The Regulation of Tissue N-Acylethanolamine and Arachidonoylglycerol Concentrations By

Home , Cohune oil, Conjugated linoleic acid, Docosatetraenoic acid, Eicosapentaenoic acid, Glucose uptake, Margaric acid

The regulation of tissue n-acylethanolamine and arachidonoylglycerol concentrations by

diet, ischemia, n-acyl phosphatidylethanolamine-specific phospholipase D and

fatty acid amide hydrolase

Lin Lin

A thesis submitted in conformity with the requirements for the degree of Doctorate of Philosophy

Department of Nutritional Sciences University of Toronto

The regulation of tissue n-acylethanolamine and arachidonoylglycerol levels by diet, ischemia, n-acyl phosphatidylethanolamine-specific phospholipase D and fatty acid amide hydrolase

Lin Lin Doctor of Philosophy Department of Nutritional Sciences University of Toronto 2018 Abstract

Dietary fatty acids (DFAs) can modulate levels of n-acylethanolamines (NAEs) and arachidonoylglycerols (AGs) via the synthetic enzyme, n-acylphosphatidylethanolamine- specific phospholipase D (NAPE-PLD) and degradation enzyme, fatty acid amide hydrolase

(FAAH). However, other parallel pathways maybe also involved. This research hypothesizes that the absence of NAPE-PLD will lower tissue NAEs; while the absence of FAAH will increase NAEs. Also, basal NAE and AG levels are lower than ischemia-induced levels, while

DFA can modulate basal NAE levels. In study 1, wild-type (C57BL/6), NAPE-PLD (-/+) and

NAPE-PLD (-/-) mice were fed AIN-93G diets enriched with beef tallow, canola oil, corn oil or fish oil for nine weeks then killed by microwave fixation. Results showed that NAPE-PLD

(-/-) had lower levels of plasma and jejunum oleoylethanolamide (OEA), lower food intake, body weight, and body fat composition than wild-type. Also, corn oil diet elevated arachidonoylethanolamide (AEA) and AGs; while fish oil elevated docosahexaenoylethanolamide (DHEA) in all genotypes. Therefore, NAPE-PLD is specific to

OEA, but not other NAE, which may play a role in regulating body composition and weight management. Also, DFA can still elevate NAE levels without NAPE-PLD suggesting it is not necessary for dietary increased NAEs. In study 2, wild-type and FAAH-KO mice were fed a standard chow diet for 9 weeks; then killed by control (microwave fixation) or partial CO2- induced ischemia or CO2-induced ischemia. Brain NAE levels were higher in the FAAH-KO than the wild-type. Interestingly, CO2-induced ischemia increased all NAEs and AGs in the wild-type mice, but only DHEA was increased in the FAAH-KO mice. Thus, FAAH may play

II an essential role in regulating lipid metabolism in response to ischemia. In conclusion, this work provides evidence that diet, NAPE-PLD, FAAH and ischemia are independently involved in maintaining tissue NAE levels.

III

Acknowledgment Beyond the persistent, cogent and passionate scientific research, this thesis is also fulfilled with my inner growth and the spirit of precious intellectual, psychological and emotional support provided by many people. I would like to share many thanks and many moments, which built the backbone of my strength to accomplish my Ph.D. degree.

Dr. Richard Bazinet: I feel honored to share my sincere gratitude to my supervisor, Dr. Richard Bazinet, for the opportunity you provided me to learn and grow under your guidance, advice, and support. I have been fortunate to have you as my supervisor who genuinely cared about my research and who taught me how to think critically and write effectively to become a better scientist. I also inspired by your research ethics and philosophy: “If you followed every step carefully to test your hypothesis, then don’t worry about what the results might say. A result is a result; this is science.”

Dr. Peter Jones: I sincerely thank my co-supervisor, Dr. Peter Jones, for your continuous support and the guidance you gave to me during my Ph.D. Your countless encouragement and mentorship since my Master degree at the University of Manitoba. Your support and belief in my ability on researching the field of nutrition and endocannabinoids for the past ten years have been invaluable. I become a better researcher since your first advice: “Be on time; use more English, and get your work published.”

Dr. Harvey Anderson: I sincerely thank my advisory committee member, Dr. Harvey Anderson for the knowledge and experience you guided to me and the tough questions you challenged me to understand my project and the general concept of science in an exceptional training.

I thank Dr. Elena Comelli, my advisory committee member, for generously given your time and insights to guide me on my graduate studies and thesis writing during the past five years.

I thank Dr. Deborah O’Connor and Dr. Mary R. L’Abbé for acting as my Departmental Appraiser. Dr. Thomas Wolever for serving as my Departmental Chair; and Dr. Harold Aukema for acting as my External Appraiser from the University of Manitoba. I feel honored to have been among such a distinguished group of scientists.

IV “Few are those who see with their own eyes and feel with their own hearts” – Albert Einstein

My lab family: Everyone in Dr. Bazinet lab lent their helping hands to my research. I especially would like to thank Shoug Alashmali, who has been a fantastic supportive lab mate and a wonderful friend during my research journey. I would like to thank Vanessa Guiliano, Scott Lacombe, Adam Metherel, Raphael Chounard-Watkins and Maha Irfan Sarah Orr, Chuck Chen, Marco Trepanier, Kathyrn Hopperton Alex Kitson, Lauren Lin, Kayla Hildebrand, for the fun time, supportive research environment and advice to help me become a confident speaker and researcher.

I also gave my special thanks to Louisa Kung and Emelia D’Souza, who has given the best administrative support in such to ensure my research journey on time and well-performed.

I want to thank my teaching mentor Debbie Gurfinkel, for guiding me to be a better teaching assistant and an instructor to deliver research in the past four years. I want to thank Fiona Wallace, who is a definite hard-working role model for me as my mentor.

I would like to thank Kate Banks, Tracy McCook, Nancy Tomas, AJ Wang Warren Foltz, Michael Leadley and Ashley St.Pierre for ensuring my experiments went well. Also, I want to thank many collaborators, including Sophie Laye, Cobol Su, Laura Best, Mathieu Di-Miceli, Mandy H, Ruslan Kubant, Ryan Bradley, Cigdem Sahin, Ivana Prce who shared their knowledge and expertise in the different area of research, which opened my view on conducting research containing diverse programs.

In addition, I want to thank Shirley Vien, Diana Sanchez Hernandez, Amel Taibi, Paraskevi Masssara, Kit-Yi Yam and many others who are my colleagues and friends going through this fantastic post-graduate journey together.

I want to thank Daniela D’Aniello and my friends, Aleese Smith, Melanie A, Marianthe Maroulis, Tad Ferreira, Gursimran Sethi, Anamika Ray Elie Lee Livia Li, Dobrachina Zubek and others who lived in Knox College and New College. We shared so many wonderful conversations, meals, events, and parties which made our experience in the University of Toronto enjoyable.

I want to thank my piano teacher Arsha Nersessian and Ed Jesus and my piano students for sharing our joyful weekly music sections in the past five years V “When I walk along with two others, from at least one I will be able to learn”-Confucius

I want to thank my dear friends Alex (Chen) Wang, Yanlin Zhao, Xin Yi, Ray Zhang, Dom Kowk, Linda (Dan) Li, Dongfang Chao, and many others for sharing your wise words, life and work experience with me through our numerous conversations. Life is a continuous line filled with different colors of flowers that we observe together.

To my amazing parents, thank you so much for providing mental, emotional and material support to cherish my life. You taught me to how to work hard while balanced, how to respect people and keep smiling even in the face of adversity. Mother, you have a kind heart and a strong spirit to lift me up in any circumstance. Whenever I had doubt, I remember you said: “Don’t be afraid.” Father, you taught me how to be a hard worker with gratitude. You are a great listener and a patient achiever. I learn from you and make you proud.

Through my fascinating Ph.D. journey, I inspired by some famous scholars and many ordinary but fantastic people by their perception of life through their experience. I also came up with my own understanding of science and life. These invaluable inspiration quotations are shared in the footnotes of the acknowledgment and reference sections, that opened my mind and heart.

VI “There is always a deeper emotion which triggers all the motives seeing in people's ostensible needs.” - Agatha Christie, 《Five Little Pigs》 Financial Disclosure

This research was funded by the Natural Sciences and Engineering Research Council of

Canada (NSERC). Lin Lin also received scholarships from Ontario Graduate Scholarship,

Ontario Student Opportunity Trust Fund and Doctoral Completion Award.

VII “Pleasure in the job puts perfection in the work”-Aristotle

Contributions

Lin Lin contributed to and/or performed all experimental designs, method development, procedures, data collection, analyses, and wrote the first draft, with the exception of the following contributions:

Co-authors for submitted / published manuscripts

Dr. Adam Metherel - Contributed to statistical analysis and provided feedback on manuscript editions (Chapter 4 & 5).

Dr. Alex Kitson - Contributed to database extraction and confirmation (Chapter 2); Data validation (Chapter 4)

Dr. Kathryn E Hopperton and Dr. Marc-Olivier Trépanier - Contributed to animal tissue collections and provided feedback on manuscript editions (Chapter 4).

Dr. Peter J Jones: contributed to the study designs, assisted in the data collection, provided technical support and edited manuscript revisions (Chapter 2, 4, 5)

Dr. Richard P Bazinet: supervised the development of research ideas, provided laboratory resources, and comprehensively edited manuscripts (Chapter 2, 4, 5)

Ms. Shoug Alashmali - Contributed to the risk of biases assessment and data validation

(Chapter 2); helped with sample preparation and provided feedback on manuscript editions

(Chapter 2 & 4)

VIII “A successful relationship is an efficient relationship, which aims to a mutual direction and motivates each other through the journey.”-Alex (Chen) Wang (Friend) Table of Contents

Acknowledgement ...... IV

Financial Disclosure ...... VII

Contributions ...... VIII

Table of Contents ...... IX

List of Tables ...... XIV

List of Figures ...... XV

List of Abbreviations ...... XVI

CHAPTER 1. Introduction ...... 1

CHAPTER 2. Literature review ...... 6

2.1. Purpose of this literature review...... 7

2.2. Background ...... 7

2.3. Identification and quantification of n-acylethanolamines (NAEs) and

monoacylglycerols (MAGs) ...... 8

2.4. The enzymatic synthesis of NAEs and MAGs ...... 9

2.4.1.The biosynthesis and degradation of NAEs ...... 9

2.4.2. The biosynthesis and degradation of MAGs (e.g., 2-arachidonoylglycerol) ...... 12

2.5. Research gap ...... 14

2.6. Study design ...... 14

2.6.1. Study eligibility criteria ...... 14

2.6.2. Literature search strategy ...... 15

2.6.3. Data extraction ...... 15

2.6.4. Data synthesis and statistical analysis...... 18

2.7. Results ...... 18

IX “"There's a way to do it better - find it."-Thomas Edison

2.7.1. Risk of bias assessments ...... 21

2.7.2. Study background / designs ...... 21

2.7.3. Clinical trials: subject characteristic ...... 21

2.7.4. Animal studies: strains, types ...... 23

2.7.5. The impact of dietary fatty acids on human NAE and MAG levels ...... 26

2.7.6. The impact of dietary fatty acids on animal NAE and MAG levels ...... 29

2.7.6.1. Blood NAE and MAG levels ...... 29

2.7.6.2. Brain NAE and MAG levels ...... 34

2.7.6.3. Other organs NAE and MAG levels ...... 43

2. 8. Summary and discussion ...... 49

2.8.1. Impact of dietary fatty acids on NAE and MAG levels ...... 49

2.8.2. Other factors may affect on NAE and MAG levels in biological samples ...... 60

2.9. Strengths and limitations...... 60

2.10. Conclusion ...... 62

CHAPTER 3. Rationale, Hypothesis and objectives ...... 64

3.1. Rational and research gap ...... 65

3.2. Overall hypothesis ...... 67

3.3. Specific hypothesis ...... 67

3.4. Overall objectives ...... 67

3.5. Specific objectives...... 67

CHAPTER 4. (Study 1) Dietary fatty acids augment tissue levels of n-acylethanolamines in n-acylphosphatidylethanolamine phospholipase D (NAPE-PLD) knockout mice ...... 68

4.1. Abstract ...... 69

4.2. Introduction ...... 69

X ““Changes will only make people stronger, that’s how adaptation works.”-Dongfang Chao (Friend)

4.3. Methods ...... 71

4.3.1 Diets ...... 71

4.3.2 Animals ...... 75

4.3.3. The percentage of fat oxidation ...... 75

4.3.4. Body fat composition ...... 76

4.3.5. Euthanasia and sample collection ...... 76

4.3.6. Tissue lipid extraction and gas chromatography-mass spectrometry ...... 77

4.3.7. Extractions of NAEs and arachidonoylglycerols (AGs) ...... 77

4.3.8. Identification and separation using high-performance liquid chromatography-mass

spectrometry ...... 78

4.4. Statistics ...... 79

4.5. Results ...... 79

4.5.1. Tissue fatty acid composition ...... 79

4.5.2. Tissue NAEs, 1-AG and 2-AG levels ...... 83

4.5.2.1. Plasma...... 83

4.5.2.2. Liver ...... 85

4.5.2.3. Jejunum...... 87

4.5.2.4. Brain ...... 89

4.5.3. Food intake and body weight ...... 91

4.5.4. Fat oxidation and fat composition ...... 93

4.6. Discussion ...... 95

CHAPTER 5. Study 2 Fatty acid amide hydrolase (FAAH) regulates hypercapnia/ischemia- induced increases in n-acylethanolamines in mouse brain ...... 100

5.1. Abstract ...... 101

XI “A question that sometimes drives me hazy: am I or others crazy?”-Albert Einstein

5.2. Introduction ...... 102

5.3. Methods ...... 103

5.3.1. Diets ...... 103

5.3.2. Animals...... 103

5.3.3. Kill methods ...... 104

5.3.4. Whole brain lipid extraction and gas chromatography-mass spectrometry ...... 106

5.3.5. NAEs and 1-AG and 2-AG sample extractions ...... 107

5.3.6. Identification and separation using high-performance liquid chromatography mass

spectrometry ...... 107

5.4. Statistics ...... 110

5.5. Results ...... 110

5.5.1. Whole brain total lipids are not altered by CO2-induced hypercapnia/ischemia or

FAAH-KO ...... 110

5.5.2. Unesterified lipids are elevated upon CO2-induced hypercapnia/ischemia ...... 112

5.5.3. OEA, AEA and DHEA are elevated in the FAAH-KO mice ...... 114

5.5.4. OEA, AEA, and DHEA are elevated with CO2-induced hypercapnia/ischemia in

wild-type, but only DHEA is elevated in FAAH-K mice ...... 114

5.5.5. 1-AG and 2-AG are elevated upon CO2-induced hypercapnia/ischemia ...... 116

5.6. Discussion ...... 118

CHAPTER 6. General discussion ...... 122

6.1. Summary of research findings...... 123

6.2. Regulatory pathway associated with NAE biosynthesis and degradation ...... 124

6.3. The effect of dietary fatty acids on NAE synthesis with/without NAPE-PLD ...... 125

.4. The effect of ischemia on NAE degradation with/without FAAH...... 125

XII “I don’t give up, unless walking away is a better solution in that situation”-Lin Lin

6.5. Strengths and Limitations...... 126

6.6. Significance and implications ...... 127

6.7. Specific conclusion ...... 128

6.8. Overall conclusion...... 129

CHAPTER 7. Future directions ...... 130

CHAPTER 8. Reference ...... 135

CHAPTER 9. Appendices ...... 149

Appendix 1. Search term for Medline 1946-October week1 2017 and Medline in-progress

and non-indexed citations. (Chapter 2) ...... 150

Appendix 2. Search Term for Embase October 14th, 2017 (Chapter 2) ...... 153

Appendix 3. PRISM checklist 2009 ...... 158

Appendix 4. The risk of bias assessment for human studies (Chapter 2) ...... 161

Appendix 5. The risk of bias assessment for animal studies (Chapter 2) ...... 163

XIII “Only one who loves can remember so well.” –Anton Chekhov

List of Tables

Table 2.1. Description of the PICOS criteria used to perform this systematic review ...... 17

Table 2.2. Human study design and subject characteristics of individual studies ...... 22

Table 2.3. Animal studies ...... 24

Table 2.4. Endpoint plasma NAEs and MAGs concentrations in human studies ...... 28

Table 2.5. Endpoint blood NAEs and MAGs concentrations in animal studies ...... 32

Table 2.6. NAEs and MAGs in brain or brain regions ...... 38

Table 2.7. NAEs and MAGs in the body ...... 46

Table 2.8. The association of dietary fat-induced NAEs and MAGs on the change of physiological outcomes ...... 54

Supplemental Table 2.1. Baseline NAEs and MAGs concentration in clinical studies ...... 63

Table 4.1. Macronutrient composition of diets ...... 73

Table 4.2. Dietary fatty acid composition ...... 74

XIV “Be happy in the moment, that’s enough. Each moment is all we need, not more”-Mother Teresa

List of Figures

Figure 1.1. Internal and external factors influence n-acylethanolamine levels ...... 5

Figure 2.1. Biosynthesis and degradation of NAE pathways ...... 11

Figure 2.2. Biosynthesis and degradation of 2-AG pathways ...... 13

Figure 2.3. Flow diagram of the systematic review ...... 20

Figure 3.1. Rational and research gap ...... 66

Figure 4.1. Fatty acid composition of liver, duodenum, and whole brain...... 81

Figure 4.2. Plasma NAE, 1-AG, and 2-AG levels ...... 84

Figure 4.3. Liver NAE, 1-AG, and 2-AG levels 2.7 Results ...... 86

Figure 4.4. Jejunum NAE, 1-AG, and 2-AG levels ...... 88

Figure 4.5. Brain NAE, 1-AG, and 2-AG levels ...... 90

Figure 4.6. The progression of accumulated food intake and weekly body weight ...... 92

Figure 4.7. Ratio of fat volume to body weight ...... 94

Supplemental Figure 4.1. Example of one magnetic resonance imaging slice of total body composition ...... 97

Figure 5.1. Study design ...... 105

Figure 5.2. The multiple reaction monitoring chromatography of NAEs and AGs ...... 109

Figure 5.3. Whole brain total lipids ...... 111

Figure 5.4. Whole brain unesterified fatty acid concentrations ...... 113

Figure 5.5. Whole brain NAE concentrations ...... 115

Figure 5.6. Whole brain 1, 2-AG concentrations ...... 117

Supplemental Figure 5.2. The schematic diagram of this study ...... 121

Figure 7.1. Overview of thesis for future research directions ...... 13

XV “There are and will be a thousand princes; there is only one Beethoven.”-Ludwig Van Beethoven

List of Abbreviations

ABH4, or 6 or 12, alpha/beta domain-containing hydrolase 4, or 6 or 12; ACC1, Acetyl-CoA carboxylase1; ACT-1, alpha serine/threonine protein kinase-1; AEA, arachidonoylethanolamide; ALA, α-linoleic acid; ALEA, linolenoylethanolamide; AMPKa1,2: AMP-activated protein kinase catalytic subunit alpha-1,2; ARA, arachidonic acid; AT, acyltransferase; BDNF, brain derived neurotrophic factor; BMI, body mass index; CAO, canola oil; CA2+-NAT, Ca-depedent-n-acyl-transacylase; Camkk2: Ca2+/calmodulin-dependent protein kinase; CB1 or 2, cannabinoid receptor 1 or 2; CNS, central nervous system CLA, conjugated linoleic acid; CREBH, cyclic AMB-responsive element-binding protein 3-like3 hepatocyte specific; CRP, C-reactive protein; CO, corn oil; COX-2, cyclooxygenase-2; DAG, diacylglycerol; DAGL, diacylglycerol lipase DAGLαβ, diacylglycerol lipase-alpha, beta; DGLEA, dihomo-γ-linolenoylethanolamide; DHA, docosahexaenoic acid; DHEA, docosahexaenoylethanolamide; Di-acyl-PE, diacyl-phosphatidylethanolamine; DTEA, docosatetraenoyletaenoylethanolamide; E, Energy;

XVI “Humans are undefined.”-Nan Zou (PhD. in statistics)

EEA, eicosanoylethanolamine EPA, eicosapentaenoic acid; EPEA, eicosapentaenoylethanolamide; FFA, free fatty acids; FAAH, fatty acid amide hydrolase; FlaxO, flaxseed oil; FO, fish oil; GC, gas chromatography; GCK, glucokinase; GDE1, glycerophosphodiester phosphodiesterase 1; GLUT1,, insulin-regulated glucose transporter 1, 4; G6P, glycose 6-phosphate; GTPgS, guanosine tri phosphate gamma sulfur; Gp-NAE, glycerophospho-NAE; GPR, orphan-G protein-coupled receptor; HDL, high density lipoprotein; HFD, high-fat diet; HOCAO, high-oleic canola oil; HOMA-IR: homeostatic model assessment of insulin resistance; IL-6, interleukin-6; iSCAT, interpheromatric scattering microscopy; ITT, insulin tolerance test; KO, knockout; KrillO, krill oil; LC, liquid chromatography; LDL, low density lipoprotein; Lyso, lyso-phosphatidylinositol; LEA, linoleoylethanolamide; LFD, low-fat diet; LNA, linolenic acid; LPA, lysophosphatidic acid; LPS, lipopolysaccharide; LPL, lipoprotein lipase;

XVII “To put everything in balance is good, to put everything in harmony is better.”-Victor Hugo

MAG, monoacylglycerol; MAGL, monoacylglyceride lipase; MFD, medium-fat diet; MS, mass spectrometry; MUFA, monounsaturated fatty acids; NAAA, n-acylethanolamine-hydrolyzing acid amidase; NAE, n-acylethanolamine; NAPE, n-acylated ethanolamine phospholipid; NAPE-PLD, n-acyl-phosphatidylethanolamine-hydrolyzing phospholipase D; NAT, n-acyl-translase N-3 PUFA, omega-3 fatty acids; N-6 PUFA, omega-6 fatty acids; O-AEA, virodhamine OA, oleic acid; OEA, oleoylethanolamide; OGTT, oral glucose tolerance test; OO, olive oil; PA, palmitic acid; PC, phosphatidycholine; PE, phosphatidylethanolamine; PEA, palmitoylethanolamide;

PGE2-GE, prostaglandin E2 glycerol ester; PIP2, phosphatidylinositol bisphosphate; PLA1/2, phospholipase A1 or 2; PLC, phospholipase C;

PLCβ, phospholipase Cβ; PI, phosphatidyl inositol; PPAR-, proliferator activated receptor-alpha; PTPN22, protein tyrosine phosphatase, non-receptor type 22; PUFA, polyunsaturated fatty acids; SaffO, Safflower oil; SCAT, subcutaneous adipose tissue; SEA, stearoylethanolmide;

XVIII “What is the meaning of a sustained note in music?” “It means to hold it until you feel like to let it go.”- Piano student (6 years old) SEM, standard error SHIP1, Src homology 2 domain-containing inositol-5-phosphatase; SIRT1, sirtuin1 sn-1-PC, sn-1 position phosphatidycholine SO, soya-bean oil; SREBP-1c, sterol regulatory element-binding protein 1c; SunfO, sunflower oil; TC: total cholesterol; TG, triglycerides; THC, delta-9-tetrahydrocannabinol; TNF-α, tumour necrosis factor-alpha; TRPV1, transient receptor potential vanilloid type 1; VA, vaccenic acid; VAT, viscous adipose tissue; WC, waist circumference 1 or 2-AG, 1- or 2-arachidonoylglycerol; 2-EG, 2-eicosanoylglycerol; 2-EPG, 2-eicosapentanoylglycerol; 2-DHG, 2-docosahexenoylglycerol; 2-DTG, 2-docosatetraenoic acid glycerol 2-LG, 2-linoleoylglycerol; 1 or 2-OG, 1- or 2-oleoylglycerol; 2-PG, 2-palmitoylglycerol; 12-HPETE-G, 12- hydroxyperoxyeicosa-5,8,10,14-tetraenoic acid glycerol ester;

XIX “Feelings are as important as logic.” –Melanie Adamsons

CHAPTER 1

Introduction

Dietary oils vary substantially in fatty acid composition, for example, animal fats are rich in saturated fatty acids, canola or olive oils are high in omega-9 monounsaturated fatty acids; corn oil is high in omega-6 polyunsaturated fatty acids and fish oil high in omega-3 polyunsaturated fatty acids. Depending on their carbon chain length, chemical structure and number of double bonds, dietary fatty acids have the ability to regulate lipoprotein metabolism, fat oxidation, cholesterol metabolism and gene expression (Katan et al. 1994, Salter & Tarling 2007). However, their exact mechanisms of action in regulating the processes discussed above are not clearly understood. One of the possible mechanism is that dietary fatty acids can be converted to n-acylethanolamines (NAEs) (e.g., arachidonoylethanolamide, AEA) and monoacylglycerols (e.g., 2-arachidonoylglycerols,2-AG), which regulate lipid metabolism, fat oxidation and appetite control (Sugiura et al. 2006, DiPatrizio et al. 2011b, Fu et al. 2005, Yang et al. 2011).

NAEs and MAGs are lipid-signalling molecules. In 1957, the first NAE compound, palmitoylethanolamide (PEA) was discovered, which has anti-inflammatory properties (Coburn & Moore 1943, Kuehl et al. 1957). Although a variety of clinical trials explored the safety, and efficacy of PEA in the 1970s (Hesselink 2013b, Hesselink 2013a), interest on endocannabinoid only surged after the discovery of AEA in 1990s (Devane et al. 1992). Improvement of NAE and MAG analysis by using liquid chromatography-mass spectrometry (LC-MS) (Schmidt et al. 2006) opened the opportunity to study multiple NAE simultaneously in biological samples (Richardson et al. 2007, Lin et al. 2012, Balvers et al. 2013).

An emerging body of evidence suggests that changes in dietary fatty acid compositions can modulate NAE and MAG levels in blood and tissues (Lin et al. 2013a, Mennella et al. 2015b, Pu et al. 2016a). For example, levels of NAEs and MAGs reflect their precursor fatty acids in the diet (Artmann et al. 2008b). Other fatty acids can also alter NAE or MAG levels (Pintus et al. 2013b).

Early studies (Gillum et al. 2008b, Schwartz et al. 2008b) demonstrated the synthesis of NAEs from fatty acids using labelled ethanolamine-C14 (Colodzin et al. 1963) upon fasting and refeeding (Rodriguez de Fonseca et al. 2001, Petersen et al. 2006b), as well as sham feeding and fatty acid infusion (DiPatrizio et al. 2013a, Schwartz et al. 2008b). It is now generally accepted

that NAE can be synthesized from membrane phospholipids via multiple redundant steps. One well-studied pathway of NAE biosynthesis and degradation involves NAPE-specific phospholipase D (NAPE-PLD) and fatty acid amide hydrolase (FAAH) (LoVerme et al. 2005, Okamoto et al. 2007). MAGs are mainly synthesized using other enzymes including diacylglycerol lipase (DAGL) and monoacylglycerol lipase (MAGL). Other parallel pathways are spontaneously involved for both NAE and MAG formations. However, the specific role of NAPE-PLD and FAAH in regulating levels of NAEs upon dietary fat manipulation has not been studied. It is hypothesized that NAPE-PLD are necessary for the conversion of fatty acids to NAEs, but not MAGs (e.g. 1-AG and 2-AG) while FAAH is necessary for the catabolism of NAEs to fatty acids.

Chapter 2 of this thesis is a systematic literature review, which is adapted from a submitted study titled N-acylethanolamine and monoacylglycerol levels are regulated by dietary fatty acids in diets with isocaloric fat content: a systematic review. This literature review systematically summarizes the literature regarding the effect of dietary fatty acids on NAE and MAG levels upon consumption of a diet containing isocaloric fat in humans and animals. Additionally,

Chapter 2 addresses other factors (eg., ischemia, health status, species) that can contribute to the variation of NAE and MAG levels, which may be independently related to the effect of dietary fatty acid modulation.

The core of this Ph.D. thesis involved both internal and external factors (Figure 1.1). Internal factors, such as n-acylphosphatidylethanolamine-phospholipase D (NAPE-PLD) and fatty acid amide hydrolase (FAAH) are key contributors in the endocannabinoid metabolic pathways. External factors, such as dietary fatty acids and ischemia can also influence endocannabinoid levels. However, the interaction between internal factors and external factors on NAE levels has not been evaluated. In this thesis, Chapter 3 states the rationale, research gap, the hypotheses and objectives. The overall objective of this thesis is to examine whether external factors (dietary fatty acids or ischemia) can further influence levels of NAEs in the absence of internal factors (NAPE-PLD or FAAH).

Chapter 4 examines the biosynthesis of NAEs via the interaction between internal factor, NAPE- PLD and external factor, dietary fatty acids. This chapter is adapted from an study titled Dietary fatty acids augment tissue levels of n-acylethanolamines in n-acylphosphatidylethanolamine phospholipase D (NAPE-PLD) knockout mice, revision/publication to the Journal of Nutritional Biochemistry. Chapter 4 contains the bulk of my experimental work, which focused on the effect of different dietary fatty acids on NAE levels while NAPE-PLD was absent. This experiment was carefully designed to quantify basal brain NAE and AG levels without ischemic stress.

Chapter 5 examined the interaction of internal factor, FAAH and external factor, ischemia on levels of NAEs. This chapter is adapted from a published study titled Fatty acid amide hydrolase (FAAH) regulates hypercapnia/ischemia-induced increases in n-acylethanolamines in mouse brain, published in the Journal of Neurochemistry. Chapter 5 tested if the absence of FAAH will result in lower levels of NAEs. Also, this experiment verified if ischemic stress could elevate NAE and AG levels. Chapter 6 and 7 summarizes my thesis and suggest future directions.

Figure 1.1. Internal and external factors influence n-acylethanolamine levels

CHAPTER 2 Literature review

N-acylethanolamine and monoacylglycerol levels are regulated by dietary fatty acids in diets with isocaloric fat content: a systematic review

Adapted from: Lin Lin, Alex Kiston, Shoug M Alashmali, Peter J Jones and Richard P Bazinet. “N-acylethanolamine and monoacylglycerol levels are regulated by dietary fatty acids in diets with isocaloric fat content: a systematic review”. Under revision

2.1 Purpose of this literature review The overarching objective of this review was to determine if the isocaloric fat content in the diet explains the alternation of NAE and MAG levels in various types of human and animal samples.

2.2 Background Dietary fatty acids can be converted to a family of lipid signaling mediators including n- acylethanolamines (NAEs) and monoacylglycerols (MAGs). Previous studies demonstrate that modulating dietary fatty acids can alter levels of NAEs and MAGs in tissues and blood, which in turn can impact homeostatic signals and physiological outcomes (DiPatrizio et al. 2011b, Engeli et al. 2014, Diep et al. 2011). Although NAEs and MAGs are both synthesized from fatty acids, they have different structures. For example, NAEs are nitrogen-containing lipids, where the acyl group(s) is linked to the nitrogen atom of ethanolamine. However, MAGs contain glycerol, which links to fatty acids through ester bonds.

Importantly, NAEs and MAGs are further divided into two categories. First, arachidonoylethanolamide (AEA, also called anandamide) is an endocannabinoid (Devane et al. 1992). AEA works similar to tetrahydrocannabinol (the active compound in marijuana), which binds to cannabinoid receptor 1 (CB1) and partially to CB2 (Pertwee 2015), stimulating fear, feeding, anxiety (Alger 2004), pain sensation (Cravatt & Lichtman 2003), energy balance, appetite, memory (Blancaflor et al. 2014), sedation, euphoria (LoVerme et al. 2005) and has psychotropic effects in humans. Additionally, other polyunsaturated NAEs, such as dihomo-γ- linolenoylethanolamide (DGLEA), eicosapentaenoylethanolamide (EPEA) and docosatetraenoyletaenoylethanolamide (DHEA), as well as polyunsaturated MAGs, such as 2- AG, also act as CB1 agonists (Bradshaw & Walker 2005, Meijerink et al. 2013).

Second, some NAEs and MAGs are described as “endocannabinoid-like” compounds, because they do not bind to CB1 receptors (Cravatt & Lichtman 2003, Esposito & Cuzzocrea 2013, Lambert et al. 2002). These compounds include saturated NAEs and MAGs (e.g., palmitoylethanolamide, PEA and 2-palmitoylglycerol, 2-PG), monounsaturated NAEs and MAGs (e,g., oleoylethanolamide, OEA and 2-oleoylglycerol, 2-OG) and some polyunsaturated NAEs and MAGs (e.g., linoleoylethanolamide, LEA and 2-linoleoylglycerol, 2-LG) (Bradshaw

& Walker 2005, Zoerner et al. 2011). These compounds bind to proliferator-activated receptor- alpha (PPAR-), transient receptor potential vanilloid type 1 (TRPV1), orphan G protein- coupled receptors (GPR55 and GPR119) (Wang & Ueda 2009), which can result in anti- inflammation, satiety and anti-pain (Fu et al. 2003, Hansen et al. 1999, Kuehl et al. 1957, LoVerme et al. 2005, Ueda et al. 1993a, Schmid et al. 1990).

2.3. Identification and quantification of n-acylethanolamines (NAEs) and monoacylglycerols (MAGs) Since NAEs and MAGs have many synonyms, we used the naming system from Pubchem (https://pubchem.ncbi.nlm.nih.gov/docs/subcmpd_summary_page_help.html) and the recommendation from the International Union of Pure and Applied Chemistry (IUPAC) (Rodriguez de Fonseca et al. 2001). Thus, we refer to these compounds as arachidonoylethanolamide (AEA) and 2-archidonoylglycerol (2-AG), as examples of NAE and MAGs, respectively (Berdyshev et al. 1996, Schmid et al. 2000).

The quantification of NAEs and MAGs in biological samples is most commonly done using gas or liquid chromatography with tandem mass spectrometry (GC-MS or LC-MS). Improvements in the analysis using LC-MS (Schmidt et al. 2006) opened the opportunity to study multiple NAEs and their isomers simultaneously in various types of biological samples (Richardson et al. 2007, Lin et al. 2012, Balvers et al. 2013). In general, levels of NAE in tissues, cells or plants are in the pmol/g range (Hansen et al. 1999, Ueda et al. 2010, Blancaflor et al. 2014, Hansen & Diep 2009). Among all identified NAE, the most abundant NAEs in animal tissues are PEA, SEA, OEA, and LEA (Ueda et al. 2010, Blancaflor et al. 2014, Hansen & Diep 2009), while in human follicular fluid or plasma, the most abundant are PEA and OEA (Schuel et al. 2002). AEA generally makes up less than 10% of the total NAE concentration (Lin et al. 2013a, Ueda et al. 2010, Maccarrone et al. 2001, Jones et al. 2014). Although tissue NAE levels are low under normal physiological conditions, NAEs can accumulate during ischemia or upon cell injury (Brose et al. 2016, Lin et al. 2017, Bazinet et al. 2005, Schmid et al. 1990). Also, it has been shown that levels of AEA increase with EDTA at room temperature (Zoerner et al. 2011) or when whole blood is stored at 4°C for one to two hours, likely due to ex vivo AEA release from

erythrocytes or leukocytes (Vogeser et al. 2006, Zoerner et al. 2011, Wood et al. 2008, Jain et al. 2017).

2.4. The enzymatic synthesis of NAEs and MAGs 2.4.1. The biosynthesis and degradation of NAEs For the biosynthesis of NAEs (Figure 2.1), the first reaction is through a calcium-dependent-N- acyl-transacylase (CA2+-NAT), which transfers a (Hansen et al. 1999)(PC) onto the amine of diacyl-phosphatidylethanolamine (PE) to generate n-acylated ethanolamine phospholipid (NAPE) (Hansen et al. 1999, Leung et al. 2006). In parallel, sn-1 fatty acids from PC and PE can form NAPE via an enzyme phospholipase A1 (PLA1) or acyltransferase (AT) as alternative pathways. It is worth noting that NAPE is a generic term for all of the three forms: “n-acylated ethanolamine phospholipid species comprising the diacyl type”, “n-acyl-plasmenyl- ethanolamine” and “n-acyl-plasmanyl-ethanolamine” (Ueda et al. 2013). NAPE can be synthesized in small intestinal gut segments after fat feeding (Gillum et al. 2008a), and is present in many other tissues (Rodriguez de Fonseca et al. 2001, Petersen et al. 2006a, Hansen et al. 1999). As well, NAPE is able to enter the central nervous system (CNS) and accumulates in the hypothalamus (Gillum et al. 2008a); and it has been suggested that NAE can be synthesized in the CNS from circulating NAPE (Gillum et al. 2008a).

After NAPE is synthesized, a group of specific hydrolases or a combination of acyltransferases and hydrolases convert it into NAE (Wang & Ueda 2009, Ueda et al. 2010, Ueda et al. 2013). The most studied pathway is the conversion of NAPE to NAE via n-acylphosphatidylethanolamine-hydrolyzing phospholipase D (NAPE-PLD) (Okamoto et al. 2007, LoVerme et al. 2005, Fu et al. 2007). However, NAPE can be converted to phosphate-NAE via phospholipase C (PLC), then phosphate-NAE is further converted into NAE via protein tyrosine phosphatase, non-receptor type 22 (PTPN22) or Src homology 2 domain-containing inositol-5- phosphatase (SHIP1) (Ueda et al. 2010). Furthermore, NAPE can be converted into lyso-NAPE via alpha/beta-hydrolase 4 (ABH4) or sPLA1/A2, then lyso-NAPE is further converted into NAE via lyso-PLD (Ueda et al. 2010, Sun et al. 2004). As well, lyso-NAPE can also be converted into glycerophospho-NAE (Gp-NAE) through ABH4, then the Gp-NAE is converted to NAE via GDE1 (Wang & Ueda 2009). Because GDE1 is highly expressed in the brain, where it can

regulated NAE synthesis (Wang & Ueda 2009). Moreover, free fatty acids and ethanolamine can be converted into NAE via fatty acid amide hydrolases (FAAH), when levels of free fatty acids and ethanolamine are at high concentrations (Katayama et al. 1999, Ueda et al. 2000). NAE can be degraded into free fatty acids and ethanolamine by hydrolases (Ueda et al. 2013). The most well-studied hydrolase for NAE degradation is FAAH, which catabolizes numerous NAEs. Another lysosomal enzyme, n-acylethanolamine-hydrolyzing acid amidase (NAAA), is also involved in NAE degradation to fatty acid and ethanolamine, which selectively hydrolyzes PEA over other NAEs (Ueda et al. 2013, Piomelli 2013, Cravatt et al. 1996, Izzo et al. 2010). Moreover, NAE can also be converted to either prostamide E2 via coyclooxygenase-2 (COX-2) or to 12-hydroperoxyeicosa-5,8,10,14-tetraenoylethanolamide via lipoxygenases (Ueda et al. 2010).

Figure 2.1. Biosynthesis and degradation of NAE pathways

Ingested dietary fat/oil is broken down into phospholipids (e.g., PC and PE), thereby converting to NAEs in tissues. The 1st step is to synthesize NAPE, which is catalyzed by enzymes (e.g., CA2+-NAT or PLA1 and/or AT-1-5) in the gut. Then NAPE is able to enter other tissues including the CNS. The 2nd step involves multiple pathways. The most well-known pathway is via the enzyme NAPE- PLD. Also, in the CNS, NAPE can first be converted to phosphate-NAE via enzyme PLC, then phosphate-NAE is further converted into NAE via enzyme PTPN22 or SHIP1. In parallel, NAPE can be converted into lyso-NAPE via enzymes (e,g. ABH4 or sPLA1/A2). Then the lyso-NAPE can be further converted into NAE in multiple pathways 1) lyso-PLD in the gut or 2) ABH4 to synthesize Gp- NAE then via GDE1 to synthesize NAE. In addition, FFA and ethanolamine can be converted into NAE via FAAH at high concentrations. Spontaneously, NAE can be broken down into FFA and ethanolamine via FAAH or NAAA (target to PEA only). Moreover, NAE can also be broken down into prostamide E2 via COX2 or 12-hydroperoxyeicosa-5,8,10,14-tetraenoylethanolamide via lipoxygenases. Source from (Hansen et al. 1999, Gillum et al. 2008a, Ueda et al. 2013, Wang & Ueda 2009, Okamoto et al. 2007, Lo Verme et al. 2005, Sun et al. 2004, Katayama et al. 1999, Izzo et al. 2010, Ueda et al. 2010)

2.4.2. The biosynthesis and degradation of MAGs (e.g., 2-arachidonoylglycerol) After ingestion, dietary fatty acids are converted to phospholipids in tissues, which can be further synthesized to MAGs. 2-AG is the most studied compounds in this family. The most accepted route of biosynthesis pathways of 2-AG is illustrated in Figure 2.2. Specifically, PI can be converted to 2-AG through the formation of lyso-PI by the enzyme PLA1 (Tsutsumi et al. 1994, Ueda et al. 1993a). PC also serves as a precursor of 2-AG after being converted to 2- arachidonoyl-DAG (Oka et al. 2005). As well, 2-arachidonoyl-lysophosphatidic acid (LPA) can be converted to 2-AG (Nakane et al. 2002, Ueda et al. 1993a). PLC-diacylglycerol lipase (DAGL) is used for 2-AG synthesis, which is in postsynaptic neurons in response to depolarization and stimulation of G q/11-coupled receptors (Ahn et al. 2008). For instance, 2- arachidonoyl phosphatidyl inositol-arachidonoyl-containing phosphatidylinositol bisphosphate (PIP2) is hydrolyzed by PLCβ to ARA-containing diacylglycerol (DAG)  or β (Ueda et al. 2013). So far, DAG is thought to be the isoform responsible for 2-AG production in the central nervous system of adult mice (Lu & Mackie 2016, Pacher et al. 2006, Tanimura et al. 2010, Gao et al. 2010).

2-AG hydrolysis to glycerol and arachidonic acid occurs via one of four hydrolytic enzymes: 1) monoacyl glycerol lipase (MAGL), 2) ABH6, 3) ABH12 or 4) FAAH. Additionally, 2-AG can be oxidized by COX-2 to prostaglandin E2 glycerol ester (PGE2-GE) or by lipoxygenases to 12- hepete-G (Lu & Mackie 2016, Pacher et al. 2006). FAAH can hydrolyze 2-AG into ARA and glycerol in vitro conditions (Goparaju et al. 1998), but in vivo studies suggest that 2-AG is mainly hydrolyzed by MAGL and that 2-AG is not a substrate for FAAH (Pacher et al. 2006,

Osei-Hyiaman et al. 2005a, Sugiura et al. 1995). Furthermore, 2-AG can be oxidized to PGE2- GE or 12-hepete-G via lipoxygenases (Lu & Mackie 2016, Pacher et al. 2006) (Figure 2).

Figure 2.2. Biosynthesis and degradation of 2-AG pathways

In postsynaptic neurons, arachidonoyl-containing PIP2 can first convert to DAG via PLCβ, then this synthesized DAG is further converted to 2-AG via DAGL. Spontaneously, this PIP2 can be converted to lyso-PI via PLA1, then the lyso-PI is further converted to 2-AG via lyso-PI-PLC. Not only PIP2, but also PC can also form DAG, in which DAG is converted to 2-AG via DAGL. In presynaptic neurons, 2-AG can be broken down to ARA and glycerol via 4 different enzymes (e,g., MAGL, FAAH, ABH6 and ABH12). Additionally, in postsynaptic neurons, 2-AG can also be broken down into PGE2-GE via COX-2 or degraded into 12-HPETE-G via lipoxygenases. Source from (Oka et al. 2005, Nakane et al. 2002, Ahn et al. 2008, Gao et al. 2010, Ueda et al. 2013, Pacher et al. 2006, Goparaju et al. 1998, Lu & Mackie 2016, Osei-Hyiaman et al. 2005a, Ueda et al. 1993a)

2.5. Research gap Many studies reviewed the biosynthesis and degradation pathways of NAEs and MAGs. Dietary modification is one significant component in regulating lipid metabolism and physiological functions. Therefore, dietary modulation can thrive through multiple constantly competitive pathways and unsustainable enzyme activities. It is still unknown how dietary fat/oil modulation can impact levels of NAEs and MAGs.

In order to provide evidence-based information in this area, the present report aims to conduct a systematic review of human and animal studies which have examined the effects of modulating the composition of dietary fatty acids under isocaloric conditions on levels of NAEs and MAGs. Also, this review aims to provide information on NAE and MAG concentrations from human and animal biological samples, which will map their range in concentrations after dietary interventions, which may further explain the feasibility of adjusting NAE and MAG levels via dietary fat modulation.

2.6. Study design Although there is no standard methodology to systematically review animal experiments, this current paper adapted the tools used for clinical systematic reviews. This review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) (Moher et al. 2009) and The Cochrane Handbook for Systematic Reviews of Interventions for planning and conducting systematic search strategies for both clinical and animal studies (Higgins et al. 2008).

2.6.1 Study eligibility criteria The study eligibility criteria are based on the PICOS (population, intervention, comparison, outcome, setting) aspects of the Cochrane review format (Higgins et al. 2008), in which the specification of this review inclusion and exclusion criteria are shown in Table 2.1. Briefly, human and animal dietary intervention studies were included in this review, which reported the effect of isocaloric dietary fats intake on NAE and MAG levels in biological samples (e.g. brain, brain regions, liver, intestine, kidney, blood, etc.). Non-peer-reviewed articles, reviews/meta- analyses and in vitro studies were excluded. In addition, this study excluded articles with 1) no dietary fat treatments or comparison of different dietary fats, oils or fatty acids, or 2) non-

isocaloric fat diets (e.g. high fat vs low fat). Furthermore, this review also excluded studies comparing the effect of dietary fat from whole foods or dietary oils (e.g. olive oil or fish oil) with pure fatty acids (e.g. OA or DHA) (Diep et al. 2014). This review also excluded studies comparing dietary fat to other dietary nutrients (e.g. proteins, carbohydrates) under isocaloric diets (DiPatrizio et al. 2011b, Schwartz et al. 2008a). In addition, this article excluded studies examining the effect of liquid diets (DiPatrizio et al. 2013b) and the effect of fatty acid position in triacylglycerol (Carta et al. 2015b, Rossmeisl et al. 2012).

2.6.2. Literature search strategy The literature search was conducted using three online sources: 1) MEDLINE (Ovid)-1946 to October week 1 2017; 2) Embase-1947 to October 14th, 2017 and 3) MEDLINE-not-indexed citation articles until October 14th, 2017. The electronic search strategy in Medline and -not- indexed citation articles included both Medical Subject Headings (MeSH) and keywords in search fields including abstract (ab), title (ti), keyword heading (kw) and Medline Subject Heading (sh) following a three-section combination, step-by-step approach with ((“diet*” [MeSH]) or (“diet*.ab.ti.kw.sh.”)) AND ((“fat” [MeSH]) or (“fat*.ab.ti.kw.sh”)) AND ((ethanolamine” [MeSH]) or (“ethanolamine*.ab.ti.kw.sh”)) as an example. Similar terms in Embase included EMTREE and keywords in the same search fields ((“lipid diet” [Embase]) or (“diet*.ab.ti.kw.sh.”)) AND ((“fats” [Embase]) or (“fat*.ab.ti.kw.sh.”)) AND ((ethanolamine” [Embase]) or (“ethanolamine*.ab.ti.kw.sh”)) as an example. Overall, all the synonyms commonly used were searched in this review. For more details, the complete search terms are listed in Appendix 1,2. Microsoft Excel, 2007 was used to eliminate duplicate reports based on title and abstract, and to manage the included complete reference. Also, manual searches were performed by checking reference lists and related reviews to identify further relevant papers.

2.6.3. Data extraction, risk of bias The selected abstracts were analyzed by the primary author (LL) to identify all relevant articles. The studies were selected or excluded according to the PICOS eligibility criteria (Table 2.1). Two investigators (LL and AK) independently reviewed all the included studies. A modified study parameter form (PROFORMA) was used to extract relevant information. Also, this review

is prepared in accordance with the existing guidelines of PRISM. The PRISM 2009 Checklist (Moher et al. 2009) was used and provided in Appendix 3.

The investigators (LL and SA) evaluated the risk of bias assessments. The Cochrane Risk of Bias Tool was used for human studies (Higgins et al. 2008). We modified the Cochrane Handbook (Higgins et al. 2008) and the SYRCLE’S risk of bias tool (Hooijmans et al. 2014) for animal studies.

Table 2.1. Description of the PICOS criteria used to perform this systematic review

Components Inclusion criteria Exclusion criteria 1) Human interventional studies (Not restricted to randomized clinical trials 1) Cell-culture; only, No restriction on health status, 2) Reviews, case-reports, abstract-only, newspaper P = Population age, gender, BMI and ethnic groups at articles, non-peer reviewed articles, letters baseline); 2) Animals (No restriction on species, sex, age); 3) English only 1) Fat type in meals (vegetable oils, animal fats) I = Intervention 2) Dietary fat composition (saturated, 1) Not dietary fat/ oil studies monounsaturated, polyunsaturated fatty acids,) The comparison focuses on the effect Comparison focused on NAE and MAG levels of consumption of dietary fatty acids upon 1) non-isocaloric fat diets or just standard on blood and tissue NAE and MAG diets, 2) fasting vs. refeeding, 3) not-regular C = Comparison levels upon isocaloric fat diets. feeding (e.g., gavage, sham, emulsions, oral (Regular dietary feeding only, no administration or injection), 4) taking substance substance, no restriction on the types of (e.g., alcohol, lipopolysaccharide, pharmacological isocaloric fat diets) doses of NAEs or MAGs) 1) Levels of NAEs and MAGs in biological samples upon isocaloric fat Outcomes did not include levels of NAE and O = Outcomes diet interventions, 2) Associated tests MAG, but focused on 1) behavior tests, 2) (e.g., physiological outcomes or biomarker tests, 3) enzyme activities biochemical outcomes) No restriction on settings (e.g., multicenter, community, hospital, nursing S = Setting home, chronic care institution, outpatients or geographical regions)

2.6.4. Data synthesis and statistical analysis The specific values for NAEs and MAGs levels were extracted based on the article’s reports or estimated from figures using the PlotDigitizer (verson 2.1) software. To keep the values of NAEs and MAGs consistent between articles, this review converted all of the values into pmol/ml for blood samples and pmol/g for tissues samples. The mean ± SEM values were reported for NAE and MAG levels in Table 2.3-2.7. All of the NAE data were double checked by two investigators (LL and AK). The risk of bias assessment for human and animal studies were analysed using Review Manager (RevMan), version 5.3 (The Nordic Cocharne Centre, The Cochrane Collaboration, Copenhagen, Denmark).

This review focuses on the modulation of dietary fats / oils upon consumption of diet containing isocaloric fat content. We applied one-way ANOVA statistical analysis to the endpoint NAEs and MAGs concentrations in isocaloric groups, if the original studies only compared the effect between baseline and endpoint NAEs and MAGs (Banni et al. 2011). We also conducted one- way ANOVA statistical analysis only between isocaloric groups, when the original paper mixed with different fat content (e.g., low-fat vs high-fat) (Artmann et al. 2008a, Alvheim et al. 2012b, Alvheim et al. 2014a). Moreover, we conducted one-way ANOVA or paired T-test to 100% sufficient diet groups, if the original studies combined the 100% sufficient diet groups with diet restricted groups (Avraham et al. 2011, Liisberg et al. 2016). In addition, we conducted paired T- test in between the formula groups, but did not include the sow’s milk group in the study by Berger et.al (Berger et al. 2001a) because it was not isocaloric.

2.7. Results A flow diagram depicting the search and selection process is shown in Figure 2.3. The search identified a total of 6696 articles. After removing articles which were reviews, not in English or not journal articles (1669), there were 5027 studies left. Then, we removed 1243 duplicates. 3784 articles were screened by title and abstract. After excluding the irrelevant articles (3619), the remaining 165 articles were retrieved and screened in full-text, of which 141 were excluded. A total of 24 articles containing 6 human and 18 animal dietary interventional studies were selected. It is worth mentioning that within the included 24 studies, one experiment from Avraham et al. (2011) were excluded because the outcome may have been driven by energy

restriction rather than the fatty acid intervention. Also, an experiment that measured NAE levels in human plasma from Balvers et al. (2013) was removed because it did not have any dietary fat intervention. As well, this review only focuses on the early low-fat diet period in the study by Demizieux et al. (2016) and excluded the high-fat diet period which does not have compared group. In addition, we only report the value in the formula groups in the Berger et al. (2001a) as not the sow’s milk group.

Figure 2.3. Flow diagram of the systematic review

Electronic searching databases: Medline 1946-October week 1 2017; Medline in- progress and not-indexed citation; Embase

6494 studies identified (through October 14th, 2017): 2111 Medline; 4189 Embase; 1669 removed articles:

194 Medline-not-indexed citation -71 Not-journal articles (12 Medline, 1 Medline in-progress and not-indexed citation, 58 Embase;); -222 Not-English articles (71 Medline, 1 Medline in-progress and not-indexed citation, 150 Embase); 4825 articles left -1376 Review articles (396 Medline, 36 Medline in-progress and not-indexed citation, 944 Embase); 1041 duplicated articles (Searched by title /abstract using Excel)

3784 articles for title and abstract 3619 articles were excluded based on the screening PICOS criteria -cell-culture studies; - case-reports, abstract-only, newspaper articles, non-peer reviewed articles, letters- study design: 1) non-isocaloric fat diets or just standard diets, 2) fasting vs. refeeding, 3) not- regular feeding (e.g., gavage, sham, emulsions, oral administration or injection) 4) taking substance (e.g., alcohol, lipopolysaccharide, pharmacological doses of NAEs or MAGs), 5) behavior test only; 6) 165 articles left for full-text screening enzyme activities only;

141 articles were excluded based on PICOS criteria

24 articles were included in this review

2.7.1. Risk of bias assessments All of the human studies and the majority of animal studies were judged as having a “low” or “unclear risk of bias” using the Cochrane Risk of Bias tool (Appendix 4,5).

2.7.2. Study background / designs Six clinical studies and 18 animal studies are included in this review. The summary characteristics of the 24 studies are presented in Table 2.2 and 2.3.

2.7.3. Clinical trials: subject characteristics Subject characteristics are presented in Table 2.2. All study subjects had an overnight fast before blood sample collection, except for one study which did not provide details of their fasting duration (Ramsden et al. 2015a). Of the 6 human studies, all trials were randomized and controlled. The sample size varied from 15 subjects (Mennella et al. 2015b) to 130 subjects (Pu et al. 2016b). One study (Ramsden et al. 2015a) examined chronic headache patients. One study (Mennella et al. 2015b) examined the acute effect of dietary fats. The duration of the long-term studies was typically one month, except one study, which was 12 weeks (Ramsden et al. 2015a). Four clinical studies reported their baseline NAE and MAG levels, which are summarized in Supplemental Table 2.1.

Table 2.2. Human study designs and subject characteristics of individual studies

Authors, year Duration / Subject characteristics Sample size /group Fasting Fat content Experimental diets / countries Study design Normal: OO: 4; KO: 63 subjects (♂: wc > 7; Menhaden oil: 4; 1) OO; 2) KO (216 mg/d EPA + 90 mg/d 102cm; ♀: wc > 88cm; 35- Overweight: OO: 7; Supplemental oil: Banni et al. 4 wks / R, P, Fasted(Maki DHA); 64yrs; ) normal: BMI <25; KO: 5; Menhaden 2g/d (500 mg / (2011) / Italy C, DB et al. 2009) 3) Menhaden oil (212 mg/d EPA + 178 mg/d overweight: 25 < BMI < oil: 7; Obese: OO: capsule x 4) DHA) 30; obese: 30 < BMI <35 8; KO: 9; Menhaden oil:12 36 hyper-holesterolemia Jones et al. 29 ds / R, C, (♂, ♀; 18-65 yrs, BMI: 1) Western diet; 2) HOCAO; 3) HOCAO/ (2014) / SB, cross- 36 12 hrs 35% E ~28;) (Gillingham et al. FlaxO Canada over, 2011) Mennella et al. 15 healthy (♂, ♀; 22-40yr, 15 mins / R, 75.9% E 15 10 hrs 1) SunfO; 2) HOSunfO; 3) Virgin OO (2015b) / Italy BMI: 18.1-25, healthy) cross-over (from 30ml oil) 26% fat by wt/wt in Pintus et al. 42 hyper-cholesterolaemia 3 wks / R, SB, 1) Control cheese (rich in SFA); 42 12 hrs 90g/d sheep cheese (2013a) / Italy (♂, ♀; 30-60yr; BMI ≤ 30) cross-over 2) CLA-enriched cheese (Mele et al. 2011) 130 subjects (♂: wc > 90cm and LDL < 1mmol/L, ♀: wc >84cm and LDL < 1) CAO (60% OA, 20% LA, 10% ALA; Pu et al. 1.3mmol/L; ♂&♀: TC ≥ 30 ds / R, C, 35% E 2) HOCAO (72% OA, 15% LNA, 2%ALA); (2016b) / 1.7mmol/L or, BP cross-over, 130 12 hrs (Senanayake et al. 3) DHA+ CAO (63% OA, 13% LA, 6%DHA); Canada ≥130mmHg systolic, ≥ DB 2014) 4) CO/SaffO (18% OA, 69% LA); 85mmHg diastolic or 5) FlaxO/SaffO (18% OA, 38% LA, 32%ALA) glucose ≥ 5.5mmol/L; 46.5yr; BMI ~29) 1) Low n-6 PUFA (< 2.5% E LNA, 60mg/d Ramsden et al. Low n-6 PUFA: 27; > 10 30.4-33.6%E 55 chronic headache 12 wks / R, P, ARA, 0.6% E ALA+EPA+DHA); (2015a) / Low n-6 PUFA + hrs(Ramsden (MacIntosh et al. patients (♂, ♀; >18 yr old) SB 2) Low n-6 PUFA + High n-3 PUFA (< 2.5%E Unite States high n-3 PUFA: 28 et al. 2013) 2013) LNA,1.5%E ALA,1000mg/d EPA+DHA)

2.7.4. Animal studies: strains, types Of the 18 animal studies, more than half were conducted using mice (Table 2.3). Other animals included salmon (Alvheim et al. 2013), hamsters (Lin et al. 2013a), pigs (Berger et al. 2001a) and rats (Artmann et al. 2008a, Piras et al. 2015, Carta et al. 2015a). All animal studies examined relatively long-term effects, in which the shortest duration was one week and the longest was six months. Tissue samples of animal studies were collected with various euthanasia methods including anesthesia, decapitation, cervical dislocation or combinations.

Table 2.3. Animal studies

Authors, Type of animals Sample size / Duration Fasting Euthanasia Fat content Experimental diets year (sex, age) housing Alvheim et C57BL/6J mice 35% E for MFD; 1) 1% E LA; 2) 8% E LA; 14 wks 9-10 / 2 Unknown Unknown al. (2012b) (♂, ~23 ds) 60% E for HFD 3) 8% E LA + 1% E EPA/DHA; Expt. 1: Atlantic Sharp blow to the 6 mos 3 / 6 24 hrs 51% E 1) SO diet; 2) FO diet salmon (~340g) head Alvheim et Expt. 2: 1) SO-based fillet (8% E LA); al. (2013) Isoflurane, then 32% E (SO diet); C57BL/6J mice 16 wks 8-9 / single Unknown 2) FO-based fillet (1% E LA + 2.7% E decapitation 35% E (FO diet); (♂; ~6 wks) EPA/DHA) Alvheim et C57BL/6J mice Isoflurane, then LFD: 11% E; 16 wks 9 / single Unknown 1) 1% E LNA diet; 2) 8% E LNA diet al. (2014a) (♂; ~ 6 wks) decapitation MFD: 32% E; Experimental diets: Artmann et Sprague Dawley 1) Palm oil diet; 2) OA diet; 3) Saf-O diet; 1 wk 8 / single Unknown Decapitation 45% E (Piscitelli et al. al. (2008a) rats (♂; ~250g) 4) Pure AA + OO diet; 5) FO + OO diet, 2011) 100% sufficient energy: Avraham et Sabra mice (♀; 12.4% E (calculated 1) 5% wt/wt CAO diet; 12 ds 10 / unknown Unknown Decapitation al. (2011) ~30g) from 5% fat wt/wt 2) 4% wt/wt FO + 1% wt/wt CAO diet Cat.901682) C57BL/6J mice 9% E (Calculated from 1) 1% SO + 3% HOSunfO; Balvers et (♂; 6 wks) 6 wks 8 / 2-3 Unknown Anesthetized 4% wt/wt fat in AIN93- 2) 1% SO +2% HOSunfO + 1% FO; al. (2013) (Verhoeckx et al. M diet) 3) 1% SO + 3% FO(Verhoeckx et al. 2011) 2011) Sodium 1) Blended oils (SunfO, rapeseed oil, coconut 11.2% E (calculated Batetta et Zucker rats (♂; 4 pentobarbital oil, linseed oils (0.0% E EPA / DHA); 4 wks 6 / unknown Fasted from 7g/100g oil in al. (2009) wks) anesthesia, 2) FO (0.8% E EPA+DHA); 3) KO (0.8% E AIN-93 diet ) kill unknown EPA+DHA) Intra-cardiac piglets (♂; < 1 1) Formula (no ARA+DHA); Berger et injection of 57.9g of fat/liter, kg, <12 hours 18 ds 3 / unknown 3-4 hrs 2) Formula (Added 0.4% ARA+0.3% DHA); al. (2001a) ketamine-HCl and (4.143MJ/liter) old) (only included the formula diets) xylazine-HCl Demizieux 1) Lard (5% pork fat), 2) SaffO (2.5 % pork fat, C57BL/6J mice et al. 10 wks 7 / unknown Unknown Anesthesia 5% wt/wt 2.475% SaffO, 0.025% linseed oil), 3) Linseed (♂;3 wks) (2016) oil (4.5% pork fat, 0.5% linseed oil);

34% E (calculated from Jacome- 1) Control (high SFA and OO); 2) VA (2% E JCR:LA-cp rat 15% fat by wt/wt) and Sosa et al. 8 wks 5 / unknown Unknown Not specific VA); 3) CLA (2% E cis-9, trans-11 CLA); 4) (♂; 8 wks) 1% cholesterol(Jacome- (2016) VA/CLA (1% E VA, 0.5% E CLA) Sosa et al. 2014) 39.1% E (calculated Liisberg et C57BL/6J mice Isoflurane, then 1) Cod in western diet; 12 wks 9 / single Unknown from 19.8% fat by al. (2016) (♂; 9 wks) cardiac puncture 2) Pork in western diet wt/wt) Isoflurane, then Kim et al. C57BL/6J mice 62 ds or 22.8% E (calculated 1) Control diet (SaffO); 9 / unknown 8 hrs cervical (2016) (♂; 3 wks) 118 ds from 11% fat by wt/wt) 2) DHA (~1100mg/kg DHA administered) dislocation Syrian Golden 23.4% E (calculated Lin et al. Hamster (♂; Isofluorane, then 1) CO diet; 2) HOCAO diet; 4 wks 12 / single 12 hrs from 10% fat by w/w in (2013a) ~80-100g, plus phlebotomy 3) FO diet; 4) HOCAO+ DHA diet AIN-93 diet) 1wk acclimation) 1) Control (11.25% ovine fat, 3.75%SunfO, 2% cholesterol wt/wt); Piras et al. Zucker rats (♂; 6 Isofluorane, then 33.1% E (Calculated 14 wks 8 / single Unknown 2) CLA (11.25% ovine fat, 2.53% SunfO, (2015) wks) Decapitation from 16.6% wt/wt fat) 1.22% of CLA, 2% of cholesterol wt/wt) (Martins et al. 2012) 13.2% E (Calculated 1) LNA (3% wt/wt from SaffO /kg); Tsuyama et Jcl: ICR mice (♂, LNA: 8; CLA: Cervical 4 wks Unknown from 5.3% wt/wt in 2) CLA (3% wt/wt from CLA triacylglycerol oil al. (2009) ♀; 4 wks old) 9 / unknown dislocation diet) /kg) n-3 sufficient: Study1: ddy mice 4; 11.5% E (calculated 1) n-3 PUFA sufficient (5% wt/wt SaffO + Utero to Watanabe (♂; starts at n-3 from AIN-76 diet by DHA ethylester); 10 wks old Diethylether, then et al. utero) deficient :3 / Unknown wt) 2) n-3 PUFA deficient (5% wt/wt SaffO) decapitated (2003) unknown Study 2: ddy 23% E (calculated from 1) 10% wt/wt LA from SaffO; 4 wks 6 / unknown mice (♂;4wks) AIN-76 diet by wt) 2) 10% wt/wt DHA from FO; CD1 mice (♂; Cervical 13.7% E (Standard 1) Pork fat (0.25% by wt n-3 PUFA); Wood et al. ~16-18g, plus 1 2 wks 10 / single Unknown dislocation, then Purina 5P00 prolab 2) DHA-rich FO diet (1.45% by wt n-3 PUFA: (2010) wk acclimation) decapitated RMH300 diet) 8.9mg EPA, 59mg DHA/d) Zamberletti Sprague-Dawley Utero to 6% fat (unknown by wt 1) Control (3% peanut+ 3% rapeseed oils); et al. rats (♂; starts at postnatal 8 / unknown Unknown Unknown or by E; from Dr. 2) n-3 PUFA deficient (6% peanut oil); (2017) utero) 75 days Piccioni Laboratory) 3) n-3 PUFA enriched (6% rapeseed oil)

2.7.5. The impact of dietary fatty acids on human NAE and MAG levels Six human studies (Mennella et al. 2015b, Pintus et al. 2013a, Banni et al. 2011, Ramsden et al. 2015a, Jones et al. 2014, Pu et al. 2016b) are included in this study (Table 2.4). The dietary fatty acids interventions on NAEs and MAGs can generally be categorized into 1) dietary n-3 polyunsaturated (PUFA), 2) dietary n-6 to n-3 balance, 3) dietary n-9 monounsaturated fatty acid (MUFA), 4) mixed dietary fatty acid composition (Western diet, Mediterranean diet), or 5) dietary CLA consumption studies.

Dietary n-3 PUFA: Banni et al. (2011) compared plasma AEA and 2-AG levels after consumption of diets enriched with long-chain n-3 PUFA (Menhaden oil or krill oil) or with MUFA, olive oil, in normal weight, overweight, and obese subjects. Results showed that obese subjects consuming krill, but not Menhaden, oil had significantly decreased plasma 2-AG levels compared to baseline. In addition, we evaluated the impact of dietary fatty acids on the endpoint plasma AEA and 2-AG levels (mean ± SEM, one-way ANOVA) across the three diets, but did not find any statistical differences.

Dietary n-6 to n-3 balance: Ramsden et al. (2015a) examined the effect of modulating the dietary n-6/n-3 PUFA ratio on plasma NAEs and MAGs in chronic daily headache patients. Results showed that high n-3 /low n-6 PUFA diet had higher levels of DHEA, 2- docosaheaenoylglycerol (2-DHG) and lower levels of 2-AG compared to a low n-6 PUFA diet. However, plasma levels of AEA, OEA, PEA, and 2-OG were not different between diets.

Dietary n-9 MUFA: Mennella et al. (2015b) examined the acute dietary effects of subjects consuming isocaloric diets containing either sunflower oil, high in n-6 PUFA, high-oleic sunflower oil, high in n-9 MUFA or virgin olive oil, high in n-9 MUFA, in normal weight subjects. Results showed that both n-9 MUFA groups had higher OEA than the n-6 MUFA group at 120-mins post-meal, but plasma PEA, LEA, AEA, and 2-AG levels were not different across all three groups at any time point.

Dietary conjugated linoleic acid (CLA): Pintus et al. (2013a) compared plasma AEA and 2-AG levels upon consumption of cheese rich either in CLA, α-linolenic acid (ALA), vaccenic acid

(VA), or in saturated fat (control). After three weeks consumption of 90g cheese/ day, AEA was lower in the CLA enriched-cheese group compared to high saturated fat control diet. Notably, when subjects consumed 45g cheese/day for three weeks, there was no difference in AEA and 2- AG levels between these two diets.

Mixed dietary fatty acid composition: Two studies (Jones et al. 2014, Pu et al. 2016b) examined the modulation of plasma NAEs and MAGs by comparing a Western diet, Mediterranean diet, and habitual diet. First, Jones et al. (2014) examined 36 subjects on a cross-over study, which showed that 4 weeks consumption of a Mediterranean-like diet, containing high-oleic canola oil, led to significantly higher plasma OEA levels compared with a Western diet, enriched with LA, or with a diet enriched with blended oils, containing blended high-oleic canola oil and flaxseed oil. Also, this study showed that the blended diet had higher levels of ALEA compared to the Western diet. Diets had no effect on PEA, LEA, AEA or DHEA levels.

Second, Pu et al. (2016b) examined 130 subjects on a cross-over study, which included five diets containing different combinations of conventional oils. The results showed that OEA levels were higher in diets enriched with n-9 MUFA, such as canola oil, high-oleic canola oil and high-oleic canola/DHA oil than diets enriched with n-3 or n-6 PUFA, such as corn/safflower oil and flax/safflower oil, respectively. Particularly, the diets enriched with n-9 PUFA, high-oleic canola oil increased OEA levels more than the high-oleic canola/DHA diet. Both LEA and AEA were higher in the corn/safflower oil and flax/safflower oil diets. DHEA was higher in DHA enriched n-3 PUFA diet, high-oleic canola/DHA oil; while ALEA was higher in the ALA enriched flax/safflower oil and canola oil diets.

Table 2.4. Endpoint plasma NAEs and MAGs concentrations in human studies Author / Characters/ NAEs (pmol/ml) MAGs (pmol/ml) Diets year Duration PEA OEA LEA AEA ALEA DHEA 2-AG Other MAGs KO 4.8 ± 0.5 67.4 ± 13.1 Normal weight / Menhaden oil 4.5 ± 0.8 45.4 ± 15.1 4 wks OO 6.2 ± 0.8 49.2 ± 17.6 Banni et KO 6.0 ± 1.0 96.0 ± 25.7 Over-weight / 4 al. Menhaden oil 5.8 ± 0.5 143.8 ± 35.8 wks (2011)*δ OO 8.2 ± 1.3 97.3 ± 14.9 KO 6.3 ± 0.7 61.3 ± 10.7 Menhaden oil Obese / 4 wks 6.2 ± 0.5 84.8 ± 11.4 OO 5.5 ± 0.7 72.4 ± 13.2 Western diet Hyper- 7 ± 0.7 6.5 ± 0.3b 2.9 ± 0.1 2.3 ± 0.1 0.1 ± 0.0a 1.7 ± 0.1 Jones et HOCAO/FlaxO cholesterolemia / 6.3 ± 0.7 6.5 ± 0.6 b 3 ± 0.2 2.2 ± 0.1 0.3 ± 0.1b 1.8 ± 0.1 al. (2014)δ HOCAO 4 wks 7.7 ± 2.0 8.3 ± 0.6 a 3.2 ± 0.2 2.5 ± 0.1 0.1 ± 0.0a 2.2 ± 0.3 SunfO Normal weight / 15.3 ± 2.0 6.8 ± 0.8 4.2 ± 0.6 3.0 ± 0.1 10.6 ± 2.4 HOSunfO Post-meal 30 16.0 ± 2.1 7.8 ± 0.8 4.6 ± 0.9 3.2 ± 0.2 11.4 ± 3.0 Virgin OO mins 16.6 ± 1.9 8.1 ± 1.0 4.5 ± 0.8 3.1 ± 0.2 12.1 ± 3.4 Mennella SunfO Normal weight / 15.1 ± 2.2 6.1 ± 0.3 4.2 ± 0.5 2.6 ± 0.1 9.0 ± 3.0 et al. HOSunfO Post-meal 60 15.1 ± 1.3 7.9 ± 0.6 4.4 ± 0.5 3.2 ± 0.2 11.2 ± 3.7 (2015b) Virgin OO mins 15.1 ± 1.0 7.2 ± 1.1 4.0 ± 0.7 3.0 ± 0.3 11.7 ± 2.9 SunfO Normal weight / 12.6 ± 2.2 4.2 ± 0.9a 4.1 ± 0.5 2.1 ± 0.3 9.1 ± 2.7 HOSunfO Post-meal 120 12.9 ± 1.8 7.7 ± 0.8b 3.8 ± 0.5 2.6 ± 0.3 8.5 ± 2.4 Virgin OO mins 13.6 ± 1.4 7.1 ± 0.9b 4.0 ± 0.6 2.5 ± 0.2 8.9 ± 2.5 control-cheese Hyper- 84.9 ± 7.9a 49.4 ± 7.8 Pintus et cholesterolaemia al. (2013a) CLA cheese 56.8 ± 6.8b 40.9 ± 6.7 / 3 wks CAO 10.7 ± 0.3 6.5 ± 0.1ab 2.5 ± 0.2ab 1.4 ± 0.0ab 0.2 ± 0.0a 3.0 ± 0.1a HOCAO 11.0 ± 0.3 6.8 ± 0.1a 2.5 ± 0.2ab 1.5 ± 0.1a 0.1 ± 0.0ab 3.0 ± 0.1a Pu et al. HOCAO-DHA Healthy / 4 wks 10.7 ± 0.3 6.1 ± 0.2b 2.2 ± 0.2c 1.3 ± 0.1b 0.1 ± 0.0b 5.1 ± 0.1b (2016b)δ FlaxO/SaffO 11.0 ± 0.3 5.5 ± 0.1c 2.8 ± 0.1bd 1.3 ± 0.1b 0.3 ± 0.0c 2.7 ± 0.2a CO/SaffO 10.7 ± 0.3 5.5 ± 0.2c 3.0 ± 0.9d 1.3 ± 0.0b 0.1 ± 0.0b 2.7 ± 0.1a 6511 ± 1034 (2-OG); Ramsden Low n-6 PUFA Chronic 11.8 ± 2.8 8.3 ± 2.9 1.4 ± 0.3 1.2 ± 0.3b 1,857 ± 687b 428 ± 134b (2-DHG); et al. headache patients High n-3 / Low 5179 ± 2851 (2-OG); (2015a)δ / 12 wks 10.8 ± 2.2 8.3 ± 2.2 1.3 ± 0.2 2.2 ± 1.1a 1,471 ± 536a n-6 PUFA 656 ± 338a (2-DHG); The difference of statistical significance was indicated with different alphabet letters. * indicates the reanalysed articles. 28

2.7.6. The impact of dietary fatty acids on animal NAE and MAG levels The following sections will discuss the effect of dietary fat intake on NAE and MAG levels in different biological samples.

2.7.6.1 Blood NAE and MAG levels Six animal studies (Balvers et al. 2013, Lin et al. 2013a, Wood et al. 2010, Jacome-Sosa et al. 2016, Kim et al. 2016, Liisberg et al. 2016) measured the effects of dietary fats on NAE and MAG levels in the blood, in which four studies (Balvers et al. 2013, Wood et al. 2010, Kim et al. 2016, Liisberg et al. 2016) examined the effect of diets enriched with long-chain n-3 PUFA; while two studies (Lin et al. 2013a, Jacome-Sosa et al. 2016) focused on the effects of mixed dietary fatty acids (Table 2.5).

Dietary long-chain n-3 PUFA: Kim et al. (2016) aimed to test the effect of a diet-enriched with DHA on modulating NAE and MAG levels in plasma of mice after 62 and 118 days of feeding. The results showed that the diet-enriched with DHA significantly increased plasma DHEA levels compared to the control diet rich in safflower oil. This DHA diet also increased n-3 PUFA- derived DHEA. The diet-enriched with DHA reduced levels of n-6 PUFA-derived NAEs and MAGs including LEA, AEA, DGLEA, DTEA, 1-AG and 2-AG, 1-LG only at day 62 and 2-LG only at day 118. The diet-enriched with DHA had lower n-9 MUFA-derived OEA and 1-OG. Interestingly, the diet-enriched with DHA had significantly higher 2-OG levels than the control diet after 62 days of feeding, while levels of 2-OG were lower on the DHA diet after 118 days of feeding. As well, the diet-enriched with DHA increased PEA levels only at day 62 and decreased SEA levels.

Liisberg et al. (2016) investigated the effect of replacing lean pork meat with lean cod meat (rich in DHA) on NAE and MAG levels. Additionally, this study also included a third group which was calorie restricted, that is not included in this review. The results demonstrated that the diet- enriched with DHA increased plasma DHEA and reduced AEA and 2-AG levels after 12 weeks of feeding.

Wood et al. (2010) showed that a DHA-enriched diet (1.25% DHA) increased DHEA and 2- eicosapentanoylglycerol (2-EPG), but decreased OEA, 2-AG, and 2-OG, compared with a pork- fat diet (0.25% DHA) in mice. There were no differences in PEA, AEA, eicosanoylethanolamine (EEA), as well as 2-PG, 2-eicosanoylglycerol (2-EG) or 2-DHG levels.

Balvers et al. (2013) showed that in the free plasma pool, both fish oil diets, containing 1 % or 3%, by weight significantly increased DHEA, but decreased AEA, and OEA compared to the control, 3% high-oleic sunflower oil diet. However, only the 3% fish oil diet reduced SEA compared to the control. This study did not report the difference between the two fish oil diets, so we evaluated NAE and MAG levels across all three diets (mean ± SEM, n, one-way ANOVA followed by Tukey’s test). The analysis revealed that the 1% of fish oil diet had higher levels of free pool OEA and SEA than the 3% of fish oil diet. Levels of PEA, AEA, and DHEA were not statistically different across three diets.

In the free + esterified NAE plasma pool, both the 1% and 3% fish oil diets decreased OEA, AEA, DGLEA, DHEA, and SEA. Both diets have higher PEA, AEA, DHEA and DGLEA than control. However, only 3% fish oil diet has higher OEA levels than control. In addition, our statistical analysis showed that the 1% fish oil group had higher free + esterified OEA, AEA, DGLEA than the 3% fish oil group. Levels of PEA, DHEA, SEA did not differ between the two fish oil diets.

Furthermore, in blood cells, this study demonstrated that both fish oil diets increased DHEA and decreased AEA, PEA and DGLEA levels, but only the 3% fish oil decreased OEA compared to the control. Levels of SEA were similar across all three diets. Moreover, 1% fish oil had significantly higher DGLEA levels than the 3% fish oil group. Levels of PEA, OEA, AEA, DHEA, SEA did not differ between the two fish oil diets.

Mixed dietary fatty acid composition: Lin et al. (2013a) examined the effect of mixed dietary fatty acids on plasma NAE levels in hamsters. Animals were fed diets mixed with high-oleic canola oil (rich in MUFA), corn oil (rich in n-6 PUFA), DHA enriched high-oleic canola oil (rich in n-3 PUFA and MUFA), or fish oil (rich in n-3 PUFA) for 4 weeks. The results demonstrated

that the fish oil diet increased plasma PEA and OEA levels, but decreased AEA levels compared to the other three diets. Also, a study by Jacome-Sosa et al. (2016) tested if replacing olive oil with VA, CLA, or VA/CLA could change PEA, OEA, AEA and 2-AG levels in rats. The results showed no statistical difference in PEA, OEA, or AEA levels in plasma across diets.

Table 2.5. Endpoint blood NAEs and MAGs concentrations in animal studies Author / Animal / NAEs (pmol/ml) MAGs (pmol/ml) Diets year Blood SEA PEA OEA AEA DHEA Other NAEs 2-AG Other MAGs 1% SO + 3% 0.2 ± 0.0 7.9 ± 0.9a 11.2 ± 0.7 15.3 ± 1.1a 0.9 ± 0.1a 0.9 ± 0.1a HOSunfO Mice / (DGLEA) 1% SO + 2% Plasma: 8.4 ± 0.4a 9.8 ± 0.4 9.9 ± 0.7b 0.2 ± 0.01b 1.8 ± 0.1b ND (DGLEA) HOSunfO+1% FO free pool 1% SO + 3% FO 4.8 ± 0.4b 7.4 ± 0.3 5.2 ± 0.4c 0.1 ± 0.01b 1.6 ± 0.2b ND (DGLEA) 1% SO + 3% 2.5 ± 0.1a Mice / 449.9 ± 32.0a 476.6 ± 15.8 256.1 ± 4.5a 20.1 ± 0.7a 23.2 ± 0.7a HOSunfO (DGLEA) Plasma: Balvers 1% SO + 2% 137.6 ± 33.7 ± 1.0 ± 0.1b free+ 261.4 ± 21.5b 572.5 ± 53.4 5.4 ± 0.6b et al. HOSunfO +1% FO 12.1b 3.1b (DGLEA) δ* esterified (2013) 29.6 ± 0.5 ± 0.1c 1% SO + 3% FO pool 179.1 ± 13.7b 477.4 ± 33.1 69.8 ± 4.0c 3.3 ± 0.2c 1.4ab (DGLEA) 1% SO + 3% 158.2 ± 30.7 ± 5.7 ± 0.4a 100.4 ± 7.8 189.6 ± 13.8a 66.5 ± 5.3a HOSunfO 12.3a 2.7a (DGLEA) Mice / 1% SO +2% 68.1 ± 3.4 ± 0.2b Blood 98.3 ± 3.3 289.5 ± 9.0b 133.3 ± 5.8ab 32.5 ± 0.6b HOSunfO+1% FO 2.3b (DGLEA) cell 72.7 ± 2.0 ± 0.3c 1% SO+3% FO 88.8 ± 4.2 302.5 ± 15.4b 110 ± 11.0b 23.3 ± 2.1b 4.2b (DGLEA) CO 19.7 ± 1.4a 17.8 ± 1.4a 2.0 ± 0.2a Lin et al. HOCAO Hamster / 18.8 ± 1.8a 18.5 ± 1.3a 2.1 ± 1.6a (2013a)δ HOCAO-DHA Plasma 20.6 ± 0.9a 17.3 ± 1.0a 1.7 ± 0.1a FO 24.7 ± 1.2b 23.2 ± 1.4b 1.5 ± 1.6b Liisberg Pork diet 0.3 ± 0.0a 0.2 ± 0.0a 3.6 ± 0.4a Mice/ et al. 0.1 ± 0.0b Cod diet Plasma 0.5 ± 0.0b 2.6 ± 0.3b (2016)*δ 145,900 ± 62,100 ± 8,900 ± OO ND 11,100 6,800 4,100 Jacome- 145,200 ± 59,300 ± VA 5,600 ± 600 ND Sosa et Rat / 2,500 1,800 al. Plasma 144,100 ± 59,000 ± CLA 7,900 ± 400 ND (2016)δ 6,000 6,500 146,900 ± 57,900 ± 6,400 ± VA+CLA ND 3,300 3,100 1,900 5.0 ± 0.1a 322 ± 3.3a (1-OG); 27 ± 0.3a Mice / (1-AG) 2,730 ± 18a (2-OG); 62 ds-SaffO 460 ± 3.8a 23.4 ± 0.2a 312 ± 5.2a 3.6 ± 0.0a 0.3 ± 0.0a (LEA); Plasma 156 ± 1.3a 41.7 ± 0.5a (1-LG); NR (-LEA); (2-AG) 5,870 ± 19.9 (2-LG) 32

Kim et 4.0 ± 0.0a al. (DGLEA); (2016)*δ 8.6 ± 0.1a (DTEA) 21.3 ± 0.3b (LEA); 2.0 ± 0.1b 172 ± 2.1b (1-OG); NR (-LEA); (1-AG) 15,820 ± 20.4b (2-OG); 62 ds-DHA 258 ± 3.2b 26.1 ± 0.3b 187 ± 3b 1.3 ± 0.0b 3.4 ± 0.0b 2.3 ± 0.0b 102 ± 0.8b 83.7 ± 2.2b (1-LG); (DGLEA); 0.9 (2-AG) 5,930 ± 75 (2-LG) ± 0.0b (DTEA) 42.5 ± 0.4a (LEA); 0.02 ± 0.0 (- 7.2 ± 0.2a 514 ± 6.8a (1-OG); 118 ds LEA); (1-AG) 3,900 ± 47a (2-OG); 838 ± 8a 32.4 ± 4.7 846 ± 23.9a 6.0 ± 0.1a 0.3 ± 0.0a SaffO 5.5 ± 0.1 181 ± 2.6a 60 ± 2.5 (1-LG); (DGLEA); (2-AG) 6,350 ± 57 (2-LG) 13.6 ± 0.3a (DTEA) 29 ± 0.2b (LEA); 0.02 ± 0.0 1.8 ± 0.0b 213 ± 1.3b (1-OG); 118 ds (-LEA); (1-AG) 1,880 ± 10b (2-OG); 470 ± 5b 35.4 ± 0.2 347 ± 7.4b 1.8 ± 0.0b 4.2 ± 0.0b DHA 3.4 ± 0.0 82.7 ± 0.7b 53.1 ± 0.8 (1-LG); (DGLEA); (2-AG) 5,390 ± 37 (2-LG) 1.4 ± 0.0b (DTEA) 25,900 ± 6415 (2-PG); 11,978 ± 7,237a (2-OG); 0.8 ± 0.7 4,150 ± Pork-fat 23.6 ± 7.7 23 ± 6.6a 1.6 ± 1.1 3.4 ± 0.8a 542 ± 300a (2-EPG); (EEA) 1983a 541 ± 117 (2-EG); Wood et Mice / 5,342 ± 2,460 (2-DHG) al. Plasma 21,664 ± 5,719 (2-PG); (2010) 5,140.4 ± 2,356.2b (2-OG) 0.6 ± 0.6 2,316 ± FO 20.2 ± 4.4 10 ± 2.3b 1.1 ± 0.7 6.3 ± 1.6b 1,288 ± 821b (2-EPG); (EEA) 1,588b 616 ± 174 (2-EG); 6,338 ± 2,559 (2-DHG) The difference of statistical significance was indicated with different alphabet letters. * indicates the reanalysed articles.

2.7.6.2 Brain NAE and MAG levels Eleven studies are included in this section (Alvheim et al. 2014a, Alvheim et al. 2012b, Artmann et al. 2008a, Alvheim et al. 2013, Watanabe et al. 2003, Wood et al. 2010, Berger et al. 2001a, Tsuyama et al. 2009, Avraham et al. 2011, Zamberletti et al. 2017, Demizieux et al. 2016) and detailed NAE and MAG concentrations are presented in Table 2.6.

Dietary n-6 PUFA: We identified three studies (Alvheim et al. 2014a, Alvheim et al. 2013, Alvheim et al. 2012b) examining the effect of dietary LNA on brain AEA and 2-AG concentrations. Alvheim et al. (2012b) showed that after 14 weeks of isocaloric high-fat diets (HFD), containing 60% of energy (E) from fat, the 8% E LNA diet, also called the high LNA diet, increased combined 1-AG + 2-AG levels compared to the 1% E LNA diet, also called low LNA diet, and 8% E LNA+ 1% E EPA/DHA in mice whole brain. Based on our re-analysed statistics across all three groups, this effect was not seen in the isocaloric medium-fat diets (MFD), containing 35% E from fat.

Alvheim et al. (2013) showed upon a MFD, containing 35% E from fat, levels of cerebral cortex AEA and 2-AG were not different between mice fed fish oil comprised of 2.7% E EPA/DHA, or soya-bean oil comprised of 8% E LNA.

This review re-analyzed the effect of LNA on NAE and MAG levels upon isocaloric MFDs and LFDs, separately in the Alvheim et al. (2014a) study. The results demonstrated that increasing dietary LNA from 1% E to 8% E did not change the whole brain 2-AG levels in isocaloric MFDs or isocaloric LFDs, containing 12.5 % E. Dietary n-3 PUFA: Two studies (Wood et al. 2010, Avraham et al. 2011) examined the effect of dietary n-3 PUFA on NAEs and MAGs. For instance, Wood et al. (2010) showed that mice fed a n-3 PUFA diet, containing 1.25% DHA by weight for two weeks increased DHEA and 2-EPG (eicosapentaenoyl glycerol) and had decreased AEA compared to the control, a pork fat diet, containing 0.25% DHA by weight, in the whole brain. Although this study measured other NAEs, including PEA, OEA, EEA, and MAGs, including 2-OG, 2-EG, 2-PG, 2-DHG, the authors found no statistical difference between diets. In addition, Avraham et al. (2011) fed mice with different types of dietary n-3 PUFA. The results showed that hypothalamus and

hippocampus levels of AEA and 2-AG were similar between diet enriched with canola oil or a diet enriched with fish oil.

Dietary n-6 to n-3 balance: Watanabe et al. (2003) conducted two experiments to examine how dietary n-3 status influences NAE and MAG levels in mouse whole brain. In the first experiment, from utero to 10 weeks of age mice consuming an n-3 PUFA sufficient diet, containing 5% fish oil by weight, had higher n-3 PUFA MAGs including 2-DHG and 1(3)-DHG compared with mice fed an n-3 PUFA deficient diet, containing 5% safflower oil by weight. As expected, the n- 3 PUFA deficient diet group, which contained higher n-6 PUFA (LA), had higher levels of n-6 PUFA corresponding MAGs including 2-AG, 2-PG+1(3)-OG, and higher n-9 MUFA derivative 2-OG, as well as higher 1 (3)-SG than the n-3 PUFA sufficient diet. In the second experiment, mice were fed either a high n-3 PUFA diet comprised of DHA-rich fish oil or a low n-3 PUFA diet comprised of LA rich safflower oil for four weeks. The results showed that the high n-3 PUFA diet reduced levels of 2-AG, 1(3)-AG and 2-PG+1(3)-OG compared with the low n-3 PUFA diet. However, 2-DHG, 1(3)-DHG, 1(3)-PG, 2-OG and 1(3)-SG were the same between the two diets.

A study by Demizieux et al. (2016) investigated the role of the dietary n-6 and n-3 PUFA ratio in relationship to endocannabinoid tone at early life. Three-week-old C57BL/6 male mice received diets with 5% by weight fat content, enriched with either saturated fat, using lard fat, LNA, using safflower oil or, ALA, using linseed oil for 10 weeks. In the brain, the diet-enriched with ALA reduced AEA and 2-AG levels compared to the diet-enriched with LNA.

Zamberletti et al. (2017) examined NAE and MAG levels in rat brain regions after feeding a standard control diet comprised of 3% of peanut oil and 3% rapeseed oil or an n-3 PUFA deficient diet with 6% peanut oil, or an n-3 PUFA enriched diet comprised of 6% rapeseed oil th from birth to the 75 day. This study compared each n-3 PUFA diet separately with the control diet. Thus, we conducted one-way ANOVA to analyse the statistical difference between control, n-3 PUFA deficient and enriched diets. Results showed that in the prefrontal cortex, AEA levels were statistically lower in n-3 PUFA enriched diet than the other two diets. Also, prefrontal cortex AEA and 2-AG levels were reduced in n-3 PUFA enriched diet by 40% compared to the

n-3 PUFA deficient and control diet. The levels of 2-DHG were higher in control diet than both n-3 PUFAs groups. However, prefrontal cortex PEA, OEA, DHEA, LEA, and EPEA levels were not statistically different between the diets. In the hippocampus, one-way ANOVA analysis showed no statistical difference in AEA levels between the three diets. DHEA levels were higher in the n-3 PUFA enriched diet compared to the other two diets. 2-AG levels were higher in the n- 3 deficient diet than the n-3 PUFA enriched diet. 2-DHG levels were higher in the standard diet than n-3 PUFA enriched diet hippocampus PEA, OEA, LEA and EPEA levels were not statistically different between diets.

Berger et al. (2001a) examined the effect of formula enriched with ARA (0.2% E) and DHA (0.16% E) on levels of NAEs and MAGs in piglet brain. This review focuses on the comparison between the formula without ARA and DHA and formula with ARA and DHA groups only. This study showed that whole brain levels of AEA, DTEA (docosatetraenoylethanolamide, C22:4 n- 6), EPEA, DPEA, DHEA, and DHG were lower in the formula without ARA and DHA diet compared to the formula with ARA and DHA diet. Levels of 2-AG and 2-DTG (2- docosatetraenoylglycerol, C22:4 n-6) were similar between this two diets. In addition, the levels of AEA and DHEA in specific brain regions were also measured. Levels of DHEA were lower in the formula without ARA and DHA in the brainstem, auditory cortex, and striatum compared to the formula with ARA and DHA. However, AEA levels in brain regions were not different between these two diets.

Dietary CLA: One study (Tsuyama et al. 2009) examined the effects of 3% dietary LNA by weight, and 3% CLA by weight in a LFD (13.2% E), on the level of brain NAEs and MAGs in mice. Mice on the CLA-enriched diet had lower levels of 2-AG than the LNA enriched diet in the cerebral cortex, but not hypothalamus. Also, this study found no statistical difference in PEA, AEA and OEA levels between diets.

Mixed dietary fatty acid compositions: Artmann et al. (2008a) investigated a Mediterranean diet, Western diet, and a high-ARA diet (13% wt/wt) on NAE and MAG levels. One week feeding a control LFD, containing 11% E fat, or five experimental isocaloric HFDs containing 45% E (Piscitelli et al. 2011), influenced the brain NAE and MAG levels. In this review, we only

analyzed the effects of the five isocaloric experimental diets on brain NAEs and MAGs using one-way ANOVA followed by Tukey’s post-hoc test (mean ± SEM, n=7 or 8/group). The ARA diet had higher levels of OEA and 2-AG than the palm oil diet (high in PA, saturated fat). Also, olive oil, ARA and fish oil rich diets led to significantly higher levels of AEA than the palm oil and safflower oil diets. In addition, olive oil and safflower oil diets resulted in higher levels of LEA compared with the palm oil diet. There was no statistical difference between PEA, SEA, and DHEA across all groups.

Table 2.6. NAE and MG levels in brain or brain regions

Author/ Species / NAEs (pmol/g) MAGs (pmol/g) Diets year Brain samples PEA OEA AEA DHEA Other NAEs 2-AG (or AGs&) Other MAGs MFD:1%E ND 11,095 ± 1,057& LNA+7%E SFA Mice / Whole MFD 8%E LNA ND 12,414 ± 528& brainδ MFD 8%E LNA Alvheim et ND 10,829 ± 793& +1%E EPA/DHA al. (2012b)* HFD 1%E LNA + ND 10,565 ± 528&a 7%E SFA Mice / Whole HFD 8%E LNA ND 13,474 ± 793&b brainδ HFD 8%E LNA ND 10,304 ± 528&a +1%E EPA/DHA Mice / 12,946 ± Alvheim et SO 36.5 ± 5.5 Cerebral 10,567& al. (2013) FO cortexδ 27.3 ± 2.0 11,358 ± 1,057& LFD 1%E LNA Mice / 39.0 ± 3.8 LFD 8%E LNA Cerebral 38.5 ± 4.9 MFD 1%E LNA cortexδ 25.9 ± 5.1 Alvheim et MFD 8%E LNA 32.1 ± 3.3 al. (2014a)* LFD 1%E LNA 12,229 ± 1,241 Mice / Whole LFD 8%E LNA 13,022 ± 872 brainδ MFD 1%E LNA 11,516 ± 1,268

MFD 8%E LNA 12,546 ± 660 171.2 ± 12.0 (SEA); Palm oil 324.0 ± 9.1 173.0 ± 8.1a 12.5 ± 0.6a 25.3 ± 2.7 2.2 ± 0.5a (LEA); 24,400 ± 2,500a ND (EPEA) 161.4 ± 11.9 (SEA); 208.8 ± OO 332.2 ± 7.6 17.6 ± 1.4b 35.1 ± 4.4 7.5 ± 1.9b (LEA); 53,900 ± 8,600ab 10.1ab Artmann et Rats / Whole ND (EPEA) al. (2008a)* brainδ (2-AG) 140.8 ± 23.6 (SEA); 336.9 ± 201.6 ± 51,300 ± SaffO 13.6 ± 0.8a 31.0 ± 2.1 8.2 ± 1.5b (LEA); 11.3 12.3ab 10,400ab ND (EPEA) 137.1 ± 14.8 (SEA); ARA* 342.5 ± 8.5 213.2 ± 8.5b 17.7 ± 0.8b 28.3 ± 2.1 5.8 ± 1.2ab (LEA); 69,100 ± 9,800b ND (EPEA) 38

138.6 ± 12.4 (SEA); 319.1 ± FO 194.1 ± 5.1ab 14.2 ± 0.7b 29.9 ± 1.8 5.2 ± 1.2ab (LEA); 53,300 ± 7,000ab 10.8 ND (EPEA) CAO Mice / 13.5 ± 1.0 8,540 ± 1,420 Hypothalamusδ CAO+FO 11.4 ± 1.4 10,380 ± 1,750 Avraham et (2-AG) al. (2011)* CAO Mice / 12.3 ± 1.6 8,590 ± 1,410 Hippocampus δ CAO+FO 10.9 ± 1.5 10,390 ± 1,650 (2-AG) 3,530 ± 650 (2- 740 ± 30a (DTEA); Formula no DTG); 1,020 ± 340a 950 ± 210a 180 ± 90a (DPEA); 66,000 ± 9,380 ARA+DHA 3,870 ± 30a (2- Piglets / 330 ± 120a (EPEA) DHG) Whole brainδ 6,230 ± 1,820 1,150 ± 20b (DTEA); Formula added (2-DTG); 5,930 3,960 ± 910b 9,030 ± 360b 1,690 ± 490b (DPEA); 44,400 ± 16,130 ARA+DHA ± 1,370b (2- 1,740 ± 470b (EPEA) DHG) Formula no 1,100 ± 90 700 ± 60a ARA+DHA Piglets / Brain- Formula added stemδ 4,900 ± 1,600 ± 230 ARA+DHA 1,790b Formula no Piglets / 3,000 ± 230 800 ± 230a Berger et al. ARA+DHA Auditory (2001a)* Formula added cortexδ 4,000 ± 690 4,400 ± 870b ARA+ DHA Formula no 3,500 ± 350 3,300 ± 170 ARA+DHA Piglets / Visual Formula added cortexδ ND ND ARA+DHA Formula no ARA+ 500 ± 60 500 ± 170 DHA Piglets / Formula added Cerebellumδ ND ND ARA+ DHA Formula no 1,800 ± 20 1,000 ± 120a ARA+DHA Piglets / Formula added Striatumδ 1,500 ± 60 2,000 ± 290b ARA+DHA

Formula no Piglets / 2,200 ± 460 1,200 ± 120 ARA+DHA Hippocampusδ Formula added 2,200 ± 170 900 ± 60 ARA+DHA Lard fat 23.7 ± 3.8ab 3,490 ± 600ab Demizieux Mice / Whole a a et al. (2016) SaffO brainδ (2-AG) 31.0 ± 1.7 4,290 ± 200 Linseed oil 17.3 ± 3.8b 3,120 ± 450b 383.2 ± LNA Mice / 213.1± 1.7 13.3 ± 1.9 23,000 ± 510a 49.4 Cerebral 518.9 ± CLA Cortexδ(2-AG) 268.3 ± 30 13.9 ± 1.2 18,492 ± 667b Tsuyama et 52.1 al. (2009) 885.8 ± LNA Mice / 443.0 ± 48.3 ND 49,900 ± 3,700 23.9 Hypothalamus 710.4 ± CLA δ(2-AG) 381.9 ± 31 ND 47,605 ± 6,687 65.5 385 ± 26 (1(3)- PG); 316 ± 32a (2-PG+1(3)- OG); 460 ±100a 107 ± NAa (2- n-3 PUFA (2-AG); OG); sufficient 380 ± 40 (1(3)- 490 ± 81a (1(3)- AG) SG); 71 ± NAa (2- Watanabe et Mice / Whole DHG) al. (2003) brainδ 130 ± 22a (1(3)- (study1) DHG) 435 ± 27 (1(3)- PG); b 1,560 ± 300b (2- 1,001 ± 124 (2-PG+1(3)- n-3 PUFA AG); OG); deficient 500 ± 70 (1(3)- 242 ± 27b (2- AG) OG); 782 ± 22b (1(3)- SG);

17 ± 11b (2- DHG); NDb (1(3)-DHG)

272 ± 27 (1(3)- PG); 809 ± 151a (2-PG+1(3)- 960 ± 240a (2- OG); AG); 311 ± 97 (2- Low n-3 PUFA 210 ± 50a (1(3)- OG); AG) 1,225 ± 242 (1(3)-SG); 1,603 ± 255 (2-DHG); 59 ± 8 (1(3)- Watanabe et Mice / Whole DHG); al. (2003) brainδ 217 ± 16 (1(3)- (Study 2) PG); 419 ± 38b (2-PG+1(3)-OG; 390 ± 30b (2- 186 ± 43 (2- AG); OG); High DHA 90 ± 10b (1(3)- 1,084 ± 237 AG) (1(3)-SG); 1,473 ± 11 (2- DHG); 85 ± NA (1(3)- DHG); 5,557 ± 1,571 (2-OG); Wood et al. Mice / Whole 186.3 ± 2,809 ± 546 (2- Pork fat 126.3 ± 94.9 5.4 ± 1.3a 10.1 ± 1.8a 13 ± 2.9 (EEA) 15,067 ± 5,550 (2010) brainδ 70.8 EG); 5,961 ± 2,996 (2-PG);

143.7 ± 54a (2- EPG); 6,210 ± 3,654 (2-DHG)

5,637 ± 1,881 (2-OG); 3,199 ± 572 (2- EG); 218.7 ± 6,233 ± 3,389 FO 98 ± 21.8 4.2 ± 1.2b 12.4 ± 2b 12.4 ± 2.8 (EEA) 14,486 ± 5,155 77.4 (2-PG); 545 ± 205b (2- EPG); 7,383 ± 4,074 (2-DHG) 14,660 ± 3,500 3% peanut oil / 47,050 ± (LEA); 3,900 ± 200a (2- 7,220 ± 730 620 ± 30 92.4 ± 14.4a 4,700 ± 320a 3% rapeseed oil 2,030 33,170 ± 5,150 DHG) (EPEA) 11,170 ± 4,560 Rats / n-3 PUFA 63,380 ± (LEA); 2,390 ± 490b (2- Prefrontal 6,210 ± 960 580 ± 170 86.3 ± 18a 3,120 ± 630a deficient 14,150 51,900 ± 12,100 DHG) cortexδ (EPEA) 12,040 ± 2,230 n-3 PUFA 8,730 ± 49,920 ± (LEA); 2,000 ± 100b (2- Zamberletti 450 ± 100 24 ± 12b 2,780 ± 200b enriched 1,700 6,060 45,270 ± 2,730 DHG) et al. (EPEA) (2017)* 23,420 ± 8,290 3% peanut oil / 50,930 ± 3,390 ± 50a (2- 1,080 ± 40 1,050 ± 60 50.7 ± 1.7 (LEA); 7,640 ± 500ab 3% rapeseed oil 2,770a DHG) 7,420 ± 2,010 (EPEA) 45,050 ± 12,960 n-3 PUFA 36,570 ± 2,770 ± 290ab (2- Rats / 870 ± 100 900 ± 50 49.4 ± 6.5 (LEA); 8,260 ± 1,110a deficient 3,240a DHG) Hippocampus 6,710 ± 1,100 (EPEA) 40,720 ± 6,120 n-3 PUFA 81,940 ± (LEA); 1,380 ± 540b (2- 860 ± 170 710 ± 140 34.8 ± 3.1 3,900 ± 1,440b enriched 11,110b 10,360 ± 2,700 DHG) (EPEA) The difference of statistical significance was indicated with different alphabet letters. * indicates the reanalysed articles. 42

2.7.6.4. Other organs NAE and MAG levels Nine studies measured non-central nervous system tissues, in which all studies measured liver NAE and MAG levels (Alvheim et al. 2012b, Alvheim et al. 2014a, Alvheim et al. 2013, Artmann et al. 2008a, Batetta et al. 2009, Lin et al. 2013a, Demizieux et al. 2016, Piras et al. 2015), three of the nine studies (Artmann et al. 2008a, Lin et al. 2013a, Jacome-Sosa et al. 2016) examined the small intestine, three studies (Jacome-Sosa et al. 2016, Batetta et al. 2009, Demizieux et al. 2016) examined adipose tissue, two studies examined muscle (Jacome-Sosa et al. 2016, Demizieux et al. 2016), and one study (Batetta et al. 2009) examined the heart (Table 2.7).

Dietary n-6 PUFA: Three studies examined the effect of increasing dietary LNA on modulating levels of NAEs and MAGs in the liver (Alvheim et al. 2012b, Alvheim et al. 2014a, Alvheim et al. 2013). We re-analyzed two studies (Alvheim et al. 2012b, Alvheim et al. 2013) to test the isocaloric diet groups. Alvheim et al. (2012b) showed that in HFD group, mice fed the 8% E LNA diet had higher liver levels of AEA and combined 1-AG+2-AG in the liver compared with 1% E LNA diet. However, in the MFD, mice fed the 8% E LNA diet had no statistical effects on combined 1- AG+2-AG or AEA in the liver compared with the 1% E LNA diet, In addition, (Alvheim et al. 2013) showed that Atlantic salmon fed a soya-bean oil diet had higher hepatic AEA, but not the combined 1-AG + 2-AG than Atlantic salmon fed a fish oil diet. In the mouse experiment, mice consuming fish fillets enriched with soya-bean oil had higher levels of combined 1-AG + 2-AG, but not AEA compared with the mice consuming fish fillet enriched with the fish oil in the liver. Also, Alvheim et al. (2014a) showed that the 8% E LNA, diet increased levels of liver AEA and 2-AG compared to 1% E LNA for both the MFD and LFD, containing 12.5 % E fat content.

Dietary n-3 PUFA: Batetta et al. (2009) examined the effect of dietary n-3 PUFA on rat liver AEA and 2-AG levels. Results showed that both n-3 PUFA enriched diets, fish oil or krill oil, reduced levels of AEA compared with control, blended oils containing 0% EPA/DHA, in the liver and visceral adipose tissue. Compared to the control, the krill oil diet decreased AEA in the heart; and decreased 2-AG in visceral adipose tissue. In contrast, the krill oil diet had higher 2- AG levels in the liver and heart than control.

A study by Demizieux et al. (2016) investigated the role of dietary n-6 and n-3 ratio in the liver, muscle, viscous adipose tissue (VAT) and subcutaneous adipose tissue (SCAT) on early life. Results showed that the diet-enriched with ALA significantly reduced AEA levels in the liver, muscle, and SCAT compared to a diet-enriched with LNA and saturated fat. The diet-enriched with ALA had lower levels of 2-AG in the liver and muscle in comparison with the diet-enriched with LNA. In addition, the diet-enriched with saturated fat also had lower 2-AG than the diet- enriched with LNA in the muscle. Among all three diets, levels of AEA were the same in VAT, while levels of 2-AG were the same in both VAT and SCAT.

Dietary CLA: One study (Piras et al. 2015) examined the effect of four weeks of a CLA-enriched diet on OEA and PEA levels in the rat liver. This study demonstrated that the CLA-enriched diet increased levels of PEA and OEA, respectively, compared with the control which contained sunflower oil, rich in LNA.

Mixed fatty acid composition: Three studies (Artmann et al. 2008a, Lin et al. 2013a, Jacome- Sosa et al. 2016) are included in this section. As we mentioned earlier, we only considered the five experimental oil groups in the study by Artmann et al. (2008a). The diets enriched with olive oil, ARA or safflower oil resulted in higher levels of liver OEA, AEA, and LEA, as compared to the other diets, respectively. The ARA diet led to higher liver levels of 2-AG than the safflower oil and fish oil diets. In contrast, the fish oil diet resulted in the lowest levels of all NAEs, except for DHEA, which was higher compared to the safflower oil diet and ARA diet. Liver levels of PEA and SEA were higher in mice fed the palm oil diet and olive oil diet compared to the ARA or fish oil diet. In the jejunum, NAE levels reflected their respective precursor fatty acids, in which fish oil diet had higher DHEA, palm oil diet had higher PEA compared to ARA and fish oil and the olive oil diet had higher OEA compared to fish oil and ARA diet had higher AEA. SEA was higher in mice fed the palm oil, safflower oil, and olive oil compared to the ARA or fish oil. Levels of LEA were the same across all experimental diet.

In addition, Lin et al. (2013a) demonstrated that in liver and proximal intestine, a fish oil diet decreased AEA levels compared to the other three diets; while OEA levels were higher upon the n-9 MUFA enriched diets (high-oleic canola oil and DHA-enriched high-oleic canola oil),

compared with the n-6 and n-3 PUFA enriched diets (corn oil and fish oil). There was no statistical difference in PEA levels across all diets in the liver or proximal intestine.

Furthermore, a study by Jacome-Sosa et al. (2016) demonstrated that NAEs and 2-AG levels in rats can be modulated by dietary fatty acids. For example, the diet-enriched with olive oil led to higher PEA levels in skeletal muscle than the VA diet, but liver PEA levels were only higher than with the VA and VA/CLA diets. Upon consuming a diet enriched with olive oil, rats had significantly higher skeletal muscle OEA compared to other diets. In jejunal mucosa, OEA levels were higher in the VA diet than the VA/CLA diet and olive oil diet. Also, in VAT, OEA levels were significantly higher in rats consuming the diets rich in olive oil and CLA compared with diets rich in VA and VA/CLA. In VAT, AEA levels were significantly higher in the CLA diet than the VA diet, but not in skeletal muscle. In jejunal mucosa, the VA diet led to higher AEA than the olive oil or CLA diets. In both liver and VAT, 2-AG levels were significantly higher in the diet-enriched with olive oil, but this did not occur in the in skeletal muscle and jejunal mucosa.

Table 2.7. NAEs and MAGs in the body NAEs (pmol/g) 2-AG (or AGs&) Author/ year Diets Species / Organs PEA OEA AEA DHEA Other NAEs (pmol/g) MFD 1%E 77.7 ± 11.5 1,849 ± 264& LNA+7%E SFA MFD 8%E LNA 115.1 ± 11.5 3,962 ± 528& MFD 8%E LNA Alvheim et al. 74.8 ± 17.3 3,170 ± 1,321& +1%E EPA/DHA Mice / Liverδ (2012b)* HFD1%E LNA 69.1 ± 8.6a 1,585 ± 159&a HFD 8%E LNA 161.2 ± 31.7b 5,019 ± 528&b HFD 8%E LNA 103.6 ± 5,283 ± 1,057&b +1%E EPA/DHA 28.8ab Expt.1: SO Atlantic salmon / 6.0 ± 0.3a 3,697,834 ± 792,393 Expt.1: FO Liverδ 4.0 ± 0.6b 2,113,048 ± 79,239 Alvheim et al. Expt.2: salmon (2013) NA 290.5 ± 26.4&a fillet-fed SO Mice / Liverδ Expt.2: salmon NA 158.5 ± 26.4&b fillet-fed FO Alvheim et al. LFD 1%E LNA 17.4 ± 2.5a 245.6 ± 18.5a (2014a)* LFD 8%E LNA 29.6 ± 3.7b 491.3 ± 81.9b MFD 1%E LNA Mice / Liverδ 13.8 ± 1.7a 250.9 ± 23.8a b b MFD 8%E LNA 23.6 ± 4 361.3 ± 31.7 136.1 ± 7.4a (SEA); 145.2 ± Palm oil 149.3 ± 8.0a 15.7 ± 1.0a 8.6 ± 0.8a 392.7 ± 58.7 (LEA); 33,400 ± 4,400a 11.4ab ND (EPEA) 113.1 ± 3.7ab (SEA); OO 128.1 ± 4.2ab 194.1 ± 7.2a 14.6 ± 0.8a 7.7 ± 1.1a 511.7 ± 35.9 (LEA); 35,300 ± 3,800a ND (EPEA) Artmann et al. Rats / Jejunum δ(2- 126.2 ± 11.0a (SEA); 122.8 ± 139.4 ± (2008a)* SaffO AG) 12.5 ± 0.9a 5.8 ± 0.8a 542.8 ± 71.3 (LEA); 24,000 ± 2,300a 12.7ab 18.6ab ND (EPEA) 94.1 ± 8.8b (SEA); 154.0 ± ARA 101.4 ± 11.8b 35.9 ± 4.4b 5.3 ± 0.8a 351.6 ± 72.6 (LEA); 65,600 ± 8,900b 19.5ab ND (EPEA) 86.7 ± 4.5b (SEA); FO 105.8 ± 7.5b 133.4 ± 13.3b 9.2 ± 0.6a 26.6 ± 2.0b 39,400 ± 5,700a 390.6 ± 59.2 (LEA);

21.6 ± 2.4 (EPEA) 34.5 ± 6.4ab (SEA); Palm oil 85.9 ± 17.5a 16.5 ± 2.3a 3.7 ± 0.7ab 3.6 ± 0.4ac 1.8 ± 0.3ab (LEA); 5,300 ± 300a ND (EPEA) 41.9 ± 4.4a (SEA); OO 66.8 ± 6.9ab 29.5 ± 2.9b 4.2 ± 0.6ab 3.8 ± 0.3ac 1.9 ± 0.2ab (LEA); 5,000 ± 300a ND (EPEA) 37.7 ± 6.8ab (SEA); SaffO Rats / Liverδ(2-AG) 46.2 ± 7.1bc 12.0 ± 1.6a 3.1 ± 0.5a 2.5 ± 0.6ab 2.7 ± 0.4a (LEA); 3,300 ± 300b ND (EPEA) 22.2 ± 1.4bc (SEA); ARA 33.1 ± 2.5bc 13.4 ± 1.0a 7.4 ± 1.4b 1.2 ± 0.2b 1.0 ± 0.1b (LEA); 5,700 ± 500a ND (EPEA) 17.3 ± 2.5c (SEA); FO 28.5 ± 4.8c 11.5 ± 1.4a 1.9 ± 0.2a 5.2 ± 0.5c 0.9 ± 0.1b (LEA); 1,600 ± 300c ND (EPEA) Blended oils 38.4 ± 4.4a 349.4 ± 67a FO Rats / VAT 24.9 ± 11.4b 349.1 ± 215.8a KO 16.3 ± 2.6b 90.6 ± 12.3b Batetta et al. Blended oils 1,100 ± 200a 5,600 ± 2,800a (2009) FO Rats/ Liverδ 750 ± 100b 5,800 ± 2,200a KO 200 ± 70c 10,100 ± 1,800b Blended oils 1,890 ± 330a 9,760 ± 4,350a FO Rats / Heartδ 1,560 ± 160a 12,760 ± 6,680ab KO 490 ± 220b 19,740 ± 2,630b Lard fat 3.7 ± 0.3a 1,395.6 ± 190a SaffO Mice / Liverδ(2-AG) 4.2 ± 0.4a 2,442.5 ± 320b Linseed oil 1.7 ± 0.4b 1,082.9 ± 50a Lard fat 24.5 ± 1.4a 1,444.4 ± 103a Mice / Muscleδ(2- SaffO 30.9 ± 1.8a 1,852.3 ± 37a Demizieux et AG) Linseed oil 16.3 ± 2.5b 904.6 ± 139b al. (2016) Lard fat 17.6 ± 2.2 443 ± 61

SaffO Mice / VATδ(2-AG) 34.1 ± 7.0 758.7 ± 140 Linseed oil 19.2 ± 4.0 1,709.3 ± 647.4 Lard fat 122.7 ± 8.5a 3,516 ± 437 SaffO Mice / ScAT δ(2-AG) 149.3 ± 22.4a 3,480.5 ± 847 Linseed oil 82.7 ± 12.7b 3,914 ± 757

CO 392.4 ± 39.3 261.1 ± 22.2a 5.8 ± 0.3a HOCAO Hamsters 442.1 ± 38.1 362.6 ± 40.2b 6.3 ± 0.5a HOCAO-DHA / Liverδ 431.9 ± 42.5 317.9 ± 26.9b 5.3 ± 0.5a Lin et al. FO 380.9 ± 32.1 260.2 ± 17.1a 3.7 ± 0.3b (2013a) Corn oil 689.8 ± 29.5 308.9 ± 28.8a 23.4 ± 2.9a HOCAO Hamsters / Prox- 657.6 ± 28.9 417.9 ± 27.6b 20.3 ± 0.9a HOCAO-DHA small intestine δ 687.0 ± 26.3 392 ± 26.6b 27.5 ± 3.1a FO 692.9 ± 39.1 355.2 ± 16.3a 14.4 ± 2.9b OO 222.1 ± 8.1b 27.4 ± 3.3 NR 3,977 ± 572c VA Rats / 95.8 ± 7a ND NR 653 ± 84a CLA Liver δ 197.2 ± 28.8b 22 ± 7.3 NR 2,063 ± 336b VA+CLA 97 ± 2.2a ND NR 1,000 ± 168ab OO 378.8 ± 38.9a 118.4 ± 18.5a 84.3 ± 21.2a 158,600 ± 53,200 VA 725.7 ± 70.8b 203.7 ± 35.8b 311.0 ± 72.5b 103,567 ± 12,430 Rats / Jejunal CLA 470.8 ± 56.6a 141.6 ± 9.3ab 121.6 ± 19.5a 164,783 ± 56,502 mucosaδ 154.5 ± VA+CLA 421.2 ± 49.6a 120.7 ± 5.3a 142,849 ± 37,809 30.1ab Jacome-Sosa et OO 251.5 ± 8.1b 179.5 ± 6.2b 24.4 ± 3.8 4,500 ± 357 al. (2016) VA 199.1 ± 16.0a 143.8 ± 9.8a 22.2 ± 2.2 4,464 ± 500 Rats / Skeletal 222.8 ± CLA muscleδ 144.6 ± 0.9a 23.3 ± 1.4 4,321 ± 214 10.7ab

219.7 ± VA+CLA 150 ± 1.8a 28.8 ± 1.2 5,857 ± 536 8.1ab OO 354.6 ± 28.9 247.5 ± 12.1b 26.2 ± 3.1ab 2,058 ± 945b VA 321.7 ± 19.9 102.2 ± 44.4a 20.5 ± 1.2a 281.1 ± 53.7a Rats / VATδ CLA 355.5 ± 21.6 231.4 ± 10.8b 30.0 ± 1.9b 577 ± 181ab VA + CLA 333.4 ± 10.9 67.3 ± 33.6a 25.5 ± 0.5ab 253.8 ± 53.7a Piras et al. SunfO 7.4 ± 0.6a 11.7 ± 0.4a 2.6 ± 0.2 32.9 ± 1.8 Rats / Liver (2015) CLA 16.0 ± 2.4b 15.1 ± 0.8b 2.9 ± 0.2 30.3 ± 1.3 The difference of statistical significance was indicated with different alphabet letters. * indicates the reanalysed articles.

2.8. Summary and discussion 2.8.1. Impact of dietary fatty acids on NAE and MAG levels This review examined levels of NAEs and MAGs in a variety of biological samples upon consumption of isocaloric fat in diets. As previously described, NAEs and MAGs play important roles in the endocannabinoid system, and influence physiological homeostasis via binding to CB and other receptors. Thus, we also extracted results of physiological outcomes, which associated with the physiological roles of NAEs and MAGs (Table 2.8).

PEA and 2-PG: PEA has been known for its anti-inflammatory properties (LoVerme et al. 2005). Also, it binds to PPAR-α (Lo Verme et al. 2005) and may be a potent agonist for G-coupled protein receptor (GPR) 55 (Ryberg et al. 2007). 2-PG is known as a 2-AG congener, which does not orthosterically bind to CBs, but promotes 2-AG activity via inhibiting its degradation (Murataeva et al. 2016). Four clinical studies (Jones et al. 2014, Mennella et al. 2015b, Pu et al. 2016b, Ramsden et al. 2015a) demonstrated that plasma PEA levels were not influenced by the fat content of the diet (Table 2.4), and were not correlated with physiological outcomes or biochemical outcomes (Table 2.8). Interestingly, in two animal studies (Lin et al. 2013a, Kim et al. 2016), fish oil diets increased plasma PEA levels, while an another two studies (Balvers et al. 2013, Wood et al. 2010) showed that fish oil diets had no effect on changing plasma PEA levels (Table 2.5). Also, dietary modulation had no effect on changing animal brain/brain region PEA (Artmann et al. 2008a, Tsuyama et al. 2009, Wood et al. 2010, Zamberletti et al. 2017) or 2-PG (Wood et al. 2010) levels, but 2-PG + 1(3)-OG levels were significantly lower in either n-3 PUFA sufficient diet or high n-3-PUFA diets (Watanabe et al. 2003) (Table 2.6). In the jejunum and liver, PEA levels reflected their precursor palmitic acid (Artmann et al. 2008a), and also PEA can be altered by VA or CLA diets in various peripheral tissues (Jacome-Sosa et al. 2016, Piras et al. 2015) (Table 2.7).

OEA and 2-OG: OEA is known to regulate the food intake, lipid metabolism, energy balance and glucose homeostasis, through PPAR-α (Fu et al. 2003, Cluny et al. 2009) as well as GPR55 (Pertwee 2007, Borrelli & Izzo 2009) and GPR119 (Overton et al. 2006, Borrelli & Izzo 2009). 2-OG is another congener (Tortoriello et al. 2013, De Petrocellis et al. 1999) of 2-AG, which is postulated to be a GPR119 agonist (Hansen et al. 2011) and regulates fat digestion in the

intestine (Yuan et al. 2016). Studies showed that diets rich in n-9 MUFA increased plasma OEA levels (Jones et al. 2014, Pu et al. 2016b, Mennella et al. 2015b, Balvers et al. 2013). Plasma OEA content may also fluctuate in response to dietary omega-3 PUFA, but the results were inconsistent (Kim et al. 2016, Wood et al. 2010, Ramsden et al. 2015a, Lin et al. 2013a, Balvers et al. 2013). Also, plasma 2-OG levels were not different between diets (Ramsden et al. 2015a, Wood et al. 2010). Along with this high MUFA diet and increased plasma OEA levels, and one clinical study (Mennella et al. 2015b) demonstrated higher fullness, satiety, lower energy intake and lower hunger sensation than n-6 PUFA diets, and other clinical studies found decreased body fat percentage (Jones et al. 2014) and android fat mass (Pu et al. 2016b).

Animal studies showed that high MUFA diets increased liver and intestine OEA levels (Lin et al. 2013a, Artmann et al. 2008a), but not brain OEA levels (Artmann et al. 2008a, Zamberletti et al. 2017). Also, the OEA tissue content may fluctuate in response to other fatty acids in diets (Lin et al. 2013a, Wood et al. 2010, Watanabe et al. 2003, Pu et al. 2016b, Tsuyama et al. 2009), but the results were inconsistent. For instance, a high ARA diet increased brain OEA levels more than a palm oil diet, but not in peripheral tissues (Artmann et al. 2008a). Piras et al (2015) showed that a CLA diet increased liver OEA levels, while other studies found CLA diets did not alter liver (Jacome-Sosa et al. 2016) or brain (Tsuyama et al. 2009) OEA levels. Thus, although OEA levels are associated with higher PPAR- activation properties (Artmann et al. 2008a), it is unclear whether the physiological changes after a particular diet is due to higher OEA content.

AEA and 2-AG: An overactive endocannabinoid system is involved with several features of the metabolic syndrome, obesity, and physical pain through its connection to multiple organ systems (Osei-Hyiaman et al. 2005a, Osei-Hyiaman et al. 2005b, Matias et al. 2008, De Petrocellis et al. 2012). These are negative correlations between absolute plasma AEA with android fat mass (Pu et al. 2016b, Diep et al. 2011) and an increase of obesity-related risk factors (Alvheim et al. 2012b, Alvheim et al. 2014a, Alvheim et al. 2013). However, some clinical studies found no significant effect of diets enriched with n-6 PUFA on plasma AEA levels (Banni et al. 2011, Jones et al. 2014, Mennella et al. 2015b) or 2-AG levels (Pintus et al. 2013a, Mennella et al. 2015b, Banni et al. 2011), in comparison with diets enriched with other types of fatty acids. Also, the elevation of ARA, AEA, 2-AG or 1-AG + 2-AG levels was inconsistent across tissues

and blood samples in animals (Balvers et al. 2013, Lin et al. 2013a, Alvheim et al. 2012b, Alvheim et al. 2014a, Watanabe et al. 2003, Artmann et al. 2008a) Thus, the synthesis of AEA from dietary n-6 PUFA (Banni & Di Marzo 2010, Berger et al. 2002) may not be directly related to the modulation of physiological changes.

Alternatively, diets enriched with n-3 PUFA reduce AEA and/or 2-AG levels in blood and tissues and suppress obesity-related risk markers (Alvheim et al. 2012b, Alvheim et al. 2013, Demizieux et al. 2016). However, reductions in AEA and 2-AG levels did not correlate with physical pain or psychological distress (Ramsden et al. 2015a). Animal studies (Zamberletti et al. 2017, Demizieux et al. 2016) suggest that dietary ALA can decrease brain AEA and 2-AG levels and improve mood, cognition and metabolic parameters. Thus, the results indicate that dietary fat intake can resolve AEA and 2-AG levels, but changes in levels are not directly associated with physiological changes (Zamberletti et al. 2017, Banni & Di Marzo 2010). Additionally, clinical and animal studies showed that CLA-enriched diets can reduce AEA and/or 2-AG levels in comparison with high LNA diets (Pintus et al. 2013a, Jacome-Sosa et al. 2016, Tsuyama et al. 2009). However, AEA and 2-AG levels may be independently associated with the effect of CLA and not correlated with physiological changes (Tsuyama et al. 2009).

LEA and 2-LG: LEA is a poor CB receptor ligand. However, it acts via PPAR-α to suppress appetite (Friedman et al. 2016, Ezzili et al. 2010) and has anorectic properties (Hansen 2014). 2- LG is another congener of 2-AG (Murataeva et al. 2016, Yuan et al. 2016) associated with comorbidity in psychiatric disorders (Pedraz et al. 2015). Safflower oil increased human and animal plasma LEA levels (Pu et al. 2016b, Kim et al. 2016), but not 2-LG or its isomer 1-LG (Kim et al. 2016) compared to DHA enriched diets. Also, a safflower oil diet increased animal brain and liver LEA levels, but not in the jejunum (Artmann et al. 2008a). However, plasma and tissue LEA levels were not different between other n-6 PUFA, MUFA or n-3 PUFA enriched oils in the diets (Jones et al. 2014, Mennella et al. 2015b, Zamberletti et al. 2017). Also, studies showed that LEA activated PPAR-α activation (Artmann et al. 2008a) and was negatively associated with fat composition (Pu et al. 2016b, Diep et al. 2011). In contrast, a safflower oil diet increased inflammatory markers and biomarkers of glucose uptake (Kim et al. 2016). Thus, more studies are needed to examine the anti-inflammatory properties of LEA and 2-LG.

DHEA and 2-DHG: DHEA was first identified as an AEA congener (Felder et al. 1993), and is a weak antagonist of CB1 (Bisogno et al. 1999, Sugiura et al. 1999). DHEA is also considered as an endocannabinoid-like compound, called synaptamide due to its synaptogenic properties in the brain (Kim et al. 2011). 2-DHG is another derivative from DHA, which has been studied for its role of antinociceptive, anxiolytic, and neurogenic properties (Ramsden et al. 2015a, Wood et al. 2010). Thus, studies examined if fish oil or other n-3 PUFA source can modulate DHEA and 2- DHG levels, which in turn reduce inflammation or pain. Both clinical (Ramsden et al. 2015a, Pu et al. 2016b) and animal (Balvers et al. 2013, Liisberg et al. 2016, Kim et al. 2016, Wood et al. 2010) studies showed that plasma DHEA levels can be increased upon enriched DHA diets. However, the effect of enriched DHA diets on plasma 2-DHG levels was inconsistent between human (Ramsden et al. 2015a) and animal studies (Wood et al. 2010). Diets enriched with DHA had no effect on rodent’s whole brain DHEA levels(Artmann et al. 2008a, Wood et al. 2010), but increased pig’s whole brain DHEA levels (Berger et al. 2001a). DHA enriched diets increased whole brain 2-DHG (Watanabe et al. 2003), but not in another two studies (Watanabe et al. 2003, Wood et al. 2010). A DHA enriched diet can also increase DHEA and 2-DHG levels in selective brain regions (Berger et al. 2001a, Zamberletti et al. 2017), the jejunum (Artmann et al. 2008a) and liver (Artmann et al. 2008a). Clinical study (Ramsden et al. 2015a) demonstrated that levels of plasma DHEA and 2-DHG were correlated with the improvement of psychological distress, frequency, and severity of headaches in chronic headache patients. Animal studies showed that DHA enriched diets can improve cognitive performance (Zamberletti et al. 2017), reduce body weight (Lin et al. 2013a), fat mass and feed efficiency (Liisberg et al. 2016) and biomarkers of glucose uptake biomarkers (Kim et al. 2016). In general, increasing dietary DHA can increase tissue DHEA and 2-DHG levels, which associates with several outcomes.

Other NAEs and MAGs: Accompanied by the improvements of LC-MS/MS, new NAEs and MAGs are being discovered and identified rapidly (Gouveia-Figueira & Nording 2014, Lin et al. 2012). Unfortunately, the impact of dietary fatty acids on those compounds has not been well documented so far. Some clinical studies reported that diets high in ALA had higher plasma ALEA, which was positively associated with fat oxidation (Pu et al. 2016b, Jones et al. 2014). The effect of n-3-enriched diets on EPEA and 2-EPG is still unclear because they are generally undetectable, except in one animal study (Zamberletti et al. 2017).

Overall, the current data suggest that dietary fatty acids can affect NAE and MAG levels, but changes in those compounds are not consistently associated with physiological parameters. Thus, further studies are needed to examine mechanisms connecting changes in NAE and MAGs upon dietary interventions, with physiological outcomes.

Table 2.8. The association of dietary fat-induced NAEs and MAGs on the change of physiological outcomes

Authors, Experiment diet Change of NAEs and MAGs Physiological outcome Biochemical outcome year Alvheim et HFD: 60% E Brain: HFD: HFD: al. (2012b)* 8% E LNA HFD: ↑ 1-AG + 2-AG in 8% E LNA; 8% E LNA vs 1% E LNA: ↑ food intake 8% E LNA vs 1% E LNA: ↔ leptin 1% E LNA Liver: & body weight & adiposity index; MFD: 1% E EPA/DHA + 8% E HFD: ↑ 1-AG + 2-AG in 8% E LNA & 8% E LNA vs 1% EPA/DHA+8% E 8% E LNA vs 1% E LNA: ↔ leptin & LNA 1% E EPA/DHA + 8% E LNA LNA: adiponectin MFD: 35% E MFD: ↑ 1-AG + 2-AG in 8% E LNA vs. ↑ food intake & adiposity index & feed ↔ insulin; ↔ gene expression 8% E LNA 1% E LNA efficiency (SREBP-1c, FAS, ACC1, AMPKa1, 1% E LNA MFD: AMPKa2, resistin) 1% E EPA/DHA + 8% E 8% E LNA vs 1% E LNA: ↑ feed NA efficiency & body weight & adiposity index; 8% E LNA vs 1% EPA/DHA+8% E LNA:↑ body weight Alvheim et Atlantic Salmon exp: Liver: ↑ AEA & ↔ 1-AG + 2-AG ↔ endpoint body weight, ↔ mean ↑ TG & total lipid in liver al. (2013) SO vs FO (HFD: 51% E) visceral somatic index, ↔ whole-fish proximate composition Mice exp: Brain: ↔ AEA & 1-AG + 2-AG; ↑ weight gain, feed efficiency, adipose ↔ leptin, adiponectin, & insulin SO vs FO (MFD: 35% E) Liver: ↑ 1-AG + 2-AG tissue inflammation; ↔ energy intake, endpoint body weight, adiposity index Alvheim et MFD: 35% E Brain/brain region: ↔ AEA & 2-AG LFD: LFD & MFD: al. (2014a)* 8% E LNA vs 1% E LNA (MFD & LFD); 8% E LNA vs 1% E LNA: ↔ food 8% E LNA vs 1% E LNA: ↑ leptin, ↓ LFD: 12.5% E Liver: ↑ AEA & 2-AG (MFD & LFD) intake, ↑ weight gain & feed efficiency, adiponectin, ↔ insulin 8% E LNA vs 1% E LNA ↔ final body weight, ↔ white adipose tissue, ↔ adiposity index MFD: 8% E LNA vs 1% E LNA: ↔ food intake, ↔ feed efficiency, ↔ final body weight & weight gain, ↔ white adipose tissue, ↔ adiposity index Artmann et HFD: 45% E Brain: ↔ PEA, DHEA, SEA & EPEA; Luciferase transactivation: All NAEs al. (2008a) PO ↑ OEA in ARA vs PO; showed PPAR- α activation, OEA & OO LEA were the most potent activators. 54

SaffO ↑ AEA in OO, ARA & FO vs SaffO & ARA PO; FO ↑ LEA in OO, SaffO vs PO; ↑ 2-AG in ARA vs PO; Liver: ↑ PEA in PO vs SaffO, ARA & FO; ↑ OEA in OO; ↑ AEA in ARA vs SaffO & FO; ↑ DHEA in FO vs SaffO & ARA; ↑ SEA in OO vs ARA & FO; ↑ LEA in SaffO vs ARA & FO; ↑ 2-AG in PO, OO ARA > SaffO > FO Jejunum: ↔ LEA; ↑ PEA in PO vs ARA & FO; ↑ OEA in OO vs FO; ↑ DHEA in FO; ↑ AEA in ARA; ↑ SEA in PO & SaffO vs ARA & FO ↑ 2-AG in ARA Avraham et LFD: 12.4% E: Hypothalamus & hippocampus: ↔ relative body weight, ↔ healthy Hypothalamus: al. (2011) 100% sufficient diet of ↔ AEA & 2-AG appearance, ↔ radial arm maze Survival markers: ↓ serotonin 5-HT, ↑ CAO vs FO acquisition performance, ↔ 100% neuropeptide Y, ↔ Camkk2 survival rates Monoamine neurotransmitter: ↓ dopamine; Endocannabinoid system markers: ↑ CB1 receptor functions; ↑ GTP S binding activities Hippocampus: Survival markers: ↔ serotonin 5-HT, ↓ relative quantity of BDNF & SIRT-1 mRNA; Monoamine neurotransmitter: ↔ norepinephrine; Endocannabinoid system marker: ↑ CB1 receptor functions; ↑ GTP S binding activities

Banni et al. OO Plasma: ↔ AEA & 2-AG ↔ BMI & waist circumference ↔ glycaemia, insulinemia, HOMA-IR, (2011) KO TG, TC, HDL, LDL & CRP(Maki et al. Menhaden oil 2009), ↑ ratio of plasma n-6/n-3 in obese subjects on KO Batetta et al. SO Liver: ↑ 2-AG in KO, ↓ AEA in KO< Plasma: ↔ inflammatory response & Plasma: ↓ LDL, (2009) KO FO< SO; growth & food intake & endpoint body ↑ triglyceridemia in KO & FO; FO Heart: ↓ AEA in KO, ↑ 2-AG in KO vs weight; Liver: ↓ TG in KO > FO > SO; ↓ SO; Liver: ↓ inflammatory response MAGL activity in KO & FO; VAT: ↓ 2-AG in KO, ↓ AEA in KO & Heart: ↓ TG in KO, ↓ MAGL activity FO in KO Macrophage: ↓ TNFα secretion in the presence of LPS in KO & FO; ↓ MAGL activity in KO & FO in VAT, ↔ FAAH activity & HDL Demizieux Lard fat Brain: ↓ AEA & 2-AG in linseed vs Body composition: Liver: et al. (2016) SaffO SaffO ↑ liver weight in lard diet, ↓ total fat Endocannabinoid system markers: ↔ Linseed oil Liver: ↓ AEA in linseed oil, ↑ 2-AG in mass in linseed oil, ↔ body weight CB1; saffO; ↑ FAAH & NAPE in lard fat vs linseed Muscle: ↓ AEA & 2-AG in linseed oil; oil, Protein level: SCAT: ↓ AEA in linseed oil, ↔ 2-AG; ↑ NAPE in lard fat vs SaffO & linseed VAT: ↔ AEA & 2-AG oil; ↑ MAGL in lard fat vs SaffO; ↑ MAGL activity in lard fat vs SaffO & linseed oil, ↔ FAAH activity; Lipid metabolism: ↓ PEPCK & FAS & fatty acid synthase & SCD in linseed oil, ↓ G6P & GCK in linseed oil vs lard fat, ↑ CREBH & TNFα & FAT/CD36 & LPL in lard fat. Plasma: ↔ leptin & glucose & Pvat; ↑ insulin & fasting triglyceridemia & cholesterolemia; ↓ adiponectin in lard fat; iSCAT: lard > linseed oil> SaffO; Glycose control: ↓OGTT & ITT in linseed oil

Jacome- OO Plasma: ↔ AEA, PEA, OEA, 2-AG; ↔ food intake Jejunal mucosa: Sosa et al. VA Skeletal muscle: ↓ PEA in VA vs OO, ↑ mRNA and protein expression: ↑ (2016) CLA OEA in OO, ↔ AEA & 2-AG; FAAH in VA+CLA > OO, ↑ CB1 in VA+CLA Liver: ↑ PEA in OO & CLA vs VA & CLA vs VA; VA+CLA, ↑ 2-AG in OO vs CLA>VA; Pro-inflammatory cytokines: ↓ TNF VAT: ↔ PEA, ↑ OEA in OO & CLA vs and IL-1β in VA vs OO VA & VA+CLA; ↑ AEA in CLA vs VA, ↑ 2-AG in OO vs VA & VA+CLA; Jejunal mucosa: ↑ PEA in VA, ↑ OEA in VA & VA+CLA; ↑ AEA in VA vs OO & CLA, ↔ 2-AG Jones et al. Western diet Plasma: ↔ PEA, AEA, LEA & DHEA; NAE and body composition: (2014) HOCAO ↑ OEA in HOCAO; ↑ OEA, AEA  ↓ fat %; HOCAO/ FlaxO ↑ ALEA in HOCAO/FlaxO ↑ AEA ↓ android fat mass & gynoid fat mass, ↑ android /gynoid fat, ↑ lean mass; ↑AEA/OEA  ↑ ratio of android /gynoid fat; ↑ DHEA/OEA  ↑ fat %, fat mass, & android fat mass NAE and energy balance: ↑ AEA, ALEA, LEA ↓ fat oxidation; ↑ LEA, AEA ↑ carbohydrates oxidation; Kim et al. SaffO DHA diet on Day 62: ↔ body weight, ↔ food intake, Eepididymal fat pad: (2016) DHA Plasma: ↑ PEA, ↔ DGLEA, ↑ SEA, ↑ ↑ lean mass & ↓ epididymal fat pad mass Endocannabinoid system markers: At OEA, ↑ LEA, ↑ AEA, ↑ DTEA, ↓ DHEA; in DHA diet Day 62 & 118: ↓CB1, DAGLα-β in ↑ 1-OG, ↓ 2-OG, ↓1-LG, ↔ 2-LG, ↑ 1- DHA diet; AG, ↑ 2-AG in SaffO vs DHA; At Day 118: ↑ CB2, NAPE-PLD, DHA diet on Day 118: FAAH in DHA diet; Plasma ↔ PEA, ↔ DGLEA, ↑ SEA, ↑ Glucose uptake biomarkers: At day 62 OEA, ↑ LEA, ↑ AEA, ↑ DTEA, ↓ DHEA; & 118: ↓Akt-1, insulin receptor, ↑ 1-OG, ↑ 2-OG, ↔1-LG, ↑ 2-LG, ↑ 1- GLUT4 in DHA diet; At Day 118: ↑ AG, ↑ 2-AG in SaffO vs DHA insulin receptor substrate-1, GLUT1, adiponectin in DHA diet;

Inflammatory markers: At day 62 & 118: ↑ AMP-activated protein kinase in DHA diet; Quadriceps: at Day 62 & 118: Endocannabinoid system markers: ↑ CB1, CB2, NAPE-PLD, FAAH, DAGLα-β in DHA diet; Glucose uptake biomarkers: ↑Akt-1, insulin receptor, insulin receptor substrate-1, GLUT4, GLUT1 in DHA diet; Inflammatory markers: ↓ IL-6, TNF-, P42/P44 & P38 & JNK mitogen activated protein kinase in DHA diet; Lin et al. CO Liver & prox-small intestine: ↔ red blood cell lipogenesis of C16:0 & ↑ hepatic CD36 expression in HOCAO (2013a) HOCAO ↔ PEA, ↑ OEA in HOCAO & HOCAO- C18:1; ↔food intake; HOCAO-DHA DHA vs CO & FO; ↓ AEA in FO; ↑ hepatic lipogenesis of C16:0 & C18:1 FO Plasma: ↑ PEA and OEA in FO; ↓ AEA in CO vs HOCAO-DHA; in FO ↑ fat oxidation in HOCAO & HOCAO- DHA vs FO, CO; ↓ body weight in FO vs CO & HOCAO- DHA; ↓ lean mass in FO; ↑ hepatic-lipogenesis of C18:1↓intestinal OEA & ↑ plasma AEA; ↑ fat oxidation  ↑ intestinal OEA Liisberg et Pork diet Plasma: ↓ DHEA, ↑ AEA, ↑ 2-AG in ↑ body mass gain & fat mass & feed al. (2016) Cod diet pork diet efficiency & liver mass & liver fat in pork diet Mennella et SunFO Plasma: ↔ PEA, LEA, 2-AG; ↑ follow up lunch energy intake & the al. (2015b) HOsunfO ↑ OEA in HOSunfO & virgin OO accumulation of hunger sensation; ↓ virgin OO fullness & satiety in SunFO vs virgin OO Pintus et al. Control cheese Plasma: ↑ AEA in control cheese, ↔ 2- ↑ AEA  ↑ leptin, between diets; (2013a) CLA-rich cheese @ AG ↔ adipokines; ↔ IL-6; ↔ C-reactive 90g/day protein

Piras et al. MFD: 31% E Liver: ↑ OEA & PEA; ↔ AEA & 2-AG ↓ lipid deposition in liver, ↓ food intake, (2015) 1% CLA; SunFO in CLA diet ↓ body weight, ↓ liver weight in 1% CLA diet Pu et al. CAO Plasma: ↔ PEA; ↓ android fat mass ↑ OEA, ↑ LEA, ↑ (2016b) HOCAO ↑ OEA in HOCAO vs HOCAO+DHA & ratio of OEA/DHEA HOCAO/DHA CAO > FlaxO/saffO & CO/ SaffO; ↑ DHEA in HOCAO/DHA diet if C- FlaxO/saffO ↑ AEA in HOCAO > HOCAO/DHA & allele carries of NAPE-PLD CO/SaffO FlaxO/saffO & CO/saffO; polymorphism rs12540583 & A-allele ↑ DHEA in HOCAO/DHA; carriers of FAAH polymorphism ↑ALEA in FlaxO/saffO > rs324420 HOCAO/DHA > CAO & CO/saffO ≥ HOCAO > CAO; ↑ LEA in CO/saffO ≥ FlaxO/saffO ≥ CAO & HOCAO > HOCAO/DHA Ramsden et High n-3 PUFA Plasma: Correlation tests: al. (2015a) Low n-6 PUFA + high n- ↑ 2-DHG & DHEA, ↓ 2-AG, ↑ 2-DHG  ↓ the number of headache 3PUFA ↔ AEA & PEA & OEA in high n-3 days /month; ↓ psychological distress PUFA ↑ DHEA  ↓ number of headache days / month, ↓ the number of severe headache hours / day; ↓ AEA  ↓ the number of severe headache hours / day; ↓2-AG  ↓ the number of severe headache hours / day Zamberletti Control (3% peanut oil, Prefrontal cortex: ↔ body weight & food intake Both n-3 PUFA deficient and enriched et al. (2017) 3% rapeseed oil) ↓ AEA, 2-AG in n-3 PUFA enriched Both n-3 PUFA deficient and enriched diets: n-3 PUFA deficient ↓ 2-DTG in both n-3 PUFA diets vs diets  ↓ cognitive performance (NOR ↑ FAAH & MAGL activity, ↓ CB1 n-3 PUFA enriched control test), ↔ emotional memory test functionality, ↓ NAPE-PLD ↓ 2-AG, 2-DHG in n-3 PUFA deficient (inhibitory avoidance task); ↔ PEA, OEA, LEA, DHEA, EPEA; n-3 PUFA enriched diet: ↓social Hippocampus: ↑DHEA in n-3 PUFA behaviors test & climbing test, ↑ enriched diet; ↑2-AG in n-3 PUFA immobility test, ↔ swimming test & pre- deficient vs n-3 enriched diet; ↓ 2-DHG in pulse inhibition test n-3 deficient vs control; ↔AEA, PEA, OEA, LEA, EPEA

2.8.2. Other factors may affect on NAE and MAG levels in biological samples The data from this review reveal that although dietary modulation can significantly elevate certain NAEs and MAGs, the changes in most NAE and MAG levels were generally small across the range of dietary fat intakes assessed. Within each clinical trial, variation of OEA, DHEA, and ALEA in human plasma across dietary fats is small (Pu et al. 2016b, Ramsden et al. 2015a, Jones et al. 2014). Equivalently, plasma NAEs levels were similar across clinical studies (Table 2.4), which covered the same range of concentrations as published baseline levels (Amoako et al. 2010, Schreiber et al. 2007). With one exception (Pintus et al. 2013a), the human plasma AEA levels ranged within 1-8 pmol/ml (Mennella et al. 2015b, Banni et al. 2011, Ramsden et al. 2015a, Pu et al. 2016b, Pintus et al. 2013a). In contrast to the NAEs, plasma 2-AG levels ranged dramatically from 13-2,398 pmol/ml (Banni et al. 2011, Mennella et al. 2015b, Pintus et al. 2013a, Ramsden et al. 2015a), which may be due to the variation in their baseline levels (Supplemental Table 2.1). The baseline variation could be related to subject health conditions, such as obesity, ethnicity, neurological disorders and pregnancy (Zoerner et al. 2011, Banni et al. 2011, Zoerner et al. 2009, Jumpertz et al. 2011, Habayeb et al. 2004, De Marchi et al. 2003, Ramsden et al. 2015a), Further investigations should be considered.

As noted previously, within each animal study, plasma NAE levels were comparable across diets (Balvers et al. 2013, Lin et al. 2013a). As in the human studies, a greater range of plasma 2-AG levels was observed in animals (Zoerner et al. 2011, Wood et al. 2010). Additionally, NAE and MAG levels were generally higher in tissue samples than in blood samples (Table 5, 6, 7). Levels of NAE and MAG, particularly AEA and 2-AG, differ across animal species and types of tissue (Alvheim et al. 2013, Artmann et al. 2008a, Batetta et al. 2009, Piras et al. 2015), which may be related to the dietary fat modulation, types of euthanasia, storage, sleep pattern, and total fat content in the diet (Bazinet et al. 2005, Zoerner et al. 2011, Brose et al. 2016, Hansen 2013, Diep et al. 2011, Vaughn et al. 2010).

2.9. Strengths and limitations A significant strength of this review was to give an overall spectrum of the levels of NAEs upon consumption of isocaloric fat diets in human and animal studies. All of the six human studies are randomized clinical trials, which minimizes bias related to confounding factors. Also, this review

examined levels of NAEs in healthy subjects with one or more elevated metabolic biomarkers, and chronic headache patients, which helps future studies to examine healthy vs. disease conditions.

This review also included animal studies, which covered a variety of species. Also, the levels of NAE are reported in the blood, brain and various organs. The strength of animal studies is that they offer a level of control not possible in humans. Animal studies provide the proof-of-concept and help guide future human studies.

Furthermore, this review extracted concentrations of NAEs and MAGs in each study using the same units, which allowed us to identify the range of these compounds and account for differences between species, sample types and to restate the effect of isocaloric fat contents. In addition, this review demonstrates the effects of individual types of fatty acids on NAE and MAG levels. These results show that the change of NAEs and MAGs levels was not limited to their precursor fatty acids, but also to other fatty acids. Also, the effect of diet on levels of NAEs and MAGs was tissue selective. Although these studies were exploratory, as many mechanistic questions remain to be answered, they provided the insight on the relationship between fatty acids and NAEs and MAGs.

There are some limitations in this study. First, clinical trials often did not consider genetic variables, which may influence NAE levels. For example, one clinical study (Wangensteen et al. 2011) demonstrated that a single SNP, rs17605251, in the NAPE-PLD gene was associated with a BMI > 35. Also, Pu et al. (2016b) indicated that C-allele carriers with NAPE-PLD polymorphism rs12540583 had higher DHEA levels after a canola-DHA diet. Also, a polymorphism, rs324420, in FAAH, specifically the A-allele carriers, lead to higher DHEA levels than the CC genotype carriers. Therefore, future studies should consider these differences in genetics. Second, the levels of NAEs in different species are difficult to interpret because it is unclear if they are real species effects or due to methodological differences between labs. Third, changes of NAE and MAG levels were indicative of diet effects, but they do not inform us about functional changes or mechanisms Second, study durations, designs and species were not consistent so direct comparisons across studies should be undertaken with caution. Fourth,

although each study contained an isocaloric fat content, total fat content varied between studies. The use of different high-fat content may have masked the individual effects of fatty acids on levels of NAEs and MAGs.

2.10 Conclusion To conclude, levels of NAE and MAG in blood, brain, and other organs can be altered by their corresponding precursor fatty acids, and also respond to other fatty acids upon either acute or long-term feeding. In addition, the magnitude of change of these compounds was quantitatively small within each study. Moreover, other factors, such as baseline characteristics, animal species, and sample preparation may vary NAE and MAG levels largely from one study to another suggesting a complex mechanism for regulating NAE and MAG levels. Moreover, although studies examined the association between dietary-induced changes in NAEs and MAGs with physiological outcome improvements, future studies will need to examine the mechanism of action in more detail.

Supplemental Table 2.1. Baseline NAEs and MAGs concentration in clinical studies Author/ NAEs (pmol/ml) MAGs (pmol/ml) Diets Subjects year PEA OEA LEA AEA DHEA 2-OG 2-AG 2-DHG normal weight 5.3 ± 0.6 64.9 ± 8.4 None Overweight 6.5 ± 0.5 111.3 ± 16.8 Obese 7.1 ± 0.4 95.0 ± 9.2 KO 4.1 ± 0.6 72.2 ± 15.1 Menhaden normal weight 7.8 ± 1.4 56.1 ± 9.7 oil OO 5.0 ± 1.1 62.4 ± 10.3 Banni et al. KO 6.3 ± 1.4 145.4 ± 42 (2011) Menhaden Overweight 7.7 ± 0.5 102.8 ± 28.9 oil OO 6.0 ± 0.5 95.9 ± 21.3 KO 7.2 ± 0.8 98.8 ± 16.3 Menhaden Obese 7.7 ± 0.4 105.5 ± 18.9 oil OO 6.3 ± 0.7 75.4 ± 5.3 SunFO 15.9 ± 0.4 7.6 ± 0.1 4.3 ± 0.1 3.2 ± 0.1 13.2 ± 0.3 Mennella et HOSunFO normal weight 15.9 ± 0.4 7.6 ± 0.1 4.3 ± 0.1 3.2 ± 0.1 13.2 ± 0.3 al. (2015b) Virgin OO 15.9 ± 0.4 7.6 ± 0.1 4.3 ± 0.1 3.2 ± 0.1 13.2 ± 0.3 Pintus et al. Hyper- None 99.6 ± 9.5 48.2 ± 7.8 (2013a) cholesterolaemia 355.5 ± 87.0 Low n-6 11.2 ± 1.9 8.7 ± 2.1 1.4 ± 0.3 1.2 ± 0.4 5,816.7 ± 1,316.0 1,682.5 ± 578.4 Ramsden et chronic headache al. (2015a) High n-3 + patients 10.2 ± 2.0 8.4 ± 1.7 1.4 ± 0.1 1.2 ± 0.5 6,489.7 ± 1,775.6 2,398.3 ± 422.6 449.9 ± 151.6 Low n-6

CHAPTER 3 Rationale, Hypotheses and Objectives

3.1 Rational and research gap Internal factors, such as NAPE-PLD and FAAH are involved in NAE metabolic pathways. Thus NAE synthesis and degradation can be interrupted by knocking out these enzymes, which are thought to play important roles in NAE-mediated effects on lipid metabolism. Also, it is known that external factors, such as diets with different fatty acid profiles impact NAEs. However, these metabolic effects without NAPE-PLD or FAAH are not characterized. In addition, another external factor, ischemia, can also affect NAE levels, which may influence the conversion of fatty acids to NAEs in the tissues, but have not been evaluated. This thesis aims to understand how internal factors and external factors interact with each other on modulating NAE levels (Figure 3.1).

Figure 3.1. Rational and research gap

3.2 Overall Hypothesis

If NAPE-PLD and FAAH are the primary enzymes responsible for the synthesis and degradation of NAEs, respectively, dietary fatty acids and ischemia will not be able to further elevate NAE concentrations in the absence of these enzymes.

3.3 Specific Hypothesis Hypothesis 1: NAPE-PLD is necessary for increases in NAEs upon dietary fatty acid manipulation.

Hypothesis 2: FAAH is necessary for CO2-induced hypercapnia/ischemia increases in NAEs.

3.4 Overall Objective To examine whether levels of NAEs, and their precursor fatty acids are influenced by the absence of NAPE-PLD or FAAH upon dietary manipulation or ischemia, respectively.

3.5 Specific Objectives Objective 1: To determine if NAPE-PLD is necessary for increases in NAEs upon dietary fatty acid manipulation.

Objective 2: To determine if FAAH is necessary for CO2-induced hypercapnia/ischemia increases in NAEs.

CHAPTER 4. (Study 1) Dietary fatty acids augment tissue levels of n-acylethanolamines in n-acylphosphatidylethanolamine phospholipase D (NAPE-PLD) knockout mice

Adapted from: Lin Lin, Adam H Metherel, Alex P Kitson, Shoug M Alashmali, Kathryn E Hopperton, Marc-Olivier Trépanier, Peter J Jones and Richard P Bazinet. “Dietary fatty acids augment tissue levels of n-acylethanolamines in n-acylphosphatidylethanolamine phospholipase D (NAPE-PLD) knockout mice.” J Nntr Biochem submitted 2017

4.1 Abstract

N-acylethanolamines (NAEs) are lipid signaling mediators which can be synthesized from dietary fatty acids via N-acylphosphatidylethanolamine-phospholipase D (NAPE-PLD). Changes in dietary fatty acids can alter NAE concentrations and in turn impact physiological outcomes; however, the mechanistic details are not fully understood. Presently, we assess the role of NAPE-PLD in the biosynthesis of NAEs upon various dietary conditions. Post-weaning male wild-type (C57Bl/6), heterozygous-NAPE-PLD (-/+) and homozygous-NAPE-PLD (-/-) mice received isocaloric fat diets containing either beef tallow, corn oil, canola oil or fish oil (10% wt/wt from fat) for 9 weeks. Brain docosahexaenoic acid (DHA) levels were higher (p < 0.01) in NAPE-PLD (-/+) (10.01 ± 0.31 μmol/g) and NAPE-PLD (-/-) mice (10.89 ± 0.61 μmol/g) compared to wild-type mice (7.72 ± 0.61 μmol/g) consuming fish oil. Brain docosahexaenoylethanolamide (DHEA) levels of wild-type and NAPE-PLD (-/-) mice were 1.6 - 2-fold higher (p < 0.01) on the fish oil diet compared to all other diets. Liver and jejunum arachidonoylethanolamide, 1,2- arachidonoylglycerol levels were highest (p < 0.05) in the corn oil diet. Genotype effect showed that NAPE-PLD (-/-) mice had lower oleoylethanolamide levels in plasma and jejunum, lower (p < 0.0001) food intake, body weight, and fat composition than the wild-type mice. These results demonstrate the complex regulation underpinning the conversion of fatty acids to NAEs, particularly brain DHA to DHEA. Furthermore, while NAPE- PLD may not be necessary for NAE biosynthesis, it may be a novel target for increasing brain DHA and DHEA.

4.2 Introduction N-acylethanolamines (NAE) are a class of naturally occurring bioactive lipid-signaling molecules derived from fatty acid precursors. Changes in the dietary fatty acid composition can alter NAE concentrations in blood and tissues, which in turn impact physiological outcomes (Lin et al. 2013a, Mennella et al. 2015a, Pu et al. 2016b). The in vivo biosynthesis of NAEs from their respective FA is believed to occur through multiple parallel steps. The sn-1 position of phosphatidylcholine and the amino group of phosphatidylethanolamine are catalyzed via either calcium-dependent n-acyl-transacylase or phospholipase A1/acyltransferase (Hansen et al. 1999, Leung et al. 2006, Cadas et al. 1997, Di Marzo et al. 1994, Pacher et al. 2006, Thabuis et al.

2008) to synthesize n-acylphosphatidylethanolamine (NAPE), an anionic membrane phospholipid (Coulon & Bure 2015). NAPE can then be further metabolized by a NAPE- phospholipase D (NAPE-PLD) to form NAEs (LoVerme et al. 2005, Okamoto et al. 2007, Fu et al. 2007). Additionally, other NAE synthesis pathways have been proposed involving phospholipase A1/A2 (PLA1/2), αβ-hydrolase 4 (ABH4) and glycerophosphodiester phosphodiesterase 1 (GDE1) (Ueda et al. 2013, Sun et al. 2004, Wang & Ueda 2009).

Mice lacking NAPE-PLD illustrate that NAPE-PLD is important for the synthesis of NAEs in the brain, particularly for the longer chain (> C20) saturated NAEs (Leung et al. 2006). However, basal concentrations of brain NAEs in these NAPE-PLD deficient mice are unknown, and it is unclear whether dietary fatty acid modulation could counteract the lower synthesis of NAEs in the NAPE-PLD deficient mice. Furthermore, individual NAEs have different properties; for instance, the omega-9 fatty acid derived NAE, oleoylethanolamide (OEA) binds to peroxisome proliferator-activated receptor-alpha (PPAR-α) (Fu et al. 2003), targeted to G- coupled protein receptor (GPR) 55 (Pertwee 2007, Borrelli & Izzo 2009), GPR119 (Overton et al. 2006), and is associated with decreased food intake and body weight (Piomelli 2013, Leweke et al. 2015, Wilson-Perez et al. 2012). Conversely, the omega-6 polyunsaturated fatty acid (PUFA) derived NAE, arachidonoylethanolamide (AEA) binds to cannabinoid receptors, but not GPR55 (Brown & Hiley 2009), and is associated with increased food intake and appetite (Karlsson et al. 2015, Lima et al. 2014, Alvheim et al. 2014a, Engeli 2012, Aguirre et al. 2015). Moreover, the omega-3 PUFA derived NAE, docosahexaenoylethanolamide (DHEA), is suggested to weakly bind to cannabinoid receptors to exert anti-inflammatory effects (Meijerink et al. 2013, Ramsden et al. 2015a). DHEA is also considered to be an endocannabinoid-like compound and is often referred to as synaptamide due to its synaptogenic properties in the brain (Kim et al. 2011, Kim & Spector 2013), where it acts via GPR110 (Lee et al. 2016). In the absence of NAPE-PLD, it is not well known how individual NAE concentrations are altered, especially by diet, or how the changes may affect physiological outcomes including lipid metabolism, energy balance, body weight and body composition. In addition, monoacylglycerols such as 1- and 2-arachidonoylglycerol (1-AG, 2-AG) can be derived from phosphatidylcholine and phosphatidylinositol via different pathways involving enzymes phospholipase Cβ,

diacylglycerol lipase, phospholipase Cß or PLA1 (Ueda et al. 2013, Tsutsumi et al. 1994) , however, the manner by which NAPE-PLD affects 1- and 2-AG’s metabolism is unknown.

Therefore, the aim of the current study was to assess whether dietary fatty acid modulation can counteract reduced levels of NAEs in NAPE-PLD deficiency, and to determine the role of NAPE-PLD in the modulation of lipid metabolism, energy balance and body composition upon feeding isocaloric fat diets. Wild-type (C57BL/6), heterozygous (NAPE-PLD (-/+)), and homozygous (NAPE-PLD (-/-)) knockout mice were fed diets enriched with either saturated fatty acids, omega-3 PUFA, omega-6 PUFA or omega-9 monounsaturated fatty acids (MUFA). We hypothesized that NAPE-PLD (-/-) mice would generally have lower NAE concentrations, which in turn would affect the lipid metabolism, food intake, and body composition compared to the wild-type mice.

4.3 Methods All procedures were performed in accordance with the policies set out by the Canadian Council on Animal Care and were approved (Protocol Number: 20010346) by the Animal Ethics Committee at the University of Toronto, Ontario, Canada (Canadian Council on Animal Care 1993).

4.3.1 Diets Diets were modified based on the AIN-93G formulation (Reeves et al. 1993) and prepared by Dyets Inc. (Bethlehem, PA, USA). The diet by weight contained 10% fat (23% energy), 20.3% protein (18.6% energy), 59.9 % carbohydrate (56.7% energy), 5% cellulose, 3.5% t- butylhydroquinone mineral mix, 1% vitamin mix, and 0.25% choline bitartrate (Table 4.1). The only difference between the four diets was their fat source:1) beef tallow (9.8% beef tallow and 0.2% safflower oil, wt/wt), 2) canola oil (2.94% beef tallow, 0.06% safflower oil and 7% canola oil), 3) corn oil (2.94% beef tallow, 0.06% safflower oil, 7% corn oil), and 4) fish oil (2.94% beef tallow, 0.06% safflower oil and 7% menhaden oil). Dietary fatty acid composition was analyzed using gas chromatography-flame ionization detection (GC-FID). The detailed fatty acid composition of these diets is listed in Table 4.2. Briefly, beef tallow diet was enriched with saturated fatty acids (C16:0) and an monounsaturated fatty acid (C18:1 n-9). The canola oil diet

was enriched with monounsaturated fatty acids (C18:1 n-9) and n-3 PUFA (C18:3 n-3). The corn oil diet was enriched with n-6 PUFA (C18:2 n-6) while the fish oil diet was enriched with saturated fatty acids (C16:0) and n-3 PUFA (C20:5 n-3, C22:6 n-3).

Table 4.1. Macronutrient composition of the diets.

Diets (g/kg) Beef tallow Canola oil Corn oil Fish oil Casein 200 200 200 200 L-Cystine 3 3 3 3 Sucrose 100 100 100 100 Cornstarch 367.48 367.48 367.48 367.48 Dextrose 132 132 132 132 t-Butylhydroquinone 0.02 0.02 0.02 0.02 Cellulose 50 50 50 50 Mineral Mix # 210025 35 35 35 35 Vitamin Mix # 310025 10 10 10 10 Choline bitartrate 2.5 2.5 2.5 2.5 Beef tallow 98 29.4 29.4 29.4 Safflower oil 2 0.6 0.6 0.6 Canola oil 0 70 0 0 Corn oil 0 0 70 0 Menhaden oil 0 0 0 70 Diets were based on AIN-93G with total 10% fat by weight

Table 4.2. Dietary fatty acid composition.

Diet (FA % wt) Beef tallow diet Canola oil diet Corn oil diet Fish oil diet 12:0 0.21 ± 0.01 0.15 ± 0.00 0.13 ± 0.01 0.31± 0.02 14:0 3.79 ± 0.11 1.35 ± 0.02 1.27 ± 0.03 10.0 ± 0.25 14:1n-7 0.91 ± 0.03 0.30 ± 0.01 0.28 ± 0.01 0.44 ± 0.01 16:0 27.53 ± 0.19 12.19 ± 0.04 17.95 ± 0.01 25.03 ± 0.22 16:1n-7 3.56 ± 0.04 1.21 ± 0.19 1.16 ± 0.08 12.14 ± 0.12 18:0 16.83 ± 0.44 6.91 ± 0.14 6.53 ± 0.02 7.96 ± 0.15 18:1n-9 40.58 ± 0.67 52.85 ± 0.31 31.34 ± 0.07 20.00 ±0.14 18:1n-7 1.12 ± 0.62 2.18 ± 0.03 0.85 ± 0.03 2.92 ± 0.06 18:2n-6 4.49 ± 0.11 15.30 ± 0.08 39.07 ± 0.04 3.57 ± 0.01 18:3n-6 0.06 ± 0.02 0.32 ± 0.03 0.02 ± 0.01 0.27 ± 0.00 18:3n-3 0.37 ± 0.01 5.87 ± 0.03 0.87 ± 0.00 1.26 ± 0.00 20:0 0.23 ± 0.02 0.43 ± 0.00 0.26 ± 0.08 0.18 ± 0.15 20:1n-9 0.30 ± 0.01 0.82 ± 0.01 0.27 ± 0.01 1.15 ± 0.04 20:2n-6 0.01 ± 0.02 0.05 ± 0.05 ND 0.23 ± 0.02 20:4n-6 0.02 ± 0.02 ND ND 0.71 ± 0.01 20:5n-3 ND ND ND 8.55 ± 0.08 22:1n-9 ND 0.08 ± 0.07 ND 0.27 ± 0.23 22:5n-3 ND ND ND 1.04 ± 0.01 22:6n-3 ND ND ND 3.95 ± 0.03 ∑SFA 48.59 21.02 26.14 43.48 ∑MUFA 46.46 57.44 33.9 36.92 ∑PUFA 4.95 21.54 39.96 19.6 n6/N3 12.42 2.67 44.85 0.32 All values are expressed as means ± SEM. n6/n3, n-6 PUFA to n-3 PUFA ratio; Fatty acid composition in each diet was analyzed by Gas Chromatography-Flame ionization detection

4.3.2 Animals NAPE-PLD (-/-) and heterozygous (NAPE-PLD -/+) mice were kindly donated by Dr. Benjamin Cravatt (The Scripps Research Institute, California). C57BL/6 wild-type mice were obtained from Charles River Laboratories (Saint-Constant, QC, Canada). Animals were acclimatized for one week in an animal facility, in which temperature (21°C), humidity and light cycle (12 hours at light / dark circle) were controlled and food and water were provided ad libitum. One male and one female with the same genotype were fed regular chow diet and paired for breeding. During pregnancy and lactation, dams were fed the regular chow diet and at 21 days of age, the male pups were singly housed and immediately weaned onto one of the four treatment diets until 12 weeks of age. The treatment order was based on a computer-arranged random sequence for the cage number and diets prior to the weaning. The food was removed 12 hours before euthanasia. Animals’ weight and food intakes were recorded twice per week throughout the study (approximately at the time between 12:00 - 2:00 pm) on a scale with 0.1g accuracy. Sample size was based on previous studies (Lin et al. 2017, Leung et al. 2006, Lin et al. 2013a), however, because the exact experimental procedures had never been performed, we did not perform a formal power calculation.

4.3.3 The percentage of fat oxidation A MM100-metabolic monitor system (CWE Inc., Ardmore, OK, USA) was used to measure energy expenditure (Lin et al. 2013a). Eight-week-old mice were weighed and randomly placed into metabolic monitor chambers containing enough bedding and water but no food. The animals were acclimatized in the chambers for 6 hours (~5 fasting hours) for the rhythms of VO2, VCO2, heat production, and RER (respiratory exchange ratio) to be acquired. Calculations for indirect calorimetry were based on the formula of Lusk et al. (Lusk 1924) and has been described previously (Harding et al. 2010, Lin et al. 2013a). The percentage of fat oxidation was calculated based on the formula: Fat (%) = 100 (1.00 – RER) / 0.293. The animal body weight was manually uploaded to the machine prior to measurements to allow for automatic adjustment of the flow rate for individual animal. Furthermore, the RER varied from 1.00 (pure carbohydrate) to 0.707 (pure fat), therefore, we repeatedly measured 6 min of each fasting hour and calculated the average RER in each hour after removing the off-ranged values.

4.3.4 Body Fat composition Body fat composition was measured using magnetic resonance imaging (MRI). MR imaging was conducted using a 7 Tesla preclinical system (Biospec 70/30 USR, Bruker Corporation, Ettlingen, Denmark), equipped with a B-GA12 gradient coil insert and 7.2 cm inner diameter linearly – polarized cylindrical RF coil. Mice were anesthetized and maintained at 1.8% isoflurane in a prone position on a slider bed. Respiratory motion was monitored using the SA Instruments system (Stony Brook, NY). Fat was resolved as a hyper-intense signal using a 2D respiratory-gated T2-weighted RARE technique (Rapid Acquisition Relaxation Enhancement) providing whole-body coverage as a stack of slices in the coronal plane (echo time 7 ms; effective echo time 28 ms; RARE factor 8; repetition time governed by respiratory cycle duration; 250 x 250 micron in-plane resolution, 90 x 40 mm field-of-view, 360x160 matrix, at least 20 slices, 1 mm slice thickness, 8 averages). The acquisition time was 5 min 20 sec at a respiratory rate of 30 breaths per minute. Each animal had 20-30 sliced MRI imaging, which provided body composition data and were analyzed using MIPAV software v7.0.1 (National Institutes of Health) to segment out the fat volume (Supplemental Figure 4.1). The fat volume in each slice was recorded and the total animal fat volume was used for analysis.

4.3.5 Euthanasia and sample collection At the end of the experiment (12 weeks of age, 9 weeks of feeding), mice were fasted for 12 hours. Ten min prior to euthanasia, blood was drawn from the saphenous vein and collected into microtubes, which did not contain clot activators. The plasma and red blood cells were immediately separated, placed on ice and stored at –80 °C for further measurements. Animals were then euthanized by head-focused, high-energy microwave irradiation (Control; 1 kW, 0.88- 0.99 seconds; Cober Electronics Inc., Stratford, CT, USA) to maintain lipid integrity, and whole brain, liver and small intestine were removed, excised, and rinsed in chilled saline. The small intestine was further divided into duodenum, jejunum and ileum sections and stored separately. All tissue samples were stored at -80 °C. At a later date, the frozen tissues were pulverized under liquid nitrogen and dry ice, collected in storage tubes and restored at –80 °C for subsequent analyses. In addition, blood and pulverized tissue samples were aliquoted for future analysis. Due to limited sample amounts, fatty acid compositions were measured in the duodenum and NAE and 1-and 2-AG compositions were measured in the jejunum.

4.3.6 Tissue lipid extraction and gas chromatography-mass spectrometry Total lipid extracts (TLE) were obtained from frozen pulverized whole brain, liver and duodenum according to a method modified from Folch (Folch et al. 1957). Briefly, samples were thawed and a known amount of heptadecanoic acid (17:0, Nu-Chek Prep, Inc., Elysian, MN,

USA) was added to the tissues as an internal standard. Lipids were then extracted from the tissues with 6 ml of chloroform: methanol (2:1 by volume), and 1.75 ml of potassium chloride (0.88%) was added to separate the aqueous and organic phases. The mixtures were centrifuged at 500g for 10 min. The lower lipid-containing chloroform layer was then pipetted into new test tubes and stored at -80 oC overnight before transesterifing to fatty acid methyl esters (FAMEs).

After drying down the chloroform under nitrogen, TLE were transesterified to FAMEs by adding 1 mL 14% boron trifluoride in methanol with 0.3 mL hexane and heated at 100°C for 1 hour. The FAMEs were collected from the upper hexane layer and stored in GC vials until analysis by GC-FID. FAMEs were analyzed on a Varian 430 gas chromatograph (Bruker, Billerica, MA, USA) equipped with a DB-23 30m × 0.25 mm i.d. × 0.25 μm film thickness, (50%- Cyanopropyl)-methylpolysiloxane, capillary column (J&W Scientific from Agilent Technologies, Mississauga, ON) with helium as the carrier gas. Samples (1 µl) were introduced by a Varian CP-8400 autosampler into the injector heated to 250°C in splitless injection mode. Initial temperature was 50°C with a 2 min hold followed by a 20°C/ min ramp to 170°C with a 1 min hold, a 3°C/min ramp to 212°C with a 9 min hold for a total run time of 32 min (Domenichiello et al. 2017, Lin et al. 2017). The flame ionization detector temperature was 300°C with air and helium make-up gas flow rates of 300 and 29 ml/min, respectively, and a sampling frequency of 20 Hz. Peaks were identified by retention times through comparison to an external mixed standard sample (GLC-674, Nu Chek Prep Inc., Elysian, MN, USA).

4.3.7 Extractions of NAEs and arachidonoylglycerols

Stock solutions of AEA, OEA, DHEA, 1-AG and 2-AG as well as AEA-D8, OEA-D2, DHEA-D4, o 1-AG-D5 were reconstituted in acetonitrile and stored at -80 C before further dilution, which was performed in acetonitrile on the ice. Calibration curves were prepared using a concentration covering a range of 0.1-50 ng/ml for plasma and 0.1-200.0 ng/ml for tissues. Each point of the calibration curve was constructed by adding a constant amount of cocktail of internal standards

(50ng/ml of OEA-D2, AEA-D8, DHEA-D4 and 250ng/ml of 1-AG-D5) for quantification purposes.

Solid-phase extraction was employed for plasma samples, while liquid-phase extraction was employed for powdered tissue samples according to our previously published method (Lin et al. 2012) with modifications. Both plasma and powdered tissue samples were randomly selected and mixed with the same cocktail of internal standards used for the calibration curve.

Plasma mixed with the internal standards was filled with distilled water to make a total 1ml volume. Then the mixture was homogenized and centrifuged at 490 g for 10 min at 4 0C. The mixture was then extracted using a vacuum manifold with the Oasis HLB 1CC, 30 mg cartridge

(Waters, Canada). The eluted samples were dried under N2 gas and then dissolved in 100ul of acetonitrile. Pulverized liver, jejunum and whole brain were prepared with ice-cold acetone (2ml); homogenized and centrifuged at 490 g for 10 min at 4 0C. The supernatant was transferred to a clean tube and dried under N2 gas. Chloroform: methanol: deionized water (2:1:1 by volume) was then added, vortexed and centrifuged at 490 g for 10 min at 4 oC. The chloroform layer was transferred into a clean tube, dried under N2 gas and dissolved in 100ul of acetonitrile. All samples were stored in GC vials at -80 oC for LC-MS/MS analysis.

4.3.8 Identification and separation using high-performance liquid chromatography-mass spectrometry MS/MS was performed with a SCIEX QTrap5500 mass spectrometer (SCIEX: Framingham, Massachusetts, USA) with an Agilent 1290 HPLC system (Agilent Technologies: Santa Clara, California, USA) equipped with a Phenomenex Kinetex XB-C18 column, 50 x 4.6 mm, 2.6 µm (Phenomenex, Torrance, CA, USA) at a flow rate of 600 µL/min. The mobile phase consisted of 30% A (water + 0.1% formic acid) starting and ramping up to 75% B (acetonitrile + 0.1% formic acid) for a total analysis time of 6.5 min. Mass spectrometry was performed in positive ESI MRM mode using a source temperature of 600°C and an ion spray voltage of 5200 V. The ion mass transitions used were as follows: OEA (m/z 326.3 / 62.0, OEA-d2 (m/z 328.3 / 62.0), AEA and O-AEA (m/z 348.3 / 62.0), AEA-d8 (m/z 356.3 / 62.0), DHEA (m/z 372.3 / 62.0), DHEA-d4 (m/z 376.3/ 66.0), 1-AG and 2-AG (m/z 379.3 / 287.3), 1-AG-d5 (m/z 384.3/ 287.3). Peak

integration and data analysis were performed using Analyst software (Sciex, Framingham, Massachusetts, USA) (Lin et al. 2017).

4.4. Statistics The statistical analysis of food intake, body weight and percentage of fat oxidation were assessed using SAS (PROX MIXED coding) for three-way repeated measures ANOVA. Significant interactions were further analyzed using one-way ANOVAs with Tukey’s multiple comparisons post-hoc test. Also, the main diet effect and/or main genotype effect were also examined. Fatty acid, NAE, and AG concentrations and fat composition were assessed using GraphPad Prism 7 with a two-way ANOVA, with significant interactions further analyzed by one-way ANOVA and Tukey’s multiple comparisons post-hoc test. All results were expressed as means ± SEM. Statistical significance was set at p < 0.05 for all analyses.

4.5 Results 4.5.1 Tissue fatty acid composition Liver, duodenum, and brain fatty acid compositions were measured after 9 weeks of feeding. Liver oleic acid (OA) was higher (p < 0.01) in the beef tallow diet compared to all other diets (Figure 4.1a) with no differences (p > 0.05) in the duodenum between diets (Figure 4.1d). In liver and duodenum, ARA (Figure 4.1b, e) and docosahexaenoic acid (DHA) (Figure 4.1c, f) levels reflected the dietary pattern of the animals. Specifically, ARA in the animals fed the corn oil diet was higher (p < 0.01) compared to all other dietary groups. Similarly, DHA was highest in the fish oil diet-fed animals (p < 0.01), followed by the canola oil diet with the lowest DHA levels in the corn oil and beef tallow diet-fed animals.

Brain OA and ARA levels were higher (p < 0.01) in the NAPE-PLD (-/-) mice compared to the wild-type mice (Figure 4.1g, h). Also, ARA levels were higher (p < 0.05) in both the corn oil and beef tallow diets, compared to the canola oil and fish oil diets (Figure 4.1h). Brain DHA levels showed a significant interaction effect (genotype x diet, p < 0.05) (Figure 4.1i). NAPE-PLD (-/+) and NAPE-PLD (-/-) mice fed beef tallow diets had higher (p < 0.01) DHA levels (8.52 ± 0.47 μmol/g, 7.98 ± 0.54 μmol/g, respectively) than the wild-type mice fed the same diet (6.34 ± 0.36 μmol/g). Similarly, NAPE-PLD (-/+) and NAPE-PLD (-/-) mice fed fish oil diets (10.01 ± 0.31

μmol/g, 10.89 ± 0.61 μmol/g, respectively) had higher (p < 0.05) DHA levels than the wild-type mice fed the fish oil diet (7.72 ± 0.61 μmol/g).

NAPE-PLD (-/+) and NAPE-PLD (-/-) mice fed the fish oil diets had higher (p < 0.01) brain DHA levels than the animals fed the corn oil diets (7.21 ± 0.49 μmol/g, 8.54 ± 0.48 μmol/g, respectively). However, NAPE-PLD (-/+) mice had higher (p < 0.01) brain DHA when fed the fish oil diet compared to the canola oil diet (7.48 ± 0.36 μmol/g), and NAPE-PLD (-/-) mice fed the fish oil diet had higher (p < 0.01) brain DHA than when fed the beef tallow diet (7.98 ± 0.54 μmol/g). Wild-type mice fed canola oil diet had higher brain DHA levels (8.24 ± 0.45 μmol/g, p < 0.05) than wild-type mice fed beef tallow diet. No other cross-diet differences were found.

Figure 4.1. Fatty acid composition of liver, duodenum, and whole brain

All values are expressed as means ± SEM. Genotype x diet interaction effect on brain DHA: alphabetical letters indicate values are significantly different between diets within a genotype. *Indicated values are significantly different between genotypes upon a same diet.

4.5.2 Tissue NAEs, 1-AG and 2-AG levels 4.5.2.1 Plasma NAPE-PLD (-/-) mice had lower (p < 0.01) plasma OEA levels (16.21 ± 1.06 pmol/ml) compared to the wild-type mice (24.56 ± 1.98 pmol/ml), (Figure 4.2a). Plasma OEA was also higher (p < 0.05) in both the beef tallow (24.49 ± 1.54 pmol/ml) and canola oil (24.96 ± 2.07 pmol/ml) diets compared to either the corn oil (18.86 ± 1.49 pmol/ml) or fish oil (11.6 ± 0.81 pmol/ml) diets, (Figure 4.2a). Plasma AEA was highest (p < 0.05) in the corn oil (1.14 ± 0.08 pmol/ml) and beef tallow (0.98 ± 0.06 pmol/ml) diets, with plasma AEA also being higher in the canola oil diet (0.72 ± 0.08 pmol/ml) compared to the fish oil diet (0.31 ± 0.02 pmol/ml) (Figure 4.2b).

Plasma DHEA levels in the fish oil diet (40.05 ± 3.46 pmol/ml) were at least 3-fold higher than the corn oil (8.71 ± 0.50 pmol/ml) and beef tallow (9.79 ± 0.28 pmol/ml) diets and was also higher (p < 0.0001) than canola oil diet (25.43 ± 2.07 pmol/ml) (Figure 4.2c). In addition, plasma 1-AG levels were higher (p < 0.05) in the corn oil diet (2438 ± 462.8 pmol/ml) compared to the beef tallow (577.1 ± 85.85 pmol/ml) and fish oil (1162 ± 251.5 pmol/ml) diets, but were not different than the canola oil diet (1508 ± 240.8 pmol/ml), while 2-AG levels in the corn oil diet (5138 ± 1186 pmol/ml) was 1.3-5.2-fold higher (p < 0.05) compared to all other diets (beef tallow: 825.9 ± 110 pmol/ml, canola oil: 2215 ± 418.7 pmol/ml, fish oil 1782 ± 393.7 pmol/ml), (Figure 4.2d, e).

Figure 4.2. Plasma NAE, 1-AG, and 2-AG levels.

All values are expressed as means ± SEM.

4.5.2.2 Liver No interaction effect or genotype effect of any liver NAEs was seen (Figure 4.3). Liver OEA levels were not different between diets (Figure 4.3a). Liver AEA, DHEA, 1-AG and 2-AG levels were impacted by dietary groups (Figure 4.3b, c, d, e). Specifically, liver AEA, 1-AG, and 2-AG levels were higher (p < 0.05) in the corn oil diet compared to all other diets (Figure 4.3b, d, e), and DHEA levels were highest (p < 0.05) in the fish oil diet compared to all other diets (Figure 4.3c).

Figure 4.3. Liver NAE, 1-AG, and 2-AG levels

All values are expressed as means ± SEM.

4.5.2.3 Jejunum In the jejunum, NAPE-PLD (-/-) mice had lower (p < 0.05) OEA levels (70.34 ± 7.7 pmol/g) compared to wild-type mice (99.81 ± 8.3 pmol/g) (Figure 4.4a), and NAPE-PLD (-/+) mice had lower (p < 0.01) 2-AG levels (11551 ± 1237 pmol/g) compared to wild-type mice (18378 ± 2394 pmol/g) (Figure 4.4e). Animals fed fish oil diet (461.1 ± 40.42 pmol/g) had higher (p < 0.01) jejunum DHEA compared to the other three diets (beef tallow: 57.36 ± 13.42 pmol/g, canola oil: 294.3 ± 42.63 pmol/g, corn oil: 66.37 ± 8.71 pmol/g) (Figure 4.4c). Mice fed the corn oil diet had higher (p < 0.05) AEA, 1-AG and 2-AG levels compared to the fish oil diet (Figure 4.4b, d, e). Jejunum OEA levels were not different between any of the diets (Figure 4.4a).

Figure 4.4. Jejunum NAE, 1-AG, and 2-AG levels.

All values are expressed as means ± SEM.

4.5.2.4 Brain All NAEs measured in the brain demonstrated significant interaction effects (genotype x diet, p < 0.05). NAPE-PLD (-/+) mice had higher (p < 0.05) OEA levels (101.9 ± 7.88 pmol/g) than both NAPE-PLD (-/-) and wild-type mice fed the corn oil diets (64.13 ± 9.68 pmol/g, 65.20 ± 6.81 pmol/g, respectively), and NAPE-PLD (-/-) had lower (p < 0.05) OEA levels (53.78 ± 4.56 pmol/g) than wild-type mice fed the beef tallow diet (90.47 ± 10.04 pmol/g) (Figure 4.5a). In addition, NAPE-PLD (-/+) mice fed the corn oil diet had higher (p < 0.05) OEA levels (101.9 ± 7.88 pmol/g) than when fed canola oil (65.04 ± 3.39 pmol/g) or fish oil (62.88 ± 6.84 pmol/g) diets (Figure 4.5a).

NAPE-PLD (-/-) mice (0.83 ± 0.13 pmol/g) had lower (p < 0.05) AEA levels than the wild-type mice fed the beef tallow diet (2.18 ± 0.44 pmol/g), and NAPE-PLD (-/-) mice (1.34 ± 0.13 pmol/g) had higher (p < 0.01) AEA levels than NAPE-PLD (-/+) and wild-type mice fed the canola oil diets (0.78 ± 0.13 pmol/g, 0.72 ± 0.12 pmol/g, respectively), (Figure 4.5b). Wild-type mice fed the beef tallow diet (2.18 ± 0.44 pmol/g) had higher (p < 0.01) brain AEA levels than animals fed either canola oil (0.72 ± 0.12 pmol/g) or fish oil (0.88 ± 0.23 pmol/g) diets. In NAPE-PLD (-/+) mice, animals fed corn oil diet (1.85 ± 0.37 pmol/g) had higher (p < 0.05) brain AEA levels than animals fed the fish oil diet (0.46 ± 0.07 pmol/g). An interaction effect (genotype x diet, p < 0.05) for brain 1-AG levels was determined, in which the NAPE-PLD (-/+) mice fed corn oil diet (28.97 ± 6.78 pmol/g) had higher (p < 0.05) 1-AG levels than NAPE-PLD (-/+) mice fed the fish oil diet (7.02 ± 1.32 pmol/g), (Figure 4.5d).

In the wild-type and NAPE-PLD (-/-) mice, respectively, animals fed the fish oil diet (115.7 ± 26.18 and 84.77 ± 9.52 pmol/g) had higher (p < 0.01) brain DHEA levels than animals fed either beef tallow (55.48 ± 7.01 and 30.55 ± 2.99 pmol/g), canola oil (45.23 ± 2.89 and 52.60 ± 2.81pmol/g) or corn oil (41.43 ± 6.25 and 35.32 ± 3.36 pmol/g) diets (Figure 4.5c).

Figure 4.5. Brain NAE, 1-AG, and 2-AG levels.

All values are expressed as means ± SEM. Bars not sharing a letter indicate values are significantly different between diets within a genotype. * Indicated values are significantly different between genotypes upon the same diet.

4.5.3 Food intake and body weight Throughout the nine-week post-weaning feeding period, NAPE-PLD (-/-) mice had lower (p < 0.01) food intake (Figure 4.6a) and lower (p < 0.01) body weight (Figure 4.6b) than NAPE-PLD (-/+) and wild-type mice. Mice fed the fish oil diet had higher (p < 0.01) food intake than mice fed the canola oil diet (Figure 4.6) and this was accompanied by higher body weights (p < 0.05) in the fish oil-fed mice compared to all other diets.

Figure 4.6. The progression of accumulated food intake and weekly body weight

(a) (b)

All values are expressed as means ± SEM. (a) Accumulated average daily food intake. (b) Progression of weekly body weight. Food intake showed a significant (p < 0.01) pattern: NAPE-PLD (-/-) < NAPE (-/+) < wild-type mice, while NAPE-PLD (-/-) mice had lower (p < 0.01) body weight than both wild-type & NAPE-PLD (-/+) mice. Animals fed fish oil diet had higher (p < 0.01) food intake than canola oil diet, while fish oil fed animals had higher (p < 0.05) body weight than beef tallow, canola oil and corn oil diets.

4 Fat oxidation and fat composition Energy expenditure was measured after eight weeks of dietary intervention. Fat oxidation was measured hourly during a five-hour fasting period. There was no genotype or diet effect on fat oxidation rates during this time period, however, fat oxidation increased during the five-hour fasting period among all animals (data not shown).

During the last week of feeding, fat composition was measured using MRI. The fat volume was manually measured (Supplemental Figure 4.1). The fat volume was then divided by the animal’s body weight to calculate the ratio of the fat volume to the body weight as a representative for fat composition. Fat volume (data not shown), as well as the ratio of fat volume to body weight, were significantly affected by genotype (p < 0.0001): NAPE-PLD (-/-) < NAPE-PLD (-/+) < wild-type; but no difference in fat composition between animals on the different diets (Figure 4.7).

Figure 4.7. Ratio of fat volume to body weight

All values are expressed as means ± SEM. The ratio of fat volume to body weight had a significant (p < 0.05) pattern: Wild-type > NAPE-PLD (-/+) > NAPE-PLD (-/-).

4.6 Discussion NAEs are a group of endogenous lipid molecules that have second messenger functions within the cells (Piomelli 2013). While NAPE-PLD is used for NAE biosynthesis, other parallel pathways are also involved. Previous studies demonstrate that modulating dietary fatty acids can alter levels of NAEs in tissues and blood, which in turn can impact homeostatic signals and physiological outcomes (Engeli et al. 2014, Diep et al. 2011, DiPatrizio et al. 2011a). However, it is unknown whether modulating dietary fatty acid composition can alter NAE concentrations in the absence of NAPE-PLD. The aim of this study was to examine the role of NAPE-PLD in lipid metabolism upon dietary fatty acid modulation. Thus, this study investigated the effect of multiple sources of dietary fatty acids on the levels of NAEs, 1-AG and 2-AG using a NAPE- PLD knockout mouse model.

ARA and DHA levels in the liver and duodenum increased as expected in response to corn oil and fish oil diets, respectively, in all three mouse genotypes. Interestingly though, brain OA and ARA levels were significantly higher in the NAPE-PLD (-/-) genotype than in the wild-type mice suggesting that the absence of NAPE-PLD may restrict the conversion of these fatty acids to NAEs, which in turn accumulate in the brain. DHA is an abundant polyunsaturated fatty acid in the brain (Orr et al. 2013, Domenichiello et al. 2015) that cannot be synthesized de novo in mammals (Domenichiello et al. 2015). Fish or fish oils are rich dietary source of DHA (Patterson & Stark 2008), which is primarily supplied to the brain via the blood (Bazinet & Laye 2014). Also, DHA can also be synthesized within the body from metabolic precursors such as ALA (rich in canola oil or flaxseed oil) in very small amounts (Domenichiello et al. 2015). As such, our results found that mice fed the fish oil diet had higher brain DHA levels in NAPE-PLD (-/+) and NAPE-PLD (-/-) mice. Surprisingly, mice fed the beef tallow diets also had higher brain DHA levels in the NAPE-PLD knockouts compared to the wild-type. We believe this is the first such study to demonstrate this effect, and supports a DHA-specific role for NAPE-PLD in the brain. Our results illustrated that upon high saturated and low omega-3 PUFA diets, NAPE-PLD mice may obtain and store a higher amount of DHA in the brain to maintain sufficient conversion to their functional metabolites (such as DHEA).

Although a few studies (Simon & Cravatt 2010, Leung et al. 2006, Simon & Cravatt 2006) have examined tissue levels of NAEs and their related enzymes’ activities, this is the first study that reports tissue fatty acid composition in a NAPE-PLD knockout mouse. As such, the mechanism yielding higher brain DHA levels in NAPE-PLD mice is not clear. DHA is required for proper brain development, growth and functions (Metherel et al. 2017b, Bazinet & Laye 2014, Hashimoto et al. 2014, Lo Van et al. 2016, Weiser et al. 2016), and the formation of DHEA from DHA is thought to provide anti-inflammatory, anti-nociceptive, and organ-protective effects (Brown et al. 2010, Meijerink et al. 2013, Figueroa et al. 2013). Therefore, to maintain sufficient levels of DHEA in the brain, it is possible that a larger pool of brain DHA is required for DHEA synthesis via alternate enzymatic pathways, and this could be accomplished by increased brain DHA uptake from the plasma unesterified pool or reducing brain DHA turnover (Chen & Bazinet 2015, Chen et al. 2015). While it is interesting to speculate, the concentrations of DHEA are about 100,000 times smaller than the total DHA concentration in the brain. Thus, this appears unlikely to be the only mechanism resulting in higher total lipid DHA in the NAPE-PLD- deficient mice. Future studies should examine the unesterified DHA concentration and the association between unesterified DHA and DHEA in this NAPE-PLD-deficient mouse model.

Brain NAEs, 1-AG, and 2-AG can increase rapidly during ischemia as a neurological protective mechanism (Bazinet et al. 2005, Brose et al. 2016, Lin et al. 2017). Brain basal NAEs in our wild-type mice are similar to our previous work using the same wild-type mice (Lin et al. 2017). A study by Leung et al. (Leung et al. 2006) assessing brain NAE levels from NAPE-PLD knockout and their wild-type littermates sacrificed by decapitation found similar brain OEA and AEA levels from the same strain of mice. However, our brain DHEA levels were generally higher compared to Leung et al. (Leung et al. 2006). Importantly, our mice were fed isocaloric diets containing 10% wt/wt fat for 9 weeks post-weaning, but it is unclear what the animal age and the dietary composition were in Leung et al. (Leung et al. 2006).

In general, mice fed the corn oil diet, high in omega-6 PUFA, increased AEA, and 1- and 2-AG levels in peripheral tissues. These findings are supported by others (Artmann et al. 2008a, Kim et al. 2016, Lin et al. 2013a, Alvheim et al. 2012b, Alvheim et al. 2014a), and confirm that dietary fatty acids can be converted to NAEs and AGs with or without NAPE-PLD. Future studies

aimed at assessing the role of other NAE-producing enzymes including ABH4, GDE1 and/or PLA1/A2 are warranted. Conversely, the biosynthesis of 1- and 2-AGs involve enzymes such as diacylglycerol lipase, phospholipase A1 or phospholipase Cβ (Ueda et al. 1993b, Tsutsumi et al. 1994, Oka et al. 2005, Ahn et al. 2016), but not NAPE-PLD. As such, we confirmed that plasma and liver 1- and 2-AG levels were not different between NAPE-PLD knockouts and wild-type mice. Interestingly, jejunum 2-AG levels were higher in the wild-type mice, and brain 1-AG levels were higher in the NAPE-PLD (-/+) mice fed corn oil diet than the fish oil diet. These results may be partially driven by the significant diet effect found for both duodenum and brain ARA levels.

As mentioned previously, individual NAEs appear to bind to different receptors resulting distinct physiological outcomes (Fu et al. 2003, Guzman et al. 2004, Fu et al. 2007, Brown et al. 2010, Calder 2013, Rodriguez de Fonseca et al. 2001), and our results showed that the absence of NAPE-PLD affects tissue fatty acid, NAE, and 1- and 2-AG composition upon dietary interventions. However, we wanted to understand whether these changes could affect physiological outcomes. To the best our knowledge, ours is the first study to examine the impact of NAPE-PLD deficiency on body composition and energy balance while consuming isocaloric fat diets. We have demonstrated that independent of dietary intervention, NAPE-PLD (-/-) mice had lower food intake, body weight, and fat composition compared to wild-type mice.

Leung et al. (Leung et al. 2006) demonstrated that the NAPE-PLD mice used in our study have lower brain OEA levels and long-chain saturated NAE (> C20) levels, but AEA or DHEA levels were not statistically different between the NAPE-PLD knockout and their wild-type littermates. Similarly, plasma and jejunum OEA in NAPE-PLD (-/-), but not AEA or DHEA levels were lower in the NAPE-PLD mice compared to the wild-type mice. OEA is an endogenous lipid mediator, which binds to PPAR-α that appears to contribute to the peripheral regulation of food intake. Many studies (Fu et al. 2003, Guzman et al. 2004, Oveisi et al. 2004, Igarashi et al. 2015a) have shown that elevation of OEA levels in the small intestine and other tissues inhibits food intake and reduces body weight in free-feeding animals, however, our results show an opposite trend. This may be due to the multiple changes in NAE levels induced by NAPE-PLD deletion. Recently, a clinical study found that a single SNP rs17605251 in the NAPE-PLD gene

was nominally associated with a BMI > 35 (Wangensteen et al. 2011), indicating a potential role for NAPE-PLD in humans.

Other NAPE-PLD knockout lines (Tsuboi et al. 2011, Leishman et al. 2016, Liu et al. 2008) exist, and the Deutsch line (Tsuboi et al. 2011) demonstrated similar OEA response as the one described in this study, however, the Luquet line (Liu et al. 2008) and the Deutsch line demonstrate lower brain AEA levels compared to wild-types. Therefore, the leaner phenotype on the Cravatt NAPE-PLD line may be related to the genetic modification technique, and future studies examining the physiological outcomes on other NAPE-PLD knockout lines combined with dietary modulation are warranted. Also, previous studies by Cravatt did not report the growth of NAPE-PLD knockout. Therefore, the leaner phenotype could be a result from lower food intake and less overall nutrient intake. Since we did not adjust the food intake to match wild-type and NAPE-PLD knockout mice, food intake could be a confounder in this results. Thus, NAPE-PLD knockout may be used for the potential importance of this complex lipid metabolism system as a novel target. Future studies should consider adjusting food intake to rule out this potential confounder effect. Furthermore, assessment of gene expression, enzyme activities and fatty acid and NAE specific biosynthesis upon dietary intervention may identify upregulation of alternative pathways in the absence of NAPE-PLD.

In conclusion, we found that NAPE-PLD (-/-) mice had higher brain DHA than wild-type mice with no measurable effect on brain DHEA levels, however, additional experimentation is required to assess potential neurological benefits. Furthermore, we found that NAPE-PLD knockout mice resulted in a leaner phenotype combined with lower plasma and jejunum OEA levels, which are independent of dietary fatty acid modulation. In addition, we demonstrated that dietary fatty acids can modulate plasma and tissue NAE levels in the absence of NAPE-PLD, which suggest that NAPE-PLD is not necessary for NAE synthesis, thereby highlighting the important role of alternative pathways in maintaining NAE levels. While much interest in the endocannabinoid system has focused on the regulation of DHEA, OEA, and other NAEs, understanding how NAPE-PLD and other NAE synthesis enzymes combined with dietary modulations may be involved in neuroprotection and weight management is an exciting future research avenue.

Supplemental Figure 4.1 Example of one magnetic resonance imaging slice of total body composition

Fat tissues are the white and light grey matters, which were circled using red lines. Each animal has 20-30 sliced images. The fat volume in each slice was recorded and the total animal fat volume was used for analysis.

CHAPTER 5. (Study 2) Fatty acid amide hydrolase (FAAH) regulates hypercapnia/ischemia-induced increases in n-acylethanolamines in mouse brain

Adapted from: Lin, L., A. H. Metherel, P. J. Jones and R. P. Bazinet (2017). "Fatty acid amide hydrolase (FAAH) regulates hypercapnia/ischemia-induced increases in n-acylethanolamines in mouse brain." J Neurochem:142(5):662-671

100

5.1 Abstract N-acylethanolamines (NAEs) are endogenous lipid ligands for several receptors including cannabinoid receptors and peroxisome proliferator-activated receptor-alpha (PPAR-α), which regulate numerous physiological functions. Fatty acid amide hydrolase (FAAH) is, largely, responsible for the degradation of NAEs. However, at high concentrations of ethanolamines and unesterified fatty acids, FAAH can also catalyze the reverse reaction, producing NAEs. Several brain insults such as ischemia and hypoxia increase brain unesterified fatty acids. Because

FAAH can catalyze the synthesis of NAE, we aimed to test if FAAH was necessary for CO2 - induced hypercapnia/ischemia increases in NAE. To test this, we examined levels of NAEs, 1- and 2-arachidonoylglycerols (AGs) as well as their corresponding fatty acid precursors in wild- type and mice lacking FAAH (FAAH-KO) with three kill methods: 1) head-focused, high-energy microwave irradiation (microwave), 2) 5 min CO2 followed by microwave irradiation (CO2 + microwave) and 3) 5 min CO2 only (CO2). Both CO2-induced groups increased, to a similar extent, brain levels of unesterified oleic, arachidonic and docosahexaenoic acid and AGs compared to the microwave group in both wild-type and FAAH-KO mice. Oleoylethanolamide (OEA), arachidonoylethanolamide (AEA) and docosahexaenoylethanolamide (DHEA) levels were about 8, 7, and 2.5 fold higher, respectively, in the FAAH-KO mice compared with the wild-type mice. Interestingly, the concentrations of OEA, AEA, and DHEA increased 2.5 to 4 fold in response to both CO2-induced groups in wild-type mice, but DHEA increased only in the

CO2 group in FAAH-KO mice. Our study demonstrates that FAAH is necessary for CO2- induced increases in OEA and AEA, but not DHEA. Targeting brain FAAH could impair the production of NAEs in response to brain injuries.

101

5.2 Introduction Fatty acid amide hydrolase (FAAH) is membrane-bound and localized predominantly in microsomal or mitochondrial fractions of the brain and liver (Ueda et al. 2000, Ueda et al. 1995, Schmid et al. 1985). FAAH is responsible for the termination of lipid signaling molecules from the family of n-acylethanolamines (NAEs) in the central nervous system and in peripheral tissues producing a corresponding fatty acid and ethanolamine (Porter & Felder 2001, Devane et al. 1992). Also, FAAH can reversely synthesize NAEs at high concentrations of free fatty acids and ethanolamines (Katayama et al. 1999, Ueda et al. 1995), although this does not occur under normal physiological conditions (Kurahashi et al. 1997, Ueda et al. 2000). Moreover, while in vitro studies suggested that FAAH can also hydrolyze 2-arachidonoylglycerol (2-AG) into arachidonic acid (ARA) and glycerol (Goparaju et al. 1998), in vivo studies have demonstrated that 2-AG is mainly hydrolyzed by a monoacylglycerol lipase (MAGL) and that 2-AG is not a substrate for FAAH (Sugiura et al. 1995, Osei-Hyiaman et al. 2005a, Pacher et al. 2006).

Unesterified fatty acids, NAEs and 2-AG rapidly accumulate during ischemia and post-mortem in the brain (Bazan 1970, Bazan et al. 1993, Bazinet et al. 2005, Brose et al. 2016, Trepanier et al. 2017). The rapid change of NAE and 2-AG during ischemia may be a protective mechanism, which prevents damage of nerve cell membranes and alters cationic responses in the brain in an attempt to maintain functionality (Bazinet et al. 2005, Brose et al. 2016). While ischemia- induced increases in unesterified fatty acids appear to be mediated, at least in part, by activation of calcium-dependent cytosolic phospholipase A2, the mechanism by which ischemia increases NAEs is not known. Furthermore, the speed at which NAE accumulate makes measurement of their true basal levels difficult. Head-focused, high-energy microwave irradiation rapidly stops enzyme activities and is generally accepted as the gold standard for measuring basal lipid levels in the brain (Bazinet et al. 2005, Brose et al. 2016, Murphy 2010).

Previously, ischemia was induced by decapitation into liquid nitrogen (Cenedella et al. 1975,

Lunt & Rowe 1968, Hadjipanayi & Schilling 2013, Bazan 1970). However, CO2 is commonly used in the laboratory prior to decapitation and we continued with this model, recognizing that

CO2 initially induces hypercapnia followed by ischemia (Pollock et al. 2009, Trepanier et al. 2017).

102

The aim of this study was to assess tissue levels of NAEs and their corresponding fatty acid precursors, as well as 1 and 2-AGs under either basal or CO2-induced hypercapnia followed by ischemia. In addition, we aimed to test the role of FAAH in the levels of NAEs, NAE corresponding fatty acid precursors and 1 and 2-AGs between the FAAH-KO and wild-type mice. Because, at high concentrations of fatty acid precursors and ethanolamides, FAAH can synthesize NAE, we hypothesized that FAAH would be necessary for CO2-induced increases in brain NAE levels.

5.3 Methods All procedures were performed in agreement with the policies set out by the Canadian Council on Animal Care and were approved by the Animal Ethics Committee at the University of Toronto, Ontario, Canada (Canadian Council on Animal Care 1993).

5.3.1 Diets Dams throughout breeding and lactation, and offspring mice from weaning were fed a standard chow diet (Teklad 2018, Harlan, Indianapolis, IN, USA). The diet by weight contained 6.2% fat (18% energy), 18.6% protein (24% energy) and 44.2% carbohydrate (58% energy). The detailed fatty acid composition of the diet has been previously published (Orr et al. 2013, Chen et al. 2015).

5.3.2 Animals FAAH-KO mice were kindly donated by Dr. Benjamin Cravatt (The Scripps Research Institute, California). C57BL/6 wild-type mice were obtained from Charles River Laboratories. Animals were acclimatized for 1 week in an animal facility, in which temperature (21°C), humidity and light cycle (12 hours at light / dark circle) were controlled, and had ad libitum access to food and water. One male and one female with the same genotype were paired for breeding. The pups were weaned at 21 days of age; then male pup littermates were housed up to four per cage with ear tagged ID and color-coded cage cards to indicate the genotypes. Therefore, the investigators were not blinded to the genotype of mice they were handling. All animals received the same chow diet and were arbitrarily assigned in the animal facility room without specific tools to achieve randomization until 12 weeks of age.

103

Fasting and refeeding can alter NAE and 2-AG levels in brain and peripheral tissues, in which refeeding generally increases OEA, but decreases 2-AG and AEA (DiPatrizio et al. 2015, Igarashi et al. 2015b, Fu et al. 2007, Izzo et al. 2010). Thus, we removed food 12 hours prior to kill to limit variations in acute food intake on NAEs and AGs. Because Habayeb et al. (2004) demonstrated that menstrual cycle can alter women plasma AEA levels, we excluded female mice in this study to avoid the interference of mouse estrous cycle, but future studies should consider this. The sample size (n=8) was based on previous studies (Bazinet et al. 2005, Brose et al. 2016) but because the exact experimental procedures had never been performed we did not conduct a formal power calculation.

5.3.3 Kill methods At 12 weeks of age, littermate male mice housed in the same cage were arbitrarily selected and killed by either head-focused, high-energy microwave irradiation (microwave; 1 kW, 0.88-0.99 seconds; Cober Electronics Inc., Stratford, CT, USA), CO2 + microwave (5 min CO2 then microwave fixation) or complete CO2 (CO2 5 min only) methods. Mice were arbitrarily allocated to one of the three kill methods for testing the effect of ischemia on NAE and AG levels. Removal of the whole brain took about 5min / animal and upon removal was immediately stored o at -80 C (Figure 5.1) until further analysis. While we refer to the CO2 groups, it is important to note that hypoxia and hypercapnia are both active processes in the brain with these methods.

104

Figure 5.1. Study design

After weaning (21 days of age), Both FAAH-KO and Wild-type mice consumed standard chow diet for 9 weeks. At 12 weeks of age, after 12 hours fasting, mice were euthanized with either 1) Microwave only; 2) CO2 + microwave: 5 min CO2 + microwave irradiation; 3) CO2 only: 5 min CO2 only; following euthanasia the whole brain collection took ~5 min / animal, then stored at -80 oC.

105

5.3.4 Whole brain lipid extraction and gas chromatography-mass spectrometry Total lipid extracts (TLE) were obtained from frozen powdered whole brain samples according to the method of Folch (Folch et al. 1957). Brain samples were labeled based on their ear-tag ID, so the operator was blinded to the kill methods, but aware of the genotypes when processing samples. Briefly, lipids were extracted with chloroform: methanol: potassium chloride (2:1:0.88 by volume). A known amount of ARA-deuterated-8 (D8) (Cambridge Isotope Laboratories, Inc, Tewksbury, MA, USA) was added as an internal standard for fatty acid extraction. The mixtures were vortexed, centrifuged at 490 g for 10 min, and the lower, chloroform lipid-containing layer was pipetted into a new test tube and stored at -80 °C.

TLE was then loaded on a thin-layer-chromatography (TLC) plate, which was washed with chloroform: methanol (2:1 by volume) and activated by heating at 100 oC for 1 hour. The plate was placed in a tank with solvents (heptane: diethyl ether: acetic acid; 60:40:2 by volume). After drying the plate at room temperature, the individual bands were visualized under ultraviolet light, after lightly spraying with 8-anilino-1-naphthalene sulfonic acid in methanol (0.1% by volume) (Hopperton et al. 2014, Song et al. 2010, Wang & Gustafson 1992). Bands containing free fatty acids were scraped into test tubes and extracted as previously described (Metherel et al. 2017a). The dried total lipid extracts were then dissolved in 100 μL of (1:10:1000, by volume) pentafluorobenzyl bromide: diisopropylamine: acetonitrile and heated at 60 °C for 15 min to form the fatty acid pentafluorobenzyl (PFB) esters as previously described (Pawlosky et al. 1992, Metherel et al. 2017a). Upon removal from heat, the reagent mixture was evaporated under nitrogen, dissolved in 100 μL of hexane and stored in GC vials until analysis by GC-MS. Fatty acid PFB esters were analyzed on a Agilent 5977A quadrupole mass spectrometer coupled to an Agilent 7890B gas chromatograph (Agilent Technologies, Mississauga, ON) in negative chemical ionization mode, as described previously (Pawlosky et al. 1992, Metherel et al. 2017a). The fatty acid PFB esters were injected via an Agilent 7693 auto-sampler into a DB-FFAP 30 m x 0.25 mm i.d. x 25 μm film thickness capillary column (J&W Scientific from Agilent Technologies, Mississauga, ON) interfaced directly into the ion source with helium as the carrier gas. Initial oven temperature was 80 °C, immediately followed by a 20 °C/min ramp to 185 °C and a 10 °C/min ramp to 240 °C and a 35 min hold at the end. The injector and transfer line were set at 250 °C and 280 °C, respectively, while the ion source and quadrupole were both set at

106

150 °C. Methane (99.999%) was used as the ionization gas. Fatty acids were analyzed in selected ion monitoring (SIM) mode using M-1 for parent ion identification with ion dwell times of 500 μs. The ion mass parent ions for each fatty acid are OA (281.3); ARA (303.3); DHA (327.3).

5.3.5 NAEs and 1-AG and 2-AG sample extractions Stock solutions of oleoylethanolamide (OEA), arachidonoylethanolamide (AEA), O-AEA (also called virodhamine), docosahexaenoylethanolamide (DHEA), 1- or 2-arachidonoylglycerols (1- and 2-AGs) as well as OEA-D2, AEA-D8, DHEA-D4, 1-AG-D5 were reconstituted in acetonitrile and stored at -80 oC before further dilution, which was performed in acetonitrile on ice. Calibration curves were prepared using a concentration span covering a range of 0.1–200.0 ng/ml. Each point of the calibration curve was constructed by adding a constant amount of cocktail of internal standards (50ng/ml of OEA-D2, AEA-D8, DHEA-D4 and 250 ng/ml of 1-AG-

D5) for quantification purposes.

Powdered whole brain samples were prepared for NAEs, 1-AG and 2-AG analysis according to our previously published method (Lin et al. 2012) with modifications. As aforementioned, the operator was blinded to kill methods, but was aware of the genotypes during sample preparation. Briefly, samples were mixed with a cocktail of internal standards. The mixture was prepared with ice-cold acetone (2ml); homogenized and centrifuged at 2000 rpm for 10 min at 4 oC. The supernatant was transferred to a clean tube and dried under N2 gas. Chloroform: methanol: deionized water (2:1:1 by volume) was then added into the dried samples, which was vortexed and centrifuged at 2000 rpm for 10 min at 4 oC. The chloroform layer was transferred into another clean tube, dried under N2 gas and dissolved in 100ul of acetonitrile. Finally, samples were transferred into vials for LC-MS/MS analysis.

5.3.6 Identification and separation using high-performance liquid chromatography mass spectrometry MS/MS was performed with a SCIEX QTrap5500 mass spectrometer (SCIEX: Framingham, Massachusetts, USA) with an Agilent 1290 HPLC system (Agilent Technologies: Santa Clara, California, USA). Chromatography was done on a Phenomenex Kinetex XB-C18 column,

107

50x4.6 mm, 2.6 µm (Phenomenex, Torrance, CA, USA) at a flow rate of 600 µL/min. The mobile phase consisted of 30% A (water + 0.1% formic acid) starting and ramping up to 75% B (acetonitrile + 0.1% formic acid) for a total analysis time of 6.5 min. Mass spectrometry was performed in positive ESI (electrospray ionization) MRM (multiple reaction monitoring) mode using a source temperature of 600°C and an ion spray voltage of 5200 V. The ion mass transitions used were as follows: OEA (m/z 326.3 / 62.0, OEA-d2 (m/z 328.3 / 62.0), AEA and O-AEA (m/z 348.3 / 62.0), AEA-d8 (m/z 356.3 / 62.0), DHEA (m/z 372.3 / 62.0), DHEA-d4 (m/z 376.3/ 66.0), 1-AG and 2-AG (m/z 379.3 / 287.3), 1-AG-d5 (m/z 384.3/ 287.3). Peak integration and data analysis was performed using Analyst software version 1.6 (Sciex, Framingham, Massachusetts, USA) (Figure 5.2).

The chromatography was optimized to identify the target compounds and their isomers. We found a shoulder peak of OEA in our mouse brain samples (Figure 5.2, a-right) which did not occur in the standard samples (Figure 5.2, a-left). This shoulder peak had an identical mass and fragmentation patterns as those of OEA in our standard (Figure 5.2, a-left), and has been identified as cis-vaccenic acid ethanolamide (also called “vaccenoylethanolamide”, VEA) as previously reported by Rohrig et al. (2016). Also, we were able to identify and separate AEA and its isomer (O-AEA) (Figure 5.2, b-left), which allowed us to quantify AEA levels in the brain. Moreover, 1-AG and 2-AG were separated and quantified using 1-AG-D5, because the 2- AG deuterated commercial standard sample was only 90% pure (with 10% 1-AG inside). Detailed chromatographic separation for each target NAE and their isomers (O-AEA), 1-AG and 2-AG is presented in Figure 5.2.

108

Figure 5.2. The multiple reaction monitoring chromatography of NAEs and AGs

Modified chromatography of (a) OEA, OEA-D2 and OEA isomer (vaccenoylethanolamide, VEA), the inset figure indicated the mass spectrum of OEA; (b) Separation of AEA, AEA-D8 and AEA isomer (virodhamine, O-AEA), the inset figure indicated the mass spectrum between O-AEA and AEA; (c) DHEA and DHEA-D4, the inset figure indicated the mass spectrum of DHEA; (d) 1-AG, 2-AG and 1-AG-D5, the inset figure indicated the mass spectrum of 2-AG.

109

5.4 Statistics All statistical analyses were performed with Graphpad Prism 6. Differences in fatty acid concentrations and the levels of 1-AG and 2-AG were assessed by two-way ANOVA. Then, the significant interactions were further analyzed by one-way ANOVA with Tukey’s multiple comparisons post-hoc test.

Previous studies showed that the levels of AEA in the FAAH-KO mice are dramatically higher compared with wild-type littermates (Cravatt et al. 2001). Thus, we conducted a D’Agostino & Pearson omnibus normality test, GraphPad Prism 6 on NAE and none of the NAEs were normally distributed (p < 0.0001) (Supplemental Figure 5.1). Thus, we used one-way ANOVA with Tukey’s multiple comparisons post-hoc test to compare the effects of kill methods on the FAAH-KO mice or the wild-type mice separately for NAEs. All of our results are expressed as means ± SEM. Statistical significance was set as p < 0.05 for all analyses.

5.5 Results

5.5.1 Whole brain total lipids are not altered by CO2-induced hypercapnia/ischemia or FAAH-KO There was no interaction effect between genotype and kill method on the levels of total OA,

ARA, and DHA. Also, CO2-induced hypercapnia/ischemia did not increase the levels of total OA, ARA, and DHA compared to the control. As well, the FAAH-KO mice had similar concentrations of total OA, ARA and DHA compared to the wild-type mice (Figure 5.3).

110

Figure 5.3. Whole brain total lipids

nmol/g, wet weight: wild-type and FAAH-KO mice. Statistical analysis was performed by two-way ANOVA (kill method x genotype). No statistical interaction or main effects of kill method or genotype were observed for total (a) oleic acid, (b) arachidonic acid, and (c) docosahexaenoic acid.

111

Unesterified lipids are elevated upon CO2-induced hypercapnia/ischemia There was no interaction effect between genotype and kill method on unesterified OA, ARA or DHA concentrations (Figure 5.4). The interaction effect between genotype and kill method on unesterified DHA was p = 0.054, and the main effect was found in the kill methods (Figure

5.4c). Both CO2 + microwave and CO2 groups dose-dependently increased (p < 0.0001) unesterified OA, ARA and DHA levels compared to the control. The concentrations of unesterified ARA in the CO2 + microwave and CO2 groups was 20- and 34-fold higher, respectively, compared to the microwave group (Figure 5.4b). There were no effects of genotype on unesterified OA (Figure 5.4a) or ARA (Figure 5.4b) levels. Interestingly, FAAH-KO mice in the CO2 group had higher (p = 0.0025) unesterified DHA compared with the wild-type mice in the CO2 group (Figure 5.4c).

112

Figure 5.4. Whole brain unesterified fatty acid concentrations

nmol/g, wet weight: wild-type and FAAH-KO mice. Statistical analysis was performed by two-way ANOVA (kill method x genotype). There was no interaction effect between kill method and genotype on unesterified (a) oleic acid; (b) arachidonic acid and (c) docosahexaenoic acid. The main effect of kill method between groups was found in a step-wise pattern: microwave < CO2 + microwave < CO2. No genotype effect on unesterified (a) oleic acid and (b) arachidonic acid. However, there was a significant genotype effect on unesterified (c) docosahexaenoic acid, in which FAAH-KO mice euthanized in the CO2 group had higher unesterified docosahexaenoic acid than other groups.

113

5.5.2 OEA, AEA and DHEA are elevated in the FAAH-KO mice FAAH-KO mice showed 8-fold higher OEA, 7-fold higher AEA and 2.5-fold higher DHEA levels compared to the wild-type mice (Figure. 5.5)

5.5.3 OEA, AEA, and DHEA are elevated with CO2-induced hypercapnia/ischemia in wild-type, but only DHEA is elevated in FAAH-KO mice

In the wild-type mice, both CO2 + microwave and CO2 groups increased the levels of OEA, AEA, and DHEA compared with the microwave group (p < 0.0001). Levels of OEA and DHEA (66.4 ± 4.1 pmol/g, 54.9 ± 4.9 pmol/g, respectively) in the microwave group were increased 3- fold with CO2-induced groups (Figure 5.5, a & c). Also, CO2-induced hypercapnia/ischemia resulted in a 4-fold higher AEA concentration compared to the microwave group (Figure 5.5 b). Although the levels of OEA, AEA, and DHEA were significantly higher in the FAAH-KO mice compared to the wild-type mice, CO2 did not affect the levels of OEA, AEA in the FAAH-KO

(Figure 5.5, a & b). Levels of DHEA in the CO2 only group were higher (p = 0.0153) than the microwave group in the FAAH-KO, but not the CO2 + microwave group (Figure 5.5 c).

114

Figure 5.5. Whole brain NAE concentrations

pmol/g, wet weight: wild-type and FAAH-KO mice. (a) OEA, (b) AEA and (c) DHEA. Significantly different means within a genotype are represented by different alpha letters determined by one-way ANOVA and followed by Tukey’s post-hoc test.

115 1-AG and 2-AG are elevated upon CO2-induced hypercapnia/ischemia There were no differences in the concentrations of 1-AG and 2-AG between FAAH-KO and wild-type mice. A main effect of the kill method was shown for 1-AG and 2-AG (Figure 5.6).

Both CO2 + microwave and CO2 groups dose-dependently increased (p < 0.0001) 1-AG and 2- AG levels compared to the microwave group (Figure 5.6).

116 Figure 5.6. Whole brain 1 & 2-AG concentrations

pmol/g, wet weight: wild-type and FAAH-KO mice (a) 1-arachidonoyl-glycerol and (b) 2-arachidonoyl-glycerol. Statistical analysis was performed by two-way ANOVA (kill method x genotype). There was no interaction effect between kill method and genotype or genotype effects. The main effect of the type of kill methods was found in a step-wise pattern: microwave < CO2 + microwave < CO2.

117 5.6 Discussion In this study, we used FAAH-KO mice, which lack the ability to break down NAEs into fatty acids and ethanolamines (Cravatt et al. 2001) to study its role in brain hypercapnia/ischemia. For the wild-type mice, our results were in line with previous studies (Brose et al. 2016, Bazinet et al. 2005, Trepanier et al. 2017) showing CO2-induced increases in AEA and 2-AG. For the first time, our study also demonstrated that CO2 increased OEA, DHEA, and 1-AG levels compared to non-ischemic controls. Thus, our results are consistent with several reports suggesting microwave irradiation is an important tool for measuring basal levels of unesterified fatty acids, AEA, and 2-AG (Brose et al. 2016, Bazinet et al. 2005, Trepanier et al. 2017). However, we identified an abundant OEA isomer, VEA, in the mouse brain, which matches other OEA parameters. It is unlikely that VEA is an artifact of OEA in brain samples. VEA was recently detected in human and rodent plasma by Rohrig et al. (2016), who suggested that it may be derived from the gut microbiota. However, the source or function of VEA in the brain is not clear and further research is warranted.

Under normal physiological conditions, fatty acids esterified to the sn-1 position of phospholipids and free fatty acids can serve as precursors for NAE synthesis (Okamoto et al. 2007, Artmann et al. 2008a, Köfalvi 2008). However, it is thought that the phospholipid esterified pool is the major precursor for NAE production (Balvers et al. 2013). During CO2- induced hypercapnia/ischemia, unesterified fatty acids increase rapidly and may allow for FAAH to catalyze the reverse reaction and synthesize NAEs (Patel et al. 2005, Katayama et al. 1999, Ueda et al. 2010). In wild-type mice, we observed an increase in OEA, AEA, and DHEA levels upon CO2-induced hypercapnia/ischemia. However, despite similar increases in precursor fatty acids upon CO2-induced hypercapnia/ischemia in FAAH-KO mice, there was no increase in corresponding OEA and AEA levels. This effect was also selective as levels of 1- and 2-AG increased upon CO2-induced hypercapnia in both FAAH-KO and wild-type mice. While it is possible that FAAH is catalyzing the production of OEA and AEA through the process from hypercapnia to ischemia, alternative explanations cannot be ruled out. Firstly, because basal NAE levels are higher in FAAH-KO mice, there may be a ceiling effect and thus, no further increases in OEA and AEA would be possible. The finding that DHEA was increased in the CO2 group vs the microwave group was not expected. However, there was a main effect of FAAH-

118 KO on unesterified DHA levels and the interaction effect approached statistical significance (p = 0.0545) so it is possible that increases in unesterified DHA in FAAH-KO, in part, drove the increase in DHEA in the FAAH-KO group upon CO2-induced hypercapnia/ischemia.

Furthermore, we cannot rule out that FAAH-KO were protected against aspects of the CO2- induced hypercapnia/ischemia, but this seems unlikely as responses to unesterified fatty acids and AGs were generally similar between wild-type and FAAH-KO mice. As mentioned above, one exception was unesterified DHA, which was significantly higher in FAAH-KO mice as compared to wild-type mice. Unesterified DHA and bioactive lipid mediators derived from DHA are anti-inflammatory and protective against hypercapnia in the brain (Orr et al. 2013, Bazan et al. 2011). Also, it is possible that NAPE-PLD or other unknown enzymes capable of synthesizing NAEs or metabolizing NAEs, including MAGL (Nomura et al. 2011), are up or down regulated in FAAH-KO mice or are selectively active during CO2-induced hypercapnia or ischemia.

Future research is warranted to test these possible interactions in FAAH-KO mice. Nevertheless, FAAH is necessary for the increases in NAE levels upon hypercapnia/ischemia and this effect is selective as increases in the levels of unesterified fatty acids and AGs are generally not different in FAAH-KO mice compared to wild-type mice upon CO2. This finding might be important as FAAH inhibition is being explored as a novel therapeutic for the brain (Mallet et al. 2016, Hansen et al. 2001, Patel et al. 2005). Our results suggest that inhibiting FAAH may decrease the brain's ability to produce NAE in response to stress such as hypercapnia/ischemia. Further research is needed to test the clinical significance of this finding. In conclusion, we find that FAAH-KO mice have similar concentrations of unesterified fatty acids and 1-AG, 2-AG as wild- type (C57BL/6) mice. The CO2-induced hypercapnia and post-mortem delay results in dramatic accumulation of the compounds in the brain. FAAH was necessary for the CO2-induced increases in OEA and AEA levels in the brain, but not DHEA.

119 Supplemental figure 5.1. Normality test for NAE levels in the whole brain

120

Supplemental Figure 5.2. The schematic diagram of this study

121

CHAPTER 6. General Discussion

122 6.1. Summary of research findings The thesis focuses on understanding how the interaction between internal factors (e.g., NAPE- PLD, FAAH) and external factors (e.g., dietary fatty acids, ischemia) influence NAE concentrations. I conducted two studies to examine the hypothesis that “If internal factors, NAPE-PLD and FAAH, are the primary enzymes responsible for the synthesis and degradation of NAEs, respectively, then external factors (dietary fatty acids or ischemia) will not be able to further elevate NAE concentrations in the absence of these enzymes”.

My first study (Chapter 4) examined the hypothesis that “NAPE-PLD is necessary for increases in NAPEs upon dietary fatty acid manipulation”. The results in the chapter 4 demonstrated that dietary fatty acids could increase tissue NAE concentrations with the absence of NAPE-PLD. The finding in this study showed that NAPE-PLD is not necessary for increases in NAEs upon dietary fatty acid manipulation.

My second study (Chapter 5) examined the hypothesis that “FAAH is necessary for CO2-induced hypercapnia/ischemia increases in NAEs”. The results in the Chapter 5 demonstrated that CO2- induced hypercapnia/ischemia could increase whole brain unesterified fatty acids, 1,2-AGs in both wild-type and FAAH-knockout. However, CO2-induced hypercapnia/ischemia increased all measure NAE levels in the wild-type, but only DHEA were increased in the FAAH-knockout.

This finding indicated that FAAH is necessary for CO2-induced hypercapnia/ischemia increases in OEA, AEA, but not DHEA.

These studies further our general understanding of the conversion of dietary fatty acids to NAEs, 1-AG, and 2-AG via NAPE-PLD, and their subsequent degradation via FAAH. First, our research demonstrated that NAPE-PLD is OEA selective, which is in-line with findings in the brain from the previous study (Leung et al. 2006). However, the conversion of dietary fatty acids to NAEs was only slightly impacted by the absence of NAPE-PLD suggesting other parallel pathways contributed to the synthesis process. Second, another internal factor, FAAH, played an important role in regulating NAE degradation pathways because the absence of FAAH resulted in a significant increase of NAE levels. More importantly, FAAH is necessary for CO2-induced hypercapnia/ischemia increases in OEA and AEA but not DHEA in the brain. Third, the research

123 presented herein demonstrated that external factors (dietary fatty acids and CO2-induced hypercapnia/ischemia) elevate NAE and AG levels in the wild-type mice suggesting that external factor can independently play a significant role on regulating NAE concentrations

6.2 Regulatory pathway associated with NAE biosynthesis and degradation The studies from this thesis aimed to examine the role of NAPE-PLD and FAAH on the conversion of fatty acids to NAEs. Our results showed that this Cravatt NAPE-PLD knockout model is OEA selective, which was in line with the original work by Cravatt. Interestingly, NAPE-PLD knockout had also lower food intake, body weight, and fat composition than the wild-type mice. These results demonstrate a sophisticated regulation is underpinning the conversion of fatty acids to NAEs, particularly brain DHA to DHEA. The absent NAPE-PLD technique may play a role in modulating physiological parameters. However, previous studies did not report the growth of NAPE-PLD knockout. Therefore, the leaner phenotype could be a result from lower food intake. Since we did not adjust the food intake to match wild-type and NAPE-PLD knockout mice, food intake could be a confounder which play a role in the leaner phenotype. Thus, NAPE-PLD knockout may be used for the potential importance of this complex lipid metabolism system as a novel target. Future studies should consider adjusting food intake to rule out this potential confounder effect.

Following up, to assess the importance of FAAH on NAE degradation process, this thesis compared levels of NAEs in between FAAH knockout mice and the wild-type mice. The results showed that FAAH knockout mice had significantly higher NAE levels, but not 1-AG or 2-AG levels than the wild-type mice. This experiment provides new evidence to support the hypothesis that FAAH plays an essential role in NAE degradation. However, FAAH does not mediate 1-AG and 2-AG degradation suggesting FAAH may be NAE specific. Thus, FAAH, as a drug target for chemical inhibitor development since 1994 (Koutek et al. 1994), our results suggest that the inhibition of FAAH could lead to a more complex spectrum, which suggests that targeting brain FAAH could impair the production of NAEs in response to brain injuries.

124 6.3 The effect of dietary fatty acids on NAE synthesis with/without NAPE-PLD Fatty acids can be converted to NAE via NAPE-PLD. Evidence showed that modulation of dietary fatty acid composition could adjust their corresponding NAE levels, which in turn impact on physiological outcomes. However, no reports existed on tissue NAE levels when NAPE-PLD is deficient during dietary modulation. This thesis examined the effect of different dietary fatty acid composition in isocaloric fat diets on NAE levels in wild-type (C57BL/6), heterozygous (NAPE-PLD -/+) and NAPE-PLD (-/-) mice from weaning (21 days old) until 12 weeks of age. This thesis showed that the dietary fatty acids were reflected in the animal tissue fatty acid composition, independent of the genotype. Also, a diet enriched with LA increased tissue ARA, AEA, 1-AG, and 2-AG levels, while a diet enriched with DHA increased tissue. DHA and DHEA levels, even when NAPE-PLD was absent. Also, NAPE-PLD knockout mice fed fish oil diet had higher brain DHEA levels than the wild-type mice fed the same diet. In liver and jejunum AEA and 1, 2-AG levels were higher in the corn oil diet. The results suggested that NAPE-PLD does not affect tissue fatty acid composition, and is not necessary for converting dietary fats to NAEs. Thus, in addition to NAPE-PLD, other parallel pathways are spontaneously involved in the biosynthesis of NAEs. Notably, the fat percentage used in study 1 is similar to American Heart Association recommendations, where saturated fat should be no more than 5- 6 % of total daily calories and replaced with unsaturated oils (American Heart Association Nutrition et al. 2006, Go et al. 2014, Mozaffarian et al. 2015, Mozaffarian et al. 2016). Thus, it is also achievable for humans via modulating dietary fatty acid composition to alter NAE levels.

6.4 The effect of ischemia on NAE degradation with/without FAAH The literature review in this thesis suggested that living organisms may have lower NAE levels than the post-mortem tissue NAE levels due to ischemia effects (Brose et al. 2016, Bazinet et al. 2005). This thesis consistently showed basal brain NAE levels used microwave fixation. Also, this thesis specifically examined the effect of CO2-induced hypercapnia/ischemia on NAE levels with and without FAAH on NAE concentrations (study 2). Our results demonstrated that CO2- induced hypercapnia/ischemia significantly increased brain NAE levels in the wild-type mice. However, OEA and AEA may saturate in the FAAH-KO mice, which did not further elevate upon CO2-induced hypercapnia/ischemia as the wild-type mice do. Also, our study confirmed that unesterified fatty acids, NAEs, 1-AG and 2-AG levels could be affected by ischemia.

125 6.5 Strengths and limitations The overarching objective of this thesis was to determine if levels of NAEs are influenced by the absence of NAPE-PLD or FAAH upon dietary manipulation or ischemia. Accordingly, both studies were designed based on our previous research that has shown that dietary fatty acid can alter NAE levels in human and hamsters (Lin et al. 2013a, Jones et al. 2014). This thesis built on this previous knowledge. However, there are strengths and limitations in our studies.

There are three main strengths of this study design. First, this work used specific knockout models model, which allowed us to identify the biosynthesis and degradation of NAE via fatty acid more precisely and comprehensively. Chemical inhibitors are also used to study the metabolic pathways of NAEs. However, the chemical inhibitors, such as NAPE-PLD inhibitor(ARN19874) (Castellani et al. 2017) or FAAH inhibitor (BIA 10-2474, URB597 or PF- 3845) only target on specific molecules (e.g., AEA uptake, PPAR-α reduction). Also, these inhibitors only work for specific or particular treatments, such as neuronal injury, neurodegeneration or pain. However, our model can examine the overall metabolic changes in completed animal system. Our results indicated that this NAPE-PLD knockout (from the 2nd or the 3rd generation offspring generated from intercrosses of 129SvJ-C57BL/6 PLD +/-) is OEA specific, which suggest other synthetic enzymes may be in favor of other NAEs. Also, our result indicated that FAAH is necessary for OEA and AEA degradation suggesting a complicated lipid metabolism involved with FAAH. Secondly, dietary fatty acids can modulate NAE levels with the absence of NAPE-PLD indicating other parallel pathways are involved in the biosynthesis process. Third, ischemia affect NAE levels, which may be partially answered the mechanistic question on the variation of NAE levels from the literature reviews.

There are some limitations in my two studies. First, a mouse model was used for my experiment, particularly multiple knockout models. However, Chapter 2 thoroughly reviewed the effect of dietary fatty acids on modulating levels of NAEs or MAGs in rats (Alvheim et al. 2012a, Alvheim et al. 2014b, Alvheim et al. 2013), hamsters (Lin et al. 2013b), pigs (Berger et al. 2001b) and humans (Ramsden et al. 2015b, Pintus et al. 2013b), which provided a broad spectrum of NAE levels in biosamples. In addition, the results from my two studies added new knowledge in this research area on top of existing evidence (Leung et al. 2006, Cravatt et al.

126 2001). While this is a standard approach to study metabolic pathways, the generalizability to humans is not known.

Both studies were conducted using established GC-FID and GC/LC-MS methods; fasted, microwave-euthanized, same feeding duration and sex. However, it is still impossible to conclude that the changes of NAEs, 1-AG or 2-AG were directly from dietary fatty acid modulation. Because we use the head-focused microwave to measure the basal levels of NAEs, 1-AG and 2-AG levels in the brain, we could not deliver the same technique to other organs or blood. Therefore, further studies are needed. Third, besides our Cravatt NAPE-PLD knockout line, other types of NAPE-PLD knockout lines (Tsuboi et al. 2011, Leishman et al. 2016, Liu et al. 2008) have also been generated. The difference is that, Cravatt NAPE-PLD knockout are selective to OEA and saturated-fatty acid-ethanolamides, such as PEA. However, other knockouts can reduce both OEA and AEA. Thus, it is not clear if my findings are generalizable to NAPE-PLD or are specific to the model used presently. Also, a clinical study (Wangensteen et al. 2011) identified a common haplotype in NAPE-PLD that was associated with severe obesity. To further explore the relationship between NAPE-PLD and obesity, using different NAPE-PLD models, adjust food intake will be one approach in the future studies. Fourth, it is unclear if other enzymes compensated for the absence of NAPE-PLD or FAAH via parallel pathways. Fifth, because menstrual cycle can alter AEA levels (Habayeb et al. 2004), we did not use female mice in our studies. Research in female mice is needed to fully understand the mechanism of action on NAE, 1-AG and 2-AG levels.

6.6 Significance and implications The significance of this thesis is the observation that using transgenic models can provide insight into the complexity of regulatory pathways of NAE biosynthesis and degradation. Also, understanding the effect of dietary fatty acid modulation and the effect of ischemia on NAE levels suggest other external factors could independently interfere with the conversion process. Despite some limitations of this research and that the relevance of these findings to humans is uncertain, these observations may impact on dietary recommendations in health guidelines. By understanding the action of dietary fats in the management of NAE and MAG synthesis, it will

127 be possible to demonstrate that fatty acids have health benefits that go beyond their essential nutrients role and could play an important role in lipid metabolism and health status.

Furthermore, results of this work have a direct contribution to understanding the mechanism of actions of these compounds, which could be beneficial for improving research on dietary recommendations, drug targets or disease prevention/treatments. For example, first, our work demonstrated that chronic feeding of dietary fatty acids can still elevate NAE levels with the absence of NAPE-PLD suggesting that parallel pathways are spontaneously involved in the synthesis of NAEs. Thus, our work highlights new mechanisms that could be considered when developing chemical blockers aimed at modulating NAE levels. Second, FAAH is necessary for the OEA, AEA, but not DHEA, while CO2 –induced hypercapnia/ischemia can elevate NAE, unesterified fatty acids, 1,2,-AG levels in the wild-type. The accumulated altered lipids may interrupt emostasis which may further interfere with other systems. Thus, it could be one reason why a phase one clinical trial of a FAAH inhibitor resulted in brain death. Overall, our work sheds new insight into the understanding of the biological pathway involving NAPE-PLD and FAAH, which helps increase the awareness of this complex endocannabinoid system. Our work also suggests that blocking FAAH or NAPE-PLD will affect many NAEs, yet is not a simple solution towards regulating energy balance, pain or other neurological disorders. Furthermore, the results of this research may lead to future research to consider other pathways, enzyme activities, and clinical relevance.

Specific conclusions Study 1: The absence of NAPE-PLD resulted in lower levels of OEA and a leaner phenotype, independent of dietary fatty acid modulation. However, NAPE-PLD is not necessary for NAE synthesis upon dietary fat intervention.

Study 2: The absence of FAAH resulted in higher levels of NAEs, but not AGs among all mice.

Also, FAAH was necessary for the CO2-induced hypercapnia/ischemia increases in OEA and AEA, but not DHEA.

128 6.7 Overall conclusion In conclusion, a complex regulation underpins the conversion of fatty acids to NAEs. NAPE- PLD explains little of the synthesis of NAE during dietary fat modulation, while FAAH demonstrates the essential role on NAE degradation upon the same diet, but other factors, such as CO2-induced ischemia can also contribute to the variation of NAE levels. Thus, the leaner phenotype found in NAPE-PLD knockouts is independent of dietary fatty acid composition, while preventing the raising of FAAH during ischemia will reveal basal levels of NAEs in living organisms.

129

CHAPTER 7 Future Directions

130 NAEs constitute a large and diverse group of signaling lipids, which can be biosynthesized and degraded on-demand prior to signaling (Leung et al. 2006). Understanding enzymes involved in NAE metabolism is imperative; and to that end this thesis has provided new evidence concerning levels of NAEs and AGs using the absence of NAPE-PLD or FAAH models. Also, this thesis further demonstrated that dietary fatty acids still can alter levels NAEs while NAPE-PLD was absent. Thus, the results add some molecular plausibility to the mechanism of actions found between NAEs, AGs, and lipid metabolism.

Further elucidating the mechanism of NAEs of action on enzymes’ activities upon dietary intervention is one line of research that would help to define the role of NAPE-PLD more fully. For instance, Chapter 4 was in agreement with the original study by Cravatt group (Leung et al. 2006), it has been demonstrated that NAPE-PLD (-/-) mice do not only decrease saturated or monounsaturated –NAEs, but also significantly increase saturated and monounsaturated n-acyl- NAPE levels in the decapitated brain compared to their wild-type littermates. Therefore, further mechanistic studies are warranted to examine the roles of NAPE-PLD as well as other enzymes (including n-acyl-NAPE) (Figure 7.1). Tests such as the gene expression, enzyme activity and fatty acid and NAE specific biosynthesis measurements as a function of dietary intervention would be informative.

Also, the original work by the Cravatt group, which measured levels of NAEs after brains were decapitated (Leung et al. 2006) demonstrated that NAPE-PLD (-/-) mice do not only decrease saturated or monounsaturated – derived NAEs, but also significantly increase saturated and monounsaturated n-acyl-NAPE levels in decapitated brain, compared to the littermates. Therefore, further mechanistic studies are warranted to examine the roles of NAPE-PLD as well as other enzymes (including n-acyl-NAPE) (Figure 7.1). Tests such as gene expression, enzyme activity and fatty acid and NAE specific biosynthesis measurements as a function of dietary intervention would likewise be informative. In addition, the original work by Cravatt group (Leung et al. 2006) and other groups (Leishman et al. 2016, Tsuboi et al. 2011) measured levels of NAEs after brains were decapitated. Chapter 5 and other studies have shown that ischemia can elevate levels of NAEs (Brose et al. 2016, Bazinet et al. 2005). Therefore, the basal levels of NAEs in NAPE-PLD knockout are still to be identified. Also, future studies should examine the

131 effect of ischemia on levels of NAEs and their precursor fatty acids in this Cravatt-NAPE-PLD knockout model fed isocaloric fat diet.

Additionally, other types of NAPE-PLD knockouts (Leishman et al. 2016, Tsuboi et al. 2011) can also reduce AEA and other polyunsaturated-NAEs, which this Cravatt-line cannot. It would be beneficial to examine the effect of dietary intervention on different types of NAPE-PLD knockouts and related lipid metabolism and physiological changes in the future. Moreover, Chapter 4 identified that Cravatt-NAPE-PLD knockout resulted in a leaner phenotype, independent of dietary fatty acid intervention. Also, as aforementioned, the change is NAE specific (OEA), and tissue-specific (jejunum or brain) in this NAPE-PLD knockout line. OEA has been implicated in the control of satiety (Rodriguez de Fonseca et al. 2001, Schwartz et al. 2008b) and lipid absorption (Yang et al. 2007) via the activation of PPAR-α in the small intestine (Bunger et al. 2007, Fu et al. 2007). Thus OEA signaling could serve as a fat-sensing mechanism that cooperates with insulin (Schwartz et al. 2008a) or peptide hormones, such as cholecystokinin (CCK) and peptide tyrosine tyrosine (PYY) (Igarashi et al. 2015b). In addition, a clinical study demonstrated that a common haplotype in NAPE-PLD is associated with severe obesity (Wangensteen et al. 2011). Thus, future studies should examine the association between modulation of NAE levels, particularly OEA levels, in this Cravat-line knockout with the modulation of other satiety hormones (Naslund & Hellstrom 2007) to further explore the mechanism of actions on lipid metabolism and energy balance.

Chapter 5 provides essential evidence on the role of FAAH during ischemia. Complementary to determining how FAAH affects NAE levels upon dietary intervention could add insight on top of the findings from Chapter 4. Therefore, future studies should examine the effect of dietary fatty acids on NAE levels in the FAAH knockout model. Also, FAAH can degrade multiple NAEs spontaneously, while NAAA is notably regulates the degradation of PEA (Ueda et al. 2013). NAEs can be degraded to fatty acids and ethanolamines and also be converted prostamide via COX-2 or 12-hydroperoxyeicosa-5,8,10,14-tetraenoylethanolamide via lipoxygenases. Thus, measuring the activities of NAAA, FAAH, COX-2 and lipoxygenase following dietary intervention could further explain the fatty acid specific mechanism of actions during NAE degradation (Figure 7.1).

132

Finally, our work demonstrated that dietary interventions can alter NAE or MAG levels over a small range compared to other factors: obesity vs normal body weight, chronic headache patients vs healthy participants, sample types (blood vs tissues, free pool vs esterified pool), animal species, fasting vs refeeding, ischemia, stress etc. Therefore, future work, particularly clinical work, should be well designed and controlled for multiple confounders which may potentially interfere with the research questions (Figure 7.1).

133 e 7.1. Overview of thesis for future research directions

134

CHAPTER 8 References

135

Aguirre, C. A., Castillo, V. A. and Llanos, M. N. (2015) The endocannabinoid anandamide during lactation increases body fat content and CB1 receptor levels in mice adipose tissue. Nutr Diabetes, 5, e167. Ahn, B., Soundarapandian, M. M., Sessions, H. et al. (2016) MondoA coordinately regulates skeletal myocyte lipid homeostasis and insulin signaling. J Clin Invest, 126, 3567-3579. Ahn, K., McKinney, M. K. and Cravatt, B. F. (2008) Enzymatic pathways that regulate endocannabinoid signaling in the nervous system. Chem Rev, 108, 1687-1707. Alger, B. E. (2004) Endocannabinoids: Getting the message across. Proc. Natl. Acad. Sci. U.S.A., 101, 8512-8513. Alvheim, A. R., Malde, M. K., Osei-Hyiaman, D., Hong Lin, Y., Pawlosky, R. J., Madsen, L., Kristiansen, K., Froyland, L. and Hibbeln, J. R. (2012a) Dietary linoleic acid elevates endogenous 2-AG and anandamide and induces obesity. Alvheim, A. R., Malde, M. K., Osei-Hyiaman, D., Lin, Y. H., Pawlosky, R. J., Madsen, L., Kristiansen, K., Froyland, L. and Hibbeln, J. R. (2012b) Dietary linoleic acid elevates endogenous 2-AG and anandamide and induces obesity. Obesity (Silver Spring), 20, 1984-1994. Alvheim, A. R., Torstensen, B. E., Lin, Y. H., Lillefosse, H. H., Lock, E. J., Madsen, L., Froyland, L., Hibbeln, J. R. and Malde, M. K. (2014a) Dietary linoleic acid elevates the endocannabinoids 2- AG and anandamide and promotes weight gain in mice fed a low fat diet. Lipids, 49, 59-69. Alvheim, A. R., Torstensen, B. E., Lin, Y. H., Lillefosse, H. H., Lock, E. J., Madsen, L., Froyland, L., Hibbeln, J. R. and Malde, M. K. (2014b) Dietary linoleic acid elevates the endocannabinoids 2- AG and anandamide and promotes weight gain in mice fed a low fat diet. Alvheim, A. R., Torstensen, B. E., Lin, Y. H., Lillefosse, H. H., Lock, E. J., Madsen, L., Hibbeln, J. R. and Malde, M. K. (2013) Dietary linoleic acid elevates endogenous 2-arachidonoylglycerol and anandamide in Atlantic salmon (Salmo salar L.) and mice, and induces weight gain and inflammation in mice. Br J Nutr, 109, 1508-1517. Amoako, A. A., Marczylo, T. H., Lam, P. M., Willets, J. M., Derry, A., Elson, J. and Konje, J. C. (2010) Quantitative analysis of anandamide and related acylethanolamides in human seminal plasma by ultra performance liquid chromatography tandem mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci, 878, 3231-3237. Artmann, A., Petersen, G., Hellgren, L. I., Boberg, J., Skonberg, C., Nellemann, C., Hansen, S. H. and Hansen, H. S. (2008a) Influence of dietary fatty acids on endocannabinoid and N- acylethanolamine levels in rat brain, liver and small intestine. Biochim Biophys Acta, 1781, 200- 212. Artmann, A., Petersen, G., Hellgren, L. I., Boberg, J., Skonberg, C., Nellemann, C., Hansen, S. H. and Hansen, H. S. (2008b) Influence of dietary fatty acids on endocannabinoid and N- acylethanolamine levels in rat brain, liver and small intestine. Avraham, Y., Saidian, M., Burston, J. J. et al. (2011) Fish oil promotes survival and protects against cognitive decline in severely undernourished mice by normalizing satiety signals. J nutr biochem, 22, 766-776. Balvers, M. G., Wortelboer, H. M., Witkamp, R. F. and Verhoeckx, K. C. (2013) Liquid chromatography- tandem mass spectrometry analysis of free and esterified fatty acid N-acyl ethanolamines in plasma and blood cells. Anal Biochem, 434, 275-283. Banni, S., Carta, G., Murru, E. et al. (2011) Krill oil significantly decreases 2-arachidonoylglycerol plasma levels in obese subjects. Nutr Metab (Lond), 8, 7. Banni, S. and Di Marzo, V. (2010) Effect of dietary fat on endocannabinoids and related mediators: consequences on energy homeostasis, inflammation and mood. Mol Nutr Food Res, 54, 82-92. Batetta, B., Griinari, M., Carta, G. et al. (2009) Endocannabinoids may mediate the ability of (n-3) fatty acids to reduce ectopic fat and inflammatory mediators in obese Zucker rats. J Nutr, 139, 1495- 1501.

136 “Love me or hate, both are in my favour. If you love me, I’ll always be in your heart; If you hate me, I’ll always be in your mind.”-William Shakespeare Bazan, N. G., Allan, G. and Rodriguez de Turco, E. B. (1993) Role of phospholipase A2 and membrane- derived lipid second messengers in membrane function and transcriptional activation of genes: implications in cerebral ischemia and neuronal excitability. Prog Brain Res, 96, 247-257. Bazan, N. G., Jr. (1970) Effects of ischemia and electroconvulsive shock on free fatty acid pool in the brain. Biochim Biophys Acta, 218, 1-10. Bazan, N. G., Musto, A. E. and Knott, E. J. (2011) Endogenous signaling by omega-3 docosahexaenoic acid-derived mediators sustains homeostatic synaptic and circuitry integrity. Mol Neurobiol, 44, 216-222. Bazinet, R. P. and Laye, S. (2014) Polyunsaturated fatty acids and their metabolites in brain function and disease. Nat Rev Neurosci, 15, 771-785. Bazinet, R. P., Lee, H. J., Felder, C. C., Porter, A. C., Rapoport, S. I. and Rosenberger, T. A. (2005) Rapid high-energy microwave fixation is required to determine the anandamide (N- arachidonoylethanolamine) concentration of rat brain. Neurochem Res, 30, 597-601. Berdyshev, E. V., Boichot, E. and Lagente, V. (1996) Anandamide--a new look on fatty acid ethanolamides. J Lipid Mediat Cell Signal, 15, 49-67. Berger, A., Crozier, G., Bisogno, T., Cavaliere, P., Innis, S. and Di Marzo, V. (2001a) Anandamide and diet: inclusion of dietary arachidonate and docosahexaenoate leads to increased brain levels of the corresponding N-acylethanolamines in piglets. Proc Natl Acad Sci U S A, 98, 6402-6406. Berger, A., Crozier, G., Bisogno, T., Cavaliere, P., Innis, S. and Di Marzo, V. (2001b) Anandamide and diet: inclusion of dietary arachidonate and docosahexaenoate leads to increased brain levels of the corresponding N-acylethanolamines in piglets.[Erratum appears in Proc Natl Acad Sci U S A 2001 Jun 19;98(13):7647]. Berger, A., Mutch, D. M., German, J. B. and Roberts, M. A. (2002) Dietary effects of arachidonate-rich fungal oil and fish oil on murine hepatic and hippocampal gene expression. Lipids Health Dis, 1, 2. Bisogno, T., Delton-Vandenbroucke, I., Milone, A., Lagarde, M. and Di Marzo, V. (1999) Biosynthesis and inactivation of N-arachidonoylethanolamine (anandamide) and N- docosahexaenoylethanolamine in bovine retina. Arch Biochem Biophys, 370, 300-307. Blancaflor, E. B., Kilaru, A., Keereetaweep, J., Khan, B. R., Faure, L. and Chapman, K. D. (2014) N- Acylethanolamines: lipid metabolites with functions in plant growth and development. Plant J, 79, 568-583. Borrelli, F. and Izzo, A. A. (2009) Role of acylethanolamides in the gastrointestinal tract with special reference to food intake and energy balance. Best Pract Res Clin Endocrinol Metab, 23, 33-49. Bradshaw, H. B. and Walker, J. M. (2005) The expanding field of cannabimimetic and related lipid mediators. Br. J. Pharmacol., 144, 459-465. Brose, S. A., Golovko, S. A. and Golovko, M. Y. (2016) Brain 2-arachidonoylglycerol levels are dramatically and rapidly increased under acute ischemia-injury which is prevented by microwave irradiation. Lipids, 51, 487-495. Brown, A. J. and Hiley, C. R. (2009) Is Gpr55 an Anandamide Receptor? Vitamins and Hormones: Anandamide an Endogenous Cannabinoid, 81, 111-137. Brown, I., Cascio, M. G., Wahle, K. W., Smoum, R., Mechoulam, R., Ross, R. A., Pertwee, R. G. and Heys, S. D. (2010) Cannabinoid receptor-dependent and -independent anti-proliferative effects of omega-3 ethanolamides in androgen receptor-positive and -negative prostate cancer cell lines. Carcinogenesis, 31, 1584-1591. Bunger, M., van den Bosch, H. M., van der Meijde, J., Kersten, S., Hooiveld, G. J. and Muller, M. (2007) Genome-wide analysis of PPARalpha activation in murine small intestine. Physiol Genomics, 30, 192-204. Cadas, H., di Tomaso, E. and Piomelli, D. (1997) Occurrence and biosynthesis of endogenous cannabinoid precursor, N-arachidonoyl phosphatidylethanolamine, in rat brain. J Neurosci, 17, 1226-1242.Calder, P. C. (2013) Omega-3 polyunsaturated fatty acids and inflammatory processes: nutrition or pharmacology? Br J Clin Pharmacol, 75, 645-662. 137 “It is time to introduce the emotional Lin and the logic Lin to meet; then you will be complete." –Dongfang Chao (Friend) Carta, G., Demurtas, D., Lisai, S., Muredda, L., Melis, S. B. M. and Barbarossa, I. T. (2015a) Evaluation of some lipid metabolism parameters in obese subjects for developing targeted nutritional strategies. Obes Surg, 25, S258-S259. Carta, G., Murru, E., Lisai, S., Sirigu, A., Piras, A., Collu, M., Batetta, B., Gambelli, L. and Banni, S. (2015b) Dietary triacylglycerols with palmitic acid in the sn-2 position modulate levels of N- acylethanolamides in rat tissues. PLoS ONE, 10, e0120424. Castellani, B., Diamanti, E., Pizzirani, D. et al. (2017) Synthesis and characterization of the first inhibitor of N-acylphosphatidylethanolamine phospholipase D (NAPE-PLD). Chem Commun (Camb), 53, 12814-12817. Cenedella, R. J., Galli, C. and Paoletti, R. (1975) Brain free fatty levels in rats sacrificed by decapitation versus focused microwave irradiation. Lipids, 10, 290-293. Chen, C. T. and Bazinet, R. P. (2015) beta-oxidation and rapid metabolism, but not uptake regulate brain eicosapentaenoic acid levels. Prostaglandins Leukot Essent Fatty Acids, 92, 33-40. Chen, C. T., Kitson, A. P., Hopperton, K. E. et al. (2015) Plasma non-esterified docosahexaenoic acid is the major pool supplying the brain. Sci Rep, 5, 15791. Cluny, N. L., Keenan, C. M., Lutz, B., Piomelli, D. and Sharkey, K. A. (2009) The identification of peroxisome proliferator-activated receptor alpha-independent effects of oleoylethanolamide on intestinal transit in mice. Neurogastroenterol Motil, 21, 420-429. Coburn, A. F. and Moore, L. V. (1943) Nutrition as a conditioning factor in the rheumatic state. Am J Dis Child, 65, 744-756. Colodzin, M., Bachur, N. R., Weissbach, H. and Udenfriend, S. (1963) Enzymatic formation of fatty acid amides of ethanolamine by rat liver microsomes. Biochem Biophys Res Commun, 10, 165-170. Coulon, D. and Bure, C. (2015) Acylphosphatidylglycerol (acyl-PG) or N-acylphosphatidylethanolamine (NAPE). J Mass Spectrom, 50, 1318-1320. Cravatt, B. F., Demarest, K., Patricelli, M. P., Bracey, M. H., Giang, D. K., Martin, B. R. and Lichtman, A. H. (2001) Supersensitivity to anandamide and enhanced endogenous cannabinoid signaling in mice lacking fatty acid amide hydrolase. Proc Natl Acad Sci U S A, 98, 9371-9376. Cravatt, B. F., Giang, D. K., Mayfield, S. P., Boger, D. L., Lerner, R. A. and Gilula, N. B. (1996) Molecular characterization of an enzyme that degrades neuromodulatory fatty-acid amides. Nature, 384, 83-87. Cravatt, B. F. and Lichtman, A. H. (2003) Fatty acid amide hydrolase: an emerging therapeutic target in the endocannabinoid system. Curr Opin Chem Biol, 7, 469-475. De Marchi, N., De Petrocellis, L., Orlando, P., Daniele, F., Fezza, F. and Di Marzo, V. (2003) Endocannabinoid signalling in the blood of patients with schizophrenia. Lipids Health Dis, 2, 5. De Petrocellis, L., Melck, D., Bisogno, T., Milone, A. and Di Marzo, V. (1999) Finding of the endocannabinoid signalling system in Hydra, a very primitive organism: possible role in the feeding response. De Petrocellis, L., Schiano Moriello, A., Imperatore, R., Cristino, L., Starowicz, K. and Di Marzo, V. (2012) A re-evaluation of 9-HODE activity at TRPV1 channels in comparison with anandamide: enantioselectivity and effects at other TRP channels and in sensory neurons. Br J Pharmacol, 167, 1643-1651. Demizieux, L., Piscitelli, F., Troy-Fioramonti, S. et al. (2016) Early low-fat diet enriched with linolenic acid reduces liver endocannabinoid tone and improves late glycemic control after a high-fat diet challenge in mice. Diabetes, 65, 1824-1837. Devane, W. A., Hanus, L., Breuer, A. et al. (1992) Isolation and structure of a brain constituent that binds to the cannabinoid receptor. Science, 258, 1946-1949. Di Marzo, V., Fontana, A., Cadas, H., Schinelli, S., Cimino, G., Schwartz, J. C. and Piomelli, D. (1994) Formation and inactivation of endogenous cannabinoid anandamide in central neurons. Nature, 372, 686-691.

138 “Success is not to reach to the sky, it is to stay in there.”-Shoug Alashmali (PhD student in Nutrition)

Diep, T. A., Madsen, A. N., Holst, B., Kristiansen, M. M., Wellner, N., Hansen, S. H. and Hansen, H. S. (2011) Dietary fat decreases intestinal levels of the anorectic lipids through a fat sensor. FASEB J, 25, 765-774. Diep, T. A., Madsen, A. N., Krogh-Hansen, S., Al-Shahwani, M., Al-Sabagh, L., Holst, B. and Hansen, H. S. (2014) Dietary non-esterified oleic Acid decreases the jejunal levels of anorectic N- acylethanolamines. PLoS ONE, 9, e100365. DiPatrizio, N. V., Astarita, G., Schwartz, G., Li, X. and Piomelli, D. (2011a) Endocannabinoid signal in the gut controls dietary fat intake. DiPatrizio, N. V., Astarita, G., Schwartz, G., Li, X. S. and Piomelli, D. (2011b) Endocannabinoid signal in the gut controls dietary fat intake. Proc. Natl. Acad. Sci. U.S.A., 108, 12904-12908. DiPatrizio, N. V., Igarashi, M., Narayanaswami, V., Murray, C., Gancayco, J., Russell, A., Jung, K. M. and Piomelli, D. (2015) Fasting stimulates 2-AG biosynthesis in the small intestine: role of cholinergic pathways. Am J Physiol Regul Integr Comp Physiol, 309, R805-813. DiPatrizio, N. V., Joslin, A., Jung, K. M. and Piomelli, D. (2013a) Endocannabinoid signaling in the gut mediates preference for dietary unsaturated fats. DiPatrizio, N. V., Joslin, A., Jung, K. M. and Piomelli, D. (2013b) Endocannabinoid signaling in the gut mediates preference for dietary unsaturated fats. FASEB J, 27, 2513-2520. Domenichiello, A. F., Kitson, A. P. and Bazinet, R. P. (2015) Is docosahexaenoic acid synthesis from alpha-linolenic acid sufficient to supply the adult brain? Prog Lipid Res, 59, 54-66. Domenichiello, A. F., Kitson, A. P., Metherel, A. H., Chen, C. T., Hopperton, K. E., Stavro, P. M. and Bazinet, R. P. (2017) Whole-body docosahexaenoic acid synthesis-secretion rates in rats are constant across a large range of dietary alpha-linolenic acid intakes. J Nutr, 147, 37-44. Engeli, S. (2012) Central and peripheral cannabinoid receptors as therapeutic targets in the control of food intake and body weight. Handb Exp Pharmacol, 357-381. Engeli, S., Lehmann, A. C., Kaminski, J., Haas, V., Janke, J., Zoerner, A. A., Luft, F. C., Tsikas, D. and Jordan, J. (2014) Influence of dietary fat intake on the endocannabinoid system in lean and obese subjects. Obesity (Silver Spring), 22, E70-76. Esposito, E. and Cuzzocrea, S. (2013) Palmitoylethanolamide in homeostatic and traumatic central nervous system injuries. CNS Neurol Disord Drug Targets, 12, 55-61. Ezzili, C., Otrubova, K. and Boger, D. L. (2010) Fatty acid amide signaling molecules. Bioorg Med Chem Lett, 20, 5959-5968. Felder, C. C., Briley, E. M., Axelrod, J., Simpson, J. T., Mackie, K. and Devane, W. A. (1993) Anandamide, an endogenous cannabimimetic eicosanoid, binds to the cloned human cannabinoid receptor and stimulates receptor-mediated signal transduction. Proc Natl Acad Sci U S A, 90, 7656-7660. Figueroa, J. D., Cordero, K., Serrano-Illan, M., Almeyda, A., Baldeosingh, K., Almaguel, F. G. and De Leon, M. (2013) Metabolomics uncovers dietary omega-3 fatty acid-derived metabolites implicated in anti-nociceptive responses after experimental spinal cord injury. Neuroscience, 255, 1-18. Folch, J., Lees, M. and Sloane Stanley, G. H. (1957) A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem, 226, 497-509. Friedman, A. N., Kim, J., Kaiser, S., Pedersen, T. L., Newman, J. W. and Watkins, B. A. (2016) Association between plasma endocannabinoids and appetite in hemodialysis patients: A pilot study. Fu, J., Astarita, G., Gaetani, S., Kim, J., Cravatt, B. F., Mackie, K. and Piomelli, D. (2007) Food intake regulates oleoylethanolamide formation and degradation in the proximal small intestine. J Biol Chem, 282, 1518-1528. Fu, J., Gaetani, S., Oveisi, F. et al. (2003) Oleylethanolamide regulates feeding and body weight through activation of the nuclear receptor PPAR-alpha. Nature, 425, 90-93.

139 “You can’t demand respect-you have to earn it.”-Anonymous

Fu, J., Oveisi, F., Gaetani, S., Lin, E. and Piomelli, D. (2005) Oleoylethanolamide, an endogenous PPAR- alpha agonist, lowers body weight and hyperlipidemia in obese rats. Neuropharmacology, 48, 1147-1153. Gao, Y., Vasilyev, D. V., Goncalves, M. B. et al. (2010) Loss of retrograde endocannabinoid signaling and reduced adult neurogenesis in diacylglycerol lipase knock-out mice. J Neurosci, 30, 2017- 2024. Gillingham, L. G., Gustafson, J. A., Han, S. Y., Jassal, D. S. and Jones, P. J. (2011) High-oleic rapeseed (canola) and flaxseed oils modulate serum lipids and inflammatory biomarkers in hypercholesterolaemic subjects. Br J Nutr, 105, 417-427. Gillum, M. P., Zhang, D., Zhang, X. M. et al. (2008a) N-acylphosphatidylethanolamine, a gut- derived circulating factor induced by fat ingestion, inhibits food intake. Cell, 135, 813-824. Gillum, M. P., Zhang, D., Zhang, X. M. et al. (2008b) N-acylphosphatidylethanolamine, a gut- derived circulating factor induced by fat ingestion, inhibits food intake. Goparaju, S. K., Ueda, N., Yamaguchi, H. and Yamamoto, S. (1998) Anandamide amidohydrolase reacting with 2-arachidonoylglycerol, another cannabinoid receptor ligand. FEBS Lett, 422, 69- 73. Gouveia-Figueira, S. and Nording, M. L. (2014) Development and validation of a sensitive UPLC-ESI- MS/MS method for the simultaneous quantification of 15 endocannabinoids and related compounds in milk and other biofluids. Guzman, M., Lo Verme, J., Fu, J., Oveisi, F., Blazquez, C. and Piomelli, D. (2004) Oleoylethanolamide stimulates lipolysis by activating the nuclear receptor peroxisome proliferator-activated receptor alpha (PPAR-alpha). J Biol Chem, 279, 27849-27854. Habayeb, O. M., Taylor, A. H., Evans, M. D., Cooke, M. S., Taylor, D. J., Bell, S. C. and Konje, J. C. (2004) Plasma levels of the endocannabinoid anandamide in women--a potential role in pregnancy maintenance and labor? J Clin Endocrinol Metab, 89, 5482-5487. Hadjipanayi, E. and Schilling, A. F. (2013) Hypoxia-based strategies for angiogenic induction: the dawn of a new era for ischemia therapy and tissue regeneration. Organogenesis, 9, 261-272. Hansen, H. H., Schmid, P. C., Bittigau, P. et al. (2001) Anandamide, but not 2-arachidonoylglycerol, accumulates during in vivo neurodegeneration. J Neurochem, 78, 1415-1427. Hansen, H. S. (2013) Effect of diet on tissue levels of palmitoylethanolamide. Hansen, H. S. (2014) Role of anorectic N-acylethanolamines in intestinal physiology and satiety control with respect to dietary fat. Pharmacol Res, 86, 18-25. Hansen, H. S. and Diep, T. A. (2009) N-acylethanolamines, anandamide and food intake. Biochemical Pharmacology, 78, 553-560. Hansen, H. S., Moesgaard, B., Hansen, H. H., Schousboe, A. and Petersen, G. (1999) Formation of N- acyl-phosphatidylethanolamine and N-acylethanolamine (including anandamide) during glutamate-induced neurotoxicity. Lipids, 34 Suppl, S327-330. Hansen, K. B., Rosenkilde, M. M., Knop, F. K., Wellner, N., Diep, T. A., Rehfeld, J. F., Andersen, U. B., Holst, J. J. and Hansen, H. S. (2011) 2-Oleoyl glycerol is a GPR119 agonist and signals GLP-1 release in humans. J Clin Endocrinol Metab, 96, E1409-1417. Harding, S. V., Rideout, T. C. and Jones, P. J. (2010) Hepatic nuclear sterol regulatory binding element protein 2 abundance is decreased and that of ABCG5 increased in male hamsters fed plant sterols. J Nutr, 140, 1249-1254. Hashimoto, M., Maekawa, M., Katakura, M., Hamazaki, K. and Matsuoka, Y. (2014) Possibility of polyunsaturated fatty acids for the prevention and treatment of neuropsychiatric illnesses. J Pharmacol Sci, 124, 294-300. Hesselink, J. M. (2013a) Chronic idiopathic axonal neuropathy and pain, treated with the endogenous lipid mediator palmitoylethanolamide: a case collection. Int Med Case Rep J, 6, 49-53. Hesselink, J. M. (2013b) Evolution in pharmacologic thinking around the natural analgesic palmitoylethanolamide: from nonspecific resistance to PPAR-alpha agonist and effective nutraceutical. J Pain Res, 6, 625-634. 140 “If your god is a good god, he wants you to be happy without hurting anyone” -Diana Sanchez Hernandaz Higgins, J. P. T., Green, S. and Cochrane Collaboration. (2008) Cochrane handbook for systematic reviews of interventions. In: Cochrane book series, pp. xxi, 649 p. Wiley-Blackwell, Chichester, England ; Hoboken, NJ. Hooijmans, C. R., Rovers, M. M., de Vries, R. B., Leenaars, M., Ritskes-Hoitinga, M. and Langendam, M. W. (2014) SYRCLE's risk of bias tool for animal studies. BMC Med Res Methodol, 14, 43. Hopperton, K. E., Duncan, R. E., Bazinet, R. P. and Archer, M. C. (2014) Fatty acid synthase plays a role in cancer metabolism beyond providing fatty acids for phospholipid synthesis or sustaining elevations in glycolytic activity. Exp Cell Res, 320, 302-310. Igarashi, M., DiPatrizio, N. V., Narayanaswami, V. and Piomelli, D. (2015a) Feeding-induced oleoylethanolamide mobilization is disrupted in the gut of diet-induced obese rodents. Biochim Biophys Acta, 1851, 1218-1226. Igarashi, M., DiPatrizio, N. V., Narayanaswami, V. and Piomelli, D. (2015b) Feeding-induced oleoylethanolamide mobilization is disrupted in the gut of diet-induced obese rodents. Bba-Mol Cell Biol L, 1851, 1218-1226. Izzo, A. A., Piscitelli, F., Capasso, R., Marini, P., Cristino, L., Petrosino, S. and Di Marzo, V. (2010) Basal and fasting/refeeding-regulated tissue levels of endogenous PPAR-alpha ligands in Zucker rats. Obesity (Silver Spring), 18, 55-62. Jacome-Sosa, M., Vacca, C., Mangat, R. et al. (2016) Vaccenic acid suppresses intestinal inflammation by increasing anandamide and related N-acylethanolamines in the JCR:LA-cp rat. J Lipid Res, 57, 638-649. Jacome-Sosa, M. M., Borthwick, F., Mangat, R. et al. (2014) Diets enriched in trans-11 vaccenic acid alleviate ectopic lipid accumulation in a rat model of NAFLD and metabolic syndrome. J nutr biochem, 25, 692-701. Jain, M., Kennedy, A. D., Elsea, S. H. and Miller, M. J. (2017) Analytes related to erythrocyte metabolism are reliable biomarkers for preanalytical error due to delayed plasma processing in metabolomics studies. Clin Chim Acta, 466, 105-111. Jones, P. J., Lin, L., Gillingham, L. G., Yang, H. and Omar, J. M. (2014) Modulation of plasma N- acylethanolamine levels and physiological parameters by dietary fatty acid composition in humans. J Lipid Res, 55, 2655-2664. Jumpertz, R., Guijarro, A., Pratley, R. E., Piomelli, D. and Krakoff, J. (2011) Central and peripheral endocannabinoids and cognate acylethanolamides in humans: association with race, adiposity, and energy expenditure. J Clin Endocrinol Metab, 96, 787-791. Karlsson, C., Hjorth, S., Karpefors, M., Hansson, G. I. and Carlsson, B. (2015) Baseline anandamide levels and body weight impact the weight loss effect of CB1 receptor antagonism in male rats. Endocrinology, 156, 1237-1241. Katan, M. B., Zock, P. L. and Mensink, R. P. (1994) Effects of fats and fatty acids on blood lipids in humans: an overview. Am J Clin Nutr, 60, 1017S-1022S. Katayama, K., Ueda, N., Katoh, I. and Yamamoto, S. (1999) Equilibrium in the hydrolysis and synthesis of cannabimimetic anandamide demonstrated by a purified enzyme. Biochim Biophys Acta, 1440, 205-214. Kim, H. Y. and Spector, A. A. (2013) Synaptamide, endocannabinoid-like derivative of docosahexaenoic acid with cannabinoid-independent function. Prostaglandins Leukot Essent Fatty Acids, 88, 121- 125. Kim, H. Y., Spector, A. A. and Xiong, Z. M. (2011) A synaptogenic amide N- docosahexaenoylethanolamide promotes hippocampal development. Prostaglandins Other Lipid Mediat, 96, 114-120. Kim, J., Carlson, M. E., Kuchel, G. A., Newman, J. W. and Watkins, B. A. (2016) Dietary DHA reduces downstream endocannabinoid and inflammatory gene expression and epididymal fat mass while improving aspects of glucose use in muscle in C57BL/6J mice. Int J Obes (Lond), 40, 129-137. Köfalvi, A. (2008) Cannabinoids and the brain, chapter 2. pp. xv, 583 p. Springer, New York.

141 “The great things will happen when intuitively unrelated things combined.”-Livia Li (Undergraduate)

Kuehl, F. A., Jacob, T. A., Ganley, O. H., Ormond, R. E. and Meisinger, M. A. P. (1957) The Identification of N-(2-Hydroxyethyl)-Palmitamide as a Naturally Occurring Anti-Inflammatory Agent. J Am Chem Soc, 79, 5577-5578. Kurahashi, Y., Ueda, N., Suzuki, H., Suzuki, M. and Yamamoto, S. (1997) Reversible hydrolysis and synthesis of anandamide demonstrated by recombinant rat fatty-acid amide hydrolase. Biochem Biophys Res Commun, 237, 512-515. Lambert, D. M., Vandevoorde, S., Jonsson, K. O. and Fowler, C. J. (2002) The palmitoylethanolamide family: a new class of anti-inflammatory agents? Curr Med Chem, 9, 663-674. Lee, J. W., Huang, B. X., Kwon, H., Rashid, M. A., Kharebava, G., Desai, A., Patnaik, S., Marugan, J. and Kim, H. Y. (2016) Orphan GPR110 (ADGRF1) targeted by N-docosahexaenoylethanolamine in development of neurons and cognitive function. Nat Commun, 7, 13123. Leishman, E., Mackie, K., Luquet, S. and Bradshaw, H. B. (2016) Lipidomics profile of a NAPE-PLD KO mouse provides evidence of a broader role of this enzyme in lipid metabolism in the brain. Biochim Biophys Acta, 1861, 491-500. Leung, D., Saghatelian, A., Simon, G. M. and Cravatt, B. F. (2006) Inactivation of N-acyl phosphatidylethanolamine phospholipase D reveals multiple mechanisms for the biosynthesis of endocannabinoids. Biochemistry-Us, 45, 4720-4726. Leweke, F. M., Schwarz, E., Grosshans, M. et al. (2015) Oylethanolamide modulates human neural responses to food stimuli in obesity. Biological Psychiatry, 77, 90s-91s. Liisberg, U., Fauske, K. R., Kuda, O. et al. (2016) Intake of a Western diet containing cod instead of pork alters fatty acid composition in tissue phospholipids and attenuates obesity and hepatic lipid accumulation in mice. J nutr biochem, 33, 119-127. Lima, P., Machado, F. and Coimbra, C. (2014) Thermal effect induced by anandamide is dissociated of locomotor and feeding behavior. Faseb Journal, 28. Lin, L., Metherel, A. H., Jones, P. J. and Bazinet, R. P. (2017) Fatty acid amide hydrolase (FAAH) regulates hypercapnia/ischemia-induced increases in n-acylethanolamines in mouse brain. J Neurochem, 142, 662-671. Lin, L., Rideout, T., Yurkova, N., Yang, H., Eck, P. and Jones, P. J. (2013a) Fatty acid ethanolamides modulate CD36-mRNA through dietary fatty acid manipulation in Syrian Golden hamsters. Appl Physiol Nutr Metab, 38, 870-878. Lin, L., Rideout, T., Yurkova, N., Yang, H., Eck, P. and Jones, P. J. (2013b) Fatty acid ethanolamides modulate CD36-mRNA through dietary fatty acid manipulation in Syrian Golden hamsters. Lin, L., Yang, H. and Jones, P. J. (2012) Quantitative analysis of multiple fatty acid ethanolamides using ultra-performance liquid chromatography-tandem mass spectrometry. Prostaglandins Leukot Essent Fatty Acids, 87, 189-195. Liu, J., Wang, L., Harvey-White, J. et al. (2008) Multiple pathways involved in the biosynthesis of anandamide. Neuropharmacology, 54, 1-7. Lo Van, A., Sakayori, N., Hachem, M., Belkouch, M., Picq, M., Lagarde, M., Osumi, N. and Bernoud- Hubac, N. (2016) Mechanisms of DHA transport to the brain and potential therapy to neurodegenerative diseases. Biochimie, 130, 163-167. Lo Verme, J., Fu, J., Astarita, G., La Rana, G., Russo, R., Calignano, A. and Piomelli, D. (2005) The nuclear receptor peroxisome proliferator-activated receptor-alpha mediates the anti-inflammatory actions of palmitoylethanolamide. Mol Pharmacol, 67, 15-19. LoVerme, J., La Rana, G., Russo, R., Calignano, A. and Piomelli, D. (2005) The search for the palmitoylethanolamide receptor. Life Sci, 77, 1685-1698. Lu, H. C. and Mackie, K. (2016) An Introduction to the Endogenous Cannabinoid System. Biological Psychiatry, 79, 516-525. Lunt, G. G. and Rowe, C. E. (1968) The production of unesterified fatty acid in brain. Biochim Biophys Acta, 152, 681-693. Lusk, G. (1924) Animal calorimetry. Twenty-fourth paper. Analysis of the oxidation of mixtures of carbohydrate and fat. A correction. Journal of Biological Chemistry, 59, 41-42. 142 "Do it or not to do it, never just try it.”-Paraskevi Massara (PhD student in nutrition)

Maccarrone, M., Attina, M., Bari, M., Cartoni, A., Ledent, C. and Finazzi-Agro, A. (2001) Anandamide degradation and N-acylethanolamines level in wild-type and CB1 cannabinoid receptor knockout mice of different ages. J Neurochem, 78, 339-348. MacIntosh, B. A., Ramsden, C. E., Faurot, K. R., Zamora, D., Mangan, M., Hibbeln, J. R. and Mann, J. D. (2013) Low-n-6 and low-n-6 plus high-n-3 diets for use in clinical research. Br J Nutr, 110, 559-568. Maki, K. C., Reeves, M. S., Farmer, M., Griinari, M., Berge, K., Vik, H., Hubacher, R. and Rains, T. M. (2009) Krill oil supplementation increases plasma concentrations of eicosapentaenoic and docosahexaenoic acids in overweight and obese men and women. Nutr Res, 29, 609-615. Mallet, C., Dubray, C. and Duale, C. (2016) FAAH inhibitors in the limelight, but regrettably. Int J Clin Pharmacol Ther, 54, 498-501. Martins, S. V., Lopes, P. A., Alves, S. P., Alfaia, C. M., Castro, M. F., Bessa, R. J. and Prates, J. A. (2012) Dietary CLA combined with palm oil or ovine fat differentially influences fatty acid deposition in tissues of obese Zucker rats. Lipids, 47, 47-58. Matias, I., Carta, G., Murru, E., Petrosino, S., Banni, S. and Di Marzo, V. (2008) Effect of polyunsaturated fatty acids on endocannabinoid and N-acyl-ethanolamine levels in mouse adipocytes. Biochim Biophys Acta, 1781, 52-60. Meijerink, J., Balvers, M. and Witkamp, R. (2013) N-Acyl amines of docosahexaenoic acid and other n-3 polyunsatured fatty acids - from fishy endocannabinoids to potential leads. Br J Pharmacol, 169, 772-783. Mele, M., Contarini, G., Cercaci, L. et al. (2011) Enrichment of Pecorino cheese with conjugated linoleic acid by feeding dairy ewes with extruded linseed: Effect on fatty acid and triglycerides composition and on oxidative stability. Int Dairy J, 21, 365-372. Mennella, I., Ferracane, R., Zucco, F., Fogliano, V. and Vitaglione, P. (2015a) Food liking enhances the plasma response of 2-arachidonoylglycerol and of pancreatic polypeptide upon modified sham feeding in humans. J Nutr, 145, 2169-2175. Mennella, I., Savarese, M., Ferracane, R., Sacchi, R. and Vitaglione, P. (2015b) Oleic acid content of a meal promotes oleoylethanolamide response and reduces subsequent energy intake in humans. Food Funct, 6, 204-210. Metherel, A. H., Domenichiello, A. F., Kitson, A. P., Lin, Y. H. and Bazinet, R. P. (2017a) Serum n-3 Tetracosapentaenoic Acid and Tetracosahexaenoic Acid Increase Following Higher Dietary alpha-Linolenic Acid but not Docosahexaenoic Acid. Lipids, 52, 167-172. Metherel, A. H., Kitson, A. P., Domenichiello, A. F., Lacombe, R. J. S., Hopperton, K. E., Trepanier, M. O., Alashmali, S. M., Lin, L. and Bazinet, R. P. (2017b) Maternal liver docosahexaenoic acid (DHA) stores are increased via higher serum unesterified DHA uptake in pregnant long Evans rats. J nutr biochem, 46, 143-150. Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G. and Group, P. (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. J Clin Epidemiol, 62, 1006-1012. Murataeva, N., Dhopeshwarkar, A., Yin, D., Mitjavila, J., Bradshaw, H., Straiker, A. and Mackie, K. (2016) Where's my entourage? The curious case of 2-oleoylglycerol, 2-linolenoylglycerol, and 2- palmitoylglycerol. Pharmacol Res, 110, 173-180. Murphy, E. J. (2010) Brain fixation for analysis of brain lipid-mediators of signal transduction and brain eicosanoids requires head-focused microwave irradiation: an historical perspective. Prostaglandins Other Lipid Mediat, 91, 63-67. Nakane, S., Oka, S., Arai, S., Waku, K., Ishima, Y., Tokumura, A. and Sugiura, T. (2002) 2- Arachidonoyl-sn-glycero-3-phosphate, an arachidonic acid-containing lysophosphatidic acid: occurrence and rapid enzymatic conversion to 2-arachidonoyl-sn-glycerol, a cannabinoid receptor ligand, in rat brain. Arch Biochem Biophys, 402, 51-58. Naslund, E. and Hellstrom, P. M. (2007) Appetite signaling: from gut peptides and enteric nerves to brain. Physiol Behav, 92, 256-262.

143 " “I am reinventing myself in the journey call life." Anamika Ray (MSc in forestry, PhD in plant science)

Nomura, D. K., Morrison, B. E., Blankman, J. L. et al. (2011) Endocannabinoid hydrolysis generates brain prostaglandins that promote neuroinflammation. Science, 334, 809-813. Oka, S., Yanagimoto, S., Ikeda, S., Gokoh, M., Kishimoto, S., Waku, K., Ishima, Y. and Sugiura, T. (2005) Evidence for the involvement of the cannabinoid CB2 receptor and its endogenous ligand 2-arachidonoylglycerol in 12-O-tetradecanoylphorbol-13-acetate-induced acute inflammation in mouse ear. J Biol Chem, 280, 18488-18497. Okamoto, Y., Wang, J., Morishita, J. and Ueda, N. (2007) Biosynthetic pathways of the endocannabinoid anandamide. Chem Biodivers, 4, 1842-1857. Orr, S. K., Palumbo, S., Bosetti, F., Mount, H. T., Kang, J. X., Greenwood, C. E., Ma, D. W., Serhan, C. N. and Bazinet, R. P. (2013) Unesterified docosahexaenoic acid is protective in neuroinflammation. J Neurochem, 127, 378-393. Osei-Hyiaman, D., Depetrillo, M., Harvey-White, J., Bannon, A. W., Cravatt, B. F., Kuhar, M. J., Mackie, K., Palkovits, M. and Kunos, G. (2005a) Cocaine- and amphetamine-related transcript is involved in the orexigenic effect of endogenous anandamide. Neuroendocrinology, 81, 273-282. Osei-Hyiaman, D., DePetrillo, M., Pacher, P. et al. (2005b) Endocannabinoid activation at hepatic CB1 receptors stimulates fatty acid synthesis and contributes to diet-induced obesity. Journal of Clinical Investigation, 115, 1298-1305. Oveisi, F., Gaetani, S., Eng, K. T. and Piomelli, D. (2004) Oleoylethanolamide inhibits food intake in free-feeding rats after oral administration. Pharmacol Res, 49, 461-466. Overton, H. A., Babbs, A. J., Doel, S. M. et al. (2006) Deorphanization of a G protein-coupled receptor for oleoylethanolamide and its use in the discovery of small-molecule hypophagic agents. Cell Metab, 3, 167-175. Pacher, P., Batkai, S. and Kunos, G. (2006) The endocannabinoid system as an emerging target of pharmacotherapy. Pharmacol Rev, 58, 389-462. Patel, S., Carrier, E. J., Ho, W. S., Rademacher, D. J., Cunningham, S., Reddy, D. S., Falck, J. R., Cravatt, B. F. and Hillard, C. J. (2005) The postmortal accumulation of brain N- arachidonylethanolamine (anandamide) is dependent upon fatty acid amide hydrolase activity. J Lipid Res, 46, 342-349. Patterson, A. C. and Stark, K. D. (2008) Direct determinations of the fatty acid composition of daily dietary intakes incorporating nutraceuticals and functional food strategies to increase n-3 highly unsaturated fatty acids. J Am Coll Nutr, 27, 538-546. Pawlosky, R. J., Sprecher, H. W. and Salem, N., Jr. (1992) High sensitivity negative ion GC-MS method for detection of desaturated and chain-elongated products of deuterated linoleic and linolenic acids. J Lipid Res, 33, 1711-1717. Pedraz, M., Araos, P., Garcia-Marchena, N. et al. (2015) Sex differences in psychiatric comorbidity and plasma biomarkers for cocaine addiction in abstinent cocaine-addicted subjects in outpatient settings. Front Psychiatry, 6, 17. Pertwee, R. G. (2007) GPR55: a new member of the cannabinoid receptor clan? Br J Pharmacol, 152, 984-986. Pertwee, R. G. (2015) Endocannabinoids. In: Handbook of experimental pharmacology;, pp. xii, 475 p. Springer, Cham ; New York. Petersen, G., Sorensen, C., Schmid, P. C., Artmann, A., Tang-Christensen, M., Hansen, S. H., Larsen, P. J., Schmid, H. H. and Hansen, H. S. (2006a) Intestinal levels of anandamide and oleoylethanolamide in food-deprived rats are regulated through their precursors. Biochim Biophys Acta, 1761, 143-150; discussion 141-142. Petersen, G., Sorensen, C., Schmid, P. C., Artmann, A., Tang-Christensen, M., Hansen, S. H., Larsen, P. J., Schmid, H. H. and Hansen, H. S. (2006b) Intestinal levels of anandamide and oleoylethanolamide in food-deprived rats are regulated through their precursors.

144 “In Great endeavours, it is imperative to understand that your success depends on your willpower to keep fighting, even when everything seems unattainable.” Tad Ferreira (PhD student in statistics) Pintus, S., Murru, E., Carta, G. et al. (2013a) Sheep cheese naturally enriched in alpha-linolenic, conjugated linoleic and vaccenic acids improves the lipid profile and reduces anandamide in the plasma of hypercholesterolaemic subjects. Br J Nutr, 109, 1453-1462. Pintus, S., Murru, E., Carta, G. et al. (2013b) Sheep cheese naturally enriched in alpha-linolenic, conjugated linoleic and vaccenic acids improves the lipid profile and reduces anandamide in the plasma of hypercholesterolaemic subjects. Piomelli, D. (2013) A fatty gut feeling. Trends Endocrinol Metab, 24, 332-341. Piras, A., Carta, G., Murru, E., Lopes, P. A., Martins, S. V., Prates, J. A. and Banni, S. (2015) Effects of dietary CLA on n-3 PUFA score and n-acylethanolamides biosynthesis in the liver of obese Zucker rats. Prostaglandins Leukot Essent Fatty Acids, 98, 15-19. Piscitelli, F., Carta, G., Bisogno, T. et al. (2011) Effect of dietary krill oil supplementation on the endocannabinoidome of metabolically relevant tissues from high-fat-fed mice. Nutr Metab (Lond), 8, 51. Pollock, J. M., Deibler, A. R., Whitlow, C. T., Tan, H., Kraft, R. A., Burdette, J. H. and Maldjian, J. A. (2009) Hypercapnia-induced cerebral hyperperfusion: an underrecognized clinical entity. AJNR Am J Neuroradiol, 30, 378-385. Porter, A. C. and Felder, C. C. (2001) The endocannabinoid nervous system: unique opportunities for therapeutic intervention. Pharmacol Therapeut, 90, 45-60. Pu, S., Eck, P., Jenkins, D. J., Connelly, P. W., Lamarche, B., Kris-Etherton, P. M., West, S. G., Liu, X. and Jones, P. J. (2016a) Interactions between dietary oil treatments and genetic variants modulate fatty acid ethanolamides in plasma and body weight composition. Pu, S. H., Eck, P., Jenkins, D. J. A., Connelly, P. W., Lamarche, B., Kris-Etherton, P. M., West, S. G., Liu, X. R. and Jones, P. J. H. (2016b) Interactions between dietary oil treatments and genetic variants modulate fatty acid ethanolamides in plasma and body weight composition. Brit J Nutr, 115, 1012-1023. Ramsden, C. E., Faurot, K. R., Zamora, D. et al. (2013) Targeted alteration of dietary n-3 and n-6 fatty acids for the treatment of chronic headaches: a randomized trial. Pain, 154, 2441-2451. Ramsden, C. E., Zamora, D., Makriyannis, A. et al. (2015a) Diet-induced changes in n-3- and n-6-derived endocannabinoids and reductions in headache pain and psychological distress. J Pain, 16, 707- 716. Ramsden, C. E., Zamora, D., Makriyannis, A. et al. (2015b) Diet-induced changes in n-3- and n-6-derived endocannabinoids and reductions in headache pain and psychological distress. Reeves, P. G., Rossow, K. L. and Lindlauf, J. (1993) Development and testing of the AIN-93 purified diets for rodents: results on growth, kidney calcification and bone mineralization in rats and mice. J Nutr, 123, 1923-1931. Richardson, D., Ortori, C. A., Chapman, V., Kendall, D. A. and Barrett, D. A. (2007) Quantitative profiling of endocannabinoids and related compounds in rat brain using liquid chromatography- tandem electrospray ionization mass spectrometry. Anal. Biochem., 360, 216-226. Rodriguez de Fonseca, F., Navarro, M., Gomez, R. et al. (2001) An anorexic lipid mediator regulated by feeding. Nature, 414, 209-212. Rohrig, W., Waibel, R., Perlwitz, C., Pischetsrieder, M. and Hoch, T. (2016) Identification of the oleic acid ethanolamide (OEA) isomer cis-vaccenic acid ethanolamide (VEA) as a highly abundant 18:1 fatty acid ethanolamide in blood plasma from rats and humans. Anal Bioanal Chem, 408, 6141-6151. Rossmeisl, M., Jilkova, Z. M., Kuda, O. et al. (2012) Metabolic Effects of n-3 PUFA as Phospholipids Are Superior to Triglycerides in Mice Fed a High-Fat Diet: Possible Role of Endocannabinoids. PLoS ONE, 7. Ryberg, E., Larsson, N., Sjogren, S. et al. (2007) The orphan receptor GPR55 is a novel cannabinoid receptor. Br J Pharmacol, 152, 1092-1101.

145 “Do what you can’t.”-Gursimran Singh Sethi (MSc student in electrical and computer engineering)

Salter, A. M. and Tarling, E. J. (2007) Regulation of gene transcription by fatty acids. Animal, 1, 1314- 1320. Schmid, H. H., Schmid, P. C. and Natarajan, V. (1990) N-acylated glycerophospholipids and their derivatives. Prog Lipid Res, 29, 1-43. Schmid, P. C., Schwartz, K. D., Smith, C. N., Krebsbach, R. J., Berdyshev, E. V. and Schmid, H. H. O. (2000) A sensitive endocannabinoid assay. The simultaneous analysis of N-acylethanolamines and 2-monoacylglycerols. Chem Phys Lipids, 104, 185-191. Schmid, P. C., Zuzarte-Augustin, M. L. and Schmid, H. H. (1985) Properties of rat liver N- acylethanolamine amidohydrolase. J Biol Chem, 260, 14145-14149. Schmidt, A., Brune, K. and Hinz, B. (2006) Determination of the endocannabinoid anandamide in human plasma by high-performance liquid chromatography. Biomed Chromatogr, 20, 336-342. Schreiber, D., Harlfinger, S., Nolden, B. M. et al. (2007) Determination of anandamide and other fatty acyl ethanolamides in human serum by electrospray tandem mass spectrometry. Anal Biochem, 361, 162-168. Schuel, H., Burkman, L. J., Lippes, J., Crickard, K., Forester, E., Piomelli, D. and Giuffrida, A. (2002) N- Acylethanolamines in human reproductive fluids. Chem Phys Lipids, 121, 211-227. Schwartz, G. J., Fu, J., Astarita, G., Li, X., Gaetani, S., Campolongo, P., Cuomo, V. and Piomelli, D. (2008a) The lipid messenger OEA links dietary fat intake to satiety. Cell Metab, 8, 281-288. Schwartz, G. J., Fu, J., Astarita, G., Li, X., Gaetani, S., Campolongo, P., Cuomo, V. and Piomelli, D. (2008b) The lipid messenger OEA links dietary fat intake to satiety. Senanayake, V. K., Pu, S., Jenkins, D. A. et al. (2014) Plasma fatty acid changes following consumption of dietary oils containing n-3, n-6, and n-9 fatty acids at different proportions: preliminary findings of the Canola Oil Multicenter Intervention Trial (COMIT). Trials, 15, 136. Simon, G. M. and Cravatt, B. F. (2006) Endocannabinoid biosynthesis proceeding through glycerophospho-N-acyl ethanolamine and a role for alpha/beta-hydrolase 4 in this pathway. J Biol Chem, 281, 26465-26472. Simon, G. M. and Cravatt, B. F. (2010) Characterization of mice lacking candidate N-acyl ethanolamine biosynthetic enzymes provides evidence for multiple pathways that contribute to endocannabinoid production in vivo. Mol Biosyst, 6, 1411-1418. Song, B. J., Elbert, A., Rahman, T., Orr, S. K., Chen, C. T., Febbraio, M. and Bazinet, R. P. (2010) Genetic Ablation of CD36 Does not Alter Mouse Brain Polyunsaturated Fatty Acid Concentrations. Lipids, 45, 291-299. Sugiura, T., Kishimoto, S., Oka, S. and Gokoh, M. (2006) Biochemistry, pharmacology and physiology of 2-arachidonoylglycerol, an endogenous cannabinoid receptor ligand. Prog Lipid Res, 45, 405- 446. Sugiura, T., Kodaka, T., Nakane, S. et al. (1999) Evidence that the cannabinoid CB1 receptor is a 2- arachidonoylglycerol receptor - Structure-activity relationship of 2-arachidonoylglycerol ether- linked analogues, and related compounds. Journal of Biological Chemistry, 274, 2794-2801. Sugiura, T., Kondo, S., Sukagawa, A., Nakane, S., Shinoda, A., Itoh, K., Yamashita, A. and Waku, K. (1995) 2-Arachidonoylglycerol: a possible endogenous cannabinoid receptor ligand in brain. Biochem Biophys Res Commun, 215, 89-97. Sun, Y. X., Tsuboi, K., Okamoto, Y., Tonai, T., Murakami, M., Kudo, I. and Ueda, N. (2004) Biosynthesis of anandamide and N-palmitoylethanolamine by sequential actions of phospholipase A2 and lysophospholipase D. Biochem J, 380, 749-756. Tanimura, A., Yamazaki, M., Hashimotodani, Y. et al. (2010) The endocannabinoid 2- arachidonoylglycerol produced by diacylglycerol lipase alpha mediates retrograde suppression of synaptic transmission. Neuron, 65, 320-327. Thabuis, C., Tissot-Favre, D., Bezelgues, J. B., Martin, J. C., Cruz-Hernandez, C., Dionisi, F. and Destaillats, F. (2008) Biological functions and metabolism of oleoylethanolamide. Lipids, 43, 887-894.

146 “At the darkest part of my journey, I found my peace; it was hiding in the sound of silence.” -Lin Lin Tortoriello, G., Rhodes, B. P., Takacs, S. M. et al. (2013) Targeted lipidomics in Drosophila melanogaster identifies novel 2-monoacylglycerols and N-acyl amides. PLoS ONE, 8, e67865. Trepanier, M. O., Eiden, M., Morin-Rivron, D., Bazinet, R. P. and Masoodi, M. (2017) High-resolution lipidomics coupled with rapid fixation reveals novel ischemia-induced signaling in the rat neurolipidome. J Neurochem, 140, 766-775. Tsuboi, K., Okamoto, Y., Ikematsu, N. et al. (2011) Enzymatic formation of N-acylethanolamines from N-acylethanolamine plasmalogen through N-acylphosphatidylethanolamine-hydrolyzing phospholipase D-dependent and -independent pathways. Biochim Biophys Acta, 1811, 565-577. Tsutsumi, T., Kobayashi, T., Ueda, H., Yamauchi, E., Watanabe, S. and Okuyama, H. (1994) Lysophosphoinositide-specific phospholipase-C in rat-brain synaptic plasma-membranes. Neurochem Res, 19, 399-406. Tsuyama, S., Oikawa, D., Tsuji, Y., Akimoto, Y., Jikuya, H. and Furuse, M. (2009) Dietary conjugated linoleic acid modifies the brain endocannabinoid system in mice. Nutr Neurosci, 12, 155-159. Ueda, H., Kobayashi, T., Kishimoto, M., Tsutsumi, T. and Okuyama, H. (1993a) A Possible Pathway of Phosphoinositide Metabolism through Edta-Insensitive Phospholipase-a(1) Followed by Lysophosphoinositide-Specific Phospholipase-C in Rat-Brain. J. Neurochem., 61, 1874-1881. Ueda, H., Kobayashi, T., Kishimoto, M., Tsutsumi, T. and Okuyama, H. (1993b) A possible pathway of phosphoinositide metabolism through EDTA-insensitive phospholipase A1 followed by lysophosphoinositide-specific phospholipase C in rat brain. J Neurochem, 61, 1874-1881. Ueda, N., Kurahashi, Y., Yamamoto, S. and Tokunaga, T. (1995) Partial purification and characterization of the porcine brain enzyme hydrolyzing and synthesizing anandamide. J Biol Chem, 270, 23823- 23827. Ueda, N., Puffenbarger, R. A., Yamamoto, S. and Deutsch, D. G. (2000) The fatty acid amide hydrolase (FAAH). Chem Phys Lipids, 108, 107-121. Ueda, N., Tsuboi, K. and Uyama, T. (2010) Enzymological studies on the biosynthesis of N- acylethanolamines. Bba-Mol Cell Biol L, 1801, 1274-1285. Ueda, N., Tsuboi, K. and Uyama, T. (2013) Metabolism of endocannabinoids and related N- acylethanolamines: canonical and alternative pathways. FEBS J, 280, 1874-1894. Vaughn, L. K., Denning, G., Stuhr, K. L., de Wit, H., Hill, M. N. and Hillard, C. J. (2010) Endocannabinoid signalling: has it got rhythm? Br J Pharmacol, 160, 530-543. Verhoeckx, K. C., Voortman, T., Balvers, M. G., Hendriks, H. F., H, M. W. and Witkamp, R. F. (2011) Presence, formation and putative biological activities of N-acyl serotonins, a novel class of fatty- acid derived mediators, in the intestinal tract. Biochim Biophys Acta, 1811, 578-586. Vogeser, M., Hauer, D., Christina Azad, S., Huber, E., Storr, M. and Schelling, G. (2006) Release of anandamide from blood cells. Clin Chem Lab Med, 44, 488-491. Wang, J. and Ueda, N. (2009) Biology of endocannabinoid synthesis system. Prostaglandins Other Lipid Mediat, 89, 112-119. Wang, W. Q. and Gustafson, A. (1992) One-dimensional thin-layer chromatographic separation of phospholipids and lysophospholipids from tissue lipid extracts. J Chromatogr, 581, 139-142. Wangensteen, T., Akselsen, H., Holmen, J., Undlien, D. and Retterstol, L. (2011) A common haplotype in NAPEPLD is associated with severe obesity in a Norwegian population-based cohort (the HUNT study). Obesity (Silver Spring), 19, 612-617. Watanabe, S., Doshi, M. and Hamazaki, T. (2003) n-3 Polyunsaturated fatty acid (PUFA) deficiency elevates and n-3 PUFA enrichment reduces brain 2-arachidonoylglycerol level in mice. Prostaglandins Leukot Essent Fatty Acids, 69, 51-59. Weiser, M. J., Butt, C. M. and Mohajeri, M. H. (2016) Docosahexaenoic acid and cognition throughout the lifespan. Nutrients, 8, 99. Wilson-Perez, H. E., Cota, D., Matias, I., Seeley, R. and Obici, S. (2012) Potential role of the intestinal lipid messenger OEA in the effects of sleeve gastrectomy on food intake and preference. Diabetes, 61, A498-A498.

147 "After endless mountains and rivers that leave doubt to way-out, suddenly one encounters the shade of a willow, bright flowers and a lovely village."-You Lu 《游山西村》 Wood, J. T., Williams, J. S., Pandarinathan, L., Courville, A., Keplinger, M. R., Janero, D. R., Vouros, P., Makriyannis, A. and Lammi-Keefe, C. J. (2008) Comprehensive profiling of the human circulating endocannabinoid metabolome: clinical sampling and sample storage parameters. Clin Chem Lab Med, 46, 1289-1295. Wood, J. T., Williams, J. S., Pandarinathan, L., Janero, D. R., Lammi-Keefe, C. J. and Makriyannis, A. (2010) Dietary docosahexaenoic acid supplementation alters select physiological endocannabinoid-system metabolites in brain and plasma. J Lipid Res, 51, 1416-1423. Yang, R., Fredman, G., Krishnamoorthy, S., Agrawal, N., Irimia, D., Piomelli, D. and Serhan, C. N. (2011) Decoding functional metabolomics with docosahexaenoyl ethanolamide (DHEA) identifies novel bioactive signals. J Biol Chem, 286, 31532-31541. Yang, Y., Chen, M., Georgeson, K. E. and Harmon, C. M. (2007) Mechanism of oleoylethanolamide on fatty acid uptake in small intestine after food intake and body weight reduction. Yuan, D., Wu, Z. and Wang, Y. (2016) Evolution of the diacylglycerol lipases. Prog Lipid Res, 64, 85-97. Zamberletti, E., Piscitelli, F., De Castro, V. et al. (2017) Lifelong imbalanced LA/ALA intake impairs emotional and cognitive behavior via changes in brain endocannabinoid system. J. Lipid Res., 58, 301-316. Zoerner, A. A., Gutzki, F. M., Batkai, S., May, M., Rakers, C., Engeli, S., Jordan, J. and Tsikas, D. (2011) Quantification of endocannabinoids in biological systems by chromatography and mass spectrometry: a comprehensive review from an analytical and biological perspective. Biochim Biophys Acta, 1811, 706-723. Zoerner, A. A., Gutzki, F. M., Suchy, M. T., Beckmann, B., Engeli, S., Jordan, J. and Tsikas, D. (2009) Targeted stable-isotope dilution GC-MS/MS analysis of the endocannabinoid anandamide and other fatty acid ethanol amides in human plasma. J Chromatogr B Analyt Technol Biomed Life Sci, 877, 2909-2923.

148 “Accomplishing a Ph.D. degree was like navigating on a long river to arrive at the ocean. The destination is just a boarding pass of an adventure ship to road to eminence."-Lin Lin

CHAPTER 9 Appendices

149

Appendix 1: Search terms from Medline 1946-October week1 2017 and Medline in-progress and non-indexed citations (Chapter 2)

MedLine-1946-October wk 1 2017 and Medline in-progress and not-indexed citation 1. exp Diet/ 2. diet*.ab,ti,kw,sh. 3. exp Food/ 4. (food*or nutrient*or meal* or dinner* or dinner time* or time* dinner or supper*or mealtime* or time* meal or lunch or lunch time* or breakfast or brunch or dessert or feast or picnic or snack or emulsion or chow or sham).ab,ti,kw,sh. 5. exp Animal Feed/ 6. exp eating/ 7. exp energy intake/ 8. (intake or dietary influence or dietary or dieting or diet or food or feed mixture or foodstuff or dairy product or food composition or food crop or food grain or nutrient or eat or eating or feed or feeding or consume or consumping or consumption or calories or calorie content or isocaloric).ab,kw,ti,sh. 9. 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 10. exp Fats/ 11. exp Oils/ 12. (dietary fat* or fat dietary* or dietary oil* or oil* dietary or unsaturated or saturated or monounsaturated or polyunsaturated or high-fat).ab,ti,kw,sh. 13. (((dietary adj2 fat*) or dietary) adj2 oil*).ab,ti,kw,sh. 14. (fat* or oil* or ghee* or butter or lard or egg yolk or nut* or seed* or animal or processed or chicken or duck or turkey or deer or beef or lamb or pork or poultry or tallow or cream or cheese or milk or margarine or shortening or trans or fish or salmon or mackerel or herring or trout or menhaden or krill or cod or codliver or avocado* or corn or palm or sunflower or safflower or high-oleic or high oleic or shrimp or grape or canola or rapeseed or palm or rice bran or olive or flaxseed or coca or coconut or pistachio or linseed or hemp or poppyseed or wheat germ or cottonseed or soybean or walnut or peanut or peanut or safflower or almon or brazil nut or cashew or hazelnut or macadamia or mongongo or pecan or pin nut or pistachio or grapefruit seed oil or lemon oil or orange oil or pumpkin or walermelon or momordica charantia or buffalo gourd oil or aci or black seed or borage seed or evening primrose or blackcurrant or amaranth or apricot or apple seed or argan oil or babassu oil or ben oil or borneo tallow nut oil or cohune oil or cocklebur or coriander or poppyseed or chestnut or yangu oil or dika oil or grape seed oil or hemp oil or kapok seed oil or lailemantia oil or kenaf seed oil or mafura oil or marula oil or meadowfoam seed or mustard oil or niger seed oil or nutmeg butter or okra seed oil or papaya or perilla or persimmon or pequi or pilli or pracaxi or prune or quinoa or ramtil or rice bran or royle oil or sacha inchi or sapote or seje or shea or taramira or tea seed or thistle or tigenut or tobacco or macadamia or plant or vegetable or dairy product* or edible oil).ab,kw,ti,sh. 15. (fatty acid* or acid* saturated or polyunsaturated or monounsaturated or aliphatic or aliphatic acid* or omega-3 or omega 3 or n-3 fatty acid* or omega-6 or omega 6 or n-6 fattty acid* or omega-9 or n-9 or n 9 or n-3 or n 3 or n-6 or n 6 or conjugated linoleic acid).ab,ti,kw,sh. 16. (palmitic acid or palmitate or hexadecanoic or palmitoleic acid).ab,ti,kw,sh.

150

17. (Heptadecanoic acid or Margaric acid).ab,ti,kw,sh. 18. (Stearic acid or dihydrooleic or octadecanoic or tetrahydrolinoleic).ab,ti,kw,sh. 19. (oleic acid or octadecenoic or oleate).ab,ti,kw,sh. 20. (linoleic acid or octadecadienoic or linoelaidic acid or linoleate).ab,ti,kw,sh. 21. (arachidonic acid or arachidonate or eicosatetraenoic).ab,ti,kw,sh. 22. (alpha-linolenic acid or alpha linolenic acid or linolenate).ab,ti,kw,sh. 23. (eicosapentaenoic acid or timnodonic or icosapentaenoic).ab,ti,kw,sh. 24. (docosatetraenoic acid or adrenic acid).ab,ti,kw,sh. 25. (docosahexaenoic acid or docosahexaenoate).ab,kw,ti,sh. 26. 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17 or 18 or 19 or 20 or 21 or 22 or 23 or 24 or 25 27. exp Endocannabinoids/ 28. (endocannabinoid* or ethanolamide* or n-acylethanolamide or n acylethanolamide or fatty acid ethanolamide or cannabinoid* or endocannabinoid-like or endocannabinoid like or acylethanolamine).ab,kw,ti,sh. 29. (Palmitoyl ethanolamide or Palmitoylethanolamide or Palmitoylethanolamine or n- palmitoylethanolamine or n palmitoylethanolamine or palmidrol or Hydroxyethylpalmitamide).ab,ti,kw,sh. 30. (Stearoyl ethanolamide or Stearoylethanolamide or Stearoylethanolamine or n- stearoylethanolamine or n Stearoyl ethanolamine or Stearoyl ethanolamine).ab,ti,kw,sh. 31. (Heptadecanoyl ethanolamide or Heptadecanoylethanolamide or Heptadecanoylethanolamine or n-heptadecanoylethanolamine or n heptadecanoylethanolamine or Heptadecanoyl ethanolamine).ab,ti,kw,sh. 32. (Oleoyl ethanolamide or Oleoylethanolamide or Oleoylethanolamine or n- oleoylethanolamine or n Oleoylethanolamine or Oleoylethanolamine).ab,ti,kw,sh. 33. (vaccenoylethanolamide or vaccenic acid ethanolamide or cis-vaccenic acid ethanolamide or n-vaccenoylethanolamine or n-vaccenoylethanolamide).ab,kw,ti,sh. 34. (Anandamide or AEA or n-arachidonoylethanolamine or arachidonoyl ethanolamide or arachidonoylethanolamide or Eicosadienoyl ethanolamide or Eicosadienoylethanolamine or n arachidonoylethanolamine or Arachidonoyl ethanolamine).ab,ti,kw,sh. 35. (Alpha-linolenoyl ethanolamide or alpha-linolenoylethanolamine or n-alpha-linolenoyl ethanolamine or Alpha-linolenoyl ethanolamine or n alpha-linolenoyl ethanolamine or alpha- linolenoyl ethanolamine).ab,ti,kw,sh. 36. (Linoleoyl ethanolamide or Linoleoylethanolamide or Linoleoylethanolamine or n- Linoleoylethanolamine or n Linoleoylethanolamine or Linoleoyl ethanolamine).ab,ti,kw,sh. 37. (Docosatetraenoyl ethanolamide or Docosatetraenoylethanolamide or Docosatetraenoylethanolamine or n-docosatetraenoylethanolamine or n docosatetraenoylethanolamine or Docosatetraenoyl ethanolamine or Adrenoyl ethanolamide).ab,ti,kw,sh. 38. (Eicosapentanoyl ethanolamide or Eicosapentanoylethanolamide or Eicosapentanoylethanolamine or n-Eicosapentanoylethanolamine or n Eicosapentanoylethanolamine or Eicosapentanoyl ethanolamine).ab,ti,kw,sh. 39. (dihomo-gama-linolenoyl ethanolamide or dihomo-gama-linolenoyl ethanolamine or dihomo gama linolenoyl ethanolamide).ab,ti,kw,sh. 40. (Docosapentaenoyl ethanolamide or Docosapentaenoylethanolamide or Docosapentaenoylethanolamine or n-docosapentaenoylethanolamine or n docosapentaenoylethanolamine or docosapentaenoyl ethanolamine).ab,ti,kw,sh.

151

41. (eicosanoylethanolamide, or n-eicosanoylethanolamide or n- eicosanoylethanolamine).ab,kw,ti,sh. 42. (Docosahexanoylethanolamide or Docosahexanoyl ethanolamide or Docosahexanoylethanolamine or n-docosahexanoylethanolamine or n docosahexanoylethanolamine or docosahexanoyl ethanolamin or Synaptamide).ab,ti,kw,sh. 43. exp Monoglycerides/ 44. (2-palmitoylglycerol or 2 palmitoylglycerol or palmitoylglycerol* or 2 palmitoyl glycerol).ab,kw,ti,sh. 45. (2-oleoylglycerol or 2 oleoylglycerol or oleoylglycerol* or 2 oleoyl glycerol).ab,kw,ti,sh. 46. (2-eicosanoylglycerol or 2-eicosanoylglycerol or eicosanoylglycerol* or 2 eicosanoyl glycerol).ab,kw,ti,sh. 47. (2-eicosapentaenoylglycerol or 2 eicosapentaenoylglycerol or eicosapentaenoylglycerol* or 2 eicosapentaenoyl glycerol).ab,kw,ti,sh. 48. (2-docosahexaenoylglycerol or 2 docosahexaenoylglycerol or docosahexaenoylglycerol * or 2-docosahexaenoyl glycerol or 2 docosahexaenoyl glycerol).ab,ti,kw,sh. 49. (2-docosatetraenoylglycerol or 2 docosatetraenoylglycerol or 2-docosatetraenoyl glycerol or 2 docosatetraenoylglycerol*).ab,ti,kw,sh. 50. (2-arachidonoylglycerol or 2-arachidonoyl glycerol or arachidonoylglycerol* or 2-AG or 2AG).ab,ti,kw,sh. 51. (Monoacylglycerol* or monoglyceride*).ab,kw,ti,sh. 52. 27 or 28 or 29 or 30 or 31 or 32 or 33 or 34 or 35 or 36 or 37 or 38 or 39 or 40 or 41 or 42 or 43 or 44 or 45 or 46 or 47 or 48 or 49 or 50 or 51 53. 9 and 26 and 52 54. limit 53 to journal article 55. limit 54 to english language 56. limit 55 to (meta analysis or "review" or "scientific integrity review" or systematic reviews) 57. 55 not 56

152

Appendix 2. Search terms from Embase October 14th, 2017 (Chapter 2)

Embase NAE-update October 14th 2017 1. (intake).ab,kw,ti,sh. 2. exp caloric intake/ or exp dietary reference intake/ or exp dietary intake/ or exp food intake/ or exp fat intake/ 3. (food or eat* or diet).ab,kw,ti,sh. 4. exp diet/ 5. exp energy consumption/ 6. exp feeding/ 7. exp eating/ 8. (Supplementation or dietary supplementation or nutritional supplementation or utrient or eat or eating or feed or feeding or consume or consuming or calories or calorie content).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 9. isocaloric.mp. 10. isocaloric.ab,ti,kw,sh. 11. 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9 or 10 12. exp fatty acid/ or exp fat/ or exp edible oil/ or exp vegetable oil/ or exp omega-3 fatty acids/ or exp saturated fatty acid/ or exp unsaturated fatty acid/ or exp monounsaturated fatty acids/ or exp polyunsaturated fatty acid/ or exp lipid composition/ or exp long chain fatty acid/ or exp essential fatty acid/ or daily products.mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 13. (fatty acid or fat or edible oil or vegetable oil or omega-3 fatty acids or saturated fatty acid or unsaturated fatty acid or monounsaturated fatty acids or polyunsaturated fatty acid or lipid composition or long chain fatty acid or essential fatty acid or daily products or palmitic acid or hexadecanoic acid or palm oil or margaric acid or heptadecanoic acid or stearic acid or octadecanoic acid or oleic acid or octadecenoic acid or linoleic acid or octadecadienoic acid or alpha-linolenic acid or arachidonic acid or eicosadienoic acid or eicosapentaenoic acid or timnodonic acid or docosatetraenoic acid or adrenic acid or docosapentaenoic acid or clupanodonic acid or docosahexaenoic acid).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 14. (fatty acid or free fatty acids or FFA or non esterified fatty acid or nonesterified fatty acid or nonesterized fatty acid or phosphatide fatty acid or phospholipid fatty acid or triglyceride fatty acid or unesterified fatty acid or Fat or fats or oil or edible oil or dietary fats or vegetable oil or plant oil or plant oils or seed oil or vegetable fat or dietary oil or edible or oil or saturated fat or saturated fatty acid or unsaturated fatty acid or UFA or unsaturated lipid or unsaturated fat or monounsaturated fatty acids or monounsaturated or polyunsaturated fatty acid or poly unsaturated fatty acid or polyunsaturated fat or lipid composition or fat composition or fatty acid composition or long chain fatty acid or fish oil or essential fatty acid or fat or oil or ghee or butter or lard or egg yolk or nut or seed or animal or processed or chicken or duck or turkey or deer or beef or lamb or pork or poultry or tallow or cream or cheese or milk or margarine or shortening or trans or fish or salmon or mackerel or herring or trout or menhaden or krill or cod or codliver or avocado* or corn or palm or sunflower or safflower or high-oleic or high oleic or

153

shrimp or grape or canola or rapeseed or palm or rice bran or olive or flaxseed or coca or coconut or pistachio or linseed or hemp or poppyseed or wheat germ or cottonseed or soybean or walnut or peanut or peanut or safflower or almon or brazil nut or cashew or hazelnut or macadamia or mongongo or pecan or pin nut or pistachio or grapefruit seed oil or lemon oil or orange oil or pumpkin or walermelon or momordica charantia or buffalo gourd oil or aci or black seed or borage seed or evening primrose or blackcurrant or amaranth or apricot or apple seed or argan oil or babassu oil or ben oil or borneo tallow nut oil or cohune oil or cocklebur or coriander or poppyseed or chestnut or yangu oil or dika oil or grape seed oil or hemp oil or kapok seed oil or lailemantia oil or kenaf seed oil or mafura oil or marula oil or meadowfoam seed or mustard oil or niger seed oil or nutmeg butter or okra seed oil or papaya or perilla or persimmon or pequi or pilli or pracaxi or prune or quinoa or ramtil or rice bran or royle oil or sacha inchi or sapote or seje or shea or taramira or tea seed or thistle or tigenut or tobacco or macadamia or plant or vegetable or dairy product or palmitic acid or hexadecanoic acid or palm oil or margaric acid or heptadecanoic acid or stearic acid or octadecanoic acid or oleic acid or octadecenoic acid or linoleic acid or octadecadienoic acid or alpha-linolenic acid or arachidonic acid or eicosadienoic acid or eicosapentaenoic acid or timnodonic acid or docosatetraenoic acid or adrenic acid or docosapentaenoic acid or clupanodonic acid or docosahexaenoic acid).ab,ti,kw,sh. 15. 12 or 13 or 14 16. (endocannabinoid*or n-acylethanolamine or n-acylethanolamide or acylethanolamine or acylethanolaminde or ethanolamine or ethanolamide or endocannabinoid-like or fatty acid ethanolamine or fatty acid ethanolamide or monoacylglycerol or monoacylglyceride or cannabinoid*).ab,ti,kw,sh. 17. exp Endocannabinoids/ 18. (endocannabinoid*or n-acylethanolamine or n-acylethanolamide or n acylethanolamine or acylethanolamine or acylethanolaminde or ethanolamine or ethanolamide or endocannabinoid- like or fatty acid ethanolamine or fatty acid ethanolamide or monoacylglycerol or monoacylglyceride or cannabinoid*).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 19. (palmitoyl ethanolamide or palmitoylethanolade or palmitoylethanolamine or n- palmitoylethanolamine or n palmitoylethanolamine).ab,kw,ti,sh. 20. (palmitoyl ethanolamide or palmitoylethanolade or palmitoylethanolamine or n- palmitoylethanolamine or n palmitoylethanolamine).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 21. (heptadecanoyl ethanolamide or Heptadecanoylethanolamide or Heptadecanoylethanolamine or n-heptadecanoylethanolamine or n heptadecanoylethanolamine).ab,kw,ti,sh. 22. (heptadecanoyl ethanolamide or Heptadecanoylethanolamide or Heptadecanoylethanolamine or n-heptadecanoylethanolamine or n heptadecanoylethanolamine).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 23. (Stearoyl ethanolamide or Stearoylethanolamide or Stearoylethanolamine or n- stearoylethanolamine or n stearoylethanolamine).ab,ti,kw,sh. 24. (Stearoyl ethanolamide or Stearoylethanolamide or Stearoylethanolamine or n- stearoylethanolamine or n stearoylethanolamine).mp. [mp=title, abstract, heading word, drug

154

trade name, original title,device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 25. (Oleoyl ethanolamide or Oleoylethanolamide or Oleoylethanolamine or n- oleoylethanolamine or n oleoylethanolamine).ab,ti,kw,sh. 26. (Oleoyl ethanolamide or Oleoylethanolamide or Oleoylethanolamine or n- oleoylethanolamine or n oleoylethanolamine).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 27. (Linoleoyl ethanolamide or Linoleoylethanolamide or Linoleoylethanolamine or n- Linoleoylethanolamine or n Linoleoylethanolamine).ab,kw,ti,sh. 28. (Linoleoyl ethanolamide or Linoleoylethanolamide or Linoleoylethanolamine or n- Linoleoylethanolamine or n Linoleoylethanolamine).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 29. (Alpha-linolenoyl ethanolamide or alpha-linolenoylethanolamine or N-alpha-linolenoyl ethanolamine or Alpha-linolenoyl ethanolamine or n alpha-linolenoyl ethanolamine).ab,ti,kw,sh. 30. (Alpha-linolenoyl ethanolamide or alpha-linolenoylethanolamine or N-alpha-linolenoyl ethanolamine or Alpha-linolenoyl ethanolamine or n alpha-linolenoyl ethanolamine).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 31. (Anandamide or n-arachidonoylethanolamine or arachidonoyl ethanolamide or arachidonoylethanolamide or Eicosadienoyl ethanolamide or Eicosadienoylethanolamine or n arachidonoylethanolamine).ab,ti,kw,sh. 32. (Anandamide or N-arachidonoylethanolamine or arachidonoyl ethanolamide or arachidonoylethanolamide or Eicosadienoyl ethanolamide or Eicosadienoylethanolamine or n arachidonoylethanolamine).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 33. (Docosatetraenoyl ethanolamide or Docosatetraenoylethanolamide or Docosatetraenoylethanolamine or n-docosatetraenoylethanolamine or n docosatetraenoylethanolamine).ab,ti,kw,sh. 34. (Docosatetraenoyl ethanolamide or Docosatetraenoylethanolamide or Docosatetraenoylethanolamine or n-docosatetraenoylethanolamine or n docosatetraenoylethanolamine).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 35. (Eicosapentanoyl ethanolamide or Eicosapentanoylethanolamide or Eicosapentanoylethanolamine or n-Eicosapentanoylethanolamine or n- Eicosapentanoylethanolamine).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 36. (Docosapentaenoyl ethanolamide or Docosapentaenoylethanolamide or Docosapentaenoylethanolamine or n-docosapentaenoylethanolamine or n docosapentaenoylethanolamine).ab,ti,kw,sh. 37. (Docosapentaenoyl ethanolamide or Docosapentaenoylethanolamide or Docosapentaenoylethanolamine or N-docosapentaenoylethanolamine or n docosapentaenoylethanolamine).mp. [mp=title,

155

abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 38. (Docosahexanoylethanolamide or Docosahexanoyl ethanolamide or Docosahexanoylethanolamine or n-docosahexanoylethanolamine or n docosahexanoylethanolamine or synaptamide).ab,ti,kw,sh. 39. (Docosahexanoylethanolamide or Docosahexanoyl ethanolamide or Docosahexanoylethanolamine or n-docosahexanoylethanolamine or n docosahexanoylethanolamine or synaptamide).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 40. (2-palmitoylglycerol or 2 palmitoylglycerol or 2 palmitoyl glycerol).ab,kw,ti,sh. 41. (2-oleoylglycerol or 2 oleoylglycerol or 2 oleoyl glycerol).ab,kw,ti,sh. 42. (2-palmitoylglycerol or 2 palmitoylglycerol or 2 palmitoyl glycerol).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 43. (Docosahexanoylethanolamide or Docosahexanoyl ethanolamide or Docosahexanoylethanolamine or n-docosahexanoylethanolamine or n docosahexanoylethanolamine).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 44. (2-oleoylglycerol or 2 oleoylglycerol or 2 oleoyl glycerol).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 45. (2-eicosanoylglycerol or 2-eicosanoylglycerol or 2 eicosanoyl glycerol).ab,kw,ti,sh. 46. (2-eicosanoylglycerol or 2-eicosanoylglycerol or 2 eicosanoyl glycerol).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 47. (2-eicosapentaenoylglycerol or 2 eicosapentaenoylglycerol or 2 eicosapentaenoyl glycerol).ab,kw,ti,sh. 48. (2-eicosapentaenoylglycerol or 2 eicosapentaenoylglycerol or 2 eicosapentaenoyl glycerol).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 49. (vaccenoylethanolamide or vaccenic acid ethanolamide or cis-vaccenic acid ethanolamide or n-vaccenoylethanolamine or n vaccenoylethanolamine).ab,kw,ti,sh. 50. (vaccenoylethanolamide or vaccenic acid ethanolamide or cis-vaccenic acid ethanolamide or n-vaccenoylethanolamine or n vaccenoylethanolamine).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 51. (2-docosahexaenoylglycerol or 2 docosahexaenoylglycerol or 2-docosahexaenoyl glycerol or 2 docosahexaenoyl glycerol).ab,ti,kw,sh. 52. (2-docosahexaenoylglycerol or 2 docosahexaenoylglycerol or 2-docosahexaenoyl glycerol or 2 docosahexaenoyl glycerol).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word]

156

53. (2-docosatetraenoylglycerol or 2 docosatetraenoylglycerol or 2-docosatetraenoyl glycerol or 2 docosatetraenoylglycerol*).ab,ti,kw,sh. 54. (2-docosatetraenoylglycerol or 2 docosatetraenoylglycerol or 2-docosatetraenoyl glycerol or 2 docosatetraenoylglycerol*).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 55. (2-arachidonoylglycerol or 2-arachidonoyl glycerol or arachidonoylglycerol).ab,ti,kw,sh. 56. (2-arachidonoylglycerol or 2-arachidonoyl glycerol or arachidonoylglycerol).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word] 57. 16 or 17 or 18 or 19 or 20 or 21 or 22 or 23 or 24 or 25 or 26 or 27 or 28 or 29 or 30 or 31 or 32 or 33 or 34 or 35 or 36 or 37 or 38 or 39 or 40 or 41 or 42 or 43 or 44 or 45 or 46 or 47 or 48 or 49 or 51 or 52 or 53 or 54 or 55 or 56 58. 11 and 15 and 57 59. limit 58 to english language 60. limit 59 to journal 61. limit 60 to "review" 62. 60 not 61

157

Appendix 3. PRISM Checklist 2009 (Chapter 2)

Reported Section/topic # Checklist item on# TITLE Title 1 Identify the report as a systematic review, meta-analysis, or both. 22 ABSTRACT Structured summary 2 Provide a structured summary including, as applicable: background; objectives; data sources; study 23 eligibility criteria, participants, and interventions; study appraisal and synthesis methods; results; limitations; conclusions and implications of key findings; systematic review registration number. INTRODUCTION Rationale 3 Describe the rationale for the review in the context of what is already known. 24-32 Objectives 4 Provide an explicit statement of questions being addressed with reference to participants, interventions, 33 comparisons, outcomes, and study design (PICOS). METHODS Protocol and registration 5 Indicate if a review protocol exists, if and where it can be accessed (e.g., Web address), and, if N/A available, provide registration information including registration number. Eligibility criteria 6 Specify study characteristics (e.g., PICOS, length of follow-up) and report characteristics (e.g., years 35, 36, 37 considered, language, publication status) used as criteria for eligibility, giving rationale. Information sources 7 Describe all information sources (e.g., databases with dates of coverage, contact with study authors to 34 identify additional studies) in the search and date last searched. Search 8 Present full electronic search strategy for at least one database, including any limits used, such that it 12, 178-187 could be repeated. Study selection 9 State the process for selecting studies (i.e., screening, eligibility, included in systematic review, and, if 34 applicable, included in the meta-analysis). Data collection process 10 Describe method of data extraction from reports (e.g., piloted forms, independently, in duplicate) and 34 any processes for obtaining and confirming data from investigators. Data items 11 List and define all variables for which data were sought (e.g., PICOS, funding sources) and any 35-36 assumptions and simplifications made.

158

Risk of bias in 12 Describe methods used for assessing risk of bias of individual studies (including specification of 36 individual studies whether this was done at the study or outcome level), and how this information is to be used in any data synthesis. Summary measures 13 State the principal summary measures (e.g., risk ratio, difference in means). 36 Synthesis of results 14 Describe the methods of handling data and combining results of studies, if done, including measures of 36 consistency (e.g., I2) for each meta-analysis. Risk of bias across 15 Specify any assessment of risk of bias that may affect the cumulative evidence (e.g., publication bias, N/A studies selective reporting within studies). Additional analyses 16 Describe methods of additional analyses (e.g., sensitivity or subgroup analyses, meta-regression), if N/A done, indicating which were pre-specified. RESULTS Study selection 17 Give numbers of studies screened, assessed for eligibility, and included in the review, with reasons for 38, 39 exclusions at each stage, ideally with a flow diagram. Study characteristics 18 For each study, present characteristics for which data were extracted (e.g., study size, PICOS, follow-up 40-44 period) and provide the citations. Risk of bias within 19 Present data on risk of bias of each study and, if available, any outcome level assessment (see item 12). 40, 190-192 studies Results of individual 20 For all outcomes considered (benefits or harms), present, for each study: (a) simple summary data for 45-70 studies each intervention group (b) effect estimates and confidence intervals, ideally with a forest plot. Synthesis of results 21 Present results of each meta-analysis done, including confidence intervals and measures of consistency. N/A Risk of bias across 22 Present results of any assessment of risk of bias across studies (see Item 15). N/A studies Additional analysis 23 Give results of additional analyses, if done (e.g., sensitivity or subgroup analyses, meta-regression [see N/A Item 16]). DISCUSSION Summary of evidence 24 Summarize the main findings including the strength of evidence for each main outcome; consider their 71-82 relevance to key groups (e.g., healthcare providers, users, and policy makers). Limitations 25 Discuss limitations at study and outcome level (e.g., risk of bias), and at review-level (e.g., incomplete 82 retrieval of identified research, reporting bias). Conclusions 26 Provide a general interpretation of the results in the context of other evidence, and implications for 83 future research. 159

FUNDING Funding 27 Describe sources of funding for the systematic review and other support (e.g., supply of data); role of 83 funders for the systematic review.

160

Appendix 4: The risk of bias assessment for human studies

a. Risk of bias summery for human studies b Risk of bias details for human studies

161

Appendix 5. The risk of bias assessment for animal studies (Chapter 2) a. Risk of bias summery for animal studies

162

b Risk of bias details animal studies

163