Analysis

Total serum cholesterol, triglycerides, phospholipids, SM, glucose and nonesterified fatty acids

(NEFA) were measured using enzymatic assays according to manufacturer instructions.

Cholesterol, triglycerides, NEFA and choline-containing phospholipids (choline oxidase method) were measured using kits obtained from Wako Diagnostics (Richmond, VA). Serum SM was analyzed using an enzymatic kit obtained from Abcam (Cambridge, MA). Serum phosphatidylcholine (PC) was estimated by subtracting serum SM values from the serum choline- containing phospholipid values. Glucose was measured using a hexokinase kit obtained from

Pointe Scientific, Inc. (Canton, MI). Serum LPS was assessed by using a chromogenic limulus amebocyte lysate (LAL) assay (QCL-1000) obtained from Lonza (Basel, Switzerland).

3.3.3. Hepatic Lipid Extraction and Analysis

Hepatic lipids were extracted using methods previously reported [28]. Briefly, liver lipids were extracted with chloroform:methanol (2:1), dried under nitrogen at 60°C and solubilized in Triton

X-100. The solubilized lipid isolates were analyzed for total cholesterol and triglycerides by enzymatic methods.

3.3.4. RNA Isolation, cDNA Synthesis and qRT-PCR

Total RNA was isolated using TRIzol reagent (Life Technologies, Carlsbad, CA) from liver, skeletal muscle, proximal small intestine (SI-I and SI-II segments) and distal small intestine (SI-

IV segment). Total RNA was treated with DNase I and reverse transcribed using iScript cDNA synthesis kit (Bio-Rad, Hercules, CA). Gene expression was assessed by quantitative real-time polymerase chain reaction (qRT-PCR) using iTaq Universal SYBR Green Supermix (Bio-Rad) and a CFX96 real-time PCR detection system (Bio-Rad). Gene expression was normalized to the geometric mean of the reference genes, glyceraldehyde 3-phosphate dehydrogenase, β-actin and ribosomal protein, large, P0 using the 2− ΔΔCt method. Primer sequences used for qRT-PCR analysis are listed in Appendix A.

3.3.5. Gut Permeability Analysis

Intestinal permeability to macromolecules was examined using a low-molecular-weight (4-kDa) fluorescent dextran (Sigma-Aldrich, St. Louis, MO). After 4 weeks on experimental diets, mice were fasted for 6 h and administered fluorescein isothiocyanate (FITC)-dextran by gavage

(600 mg/kg body weight, 125 μg/μl). After 1 h, ~ 100-μl blood was collected from the tail vein.

Plasma was isolated, diluted 1:2 in PBS (pH 7.4) and analyzed for FITC-dextran on a fluorescence spectrophotometer (BioTek Instruments, Inc. Winooski, VT) at excitation 485 nm and emission

535 nm. A standard curve was generated by diluting FITC-dextran in untreated, diluted mouse plasma.

3.3.6. Gut Microbiota Analysis

Fecal samples were collected directly from separately housed mice and submitted to the University of Connecticut-Storrs Microbial Analysis, Resources and Services facility for microbiota characterization utilizing 16S V4 analysis. DNA was extracted from 0.25 g of fecal sample using the MoBio PowerMag Soil 96-well kit (MoBio Laboratories, Inc., Carlsbad, CA) according to the manufacturer's protocol for the Eppendorf epMotion liquid handling robot. DNA extracts were quantified using the Quant-iT PicoGreen kit (Thermo Fisher Scientific, Waltham, MA). Partial

77 bacterial 16S rRNA genes (V4) were amplified using 30 ng extracted DNA as template. The V4 region was amplified using 515F and 806R with Illumina adapters and golay indices on the 3′ end [29]. Samples were amplified in triplicate using Phusion High-Fidelity PCR master mix (New

England BioLabs, Ipswich, MA) with the addition of 10 μg BSA (New England BioLabs). PCR products were pooled for quantification and visualization using the QIAxcel DNA Fast Analysis

(Qiagen, Venlo, Netherlands). PCR products were normalized based on the concentration of DNA from 250 to 400 bp then pooled using the QIAgility liquid handling robot. The pooled PCR products were cleaned using the Gene Read Size Selection kit (Qiagen) according to the manufacturer's protocol. The cleaned pool was sequenced on the MiSeq using v2 2 × 250 basepair kit.

Demultiplexing of the sequences was done with a custom script [30] requiring 0 mismatches in the index sequences and Q25 minimum. Sequences were merged using SeqPrep then filtered for length (maximum 300 bp) before clustering into OTUs with QIIME (for exact commands, see https://github.com/mcnelsonphd/16S-RDS/blob/master/Qiime_Process).

3.3.7. Statistical Analysis

Differences between groups were evaluated by one-way ANOVA with post hoc comparisons

(Holm-Sidak) or by independent t tests (P <.05 deemed significant). Bivariate Pearson correlations were used to assess relationships between variables. All statistical analyses were conducted using

GraphPad Prism version 6 software. Data are reported as mean ± SEM.

3.4. Results

3.4.1. Body and Tissue Weights

Body and tissue weights are shown in Table 1. No significant changes in food intake, week 4 body weight or liver weight were observed between the groups. However, MSM reduced the total

78 amount of weight gained over the 4 weeks. ESM experienced an increase in epididymal adipose both in absolute weight and as a percentage of body weight compared with both CTL and MSM groups.

3.4.2. MSM Reduced Cholesterol, While ESM Increased Serum Lipids

Fasting serum lipids and glucose values are presented in Table 3.2. MSM had significantly lower serum cholesterol (−15%) compared to CTL. In contrast, ESM had significantly higher serum cholesterol (+15%) and serum triglycerides (+30%) compared to CTL. Serum NEFA were 23% lower in both SM groups compared to CTL (P =.05). ESM had significantly elevated serum choline-containing phospholipids compared to CTL (+ 19%) and MSM groups (+ 31%), which was primarily accounted for by higher serum SM levels

Table 3.2. Body and tissue weights of C57BL/6J mice after 4 weeks of high-fat diets Parameter Control Egg SM Milk SM Body weight (g) 29.47 ± 0.77 28.53 ± 0.68 27.26 ± 0.74 Weight change (g) 5.14 ± 0.58a 4.48 ± 0.34ab 3.35 ± 0. 25b Food intake (g/day) 3.80 ± 0.20 4.08 ± 0.09 4.06 ± 0.10 Liver weight (g) 1.19 ± 0.05 1.25 ± 0.04 1.12 ± 0.05 % Liver weight 4.07 ± 0.20 4.14 ± 0.13 4.09 ± 0.12 Epididymal fat (g) 0.62 ± 0.11a 0.88 ± 0.07b 0.43 ± 0.04a % Epididymal weight 2.05 ± 0.32a 3.07 ± 0.22b 1.44 ± 0.10a

Each value represents the mean ± SEM of the values for n = 10 mice. Superscripts with different letters indicate differences at P <.05 using post hoc comparisons.

Table 3.3. Serum markers of C57BL/6J mice after 4 weeks of high-fat diets Parameter Control Egg SM Milk SM Total cholesterol (mmol/L) 3.47 ± 0.12a 4.03 ± 0.18b 2.96 ± 0.17c Triglycerides (mmol/L) 0.61 ± 0.04a 0.80 ± 0.05b 0.64 ± 0.05ab NEFA (mmol/L) 0.68 ± 0.05 0.49 ± 0.03# 0.49 ± 0.04# Total phospholipids (mmol/L) 2.36 ± 0.10a 2.85 ± 0.10b 2.25 ± 0.13a SM (mmol/L) 1.15 ± 0.09a 1.55 ± 0.10b 1.23 ± 0.05a PC (mmol/L) 1.22 ± 0.09 1.29 ± 0.13 0.93 ± 0.08 SM:PC ratio 0.845 ± 0.07 1.37 ± 0.19 1.31 ± 0.17 Fasting glucose (mmol/L) 8.02 ± 0.66 8.59 ± 0.64 7.31 ± 0.46

Each value represents the mean ± SEM of the values for n = 10 mice. Superscripts with different letters indicate differences at P <.05 using post hoc comparisons. #P =.05 vs. control using post hoc comparisons.

(+25–35%). Serum SM was positively associated with serum cholesterol across groups

(r = 0.623, p =.0002; Figure 3.1A), whereas serum PC was not correlated with serum cholesterol

(r =− 0.011, P =.956; Figure 3.1B). No significant differences were found for fasting glucose levels between groups.

Figure 3.1. Serum cholesterol is correlated with serum SM but not serum PC. Bivariate Pearson correlations between serum cholesterol and serum SM (A) or serum PC (B) are shown

3.4.3. MSM Alters Intestinal mRNA Expression of Cholesterol Transporters

In the proximal small intestine, MSM significantly increased Niemann-Pick C1-like 1 (Npc1l1) mRNA compared to ESM (+110%) (Figure 3.2). Abcg5 mRNA expression was reduced by 42% in MSM compared to CTL. Both Npc1l1 and Abcg5 are major transporters of free cholesterol on the apical surface of epithelial cells. No differences were detected in the expression of genes related to chylomicron formation or triglyceride absorption (Figure 3.2). There were also no differences between groups in gene expression of tight junction-related proteins involved in gut barrier function in the distal intestine (Figure 3.2).

Figure 3.2. MSM significantly altered gene expression related with intestinal cholesterol flux. Small intestine mRNA was measured by qRT-PCR. Data were normalized to endogenous reference gene expression (n = 10 per group, mean ± SEM). ABCG5, ATP-binding cassette subfamily G member 5; APOA4, apolipoprotein A-IV; DGAT2, diacylglycerol O-acyltransferase 2; FABP2, fatty-acid-binding protein 2, intestinal; MTTP, microsomal triglyceride transfer protein; OCLN, occludin; TJP1, tight junction protein 1. Values with different superscripts are P <.05 using post hoc comparisons.

3.4.4. MSM Reduced, While ESM Increased Hepatic Lipids

MSM reduced hepatic triglyceride accumulation by 47% compared with CTL, while ESM increased triglycerides by 37% (Figure 3.3A). No significant differences were observed for liver cholesterol between groups. MSM had significant inductions in both hepatic Srebp2 mRNA

(+50%) and HMG-CoA reductase (Hmgcr) mRNA (+190%) compared with CTL (Figure 3.3B), which may have impacted liver cholesterol via compensatory induction in cholesterol biosynthesis. Further within cholesterol-regulated genes, both SM groups tended to lower hepatic

IDOL expression compared with CTL, but these changes were not significant (p = 0.1). ESM increased stearoyl-CoA desaturase-1 (Scd1) mRNA expression by 3-fold compared to CTL. Scd1 is an enzyme involved in de novo lipogenesis. No other differences between groups were detected for mRNA expression of genes related to lipid metabolism (Srebp1c, Lxrα, Acox and Pparα), lipoprotein uptake (Ldlr and Pcsk9) or inflammation (Mcp1). These data show that MSM induced the hepatic expression of a majority of the cholesterol-responsive genes examined, while ESM induced Scd1 gene expression, which may explain the elevated hepatic triglycerides in this group.

3.4.5. SM Increases Skeletal Muscle GLUT4 mRNA Expression with No Effects on Fatty Acid Oxidation-related Gene Expression

Skeletal muscle (quadriceps femoris) was examined for mRNA expression of genes involved in lipid metabolism, inflammation and glucose disposal (Figure 3.4). Both SM-fed groups experienced a 3-fold increase in Glut4 mRNA. Interleukin-6 (Il-6) mRNA expression was 70% lower in MSM compared to CTL. There were no other differences observed between groups for other gene expression related to fatty acid oxidation (Acad, Acox, Pgc-1α and Lpl).

3.4.6. MSM Reduce Serum LPS but Does Not Change Gut Permeability to FITC-dextran

High-fat-diet feeding of mice has been shown to increase plasma LPS after only 4 weeks [3]. Since milk SM feeding lowered serum and liver lipids, we examined its effects on serum LPS. MSM reduced serum LPS by 35% compared with CTL, while no difference was observed with ESM

Figure 3.3. MSM reduces hepatic triglycerides and alters cholesterol-responsive gene expression. Hepatic lipids were extracted with chloroform:methanol (2:1), dried under nitrogen at 60°C and solubilized in Triton X-100 as described in Materials and Methods. Cholesterol and triglyceride were measured by enzymatic methods (A) (n = 10 per group, mean ± SEM). Hepatic mRNA expression was measured by qRT-PCR (B). Data were normalized to endogenous reference gene expression (n = 10 per group, mean ± SEM). ACOX, acyl-CoA oxidase 1; IDOL, inducible degrader of the LDL receptor; LDLr, LDL receptor; LXRα, liver X receptor α; MCP1, monocyte chemotactic protein 1; PCSK9, proprotein convertase subtilisin/kexin type 9; PPARa, peroxisome proliferator-activated receptor α; SREBP1c, sterol regulatory element-binding protein 1c; SREBP2, sterol regulatory element-binding protein 2. Values with different superscripts are P <.05 using post hoc comparison

(Figure 3.5A). Serum LPS was significantly associated with the amount of weight gain over the

4 weeks (Figure 3.5B). We examined effects of milk SM on gut permeability in fasting mice by the appearance of FITC-dextran (4 kDa) in plasma after gavage. Gut permeability to FITC-dextran was 5% lower on average in MSM (Figure 3.5C); however, this was not significantly different compared to

average in MSM (Figure 3.5C); however, this was not significantly different compared to CTL.

Therefore, coinciding with the lack of differences in small intestine tight junction- related gene expression (Fig. 1), it appeared that differences in serum LPS were not due to damage to the gut barrier. Since sphingosine has been shown to display bactericidal activity against several Gram- negative strains [21], we next examined if there were differences in distal gut microbiota between

MSM and CTL.

3.4.7. MSM Alters Distal Gut Microbiota Phylogenetic Abundance and Gram-negative Bacteria

Fecal microbiota composition from separately caged mice was examined by 16S rRNA sequence analysis and results are presented in Figure 3.6. Comparing the major bacterial phyla (Figure

3.6B), MSM had significantly higher phylogenetic relative abundance of the predominately Gram- positive Firmicutes phylum (p<0.05) and significantly lower of the Gram-negative Bacteroidetes phylum (p <0.05). Comparing the minor bacterial phyla (Figure 3.6C), MSM had significantly higher relative abundance of the Gram-positive Actinobacteria phylum (p <0.001) and less of the

Gram-negative Tenericutes phylum (p <0.001). The relative abundance of the Gram-negative

Proteobacteria phylum did not differ between MSM and CTL groups. LPS is a component of the outer membrane of the cell wall of Gram-negative bacteria. When classified based on Gram staining (Figure 3.6D), MSM had significantly lower relative abundance of Gram-negative bacteria and a reciprocal increase in the predominately Gram-positive bacteria (p <0.05). These

Figure 3.4. MSM increases skeletal muscle GLUT4 and decreases IL-6 mRNA. Skeletal muscle mRNA (quadriceps femoris) was measured by qRT-PCR. Data were normalized to endogenous reference gene expression (n = 10 per group, mean ± SEM). ACAD, acyl-CoA dehydrogenase; ACOX, acyl-CoA oxidase 1; CD68, cluster of differentiation 68; GLUT4, glucose transporter type 4; LPL, lipoprotein lipase; PGC-1α, peroxisome proliferator- activated receptor gamma coactivator 1α. Values with different superscripts are P <.05 using post hoc comparisons.

data suggest that the lower serum LPS levels may be due to a reduction in LPS-containing Gram- negative bacteria in the gut microbiota. At the genus level (Figure 3.6E), MSM had significantly higher relative abundance of the beneficial bacteria, Bifidobacterium (p <0.001), although no difference in relative abundance of Lactobacillus was observed between groups. Interestingly,

MSM tended to have higher relative abundance of Bacteroides (p = 0.06), one of the few microbes that synthesize and utilize sphingolipids [31], which was absent in CTL.

Figure 3.5. MSM reduces serum LPS but does not influence gut permeability. Serum LPS was measured as described in Materials and Methods using an endpoint chromogenic LAL assay (A) (n = 10 per group, mean ± SEM). Values with different superscripts are P <.05 using post hoccomparisons. A bivariate Pearson correlation between serum LPS and 4-week weight change is shown (B). Mice (n = 6 per group, mean ± SEM) were fasted for 6 h prior to a gavage of 4 kDa FITC-dextran (600 mg/kg body weight, 125 μg/μl) as described in Materials and Methods. Gut permeability to FITC-dextran in fasted mice was measured in plasma by fluorometric methods (C)

3.5. Discussion

High-fat diet has been shown to induce dyslipidemia [32] and gut dysbiosis [33], which are both associated with the development of diet-induced obesity. Sphingolipids are bioactives found in food, which have been shown to reduce lipid absorption in rodents [15–18] and have potential to alter gut microbiota [21]. In this study, C57BL/6J mice were fed SM derived from milk or egg for

4 weeks and then studied for effects on lipid metabolism. Surprisingly, egg SM increased several serum lipids and hepatic triglycerides. In contrast, milk SM was shown to reduce serum cholesterol, serum LPS and hepatic triglycerides. Serum LPS can be influenced by differences in gut microbiota [8], chylomicron production [6] and differences in gut permeability [7]. In this study, milk SM was shown to alter the gut microbiota, although no differences were found for in vivo gut permeability in the fasted state. This study suggests that milk SM can ameliorate the negative metabolic effects of a high-fat diet in mice.

SM and its hydrolytic byproducts have been shown to reduce the intestinal absorption of cholesterol, triglycerides and fatty acids in rodents [15–18,20]. SM interacts with high affinity to cholesterol and is thought to influence its micellar solubilization. Eckhardt et al. [17] observed that feeding mice diets with milk SM at 0.1%, 0.5% or 5% (wt/wt) reduced intestinal cholesterol absorption measured by dual-isotope method by approximately 20%, 54% and 86%, respectively.

Intestinal hydrolysis of SM is initiated by alkaline sphingomyelinase that hydrolyzes the phosphorylcholine head, resulting in ceramide [34]. Alkaline sphingomyelinase is inhibited by acylglycerides (monoglycerides, diglycerides and triglycerides), free fatty acids, cholesterol and ceramide [35], all of which are highly present in the duodenum and proximal jejunum where the majority of lipid absorption occurs. Feng et al. [36] showed that ceramide and sphingosine can also inhibit the absorption of cholesterol by Caco-2 cells. Ceramide is a major signal for apoptosis

87 and is shown to be beneficial in reducing chemically induced colorectal cancer [37]. Neutral ceramidase functions through hydrolysis of the fatty acid ester linkage of ceramide, releasing the sphingoid backbone, which is predominantly sphingosine [38]. Sphingosine may act as an antimicrobial [21] and is absorbed in higher quantities than ceramide and SM [39]. As SM moves down the intestinal tract, the activity of sphingomyelinase and ceramidase will increase due the depletion of luminal lipids. This would likely result in the cecal and colonic bacteria being exposed to appreciable levels of sphingosine and ceramide.

In the current study, MSM decreased (−15%) total serum cholesterol compared to the high-fat control. In contrast, ESM increased cholesterol (+15%), triglycerides (+30%), SM (+ 35%) and total phospholipids

(+20%). Serum lipids in mice fed dietary sphingolipids has been shown to decrease [18], not change [40], and even increase in some studies [41]. Li et al. [41] reported that feeding LDLr−/− mice for 12 weeks a modified AIN-76A diet supplemented with 1% (wt/wt) ox brain sphingolipids (28% SM) increased plasma cholesterol (+ 62%), plasma SM (+50%) and aortic atherosclerosis. Recently, Chung et al. [40] reported that egg SM feeding (0.6%, wt/wt) for 18 days reduced hepatic lipid levels and increased fecal cholesterol output in mice fed a high-fat diet (21% by weight butter fat; 0.15% by weight added cholesterol). Chung et al. [40] also reported that egg SM feeding led to an ~ 30% reduction in cholesterol absorption, although it did not affect plasma cholesterol or triglyceride levels. Duivenvoorden et al. [18] observed significant reductions in plasma cholesterol and triglycerides when 0.2–0.4% egg SM (wt/wt) was added to a

Western-type diet (15% by weight cocoa butter; 0.25% by weight cholesterol added)and fed to

APOE3*Leiden female mice for 3 weeks. Overall, studies in mice where large amounts of cholesterol were added to diets in addition to sphingolipids displayed reductions in serum cholesterol and hepatic lipids, whereas those that did not observed elevations in serum and hepatic lipids (including the current study).

Therefore, it is possible that the weaker effect of egg SM on lipid absorption compared with milk SM [16]

Figure 3.6. MSM group distal gut microbiota has significantly reduced Gram-negative bacteria. Fecal samples were collected directly from separately housed mice and microbiota composition was assessed by 16S rRNA sequencing as described in Materials and Methods. Phyla phylogenetic abundance (% of total sequences) of individual mice (A). Phylogenetic relative abundance of major phyla Firmicutes and Bacteroidetes (B), and minor phyla Actinobacteria, Tenericutes and Proteobacteria (C). Relative abundance of Gram- negative and Gram-positive phyla (D). Relative abundance of Bifidobacterium, Lactobacillus and Bacteroides genera (n = 3 per group, mean ± SEM, *P <.05 using t test).

may require excess dietary cholesterol to influence serum cholesterol and hepatic lipids. Another explanation for discrepancies in serum and hepatic cholesterol responses in the studies discussed may be due to the different strains of mice used (i.e., wild type or transgenic). Recently,

Ramprasath et al. [42] reported that 4-week supplementation of milk SM (1 g/day) in healthy humans significantly increased plasma HDL-C in a small crossover study (n = 10), although no effects on non-HDL-cholesterol and cholesterol absorption were found. Lack of effects of milk

SM on cholesterol absorption in this human study may be related to greater SM digestion efficiency in humans compared with rodents [19] or potentially related to the healthy population examined.

The differences in serum cholesterol between groups were supported by changes in cholesterol-regulated gene expression in the proximal intestine. Intestinal NPC1L1 mRNA, which was increased in MSM (+110% vs. ESM), is responsible for cholesterol uptake at the apical membrane of intestinal epithelial cells. Intestinal NPC1L1 mRNA has been shown to be reduced by excessive cellular cholesterol and induced by cholesterol depletion [43]. ABCG5, which was reduced in MSM (-42% vs. CTL), constitutes half of the ABCG5/8 heterodimer that enables free cholesterol and phytosterol efflux from the intestinal cell back into the lumen [44]. The observed changes in the gene expression of these transporters in MSM-fed mice suggest that the intestines were compensating for cellular cholesterol depletion, likely due to lower cholesterol absorption.

Noh and Koo [15,16] reported that dietary egg SM and milk SM could both acutely reduce intestinal lipid absorption in rodents, with milk SM being a more potent inhibitor [16].

Inhibition of intestinal cholesterol absorption through ezetimibe, a pharmacological inhibitor of NPC1L1, can lower serum cholesterol levels and reduce hepatic lipids in mice [45].

The current study demonstrated significantly lower triglycerides in MSM livers (−47%) and higher

90 in ESM (+37%), compared to CTL. Hepatic cholesterol biosynthesis may have increased with SM treatment, suggested by the 190% increase in HMGCR mRNA in MSM and the increasing trend

(p=0.07) in ESM. Additionally, hepatic mRNA expression of SREBP2, a transcription factor that induces genes for cholesterol biosynthesis, was increased in MSM (+54% vs. ESM, 51% vs. CTL).

In this study, compensatory increases in cholesterol biosynthesis potentially masked differences in hepatic cholesterol levels between groups. Interestingly, egg SM increased hepatic SCD1 mRNA (+189% vs. CTL), which was not seen with milk SM feeding. SCD1 is involved in de novo lipogenesis, enhancing hepatic triglyceride formation [32].

Skeletal muscle GLUT4 expression increased in ESM (+172%) and MSM (+ 162%) compared to CTL. Skeletal muscle GLUT4 expression is regulated by the transcription factor myocyte enhancer factor 2 (MEF2) [46]. PGC-1α serves as a coactivator of MEF2 transcriptional activity and has been shown to participate in an autoregulatory loop where it induces PGC-1α expression in skeletal muscle [47]. Milk SM (0.2% by weight) fed for 12 weeks to BALB/c mice was reported to increase endurance capacity, associated with induction of skeletal muscle PGC-1α mRNA [48]. We did not observe a change in PGC-1α at the mRNA level, which may be due to differences in background diet, duration or strain of mice. There is also a possibility that PGC-1α coactivator activity was altered, which we did not measure. Further studies examining the effects of dietary SM on skeletal muscle lipid and glucose metabolism are warranted.

In the current study, milk SM reduced serum cholesterol and NEFA (p =0.05), suggesting a reduction in dietary lipid absorption. Dietary lipid absorption may increase the translocation of gut-derived LPS by enhancing transcellular transport via chylomicrons secretion [6] and/or paracellular leakage across the gut barrier [7]. Based on these relationships, it is not surprising that we observed 35% lower serum LPS in MSM compared to CTL. This difference did not appear to

91 be due to differences in fasting gut permeability. LPS is a major constituent of the outer membrane of cell walls of Gram-negative bacteria. Sphingosine has been shown to have strong bactericidal effects on Gram-negative and Gram-positive bacterial pathogens in vitro [21]. As dietary SM is slowly digested and absorbed, it is likely that much of it reaches the colon where bacteria are most abundant. Remarkably, dietary milk SM strongly altered fecal phyla composition, with significantly reduced relative abundance of Gram-negative phyla, such as Bacteroidetes and

Tenericutes, and reciprocal increases in predominately Gram-positive phyla, such as Firmicutes and Actinobacteria. Interestingly, one study reported that nonobese mice fed a high-fat diet, which are diabetes-resistant, have lower plasma LPS and gut permeability and increased cecal Firmicutes and Actinobacteria than diabetes-prone mice [49]. Notably, milk-SM-fed mice had significantly greater relative abundance of fecal Bifidobacterium, a genus of mutualistic “beneficial” bacteria.

Fecal Bifidobacterium are known to be dramatically reduced in mice fed high-fat diets, and selective increases are associated with reduced plasma LPS and improvements in metabolic parameters in mice [50]. Bifidobacteria can protect against enteropathogens [51] and are well known to ferment prebiotic oligosaccharides and increase the production of short-chain fatty acids

(SCFA) in the colon [52]. In addition to bifidobacteria, milk-SM-fed mice tended to have higher relative abundance of fecal Bacteroides (p =0.06), one of the few microbes that synthesize and utilize sphingolipids [31]. Species of the Bacteroides genus are also known to ferment prebiotics to SCFA [52], with some species shown to prevent metabolic complications of obese mice induced with a high-fat diet when used as a probiotic [53].

The current study has several strengths. We used purified, natural sources of milk and egg

SM (> 99% pure), as opposed to a mixed phospholipid concentrate or milk globule membrane extract, which have been studied previously [54,55]. Thus, using purified SM allows us to attribute

92 the effects observed on lipid metabolism and gut microbiota directly to the additional dietary SM ingestion. We also examined effects of dietary SM on markers of metabolic endotoxemia and gut microbiota, which to our knowledge have never been reported. Although egg SM did not reduce serum LPS, we did not examine effects of egg SM on gut permeability and gut microbiota, which could be a limitation of the current study. The dose chosen for this study roughly equates to an amount of SM (1.5 g/day) which is above the estimated sphingolipid intake (300–400 mg/day) of

Western diets. This high dosage may also be a limitation. However, it is possible that this dose could be achievable through supplementation or through careful selection of SM-rich foods.

Overall, we observed significant differences in the effects of milk SM and egg SM on lipid metabolism in mice fed a high-fat diet. Milk SM reduced serum cholesterol and altered gene expression consistent with intestinal and hepatic cholesterol depletion. Milk SM also reduced hepatic triglycerides and induced gene expression associated with cholesterol biosynthesis.

Unexpectedly, egg SM increased several serum lipids and hepatic triglycerides. A possible explanation for the increase in lipids in egg-SM-fed mice was increased hepatic de novo lipogenesis; however, the mechanism is not yet clear. Both SM groups demonstrated a strong induction of GLUT4 mRNA in skeletal muscle, suggesting an increase in glucose utilization. Milk

SM also significantly lowered serum LPS and was associated with bifidogenic effects and alterations in distal gut microbiota. In conclusion, milk SM appears to have beneficial effects in protecting against the deleterious effects of a high-fat diet in mice. Further research is warranted to examine the mechanisms of action of milk SM on serum LPS and alterations in gut microbiota.

3.6. References

1. Schrauwen P, Westerterp KR: The role of high-fat diets and physical activity in the regulation of body weight. Br J Nutr 2000, 84:417-427. 2. Hariri N, Thibault L: High-fat diet-induced obesity in animal models. Nutr Res Rev 2010, 23:270-299. 3. Cani PD, Amar J, Iglesias MA, Poggi M, Knauf C, Bastelica D, Neyrinck AM, Fava F, Tuohy KM, Chabo C, et al: Metabolic endotoxemia initiates obesity and insulin resistance. Diabetes 2007, 56:1761-1772. 4. Ding S, Lund PK: Role of intestinal inflammation as an early event in obesity and insulin resistance. Curr Opin Clin Nutr Metab Care 2011, 14:328-333. 5. Neal MD, Leaphart C, Levy R, Prince J, Billiar TR, Watkins S, Li J, Cetin S, Ford H, Schreiber A, Hackam DJ: Enterocyte TLR4 mediates phagocytosis and translocation of bacteria across the intestinal barrier. J Immunol 2006, 176:3070-3079. 6. Ghoshal S, Witta J, Zhong J, de Villiers W, Eckhardt E: Chylomicrons promote intestinal absorption of lipopolysaccharides. J Lipid Res 2009, 50:90-97. 7. Brun P, Castagliuolo I, Di Leo V, Buda A, Pinzani M, Palu G, Martines D: Increased intestinal permeability in obese mice: new evidence in the pathogenesis of nonalcoholic steatohepatitis. Am J Physiol Gastrointest Liver Physiol 2007, 292:G518-525. 8. Cani PD, Bibiloni R, Knauf C, Waget A, Neyrinck AM, Delzenne NM, Burcelin R: Changes in gut microbiota control metabolic endotoxemia-induced inflammation in high-fat diet-induced obesity and diabetes in mice. Diabetes 2008, 57:1470-1481. 9. Murphy EF, Cotter PD, Healy S, Marques TM, O'Sullivan O, Fouhy F, Clarke SF, O'Toole PW, Quigley EM, Stanton C, et al: Composition and energy harvesting capacity of the gut microbiota: relationship to diet, obesity and time in mouse models. Gut 2010, 59:1635-1642. 10. Henao-Mejia J, Elinav E, Jin C, Hao L, Mehal WZ, Strowig T, Thaiss CA, Kau AL, Eisenbarth SC, Jurczak MJ, et al: Inflammasome-mediated dysbiosis regulates progression of NAFLD and obesity. Nature 2012, 482:179-185. 11. Howell WH, McNamara DJ, Tosca MA, Smith BT, Gaines JA: Plasma lipid and lipoprotein responses to dietary fat and cholesterol: a meta-analysis. Am J Clin Nutr 1997, 65:1747-1764. 12. Subramanian S, Han CY, Chiba T, McMillen TS, Wang SA, Haw A, 3rd, Kirk EA, O'Brien KD, Chait A: Dietary cholesterol worsens adipose tissue macrophage accumulation and atherosclerosis in obese LDL receptor-deficient mice. Arterioscler Thromb Vasc Biol 2008, 28:685-691. 13. Subramanian S, Goodspeed L, Wang S, Kim J, Zeng L, Ioannou GN, Haigh WG, Yeh MM, Kowdley KV, O'Brien KD, et al: Dietary cholesterol exacerbates hepatic steatosis and inflammation in obese LDL receptor-deficient mice. J Lipid Res 2011, 52:1626-1635. 14. Wouters K, van Gorp PJ, Bieghs V, Gijbels MJ, Duimel H, Lutjohann D, Kerksiek A, van Kruchten R, Maeda N, Staels B, et al: Dietary cholesterol, rather than liver steatosis, leads to hepatic inflammation in hyperlipidemic mouse models of nonalcoholic steatohepatitis. Hepatology 2008, 48:474-486. 15. Noh SK, Koo SI: Egg sphingomyelin lowers the lymphatic absorption of cholesterol and alpha-tocopherol in rats. J Nutr 2003, 133:3571-3576. 16. Noh SK, Koo SI: Milk sphingomyelin is more effective than egg sphingomyelin in inhibiting intestinal absorption of cholesterol and fat in rats. J Nutr 2004, 134:2611-2616.

17. Eckhardt ER, Wang DQ, Donovan JM, Carey MC: Dietary sphingomyelin suppresses intestinal cholesterol absorption by decreasing thermodynamic activity of cholesterol monomers. Gastroenterology 2002, 122:948-956. 18. Duivenvoorden I, Voshol PJ, Rensen PC, van Duyvenvoorde W, Romijn JA, Emeis JJ, Havekes LM, Nieuwenhuizen WF: Dietary sphingolipids lower plasma cholesterol and triacylglycerol and prevent liver steatosis in APOE*3Leiden mice. Am J Clin Nutr 2006, 84:312-321. 19. Ohlsson L, Hertervig E, Jonsson BA, Duan RD, Nyberg L, Svernlov R, Nilsson A: Sphingolipids in human ileostomy content after meals containing milk sphingomyelin. Am J Clin Nutr 2010, 91:672-678. 20. Garmy N, Taieb N, Yahi N, Fantini J: Interaction of cholesterol with sphingosine: physicochemical characterization and impact on intestinal absorption. J Lipid Res 2005, 46:36-45. 21. Sprong RC, Hulstein MF, Van der Meer R: Bactericidal activities of milk lipids. Antimicrob Agents Chemother 2001, 45:1298-1301. 22. Vesper H, Schmelz EM, Nikolova-Karakashian MN, Dillehay DL, Lynch DV, Merrill AH, Jr.: Sphingolipids in food and the emerging importance of sphingolipids to nutrition. J Nutr 1999, 129:1239-1250. 23. Nyberg L, Duan RD, Nilsson A: A mutual inhibitory effect on absorption of sphingomyelin and cholesterol. J Nutr Biochem 2000, 11:244-249. 24. Ramstedt B, Leppimaki P, Axberg M, Slotte JP: Analysis of natural and synthetic sphingomyelins using high-performance thin-layer chromatography. Eur J Biochem 1999, 266:997-1002. 25. Jensen RG: The composition of bovine milk lipids: January 1995 to December 2000. J Dairy Sci 2002, 85:295-350. 26. Fedotova Y, Lencki RW: The Effect of Phospholipids on Milkfat Crystallization Behavior. Journal of the American Oil Chemists' Society 2008, 85:205-212. 27. Osawa Y, Uchinami H, Bielawski J, Schwabe RF, Hannun YA, Brenner DA: Roles for C16-ceramide and Sphingosine 1-Phosphate in Regulating Hepatocyte Apoptosis in Response to Tumor Necrosis Factor-α. Journal of Biological Chemistry 2005, 280:27879- 27887. 28. Farrell N, Norris G, Lee SG, Chun OK, Blesso CN: Anthocyanin-rich black elderberry extract improves markers of HDL function and reduces aortic cholesterol in hyperlipidemic mice. Food Funct 2015, 6:1278-1287. 29. Caporaso JG, Lauber CL, Walters WA, Berg-Lyons D, Huntley J, Fierer N, Owens SM, Betley J, Fraser L, Bauer M, et al: Ultra-high-throughput microbial community analysis on the Illumina HiSeq and MiSeq platforms. ISME J 2012, 6:1621-1624. 30. Nelson MC, Morrison HG, Benjamino J, Grim SL, Graf J: Analysis, optimization and verification of Illumina-generated 16S rRNA gene amplicon surveys. PLoS One 2014, 9:e94249. 31. An D, Na C, Bielawski J, Hannun YA, Kasper DL: Membrane sphingolipids as essential molecular signals for Bacteroides survival in the intestine. Proc Natl Acad Sci U S A 2011, 108 Suppl 1:4666-4671. 32. Sampath H, Miyazaki M, Dobrzyn A, Ntambi JM: Stearoyl-CoA desaturase-1 mediates the pro-lipogenic effects of dietary saturated fat. J Biol Chem 2007, 282:2483-2493.

33. Murphy EA, Velazquez KT, Herbert KM: Influence of high-fat diet on gut microbiota: a driving force for chronic disease risk. Curr Opin Clin Nutr Metab Care 2015, 18:515-520. 34. Duan RD, Hertervig E, Nyberg L, Hauge T, Sternby B, Lillienau J, Farooqi A, Nilsson A: Distribution of alkaline sphingomyelinase activity in human beings and animals. Tissue and species differences. Dig Dis Sci 1996, 41:1801-1806. 35. Liu JJ, Nilsson A, Duan RD: In vitro effects of fat, FA, and cholesterol on sphingomyelin hydrolysis induced by rat intestinal alkaline sphingomyelinase. Lipids 2002, 37:469-474. 36. Feng D, Ohlsson L, Ling W, Nilsson A, Duan RD: Generating ceramide from sphingomyelin by alkaline sphingomyelinase in the gut enhances sphingomyelin-induced inhibition of cholesterol uptake in Caco-2 cells. Dig Dis Sci 2010, 55:3377-3383. 37. Garcia-Barros M, Coant N, Truman JP, Snider AJ, Hannun YA: Sphingolipids in colon cancer. Biochim Biophys Acta 2014, 1841:773-782. 38. Olsson M, Duan RD, Ohlsson L, Nilsson A: Rat intestinal ceramidase: purification, properties, and physiological relevance. Am J Physiol Gastrointest Liver Physiol 2004, 287:G929-937. 39. Garmy N, Taieb N, Yahi N, Fantini J: Apical uptake and transepithelial transport of sphingosine monomers through intact human intestinal epithelial cells: physicochemical and molecular modeling studies. Arch Biochem Biophys 2005, 440:91-100. 40. Chung RW, Kamili A, Tandy S, Weir JM, Gaire R, Wong G, Meikle PJ, Cohn JS, Rye KA: Dietary sphingomyelin lowers hepatic lipid levels and inhibits intestinal cholesterol absorption in high-fat-fed mice. PLoS One 2013, 8:e55949. 41. Li Z, Basterr MJ, Hailemariam TK, Hojjati MR, Lu S, Liu J, Liu R, Zhou H, Jiang XC: The effect of dietary sphingolipids on plasma sphingomyelin metabolism and atherosclerosis. Biochim Biophys Acta 2005, 1735:130-134. 42. Ramprasath VR, Jones PJ, Buckley DD, Woollett LA, Heubi JE: Effect of dietary sphingomyelin on absorption and fractional synthetic rate of cholesterol and serum lipid profile in humans. Lipids Health Dis 2013, 12:125. 43. Alrefai WA, Annaba F, Sarwar Z, Dwivedi A, Saksena S, Singla A, Dudeja PK, Gill RK: Modulation of human Niemann-Pick C1-like 1 gene expression by sterol: Role of sterol regulatory element binding protein 2. Am J Physiol Gastrointest Liver Physiol 2007, 292:G369- 376. 44. Berge KE, Tian H, Graf GA, Yu L, Grishin NV, Schultz J, Kwiterovich P, Shan B, Barnes R, Hobbs HH: Accumulation of dietary cholesterol in sitosterolemia caused by mutations in adjacent ABC transporters. Science 2000, 290:1771-1775. 45. Umemoto T, Subramanian S, Ding Y, Goodspeed L, Wang S, Han CY, Teresa AS, Kim J, O'Brien KD, Chait A: Inhibition of intestinal cholesterol absorption decreases atherosclerosis but not adipose tissue inflammation. J Lipid Res 2012, 53:2380-2389. 46. Thai MV, Guruswamy S, Cao KT, Pessin JE, Olson AL: Myocyte enhancer factor 2 (MEF2)-binding site is required for GLUT4 gene expression in transgenic mice. Regulation of MEF2 DNA binding activity in insulin-deficient diabetes. J Biol Chem 1998, 273:14285- 14292. 47. Handschin C, Rhee J, Lin J, Tarr PT, Spiegelman BM: An autoregulatory loop controls peroxisome proliferator-activated receptor gamma coactivator 1alpha expression in muscle. Proc Natl Acad Sci U S A 2003, 100:7111-7116.

48. Haramizu S, Ota N, Otsuka A, Hashizume K, Sugita S, Hase T, Murase T, Shimotoyodome A: Dietary milk fat globule membrane improves endurance capacity in mice. Am J Physiol Regul Integr Comp Physiol 2014, 307:R1009-1017. 49. Serino M, Luche E, Gres S, Baylac A, Berge M, Cenac C, Waget A, Klopp P, Iacovoni J, Klopp C, et al: Metabolic adaptation to a high-fat diet is associated with a change in the gut microbiota. Gut 2012, 61:543-553. 50. Cani PD, Neyrinck AM, Fava F, Knauf C, Burcelin RG, Tuohy KM, Gibson GR, Delzenne NM: Selective increases of bifidobacteria in gut microflora improve high-fat-diet-induced diabetes in mice through a mechanism associated with endotoxaemia. Diabetologia 2007, 50:2374-2383. 51. Lievin V, Peiffer I, Hudault S, Rochat F, Brassart D, Neeser JR, Servin AL: Bifidobacterium strains from resident infant human gastrointestinal microflora exert antimicrobial activity. Gut 2000, 47:646-652. 52. Cummings JH, Macfarlane GT: Gastrointestinal effects of prebiotics. Br J Nutr 2002, 87 Suppl 2:S145-151. 53. Gauffin Cano P, Santacruz A, Moya A, Sanz Y: Bacteroides uniformis CECT 7771 ameliorates metabolic and immunological dysfunction in mice with high-fat-diet induced obesity. PLoS One 2012, 7:e41079. 54. Wat E, Tandy S, Kapera E, Kamili A, Chung RW, Brown A, Rowney M, Cohn JS: Dietary phospholipid-rich dairy milk extract reduces hepatomegaly, hepatic steatosis and hyperlipidemia in mice fed a high-fat diet. Atherosclerosis 2009, 205:144-150. 55. Zhou AL, Hintze KJ, Jimenez-Flores R, Ward RE: Dietary fat composition influences tissue lipid profile and gene expression in Fischer-344 rats. Lipids 2012, 47:1119-1130.

Chapter 4. Dietary Sphingomyelin Attenuates Hepatic Steatosis and Adipose Tissue Inflammation in High-fat-diet- induced Obese C57BL/6J Mice

This chapter is reprinted from “Norris GH, Porter CM, Jiang C, Millar CL, Blesso CN: Dietary sphingomyelin attenuates hepatic steatosis and adipose tissue inflammation in high-fat-diet- induced obese mice. J Nutr Biochem 2017, 40:36-43.” As the author of this Elsevier article, Gregory Norris retains the right to include this article in a dissertation, permission from the publisher is not required

4.1. Abstract

Western-type diets can induce obesity and related conditions such as dyslipidemia, insulin resistance, and hepatic steatosis. We evaluated the effects of milk sphingomyelin (SM) and egg

SM on diet-induced obesity, the development of hepatic steatosis, and adipose inflammation in

C57BL/6J mice fed a high fat, cholesterol-enriched diet for 10 weeks. Mice were fed either a low- fat diet (10% kcal from fat) (LFD, n = 10), a high fat diet (60% kcal from fat) (HFD, n = 14), or a high fat diet modified to contain either 0.1% (w/w) milk SM (HFMSM, n = 14) or 0.1% (w/w) egg

SM (HFESM, n = 14). After 10 weeks, egg SM ameliorated weight gain, hypercholesterolemia, and hyperglycemia induced by HFD. Both egg SM and milk SM attenuated hepatic steatosis development, with significantly lower hepatic triglycerides (TG) and cholesterol relative to HFD.

This reduction in hepatic steatosis was stronger with egg SM supplementation relative to milk SM.

Reductions in hepatic TG observed with dietary SM were associated with lower hepatic mRNA expression of PPARγ-related genes; Scd1 and Pparg2 in both SM groups, and Cd36 and Fabp4 with egg SM. Egg SM, and to a lesser extent milk SM, reduced inflammation and markers of macrophage infiltration in adipose tissue. Egg SM also reduced skeletal muscle TG content compared to HFD. Overall, the current study provides evidence of dietary SM improving metabolic complications associated with diet-induced obesity in mice. Further research is warranted to understand the differences in bioactivity observed between egg and milk SM.

4.2. Introduction

The National Center for Health Statistics reported that 36.5% of American adults were considered obese from 2011-2014 [1]. Obesity can contribute to the development of several non- communicable diseases, including non-alcoholic fatty liver disease (NAFLD) and type 2 diabetes mellitus (T2DM). Concurrent with rises in obesity and an aging population, the prevalence of

100

T2DM and NAFLD have increased dramatically in recent decades [2, 3]. Obesity, particularly abdominal obesity, is often accompanied by chronic low-grade inflammation in both adipose tissue and systemic circulation, which may result in metabolic abnormalities such as insulin resistance

[4, 5]. Localized adipose tissue inflammation is thought to arise from a dysfunction in the normal metabolic handling of nutrients by adipocytes, as a consequence of excessive adipocyte enlargement, insulin resistance, and infiltration of adipose tissue with macrophages [4, 5]. Insulin resistance and metabolic defects in lipid metabolism are associated with NAFLD, related to an increased flux of non-esterified fatty acids (NEFA) from inflamed adipose tissue to the liver [6].

NAFLD consists of a broad spectrum of conditions occurring in the absence of alcohol use, including steatosis, non-alcoholic steatohepatitis (NASH), and liver cirrhosis. Hepatic steatosis has been estimated to be present in over two-thirds of the obese population [7-9]. Simple steatosis is relatively benign; however, once steatosis has developed, the liver is “sensitized” to various inflammatory stimuli, which can precipitate NASH [10]. Consequently, NASH can progress to cirrhosis and may result in hepatocellular carcinoma or liver failure [11]. Currently, there is a lack of effective management options for NAFLD besides weight reduction, which can be difficult to maintain long-term.

Dietary lipids, such as cholesterol and triglycerides (TG), have been shown to exacerbate adipose tissue inflammation and NAFLD in animal models [12-16]. Dietary phospholipids, however, are a potential source of bioactive lipids which may improve cardiometabolic health [17].

Dietary sphingomyelin (SM), a sphingolipid found exclusively in animals, has been shown to dose- dependently reduce the absorption of cholesterol, TG, and fatty acids both in vitro [18, 19] and in rat models [20, 21]. Although dietary SM is often overlooked, it demonstrates potential to influence chronic diseases, such as obesity and NAFLD. Only a few studies have evaluated the

101 effects of chronic dietary SM intake on lipid metabolism in rodents [22-25]. Little is known about an optimal intake of SM and whether an increased dietary intake of SM is beneficial to health.

Consumption of dietary sphingolipids in Western diets is relatively common (0.3-0.4 g/day) and comes mainly from animal-based foods, including eggs and dairy [26, 27].

Previously, Chung et al. [22] reported that supplementation with egg-derived SM at 0.3%,

0.6%, and 1.2% (w/w) reduced hepatic TG in a dose-dependent manner in mice on a 4-week

Western-type diet. We have recently reported that dietary milk-derived SM (0.25% w/w) reduced both serum lipids and hepatic TG in mice on a 4-week high fat diet (HFD) [23]. However, since both studies were relatively short term, the efficacy of dietary SM in preventing diet-induced obesity, adipose tissue inflammation, and hepatic steatosis were not fully examined. Furthermore, egg SM and milk SM differ in their amide-linked fatty acid and sphingoid base compositions [28-

30], and therefore, may have different biological effects on lipid metabolism in vivo [21, 23]. Thus, the current study evaluated the effects of dietary SM from both egg and milk on hepatic steatosis and obesity in C57BL/6J mice fed a cholesterol-enriched HFD for 10 weeks. We hypothesized that both sources of dietary SM would attenuate hepatic steatosis and obesity development, due to their abilities to inhibit lipid absorption. Furthermore, we expected milk SM to be more effective than egg SM, based on its greater potency to inhibit lipid absorption in rodent models [21].

4.3. Materials and Methods

4.3.1 Animals and Diets

Male C57BL/6J mice (6 weeks old) were obtained from Jackson Laboratory (Bar Harbor, ME) and allowed to acclimate for 2 weeks before being fed 1 of 4 lard-based diets for 10 weeks: low fat diet control (LFD; 10% kcal from fat; n = 10); high fat, high cholesterol diet control (HFD;

60% kcal from fat, 0.15% cholesterol added by weight; n = 14); HFD modified to contain 0.1% of

102 milk SM added by weight (HFMSM; n = 14); or HFD with 0.1% of egg SM added by weight

(HFESM; 60% kcal from fat; n = 14). Diet composition details are provided in Table 4.1.

Experimental diets were made using ingredients obtained from Dyets, Inc. (Bethlehem, PA). Milk

SM (bovine; >99% purity) and egg SM (chicken egg; >99% purity) powders were obtained from

Avanti Polar Lipids, Inc. (Alabaster, AL) and were substituted for an equal weight of lard in their respective experimental groups. Fatty acid composition varied between the milk SM (34% C23:0,

21% C24:0, 20%, C22:0, 16% C16:0, 3% C24:1 and 6% unknown) and egg SM (86% C16:0, 6%

C18:0, 3%, C22:0, 3% C24:1 and 2% unknown) powders. Egg SM and milk SM also vary in the distribution of sphingoid bases; egg SM contains almost entirely sphingosine (d18:1) bases, while milk SM contains a more varied distribution (d16:0 to d19:0) [30]. Based on body surface normalization and accounting for weight gain throughout the study, the experimental diets provided the equivalent of consuming approximately 405-670 mg/day in a 70 kg human [31].

Fresh food was provided biweekly and mice were allowed to eat ad libitum. Body weights were recorded on a weekly basis, while food intake was estimated biweekly. After 10 weeks on their respective diets, mice were fasted for 6-8 hours prior to blood collection by cardiac puncture following euthanasia. Blood was allowed to clot at room temperature for 30 minutes before serum was isolated by centrifugation (10,000 x g for 10 minutes at 4°C) and then stored at -80°C. Tissues were perfused with sterile saline before being harvested, snap-frozen in liquid nitrogen and stored at -80˚C. The liver, skeletal muscle (quadriceps femoris), and epididymal adipose tissues were collected from animals. Pieces of liver were harvested from the left lateral lobe for histology and lipid analysis. Liver and epididymal adipose tissues were fixed in 10% neutral-buffered formalin for at least 48 hours prior to histological processing. Mice were housed in the University of

Connecticut-Storrs vivarium in a temperature-controlled room and maintained in a 12-hour

103 light/12-hour dark cycle. The Animal Care and Use Committee of the University of Connecticut-

Storrs approved all procedures used in the current study.

Table 4.1. Diet composition (g/kg) Diet Component HFMSM / LFD HFD HFESM Casein 210 265 265 L-Cystine 3 4 4 Corn Starch 467 0 0 Maltodextrin 100 0 0 Sucrose 90 253.5 253.5 Lard 20 310 309 Soybean Oil 20 30 30 Cellulose 37.15 64 64 Mineral Mix, AIN-93G-MX (94046) 35 48 48 Vitamin Mix, AIN-93-VX (94047) 15 21 21 Choline Bitartrate 2.75 3 3 Cholesterol 0 1.5 1.5 Milk or Egg Sphingomyelin 0 0 1

4.3.2. Serum Biochemical Analysis

Total serum cholesterol, TG, NEFA, liver enzymes, and glucose concentrations were measured using commercial assays according to manufacturer instructions. Serum cholesterol, NEFA, and

TG were measured using enzymatic kits obtained from Wako Diagnostics (Richmond, VA).

Fasting glucose, serum alanine transaminase (ALT), and aspartate transaminase (AST) were measured using kits obtained from Pointe Scientific, Inc. (Canton, MI). Serum insulin and C-C motif chemokine ligand 2 (CCL2) were measured by ultra-sensitive ELISA (Crystal Chem, Inc.,

Downers Grove, IL) and magnetic bead-based assay (EMD Millipore, Billerica, MA),

104 respectively. The homeostasis model assessment (HOMA-IR) equation was used to estimate insulin resistance based on fasting serum insulin and glucose concentrations [32].

4.3.3. Tissue Lipid Extraction and Analysis

Hepatic and skeletal muscle lipids were extracted using a modified Folch method as previously reported [33], with solubilization and enzymatic analysis of lipid extracts by the methods of Carr et al. [34]. Briefly, tissue lipids were extracted with chloroform:methanol (2:1), dried under nitrogen, and solubilized in Triton X-100. The solubilized lipid extracts were analyzed for total cholesterol (TC), free cholesterol (FC), choline-containing phospholipids (PL), and TG by enzymatic methods. Cholesteryl ester (CE) mass was calculated as (TC – FC) X 1.67 to account for the fatty acid moiety contribution to CE [35].

4.3.4. Tissue Histology

Formalin-fixed liver and epididymal adipose tissues were embedded in paraffin and cut into 5-μm sections prior to staining with hematoxylin and eosin (H&E). All histological procedures were conducted at the Connecticut Veterinary Medical Diagnostic Laboratory (Storrs, CT). The stained tissue sections were viewed under bright field microscopy at ×200 magnification and images were taken with an AxioCam ICc3 camera (Zeiss, Thornwood, NY, USA). The infiltration of macrophages into adipose tissue was assessed by the counting of crown-like structures (3 slides per animal) by a technician blinded to group assignment.

4.3.5. RNA Isolation, cDNA Synthesis, and Real Time qRT-PCR

Total RNA from liver, epididymal adipose, and skeletal muscle was isolated using TRIzol reagent

(Life Technologies, Carlsbad, CA). Total RNA was treated with DNase I, and reverse transcribed using iScript cDNA synthesis kit (Bio-Rad, Hercules, CA). Real-time qRT-PCR was performed

105 using iTaq Universal SYBR Green Supermix (Bio-Rad) on a CFX96 real-time PCR detection system (Bio-Rad). For liver, gene expression was normalized to the geometric mean of the reference genes, glyceraldehyde 3-phosphate dehydrogenase (Gapdh) and beta actin, using the 2-

ΔΔCt method. The geometric mean of Gapdh and ribosomal protein, large, P0 (36B4) was used as a reference gene control for both adipose tissue and skeletal muscle. Primer sequences used for qRT-PCR analysis are listed in Appendix A.

4.3.6. SDS-PAGE and Immunoblotting

Hepatic protein was isolated and analyzed by SDS-PAGE and immunoblotting as previously described [36]. Briefly, livers were homogenized in a tissue lysate RIPA buffer (Cell Signaling

Technologies, Beverly, MA) containing 2 mM phenylmethylsulfonyl fluoride (PMSF) (Sigma

Aldrich, St. Louis, MO). Total protein concentrations of the lysates were determined by bicinchoninic acid (BCA) assay (Cell Signaling Technologies). Sixty µg of total protein was mixed with loading buffer containing dithiothreitol (DTT) (BioRad) and loaded onto 4-20% SDS Stain-

Free gels (BioRad). Gels were activated using the Stain-Free protocol on a ChemiDoc XRS+ imager (BioRad). Protein was then transferred to a PVDF membrane (Thermo Fisher Scientific,

Waltham, MA) through a semi-dry protocol using a Trans-Blot Turbo System (BioRad). Stain-

Free blot images were captured and membranes were blocked for 1 hour at room temperature with blocking buffer (20 mM Tris base, 150 mM NaCl, 0.05% Tween-20, 5% w/v non-fat dry milk).

Primary rabbit anti-mouse PPARγ antibody was obtained from Santa Cruz Biotechnology (Dallas,

TX), diluted according to manufacturer’s recommendation, and incubated overnight at 4°C. After washing the membrane with TBS buffer (20 mM Tris base, 150 mM NaCl, 0.05% Tween-20), a

1:10,000 dilution of secondary HRP-anti-rabbit antibody (Thermo Fisher Scientific) was incubated with the membrane at room temperature for 1 hour. Pierce ECL Western Blotting Substrate

106

(Thermo Fisher Scientific) was used for the final imaging of the blot. PPARγ relative abundance was quantified after standardizing to total protein images obtained from the Stain-Free blot using

ImageLab Software V5.1 (BioRad).

4.3.7. Statistical Analysis

Differences between groups were evaluated by one-way ANOVA with post hoc comparisons

(Holm-Sidak). Bivariate Pearson correlations were used to assess relationships between variables.

All statistical analyses were conducted using GraphPad Prism version 6 software. Data are reported as mean ± SEM.

4.4. Results

4.4.1. Egg SM Reduced HFD-induced Weight Gain and Increases in Adiposity

Ten weeks of a high fat, high cholesterol diet induced significant gains in body weight, epididymal fat, and liver weight in the HFD control group compared to LFD (Figure 4.1). Egg SM supplementation attenuated these effects, significantly decreasing body weight (Figure 4.1B) and epididymal fat pad mass (Figure 4.1D) compared to HFD by 14% and 19%, respectively. Liver weight tended to be lower in HFESM compared to HFD, however, the weight change did not reach statistical significance (Figure 4.1D). No differences were observed in food intake (Figure 4.1C) between HFD groups.

4.4.2. Egg SM Reduced Serum Total Cholesterol and Fasting Glucose

Serum lipids and glycemic markers are presented in Table 4.2. Serum cholesterol (+135%), fasting glucose (+108%), fasting insulin (+340%), and HOMA-IR (+848%) were significantly increased with HFD relative to LFD. Supplementation of egg SM significantly reduced serum cholesterol (-

22%) and fasting glucose (-28%) compared to HFD. Serum NEFA were significantly reduced in

107

Figure 4.1. Egg SM reduces body weight and epididymal fat with no change in food intake. Body weight was assessed weekly (A, B). Food intake (C) was measured biweekly and daily food intake calculated. Wet tissue weights were taken at the time of sacrifice (D). Each value represents the mean + SEM for n = 10 mice for low fat diet control (LFD) and n = 13-14 mice each for high fat diet control (HFD), HFD + milk SM (HFMSM), and HFD + egg SM (HFESM). Mean values with unlike letters indicate differences at P < 0.05 using post hoc comparisons.

108 both dietary SM groups compared to HFD. No significant differences were detected in serum

TG, fasting insulin, or HOMA-IR between HFD groups.

Table 4.2. Serum markers of C57BL/6J mice after 10 weeks of diets.

Parameter LFD HFD HFMSM HFESM

Total Cholesterol b b c 1.94 ± 0.09a 4.56 ± 0.27 4.33 ± 0.33 3.54 ± 0.26 (mmol/L) Triglycerides 0.29 ± 0.03 0.29 ± 0.03 0.36 ± 0.05 0.38 ± 0.04 (mmol/L) NEFA (mmol/L) 0.44 ± 0.06a 0.46 ± 0.07a 0.28 ± 0.02b 0.31 ± 0.02b Fasting Glucose 8.86 ± 0.61a 18.48 ± 1.68b 16.03 ± 0.66bc 13.24 ± 0.73c (mmol/L) Insulin (ng/mL) 0.15 ± 0.01a 0.66 ± 0.09b 0.53 ± 0.06b 0.55 ± 0.10b

HOMA-IR 1.61 ± 0.22a 15.26 ± 3.40b 10.86 ± 1.88b 9.34 ± 1.79b Each value represents the mean ± SEM of the values for n = 10-14 mice. NEFA, non-esterified fatty acids; HOMA-IR, homeostasis model assessment-insulin resistance. Superscripts with different letters indicate differences at p < 0.05 using post-hoc comparisons.

4.4.3. Dietary SM Attenuated HFD-induced Hepatic Steatosis and Reduced PPARγ-related mRNA Expression

H&E stained livers showed extensive lipid droplet accumulation in the HFD group (Figure 4.2A).

Milk SM attenuated the accumulation of hepatic lipid droplets compared to HFD, while egg SM further reduced the hepatic steatosis observed with HFD. These observations were confirmed by liver lipid analysis; HFD significantly increased hepatic TG, TC, and CE compared to LFD

(Figure 4.2B). Egg SM strongly reduced hepatic TG (-60%) compared to HFD, while milk SM lowered hepatic TG to a lesser extent (-33%). Both dietary SM groups lowered hepatic TC and CE by ~25-30%. HFD increased serum ALT (+43%) compared to LFD, but this effect was not significantly altered by dietary SM (Figure 4.2C).

109

Figure 4.2. Dietary SM attenuates hepatic steatosis and reduces hepatic PPARγ-related gene expression. Liver sections were formalin-fixed, sectioned and H&E stained (A) for visualizing hepatic steatosis. Liver lipids were extracted using a modified Folch extraction and measured by enzymatic methods (B), (triglycerides, TG; total cholesterol, TC; cholesteryl ester, CE; free cholesterol, FC; choline-containing phospholipids, PL) (n = 10-14 per group, mean + SEM). Serum activities of hepatic enzymes (alanine transaminase, ALT; aspartate transaminase, AST) were determined using a kinetic assay (C) (n = 10-14 per group, mean + SEM). Hepatic mRNA expression was determined using real-time qRT-PCR and standardized to the geometric mean of Gapdh and β-actin reference genes (-ΔΔCt) using the 2 method (D) (n = 13-14 per group, mean + SEM). Mean values with unlike letters indicate differences at P < 0.05 using post hoc comparisons.

110

Compared to HFD, both egg SM (-54%) and milk SM (-33%) significantly reduced Pparg2 mRNA (Figure 4.2D). This corresponded with reductions in the expression of PPARγ target genes related to TG accumulation; Scd1 was reduced in both SM groups, while Cd36 and Fabp4 (p =

0.05) were reduced in egg SM. Egg SM also reduced Ccl2 mRNA expression compared to HFD

(p = 0.05). No other significant differences were detected in the hepatic expression of other lipogenic or gluconeogenic genes. Hepatic Pparg2, Scd1, and Cd36 mRNA expression were significantly associated with hepatic TG among all mice fed HFD (Figure 4.3). Thus, changes in the expression of these genes were associated with hepatic steatosis. However, there were no significant differences in PPARγ protein abundance between HFD-fed groups; milk SM tended to be higher, while egg SM tended to be lower than HFD control (Figure 4.4A,B,C). Total PPARγ protein abundance had no correlation with hepatic TG (Figure 4.4D). Thus, inconsistent findings with hepatic PPARγ protein across SM-fed groups suggests that differences in the expression of

PPARγ target genes may be related to altered PPARγ activity, or possibly due to other transcriptional regulators.

Figure 4.3. Hepatic PPARγ-related lipogenic gene expression positively correlates with hepatic TGs. Pearson correlations between hepatic Pparg2 (A), Scd1 (B) and Cd36 (C) mRNA with hepatic TGs for all mice fed high-fat diets (n = 41).

111

Figure 4.4. Dietary SM does not alter total PPARγ protein in liver. Hepatic PPARγ protein expression was visualized by immunoblotting. (A), and total protein was imaged from Stain-Free protein blot (B). Quantitative relative abundance was measured through densitometry analysis (C) after normalizing to Stain-Free total protein images. Each value represents the mean ± S.E.M. for n = 12 mice per group. Pearson correlation of total hepatic PPARγ protein with hepatic TGs for HFD, HFMSM and HFESM (D)

112

4.4.4. Dietary Egg SM, and to a Lesser Extent Milk SM, Reduces Epididymal Adipose Tissue Inflammation

Compared to LFD, the HFD control group showed a noticeable enlargement in adipocyte size and increased macrophage infiltration into epididymal adipose tissue, indicated by the number of crown-like structures (CLS) (Figure 4.5A,B). The number of CLS was markedly attenuated with egg SM feeding (-68%) but was unchanged with milk SM feeding (Figure 4.5B). However, both dietary SM fed groups significantly reduced serum CCL2 compared to HFD control (Figure4.

5C). Compared to HFD control, markers of macrophage infiltration were reduced with egg SM feeding (F4/80, Cd68, Cd11c) and to a lesser extent milk SM (F4/80) (Figure 4.5D). For inflammatory gene expression, egg SM reduced both Ccl2 and Tnfa mRNA, whereas milk SM only reduced Tnfa. Egg SM also increased the expression of GLUT4 and adiponectin (Adipoq) mRNA, genes that are regulated by PPARγ and CCAAT/enhancer-binding protein alpha

(C/EBPα). Correspondingly, egg SM significantly increased C/EBPα (Cebpa) mRNA and tended to increase PPARγ (Pparg) mRNA (p = 0.07). Overall, markers of adipose tissue inflammation were reduced with egg SM feeding and to a lesser extent with milk SM.

4.4.5. Dietary Egg SM Reduces Skeletal Muscle Lipid Accumulation Induced by HFD

Total lipids were extracted from skeletal muscle (quadriceps femoris) and examined for TG content. HFD significantly increased TG concentration of muscle, whereas only egg SM attenuated this effect (Figure 4.6A). Both dietary SM groups had significantly lower expression of β- oxidative genes in muscle compared to HFD (Figure 4.6B), whereas only egg SM reduced Cd36 and Acad expression. No differences were observed for GLUT4 and Lpl mRNA expression between HFD group.

113

Figure 4.5. Dietary SM attenuates inflammation in adipose tissue. Epididymal adipose tissue sections were formalin-fixed, sectioned and H&E stained (A). Crown-like structures (CLS) were manually counted from adipose H&E stains and averaged across 3 random 200X high-powered fields (HPF) (B) (n = 4 per group, mean + SEM). Serum CCL2 was determined by magnetic bead-based assay (C) (n = 13-14 per group, mean + SEM). Epididymal adipose tissue mRNA expression was determined using real-time qRT-PCR and (-ΔΔCt) standardized to the geometric mean of Gapdh, 36B4, and β-actin reference genes using the 2 method (D) (n = 13-14 per group, mean + SEM). Mean values with unlike letters indicate differences at P < 0.05 using post hoc comparisons.

114

4.5. Discussion

High fat and cholesterol-enriched diets have been linked with the development of obesity, hepatic steatosis, and dyslipidemia in animal models [12-15]. Sphingolipids are dietary bioactives shown to reduce lipid absorption and alter lipid metabolism both in vitro and in rodent models [18-25].

In the current study, we examined the development of diet-induced obesity and hepatic steatosis in mice while supplementing with dietary SM. Both dietary egg SM and milk SM supplementation were able to attenuate the development of hepatic steatosis. Hepatic TG reductions were confirmed with histological changes in steatosis and reductions in PPARγ-related gene expression. Adipose tissue inflammation and markers of macrophage infiltration were also decreased with dietary SM.

Contrary to our hypothesis, egg SM reduced serum cholesterol, adipose tissue inflammation, hepatic TG, and skeletal muscle TG more effectively than milk SM. Overall, the current findings show dietary SM can attenuate hepatic steatosis and adipose tissue inflammation observed with diet-induced obesity, with a substantial difference in potency between egg SM and milk SM.

In the current study, egg SM attenuated hypercholesterolemia and hyperglycemia induced by HFD. Dietary SM is well-documented to acutely reduce cholesterol absorption [18-21], and in some cases, the absorption of other lipids and fat-soluble compounds [20, 21]. This inhibitory effect provides the rationale for studies examining chronic intake of dietary SM and lipid homeostasis [22-25]. We predicted that milk SM would be more effective than egg SM in preventing diet-induced obesity complications, based on previous studies in rats showing milk SM had greater inhibition of lipid absorption than

115

Figure 4.6 Egg SM attenuates skeletal muscle triglyceride accumulation with HFD. Skeletal muscle (quadriceps femoris) lipids were extracted using a modified Folch extraction and TG was measured by enzymatic methods (n = 6 per group, mean + SEM) (A). Muscle mRNA was measured by real-time qRT-PCR (B). Data was normalized to endogenous reference gene expression (n = 13-14 per group, mean + SEM). Mean values with unlike letters indicate differences at P < 0.05 using post hoc comparisons

116 egg SM [21]. However, this was not observed, and may be related to differences in study design, such as duration or animal model. The ratio of cholesterol and SM in lipid emulsions is known to impact the magnitude of cholesterol absorption in vitro and in vivo [18]. Cholesterol and SM mutually inhibit each other’s absorption [19]. As little as 0.1% (w/w) dietary milk SM was shown to reduce cholesterol absorption by 20% in mice on a chow diet [18]. Egg SM (0.2% w/w) supplemented over 6 weeks to a high fat, high cholesterol diet lowered serum lipids in APOE*3-

Leiden mice [25]. We have previously shown that milk SM (0.25% w/w) supplementation of a

HFD for 4 weeks, without added cholesterol, was able to reduce serum and hepatic lipids in mice, while egg SM supplementation (0.25% w/w) surprisingly increased serum and hepatic lipids [23].

Serum lipids in rodents fed various animal-sourced sphingolipids have been shown to decrease

[23-25], not change [22], and even increase in some studies [23, 37]. In general, sphingolipid supplementation in rodents consuming large amounts of dietary cholesterol typically reduces serum cholesterol and hepatic lipids, whereas in the absence of large doses of cholesterol, supplementation with sphingolipids resulted in increased serum and hepatic lipids in some studies

[23, 37]. While in the current study milk SM did not lower serum cholesterol, it successfully reduced hepatic lipids. Dietary SM has been previously reported to reduce hepatic lipids independent of changes in serum lipids [22]. More research is required to determine the optimal conditions for dietary sources of SM to reduce serum and hepatic lipids.

The most striking observation between dietary SM-supplemented groups and HFD control were the histological and biochemical differences in hepatic steatosis. Both SM groups had lower hepatic TG and cholesterol concentrations compared to HFD; however, hepatic TG were reduced most prominently with egg SM. It has been estimated that up to 60% of hepatic TG are derived from serum NEFA [38]. In the current study, both SM groups lowered serum NEFA concentrations

117 compared to HFD. Reduced NEFA uptake by the liver is further supported by the reduction in hepatic Cd36 and Fabp4 mRNA observed in the egg SM group. CD36 mediates long-chain fatty acid uptake into cells to be used for TG synthesis [39]. FABP4 is responsible for binding fatty acids and transporting them inside cells [40]. Differences in oxidation, synthesis, or secretion could also contribute to the observed reductions in hepatic TG and cholesterol. In the current study, no differences in hepatic expression of various genes related to β-oxidation, de novo fatty acid synthesis, or cholesterol biosynthesis were detected, suggesting that fatty acid oxidation and fatty acid/cholesterol biosynthesis were unchanged. However, Scd1 mRNA was significantly reduced in both SM groups. SCD1 is an important enzyme in TG synthesis, enzymatically desaturating stearic and palmitic acid to oleic and palmitoleic acid, respectively [41]. Dietary fatty acids can also be a source of NEFA for hepatic TG synthesis [38]. Dietary sphingolipids have been previously shown to reduce the absorption of dietary fat in rodents [20, 21, 25]. While we observed differences between HFD groups in hepatic steatosis, an early marker of NAFLD, we did not examine long-term sequelae, such as inflammation and fibrosis. Longer studies are necessary to determine any effects of dietary SM on more severe stages of NAFLD, such as NASH and fibrosis.

Endogenous ceramide content has been linked with altered hepatic function. Hepatic insulin sensitivity is reduced in mice lacking ceramide synthase 2 (CerS2), an enzyme required for very- long chain ceramide (C22-C24) synthesis in hepatocytes [42-44]. C16:0-ceramide and the expression of its major synthetic enzymes, ceramide synthase 5 and 6, markedly increase in the livers of obese mice [44, 45]. Other studies have reported reductions in very-long chain ceramides

(C22-C24) in the livers of HFD-fed mice [45, 46]. Thus, it will be important in future studies to determine if there are alterations in hepatic ceramide species with dietary SM intake.

118

The hepatic expression of Scd1 [47], Fabp4 [48], Cd36 [49], and Pparg2 [50] mRNA are known to be induced by PPARγ agonists. Hepatic PPARγ activity is associated with the accumulation of lipid droplets [51]. PPARγ overexpression in mice accelerates the development of hepatic steatosis [52]. PPARγ is a ligand-activated transcription factor, shown to be activated by long-chain fatty acids [53]. In obesity, more long-chain fatty acids would be available to the liver, offering more potential ligands for PPARγ and substrates for TG synthesis. We failed to detect significant differences in total PPARγ protein in livers of SM-supplemented mice compared to HFD control. However, PPARγ protein expression is not a complete picture of PPARγ activity.

PPARγ activity is heavily regulated by post-translational modifications and ligand availability [54,

55]. Both dietary SM-supplemented groups reduced lipid accumulation relative to HFD control.

We speculate that an overabundance of fatty acids, known ligands for PPARγ, are responsible for driving the higher hepatic expression of Pparg2, Cd36, Fabp4, and Scd1 mRNA detected in HFD control vs egg SM. Further research should investigate if dietary SM has any direct effects on altering hepatic PPARγ activity or indirect effects by reducing fatty acid ligand availability.

Obesity-related metabolic complications are associated with adipocyte dysfunction and the recruitment of inflammatory macrophages into adipose tissue, contributing to adipose tissue inflammation and insulin resistance [56]. We observed a significant attenuation of adipose tissue macrophage infiltration and inflammation by egg SM, with a smaller effect by milk SM. The recruitment of macrophages into adipose tissue is driven by increases in CCL2, a chemokine factor otherwise known as monocyte chemoattractant protein-1 (MCP-1) [57]. Both dietary SM groups reduced serum CCL2, while only egg SM had a significant effect on all markers of macrophage infiltration measured (Cd68, F4/80, Cd11c, crown-like structures). This may be related to the stronger effect of egg SM on reducing CCL2 mRNA expression in the adipose tissue. However,

119 both SM groups had significant reductions in adipose tissue F4/80 and TNFα mRNA expression, suggesting a general reduction in adipose tissue inflammation with dietary SM. Umemoto et al.

[13] had previously shown a marginal effect of the cholesterol absorption inhibitor, ezetimibe, on preventing adipose tissue macrophage infiltration and inflammation in diet-induced obese mice.

However, we observed strong effects with egg SM, thus, suggesting dietary SM works through other mechanisms besides inhibiting dietary cholesterol absorption. Additionally, the stronger effects observed with egg SM, which is reportedly less effective at inhibiting cholesterol absorption than milk SM [21], supports this assumption. Along with reduced inflammation, egg

SM increased the expression of GLUT4 and adiponectin (Adipoq) mRNA in adipose tissue, suggesting an improvement in metabolic function. While skeletal muscle is considered the major organ involved in whole-body glucose disposal [58], adipose GLUT4 has been shown to promote glucose tolerance in mice [59] and, thus, may have contributed to the reduced fasting glucose in egg SM-fed mice compared to HFD.

Skeletal muscle TG content of mice fed egg SM was reduced compared to HFD, and this may have also contributed to an improvement in fasting glucose [60]. However, we did not observe differences in skeletal muscle Glut4 mRNA expression between HFD groups and also failed to observe differences in serum insulin and HOMA-IR. Although lower skeletal muscle TG was observed with egg SM relative to HFD, the expression of several genes involved in β-oxidation was reduced (Acox1, Acadl, Cpt1b). It seems plausible that β-oxidative genes were induced in

HFD skeletal muscle to compensate for an increased fatty acid load, an effect which was reversed with egg SM.

We observed a substantial difference in potency between egg SM and milk SM in the current study. Compositional variation between egg SM and milk SM would be expected to result

120 in differences in bioactivity. Egg SM species are mostly uniform, with amide-linked palmitic acid

(16:0) and sphingosine bases (d18:1) as primary components [28, 29]. In contrast, bovine milk SM has a more variable fatty acid (16:0, 22:0-24:0) and sphingoid base (d16:0-d19:0 and d16:1-19:1) composition [28, 30]. The longer amide-linked fatty acyl chains in milk SM are thought to form tighter hydrophobic interactions with other dietary lipids, inhibiting their absorption to a greater extent than the 16:0 chains found in egg SM [21]. Saturated sphingoid bases are also more pervasive in milk SM. SM with saturated sphingoid backbones are known as dihydrosphingomyelin (DHSM). DHSM has been shown to display stronger interactions with cholesterol than SM [61]. Dihydroceramide, the hydrolytic product of DHSM, demonstrates a lower bioactivity to induce apoptosis of cells than ceramide [62]. Therefore, variations in bioactivity and affinity for cholesterol, driven by length of the amide-linked fatty acyl chains and the saturation of the sphingoid backbone, could have resulted in the differences in bioactivity of egg SM and milk SM observed in the current study. Another possible explanation for variations in bioactivity of dietary SM could be related to their hydrolysis in the gut. Alkaline sphingomyelinase’s in vitro activity is inhibited by excess lipid, especially longer chain fatty acids and higher cholesterol-to-fatty acid ratios [63]. Ceramide derived from milk SM is a potent inhibitor of in vitro cholesterol absorption [64]. Sphingosine can also inhibit cholesterol absorption but is not as effective as ceramide [64] and is readily absorbed [65]. Structural variations between egg SM and milk SM may result in differential hydrolysis in the gut. Milk SM may not be hydrolyzed as efficiently as egg SM by alkaline sphingomyelinase, due to its longer fatty acid chains. Further supporting this notion, increases in very long-chain fatty acids (22:0–24:0) have not been observed in rat mesenteric lymph after feeding milk SM, whereas palmitic acid (16:0) increased in rat mesenteric lymph after feeding egg SM [21].

121

In conclusion, the current study demonstrates a decrease in the development of hepatic steatosis and adipose tissue inflammation in diet-induced obese mice when supplemented with egg

SM, and to a lesser extent with milk SM. The visible decreases in hepatic steatosis were confirmed by reductions in both liver TG and CE in the SM-supplemented groups. These reductions in hepatic lipids were associated with the reduced expression of PPARγ-related genes involved in TG synthesis, especially with egg SM. Egg SM was more effective than milk SM in preventing hepatic steatosis, while also reducing adiposity, fasting glucose, and serum cholesterol. The reason for such differences observed is unclear, although there are likely additional bioactive effects of dietary SM beyond impairing lipid absorption. More research is warranted to determine the effects of dietary SM on other obesity-related outcomes and to delineate mechanisms underlying the differences in bioactivity between egg and milk SM.

122

4.6. References

1. Ogden CL, Carroll MD, Fryar CD, Flegal KM: Prevalence of Obesity Among Adults and Youth: United States, 2011-2014. NCHS Data Brief 2015:1-8. 2. Cowie CC, Rust KF, Byrd-Holt DD, Gregg EW, Ford ES, Geiss LS, Bainbridge KE, Fradkin JE: Prevalence of diabetes and high risk for diabetes using A1C criteria in the U.S. population in 1988- 2006. Diabetes Care 2010, 33:562-568. 3. Bellentani S, Scaglioni F, Marino M, Bedogni G: Epidemiology of non-alcoholic fatty liver disease. Dig Dis 2010, 28:155-161. 4. Despres JP, Lemieux I: Abdominal obesity and metabolic syndrome. Nature 2006, 444:881-887. 5. Gregor MF, Hotamisligil GS: Inflammatory mechanisms in obesity. Annu Rev Immunol 2011, 29:415-445. 6. Postic C, Girard J: Contribution of de novo fatty acid synthesis to hepatic steatosis and insulin resistance: lessons from genetically engineered mice. J Clin Invest 2008, 118:829-838. 7. Losekann A, Weston AC, de Mattos AA, Tovo CV, de Carli LA, Espindola MB, Pioner SR, Coral GP: Non-Alcoholic Steatohepatitis (NASH): Risk Factors in Morbidly Obese Patients. Int J Mol Sci 2015, 16:25552-25559. 8. Cordeiro A, Pereira SE, Saboya CJ, Ramalho A: Nonalcoholic Fatty Liver Disease Relationship with Metabolic Syndrome in Class III Obesity Individuals. Biomed Res Int 2015, 2015:839253. 9. Xanthakos SA, Jenkins TM, Kleiner DE, Boyce TW, Mourya R, Karns R, Brandt ML, Harmon CM, Helmrath MA, Michalsky MP, et al: High Prevalence of Nonalcoholic Fatty Liver Disease in Adolescents Undergoing Bariatric Surgery. Gastroenterology 2015, 149:623-634.e628. 10. Anstee QM, Goldin RD: Mouse models in non-alcoholic fatty liver disease and steatohepatitis research. Int J Exp Pathol 2006, 87:1-16. 11. Hardy T, Oakley F, Anstee QM, Day CP: Nonalcoholic Fatty Liver Disease: Pathogenesis and Disease Spectrum. Annu Rev Pathol 2016, 11:451-496. 12. Subramanian S, Han CY, Chiba T, McMillen TS, Wang SA, Haw A, 3rd, Kirk EA, O'Brien KD, Chait A: Dietary cholesterol worsens adipose tissue macrophage accumulation and atherosclerosis in obese LDL receptor-deficient mice. Arterioscler Thromb Vasc Biol 2008, 28:685-691. 13. Umemoto T, Subramanian S, Ding Y, Goodspeed L, Wang S, Han CY, Teresa AS, Kim J, O'Brien KD, Chait A: Inhibition of intestinal cholesterol absorption decreases atherosclerosis but not adipose tissue inflammation. J Lipid Res 2012, 53:2380-2389. 14. Subramanian S, Goodspeed L, Wang S, Kim J, Zeng L, Ioannou GN, Haigh WG, Yeh MM, Kowdley KV, O'Brien KD, et al: Dietary cholesterol exacerbates hepatic steatosis and inflammation in obese LDL receptor-deficient mice. J Lipid Res 2011, 52:1626-1635. 15. Wouters K, van Gorp PJ, Bieghs V, Gijbels MJ, Duimel H, Lutjohann D, Kerksiek A, van Kruchten R, Maeda N, Staels B, et al: Dietary cholesterol, rather than liver steatosis, leads to hepatic inflammation in hyperlipidemic mouse models of nonalcoholic steatohepatitis. Hepatology 2008, 48:474-486. 16. Timlin MT, Parks EJ: Temporal pattern of de novo lipogenesis in the postprandial state in healthy men. The American Journal of Clinical Nutrition 2005, 81:35-42. 17. Blesso CN: Egg phospholipids and cardiovascular health. Nutrients 2015, 7:2731-2747. 18. Eckhardt ER, Wang DQ, Donovan JM, Carey MC: Dietary sphingomyelin suppresses intestinal cholesterol absorption by decreasing thermodynamic activity of cholesterol monomers. Gastroenterology 2002, 122:948-956. 19. Nyberg L, Duan RD, Nilsson A: A mutual inhibitory effect on absorption of sphingomyelin and cholesterol. J Nutr Biochem 2000, 11:244-249. 20. Noh SK, Koo SI: Egg sphingomyelin lowers the lymphatic absorption of cholesterol and alpha- tocopherol in rats. J Nutr 2003, 133:3571-3576.

123

21. Noh SK, Koo SI: Milk sphingomyelin is more effective than egg sphingomyelin in inhibiting intestinal absorption of cholesterol and fat in rats. J Nutr 2004, 134:2611-2616. 22. Chung RW, Kamili A, Tandy S, Weir JM, Gaire R, Wong G, Meikle PJ, Cohn JS, Rye KA: Dietary sphingomyelin lowers hepatic lipid levels and inhibits intestinal cholesterol absorption in high-fat- fed mice. PLoS One 2013, 8:e55949. 23. Norris GH, Jiang C, Ryan J, Porter CM, Blesso CN: Milk sphingomyelin improves lipid metabolism and alters gut microbiota in high fat diet-fed mice. J Nutr Biochem 2016, 30:93-101. 24. Yunoki K, Renaguli M, Kinoshita M, Matsuyama H, Mawatari S, Fujino T, Kodama Y, Sugiyama M, Ohnishi M: Dietary sphingolipids ameliorate disorders of lipid metabolism in Zucker fatty rats. J Agric Food Chem 2010, 58:7030-7035. 25. Duivenvoorden I, Voshol PJ, Rensen PC, van Duyvenvoorde W, Romijn JA, Emeis JJ, Havekes LM, Nieuwenhuizen WF: Dietary sphingolipids lower plasma cholesterol and triacylglycerol and prevent liver steatosis in APOE*3Leiden mice. Am J Clin Nutr 2006, 84:312-321. 26. Vesper H, Schmelz EM, Nikolova-Karakashian MN, Dillehay DL, Lynch DV, Merrill AH, Jr.: Sphingolipids in food and the emerging importance of sphingolipids to nutrition. J Nutr 1999, 129:1239-1250. 27. Zeisel SH, Char D, Sheard NF: Choline, phosphatidylcholine and sphingomyelin in human and bovine milk and infant formulas. J Nutr 1986, 116:50-58. 28. Ramstedt B, Leppimaki P, Axberg M, Slotte JP: Analysis of natural and synthetic sphingomyelins using high-performance thin-layer chromatography. Eur J Biochem 1999, 266:997- 1002. 29. Byrdwell WC, Perry RH: Liquid chromatography with dual parallel mass spectrometry and (31)P nuclear magnetic resonance spectroscopy for analysis of sphingomyelin and dihydrosphingomyelin. I. Bovine brain and chicken egg yolk. J Chromatogr A 2006, 1133:149-171. 30. Byrdwell WC, Perry RH: Liquid chromatography with dual parallel mass spectrometry and 31P nuclear magnetic resonance spectroscopy for analysis of sphingomyelin and dihydrosphingomyelin. II. Bovine milk sphingolipids. J Chromatogr A 2007, 1146:164-185. 31. (CDER) CfDEaR: Guidance for industry: estimating the maximum safe starting dose in initial clinical trials for therapeutics in adult healthy volunteers.: U.S. Department of Health and Human Services; 2005. 32. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC: Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985, 28:412-419. 33. Farrell NJ, Norris GH, Ryan J, Porter CM, Jiang C, Blesso CN: Black elderberry extract attenuates inflammation and metabolic dysfunction in diet-induced obese mice. Br J Nutr 2015, 114:1123-1131. 34. Carr TP, Andresen CJ, Rudel LL: Enzymatic determination of triglyceride, free cholesterol, and total cholesterol in tissue lipid extracts. Clin Biochem 1993, 26:39-42. 35. Bragdon JH, Eder HA, Gould RG, Havel RJ: Lipid nomenclature; recommendations regarding the reporting of serum lipids and lipoproteins made by the Committee on Lipid and Lipoprotein Nomenclature of the American Society for the Study of Arteriosclerosis. Circ Res 1956, 4:129. 36. Farrell N, Norris G, Lee SG, Chun OK, Blesso CN: Anthocyanin-rich black elderberry extract improves markers of HDL function and reduces aortic cholesterol in hyperlipidemic mice. Food Funct 2015, 6:1278-1287. 37. Musso G, Gambino R, Durazzo M, Biroli G, Carello M, Fagà E, Pacini G, De Michieli F, Rabbione L, Premoli A, et al: Adipokines in NASH: Postprandial lipid metabolism as a link between adiponectin and liver disease. Hepatology 2005, 42:1175-1183. 38. Donnelly KL, Smith CI, Schwarzenberg SJ, Jessurun J, Boldt MD, Parks EJ: Sources of fatty acids stored in liver and secreted via lipoproteins in patients with nonalcoholic fatty liver disease. J Clin Invest 2005, 115:1343-1351.

124

39. Aitman TJ, Glazier AM, Wallace CA, Cooper LD, Norsworthy PJ, Wahid FN, Al-Majali KM, Trembling PM, Mann CJ, Shoulders CC, et al: Identification of Cd36 (Fat) as an insulin-resistance gene causing defective fatty acid and glucose metabolism in hypertensive rats. Nat Genet 1999, 21:76-83. 40. Veerkamp JH, Peeters RA, Maatman RG: Structural and functional features of different types of cytoplasmic fatty acid-binding proteins. Biochim Biophys Acta 1991, 1081:1-24. 41. Enoch HG, Catala A, Strittmatter P: Mechanism of rat liver microsomal stearyl-CoA desaturase. Studies of the substrate specificity, enzyme-substrate interactions, and the function of lipid. J Biol Chem 1976, 251:5095-5103. 42. Park JW, Park WJ, Kuperman Y, Boura-Halfon S, Pewzner-Jung Y, Futerman AH: Ablation of very long acyl chain sphingolipids causes hepatic insulin resistance in mice due to altered detergent- resistant membranes. Hepatology 2013, 57:525-532. 43. Park WJ, Park JW, Merrill AH, Storch J, Pewzner-Jung Y, Futerman AH: Hepatic fatty acid uptake is regulated by the sphingolipid acyl chain length. Biochim Biophys Acta 2014, 1841:1754-1766. 44. Turpin SM, Nicholls HT, Willmes DM, Mourier A, Brodesser S, Wunderlich CM, Mauer J, Xu E, Hammerschmidt P, Bronneke HS, et al: Obesity-induced CerS6-dependent C16:0 ceramide production promotes weight gain and glucose intolerance. Cell Metab 2014, 20:678-686. 45. Cinar R, Godlewski G, Liu J, Tam J, Jourdan T, Mukhopadhyay B, Harvey-White J, Kunos G: Hepatic cannabinoid-1 receptors mediate diet-induced insulin resistance by increasing de novo synthesis of long-chain ceramides. Hepatology 2014, 59:143-153. 46. Turner N, Kowalski GM, Leslie SJ, Risis S, Yang C, Lee-Young RS, Babb JR, Meikle PJ, Lancaster GI, Henstridge DC, et al: Distinct patterns of tissue-specific lipid accumulation during the induction of insulin resistance in mice by high-fat feeding. Diabetologia 2013, 56:1638-1648. 47. Matsusue K, Haluzik M, Lambert G, Yim SH, Gavrilova O, Ward JM, Brewer B, Jr., Reitman ML, Gonzalez FJ: Liver-specific disruption of PPARgamma in leptin-deficient mice improves fatty liver but aggravates diabetic phenotypes. J Clin Invest 2003, 111:737-747. 48. Schadinger SE, Bucher NL, Schreiber BM, Farmer SR: PPARgamma2 regulates lipogenesis and lipid accumulation in steatotic hepatocytes. Am J Physiol Endocrinol Metab 2005, 288:E1195-1205. 49. Zhou J, Febbraio M, Wada T, Zhai Y, Kuruba R, He J, Lee JH, Khadem S, Ren S, Li S, et al: Hepatic fatty acid transporter Cd36 is a common target of LXR, PXR, and PPARgamma in promoting steatosis. Gastroenterology 2008, 134:556-567. 50. De Coppi P, Milan G, Scarda A, Boldrin L, Centobene C, Piccoli M, Pozzobon M, Pilon C, Pagano C, Gamba P, Vettor R: Rosiglitazone modifies the adipogenic potential of human muscle satellite cells. Diabetologia 2006, 49:1962-1973. 51. Ahmadian M, Suh JM, Hah N, Liddle C, Atkins AR, Downes M, Evans RM: PPARgamma signaling and metabolism: the good, the bad and the future. Nat Med 2013, 19:557-566. 52. Yu S, Matsusue K, Kashireddy P, Cao WQ, Yeldandi V, Yeldandi AV, Rao MS, Gonzalez FJ, Reddy JK: Adipocyte-specific gene expression and adipogenic steatosis in the mouse liver due to peroxisome proliferator-activated receptor gamma1 (PPARgamma1) overexpression. J Biol Chem 2003, 278:498-505. 53. Xu HE, Lambert MH, Montana VG, Parks DJ, Blanchard SG, Brown PJ, Sternbach DD, Lehmann JM, Wisely GB, Willson TM, et al: Molecular recognition of fatty acids by peroxisome proliferator- activated receptors. Mol Cell 1999, 3:397-403. 54. Ohshima T, Koga H, Shimotohno K: Transcriptional activity of peroxisome proliferator- activated receptor gamma is modulated by SUMO-1 modification. J Biol Chem 2004, 279:29551- 29557. 55. Hu E, Kim JB, Sarraf P, Spiegelman BM: Inhibition of adipogenesis through MAP kinase- mediated phosphorylation of PPARgamma. Science 1996, 274:2100-2103. 56. Xu H, Barnes GT, Yang Q, Tan G, Yang D, Chou CJ, Sole J, Nichols A, Ross JS, Tartaglia LA, Chen H: Chronic inflammation in fat plays a crucial role in the development of obesity-related insulin resistance. J Clin Invest 2003, 112:1821-1830.

125

57. Kanda H, Tateya S, Tamori Y, Kotani K, Hiasa K, Kitazawa R, Kitazawa S, Miyachi H, Maeda S, Egashira K, Kasuga M: MCP-1 contributes to macrophage infiltration into adipose tissue, insulin resistance, and hepatic steatosis in obesity. J Clin Invest 2006, 116:1494-1505. 58. Shulman GI, Rothman DL, Jue T, Stein P, DeFronzo RA, Shulman RG: Quantitation of muscle glycogen synthesis in normal subjects and subjects with non-insulin-dependent diabetes by 13C nuclear magnetic resonance spectroscopy. N Engl J Med 1990, 322:223-228. 59. Carvalho E, Kotani K, Peroni OD, Kahn BB: Adipose-specific overexpression of GLUT4 reverses insulin resistance and diabetes in mice lacking GLUT4 selectively in muscle. Am J Physiol Endocrinol Metab 2005, 289:E551-561. 60. Pan DA, Lillioja S, Kriketos AD, Milner MR, Baur LA, Bogardus C, Jenkins AB, Storlien LH: Skeletal muscle triglyceride levels are inversely related to insulin action. Diabetes 1997, 46:983-988. 61. Kuikka M, Ramstedt B, Ohvo-Rekila H, Tuuf J, Slotte JP: Membrane properties of D-erythro- N-acyl sphingomyelins and their corresponding dihydro species. Biophys J 2001, 80:2327-2337. 62. Bielawska A, Crane HM, Liotta D, Obeid LM, Hannun YA: Selectivity of ceramide-mediated biology. Lack of activity of erythro-dihydroceramide. J Biol Chem 1993, 268:26226-26232. 63. Liu JJ, Nilsson A, Duan RD: In vitro effects of fat, FA, and cholesterol on sphingomyelin hydrolysis induced by rat intestinal alkaline sphingomyelinase. Lipids 2002, 37:469-474. 64. Feng D, Ohlsson L, Ling W, Nilsson A, Duan RD: Generating ceramide from sphingomyelin by alkaline sphingomyelinase in the gut enhances sphingomyelin-induced inhibition of cholesterol uptake in Caco-2 cells. Dig Dis Sci 2010, 55:3377-3383. 65. Garmy N, Taieb N, Yahi N, Fantini J: Apical uptake and transepithelial transport of sphingosine monomers through intact human intestinal epithelial cells: physicochemical and molecular modeling studies. Arch Biochem Biophys 2005, 440:91-100.

126

Chapter 5. Dietary Milk Sphingomyelin Reduces Systemic Inflammation in Diet- Induced Obese Mice and Inhibits LPS Activity in Macrophages

127

This chapter is reprinted from “Norris GH, Porter CM, Jiang C, Blesso CN: Dietary Milk Sphingomyelin Reduces Systemic Inflammation in Diet-Induced Obese Mice and Inhibits LPS Activity in Macrophages. Beverages 2017, 3:37.” As the author of this MDPI article, Gregory Norris retains the right to include this article in a dissertation, permission from the publisher is not required.

128

5.1. Abstract

High fat diets increase lipopolysaccharide (LPS) activity in the blood and may contribute to systemic inflammation with obesity. We hypothesized that dietary milk sphingomyelin (SM), which reduces lipid absorption and colitis in mice, would reduce inflammation and be mediated through effects on gut health and LPS activity. C57BL/6J mice were fed high fat, high cholesterol diets (HFD, n = 14) or the same diets with milk SM (HFD-MSM, 0.1% by weight, n = 14) for 10 weeks. HFD-MSM significantly reduced serum inflammatory markers and tended to lower serum

LPS (p = 0.08) compared to HFD. Gene expression related to gut barrier function and macrophage inflammation were largely unchanged in colon and mesenteric adipose tissues. Cecal gut microbiota composition showed greater abundance of Acetatifactor genus in mice fed milk SM, but minimal changes in other taxa. Milk SM significantly attenuated the effect of LPS on pro- inflammatory gene expression in RAW264.7 macrophages. Milk SM lost its effects when hydrolysis was blocked, while long-chain ceramides and sphingosine, but not dihydroceramides, were anti-inflammatory. Our data suggest that dietary milk SM may be effective in reducing systemic inflammation through inhibition of LPS activity and that hydrolytic products of milk SM are important for these effects.

5.2. Introduction

Chronic low-grade inflammation is involved in the pathogenesis of cardiovascular disease, type 2 diabetes and non-alcoholic fatty liver disease [1]. Metabolically-related chronic low-grade inflammation may be exacerbated by circulating lipopolysaccharide (LPS), also called

“endotoxin”, a pro-inflammatory molecule found in the outer membrane of Gram-negative bacteria [2]. Recognition of bacterial LPS by host cells can trigger a pro-inflammatory signaling cascade mediated by Toll-like receptor-4 (TLR4), a pattern recognition receptor [3]. The human

129 gastrointestinal tract has a “gut barrier” to limit the passage of microbes between host intestinal cells (e.g., tight junctions between enterocytes) and into the bloodstream [4]. These features of the gut barrier function to reduce permeability to LPS, which may otherwise trigger host inflammation and disease [5]. However, LPS translocation from the gut to circulation is enhanced by the absorption of dietary lipids [6]. High fat diets (HFD) have been shown to promote inflammation of the distal intestine, causing an impairment of the protective gut barrier and translocation of LPS into host circulation [7]. Diets high in fat can also alter the gut microbiota [8], resulting in diminished amounts of beneficial bacteria, such as some bifidobacteria, which would promote gut barrier permeability [7]. Additionally, diets rich in triglyceride and cholesterol increase chylomicron production during digestion and absorption. It has been shown that LPS can be incorporated into chylomicrons and transported into the bloodstream of mice [6]. Overall, these effects of HFD result in the paracellular and transcellular translocation of LPS from the gut into the circulation. The presence of LPS in the circulation activates the inflammatory response of the immune system, mainly through the activation of TLR4 [3]. Lipopolysaccharide engages the

TLR4/MD-2 receptor complex via LPS-binding protein (LBP) and CD14. Toll-like receptor 4 signaling then activates the nuclear factor-kappa B (NF-κb) transcription factor resulting in the production of pro-inflammatory cytokines, such as tumor necrosis factor alpha (TNF-α) [3].

Macrophages in the colon have been shown to be the first pro-inflammatory immune responders to an HFD [9]; thus, inhibiting LPS translocation and the subsequent inflammatory responses of macrophages shows great potential in reducing the detrimental effects of a diet rich in fat and cholesterol.

Dietary phospholipids, including sphingolipids, show potential in mitigating chronic disease through effects on lipid absorption and inflammation [10, 11]. Dietary sphingolipids,

130 which include sphingomyelin (SM), ceramides, and sphingosine, are mainly found in milk, eggs, and soybeans [12]. It is estimated that the average American consumes 0.3 to 0.4 grams of sphingolipids per day [12]. Sphingomyelin is considered a zoochemical, being present in animal cell membranes but absent from plants [13]. Sphingomyelin found in milk is an important component of milk fat globule membranes [14]. Dietary SM and other sphingolipids have been studied for their effects on dyslipidemia because they interfere with the absorption of dietary fat and cholesterol [15-21]. In addition to effects on lipid absorption, dietary SM may have additional bioactive effects by reducing inflammation and LPS activity, potentially influencing chronic disease. Dietary milk SM has been shown to reduce DSS-induced colitis in mice, suggesting anti- inflammatory effects in the gut [22]. Although, the effect of dietary SM on colon inflammation is controversial, as exacerbation of colitis in mice has been observed with feeding egg-derived SM

[23, 24]. Phospholipids and sphingolipids appear to impact LPS activity, as they have been shown to dampen LPS-induced inflammation [23, 24]. The presence of LPS in circulation also results in the liver upregulating sphingolipid biosynthesis via the sphingolipid rate-limiting biosynthetic enzyme, serine palmitoyltransferase (SPT), possibly as a compensatory mechanism [25]. We have previously shown that feeding 0.25% (w/w) milk SM reduced serum LPS and altered the gut microbiota in HFD-fed mice after 4 weeks [26]. We also recently reported that feeding dietary SM

(0.1% w/w) attenuated hepatic steatosis and adipose tissue inflammation in diet-induced obese mice [27]. The current study aimed to assess the effects of dietary milk SM on systemic and gut inflammation using a high fat diet-induced obese mouse model. We also sought to elucidate whether milk SM and its hydrolytic products (ceramides, dihydroceramides, and sphingosine) directly affect the inflammatory response of macrophages stimulated by LPS.

131

5.3. Materials and Methods

5.3.1. Animals and Diets

Male C57BL/6J mice were purchased from the Jackson Laboratory (Bar Harbor, ME) at 6 weeks of age. Mice were housed in a temperature-controlled room and maintained in a 12-h light/12-h dark cycle within the University of Connecticut-Storrs vivarium. The Animal Care and Use

Committee of the University of Connecticut-Storrs approved all procedures used in the current study. Mice were acclimated to the facility for two weeks before being placed on either a lard- based high fat, high cholesterol diet (HFD; 60% kcal from fat, 0.15% cholesterol added by weight; n = 14) or HFD supplemented with 0.1% of milk SM added by weight (HFD-MSM; n = 14).

Detailed diet compositions are presented in Table 4.1. Experimental diets were prepared using purified ingredients commercially available from Dyets, Inc. (Bethlehem, PA). Milk SM (bovine;

>99% purity) was obtained from Avanti Polar Lipids, Inc. (Alabaster, AL) and then substituted for an equal weight of lard in the treatment group. Accounting for weight gained throughout the study,

HFD-MSM provided the equivalent of consuming approximately 405-670 mg milk SM/day in a

70-kg human based on body surface normalization [28].

Mice consumed the diets ad libitum and fresh food was provided twice per week. Body weight was assessed weekly, while food intake was calculated at each feeding. After 10 weeks, mice were fasted for 6–8 h followed by euthanasia and subsequent blood collection via cardiac puncture. Blood clotted at room temperature for 30 min before serum isolation by centrifugation

(10,000×g for 10 min at 4°C) and then stored at -80°C. Special precautions were taken to minimize endotoxin contamination of the serum samples. Mesenteric adipose and colon tissues were isolated and snap frozen in liquid nitrogen before storage in -80°C.

132

5.3.2. Serum Biochemical Analysis

Serum IL-6, TNF-α, IFNγ, and MIP-1β were measured by Luminex/xMAP magnetic bead-based multiplexing assay using MAGPIX instrumentation from EMD Millipore (Billerica, MA). Serum

LPS was measured using a chromogenic limulus amebocyte lysate (LAL) assay (QCL-1000) obtained from Lonza (Basel, Switzerland).

5.3.3. Gut Microbiota Analysis

Cecal fecal samples were collected from mice and submitted to the University of Connecticut-

Storrs Microbial Analysis, Resources and Services (MARS) facility for microbiota characterization utilizing 16S V4 analysis. DNA was extracted from 0.25 g of fecal sample using the MoBio PowerMag Soil 96 well kit (MoBio Laboratories, Inc, Carlsbad, CA) according to the manufacturer’s protocol for the Eppendorf epMotion liquid handling robot. DNA extracts were quantified using the Quant-iT PicoGreen kit (ThermoFisher Scientific). Partial bacterial 16S rRNA

(V4) and fungal ITS2 genes were amplified using 30 ng extracted DNA as template. The V4 region was amplified using 515F and 806R with Illumina adapters and dual indices (8 basepair golay on

3’[29], and 8 basepair on the 5’ [30]). Samples were amplified in triplicate using Accuprime PFX

PCR master mix (ThermoFisher) with the addition of 10 µg BSA (New England BioLabs, Ipswich,

MA). The PCR reaction was incubated at 95˚C for 2 min., the 30 cycles of 15 s at 95.0°C, 1 min. at 55°C and 1 min. at 68°C, followed by final extension as 68°C for 5 min. PCR products were pooled for quantification and visualization using the QIAxcel DNA Fast Analysis (Qiagen, Hilden,

Germany). PCR products were normalized based on the concentration of DNA from 250-400 bp then pooled using the QIAgility liquid handling robot. The pooled PCR products were cleaned using Mag-Bind RXNPure Plus (Omega Bio-Tek, Norcross, GA) according to the manufacturer’s protocol. The cleaned pool was sequenced on the MiSeq using v2 2x250 base pair kit (Illumina,

133

Inc, San Diego, CA). Sequences were demultiplexed using onboard bcl2fastq. Demultiplexed sequences were processed in Mothur v. 1.39.4 following the MiSeq SOP [30]. Exact commands can be found here https://github.com/krmaas/bioinformatics/blob/master/mothur.batch. Merged sequences that had any ambiguities or did not meet length expectations were removed. Sequences were aligned to the Silva nr_v119 alignment [31]. Taxonomic identification of OTUs was done using the RDP Bayesian classifier [32] against the Silva nr_v119 taxonomy database.

5.3.4. Cell Culture

Murine RAW264.7 macrophages were obtained from ATCC (Manassas, VA) and cultured in a humidified incubator at 37°C and 5% CO2. Cells were maintained in DMEM (4 g/L glucose) containing sodium pyruvate, 10% fetal bovine serum (Hyclone, Logan, UT), 2 mM L-glutamine,

100 U/mL penicillin, 100 µg/mL streptomycin antibiotic (ThermoFisher Scientific, Waltham, MA) and 100 µg/mL Normocin (Invitrogen, Carlsbad, CA). Cell counts and viability were routinely measured using Trypan blue and TC-20 automated cell counter (Bio-Rad, Hercules, CA).

5.3.5. Effects of Sphingolipids on LPS Stimulation of RAW264.7 Macrophages

Milk sphingomyelin (bovine), sphingosine (d18:1), C16-ceramide (d18:1/16:0), C24-ceramide

(d18:1/24:0), C16-dihydroceramide (d18:0/16:0), and C24-dihydroceramide (d18:0/24:0) were obtained from Avanti Polar Lipids at >99% purity. Milk SM and sphingosine were dissolved in

100% ethanol (EtOH), while ceramides were dissolved in EtOH/dodecane (98.8/0.2, v/v).

RAW264.7 cells were seeded in 24-well plates and allowed to adhere overnight prior to experimentation. To examine possible anti-inflammatory effects of milk SM, RAW264.7 macrophages were pre-incubated with milk SM (0.8-8 µg/mL) or EtOH vehicle control for 1 h, followed by a 4 h co-incubation in the presence or absence of 1 ng/mL LPS (E. coli 0111:B4)

(Sigma-Aldrich, St. Louis, MO). Since milk SM comprised of a natural mixture of SM species,

134 concentrations are reported in 0.8-8 µg/mL, which is comparable to 1-10 µM of a pure SM species.

To determine if effects were dependent on SM hydrolysis, cells incubated for 2 h with 100 μM imipramine (Sigma-Aldrich) prior to 1 h incubation with milk SM or control ± 4 h co-incubation with LPS. Imipramine has been shown to reduce acid sphingomyelinase (SMase) activity to ~20% of basal levels after 2 h of incubation [33]. To test the effects of various ceramides and sphingosine,

RAW264.7 macrophages were incubated for 4 h with ceramides (10 µM), dihydroceramides (10

µM), sphingosine (1-10 µM), or vehicle control ± LPS (1 ng/mL). Ethanol was used as a vehicle control for milk SM and sphingosine treatments, while EtOH/dodecane (98.8/0.2, v/v) was used as a control for ceramide treatments. To test for cytotoxicity, Cell Counting Kit-8 (CCK-8) (Sigma-

Aldrich) was used after cells were treated with the different sphingolipids. Sodium dodecyl sulfate

(SDS, 50 μM) was used as a cytotoxic positive control. Caspase 3 activity was also determined as an indicator of apoptosis using a colorimetric protease assay kit (ThermoFisher Scientific) according to manufacturer instructions. Cisplatin (ThermoFisher Scientific) at 50 µM concentration was used on cells as a positive control to induce apoptosis.

5.3.6. RNA isolation, cDNA Synthesis, and qRT-PCR

Total RNA from small intestine, colon, mesenteric adipose tissue and RAW264.7 macrophages was isolated using TRIzol reagent (Life Technologies, Carlsbad, CA). RNA was treated with

DNase I before reverse transcription using iScript cDNA synthesis kit (Bio-Rad). Real-time qRT-

PCR using the iTaq Universal SYBR Green Supermix was performed on a CFX96 real-time-PCR detection system (Bio-Rad). For small intestine, colon, and mesenteric adipose tissues, the geometric mean of the reference genes, glyceraldehyde 3-phosphate dehydrogenase (Gapdh) and beta actin was used to standardize mRNA expression using the 2−ΔΔCt method. The macrophage

135 mRNA was standardized to Gapdh expression. All primer sequences used are listed in Appendix

5.2.7. Statistical Analysis

Data were analyzed using GraphPad Prism 6. Means were compared using Student’s t test or one- way analysis of variance with Holm-Sidak post hoc analysis where indicated. For gut microbiota analysis, alpha and beta diversity statistics were calculated by taking the average of 1000 random subsampling to 10,000 reads per sample in Mothur. NMS and Permanova were run in R 3.3.2. A subsampled species matrix was used for indicator species analysis [34]. Figures were drawn in R

3.3.2 using ggplot2 2.2.1 and RColorBrewer 1.1-2. Data are reported as mean ± SEM. Significance is reported as p < 0.05.

5.4. Results

5.4.1. Dietary milk SM reduces systemic inflammation and tends to lower circulating LPS

All mice fed the lard-based HFD were obese and insulin-resistant after 10 weeks, while supplementation of 0.1% (w/w) milk SM did not affect food intake, body or tissue weight gain compared to HFD control [27]. However, feeding milk SM strongly decreased serum inflammatory cytokines/chemokines compared to HFD control (Figure 5.1A). Milk SM tended to reduce LPS compared to HFD control (-36%), but it was not significantly different (p = 0.08)

(Figure 5.1B). Milk SM tended to increase Niemann-Pick C1-Like 1 (NPC1L1) mRNA expression (p = 0.07) in the small intestine (Figure 5.2A), which is induced by cellular cholesterol depletion [35]. No other significant changes were observed in the small intestine for gene expression related to lipid absorption. Interestingly, C-C motif chemokine ligand 2 (CCL2) mRNA expression was significantly increased in the colon by milk SM (Figure 5.2B). However, this did not appear to alter gut health, as mRNA expression of genes related to macrophage

136

Figure 5.1. Dietary milk SM reduces serum inflammation markers in diet-induced obese mice. Male C57BL/6J mice were fed high fat diet (60% kcal fat, 31% lard, 0.15% added cholesterol) (HFD) or HFD supplemented with 0.1% (w/w) milk SM (HFMSM) for 10 weeks. Serum cytokines/chemokines (A) and endotoxin concentrations (B) were by magnetic bead-based assay and chromogenic LAL assay, respectively. Mean ± SEM, n = 13-14 per group. *p < 0.05, **p < 0.01 compared to HFD.

137

Figure 5.2. Intestinal and mesenteric adipose tissue mRNA expression of diet-induced obese mice. Small intestine mRNA (A), colon mRNA (B), and mesenteric adipose tissue mRNA (C) was determined using real-time qRT-PCR and standardized to the (-ΔΔCt) geometric mean of Gapdh, 36B4, and β-actin reference genes using the 2 method. Mean ± SEM, n = 13-14 per group. **p < 0.01 compared to HFD.

138 infiltration/inflammation (F4/80, Cd68, Cd11c, Tnf) and gut barrier function (Tjp1, Alpi, Ocln) was mostly unaffected by milk SM (Figure 5.2B). Furthermore, mesenteric adipose tissue mRNA expression related to inflammation was also unchanged, while there was a trend in GLUT4 mRNA expression to be increased by milk SM (p = 0.09) (Figure 5.2C).

5.4.2. Gut microbiota composition is mostly unaffected by 0.1% (w/w) dietary milk SM

To determine if alterations in gut microbiota could explain differences in systemic inflammation in the present study, cecal feces microbiota composition was examined by 16S rRNA sequence analysis and results are presented in Figure 5.3. Verrucomicrobia was the major phylum in cecal feces, but there were no significant differences between groups in its relative abundance or that of its major genus, Akkermansia (Figure 5.3B). Additionally, no significant differences were observed in the relative abundance of Firmicutes or Bacteroidetes phyla between groups (Figure

5.3B) or when phyla were grouped as Gram-negative bacteria (HFD: 76.9% ± 1.5% vs. HFD-

MSM: 73.5% ± 2.8%, p = 0.31). However, relative abundance of Acetatifactor, a genus of the

Firmicutes phylum, was significantly higher in the group fed milk SM compared to HFD control

(Figure 5.3B). Alpha-diversity analysis (diversity within a sample) determined by inverse

Simpson index was not significantly different between groups (HFD: 2.19 ± 0.16 vs. HFD-MSM:

2.46 ± 0.32, p = 0.45). Furthermore, beta-diversity analysis (between sample diversity) by Bray-

Curtis revealed no significant clustering of samples according to diet (Figure 5.3C). Therefore, in the context of diet-induced obesity, gut microbiota composition was generally unaffected by supplementation of 0.1% (w/w) milk SM in the diet.

5.4.3. Milk SM inhibits LPS stimulation of macrophages

Since dietary milk SM may directly influence inflammatory processes beyond inhibiting lipid absorption, we examined its effects on pro-inflammatory gene expression in RAW264.7

139

Figure 5.3. Gut microbiota composition of diet-induced obese mice. Cecal fecal samples were collected and microbiota composition was assessed by 16S rRNA sequencing as described in Materials and Methods. Mean phylogenetic abundance (% of total sequences) of mice (A) and taxa comparisons (B). Beta diversity shown by Bray-Curtis based non-metric multidimensional scaling plot to visualize between sample diversity (C). Mean ± SEM, n = 10 per group. *p < 0.05 compared to HFD. Abbreviations: Lachnospiraceae_u, Lachnospiraceae_unclassified; Ruminococcaceae_u, Ruminococcaceae_unclassified

140 macrophages. Increases in TNF-α and CCL2 mRNA with 4 h LPS stimulation were both significantly reduced with milk SM (0.8 and 8 μg/mL) (Figure 5.4A,B). Milk SM in the absence of LPS did not influence pro-inflammatory gene expression. Furthermore, cell viability was not affected by milk SM at concentrations shown to affect pro-inflammatory gene expression (Figure

5.4C). With the addition of imipramine, LPS-stimulated inflammation was not reduced in the presence of milk SM (Figure 5.4D). Imipramine causes proteolytic degradation of acid sphingomyelinase, which hydrolyzes SM to ceramide and phosphorylcholine. This suggests that a hydrolytic product of SM (e.g. ceramide, sphingosine) is important for milk SM’s effect on inflammation.

5.4.4. Ceramides and sphingosine, but not dihydroceramides, inhibit LPS stimulation of macrophages

Since milk SM contains a mixture of SM species with different fatty acid chain lengths and sphingoid backbones, we tested the effects of C16-ceramide, C24-ceramide, C16- dihydroceramide, and C24-dihydroceramide on macrophage pro-inflammatory gene expression

(Figure 5.5A,B). Interestingly, both C16-ceramide and C24-ceramide at 10 µM concentrations significantly reduced TNF-α mRNA (Figure 5.5A) and CCL2 mRNA (Figure 5.5B) in LPS- stimulated macrophages, but their corresponding dihydroceramides did not. None of the ceramides tested altered inflammatory gene expression in the absence of LPS. The anti-inflammatory effects of ceramides (sphingosine base), but not dihydroceramides (sphinganine base), suggest sphingosine is important for bioactivity Supporting this notion, sphingosine significantly reduced

LPS-stimulation of TNF-α mRNA (Figure 5.6A) and CCL2 mRNA (Figure 5.6B) in macrophages. Neither ceramides nor sphingosine significantly affected cell viability (Figure

5.7A,B) or caspase 3 activity (Figure 5.7C) at the concentrations tested, suggesting cell death was not the cause of anti-inflammatory effects.

141

Figure 5.4. Milk SM inhibits LPS-activation of macrophages. RAW264.7 macrophages were pre-incubated with milk SM (MSM) or EtOH vehicle control for 1 h, followed by a 4 h co-incubation ± LPS (1 ng/mL). TNF-α mRNA (A) and CCL2 mRNA (B) were measured by real-time qRT-PCR. Cell viability determined using Cell Counting Kit-8 (C). Sodium dodecyl sulfate (SDS) (0.5 mM) was used on cells as a positive control to induce cell death, **p < 0.001 compared to control. TNF-α mRNA was measured in cells incubated for 2 h with 100 μM imipramine prior to a 1 h incubation with milk SM or control ± a 4 h co-incubation with LPS (D). Mean ± SEM, n = 3-7 independent experiments. Unlike letters p < 0.05 by Holm-Sidak post hoc test.

142

Figure 5.5. Long-chain ceramides inhibit LPS-activation of macrophages. RAW264.7 macrophages were incubated for 4 h with ceramides (10 µM), dihydroceramides (10 µM), or vehicle control ± LPS (1 ng/mL). C16-ceramide (d18:1/16:0), C24- ceramide (d18:1/24:0), C16-dihydroceramide (d18:0/16:0), and C24-dihydroceramide (d18:0/24:0) were tested. TNF-α mRNA (A) and CCL2 mRNA (B) measured by real-time qRT-PCR. Mean ± SEM, n = 3-7 independent experiments. Unlike letters p < 0.05 by Holm-Sidak post hoc test.

143

Figure 5.6. Sphingosine inhibits LPS-activation of macrophages. RAW264.7 macrophages were incubated for 4 h with sphingosine (1-10 µM) or vehicle control ± LPS (1 ng/mL). TNF-α mRNA (A) and CCL2 mRNA (B) measured by real-time qRT- PCR. Mean ± SEM, n = 3-7 independent experiments. Unlike letters p < 0.05 by Holm-Sidak post hoc test.

144

Figure 5.7. Ceramides and sphingosine do not affect cell viability or apoptosis. Cell viability determined using Cell Counting Kit-8 for ceramides (A) and sphingosine (B). Sodium dodecyl sulfate (SDS) (0.5 mM) was used on cells as a positive control to induce cell death. Caspase 3 activity measured as an indicator of apoptosis (C). Cisplatin (50 µM) was used on cells as a positive control to induce apoptosis. Mean ± SEM, n = 3-5 independent experiments. **p < 0.01 compared to control.

145

5.5. Discussion

Chronic low-grade inflammation is a common underlying factor in many diseases afflicting

Western societies, including diabetes, non-alcoholic fatty liver disease, and atherosclerosis [36].

Low-grade inflammation in these states may be exacerbated by circulating LPS, also known as endotoxin [2]. LPS translocation from the gut to circulation is enhanced by lipid absorption [6].

Dietary SM has been studied for its effects on dyslipidemia because it reduces the absorption of other lipids (e.g., fat and cholesterol), and therefore, could potentially influence systemic inflammation. In this study, the addition of 0.1% (w/w) dietary milk SM to a lard-based HFD significantly reduced systemic inflammation markers and tended to lower circulating LPS but had no effect on gene expression of inflammation or gut barrier function markers in the mesenteric adipose tissue or colon. Furthermore, microbiota composition of cecal feces was mostly unaffected by the addition of milk SM to the HFD diet, suggesting gut microbiota compositional differences did not play a major role in altering systemic inflammation markers. Milk SM directly attenuated

LPS stimulation of pro-inflammatory gene expression in RAW 264.7 macrophages. Interestingly, hydrolytic products of milk SM (C16-ceramide, C24-ceramide, and sphingosine) showed similar anti-inflammatory effects, whereas the inhibition of SM hydrolysis with imipramine prevented such effects. These results suggest that dietary SM may have additional bioactive effects beyond inhibiting lipid absorption through the lowering of macrophage inflammatory responses, potentially influencing metabolic disease.

In rodent studies, dietary patterns rich in saturated fat and cholesterol, so called “Western” diets, have been linked to chronic disease through increasing inflammation [37, 38]. Detrimental effects of Western diets in mice appear to be influenced by the presence of microbes in the GI tract, since depletion of gut microbiota with oral antibiotics reduces the inflammatory response

146

[39]. This is partly related to the effect that Western diets have on increasing LPS in circulation, termed “metabolic endotoxemia”, which has been shown to precede increases in systemic inflammation in animal models [2, 39]. There is also evidence that diets high in fat and cholesterol increase endotoxin concentration in the bloodstream of humans [40, 41]. With the current study, we fed milk SM (0.1%) for 10 weeks to HFD-fed mice (60% kcal as fat; 0.15% cholesterol added) to determine if dietary SM could affect chronic low-grade inflammation. Milk SM strongly reduced serum inflammatory cytokines/chemokines and tended to reduce serum LPS (p = 0.08).

Lower serum CCL2 and mRNA expression of inflammation markers in epididymal adipose tissue were also previously shown in mice fed milk SM [27]. We have previously reported that the 4- week dietary supplementation of a higher dose of milk SM (0.25% w/w) reduced circulating LPS by 35% in mice fed a milkfat-based high fat diet [26]. Although we did not observe any differences in mRNA expression of gut barrier markers in the small intestine, reductions in circulating LPS are expected to be partly due to the inhibitory effect of milk SM on lipid absorption. Supporting this notion, we have previously observed reductions in serum and hepatic lipids with the feeding of milk SM [26, 27].

Recently, chronic feeding of HFD to mice was shown to increase colonic pro-inflammatory macrophages and gut inflammation, which promoted LPS translocation, insulin resistance, and adipose tissue inflammation [9]. These effects were mediated by early induction of CCL2 in intestinal epithelial cells, which attracted pro-inflammatory macrophages to colon tissue [9]. In this study, although we observed significantly greater colon CCL2 mRNA expression with milk

SM intake, we did not observe any significant changes between groups in gene expression of gut barrier markers or macrophage infiltration/inflammation in colon and mesenteric adipose tissues.

This suggests that CCL2 mRNA expression was possibly induced with milk SM because of greater

147 dietary fat/cholesterol reaching the colon, but this did not induce macrophage infiltration and TNF-

α expression in the colon and surrounding adipose tissue. Although dietary SM is known to inhibit cholesterol absorption, luminal cholesterol can reciprocally inhibit SM digestion [16]. Feeding mice a high fat diet has also been shown to reduce the expression of the key enzyme in SM digestion, alkaline sphingomyelinase [45, 46]. Therefore, it is possible that the use of a high fat lard diet with added cholesterol in this study influenced ileal and colonic exposure to SM and bioactive metabolic products (ceramide, sphingosine, and sphingosine-1-phosphate), which in turn could affect colon inflammation [47, 48].

The intake of HFD has been shown to negatively alter the gut microbiota [42]. Gut dysbiosis has been linked with endotoxin-mediated chronic disease in mouse models [43, 44].

Sphingosine, a hydrolytic product of dietary SM, is known to have bactericidal effects [45]. We have previously reported that supplementation of 0.25% (w/w) milk SM to high fat diet-fed mice altered the fecal microbiota composition by reducing Gram-negative bacterial phyla and increasing

Bifidobacterium [26]. In the current study, however, we did not observe such large changes in gut microbiota composition, possibly related to the lower dose used or the more severe high fat diet challenge. We observed a significant difference in the relative abundance of Acetatifactor genus, with higher levels in mice fed milk SM. This genus was shown to be isolated from the intestine of obese mice and produces the short-chain fatty acids, acetate and butyrate [46]. The abundance of

Acetatifactor in the gut was also related to lard intake in mice and correlated with the concentrations of secondary bile acids in mouse cecum [47]. We also have observed increases in

Acetatifactor genus with 0.25% (w/w) milk SM feeding on a low-fat diet background (Norris and

Blesso, unpublished observations 2016). At this point, it is unclear if these bacteria contribute to

148 the inflammatory changes observed or in response to the effects of milk SM on lipid absorption.

Further research is needed to clarify these associations.

Milk SM significantly reduced TNF-α and CCL2 mRNA expression in LPS-stimulated

RAW264.7 macrophages. There are several potential mechanisms that could explain this observed effect. First, there could be a direct interaction between milk SM and LPS within the media, where intact SM itself is neutralizing LPS, possibly due to the formation of a complex. In our current study, this would mean that the LPS would be neutralized by SM before it could enter the cell, as part of a lipid vesicle that masked the reactive lipid-A moiety of LPS. Incubation with 100 µM bovine brain SM was reported to neutralize LPS stimulation of polymorphonuclear leukocytes

(PMN) in the presence of soluble CD14 and LBP [23]. Furthermore, SM hydrolysis can yield bioactive products, including ceramide and sphingosine. Imipramine is a tricyclic anti-depressant

(TCA) that causes proteolytic degradation of the enzyme acid-SMase [33]. Imipramine may also have other effects besides acid-SMase degradation, as it was shown that acid ceramidase was down-regulated by another TCA, desipramine [48]. Interestingly, the addition of imipramine completely abolished the anti-inflammatory effects of milk SM. This indicates that one of the hydrolytic products of SM (e.g. ceramide, sphingosine, phosphorylcholine) may be responsible for the reduction in inflammation. If milk SM were hydrolyzed by SMase and ceramidase to ceramide and sphingosine, respectively, each of those products could potentially exert anti-inflammatory effects. Longer-chain ceramides may compete with LPS and block TLR4 signaling, as they have been shown to bind to CD14 in monocytes and form a multi-molecular complex that contains

CD36, but lacks TLR4 [49]. To confirm this, ceramides containing different amide-linked fatty acid chain lengths (16:0 and 24:0) and sphingoid bases (dihydroceramides) were tested. We used ceramides with 16-carbon and 24-carbon fatty acid chain lengths because these ceramides would

149 be produced from the hydrolysis of the natural mixture of SM species that comprise milk SM.

Additionally, these fatty acid chain lengths within the ceramide have the ability to alter its bioactivity [50]. Specifically, it has been shown that intracellular C16-ceramide is particularly pro- apoptotic, whereas C24-ceramide and dihydroceramides antagonize apoptosis induced by shorter ceramides [51-53]. Interestingly, both C16-ceramide and C24-ceramide treatment of macrophages resulted in significant reductions in TNF-α and CCL2 mRNA expression, while dihydroceramides had no effect. None of the ceramides or dihydroceramides affected cell viability or the early apoptosis marker, caspase 3 activity, suggesting the effects were not due to an induction of cell death. Although intracellular ceramides have been shown to be pro-apoptotic [50], exogenously added long-chain ceramides do not affect viability of macrophages [54]. These data indicate that ceramides are responsible for the anti-inflammatory effect of milk SM and strengthens the findings of the imipramine data. However, since only the sphingosine base containing-ceramides showed anti-inflammatory effects and not sphinganine-containing dihydroceramides, we further tested the effects of sphingosine on LPS stimulation of macrophages. We observed similar effects of sphingosine (d18:1) on inhibiting TNF-α and CCL2 mRNA without affecting cell viability or apoptosis, as C16- and C24-ceramides. Sphingosine could display anti-inflammatory effects through the activation of the nuclear receptor, peroxisome proliferator-activate receptor-gamma

(PPARγ) [22]. In support of our findings, exogenous C8-ceramide and sphingosine have been previously reported to reduce TNF-α secretion from LPS-stimulated macrophages [55].

Furthermore, both C16-ceramide and membrane-permeable C2-ceramide have been shown to reduce TNF-α secretion from LPS-stimulated macrophages by post-translational mechanisms [56].

In summary, dietary milk SM reduces systemic inflammation in diet-induced obese mice, without altering gene expression of macrophage or barrier function markers in the mesenteric

150 adipose tissue or colon. Milk SM directly attenuates LPS stimulation of pro-inflammatory gene expression in RAW 264.7 macrophages. The inhibition of SM hydrolysis prevents milk SM anti- inflammatory effects, while the hydrolytic products C16-ceramide, C24-ceramide, and sphingosine showed anti-inflammatory effects. In addition to effects on lipid absorption, dietary

SM may have additional bioactive effects by lowering inflammation and LPS activity, potentially influencing metabolic disease. Further research is warranted to confirm the mechanism and effect of dietary milk SM on inflammation.

151

5.6. References

1. Hotamisligil GS: Inflammation and metabolic disorders. Nature 2006, 444:860-867. 2. Cani PD, Amar J, Iglesias MA, Poggi M, Knauf C, Bastelica D, Neyrinck AM, Fava F, Tuohy KM, Chabo C, et al: Metabolic endotoxemia initiates obesity and insulin resistance. Diabetes 2007, 56:1761-1772. 3. Lu YC, Yeh WC, Ohashi PS: LPS/TLR4 signal transduction pathway. Cytokine 2008, 42:145-151. 4. Suzuki T: Regulation of intestinal epithelial permeability by tight junctions. Cell Mol Life Sci 2013, 70:631-659. 5. Neurath MF: Cytokines in inflammatory bowel disease. Nat Rev Immunol 2014, 14:329- 342. 6. Ghoshal S, Witta J, Zhong J, de Villiers W, Eckhardt E: Chylomicrons promote intestinal absorption of lipopolysaccharides. J Lipid Res 2009, 50:90-97. 7. Serino M, Luche E, Gres S, Baylac A, Berge M, Cenac C, Waget A, Klopp P, Iacovoni J, Klopp C, et al: Metabolic adaptation to a high-fat diet is associated with a change in the gut microbiota. Gut 2012, 61:543-553. 8. Tilg H, Moschen AR, Kaser A: Obesity and the microbiota. Gastroenterology 2009, 136:1476-1483. 9. Kawano Y, Nakae J, Watanabe N, Kikuchi T, Tateya S, Tamori Y, Kaneko M, Abe T, Onodera M, Itoh H: Colonic Pro-inflammatory Macrophages Cause Insulin Resistance in an Intestinal Ccl2/Ccr2-Dependent Manner. Cell Metab 2016, 24:295-310. 10. Blesso CN: Egg phospholipids and cardiovascular health. Nutrients 2015, 7:2731-2747. 11. Norris GH, Blesso CN: Dietary sphingolipids: potential for management of dyslipidemia and nonalcoholic fatty liver disease. Nutr Rev 2017. 12. Vesper H, Schmelz EM, Nikolova-Karakashian MN, Dillehay DL, Lynch DV, Merrill AH, Jr.: Sphingolipids in food and the emerging importance of sphingolipids to nutrition. J Nutr 1999, 129:1239-1250. 13. Hannich JT, Umebayashi K, Riezman H: Distribution and functions of sterols and sphingolipids. Cold Spring Harb Perspect Biol 2011, 3. 14. Rombaut R, Dewettinck K: Properties, analysis and purification of milk polar lipids. International Dairy Journal 2006, 16:1362-1373. 15. Noh SK, Koo SI: Egg sphingomyelin lowers the lymphatic absorption of cholesterol and alpha-tocopherol in rats. J Nutr 2003, 133:3571-3576. 16. Nyberg L, Duan RD, Nilsson A: A mutual inhibitory effect on absorption of sphingomyelin and cholesterol. J Nutr Biochem 2000, 11:244-249. 17. Duivenvoorden I, Voshol PJ, Rensen PC, van Duyvenvoorde W, Romijn JA, Emeis JJ, Havekes LM, Nieuwenhuizen WF: Dietary sphingolipids lower plasma cholesterol and triacylglycerol and prevent liver steatosis in APOE*3Leiden mice. Am J Clin Nutr 2006, 84:312-321. 18. Noh SK, Koo SI: Milk sphingomyelin is more effective than egg sphingomyelin in inhibiting intestinal absorption of cholesterol and fat in rats. J Nutr 2004, 134:2611-2616. 19. Garmy N, Taieb N, Yahi N, Fantini J: Interaction of cholesterol with sphingosine: physicochemical characterization and impact on intestinal absorption. J Lipid Res 2005, 46:36-45.

152

20. Eckhardt ER, Wang DQ, Donovan JM, Carey MC: Dietary sphingomyelin suppresses intestinal cholesterol absorption by decreasing thermodynamic activity of cholesterol monomers. Gastroenterology 2002, 122:948-956. 21. Chung RW, Kamili A, Tandy S, Weir JM, Gaire R, Wong G, Meikle PJ, Cohn JS, Rye KA: Dietary sphingomyelin lowers hepatic lipid levels and inhibits intestinal cholesterol absorption in high-fat-fed mice. PLoS One 2013, 8:e55949. 22. Mazzei JC, Zhou H, Brayfield BP, Hontecillas R, Bassaganya-Riera J, Schmelz EM: Suppression of intestinal inflammation and inflammation-driven colon cancer in mice by dietary sphingomyelin: importance of peroxisome proliferator-activated receptor gamma expression. J Nutr Biochem 2011, 22:1160-1171. 23. Wurfel MM, Wright SD: Lipopolysaccharide-binding protein and soluble CD14 transfer lipopolysaccharide to phospholipid bilayers: preferential interaction with particular classes of lipid. J Immunol 1997, 158:3925-3934. 24. Parker TS, Levine DM, Chang JC, Laxer J, Coffin CC, Rubin AL: Reconstituted high- density lipoprotein neutralizes gram-negative bacterial lipopolysaccharides in human whole blood. Infect Immun 1995, 63:253-258. 25. Memon RA, Holleran WM, Moser AH, Seki T, Uchida Y, Fuller J, Shigenaga JK, Grunfeld C, Feingold KR: Endotoxin and cytokines increase hepatic sphingolipid biosynthesis and produce lipoproteins enriched in ceramides and sphingomyelin. Arterioscler Thromb Vasc Biol 1998, 18:1257-1265. 26. Norris GH, Jiang C, Ryan J, Porter CM, Blesso CN: Milk sphingomyelin improves lipid metabolism and alters gut microbiota in high fat diet-fed mice. J Nutr Biochem 2016, 30:93- 101. 27. Norris GH, Porter CM, Jiang C, Millar CL, Blesso CN: Dietary sphingomyelin attenuates hepatic steatosis and adipose tissue inflammation in high-fat-diet-induced obese mice. J Nutr Biochem 2017, 40:36-43. 28. Nair AB, Jacob S: A simple practice guide for dose conversion between animals and human. J Basic Clin Pharm 2016, 7:27-31. 29. Caporaso JG, Lauber CL, Walters WA, Berg-Lyons D, Huntley J, Fierer N, Owens SM, Betley J, Fraser L, Bauer M, et al: Ultra-high-throughput microbial community analysis on the Illumina HiSeq and MiSeq platforms. ISME J 2012, 6:1621-1624. 30. Kozich JJ, Westcott SL, Baxter NT, Highlander SK, Schloss PD: Development of a dual- index sequencing strategy and curation pipeline for analyzing amplicon sequence data on the MiSeq Illumina sequencing platform. Appl Environ Microbiol 2013, 79:5112-5120. 31. Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, Peplies J, Glockner FO: The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res 2013, 41:D590-596. 32. Wang Q, Garrity GM, Tiedje JM, Cole JR: Naive Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy. Appl Environ Microbiol 2007, 73:5261-5267. 33. Hurwitz R, Ferlinz K, Sandhoff K: The tricyclic antidepressant desipramine causes proteolytic degradation of lysosomal sphingomyelinase in human fibroblasts. Biol Chem Hoppe Seyler 1994, 375:447-450. 34. De Caceres M, Legendre P: Associations between species and groups of sites: indices and statistical inference. Ecology 2009, 90:3566-3574.

153

35. Alrefai WA, Annaba F, Sarwar Z, Dwivedi A, Saksena S, Singla A, Dudeja PK, Gill RK: Modulation of human Niemann-Pick C1-like 1 gene expression by sterol: Role of sterol regulatory element binding protein 2. Am J Physiol Gastrointest Liver Physiol 2007, 292:G369- 376. 36. Baker RG, Hayden MS, Ghosh S: NF-kappaB, inflammation, and metabolic disease. Cell Metab 2011, 13:11-22. 37. Wouters K, van Gorp PJ, Bieghs V, Gijbels MJ, Duimel H, Lutjohann D, Kerksiek A, van Kruchten R, Maeda N, Staels B, et al: Dietary cholesterol, rather than liver steatosis, leads to hepatic inflammation in hyperlipidemic mouse models of nonalcoholic steatohepatitis. Hepatology 2008, 48:474-486. 38. Wang Y, Qian Y, Fang Q, Zhong P, Li W, Wang L, Fu W, Zhang Y, Xu Z, Li X, Liang G: Saturated palmitic acid induces myocardial inflammatory injuries through direct binding to TLR4 accessory protein MD2. Nat Commun 2017, 8:13997. 39. Cani PD, Bibiloni R, Knauf C, Waget A, Neyrinck AM, Delzenne NM, Burcelin R: Changes in gut microbiota control metabolic endotoxemia-induced inflammation in high-fat diet-induced obesity and diabetes in mice. Diabetes 2008, 57:1470-1481. 40. Pendyala S, Walker JM, Holt PR: A high-fat diet is associated with endotoxemia that originates from the gut. Gastroenterology 2012, 142:1100-1101 e1102. 41. Erridge C, Attina T, Spickett CM, Webb DJ: A high-fat meal induces low-grade endotoxemia: evidence of a novel mechanism of postprandial inflammation. Am J Clin Nutr 2007, 86:1286-1292. 42. Murphy EF, Cotter PD, Healy S, Marques TM, O'Sullivan O, Fouhy F, Clarke SF, O'Toole PW, Quigley EM, Stanton C, et al: Composition and energy harvesting capacity of the gut microbiota: relationship to diet, obesity and time in mouse models. Gut 2010, 59:1635-1642. 43. Henao-Mejia J, Elinav E, Jin C, Hao L, Mehal WZ, Strowig T, Thaiss CA, Kau AL, Eisenbarth SC, Jurczak MJ, et al: Inflammasome-mediated dysbiosis regulates progression of NAFLD and obesity. Nature 2012, 482:179-185. 44. De Minicis S, Rychlicki C, Agostinelli L, Saccomanno S, Candelaresi C, Trozzi L, Mingarelli E, Facinelli B, Magi G, Palmieri C, et al: Dysbiosis contributes to fibrogenesis in the course of chronic liver injury in mice. Hepatology 2014, 59:1738-1749. 45. Sprong RC, Hulstein MF, Van der Meer R: Bactericidal activities of milk lipids. Antimicrob Agents Chemother 2001, 45:1298-1301. 46. Pfeiffer N, Desmarchelier C, Blaut M, Daniel H, Haller D, Clavel T: Acetatifactor muris gen. nov., sp. nov., a novel bacterium isolated from the intestine of an obese mouse. Arch Microbiol 2012, 194:901-907. 47. Kubeck R, Bonet-Ripoll C, Hoffmann C, Walker A, Muller VM, Schuppel VL, Lagkouvardos I, Scholz B, Engel KH, Daniel H, et al: Dietary fat and gut microbiota interactions determine diet-induced obesity in mice. Mol Metab 2016, 5:1162-1174. 48. Zeidan YH, Pettus BJ, Elojeimy S, Taha T, Obeid LM, Kawamori T, Norris JS, Hannun YA: Acid ceramidase but not acid sphingomyelinase is required for tumor necrosis factor- {alpha}-induced PGE2 production. J Biol Chem 2006, 281:24695-24703. 49. Pfeiffer A, Bottcher A, Orso E, Kapinsky M, Nagy P, Bodnar A, Spreitzer I, Liebisch G, Drobnik W, Gempel K, et al: Lipopolysaccharide and ceramide docking to CD14 provokes ligand-specific receptor clustering in rafts. Eur J Immunol 2001, 31:3153-3164. 50. Grosch S, Schiffmann S, Geisslinger G: Chain length-specific properties of ceramides. Prog Lipid Res 2012, 51:50-62.

154

51. Bielawska A, Crane HM, Liotta D, Obeid LM, Hannun YA: Selectivity of ceramide- mediated biology. Lack of activity of erythro-dihydroceramide. J Biol Chem 1993, 268:26226- 26232. 52. Stiban J, Fistere D, Colombini M: Dihydroceramide hinders ceramide channel formation: Implications on apoptosis. Apoptosis 2006, 11:773-780. 53. Stiban J, Perera M: Very long chain ceramides interfere with C16-ceramide-induced channel formation: A plausible mechanism for regulating the initiation of intrinsic apoptosis. Biochim Biophys Acta 2015, 1848:561-567. 54. Shabbits JA, Mayer LD: Intracellular delivery of ceramide lipids via liposomes enhances apoptosis in vitro. Biochim Biophys Acta 2003, 1612:98-106. 55. Jozefowski S, Czerkies M, Lukasik A, Bielawska A, Bielawski J, Kwiatkowska K, Sobota A: Ceramide and ceramide 1-phosphate are negative regulators of TNF-alpha production induced by lipopolysaccharide. J Immunol 2010, 185:6960-6973. 56. Rozenova KA, Deevska GM, Karakashian AA, Nikolova-Karakashian MN: Studies on the role of acid sphingomyelinase and ceramide in the regulation of tumor necrosis factor alpha (TNFalpha)-converting enzyme activity and TNFalpha secretion in macrophages. J Biol Chem 2010, 285:21103-21113.

155

Chapter 6. Dietary Milk Phospholipids Improve Dyslipidemia and Hepatic Steatosis in LDLr-/- Mice Fed a Western- type Diet

156

6.1. Abstract

We have previously shown that chronic feeding of purified milk sphingomyelin (SM), a phospholipid (PL) component of milk fat globule membranes (MFGM), is protective against hepatic steatosis and dyslipidemia in high fat diet-fed mice. However, it is unclear whether ingestion of the total milk PL fraction can have effects similar to purified SM. Therefore, male

LDLr-/- mice were fed ad libitum a high-fat, high-cholesterol diet (CTL; 45% kcal from fat, 0.22% cholesterol added by weight; n = 15), or the same diet supplemented with 1% milk PL (1%; n =

15) or 2% milk PL (2%; n = 15) added by weight from beta serum for 14 weeks. No differences were detected in food intake, body weight, or tissue weights between the groups. Total serum cholesterol, non-HDL cholesterol, and non-esterified fatty acids (NEFAs) were significantly lower in the 2% group compared to CTL (-51%, -56%, and -26% respectively, p < 0.01). Hepatic total cholesterol concentrations were reduced by 53% (p < 0.05) and 55% (p < 0.01) compared to CTL in the 1% and 2% groups, respectively. These coincided with increased expression of hepatic

Hmgcr mRNA, and decreased expression levels of Scd1 and Mip1b in the 2% group. Epididymal adipose Ccl2 mRNA was reduced in the 2% group. Milk PL demonstrated hypolipidemic effects by strongly reducing serum and hepatic cholesterol pools in LDLr-/- mice fed a Western-type diet.

Beta serum could be utilized as a value-added source of milk PL to potentially improve hyperlipidemia-related disease outcomes.

6.2. Introduction

Current available research and results presented within this dissertation thus far have established dietary sphingomyelin (SM) as a potential mediator of chronic metabolic diseases. However, current commercially-available purified forms of dietary SM are likely too expensive to be used

157 as a daily supplement in humans (~$300-500 per gram). Therefore, we searched for whole foods or food byproducts, rich in dietary SM, that could be incorporated into diets or used for SM purification. Table 6.1. illustrates that beta serum is at least 2 times as concentrated in SM when compared to other common dietary sources. Beta serum is a polar lipid byproduct of anhydrous milk fat production [1]. Beta serum is particularly rich in milk polar lipids, including SM [2] and thus may provide a financially-viable dietary source of SM. It should be noted that milk polar lipids (MPLs) may influence the efficacy of milk-derived SM, as dietary MPLs have been shown to enhance SM absorption in animal studies [3]. Additionally, other dietary PLs have been shown to have beneficial effects in cardiometabolic disease [4]. Therefore, assessing MPLs’ effects in chronic metabolic disease is necessary. Some studies have demonstrated chronic feeding of MPL fractions reduces serum and hepatic lipids in rodents [5-7]. Dietary MPLs also reduce the inflammatory response with acute inflammatory challenges in mice [8, 9].

Table 6.4 Dietary SM content by source Content Source Ref (mg/100g) Bovine Milk, Whole 9 [10] Beef 44-69 [11] Egg 82 [12] Cottage Cheese 139 [13] Chicken Liver 291 [14] Beta Serum 420 [2] Beta Serum 1015 Manufacturer’s Report

Multiple rodent feeding studies have examined the effects of dietary MPLs on lipid homeostasis. Feeding a MPL extract at 1.7% to KK-Ay mice improved both serum and hepatic lipids [7]. Supplementing a high-fat diet (HFD) with MPL (2.5% wt:wt of the diet) lowered serum cholesterol, hepatic total lipids, hepatic cholesterol, and hepatic triglycerides, while increasing fecal lipids in mice, suggesting a decrease in lipid absorption [5]. However, not all animal studies

158 with MPLs have shown beneficial effects. For instance, a diet supplemented with 10% milk fat globular membranes (wt:wt, MFGM) increased serum triglycerides in rats compared to an isocaloric low-fat control diet [15]. Overall, in most animal studies dietary MPLs improved serum and hepatic lipids, most likely through preventing the absorption of dietary fat and/or luminal cholesterol.

In this chapter, we will examine MPLs for their efficacy in reducing diet-induced metabolic dysfunction and inflammation in low-density lipoprotein receptor-deficient (LDLr-/-) mice. When fed cholesterol-enriched HFD for 12 weeks, total cholesterol of mice reaches over 1000 mg/dL compared to 200-300 mg/dL on a control chow diet [16]. High-fat, high-cholesterol diets given to

LDLr-/- mice also exacerbate inflammation in adipose tissue [17] and livers [18, 19]. For these reasons, LDLr-/- mice are particularly useful as a model of NAFLD [16], the hepatic manifestation of metabolic dysfunction. This mouse model is particularly sensitive to the benefits of inhibiting cholesterol absorption. Treatment of Western-type diet-fed LDLr-/- mice with ezetimibe reduced some of the metabolic dysfunction-related consequences of the diet [18]. Therefore, we hypothesized that feeding LDLr-/- mice a high-fat, high-cholesterol, high-sucrose diet supplemented with MPLs from beta serum would attenuate the detrimental effects of the Western- type diet.

6.3. Materials and Methods

6.3.1 Animals and Diets

Male LDLr-/- mice on a C57BL/6J background (4 weeks old) were purchased from Jackson

Laboratory (Bar Harbor, ME, USA) and allowed to acclimate for 2 weeks before being fed one of three milk fat-based diets over 14 weeks: high-fat control (CTL; 45% kcal from fat, 0.15% cholesterol added by weight; n=15); CTL adjusted to contain 1% (wt:wt) of PLs (1%; n=15) or

159

CTL with 2% PLs added by weight (2%; n=15). Beta serum powder was used as a rich source of

PLs and was obtained from Tatua Dairy Cooperative (Morrinsville, New Zealand). The 1% and

2% PL diets supplied 0.2% and 0.4% milk-derived SM (wt:wt of the diet), respectively. Diets were adjusted to have similar mineral content, lactose, milk fat, and cholesterol. Skim milk was added to the CTL and 1% groups to account for milk bioactives outside of the MFGMs. Diet composition details are provided in Table 6.2. and 6.3. Experimental diets were made using ingredients obtained from Dyets, Inc. (Bethlehem, PA, USA). Body weight was assessed weekly, while food intake was estimated biweekly. Mice were provided with fresh food weekly. At 20 weeks old, the mice were fasted for 6–8 h prior to blood collection by cardiac puncture following euthanasia.

Blood samples were allowed to clot at room temperature for 30 min before serum was isolated by centrifugation (10,000g for 10 min at 4°C) and then stored at −80°C. Mice were perfused with sterile saline before tissues were harvested. The liver and epididymal adipose were collected from animals, snap-frozen in liquid nitrogen, and stored at −80°C. Sections of these tissues were fixed in 10% phosphate-buffered formalin. All mice were housed in a temperature-controlled room and maintained in a 12-h light/12-h dark cycle at the University of Connecticut-Storrs vivarium. The

Animal Care and Use Committee of the University of Connecticut-Storrs approved all procedures used in the current study.

6.3.2. Serum Biochemical Analysis

Total serum cholesterol, triglycerides, glucose, non-esterified fatty acids (NEFAs), and serum hepatic enzymes (ALT and AST) were measured using enzymatic or spectrophotometric assays according to manufacturer instructions. High-density lipoprotein cholesterol (HDL-C) was measured by quantifying serum cholesterol after precipitating apolipoprotein B (ApoB)-containing lipoproteins with Mg2+/dextran sulfate [20]. Cholesterol, triglycerides, and NEFA kits were

160 obtained from Wako Diagnostics (Richmond, VA). The glucose hexokinase kit and hepatic enzyme kits were obtained from Pointe Scientific, Inc. (Canton, MI). Non-HDL-C was calculated as the difference between total cholesterol and HDL-C. Insulin and Adiponectin was measured using commercially available ELISA (Crystal Chem, Elk Grove Village, IL). HOMA-IR was calculated as in Aim 2.

Table 6.2. Diet composition by component Diet Component CTL 1% 2% (g/kg) Casein 80 80 80 L-Cystine 3 3 3 Sucrose 200 200 200 Corn Starch 20 17.5 15 Lactose 21.35 10.68 0 Anhydrous Milkfat 207 181 155 Soybean Oil 20 20 20 Cellulose 50 50 50 Mineral Mix, AIN-93G-MX (94046) 43 43 43 Vitamin Mix, AIN-93-VX (94047) 19 19 19 Choline Bitartrate 3 3 3 TBHQ, antioxidant 0.04 0.04 0.04 Cholesterol 1.5 1.31 1.13 Skim milk powder (403150) 332 166 0 Butter serum 0 205.5 411 Calcium Carbonate 0 0.68 1.36 Potassium Chloride 0 0.19 0.38 Sodium Chloride 0 0.25 0.51

6.3.3. Hepatic Lipid Extraction and Analysis

Hepatic lipids were extracted from liver samples and analyzed as in Chapters 3 and 4.

161

Table 6.3. Diet composition by macronutrient Diet Component CTL 1% 2% Total Protein (g/kg) 200.2 200.3 200.4 Total Carbohydrate (g/kg) 414.0 411.5 409.0 Total Fat (g/kg) 229.7 231.3 233.0 Total Lactose (g/kg) 194.0 194.0 194.0 Total Cholesterol (g/kg) 2.06 2.06 2.06 Total Phospholipid (g/kg) 0.5 10.3 20.0 % kcal from Protein 17.7 17.7 17.7 % kcal from Carbohydrate 36.6 36.3 36.1 % kcal from Fat 45.7 46.0 46.2 Calorie Density (kcal/g) 4.52 4.52 4.52

6.3.4. Tissue Histology

Paraffin-embedded formalin-fixed liver and epididymal adipose tissues were cut into 5 μm sections before staining with hematoxylin and eosin (H&E) by the Connecticut Veterinary Medical

Diagnostic Laboratory (Storrs, CT). Each stained section was viewed under bright field microscopy at ×200 magnification and images were taken with an AxioCam ICc3 camera (Zeiss,

Thornwood, NY, USA).

3.3.5. RNA Isolation, cDNA Synthesis and qRT-PCR

Total RNA was isolated from liver and epididymal adipose tissue and used for cDNA synthesis as in previous chapters. Gene expression was normalized to the geometric mean of the reference genes, glyceraldehyde 3-phosphate dehydrogenase, β-actin and ribosomal protein, large, P0 using the 2−ΔΔCt method [21]. Primer sequences used for qRT-PCR analysis are listed in Appendix A.

162

3.3.6. Statistical Analysis

Differences between groups were evaluated by one-way ANOVA with post hoc comparisons

(Holm-Sidak) or the Kruskal-Wallis nonparametric test for non-normally distributed data (P <

0.05 deemed significant). D’Agostino-Pearson omnibus normality test was used to assess normality for each data set. All statistical analyses were conducted using GraphPad Prism version

6 software. Data are reported as mean ± SEM.

6.4. Results

6.4.1. Dietary Phospholipids Do Not Alter Body Weight and Food Intake

There were no differences between the groups in body weight, tissue weight, or tissue weight as a percentage of body weight (Table 6.4.). Expectedly, there were no significant differences in estimated daily food intake.

Table 6.4. Body and tissue weights after 14 weeks of diets CTL 1% 2% Body Weight (g) 34.25 ± 1.49 35.27 ± 1.34 33.19 ± 0.91 Food Intake (g/day) 3.40 ± 0.12 3.69 ± 0.13 3.73 ± 0.23 Liver (g) 1.44 ± 0.11 1.48 ± 0.10 1.29 ± 0.04 % Liver Weight 4.17 ± 0.16 4.16 ± 0.14 3.88 ± 0.06 Epididymal Adipose (g) 1.29 ± 0.18 1.61 ± 0.22 1.23 ± 0.19 % Epididymal Adipose 3.79 ± 0.53 4.34 ± 0.48 3.55 ± 0.50 Values are presented as mean ± SEM (n = 15)

6.4.2. MPL Treatment Attenuated Hypercholesterolemia and Reduced Circulating NEFAs

The 2% had a 50% reduction in total serum cholesterol compared to the CTL group (Table 6.5).

Additionally, HDL-C remained unchanged between the groups. After calculating non-HDL-C, it was found that estimated non-HDL-C was reduced by 53% in the 2% treatment. Circulating

NEFAs were 25% lower in the 2% group compared to the control. Fasting triglycerides and glucose were unchanged by either treatment. Additionally, serum hepatic enzymes ALT and AST were not

163 significantly different between the groups. Lastly, insulin, HOMA-IR, resistin, and adiponectin were similar between all three groups (Table 6.5).

Table 6.5. Serum values of LDLr-/- after 14 weeks of diets CTL 1% 2% Total Cholesterol (mg/dL) 1001 ± 149a 877 ± 123ab 491 ± 69b non-HDL-C (mg/dL) 928 ± 158a 840 ± 134ab 433 ± 74b HDL-C (mg/dL) 81 ± 5 83 ± 5 84 ± 5 Triglycerides (mg/dL) 404 ± 83 433 ± 78 268 ± 41 ALT (U/L) 64 ± 6 92 ± 20 59 ± 8 AST (U/L) 137 ± 18 140 ± 15 104 ± 7 NEFA (mmol/L) 0.83 ± 0.06a 0.83 ± 0.05a 0.62 ± 0.04b Glucose (mg/dL) 177 ± 15 178 ± 10 195 ± 12 Insulin (ng/mL) 0.11 ± 0.01 0.13 ± 0.01 0.10 ± 0.01 HOMA-IR 1.31 ± 0.20 1.50 ± 0.24 1.24 ± 0.13 Adiponectin (μg/mL) 6.18 ± 0.22 6.72 ± 0.26 6.38 ± 1.28 Resistin (ng/mL) 45.1 ± 2.0 42.6 ± 2.2 48.4 ± 2.8 Values are presented as mean ± SEM (n = 15), different superscripts indicate statistical significance (p = 0.05)

6.4.3. Hepatic Cholesterol Accumulation and Inflammation were reduced by MPLs

Compared to the CTL, the 2% group demonstrated reduced hepatic total cholesterol and cholesteryl esters by 56% and 74%, respectively. The 1% treatment reduced total cholesterol (-

50%) and free cholesterol (-23%) compared to the CTL group. Hepatic triglycerides were unchanged between the groups (Figure 6.1B). Histologically, the CTL H&E-stained liver sections appeared to have more lipid droplets compared to the 2%, although the steatosis appeared to be mild (Figure 6.1A).

Notably, the observed changes in hepatic cholesterol were corroborated in the 2% group by an increase in Hmgcr mRNA (+46%) compared to the control (Figure 6.1C). Although the hepatic triglycerides were not reduced in the treatment groups, Scd1 mRNA was significantly lower in the

2% group compared to CTL (-51%). Macrophage inflammatory protein 1β (Mip1b) was significantly lower (-77%) in the 2% treatment compared to the control. Chemokine (C-C) motif

164

Figure 6.1. Dietary milk phospholipids reduced hepatic cholesterol accumulation and inflammation. Liver sections were formalin- fixed before sectioned and H&E stained (A) (CTL, Control; 1%, 1% (wt:wt) MPL; 2%, 2% (wt:wt) MPL) for visualizing hepatic steatosis. Liver lipids were extracted using a modified Folch extraction and measured by enzymatic methods (B), (triglycerides, TG; total cholesterol, TC; cholesteryl ester, CE; free cholesterol, FC; choline-containing phospholipids, PL) (n = 15 per group, mean + SEM). Hepatic mRNA expression was determined using real-time qRT-PCR and standardized to the geometric mean of Gapdh and β-actin reference genes using (-ΔΔCt) the 2 method (D) (n = 13-14 per group, mean + SEM). Mean values with unlike letters indicate differences at P < 0.05 using post hoc comparisons.

165 ligand 2 (Ccl2) mRNA data was analyzed using the Kruskal-Wallis nonparametric test because the data was not normally distributed. While the mean expression of Ccl2 was lower in the 2% treatment group by 82%, the data failed to reach significance (p = 0.1). The macrophage marker

Adgre1 (encoding F4/80 protein) remained unchanged between the groups. Genes related to lipogenesis (Cd36, Dgat2, Nr1h3, Srebf2, Srebf1, and Pparg), lipolysis (Acad1 and Acox), and

Pepck remained unchanged by the dietary intervention.

6.4.4. Consuming MPLs Reduces Ccl2 mRNA Expression in the Epididymal Adipose

The epididymal adipose mRNA expression was assessed for potential changes in metabolism and inflammation. The expression of Ccl2 was attenuated in the 2% treatment by 56% compared to the CTL group (Figure 6.2B). In addition to this, Il6 mRNA expression appeared to be lower in both treatment groups; however, this failed to reach significance (p > 0.1). The mRNA of Slc2a4 was non-significantly greater in the 2% (p > 0.1) compared to CTL. Target genes of PPARγ

(aP2, AdipoQ, Lpl) were trending to be reduced in the 1% group compared to the CTL group.

The histology between the groups revealed no major differences in H&E staining of adipose tissue (Figure 6.2A).

6.5. Discussion

High-fat, high-cholesterol diets fed to LDLr-/- will induce hypercholesterolemia and hepatic lipid accumulation [16, 22]. Non-alcoholic steatohepatitis (NASH) will develop if LDLr-/- mice are exposed to these diets for >16 weeks [16]. Dietary PLs [23] and sphingolipids have been shown to inhibit lipid absorption and thus may slow the progression of NASH induced by a lipid-rich diet.

Beta serum is a dairy byproduct, rich in both PLs and sphingolipids. In this chapter, we demonstrated the hypolipidemic effect of dietary MPLs serum in genetically hypercholesterolemic

LDLr-/- mice. The higher dose of MPLs (2%) reduced serum and hepatic cholesterol levels as well

166

A B

p = 0 .2 4

) 3

q C T L

 

- 1 % 2

( p = 0 .2 2

2 % p = 0 .1 2 n

o 2 p = 0 .1 2 i a

s p = 0 .1 6

r p

x a b E

1 e

e R 0 2 l 1 8 c y 4 2 6 Q p r l l P o e 6 1 a c I a L r d 1 a 2 ip g d p c C d C P l d C S A A

Figure 6.2. MPLs mildly reduces adipose inflammation. Epididymal adipose tissue sections were formalin-fixed, sectioned and H&E stained (A). Epididymal adipose tissue mRNA expression was determined using real-time qRT-PCR and standardized to the geometric (-ΔΔCt) mean of Gapdh, 36B4, and β-actin reference genes using the 2 method (B) (n = 13-14 per group, mean + SEM). Mean values with unlike letters indicate differences at P < 0.05 using post hoc comparisons.

167 as reduced markers of hepatic and adipose tissue inflammation compared to controls.

Unsurprisingly, the lower dose treatment (i.e. 1% (wt:wt) MPL) was less effective, as it only reduced hepatic cholesterol concentration. These effects may be suggestive of inhibitory effects on cholesterol absorption because reducing cholesterol absorption has been shown to improve dyslipidemia, hepatic steatosis, and adipose inflammation in animal studies similar to a study examining the effects of ezetimibe on this model [17].

In this study we used increasing doses of beta serum to supply 1% and 2% (wt:wt) MPLs.

This is equivalent to 6 and 12 g of dietary PL/day in a 70-kg adult human. This will contribute

0.21-0.42% milk-derived SM, or 1.2-2.4 g/day in an adult. The typical western diet consists of 6-

10 g/day of PLs and 0.3-0.4 g/day of sphingolipids [24]. The higher dose of MPLs reduced serum total cholesterol and non-HDL-C. Due to the fact that hepatic cholesterol was also reduced and the adipose phenotype was largely unchanged between the CTL and 2% group, it seems likely that some of this effect could be attributed to the inhibition of cholesterol absorption attributed to both

PLs and sphingolipids [4]. It is interesting to compare these dosages of SM to the previous studies on dietary SM. The 1% PL diet supplied ~0.21% (wt:wt of the diet) SM to the animals, a dose previously shown to be effective in reducing hyperlipidemia and steatosis [25, 26]. Yet, the 1%

PL diet did not have a strong effect on serum lipids and only reduced hepatic cholesterol levels.

This could be due to differences in diets and/or rodent models, however it is interesting to note that dietary PLs have been shown to increase the appearance of SM-derived ceramide in the lymph

[27]. Therefore, additional PLs found in the beta serum may increase the absorption of SM and decrease its efficacy by removing it from the lumen. High doses of dietary PLs have also been shown to inhibit cholesterol absorption [28-30]. The higher dose of beta serum could be providing enough PLs to inhibit lipid absorption. Alternatively, the additional SM (~0.42% wt:wt) could be

168 compensating for the increase in SM absorption. These results highlight the necessity to account for the food matrix when interpreting data concerning dietary SM. The effects of a complex PL mixture like this has yet to be tested directly in models of cholesterol or fatty acid absorption.

Therefore, directly comparing the efficacy of beta serum to its individual components could be an effective means to understand the mechanism.

There have been several studies reporting beneficial effects on lipid homeostasis similar to this study. After 5 weeks on a HFD supplemented with MPLs (2.5% wt:wt of the diet), mice in the treatment group had reduced serum cholesterol, hepatic total lipids, hepatic cholesterol, and hepatic triglycerides, while increasing fecal lipids in mice compared to HFD control [5]. KK-Ay mice supplemented with MPL also improved both serum and hepatic lipids [7]. Similarly, 8 weeks on a HFD, hepatic lipids and serum cholesterol as well as triglycerides were improved by supplementation of 2.5% milk polar lipids [6]. Mice supplemented with MFGM on a HFD have an improved lipid profile [31]. In the context of these studies, this aim further demonstrates the ability of dietary MPL to reduce diet-induced dyslipidemia and hepatic lipid accumulation.

Some clinical trials have examined the effects of MPL supplements on lipid homeostasis.

Overweight adults in a single-blind, randomized, controlled, isocaloric parallel study for 8 weeks, treated with 40 g of milkfat/day as whipping cream significantly reduced plasma LDL-C and apoB:apoA-I ratio compared to a milk fat control [32]. Whipping cream is particularly rich in PLs.

Two clinical trials feeding buttermilk found slightly different effects. First, Ohlsson et al. observed that a 4-week supplementation of a buttermilk drink enriched in MFGM, in a placebo-controlled parallel study with healthy adults had no effect on the plasma lipids. However, a non-significant reduction in plasma LDL-C (p = 0.056) in women was observed [33]. This study was likely underpowered with a small sample size and a parallel design. However, 4-week supplementation

169 with 45 g/day of buttermilk significantly reduced both plasma cholesterol and TG in healthy adults

[34]. This study was a double-blind, randomized, crossover study and indirectly measured cholesterol absorption through serum β-sitosterol concentration. The β-sitosterol concentration values were correlated with the reductions in plasma LDL-C. suggesting a reduction in cholesterol absorption may have contributed to the reduction in LDL-C [34]. This study was much more robust with 38 participants, and the crossover design improved statistical power. In aggregate, there is some clinical data that suggests the hypolipidemic effects found in animal studies described above is translatable to humans.

Hepatic mRNA analysis revealed a relative increase in Hmgcr mRNA expression in the

2% treatment group compared to the CTL. This is likely reflective of a compensatory mechanism in the liver. A high-cholesterol diet has been shown to decrease hepatic mRNA expression of

Hmgcr in mice [35]. If MPLs are reducing cholesterol absorption, the dietary cholesterol is likely not suppressing Hmgcr mRNA to the same degree as in the control diet. Similarly, Scd1 mRNA expression was reduced in the 2% group compared to CTL. High-fat diets can induce hepatic Scd1 mRNA in mice [36]. Therefore, reducing lipid absorption will cause the HFD to induce Scd1 less strongly. These findings are comparable to the results of Aims 1 and 2, where SM treatment reduced Scd1 and increased Hmgcr mRNA levels in liver. Additionally, Mip1b expression was significantly reduced by the 2% group compared to the CTL group, while Ccl2 was trending towards significance, suggesting a decrease in hepatic inflammation. However, both Mip1b and

Ccl2 mRNA expression levels were variable in the CTL, suggesting the duration of the diet may have been too short to induce hepatitis, or that the diet did not have enough lipids. Most studies of high-fat, high-cholesterol steatohepatitis are >16 weeks and have used higher fat content diets [37-

39]. Additionally, some studies of LDLr-/- mice reveal an increase locomotor activity [40].

170

LDLr-/- mice fed a HFD had increased thermogenesis, and decreased food intake [41]. This could hinder their response to a HFD compared to wild-type controls [41, 42]. Therefore, although LDLr-

/- mice are a useful model of liver disease they are more resistant to early diet-induced metabolic dysfunction and inflammation.

Analysis of the epididymal adipose mRNA for metabolic and inflammatory changes showed a reduction in Ccl2 mRNA. This suggests the adipose in this treatment group is somewhat less inflamed than in the CTL, which is corroborated by the significantly lower circulating NEFAs.

Recently, it was shown that feeding increasing doses of MFGM reduces adipose weight as well as the expression of lipogenic genes and proteins in adipose tissue [31] returning the high fat-fed mice’s phenotype similar to the low-fat control. Whether inhibiting cholesterol absorption could be solely responsible for a decrease in adipose tissue inflammation in this model remains uncertain.

Ezetimibe has been shown to lower both total cholesterol and hepatic cholesterol [17]. Adding excessive cholesterol to the diet has been shown to induce adipose tissue inflammation in LDLr-/- mice [17]. However, treatment of LDLr-/- mice with ezetimibe has shown only marginal effects on adipose tissue inflammation [43]. Therefore, an independent effect of dietary MPL may be responsible for the reduction in adipose inflammation.

The potential anti-inflammatory effects of MPL cannot be ignored. Multiple recent studies have tested MPL fractions for anti-inflammatory effects. When examining acute inflammation in mice, 5 weeks of dietary MPL reduced the cytokine response within 24 hours to LPS intraperitoneal injection, while increasing survival rates at 48 hours [8]. Following 8 weeks of a

HFD, mice fed MPL had decreased expression of adipose inflammatory markers (e.g., cluster of differentiation 68) compared to HFD control. This was in contrast to a group of mice fed soy PL, whom had increased responses of inflammatory markers [9]. In humans, adding MFGM to palm

171 oil-based drinks ingested by overweight and obese men reduced soluble intracellular adhesion molecule, while increasing IL-10 in the post-prandial state [44]. This suggests an improvement in post-prandial inflammation, which is usually observed in obese populations and has been considered a predictor of metabolic disease [45]. Ultimately, there is some evidence for the anti- inflammatory effects of MFGM or MPL-fractions, which would be beneficial to improve metabolic outcomes.

The higher dose was effective in attenuating the hypercholesterolemia and hepatic cholesterol accumulation associated with our model. Additionally, there were modest reductions in hepatic and adipose tissue inflammation, which may have been limited by the length of the study. Regardless, MPLs have potential to function as a preventative treatment for metabolic dysfunction. Therefore, more research is warranted to further understand all the potential mechanisms and the translatability to humans.

6.6. Acknowledgments

This work was supported by a research grant to C. Blesso from the National Dairy Council (Dairy

Management Inc.).

172

6.7. References

1. Vanderghem C, Bodson P, Danthine S, Paquot M, Deroanne C, Blecker C: Milk fat globule membrane and buttermilks: from composition to valorization. Biotechnologie 2010, 14:485-500. 2. Contarini G, Povolo M: Phospholipids in Milk Fat: Composition, Biological and Technological Significance, and Analytical Strategies. International Journal of Molecular Sciences 2013, 14:2808- 2831. 3. Lecomte M, Bourlieu C, Meugnier E, Penhoat A, Cheillan D, Pineau G, Loizon E, Trauchessec M, Claude M, Menard O, et al: Milk Polar Lipids Affect In Vitro Digestive Lipolysis and Postprandial Lipid Metabolism in Mice. J Nutr 2015, 145:1770-1777. 4. Blesso CN: Egg phospholipids and cardiovascular health. Nutrients 2015, 7:2731-2747. 5. Kamili A, Wat E, Chung RW, Tandy S, Weir JM, Meikle PJ, Cohn JS: Hepatic accumulation of intestinal cholesterol is decreased and fecal cholesterol excretion is increased in mice fed a high-fat diet supplemented with milk phospholipids. Nutrition & Metabolism 2010, 7:90. 6. Wat E, Tandy S, Kapera E, Kamili A, Chung RW, Brown A, Rowney M, Cohn JS: Dietary phospholipid-rich dairy milk extract reduces hepatomegaly, hepatic steatosis and hyperlipidemia in mice fed a high-fat diet. Atherosclerosis 2009, 205:144-150. 7. Watanabe S, Takahashi T, Tanaka L, Haruta Y, Shiota M, Hosokawa M, Miyashita K: The effect of milk polar lipids separated from butter serum on the lipid levels in the liver and the plasma of obese-model mouse (KK-Ay). Journal of Functional Foods 2011, 3:313-320. 8. Snow DR, Ward RE, Olsen A, Jimenez-Flores R, Hintze KJ: Membrane-rich milk fat diet provides protection against gastrointestinal leakiness in mice treated with lipopolysaccharide. J Dairy Sci 2011, 94:2201-2212. 9. Lecomte M, Couëdelo L, Meugnier E, Plaisancié P, Létisse M, Benoit B, Gabert L, Penhoat A, Durand A, Pineau G, et al: Dietary emulsifiers from milk and soybean differently impact adiposity and inflammation in association with modulation of colonic goblet cells in high-fat fed mice. Molecular Nutrition & Food Research 2016, 60:609-620. 10. Morrison WR: The distribution of phospholipids in some mammalian milks. Lipids 1968, 3:101-103. 11. Nakanishi T, Suyama K: Studies on phospholipids of meat VIII. Reaction of carbonyl compounds with phospholipids (Part 1) Formation of α, β- unsaturated aldehydes by heat treatment of a mixture of saturated aldehyde and phospholipids. Nihon Chikusan Gakkaiho 1969, 40:320-326. 12. Christie WW, Moore JH: The lipid composition and triglyceride structure of eggs from several avian species. Comparative Biochemistry and Physiology Part B: Comparative Biochemistry 1972, 41:297-306. 13. Weihrauch JL, Son YS: Phospholipid content of foods. Journal of the American Oil Chemists' Society 1983, 60:1971-1978. 14. Marion JE, Miller WO: Phospholipids and Component Fatty Acids of Chicken Tissuesa. Poultry Science 1968, 47:1453-1459. 15. Zhou AL, Hintze KJ, Jimenez-Flores R, Ward RE: Dietary Fat Composition Influences Tissue Lipid Profile and Gene Expression in Fischer-344 Rats. Lipids 2012, 47:1119-1130. 16. Bieghs V, Van Gorp PJ, Wouters K, Hendrikx T, Gijbels MJ, van Bilsen M, Bakker J, Binder CJ, Lütjohann D, Staels B, et al: LDL Receptor Knock-Out Mice Are a Physiological Model Particularly Vulnerable to Study the Onset of Inflammation in Non-Alcoholic Fatty Liver Disease. PLOS ONE 2012, 7:e30668. 17. Subramanian S, Han CY, Chiba T, McMillen TS, Wang SA, Haw A, Kirk EA, O’Brien KD, Chait A: Dietary cholesterol worsens adipose tissue macrophage accumulation and atherosclerosis in obese LDL receptor-deficient mice. Arteriosclerosis, thrombosis, and vascular biology 2008, 28:685-691.

173

18. Umemoto T, Subramanian S, Ding Y, Goodspeed L, Wang S, Han CY, Teresa AS, Kim J, O'Brien KD, Chait A: Inhibition of intestinal cholesterol absorption decreases atherosclerosis but not adipose tissue inflammation. Journal of Lipid Research 2012, 53:2380-2389. 19. Wouters K, van Gorp PJ, Bieghs V, Gijbels MJ, Duimel H, Lutjohann D, Kerksiek A, van Kruchten R, Maeda N, Staels B, et al: Dietary cholesterol, rather than liver steatosis, leads to hepatic inflammation in hyperlipidemic mouse models of nonalcoholic steatohepatitis. Hepatology 2008, 48:474-486. 20. Warnick GR, Benderson J, Albers JJ: Dextran sulfate-Mg2+ precipitation procedure for quantitation of high-density-lipoprotein cholesterol. Clin Chem 1982, 28:1379-1388. 21. Rao X, Huang X, Zhou Z, Lin X: An improvement of the 2ˆ(–delta delta CT) method for quantitative real-time polymerase chain reaction data analysis. Biostatistics, bioinformatics and biomathematics 2013, 3:71-85. 22. Wu L, Vikramadithyan R, Yu S, Pau C, Hu Y, Goldberg IJ, Dansky HM: Addition of dietary fat to cholesterol in the diets of LDL receptor knockout mice: effects on plasma insulin, lipoproteins, and atherosclerosis. Journal of Lipid Research 2006, 47:2215-2222. 23. Cohn JS, Kamili A, Wat E, Chung RWS, Tandy S: Dietary Phospholipids and Intestinal Cholesterol Absorption. Nutrients 2010, 2:116-127. 24. Vesper H, Schmelz EM, Nikolova-Karakashian MN, Dillehay DL, Lynch DV, Merrill AH, Jr.: Sphingolipids in food and the emerging importance of sphingolipids to nutrition. J Nutr 1999, 129:1239-1250. 25. Chung RW, Kamili A, Tandy S, Weir JM, Gaire R, Wong G, Meikle PJ, Cohn JS, Rye KA: Dietary sphingomyelin lowers hepatic lipid levels and inhibits intestinal cholesterol absorption in high-fat- fed mice. PLoS One 2013, 8:e55949. 26. Duivenvoorden I, Voshol PJ, Rensen PC, van Duyvenvoorde W, Romijn JA, Emeis JJ, Havekes LM, Nieuwenhuizen WF: Dietary sphingolipids lower plasma cholesterol and triacylglycerol and prevent liver steatosis in APOE*3Leiden mice. Am J Clin Nutr 2006, 84:312-321. 27. Morifuji M, Higashi S, Oba C, Ichikawa S, Kawahata K, Yamaji T, Itoh H, Manabe Y, Sugawara T: Milk Phospholipids Enhance Lymphatic Absorption of Dietary Sphingomyelin in Lymph- Cannulated Rats. Lipids 2015, 50:987-996. 28. Eckhardt ER, Wang DQ, Donovan JM, Carey MC: Dietary sphingomyelin suppresses intestinal cholesterol absorption by decreasing thermodynamic activity of cholesterol monomers. Gastroenterology 2002, 122:948-956. 29. Hollander D, Morgan D: Effect of plant sterols, fatty acids and lecithin on cholesterol absorption in vivo in the rat. Lipids 1980, 15:395-400. 30. Jiang Y, Noh SK, Koo SI: Egg phosphatidylcholine decreases the lymphatic absorption of cholesterol in rats. J Nutr 2001, 131:2358-2363. 31. Li T, Gao J, Du M, Song J, Mao X: Milk Fat Globule Membrane Attenuates High-Fat Diet- Induced Obesity by Inhibiting Adipogenesis and Increasing Uncoupling Protein 1 Expression in White Adipose Tissue of Mice. Nutrients 2018, 10:331. 32. Rosqvist F, Smedman A, Lindmark-Mansson H, Paulsson M, Petrus P, Straniero S, Rudling M, Dahlman I, Riserus U: Potential role of milk fat globule membrane in modulating plasma lipoproteins, gene expression, and cholesterol metabolism in humans: a randomized study. Am J Clin Nutr 2015, 102:20-30. 33. Ohlsson L, Burling H, Nilsson A: Long term effects on human plasma lipoproteins of a formulation enriched in butter milk polar lipid. Lipids Health Dis 2009, 8:44. 34. Conway V, Couture P, Richard C, Gauthier SF, Pouliot Y, Lamarche B: Impact of buttermilk consumption on plasma lipids and surrogate markers of cholesterol homeostasis in men and women. Nutr Metab Cardiovasc Dis 2013, 23:1255-1262. 35. Hwang S, Hartman IZ, Calhoun LN, Garland K, Young GA, Mitsche MA, McDonald J, Xu F, Engelking L, DeBose-Boyd RA: Contribution of Accelerated Degradation to Feedback Regulation of

174

3-Hydroxy-3-methylglutaryl Coenzyme A Reductase and Cholesterol Metabolism in the Liver. The Journal of Biological Chemistry 2016, 291:13479-13494. 36. Hu CC, Qing K, Chen Y: Diet-induced changes in stearoyl-CoA desaturase 1 expression in obesity-prone and -resistant mice. Obes Res 2004, 12:1264-1270. 37. Savard C, Tartaglione EV, Kuver R, Haigh WG, Farrell GC, Subramanian S, Chait A, Yeh MM, Quinn LS, Ioannou GN: Synergistic interaction of dietary cholesterol and dietary fat in inducing experimental steatohepatitis. Hepatology 2013, 57:81-92. 38. Lieber CS, Leo MA, Mak KM, Xu Y, Cao Q, Ren C, Ponomarenko A, DeCarli LM: Model of nonalcoholic steatohepatitis. Am J Clin Nutr 2004, 79:502-509. 39. Takahashi Y, Soejima Y, Fukusato T: Animal models of nonalcoholic fatty liver disease/nonalcoholic steatohepatitis. World Journal of Gastroenterology : WJG 2012, 18:2300-2308. 40. Elder GA, Ragnauth A, Dorr N, Franciosi S, Schmeidler J, Haroutunian V, Buxbaum JD: Increased locomotor activity in mice lacking the low-density lipoprotein receptor. Behavioural brain research 2008, 191:256-265. 41. Ngai YF, Quong WL, Glier MB, Glavas MM, Babich SL, Innis SM, Kieffer TJ, Gibson WT: Ldlr- /- mice display decreased susceptibility to Western-type diet-induced obesity due to increased thermogenesis. Endocrinology 2010, 151:5226-5236. 42. Karagiannides I, Abdou R, Tzortzopoulou A, Voshol PJ, Kypreos KE: Apolipoprotein E predisposes to obesity and related metabolic dysfunctions in mice. Febs j 2008, 275:4796-4809. 43. Umemoto T, Subramanian S, Ding Y, Goodspeed L, Wang S, Han CY, Teresa AS, Kim J, O'Brien KD, Chait A: Inhibition of intestinal cholesterol absorption decreases atherosclerosis but not adipose tissue inflammation. J Lipid Res 2012, 53:2380-2389. 44. Demmer E, Van Loan MD, Rivera N, Rogers TS, Gertz ER, German JB, Smilowitz JT, Zivkovic AM: Addition of a dairy fraction rich in milk fat globule membrane to a high-saturated fat meal reduces the postprandial insulinaemic and inflammatory response in overweight and obese adults. J Nutr Sci 2016, 5:e14. 45. Klop B, Proctor SD, Mamo JC, Botham KM, Castro Cabezas M: Understanding postprandial inflammation and its relationship to lifestyle behaviour and metabolic diseases. Int J Vasc Med 2012, 2012:947417.

175

Chapter 7. Summary, Conclusions, and Future Directions

176

7.1. Summary and Conclusions

The focus of this dissertation was to further define the effects of dietary SM on diet-induced metabolic dysfunction and inflammation. Our central hypothesis was that dietary SM would reduce dyslipidemia, metabolic dysfunction, and inflammation in mice fed Western-type diets. In Aim 1, we sought to determine the potential differences in the effects of dietary milk-derived and egg- derived SM on serum and hepatic lipid concentrations. As expected, milk-derived SM improved both hepatic lipid accumulation and serum cholesterol. Surprisingly, egg-derived SM increased both hepatic lipid accumulation and serum lipids. These findings led to Aim 2, where we investigated both egg- and milk-derived SM effects on preventing metabolic dysfunction in an obesogenic diet. In contrast to Aim 1, egg-derived SM reduced hypercholesterolemia and hyperglycemia. In this aim, both SM treatments reduced hepatic steatosis and adipose inflammation, although egg-derived SM was more effective than milk-derived SM. While the first two aims highlight differences in egg-derived and milk-derived SM in the context of SM and its hypolipidemic effects, for Aim 3 we targeted the anti-inflammatory effects of SM. In Aim 3, we found that feeding milk-derived SM reduced systemic cytokines and LPS in mice fed an obesogenic diet. Most in vitro studies have demonstrated anti-inflammatory effects of exogenous sphingolipids; however, the mechanism is poorly understood. Therefore, we used LPS-stimulation in RAW264.7 macrophages to examine the anti-inflammatory effects of SM, ceramide, and sphingosine. We observed all three compounds to be effective, and that SM hydrolysis was required for SM effectiveness. Purified SM was used in Aims 1-3, however, consumption of dietary SM found in foods will be accompanied by other dietary phospholipids. Therefore, we also tested MPL supplied by beta-serum, a dairy byproduct that is rich in phospholipids, including SM, on diet-induced metabolic dysfunction and inflammation in genetically hypercholesterolemic

177 mice. We found that the highest dose of MPL (2% of diet by weight) reduced hypercholesterolemia, adipose tissue inflammation, hepatic cholesterol and inflammation. A detailed review of the findings from each of the 4 specific aims follows and is described in Figure

7.1..

Aim 1: To evaluate differential effects of dietary egg-derived SM and milk-derived SM on dyslipidemia and metabolic dysfunction in mice fed a short-term high-fat diet.

Hypothesis 1: We hypothesized that both SM supplements would improve Western-type diet- induced metabolic dysfunction and endotoxemia, but milk-derived SM would be more potent compared to egg-derived SM. This difference would be due to milk-derived SM’s more potent inhibitory effect on lipid absorption compared to egg-derived SM.

Findings of Aim 1: There were clear and distinct effects of milk and egg SM. Milk-derived SM reduced weight gain and serum cholesterol compared to CTL. Conversely, egg-derived SM treatment increased serum concentrations of cholesterol, triglycerides, phospholipids, and SM compared to CTL. Milk-derived SM significantly decreased, while egg-derived SM increased, hepatic triglycerides. Milk-derived SM-fed mice displayed gene expression changes consistent with cholesterol depletion in the intestine and liver. In terms of inflammation, serum LPS was reduced in the milk-derived SM compared to CTL. Differences in serum LPS may have been related to changes in gut microbiota. The fecal Gram-negative bacteria were significantly lower, and fecal Bifidobacterium were higher, in the milk-derived SM group. There was no effect of dietary SM on fasting gut permeability. Thus, milk-derived SM has potential to reduce the early stages of diet-induced metabolic dysfunction.

178

Aim 2: To determine the effects of dietary egg-derived SM and milk-derived SM on the development of high-fat diet-induced NAFLD, inflammation, and metabolic dysfunction.

Hypothesis 2: Both SM treatment groups would reduce the metabolic dysfunction phenotype due to SM’s effects on both lipid absorption and inflammation compared to high-fat control. However, milk-derived SM would be more effective than egg-derived SM because of the stronger inhibitory effect on lipid absorption.

Findings of Aim 2: After 10 weeks, egg-derived SM ameliorated weight gain, hypercholesterolemia, and hyperglycemia induced by HFD. Both egg- and milk-derived SM attenuated hepatic steatosis development, with significantly lower hepatic TG and cholesterol relative to the HFD control. This reduction in hepatic steatosis was stronger with egg-derived SM supplementation relative to milk-derived SM. Reductions in hepatic TG observed with dietary SM were corroborated by lower hepatic mRNA expression of PPARγ-related genes; Scd1 and Pparg2 in both SM groups, and Cd36 and Fabp4 with egg SM. Egg-derived SM, and to a lesser extent milk-derived SM, reduced inflammation and markers of macrophage infiltration in adipose tissue.

Egg-derived SM also reduced skeletal muscle TG content compared to HFD control. Aim 2 highlights dietary SM’s ability to inhibit diet-induced metabolic dysfunction associated with obesity.

Aim 3: To evaluate the effects of dietary SM on HFD-induced inflammation and determine if SM and its metabolites directly affect macrophage inflammation.

Hypothesis 3: We hypothesized that dietary SM would inhibit diet-induced inflammation and exogenous SM treatment would inhibit LPS-stimulation of RAW264.7 macrophages. Due to the fact that SM metabolites inhibitory properties, we anticipated that blocking hydrolysis of SM

179 would reduce SM’s effectiveness for inhibiting LPS-stimulation. Consequently, the addition of the hydrolytic products of SM would also inhibit LPS-stimulation. Furthermore, dihydroceramide and ceramide would have differential effects on macrophage inflammatory response, as they have previously shown different bioactivities in other contexts.

Findings of Aim 3: Dietary milk-derived SM significantly reduced inflammatory serum cytokines and tended to lower serum LPS (p = 0.08) compared to HFD. Cecal gut microbiota analysis revealed that milk-derived SM increased the abundance of Acetatifactor genus, with minimal changes in other taxa. Milk SM significantly attenuated the in vitro effect of LPS on pro- inflammatory gene expression in RAW264.7 macrophages. This effect was dependent on SM hydrolysis, as blocking sphingomyelinase activity caused SM to be ineffective. Additionally, the metabolites ceramide and sphingosine inhibited the pro-inflammatory changes. Overall, this aim revealed there are possibly direct anti-inflammatory effects of dietary SM. Additionally, these anti- inflammatory effects may be independent of the effects of lipid absorption and are likely dependent on SM hydrolysis.

Aim 4: To evaluate the effects of dietary milk phospholipids, rich in SM, on the development of metabolic dysfunction in LDLr-/- mice fed a high-fat, high-cholesterol diet.

Hypothesis 4: Phospholipid supplementation would reduce dyslipidemia and insulin resistance in

Western-type diet-fed LDLr-/- mice. This would result in an improved hepatic phenotype and decreased adipose inflammation.

Findings of Aim 4: Supplementation of milk phospholipids (2% PL) reduced total serum cholesterol, non-HDL cholesterol, and NEFAs compared to CTL (-51%, -56%, and -26% respectively). Hepatic total cholesterol concentrations were reduced compared to CTL by 53% and

180

55% in the 1% and 2% groups, respectively. These changes came with an increased hepatic Hmgcr mRNA, decreased Scd1 and Mip1b in the 2% group. Histology analysis corroborated these findings although the steatosis was mild. Ccl2 mRNA in the epididymal adipose tissue was reduced in the 2% group. This aim provides further evidence for MPL to prevent diet-induced dyslipidemia and hepatic steatosis. Therefore, MPL extracts could be a useful source of dietary bioactives to potentially combat metabolic complications of obesity.

Figure 7.1. Schematic of dietary sphingomyelin’s effects on metabolic dysfunction. Bolded effects represent findings from this work. Lined effects are well-established within the literature. Italicized effects are future directions based off of findings in this work. Solid lines represent known connections between effects, while dashed lined are potential connections.

7.2. Future Directions

7.2.1. Understanding the Impact of Food Matrix on Dietary SM

Although this dissertation has contributed to the literature that supports a role for dietary SM to limit lipid absorption and influence lipid metabolism, we still do not completely understand the mechanism. Sphingomyelin’s effect on inhibiting lipid absorption has been attributed partly to an inhibition of pancreatic lipase activity [1, 2]. Despite pancreatic lipase having a sphingolipid-

181 sensing domain, SM seemingly does not bind to a synthetic mimic of pancreatic lipase [3], and therefore, it is not likely to be a direct inhibition. Alternatively, SM inhibition of pancreatic lipase may be due in part to the displacement of colipase from pancreatic lipase [2]. The mechanism of this is not completely clear, but it could be a result from increased lipid ordering caused by

SM:cholesterol interactions [4]. Order can be tested using a laurdan probe [5]. This could be tested by making micelles with varying concentrations of SM, cholesterol as well as saturated and unsaturated phosphatidylcholine to generate micelles of different rigidities measured by the laurdan probe. These micelles could be incubated with lipase and colipase and the association between them can be measured as a ratio, as previously reported [2]. Comparing degrees of ordering and colipase:lipase interactions would allow us to see if there is an effect of lipid ordering on lipase activity. Additionally, comparing micelles of similar rigidities containing different amounts of SM would elucidate if it is a specific interaction of SM or the consequence of lipid ordering. Another potential explanation of SM effects on lipid absorption is its potential for affecting proteins involved in cholesterol uptake by intestinal cells. Recently, exogenous SM was shown to reduce NPC1L1 protein expression in Caco-2 cells [6]. Although this runs counter to our in vivo mRNA findings of Aim 1 thought to be a consequence of cellular cholesterol reduction by dietary SM, it has been shown that increasing membrane rigidity through lysophosphatidylcholine acyltransferase 3 knockout, disrupts NPC1L1 protein expression in the intestine [7]. If the membrane is more rigid and prevents NPC1L1 to incorporation, it may be dislocated and degraded, a change that may not be captured by mRNA analysis. Therefore, quantifying \ the protein expression of NPC1L1 and other membrane proteins in the intestines of mice fed a long-term SM- rich diet may be of interest to confirm these recent findings [6].

182

Overall, our understanding of the interactions of the food matrix and SM’s efficacy is sparse. This is highlighted by the discrepancy in the findings between Aim 1, where milk-derived

SM was more hypolipidemic, and Aim 2, where egg-derived SM was more effective. There are several differences in these studies, including diet composition, SM dosage, and diet duration.

Since the hypolipidemic effects are thought to be due to the inhibition of lipid absorption, these discrepancies are likely due to differences among dietary-derived SM structures. Therefore, the likely important variables are the dosage or diet composition. This work started under the assumption that milk-derived SM more strongly inhibits lipid absorption than egg-derived SM [8].

However, this direct comparison may not have been fully explored. A major difference between the diet that can be explored is the compositions in Aims 1 and 2. Aim 2 added a substantial amount of cholesterol. First, the ratio of SM to cholesterol can alter cholesterol absorption [9, 10], however, this is not necessarily in a linear fashion. Therefore, the relationship may be more complex than more SM leads to a greater inhibition of lipid absorption. If this is true, there may be different optimal ratios of cholesterol:SM for different structurally-distinct sources of SM. Current absorption studies, with the exception of one [8], have only used a single source of SM. Therefore, testing the effects of several ratios of egg- and milk-derived SM to cholesterol on lipid absorption using both cell culture and rodent models could help explain differential effects in Aim 1 and 2.

Lastly, dietary phospholipids have been shown to increase the appearance of milk SM- derived ceramide species [11]. This is possibly due to increased SM hydrolysis because there is little evidence of SM being absorbed whole. However, this would run counter to previous in vitro studies where phospholipids inhibited alkaline sphingomyelinase [12]. Therefore, this study could be repeated with intestine-specific alkaline sphingomyelinase knockout mice [13]. If milk

183 phospholipids are unable to increase lymphatic milk SM-derived ceramides in this model, the milk phospholipids are likely enhancing SM digestion.

Using the dual-isotope method [9], it would be important to understand if administration of SM as a pure compound or as part of a mixture, such as in beta-serum, has differential effects on cholesterol and lipid absorption. This was suggested by the finding that 1% MPLs (wt:wt) had little effect on serum cholesterol in Aim 4, while purified SM at similar concentrations to lowered serum cholesterol in Aims 1 and 2, as well as in other studies [14]. However, direct comparison of these studies is insufficient to conclude that dietary phospholipids can alter SM’s effects on cholesterol absorption, because these studies use significantly different animal models and intestinal cholesterol absorption is not the only factor contributing to hypercholesterolemia.

Therefore, directly comparing purified milk-derived SM and beta-serum would provide evidence of how the phospholipids alter SM’s efficacy.

7.2.2 Further Elucidating the Molecular Mechanism of SM’s Anti-inflammatory Effects

The effects of exogenous and endogenous sphingolipids on inflammation are contradictory. In general, research has found most exogenous sphingolipids to be anti-inflammatory [15-17], but the mechanism is poorly understood. In Aim 3, we managed to determine that the hydrolysis of

SM is necessary for the anti-inflammatory effects. Replicating this finding in bone marrow-derived macrophages would further strengthen these findings.

One aspect that remains unclear is if ceramide digestion is required for the anti- inflammatory effects observed. Sphingosine’s capacity to inhibit inflammation certainly suggests ceramide digestion is necessary. To test this, we will need to repeat the SM experiments while inhibiting acid ceramidase, which is also secreted from RAW 264.7 macrophages [18]. We could make use of the newly synthesized ceramidase inhibitor, LCL521, which targets the lysosome, to

184 hopefully avoid off target effects [19]. This would inform us if ceramide hydrolysis is required for the anti-inflammatory effects of SM and further resolve which metabolite are responsible for the effects found in Aim 3.

Assuming sphingosine is the causal agent, it would be important to use radiolabeled sphingosine to examine the fate of sphingosine within the cell in both inflammatory and non- inflammatory conditions. To do this, we could simply repeat the experiment from Aim 3 with radiolabeled sphingosine and separate the lipids of cell lysates using thin layer chromatography.

This would provide insight to whether it is sphingosine or a metabolite of exogenous sphingosine that is formed and provide a target for further research. Finally, to address the dynamic nature of sphingolipid metabolism within a cell, these lysates should be analyzed for total sphingolipids and each sphingolipid class using lipidomics [20]. This would help to determine if there are any increases in more complex sphingolipid species that could be missed with the radiolabeled tracer method mentioned above.

Additionally, some in vivo experiments demonstrate anti-colitic effects of dietary SM [21,

22]. Due to dietary SM’s incomplete digestion, it is possible some of the anti-inflammatory effects are at the level of the intestine. One putative mechanism is inhibiting LPS translocation. Aim 1 and 3 demonstrate reductions in serum LPS activity by dietary milk-derived SM. Ghoshal et al. has reported that formation of chylomicrons is required for LPS translocation into the lymph [23].

Thus, SM inhibition of lipid absorption could potentially lower LPS translocation. In support of this, dietary gangliosides have been shown to improve gut barrier function in response to LPS challenge [24]. Dietary SM helps promote the maturation of the gut in infant rats [25]. Bovine milk phospholipids have been shown to improved Caco-2 cell barrier function in response to LPS [26].

Therefore, our lab is currently aiming to explore SM’s potential to reduce LPS translocation using

185 a Caco-2 cell transwell model. In this model, we will measure transepithelial electrical resistance,

LPS activity of the basal lateral compartment, and triglyceride accumulation. These experiments will allow us to determine if emulsions containing milk-derived SM can improve barrier function and/or reduce LPS translocation, and if they are associated with inhibition of lipid absorption.

7.2.3. Interactions between Dietary SM and Gut Microbiota

Sphingosine has been shown to be bactericidal against a variety of enteropathogens [27].

Conversely, some microbes have been identified to possess sphingolipid metabolic capacity [28].

Our results show an increase in Firmicutes, Actinobacterium, and Bifidobacterium with higher dosages of dietary milk SM (0.25% w/w of diet) in mice. In addition to our results, there are only a few studies that examine dietary SM and gut microbiota [29, 30]. However, at this point these studies are difficult to compare, as wild-type mice can have vastly different gut microbiotas between litters [31]. Additionally, contamination from non-sterile bedding or supplies can be a factor [31]. Therefore, using gnotobiotics [32], conditions of well-controlled sterility, or mice inoculated with a known culture [33] could help control for these factors. These types of studies would offer more clarity in the effects of both background diet and dietary SM on the gut microbiota.

Another aspect of sphingolipids that has gone unexplored is their impact on the gut microbiome, or the collective genome of gut microbiota. As mentioned above, there are species of bacteria that can metabolize sphingolipids. Combined with the broad-spectrum bactericidal effects of sphingosine, it is fair to hypothesize these bacteria may have a survival advantage in the context of a sphingolipid-enriched diet. Therefore, bulk genomic sequencing of the microbiome in addition to 16S ribosomal RNA sequencing of gut microbiota may help further elucidate the effects of dietary sphingolipids on gut bacteria. I would expect the genes associated with sphingolipid

186 metabolism to be increased in the mice fed sphingolipids. This idea has been illustrated by measuring the gut bacteria-dependent trimethylamine production from choline and carnitine.

Particularly, in response to carnitine challenge, omnivorous humans produced more trimethylamine N-oxide compared to strict vegans [34]. This suggests that the chronic consumption of meat increased the host gut microbiota capacity to produce trimethylamine N- oxide’s precursor.

Lastly, while Sprong et al. [27] provided proof-of-concept for the bactericidal effects of sphingosine, much more work could be done with this concept. Dietary SM and ceramide reach the distal intestines [35]. Therefore, studying their interactions with common gut bacteria in simple in vitro models, similar to the Sprong et al. publication, could provide useful information.

Additionally, more species of bacteria should be tested for sphingolipid toxicity. Some members of the phyla Bacteroidetes can synthesize sphingolipids [36, 37], which are essential for their survival in the intestines. Additionally, Bifidobacterium, part of the Actinobacteria phyla, can metabolize sphingolipids [28]. Therefore, it would be interesting to see if exogenous sphingolipids are less toxic to Bacteroidetes or Bifidobacterium. The results of Aim 1 would suggest

Bifidobacterium are less affected by dietary sphingolipids. Aim 1 revealed a decrease in

Bacteroidetes with milk-derived SM treatment. This may be a result of increased dietary fat reaching the colon because chronic HFD decreases Bacteroidetes in mice [38].

7.2.4. Dietary SM and Prevention of Chronic Disease

Currently, there is little work on dietary SM and the development of late-stage chronic metabolic diseases. The work in this dissertation lays the foundation for the investigation into the progression of non-alcoholic steatohepatitis, cardiovascular disease, and type 2 diabetes mellitus. Currently,

187 our laboratory has two studies investigating the impact of dietary SM on atherosclerosis development in hyperlipidemic mouse models.

A natural progression of Aim 4 would be to induce non-alcoholic steatohepatitis (NASH) in LDLr-/- mice while feeding beta serum. It is likely the signs of improvement with the higher dose observed in Aim 4 would result in reductions in disease severity. In this study, comparing the efficacy of an equal dose of milk-derived SM would help understand if the excess phospholipids are improving or reducing the beneficial effects of SM.

Another important distinction that needs to be made is if the effects of dietary SM are exclusively due to effects on lipid absorption. The anti-inflammatory data, effects on gut microbiota, and the fact that some milk-derived ceramides appear in the lymph suggest otherwise.

Inhibiting diet-induced inflammation may be a secondary effect of SM. A simple comparison study with ezetimibe could help deduce if there are additional anti-inflammatory effects. First, a dual- isotope study could be conducted in mice to determine equivalent hypolipidemic dosages for milk- derived SM and ezetimibe on a Western-type diet. These doses could then be used in a diet-induced obesity mouse model to see if there are significant differences on inflammation outcomes.

The gut microbiota is another important factor in metabolic dysfunction. As demonstrated in Aims 1 and 3, SM can significantly alter the gut microbiota. Whether this impacts SM’s effects on metabolic dysfunction remains to be seen. It has been recently shown that egg-derived SM had anti-atherosclerotic effects in apoE-/- mice, and these effects were attenuated when co-administered with broad-spectrum antibiotics to deplete gut microbiota [29]. The gut microbiota has implications in metabolic dysfunction as reviewed in Chapter 2. Therefore, testing the effects of dietary SM on an obesogenic diet with and without broad spectrum antibiotics would demonstrate

188 if there are implications of the SM and gut microbiota interactions on metabolic dysfunction.

Another interesting discovery made recently is the deposition of serine dipeptide lipids in human atherosclerotic plaques derived from bacteria of the Bacteroidetes phylum, namely lipid 654 and lipid 430 (L654 and L430) [39]. This association has highlighted another potential bacterial byproduct that may accumulate in tissue. Due to the hydrophobic nature and sphingolipids’ effects on microbiota it would be interesting to see if L654 and L430 levels in the liver are modulated by dietary sphingolipids. More research is needed into the characteristics of these bacterial lipids, as their transport and metabolism in tissues is not well understood.

7.2.5. Clinical Trials to Measure the Efficacy of SM to Inhibit Cholesterol Absorption in Humans

Although one study did measure phytosterols as a measure of sterol absorption when feeding a phospholipid mixture, it would be advantageous to directly measure cholesterol absorption in humans, using stable isotope tracers [40]. The additional alkaline sphingomyelinase in human bile

[41] and the minute gastric lipase activity in rodents [42] could result in differences in dietary SM effects between rodents and humans. Gastric lipase reduces SM’s inhibitory effect on pancreatic lipase activity [43]. These tracer studies need to be done with both purified SM and a phospholipid fraction, such as beta-serum. This type of work should precede any further dietary interventions to ensure efficacy in humans.

7.2.6. Does Dietary SM Correlate with Disease Outcomes?

An epidemiological study could be conducted to determine if there is an association between dietary SM intake and serum cholesterol, metabolic syndrome, cardiovascular events, or liver disease. The intakes could be estimated using the National Health and Nutrition Examination

Survey (NHANES) data in combination with multiple sources citing the abundance of SM in foods. This study would need to control for the intake of plant-based foods, fats, sugars, and

189 cholesterol, as these are some of the possible confounders that could influence the findings.

Hypothetically, if there was an association it would strengthen the hypolipidemic effects observed in preclinical and clinical trials.

190

7.3. References

1. Mathiassen Jakob H, Nejrup Rikke G, Frøkiær H, Nilsson Å, Ohlsson L, Hellgren Lars I: Emulsifying triglycerides with dairy phospholipids instead of soy lecithin modulates gut lipase activity. European Journal of Lipid Science and Technology 2015, 117:1522- 1539. 2. Patton JS, Carey MC: Inhibition of human pancreatic lipase-colipase activity by mixed bile salt-phospholipid micelles. American Journal of Physiology-Gastrointestinal and Liver Physiology 1981, 241:G328-G336. 3. Aubert-Jousset E, Garmy N, Sbarra V, Fantini J, Sadoulet MO, Lombardo D: The combinatorial extension method reveals a sphingolipid binding domain on pancreatic bile salt-dependent lipase: role in secretion. Structure 2004, 12:1437-1447. 4. Lönnfors M, Doux Jacques P, Killian JA, Nyholm Thomas K, Slotte JP: Sterols Have Higher Affinity for Sphingomyelin than for Phosphatidylcholine Bilayers even at Equal Acyl-Chain Order. Biophysical Journal 2011, 100:2633-2641. 5. Harris FM, Best KB, Bell JD: Use of laurdan fluorescence intensity and polarization to distinguish between changes in membrane fluidity and phospholipid order. Biochim Biophys Acta 2002, 1565:123-128. 6. Yang F, Chen G, Ma M, Qiu N, Zhu L, Li J: Egg-Yolk Sphingomyelin and Phosphatidylcholine Attenuate Cholesterol Absorption in Caco-2 Cells. Lipids 2018, 53:217-233. 7. Kabir I, Li Z, Bui HH, Kuo M-S, Gao G, Jiang X-C: Small Intestine but Not Liver Lysophosphatidylcholine Acyltransferase 3 (Lpcat3) Deficiency Has a Dominant Effect on Plasma Lipid Metabolism. The Journal of Biological Chemistry 2016, 291:7651-7660. 8. Noh SK, Koo SI: Milk sphingomyelin is more effective than egg sphingomyelin in inhibiting intestinal absorption of cholesterol and fat in rats. J Nutr 2004, 134:2611- 2616. 9. Eckhardt ER, Wang DQ, Donovan JM, Carey MC: Dietary sphingomyelin suppresses intestinal cholesterol absorption by decreasing thermodynamic activity of cholesterol monomers. Gastroenterology 2002, 122:948-956. 10. Nyberg L, Duan RD, Nilsson A: A mutual inhibitory effect on absorption of sphingomyelin and cholesterol. J Nutr Biochem 2000, 11:244-249. 11. Morifuji M, Higashi S, Oba C, Ichikawa S, Kawahata K, Yamaji T, Itoh H, Manabe Y, Sugawara T: Milk Phospholipids Enhance Lymphatic Absorption of Dietary Sphingomyelin in Lymph-Cannulated Rats. Lipids 2015, 50:987-996. 12. Liu JJ, Nilsson A, Duan RD: Effects of phospholipids on sphingomyelin hydrolysis induced by intestinal alkaline sphingomyelinase: an in vitro study. J Nutr Biochem 2000, 11:192-197. 13. Zhang Y, Cheng Y, Hansen GH, Niels-Christiansen L-L, Koentgen F, Ohlsson L, Nilsson Å, Duan R-D: Crucial role of alkaline sphingomyelinase in sphingomyelin digestion: a study on enzyme knockout mice. Journal of Lipid Research 2011, 52:771-781. 14. Duivenvoorden I, Voshol PJ, Rensen PC, van Duyvenvoorde W, Romijn JA, Emeis JJ, Havekes LM, Nieuwenhuizen WF: Dietary sphingolipids lower plasma cholesterol and triacylglycerol and prevent liver steatosis in APOE*3Leiden mice. Am J Clin Nutr 2006, 84:312-321.

191

15. Jozefowski S, Czerkies M, Lukasik A, Bielawska A, Bielawski J, Kwiatkowska K, Sobota A: Ceramide and ceramide 1-phosphate are negative regulators of TNF-alpha production induced by lipopolysaccharide. J Immunol 2010, 185:6960-6973. 16. Wurfel MM, Wright SD: Lipopolysaccharide-binding protein and soluble CD14 transfer lipopolysaccharide to phospholipid bilayers: preferential interaction with particular classes of lipid. J Immunol 1997, 158:3925-3934. 17. Chiba N, Masuda A, Yoshikai Y, Matsuguchi T: Ceramide inhibits LPS-induced production of IL-5, IL-10, and IL-13 from mast cells. Journal of Cellular Physiology 2007, 213:126-136. 18. Romiti E, Meacci E, Tani M, Nuti F, Farnararo M, Ito M, Bruni P: Neutral/Alkaline and Acid Ceramidase Activities Are Actively Released by Murine Endothelial Cells. Biochemical and Biophysical Research Communications 2000, 275:746-751. 19. Bai A, Mao C, Jenkins RW, Szulc ZM, Bielawska A, Hannun YA: Anticancer actions of lysosomally targeted inhibitor, LCL521, of acid ceramidase. PLOS ONE 2017, 12:e0177805. 20. Shaner RL, Allegood JC, Park H, Wang E, Kelly S, Haynes CA, Sullards MC, Merrill AH, Jr.: Quantitative analysis of sphingolipids for lipidomics using triple quadrupole and quadrupole linear ion trap mass spectrometers. J Lipid Res 2009, 50:1692-1707. 21. Furuya, Ohkawara, Nagashima, Asanuma, Hino: Dietary Sphingomyelin Alleviates Experimental Inflammatory Bowel Disease in Mice. International Journal for Vitamin and Nutrition Research 2008, 78:41-48. 22. Mazzei JC, Zhou H, Brayfield BP, Hontecillas R, Bassaganya-Riera J, Schmelz EM: Suppression of intestinal inflammation and inflammation-driven colon cancer in mice by dietary sphingomyelin: importance of peroxisome proliferator-activated receptor gamma expression. J Nutr Biochem 2011, 22:1160-1171. 23. Ghoshal S, Witta J, Zhong J, de Villiers W, Eckhardt E: Chylomicrons promote intestinal absorption of lipopolysaccharides. Journal of Lipid Research 2009, 50:90-97. 24. Park EJ, Thomson AB, Clandinin MT: Protection of intestinal occludin tight junction protein by dietary gangliosides in lipopolysaccharide-induced acute inflammation. J Pediatr Gastroenterol Nutr 2010, 50:321-328. 25. Motouri M, Matsuyama H, Yamamura J, Tanaka M, Aoe S, Iwanaga T, Kawakami H: Milk sphingomyelin accelerates enzymatic and morphological maturation of the intestine in artificially reared rats. J Pediatr Gastroenterol Nutr 2003, 36:241-247. 26. Anderson RC, MacGibbon AKH, Haggarty N, Armstrong KM, Roy NC: Bovine dairy complex lipids improve in vitro measures of small intestinal epithelial barrier integrity. PLOS ONE 2018, 13:e0190839. 27. Sprong RC, Hulstein MF, Van der Meer R: Bactericidal activities of milk lipids. Antimicrob Agents Chemother 2001, 45:1298-1301. 28. Hasegawa K, Stewart CJ, Mansbach JM, Linnemann RW, Ajami NJ, Petrosino JF, Camargo CA: Sphingolipid metabolism potential in fecal microbiome and bronchiolitis in infants: a case–control study. BMC Research Notes 2017, 10:325. 29. Chung RWS, Wang Z, Bursill CA, Wu BJ, Barter PJ, Rye K-A: Effect of long-term dietary sphingomyelin supplementation on atherosclerosis in mice. PLOS ONE 2017, 12:e0189523.

192

30. Reis MG, Roy NC, Bermingham EN, Ryan L, Bibiloni R, Young W, Krause L, Berger B, North M, Stelwagen K, Reis MM: Impact of dietary dairy polar lipids on lipid metabolism of mice fed a high-fat diet. J Agric Food Chem 2013, 61:2729-2738. 31. Hufeldt MR, Nielsen DS, Vogensen FK, Midtvedt T, Hansen AK: Variation in the Gut Microbiota of Laboratory Mice Is Related to Both Genetic and Environmental Factors. Comparative Medicine 2010, 60:336-342. 32. Williams SCP: Gnotobiotics. Proceedings of the National Academy of Sciences of the United States of America 2014, 111:1661-1661. 33. Schaedler RW, Dubos R, Costello R: THE DEVELOPMENT OF THE BACTERIAL FLORA IN THE GASTROINTESTINAL TRACT OF MICE. J Exp Med 1965, 122:59-66. 34. Koeth RA, Wang Z, Levison BS, Buffa JA, Org E, Sheehy BT, Britt EB, Fu X, Wu Y, Li L, et al: Intestinal microbiota metabolism of L-carnitine, a nutrient in red meat, promotes atherosclerosis. Nature medicine 2013, 19:576-585. 35. Ohlsson L, Hertervig E, Jonsson BA, Duan RD, Nyberg L, Svernlov R, Nilsson A: Sphingolipids in human ileostomy content after meals containing milk sphingomyelin. Am J Clin Nutr 2010, 91:672-678. 36. An D, Na C, Bielawski J, Hannun YA, Kasper DL: Membrane sphingolipids as essential molecular signals for Bacteroides survival in the intestine. Proc Natl Acad Sci U S A 2011, 108 Suppl 1:4666-4671. 37. Moye ZD, Valiuskyte K, Dewhirst FE, Nichols FC, Davey ME: Synthesis of Sphingolipids Impacts Survival of Porphyromonas gingivalis and the Presentation of Surface Polysaccharides. Front Microbiol 2016, 7:1919. 38. Zhang M, Yang X-J: Effects of a high fat diet on intestinal microbiota and gastrointestinal diseases. World Journal of Gastroenterology 2016, 22:8905-8909. 39. Nemati R, Dietz C, Anstadt EJ, Cervantes J, Liu Y, Dewhirst FE, Clark RB, Finegold S, Gallagher JJ, Smith MB, et al: Deposition and Hydrolysis of Serine Dipeptide Lipids of Bacteroidetes Bacteria in Human Arteries: Relationship to Atherosclerosis. Journal of Lipid Research 2017. 40. Bosner MS, Ostlund RE, Jr., Osofisan O, Grosklos J, Fritschle C, Lange LG: Assessment of percent cholesterol absorption in humans with stable isotopes. J Lipid Res 1993, 34:1047-1053. 41. Duan RD, Nilsson A: Purification of a newly identified alkaline sphingomyelinase in human bile and effects of bile salts and phosphatidylcholine on enzyme activity. Hepatology 1997, 26:823-830. 42. DeNigris SJ, Hamosh M, Kasbekar DK, Lee TC, Hamosh P: Lingual and gastric lipases: species differences in the origin of prepancreatic digestive lipases and in the localization of gastric lipase. Biochim Biophys Acta 1988, 959:38-45. 43. Mathiassen JH, Nejrup RG, Frøkiær H, Nilsson Å, Ohlsson L, Hellgren LI: Emulsifying triglycerides with dairy phospholipids instead of soy lecithin modulates gut lipase activity. European Journal of Lipid Science and Technology 2015, 117:1522-1539.

193

Appendix A. Primer List

194

Gene Protein* 5'-Forward Primer-3' 5'-Reverse Primer-3' Acaca ACC GTCCCGGCCACATAACTGAT CGCTCAGGTCACCAAAAAGAAT Abcg5 ABCG5 GCGTAGGTCTCCTTTACCAGTTTG GGAAACAGATTCACAGCGTTCA Acad ACAD1 AGAGCAAGTCCCCACCAATG GGCTTGCTTGGCATCAACA Acox1 ACOX1 CCCAAGACCCAAGAGTTCATTC CAGGCCACCACTTGATGGA Actb β-actin TCCTTCTTGGGTATGGAATCCT CAGCACTGTGTTGGCATAGA Adgre1 F4/80 GAGTGGAATGTCAAGATGTTA CAGTGGAAGAAGAGAAGC Adipoq Adiponectin CAACCAACAGAATCATTATG GGTAAGAGAAGTAGTAGAGT Apoa4 ApoA-IV TTCCTGAAGGCTGCGGTGCTG CTGCTGAGTGACATCCGTCTTCTG Ccl2 MCP-1 TTCCTCCACCACCATGCAG CCAGCCGGCAACTGTGA Cd36 CD36 CTTACACATACAGAGTTCGTTATC TCCAACAGACAGTGAAGG Cd68 CD68 TGTCTGATCTTGCTAGGACCG GAGAGTAACGGCCTTTTTGTGA Cebpa C/EBPα CAAGAGCCGAGATAAAGC TCATTGTCACTGGTCAAC Cpt1a CPT1α GAACCCCAACATCCCCAAAC TCCTGGCATTCTCCTGGAAT Cpt1b CPT1β GCTCTGAGACACATCTACC CTAAGGATGCCATTCTTGATTC Dgat2 DGAT2 GGAATAAGTGGGAACCAGATCA CCGCAAAGGCTTTGTGAAG Fabp2 FABP2 GTCTAGCAGACGACGGAACGGA AGAAACCTCTCGGACAGCAA Fabp4 FABP4 CACCATCCGGTCAGAGAGTACTT CGGTGATTTCATCGAATTCCA Gapdh GAPDH TGTGTCCGTCGTGGATCTGA CCTGCTTCACCACCTTCTTGAT Hmgcr HMGCR GATGATTATGTCTTTAGGCTTG CAGAGAGAAACACTTGGT Il1b IL-1β GTCACAAGAAACCATGGCACAT GCCCATCAGAGGCAAGGA Il6 IL-6 GACTGATGCTGGTGACAA GCCATTGCACAACTCTTT Itgax Cd11c CTGGATAGCCTTTCTTCTGCTG GCACACTGTGTCCGAACTC Ldlr LDLr TTCCTGTCCATCTTCTTCC GACCATCTGTCTTGAGGG Lpl LPL AAGGTCATCTTCTGTGCTAGG CCAGTGTCAGCCAGACTT Mip1b MIP-1β TCTGTGCAAACCTAACCCCG GAGGGTCAGAGCCCATTGGT Mttp MTTP CTACCAGGCCCAACAAGAC CGCTCAATTTTGCATGTATCC Mylip IDOL AGGAGATCAACTCCACCTTCTG ATCTGCAGACCGGACAGG Npc1l1 NPC1L1 CGTCTGTCCCCGCCTATACA CTAATGACACCAGCTGCTTGGT Nr1h3 LXRα TTTAGGGATAGGGTTGGAGTCA GTGGACGAAGCTCTGTCG

195

Ocln OCLN CACCTATCACTTCAGATCAACAA CAGCAGCCATGTACTCTTC Pck1 PEPCK AACTGACAGACTCGCCCTATGTG TGCCCATCCGAGTCATGAT Pcsk9 PCSK9 AGGTGGAGGTGTATCTCTTAGA CACTTGCTCGCCTGTCTG Ppara PPARα CGGCAGTGCCCTGAACA TGGTACCCTGAGGCCTTGTC Pparg PPARγ CACAATGCCATCAGGTTTGG GCTGGTCGATATCACTGGAGATC Pparg2 PPARγ2 TTCGCTGATGCACTGCCTAT GCATCTCTGTGTCAACCATGGT Ppargc1a PGC-1α AAGCTGAAGCCCTCTTGCAA ACTGTACCGGGCCCTCTTG Rplp0 RPLP0 (36B4) CCTGAAGTGCTCGACATCAC CCACAGACAATGCCAGGAC Scd1 SCD1 CAGTGCCGCGCATCTCT CCCGGGATTGAATCTTCTTG Slc2a4 GLUT4 CATGGCTGTCGCTGGTTTC AAACCCATGCCGACAATGA Srebf1 SREBP-1c GGAGCCATGGATTGCACATT GGCCCGGGAAGTCACTGT Srebf2 SREBP-2 GTGTGCGGAGGAGAAAATCC CCCATGGCCGCTGTCA Tgfb1 TGFβ CGCAACAACGCCATCTAT TGCTTCCCGAATGTCTGA Tjp1 TJP1 AGAGGAAGAGCGAATGTC TTCGGTTCTGGAAGAGTG Tnfa TNFα GGCTGCCCCGACTACGT ACTTTCTCCTGGTATGAGATAGCAAAT

196