The Pennsylvania State University

The Graduate School

The Department of Nutritional Sciences

EFFECTS OF REPLACING SATURATED FATTY ACIDS WITH WALNUTS OR

VEGETABLE OILS ON CENTRAL BLOOD PRESSURE, ARTERIAL

STIFFNESS INDICES, LIPIDS, LIPOPROTEINS, AND THE GUT MICROBIOME

IN ADULTS AT RISK FOR CARDIOVASCULAR DISEASE

A Dissertation in

Nutritional Sciences and Clinical and Translational Sciences

Alyssa M. Tindall

Submitted in Partial Fulfillment of the Requirements for the Degree of

Doctor of Philosophy

May 2019

The dissertation of Alyssa M. Tindall was reviewed and approved* by the following:

Penny M. Kris-Etherton Distinguished Professor of Nutritional Sciences Dissertation Advisor Chair of Committee

Laura E. Murray-Kolb Associate Professor of Nutritional Sciences Professor-in-Charge of the Nutritional Sciences Graduate Program

Gregory C. Shearer Associate Professor of Nutritional Sciences

David N. Proctor Professor of Kinesiology

Regina Lamendella Associate Professor of Biology at Juniata College Special Member

*Signatures are on file in the Graduate School

ABSTRACT

The purpose of this dissertation was to investigate the effects of walnut consumption on traditional and emerging cardiovascular disease (CVD) risk factors and the contributions of α linolenic acid (ALA) versus bioactives in adults at risk for CVD. The specific study objectives were (1) to determine whether the individual or combined effect of the bioactives and ALA content of walnuts play a critical role in CVD risk reduction to better understand the underlying mechanisms and (2) to understand how differing fatty acid and bioactive profiles may modulate the gut microbial community. In a randomized, three-period, crossover, controlled-feeding study, participants were fed isocaloric, weight maintenance diets for six weeks each following a two-week Western-style run-in diet. The three study diets included the walnut diet [(WD; 7% saturated fatty acids (SFA), 9% monounsaturated fatty acids (MUFA), 16% polyunsaturated fatty acids (PUFA), 2.7% (ALA)] that provided whole walnuts (57-99 g/day), the walnut fatty acid matched diet (WFMD; 7% SFA, 9% MUFA, 16% PUFA, 2.6% ALA) that had the same fatty acid profile as the WD (but devoid of walnuts), and the oleic acid replaces ALA diet (ORAD; 7% SFA, 12% MUFA, 14% PUFA) that substituted the ALA content of walnuts in the WD with oleic acid and was otherwise macronutrient matched to the other two diets. The run-in diet was formulated to represent a Western-style diet (12% SFA, 12% MUFA, 7% PUFA).

In the first study, we examined the effects of the three study diets on central and peripheral blood pressure, markers of vascular health, and lipids and lipoproteins in 45 participants. After six weeks, central diastolic blood pressure

iii decreased on the WD (-1.78±1.0 mmHg) from baseline (P=0.02) compared to the

ORAD (0.15±0.7 mmHg, P=0.04). The WD lowered bMAP (-1.44 ± 0.7 mmHg,

P=0.04) and cMAP (-1.72 ± 0.8 mmHg, P=0.02) compared to baseline, but there were no differences between the diets. TC was lowered following the WD (-15.8

± 3.2 mg/dL), WFMD (-14.6 ± 2.9 mg/dL), and ORAD (-11.2 ± 2.7 mg/dL) from baseline (P for all <0.0001). Similarly, LDL-C decreased after the WD (-13.4 ± 2.9 mg/dL), WFMD (-11.1 ± 2.3 mg/dL), and ORAD (-9.2 ± 2.3 mg/dL) from baseline

(P for all <0.0001) and HDL-C was also reduced from baseline following the WD,

WFMD, and ORAD (-1.8 ± 0.8 mg/dL, -2.1 ± 0.9, -1.1 ± 0.7, respectively, P for all

<0.0001). Non-HDL-C was reduced following the WD, WFMD, and ORAD (-14.0

± 2.9 mg/dL, P<0.006; -12.5 ± 2.5, P<0.0009; -11.0 ± 2.2, P<0.03, respectively).

There were not significant differences between diets for lipoprotein endpoints.

There were no significant differences in measures of arterial stiffness.

In the second study we investigated the effects of the study diets on the gut microbiome through 16S rRNA sequencing. Our results demonstrate enrichment of eubiotic bacteria, including butyrate producers such as Roseburia

[Relative abundance (RA)=4.2%, LDA=4.2, P=0.0008], Eubacterium eligensgroup (RA=1.4%, LDA=3.6, P=0.05) and Butyricicoccus (RA=0.3%,

LDA=3.3, P=0.01) following the WD and Roseburia (RA=3.6%, LDA=3.8,

P=0.02), Eubacterium eligensgroup (RA=1.5%, LDA=3.4, P=0.02) after the

WFMD. Our data also showed enrichment of Gordonibacter (RA=0.04%,

LDA=3.2, P=0.03), a gut bacteria involved in the metabolism of a walnut- bioactive, i.e., ellagitannins, was uniquely enriched following the WD. We also

iv observed correlations between gut bacteria and CVD risk factors after the WD;

Eubacterium eligens was associated with brachial MAP (R=-0.50; P=0.0009), central diastolic BP (-0.52; 0.0006), and central MAP (-0.47, 0.002),

Lachnospiraceae was associated with brachial MAP (R=-0.37, P=0.02), central diastolic BP (-0.32; 0.04), central MAP (-0.35; 0.02), TC (-0.35; 0.03), non-HDL-C

(-0.37; 0.02), and Leuconostocaceae was associated with brachial MAP (R=0.34;

P=0.03) and central MAP (0.34; 0.03).

Overall, these data indicate that the fatty acid profile of walnuts lowers

CVD risk and there may be additional benefits to consuming whole walnuts beyond their fatty acid profile on novel and established cardiovascular risk factors. We investigated the contributions ALA versus bioactives in walnuts and report evidence of reductions in central systolic blood pressure and favorable changes in the gut microbiome that may be protective against CVD. Together, these two studies provide information to add to the existing knowledge of the underlying mechanisms of walnut consumption associated with cardiovascular benefits.

TABLE OF CONTENTS

List of Figures…………………………………………………………………….….....vii List of Tables……….……………………………………………………………………ix Abbreviations……………………………………………………………………….…...xi Acknowledgements…………………………………………………………………....xiii Chapter 1 Introduction……………………………………………………………….....1 Chapter 2 Literature Review…………………………………………………………...3 2.1 Nutrient profile of walnuts.……………………………………………...... 3 2.1.1 Macronutrient profile………………………………………….....3 2.1.2 Bioactive profile……………………………………………….....6 2.1.2.1 Ellagitannins…………………………………………..6 2.1.2.2 γ-Tocopherol…..……………………………………...8 2.1.2.3 Melatonin.….……………………………………….....9 2.2 Evidence for walnuts reducing established and emerging cardiovascular risk factors…………………………………………………….12 2.2.1 Epidemiological evidence for nut intake and CVD…………12 2.2.2 Experimental evidence for walnut consumption and CVD risk……………………………………………………………...17 2.2.2.1 Blood lipid profile……………………………………..17 2.2.2.2 Vascular health……………………………………….25 2.2.2.3 The gut microbiome………………………………….31 2.3 Rationale for current research…………………………………………..38 Chapter 3 Replacing saturated fat with walnuts or vegetable oils improves central blood pressure and serum lipids in adults at risk for cardiovascular disease: a randomized, controlled-feeding trial……………………………………………...….37 Abstract…………………………………………………………………………37 Introduction.…………………………………………………………………….39 Methods………………………………………………………………………...41 Results……………………………………………………………………….....53 Discussion……………………………………………………………………...69 Chapter 4 Walnuts and vegetable oils differentially affect the gut microbiome and associations with cardiovascular risk factors……………………………………….78 Abstract…………………………………………………………………………78 Introduction.…………………………………………………………………….80 Methods and Statistics………………………………………………………..81 Results.….……….……………………………………………………………..90 Discussion.……………………………………………………………………111 Chapter 5 Research Summary and Future Directions…………………………..119 Appendix……………..……...…….…………………………………………………124 References…………………………………………………...………………………149

List of Figures

Figure 2-1. One potential mechanism by which α-linolenic acid (ALA) reduces plasma cholesterol levels…………………………………………………………..…18

Figure 2-2. Known and proposed interactions between diet and cardiometabolic risk……………………………………………………………………………………....33

Figure 3-1. Study design for this randomized, 3-period, crossover, controlled- feeding trial………………………………………………………………………….….42

Figure 3-2. CONSORT diagram of participant flow through the study. Abbreviations: BMI, body mass index; BP, blood pressure; LDL-C, low-density lipoprotein-cholesterol…………………………………………………………….…..54 Figure 3-3. Mean changes from baseline for blood pressure and vascular measures following each of the three 6-week study diets (n=45)…………….….59

Figure 3-4. Mean changes from baseline for lipids and lipoproteins following each of the three 6-week study diets (n=45)…………………………………..……63

Figure 3-5. Comparisons between high and low CRP groups for lipids and lipoproteins……………………………………………………………………….…….68

Figure 4-1. Alpha diversity curves and boxplots were generated from a biom- formatted ASV table in QIIME 1.9.0 to compare richness between samples and diets……………………………………………………………………………………..92

Figure 4-3. Average relative abundance (RA) of phyla in the three study diets..94

Figure 4-4. Linear discriminant analysis effect size (LefSe) plots for between-diet and study diet-run-in diet comparisons of enriched taxa………………………….96

Figure 4-5. Relative abundance of the most prominent phyla following the three study diets………………………………………………………………………...…..100

Figure 4-6. Relative abundance of the most prominent phyla in participants with overweight, obesity, and morbid obesity…………………………………………..101 vii

Figure 4-7. Co-occurrence network (CoNet) analysis for the walnut diet (a), walnut fatty acid matched diet (b), oleic acid replaces ALA diet (c) and the run-in, standard Western diet (d)………………………………………………………..….103

viii

List of Tables

Table 2-1. Average nutrient composition of tree nuts in a 42.4 g (1.5 oz) serving (modified from Tindall et al.1)………………………………………………...…….….5

Table 2-2. Details the landmark studies that show an association between nut consumption and CVD mortality……………………………………………………..15 Table 2-3. Effect of walnut consumption on lipids and lipoproteins (RCTs)….…19 Table 2-4. Walnut consumption and blood pressure (RCTs)………………….….28 Table 2-5. Effects of walnuts on the gut microbiome (RCTs)…………………….34 Table 3-1. Nutrient composition of the study diets………………………………...46 Table 3-2. Sample menu from one day of the six-day cycle menu………………47 Table 3-3. Sub-sample total plasma fatty acid analysis (n=5)……………………48

Table 3-4. Average study compliance for each participant, including participants that did not complete the entire study……………………………………………….55

Table 3-5. Baseline characteristics of participants that did not complete the study………………………………………………………………………………..56

Table 3-6. Baseline characteristics of participants randomized to study diets….57

Table 3-7. Between-diet comparisons of vascular measures…………………….61

Table 3-8. Between-diet differences in blood measures and weight…………….64

Table 3-9. Comparisons between diets by BMI classifications for cSBP and cMAP………………………………………………………………………………67

Table 4-1. Nutrient Profiles of the run-in diet and study diets………………….…82

Table 4-2. Baseline Characteristics ………………………………………………...90

Table 4-3. Linear discriminant analysis effect size (LefSe) analysis between study diets and versus the run-in diet…………………………………………….....97

Table 4-4. Correlations between enriched taxa following the walnut diet (WD) and cardiovascular risk factors that were significantly changed from baseline…………………………………………………………………………….....105

Table 4-6. Correlations between enriched taxa following the oleic acid replaces ALA diet (ORAD) and cardiovascular risk factors that were significantly changed from baseline………………………………………………………………………...107

Abbreviations

ABCA1 ATP-binding cassette transporter 1 ACSL3 acyl-CoA synthetase long chain family member 3 ACSS2 acyl-CoA synthetase short chain faimily member 2 AHA American Heart Association AI adequate intake AIx augmentation index ALA alpha-linolenic acid AP augmentation pressure ASV amplicon sequence variant AUC area under the curve bMAP brachial mean arterial pressure BMI body mass index bPP bBrachial pulse pressure cDBP central diastolic blood pressure CEHC carboxyethyl hydroxychroman CI confidence interval cMAP central mean arterial pressure CoNet co-occurrence network analysis cPP central pulse pressure CPT-1a carnitine palmitoyltransferase-1a CRC Clinical Research Center CRP C-reactive protein cSBP central systolic blood pressure CVD cardiovascular disease CYP51 lanosterol 14alpha-demethylase dada2 divisive amplicon denoising algorithm DGA Dietary Guidelines for Americans dsDNA double stranded deoxyribonucleic acid ET ellagitannins FA fatty acids FMD flow-mediated dilation FXR farnesyl X receptor gDNA genomic deoxyribonucleic acid GIT gastriintestinal tract HDL-C high-density lipoprotein choleseterol HMGCS1 hydroxymethlglutaryl-CoA synthase hsCRP high-sensitivity C-reactive protein IBD inflammatory bowel disease IL-1B interleukin 1-beta IL-6 interleukin-6 ITT intent-to-treat xi

KEGG Kyoto encyclopedia of genes and genomes KO KEGG orthology LDA linear discriminant analysis LDL-C low-density lipoprotein cholesterol LEfSe linear discriminant analysis effect size MDSC Metabolic Diet Study Center mRNA messenger ribonucleic acid MT1-3 melatonin receptor 1, 2, 3 MUFA monounsaturated fatty acids NHS Nurses' Health Study NR not reported ORAD oleic acid replaces alpha-linolenic acid diet PCoA principal coordinates analysis PCR polymerase chain reactions PICRUSt phylogenetic investigation of communities by reconstruction of unobserved states PTT pulse transit time PUFA polyunsaturated fatty acids PWV pulse wave velocity QIIME quantitative insights into microbial ecology RA relative abundance RCT randomized controlled trial RDA recommend daily allowance RR relative risk rRNA ribosomal ribonucleic acid SBP systolic blood pressure SC5D sterol-C5-desaturase SCD1 sterol-CoA desaturase-1 SCFA short chain fatty acid SEM standard error of the mean SFA saturated fatty acid SHP small heterodiamer partner SREBP sterol regulatory element-binding protein sVCAM-1 soluble vascular cell adhesion molecule-1 SWD standard Western diet T2DM Type 2 diabetes mellitus TC total cholesterol TG triglycerides TM7SF2 transmembrane 7 superfamily member 2 TMA trimethyamine WD walnut diet WFMD walnut fatty acid matched diet

xii

ACKNOWLEDGEMENTS

Graduate school and the department of nutritional sciences have helped me learn, explore, and grow as a scientist, educator, and community member.

The support I have experienced from colleagues, mentors, and friends has been overwhelming and successful completion of this program would have been impossible without them.

I first would like to acknowledge my advisor, Penny Kris-Etherton, for her guidance, patience, and encouragements during the four years I’ve spent at

Penn State in her lab. She is not only a truly amazing scientist, but a warm and thoughtful person. I would also like to thank my committee members, Drs. Laura

Murray-Kolb, Gregory Shearer, David Proctor, and Regina Lamendella for providing support, expertise, and mentorship. Kristina Petersen has also been an excellent mentor to me and I greatly appreciate her help and guidance.

I would like to acknowledge the funding sources of this research. Support for this research was provided by the California Walnut Commission and the

Penn State University Clinical and Translational Research Institute, Penn State

University Clinical and Translational Science Award, and NIH/National Center for

Advancing Translational Sciences grant no. UL1TR000127 (any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author and do not necessarily reflect the view of the NIH/National

Center for Advancing Translational Sciences).

xiii

Completion of my dissertation work would not have been possible without the nursing staff at the Clinical Research Center (Cyndi Flanagan, Christa

Oelhaf, Phyllis Martin), the Metabolic Diet Study Center manager (Marcella

Smith) and staff, Brian Harsch and Christopher McLimans for assistance in fatty acid and microbiome analyses, and I am immensely grateful to them. Current and previous PKE lab members (Kate Bowen, Valerie Sullivan, Emily Johnston,

Philip Sapp, Yujin Lee, Chesney Richter, and Danette Teeter) have been an amazing support system and wonderful friends.

I would like to thank my mom and dad, my siblings (Krista and Kyle), and the rest of my wonderful family and friends who have provided me continual encouragement. Lastly, I thank, Eric, who has been loving and supportive despite driving countless hours over the past years.

xiv

Chapter 1 Introduction

Cardiovascular disease (CVD) is the leading cause of death in both the

United States and worldwide, accounting for approximately 31% of global deaths2. Although non-modifiable risk factors exist, such as age and sex, improvements in modifiable risk factors can greatly lower the risk of CVD.

Modifiable risk factors include but are not limited to: abnormal blood lipids, hypertension, endothelial dysfunction, and arterial stiffness3. Improvements in lifestyle factors, including diet, can lower disease risk3. The 2015-2020 Dietary

Guidelines for Americans recommend three, plant-based dietary patterns to lower chronic disease risk, including CVD4.

The recommended eating patterns all contain a variety of vegetables, whole grains, and protein from plant and animal sources. Nuts and seeds are a common dietary component across all patterns and the FDA has also issued a health claim for nuts and coronary heart disease. Although the 2015-2020

Dietary Guidelines recommended 142-198 g (5-7 oz) equivalents of nuts and seeds per week (based on a 2000 calorie diet), a large portion of the US does not fall within this intake range4. Micha et al.5 estimated that 45% of cardiometabolic deaths can be attributed to suboptimal diet. Among individual dietary factors, low nuts and seeds consumption was ranked second highest and associated with an estimated 8.5% of diet-related cardiometabolic deaths (59,

374 deaths)5.

Studies of individual tree nuts have reported favorable changes in cardiovascular risk factors including lipids and lipoproteins, brachial blood pressure, and endothelial function. Walnuts are well-known for their lipid-lowering benefits, but less is known on more novel risk factors, such as central blood pressure and arterial stiffness. Further, walnuts are uniquely rich in polyunsaturated fatty acids (PUFA), including α-linolenic acid (ALA) versus all other tree nuts contain predominately monounsaturated fatty acids (MUFA). The cardiovascular benefits of walnut consumption are largely attributed to the PUFA content; however, walnuts also contain bioactive compounds, such as tocopherols, which may also contribute to walnut-related benefits. The overarching aim of this dissertation is to evaluate the effects of walnut consumption on traditional and emerging CVD risk factors and the contributions of ALA versus bioactives.

Chapter 2 Literature Review

2.1 Nutrient profile of walnuts

2.1.1 Macronutrient profile

Walnuts (Juglans regia) are classified as a tree nut in addition to: almonds, Brazil nuts, cashews, hazelnuts, macadamia nuts, pecans, pine nuts, and pistachios. The proportion of unsaturated fatty acids is greater compared to saturated fatty acids (SFA; 72-88% total fat compared to 7-24%, respectively; see Table 2-1), which makes them a source of healthy dietary fat, along with the other beneficial nutrients they provide. The 2015-2020 Dietary Guidelines for

Americans (DGA) recommended a healthy eating pattern that replaces SFA with unsaturated fatty acids to achieve a SFA intake of <10% as a strategy to reduce risk of chronic disease. The majority of the unsaturated fatty acids contained in most tree nuts are MUFA, except for walnuts, which contain a greater amount of

PUFA (73% total fat). PUFA are preferentially recommended as a SFA replacement compared to MUFA6. The presidential advisory from the American

Heart Association (AHA) states “lowering intake of saturated fat and replacing it with unsaturated fats, especially polyunsaturated fats, will lower the incidence of

CVD”.

Walnuts are also high in ALA, an omega-3 fatty acid. English walnuts contain 2.5 g ALA per 28.3 g (1 ounce), which is 156-227% of the adequate intake (AI). The AI is based on observed median intakes for the US population for which there is no deficiency and is not a recommended dietary allowance (RDA) 3 or an intake of fatty acids shown to lower risk of disease. Although there is no

RDA for ALA, as well as all omega-3 fatty acids, ALA still is an essential fatty acid and is associated with lower CVD risk7,8. The mechanisms by which ALA affects cardiovascular risk are not completely understood, but ALA has been shown to have anti-inflammatory properties that may prevent smooth muscle damage and endothelial dysfunction 7,9. Walnuts are a key plant source of ALA, since this fatty acid is not readily abundant in Western-style diets. Other plant sources of omega-3 include: flax, chia, and canola oil.

Walnuts are primarily comprised of fatty acids (90%), but also contain protein [4.3 g/28 g (4.3 g/1 oz)], carbohydrates (3.9 g/1 oz), and fiber (1.9 g/1 oz).

These macronutrients are present as a small fraction of the overall nutrient profile, but may play an important role in the benefits walnuts provide. The fiber content in 28 g (1 oz) of walnuts provides 5-8% of the RDA and given the current average daily fiber intake is only 45-69% of the RDA, even small increases could have a large health impact6. The 2015-2020 DGA recommended seafood, legumes, and nuts as desirable protein sources; therefore, the protein content in walnuts is an important source of protein6. The small amount of carbohydrate content of walnuts provides minimal energy, but functionally reaches the lower gastrointestinal tract along with other non-digestible nutrients, where it is fermented by the gut microbiome10,11. Although the protein, carbohydrate, and fiber are relatively low compared with the lipid content in walnuts, these nutrients may still play an important role in the observed benefits of walnut consumption12–

14.

Table 2-1. Average nutrient composition of tree nuts in a 42.4 g (1.5 oz) serving1 (modified from Tindall et al.1)

Energy Total SFA MUFA PUFA Protein Fiber Melatonin Total ET γ- (kcal) fat (g) (g) (g) (g) (g) (mg) phenolics (mg) tocopherol (g) (mg) (mg) Almonds 246 21.2 1.6 13.4 5.2 9.0 5.3 NR 19.9-177.7 20.8-26.8 0.01 Brazil nuts2 280 28.5 6.9 10.2 10.4 6.1 3.2 NR 47.6-131.8 NR 0.623 Cashews 235 18.7 3.3 10.1 3.3 7.8 1.4 NR 58.2-116.5 NR 2.26 Hazelnuts 267 25.8 1.9 19.4 3.4 6.4 4.1 NR 123.7-354.9 NR NR Macadamia 305 32.2 5.1 25.0 0.6 3.4 3.7 NR 19.6-66.3 NR 0.0004 nuts Pecans 294 30.6 2.6 17.4 9.2 3.9 4.1 NR 545.7-856.8 128.0 10.39 Pine nuts2 286 29.1 2.1 8.0 14.5 5.8 1.6 NR 13.6-28.9 NR 4.74 Pistachios 238 19.3 2.5 10.0 6.1 8.6 4.5 9.6-9.9 368.5-704.2 NR 8.68 Walnuts 278 27.7 2.6 3.8 20.1 6.5 2.8 0.0001- 662.2-690.6 2.6-349.8 8.86 0.0003

1Nutrients and energy values for raw nuts. 2Values are for dried nuts.

Abbreviations: Ellagitannins (ET), not reported (NR)

2.1.2 Bioactive profile1

2.1.2.1 Ellagitannins

Ellagitannins (ET) are hydrolysable tannins that may provide cardiovascular-related improvements and are relatively high in walnuts compared to other tree nuts (Table1.1)15. In-vitro evidence suggests that ellagitannins may have an atheroprotective affect through reducing adhesion to THP-1 monocytes to endothelial cells and the secretion of cellular adhesion molecule, soluble vascular cell adhesion molecule-1 (sVCAM-1) and inflammatory cytokine, interleukin-6 (IL-6)16. Catabolism of ET and production of various metabolites may be the most important step in the reported benefits. In order to obtain the

CVD risk-reducing effects from ETs, the phenolic must first be metabolized in the gastrointestinal tract (GIT) to ellagic acid. Ellagic acid can be released from ETs through GIT pH changes and/or gut microbial hydrolysis via tannin-hydrolase and lactonase17,18. Ellagic acid may also be an obesity-moderator; in-vitro studies reported ellagic acid inhibits adipogenesis, reduces lipogenesis, and alters adipocyte differentiation18. However, the translation to humans would be challenging since the estimated effective dose of ellagic acid was reported to be approximately 30–850 mg/day for a healthy individual19, which is not easily

______1Sentences in this section (2.1.2.) were extracted, with permission, from a published article in Current Developments in Nutrition [Tindall AM, Petersen KS, Lamendella R, Shearer GC, Murray-Kolb LE, Proctor DN, Kris-Etherton PM. The Microbiome as a Potential Mediator between Tree Nut Consumption and Adipose Tissue Mass. Current Developments in Nutrition. 2018; 2(11). nzy069.] Alyssa Tindall wrote sentences selected from the publication in their entirety. The final publication is available at https://doi.org/10.1093/cdn/nzy069.

6 attainable through a typical American dietary pattern. ET and ellagic acid are associated with reduced CVD risk, although the evidence is based on in-vitro studies and needs to be confirmed with clinical trials.

The gut microbiome also affects metabolism of ET and in turn, CVD risk.

Ellagic acid can undergo further catabolism to urolithins via gut bacteria. Recent evidence suggests urolithins production varies among individuals; some individuals produce different types and quantities of urolithins compared to others

20,21. There are several different urolithin conjugates, urolithin-A and urolithin-B, and cardiovascular benefits may depend on the urolithin metabolite profile produced. One in-vitro study reported increased eNOS activity in human aortic endothelial cells in the presence of a mixture of urolithin A and urolithin B metabolites22. Tomás-Barberán (2014) and colleagues analyzed urolithin content in urine and fecal samples collected from three separate acute trials that provided participants with 30 g walnuts or 1.9 g and 0.9 g pomegranate extract.

The authors determined that three phenotypes could be observed independent of the volunteers’ health status, age, gender, BMI and of amount or type of ellagitannin food source ingested; “metabotype A” produced only urolithin A conjugates, “metabotype B” produced isourolithin A and/or urolithin B in addition to urolithin A, and “metabotype 0” did not produce urolithins. These metobotypes may be important in eliciting cardiovascular benefits from ET 21 and have been correlated with weight status24. Selma et al.24 investigated differences in ellagic acid metabolism between healthy overweight-obese and normal weight individuals from two separate studies in which participants were given 30 g nuts

(n=20) or 450 mg pomegranate extract (n=49). Investigators correlated cardiometabolic risk biomarkers found in the plasma with urolithins in feces and urine and reported metabotype B was most common in overweight-obese (31%) versus normal weight (20%) individuals, while metabotype A was higher in normal weight (70%) compared to overweight-obese participants (57%). This study suggests an association with body composition and urolithin metabotype; healthy weight individuals may have the ability to obtain the greatest benefit from

ET consumption. The reductions in cardiovascular risk attributed to ET may be related to both body weight status and the gut microbiome; however, the number of in-vivo studies is limited with respect to these reported associations.

2.1.2.2 γ-Tocopherol

Vitamin E is an essential, lipid soluble vitamin that is naturally found in eight different structural forms; four tocopherols and four tocotrienols that all possess different bioactivities25. Gamma (γ)-tocopherol is the major form of vitamin E found in most tree nuts [0.0003-6.93 mg/28 g (0.0003-6.93 mg/1 oz) nuts], but is found in the greatest concentration in pecans, walnuts, and pistachios, respectively26,27. In-vitro models showed that α- and γ-tocopherol can decrease endogenous cholesterol synthesis, apo-A1-mediated cholesterol secretion, and down-regulate almost half of the genes involved in cholesterol synthesis28. Γ -tocopherol can be degraded to hydrophilic, γ-carboxyethyl hydroxychroman (CEHC) by a cytochrome P450-dependent process28. Both γ- tocopherol and γ-CEHC possess anti-inflammatory activity through inhibiting 8 prostaglandin E2 synthesis in lipopolysaccharide-stimulated macrophages and interleukin 1β (IL-1β)-activated epithelial cells25. The antioxidant and anti- inflammatory properties have been shown to have a protective effect on the vascular endothelium following postprandial hyperglycemia29, a known contributor to endothelial dysfunction and oxidative stress response30. However,

Ward et al.31 tested the effect of 500 mg mixed tocopherols (60% γ- tocopherol,

25% δ- tocopherol,15% α- tocopherol) in adults with type 2 diabetes (T2DM) for six weeks and the authors reported an increase in blood pressure compared to placebo (soybean oil stripped of tocopherols). Another study showed lower platelet activation, low-density lipoprotein cholesterol (LDL-C), platelet aggregation, and mean platelet volume after supplementation with 100-200 mg/d

γ-tocopherol32. These findings suggest the structural form and dose of vitamin E may affect cardiovascular risk. Clinical trials have examined whole foods that contain tocopherols and have reported reductions in cardiovascular risk factors, illustrating the form in which tocopherols are consumed is important33–36.

Consuming tocopherols within a food complex versus an extracted supplement may affect their bioactivity and efficacy in reducing CVD risk.

2.1.2.3 Melatonin

Melatonin, known as N-acetyl-5-methoxy tryptamine, is synthesized in the pineal gland and neuro-endocrine cells of the GIT mucosa37. Melatonin also can be obtained in the diet from foods, including two tree nuts. The melatonin content

9 in walnuts is 3.5 – 7.5 ng/g38,39 and 226,900 – 233,000 ng/g in pistachios40. The significant range in melatonin concentrations that have been reported may be explained by the use of different methods of analysis and nut varieties; standardized methodology could provide more precise results. Although foods that have melatonin contain relatively low amounts and bioavailability is only approximately 15%, bioactive compounds can evoke metabolic effects even at very small concentrations41. Melatonin acts on the melatonin-1-3 receptors (MT1,

MT2, and MT3), present on numerous human cells, including but not limited to: brain, retina, cardiac ventricular wall, aorta, coronary and cerebral arteries, liver, gallbladder, duodenal enterocytes, colon, skin, parotid gland, exocrine pancreas, kidney, immune cells, platelets, adipocytes, prostate and breast epithelial cells, ovary/granulosa cells, and myometrium42,43. Melatonin is important in promoting sleep and maintaining circadian rhythm, which affects many biological processes.

Sleep and circadian rhythm also play a role in cardiovascular risk, including body weight status and blood pressure regulation. There is a growing body of evidence examining the correlation between sleep deprivation and greater risk of obesity. Studies to date suggest that increasing melatonin concentrations via over-the-counter supplements in doses ranging from 1-10 mg44, may improve sleep outcomes and also lower obesity risk, but further investigation is needed45,46. The role that melatonin plays in circadian rhythm regulation also affects nighttime blood pressure and blood pressure dipping47–49, which is associated with cardiovascular risk50. Nighttime blood pressure is typically 10-20% lower than daytime blood pressure; however, about 30% of

10 people in the US do not experience this reduction and are at increased risk for having a cardiovascular event50. Two clinical trials examined the effect of 2-3 mg/day melatonin treatment in hypertensive adults for 3-4 weeks and reported that administration lowered nocturnal blood pressure. However, one study reported an increase in daytime blood pressure coinciding with a decrease in nighttime blood pressure in adults with coronary artery disease after 5 mg/day melatonin treatment for 90 days. Supplemental melatonin administration may only be beneficial in lower doses in individuals at risk for CVD and the safety of consuming higher doses should continue to be evaluated.

Although melatonin is produced endogenously, increased concentrations through dietary intervention may provide benefits. The effect of melatonin derived from consumption of a melatonin-rich dietary pattern on body composition and vascular measurements has not been examined; clinical trials are needed to evaluate both safety and efficacy. A review of the Mediterranean-style diet suggested melatonin may be a key factor that contributes to chronic disease risk- reducing benefits associated with the Mediterranean diet51. The Mediterranean diet contains several foods that contain melatonin including: wine, extra-virgin olive oil, tomatoes, and tree nuts. Dietary melatonin may reduce cardiovascular risk; however, the effects of a melatonin-rich diet have not been evaluated.

2.2 Evidence for walnuts reducing established and emerging cardiovascular risk factors

2.2.1 Epidemiological evidence for nut intake and CVD2

Epidemiological evidence has consistently demonstrated a beneficial association between nut consumption and heart disease risk and mortality52–55.

The four landmark studies that laid the ground work for understanding the cardiovascular benefits of nut consumption are outlined here and detailed in

Table 2-2. The Adventist Health Study was the first study to investigate nut consumption and heart disease risk in 31,208 healthy individuals52. In this landmark study, there were fewer fatal coronary events [relative risk (RR) = 0.52;

95% Confidence Interval (CI): 0.36, 0.76; P < 0.0001)] and fewer nonfatal myocardial infarctions (RR = 0.49; 95% CI: 0.28, 0.85; P < 0.005) in high nut consumers (type of nut undefined; >5 times/week) compared to low nut consumers (<1 time/week). The Iowa Women’s Health Study examined the frequency of nut intake (type of nut undefined) and heart disease death in a population of healthy postmenopausal women (n = 34,486). The authors reported that women in the highest nut consumption quartile (> 4 servings/week) had a

40% reduction in risk of fatal heart disease [RR = 0.60 (95% CI: 0.36,1.01; P =

0.016)] compared to women who did not consume nuts after a 7 year follow-up56.

______2Sentences in this section (2.2.1.) were extracted, with permission, from a book chapter titled “Nut Consumption and Coronary Heart Disease Risk and Mortality” in: Bergeron N, Siri-Tarino P, editors. Nutrition and Cardiometabolic Health. Boca Raton (FL): CRC Press; 2018. P. 449-480. Alyssa Tindall wrote sentences selected from the chapter in their entirety.

The Nurses’ Health Study (NHS) was another large, prospective study (n =

86,016) that evaluated the association between nut consumption and heart disease in middle-aged women54. The authors found a 39% decrease in risk of fatal heart disease [RR = 0.61 (95% CI: 0.35, 1.05; P = 0.007)] in high nut consumers (>5 servings nuts/week; type of nut not defined) compared to low nut consumers (<1 serving nuts/month)54. The Physicians’ Health Study reported similar results in 22,071 U.S. males55. The authors reported that men who consumed nuts (type of nut undefined) >2 times/week had a reduced risk of total heart disease death [RR = 0.53 (95% CI: 0.30, 0.92; P = 0.01)] compared to men who rarely or never consumed nuts after a 17 year follow-up. The risk reduction was largely attributable to the 47% reduction in risk of sudden cardiac death in higher nut consumers [RR = 0.53 (95% CI: 0.30, 0.92; P = 0.01])55. A pooled- analysis of these four large, U.S. cohort studies demonstrated a 35% reduced risk heart disease incidence for the group consuming nuts > 5 times/week [RR =

0.65; (95% CI: 0.47, 0.89)] and the non-fatal heart disease RR was 0.68 (95% CI:

0.47, 1.00)57. Collectively, the results from these studies demonstrate a dose- response resulting in 8.3% risk reduction in heart disease for each serving of nuts per week and overall, protective benefits of nut consumption on heart disease58.

More recent epidemiological studies mirror the favorable correlation between nut intake and reduced CVD risk reported in the first four landmark studies. The majority of prospective cohorts have been in Western countries, but recent, similar results have been reported in other countries, such as Iran and

Croatia59,60. A meta-analysis of 18 prospective studies reported the RRs for high

(≥ 3 servings/week) compared with low nut consumption were 0.81 (95% CI:

0.78, 0.84) for all-cause mortality, 0.75 (95% CI: 0.71, 0.79) for CVD mortality,

0.73 (95% CI: 0.67, 0.80) for CHD mortality, 0.82 (95% CI: 0.73, 0.91) for stroke mortality, and 0.87 (95% CI: 0.80, 0.93) for cancer mortality61. Most recently,

Guasch-Ferré et al.62 conducted an analysis including 76,364 women from the

Nurses' Health Study (1980 to 2012), 92,946 women from the Nurses' Health

Study II (1991 to 2013), and 41,526 men from the Health Professionals Follow-

Up Study (1986 to 2012) to examine the associations between the intake of total and specific types of nuts and CVD, heart disease, and stroke risk. The authors reported the pooled multivariable hazard ratios for CVD and heart disease among participants who consumed one serving of nuts (28 g) ≥5/week, compared with individuals who never or almost never consumed nuts, were 0.86

(95% CI: 0.79, 0.93; P for trend = 0.0002) and 0.80 (95% CI: 0.72, 0.89; P for trend <0.001), respectively. Consumption of peanuts and tree nuts (≥2/week) and walnuts (≥1/week) was associated with a 13% to 19% lower risk of total CVD and

15% to 23% lower risk of heart disease62. The epidemiological evidence remains strong for the association between nut consumption and reduced cardiovascular risk.

Table 2-2. Details the landmark studies that show an association between nut consumption and CVD mortality

Reference Study Population Nut type Follow-up Main Results Albert et The 22,071 Nuts 17 years RR= 0.53 (95%CI: 0.30, 0.92; P=0.01) for fatal al. 200255 Physicians’ healthy men (undefined) and 47% decrease in sudden death Health Comparison: Study >2 servings/week compared to never or <1 time/month No association for risk of diabetes. HR = 0.87(95%CI: 0.61, 1.24; P=0.99) Ellsworth Iowa 34,486 Nuts 7 years RR= 0.60 (95% CI: 0.36, 1.01; P=0.016) of fatal et al. Women’s healthy post- (undefined) heart disease 200156 Health menopausal Comparison: Study woman >4 times/week compared to 0 times/week Hu et al. Nurses’ 86, 016 1.) Nuts 14 years RR = 0.61 (95% CI: 0.35, 1.05; P=0.007) for CVD 199854 Health healthy (undefined) risk Study middle-aged Comparison: women >5 servings /week of compared to <1serving RR=0.92 (95%CI:0.74, 1.15; P=0.094) for heart 2. Peanuts disease risk RR=0.76 (95%CI: 0.51, 1.15;P=0.09) for fatal heart disease Comparison: >5 times/week compared to almost never Fraser et The 31,208 Nuts 6 years RR= 0.52 (95% CI: 0.37, 0.76; P<0.05) for non- al. 199252 Adventist healthy men (undefined) fatal and 0.49 (95%CI: 0.28, 0.85; P=0.005) Health and women Comparison: >5 times/week compared to <1 Study time/week

2.2.2 Experimental evidence for walnut consumption and CVD risk

2.2.2.1 Blood lipid profile

There is strong evidence that the replacement of SFA with PUFA has been shown to improve blood lipids and lipoproteins and reduce CVD risk63–66.

The high PUFA content of walnuts makes them an ideal SFA replacement and the evidence for an effect of walnut consumption on lipids and lipoproteins is also robust (Table 2-3). In a landmark study conducted by Sabate et al. in 199367,

SFA were decreased (6% energy on the walnut diet versus 9% on the control diet) and walnut-derived PUFA were increased (16.5% energy versus 9.5%), resulting in a 16% reduction in LDL-cholesterol. Over the last decade, two meta- analyses have been completed on the effect of walnut consumption on lipids and lipoproteins and both reported similar findings68,69. In 2009, Banel and Hu68 reported the results of a meta-analysis and systematic review of 13 studies

(n=365) that evaluated the effects of walnut consumption on blood lipids and lipoproteins. Various cholesterol-lowering diets were evaluated and walnut diets low in SFA and high in PUFA were compared with average American/Western diets, Mediterranean diets, and low-fat, high carbohydrate diets. All studies reported a cholesterol-lowering effect of the walnut diets, and the average LDL-C lowering response was 9.2 mg/dL68. Recently, Guasch-Ferré et al.69 updated their original meta-analysis and systematic review with 26 clinical trials that included 1059 participants and evaluated the effects of walnut consumption on blood lipids and other cardiovascular risk factors. The authors reported walnut- enriched diets reduced a variety of lipids and lipoproteins compared with control 16 groups including: −6.99 mg/dL (95% CI: −9.39, −4.58 mg/dL; P < 0.001; 3.25% greater reduction) for TC, −5.51 mg/dL (95% CI: −7.72, −3.29 mg/dL; P < 0.001;

3.73% greater reduction) for LDL-C, and −4.69 (95% CI: −8.93, −0.45; P = 0.03;

5.52% greater reduction) for TG concentrations. There were even more pronounced reductions in blood lipids when walnut interventions were compared with Western-style diets [−12.30 mg/dL (95% CI: −23.17, −1.43; P<0.01) for TC,

−8.28 mg/dL (95% CI: −13.04, −3.51; P < 0.01) for LDL-C, and −3.74 mg/dL

(95% CI: −6.51, −0.97); P = 0.008 for apolipoprotein B]69. The authors also noted that walnut-enriched diets did not lead to significant differences in weight change compared with control diets [−0.12 kg (95% CI: −2.12, 1.88); P = 0.90]69.

Evidence continues to demonstrate that walnut consumption reduces LDL-C and lowers CVD risk without adversely affecting weight status.

There are several proposed mechanisms for the improvements in blood lipid profile observed with walnut consumption. Zhang et al.70 showed that walnut oil significantly increased cholesterol efflux through decreasing the expression of lipogenic enzyme stearoyl CoA desaturase 1 (SCD1) in macrophage-derived foam cells. Further, the authors reported that ALA increased farnesyl X receptor

(FXR) messenger ribonucleic acid (mRNA) expression, which increased expression of its gene target, small heterodiamer partner (SHP), and resulted in over-expression of SHP and reduction in SCD1 expression. ALA also may act through alternate pathways. An in-vivo study showed that treatment of adipocytes with ALA can affect the signal transduction of cholesterol and TG biosynthesis through suppression of sterol regulatory element-binding protein

(SREBP)-1c, SREBP-1a, and SREBP-2 (Figure 2-1)71. Although the precise mechanisms are not completely understood, there is a large pool of evidence to support the favorable effects of walnut consumption on lipids and lipoproteins.

Figure 2-1. One potential mechanism by which α-linolenic acid (ALA) reduces plasma cholesterol levels. ALA treatment of adipocytes reduced expression of SREBPs (SREBP-1c, SREBP-1a, SREBP-2), which could suppresses the expression of genes related to the de novo cholesterol and triacylglycerol biosynthesis (SC5D, TM7SF2, CYP51, HMGCS1, SQLE, ACSL3, ABCA1, ACSS2). ALA also enhanced the expression of genes related to the fatty acid oxidation (CPT-1a, leptin) in adipocytes. Source: Fukumitsu et al., 201371

Table 2-3. Effect of walnut consumption on lipids and lipoproteins (RCTs)1

Reference Study Subjects Treatment Control Diet Duration Results Design Diet(s) Rock et al. Parallel Overweight and Walnut- Reduced- 6 months Reduction in TC Nutr J 2017 obese men and enriched diet energy diet + (P<0.05) and LDL-C women (n=100) (15% energy) behavioral (P<0.05) following reduced- weight loss the walnut-enriched energy diet + intervention diet behavioral weight loss intervention Zibaeenezhad Parallel Hyperlipidemic adults 15 mL walnut 15 mL distilled 90 days Reduction in TC (- et al. Nutr with T2DM (n=11) oil/day water/day 30.04, P<0.001), TG Diabetes 2017 (-15.04, P=0.021), LDL-C (-30.44, P<0.001), TC:HDL-C (-0.72, P<0.001) compared to the control group Wu et al. Crossover Healthy adults (n=40) 43 g Nut-free 8 weeks Reduction in non- Metabolism walnuts/day Western-type HDL-C (-10±3 mg/dL, 2014 diet P=0.025) and apolipoprotein-B (- 5±1.3 mg/dL, P=0.009) compared to control Aronis et al. Crossover Obese adults with 48 g Nut-free diet 4 days Increase in Metabolism metabolic syndrome walnuts/day apolipoprotein A 2012 (n=9) (P=0.03) following the walnut diet 19

Canales et al. Crossover Overweight and 150 g walnut 150 g low-fat 5 weeks Reduced Eur J Clin Nutr obese men and post- paste (in steak (in paraoxonase-1/HDL- 2011 menopausal women restructured restructured C (P<0.001) and at risk for CVD (n=22) meat) meat) paraoxonase-1/Apo A1 (P<0.001) ratios Kalgankar et Parallel PCOS patients (n=31) 31 g No control 6 weeks Reduction in LDL-C al. Eur J Clin walnuts/day or (-6%, P=0.05) and Nutr 2011 31 g apolipoprotein-B almonds/day (11%, P<0.03) following the walnut diet Torabian et al. Crossover Adults with normal to 28-64 g (12% Nut-free diet 6 months Reduction in TC Eur J Clin Nutr moderately high TC energy) (P=0.01) and TG 2010 (n=87) walnuts/day (0.01) following the walnut diet Rajaram et al. Crossover, Normal to mildly 42.5 g No fish or nuts 4 weeks Reduction in TC and Am J Clin Nutr feeding hyperlipidemic adults walnuts/day (30% total fat LDL-C following the 2009 (n=25) (30% total fat and <10% the walnut diet and <10% SFA) compared to the SFA) or 113 g control and the fish salmon diet (all P<0.0001) twice/week and the reduction in (30% total fat TC:HDL-C, LDL- and <10% C:HDL-C, SFA) apolipoprotein- B:apolipoprotein A1 were lower following the walnut diet compared to the control and the fish

diet (P<0.05)

Olmedilla- Crossover Adults at risk for CVD 120 g walnuts Restructured 5 weeks Reduction in TC (-6.8 Alanso et al. J (n=25) within meat without mg/dL, P=0.027) Am Coll Nutr restructured walnuts following the walnut 2008 meat/week diet compared to control Mukuddem- Crossover, Adults with MetS 63-108 g (20% Nut-free diet 8 weeks There were no Petersen et al. feeding (n=64) energy) significant changes in Br J Nutr 2007 walnuts/day or lipids and 63-108 g (20% lipoproteins energy) or cashews/day Schutte et al. Parallel Adults with metabolic 63-108 g (20% Habitual diet 8 weeks No changes in HDL- Am J syndrome (n=62) energy) C or TG Hypertens walnuts/day or 2006 63-108 g (20% energy) or cashews/day Zibaeenezhad Parallel Adults with TG>350 20 g Nut-free diet 8 weeks Reduction in TG (- et al. mg/dL or TC>250 walnuts/day 17.1, P= 0.02) and Angiology mg/dL (n=52) an increase in HDL-C 2005 (9%, P= 0.03) following the walnut 21

diet

Tapsell et al. Parallel Adults with T2DM 30 g Usual diet 6 months Increase in HDL- Diabetes Care (n=58) walnuts/day + (low-fat diet) C:TC (P=0.049) and 2004 dietary advice HDL (P=0.046) (<30% energy compared to the two from fat) or other groups and dietary advice 10% reduction in (<30% energy LDL-C following the from fat) devoid walnut diet (P=0.032) of walnuts Ros et al. Crossover Hypercholesterolemic Walnut- Mediterranean 4 weeks Reduction in TC (- Circulation adults (n=21) enriched diet diet 4.4±7.4%, P<0.05) 2004 (replaced 32% and LDL-C (- of MUFA in a 6.4±10.0%, P<0.05) Mediterranean diet) Zibaeenezhad Parallel Hyperlipidemic adults 3 g walnut oil Placebo (not 45 days Reduction in TG (- et al. (n=60) (capsules) detailed) 23%, P<0.05) Angiology following the walnut 2003 oil treatment Iwamoto et al. Crossover, Healthy adults (n=40) 44-58 g (12.5% Isocaloric 4 weeks Reduction in LDL-C Eur J Clin Nutr feeding energy) Japanese diet (-8.5 mg/dL, 2002 walnuts/day in devoid of P=0.0008) in women a Japanese walnuts following the walnut diet diet

Muñoz et al. J Crossover, Men with polygenic 41-56 g (35% Mediterranean 6 weeks Reduction in TC (- Lipid Res 2001 feeding hypercholesterolemia energy) diet 4.2%) and LDL-C (- (n=10) walnuts/day in 6%), but was not a statisitcally significant Mediterranean (P=0.176 and diet P=0.087). LDL uptake by Hep2G cells was correlated with the ALA-content in the LDL particle (r²=0.42, P<0.05) Zambón et al. Crossover, Men and women with 41-56 g (35% Mediterranean 6 weeks Reduction in TC (- Ann Intern feeding polygenic energy) diet 10.8 mg/dL, Med 2000 hypercholesterolemia walnuts/day in P<0.001), LDL-C (- (n=59) a 11.2 mg/dL, Mediterranean P<0.001) and diet lipoprotein A (-0.021 g/L, P=0.42) Chisholm et al. Crossover Dyslipidemic men 78 g Nut-free diet 4 weeks Reduction in TC (-9.7 Eur J Clin Nutr (n=21) walnuts/day (30% total mg/dL, P<0.05), LDL- 1998 (average) (38% energy from C (-13.9 mg/dL, total energy fat) P<0.01) and an from fat) increase HDL-C (5.8 mg/dL, P<0.01) following the walnut diet and an increase in HDL-C (4.6 mg/dL, P<0.01) following the control diet; Apolipoprotein B100

was lower following the walnut diet compared to the control (-13.3 vs. - 8.34 mg/dL, P<0.05)

Sabaté et al. N Crossover, Healthy men (n=18) 20% energy Step 1 diet 4 weeks Reduction in TC (- Engl J Med feeding from devoid of 22.4 mg/dL, 1993 walnuts/day in walnuts P<0.001), LDL-C (- a Step 1 diet 18.2 mg/dL, P<0.001), HDL-C (- 2.3 mg/dL, P=0.009), and LDL-C:HDL-C (- 0.3, P<0.001) following the walnut diet compared to the reference diet

1Search terms used in PubMed included: (("juglans"[MeSH Terms] OR "juglans"[All Fields] OR "walnuts"[All Fields]) AND ("lipids"[MeSH Terms] OR "lipids"[All Fields])) and ("lipoproteins"[MeSH Terms] OR "lipoproteins"[All Fields]) AND (Clinical Trial[ptyp] AND "humans"[MeSH Terms]). Post-prandial studies, studies that did not include lipid/lipoprotein endpoints, studies that were not controlled, and studies that provided walnuts with other nuts (mixed nuts) were excluded. Mukuddem-Petersen et al. was added to the table although it did not appear in the initial search.

2.2.2.2 Vascular health

Brachial blood pressure is the most commonly assessed measure of vascular health and even small elevations above optimal (<120/80 mmHg) result in an increased the risk of a cardiovascular event72. One meta-analysis, including

21 RCTs (n=1652), assessed the effects of tree nut, peanut, and soy nut consumption for more than two weeks on blood pressure and reported 30-108 g nut consumption reduced brachial systolic blood pressure (SBP; -1.29 mmHg;

95% CI: -2.35, -0.22; P=0.02) in participants free of T2DM73. A recent meta- analysis included 26 RCTs (n=1059) and examined the effects of walnuts on cardiovascular risk factors, including blood pressure69. The authors included eight RCTs in the analysis that tested blood pressure as an endpoint and did not observe significant differences in blood pressure following a walnut-enriched diet compared to a control diet. Only one of the eight trials included in the meta- analysis reported the effect of walnut consumption on central blood pressure and this study did not observe a significant change following consumption of a walnut-enriched diet74. Central (or aortic) blood pressure reflects the blood pressure experienced by organs and is more predictive of mortality than brachial blood pressure in clinical studies75,76. Central blood pressure can be algorithmically computed using brachial blood pressure and has been validated against an indwelling catheter77. There are overall, beneficial effects of nut consumption on brachial blood pressure, but results range from null to favorable for the effect of walnut consumption on brachial and central blood pressure

(Table 2-4).

Arterial stiffness can be estimated locally or regionally using pulse wave velocity (PWV), the speed at which the forward pressure wave is transmitted through the vascular tree. Carotid to femoral PWV is considered the gold standard for assessing arterial stiffness and can be measured noninvasively78.

There is limited evidence on the effect of nut consumption on arterial stiffness and only one clinical trial has assessed the effect of walnut consumption on

PWV74. Din et al.74 reported no significant changes in PWV following four weeks walnut supplementation (15 g/day) compared to control (no walnuts). The authors suggest that the amount of walnuts may not be sufficient to improve

PWV. At this point, there is not conclusive evidence that walnut consumption affects arterial stiffness.

Arterial endothelial dysfunction is one of the early events in atherogenesis, preceding structural atherosclerotic changes and increasing CVD risk. The gold standard for assessing vascular function is flow-mediated dilation (FMD), which uses an ultrasound to measure brachial artery dilation following a transient period of forearm ischemia. The prognostic strength of FMD was summarized in a meta- analysis that reported a 1% increase in FMD is equated to a 13% decrease in cardiovascular event risk79. A recent meta-analysis of 10 randomized, controlled trials (n=374) reported that nut consumption improved FMD (0.41%; 95% CI:

0.18%, 0.63%; P=0.001) and subgroup analyses showed that walnuts improved

FMD (0.39%; 95% CI: 0.16%, 0.63%; P=0.001)80. Three trials examined the effects of walnuts on FMD81–83; two trials were free living that provided 56 g walnuts/day, but the other study was a highly-controlled feeding trial that

26 provided walnuts as 16.4% total energy. The authors reported a 34% increase

(P=0.05) following a diet containing 37 g walnuts and 15 g walnut oil for six weeks compared to a typical American diet devoid of walnuts and walnut oil

(n=12)84. Consistent evidence has shown walnuts have a favorable effect on endothelial function assessed using FMD.

There are several proposed mechanisms by which walnuts affect blood pressure and endothelial function, but none are well-characterized. In a meta- analysis on nut consumption and blood pressure, Mohammadifard et al.73 suggested that the anti-hypertensive effects of nuts may depend on the non-fatty acid compounds, such as dietary fiber, plant proteins, antioxidants, bioactives, vitamins and minerals. Walnuts contain γ-tocopherol, a potent antioxidant that could reduce inflammatory markers, such as C-reactive protein and IL-6, and oxidative stress, leading to a reduction in oxidized LDL-C and improved endothelial function73,85. Walnuts also contain magnesium [45 mg/28 g (45 mg/1oz)] which has also been shown to reduce blood pressure as a dietary supplement86. A meta-analysis of 22 trials (n=1173) reported that 120-973 mg magnesium supplementation for 3-24 weeks results in a small, but clinically significant (3-4 mmHg reduction in SBP; 2-3 mmHg reduction in DBP). Although one oz of walnuts only contains 11% of the mean supplement dose reported in the meta-analysis, the magnesium content of walnuts may be a contributor to the blood pressure lowering effects observed in clinical trials81,87. Walnut consumption has been shown to improve markers of vascular health, but the precise mechanisms are not well understood.

Table 2-4. Walnut consumption and blood pressure (RCTs)1

Reference Study Subjects Treatment Diet(s) Control Diet Duration Results Design Rock et al. Parallel Overweight and Reduced-energy Reduced- 6 months Reduction in SBP (-6 Nutr J 2017 obese adults diet with 28-42 g energy diet mmHg, P<0.05) and (n=100) (15% energy) devoid of DBP (-5 mmHg, walnuts/day walnuts P<0.05) following the walnut diet Tapsell et al. Parallel Overweight and Multi-disciplinary Usual care 12 months No statistically BMJ 2017 obese adults intervention + 30 g (control; significant (n=178) walnuts/day or general diet differences in multi-disciplinary and exercise systolic blood intervention devoid advice) pressure of nuts Katz et al. J Crossover Overweight adults 56 g walnuts/day Nut-free diet 8 weeks Increase in FMD Am Coll Nutr at risk for metabolic following the walnut syndrome (n=46) diet (1.4±2.4%) compared to control (0.3±1.5%, P=0.019) Din et al. Eur Crossover Healthy men (n=30) 15 g walnuts/day Habitual diet 4 weeks No statistically J Clin Nutr significant 2011 differences in central BP, brachial BP, or arterial stiffness

West et al. J Crossover, Hypercholesterolem Walnuts + walnut Average 6 weeks Both walnut diets Am Coll Nutr feeding ic adults (n=20; oil (16.4% energy) American reduced diastolic 2010 FMD n=12) or walnuts + walnut diet (8.7% blood pressure (-2-3 oil + flax oil (17% energy from mmHg, P<0.05) and energy) PUFA) total peripheral devoid of resitance (-4%, nuts and nut- P<0.02); the walnut oils + walnut oil + flax increased FMD (34%, P=0.02) Wu et al. J Parallel Adults with MetS or Lifestyle counseling Lifestyle 12 weeks Reduction in SBP (- Nutr 2010 at risk for MetS + 30 g walnuts/day counseling 7.0 to -8.8 mmHg, (n=239) or lifestyle P<0.05) and DBP (- counseling + 4.2 to -5.0 mmHg, 30 g P<0.05) following all flaxseed/day interventions Olmedilla- Crossover Adults at risk for 120 g walnuts Restructured 5 weeks Reduction in SBP (- Alanso J Am CVD (n=25) within restructured meat without 10.8 mmHg, Coll Nutr meat/week walnuts P=0.001) following 2008 the control; no between-group differences Mukuddem- Crossover, Adults with MetS 63-108 g (20% Nut-free diet 8 weeks No statistically Petersen et feeding (n=64) energy) significant changes al. Br J Nutr walnuts/day or 63- in BP 2007 108 g (20% energy) or cashews/day

Sabaté et al. Crossover, Healthy men (n=18) 20% energy from Step 1 diet 4 weeks No statistically N Engl J feeding walnuts/day in a devoid of significant changes Med 1993 Step 1 diet walnuts in BP

1Search terms used in PubMed: ("juglans"[MeSH Terms] OR "juglans"[All Fields] OR "walnuts"[All Fields]) AND ("blood pressure"[MeSH Terms] OR ("blood"[All Fields] AND "pressure"[All Fields]) OR "blood pressure"[All Fields] OR "blood pressure determination"[MeSH Terms] OR ("blood"[All Fields] AND "pressure"[All Fields] AND "determination"[All Fields]) OR "blood pressure determination"[All Fields] OR ("blood"[All Fields] AND "pressure"[All Fields]) OR "blood pressure"[All Fields] OR "arterial pressure"[MeSH Terms] OR ("arterial"[All Fields] AND "pressure"[All Fields]) OR "arterial pressure"[All Fields] OR ("blood"[All Fields] AND "pressure"[All Fields])) AND (Clinical Trial[ptyp] AND "humans"[MeSH Terms]). Post- prandial studies, studies that did not include blood pressure-related endpoints, retracted studies, and studies that provided walnuts with other nuts (mixed nuts) were excluded.

2.2.2.3 The gut microbiome3

The gut microbiome is markedly influenced by diet and plays a major role in human health88–90. An imbalance of unfavorable to beneficial gut microbes, or gut dysbisosis, has been implicated in a variety of disease states, including obesity, inflammatory bowel disease (IBD), Crohn’s disease, diabetes, and

CVD88,90–93. However, the dynamic interplay between diet, microbial species, and human metabolism is not well-defined.

Numerous studies have reported alterations in the gut microbiome in response to dietary interventions94–100. Two recent human clinical trials have examined the effect of walnut consumption on the gut microbiome composition

(Table 2-5). Holscher et al.101 analyzed fecal samples from healthy men and women provided isocaloric diets containing 0 or 42 g (~1.5 oz) walnuts/day for three weeks in a randomized, crossover, controlled-feeding study (n=18). The authors reported a higher relative abundance of Faecalibacterium, Clostridium,

Dialister, and Roseburia and lower relative abundances of Ruminococcus,

Dorea, Oscillospira, and Bifidobacterium after the walnut diet compared to the control diet. Bamberger and colleagues102 evaluated the gut microbiome from healthy adults supplied 0 or 43 g (~1.5 oz) walnuts/day for eight weeks in a

______3Sentences in this section (2.2.3.- Microbiome) were extracted, with permission, from a published article in the Journal of Nutrition [Tindall AM, Petersen KS, Kris- Etherton PM. Dietary Patterns Affect the Gut Microbiome – The Link to Cardiometabolic Diseases. Journal of Nutrition. 2018; 148(9):1402-1407.] Alyssa Tindall wrote sentences selected from the publication in their entirety. The final publication is available at https://doi.org/10.1093/jn/nxy141.

31 randomized, crossover, free-living study (n=135). The authors reported that the abundance of Ruminococcaceae and Bifidobacteria increased and Clostridium

(cluster XIVa) decreased compared to the walnut-free diet. These differences may be explained by the differential background diets in these studies.

The gut microbiome is a relatively new target for CVD risk-reduction and there is limited evidence about the effects of walnut consumption on the microbial ecosystem. However, emerging evidence suggests that vegetarian and

Mediterranean dietary patterns favorably affect both the gut microbiome and reduce CVD risk103. These dietary patterns have been shown to favorably affect gut-derived metabolites that are also associated with CVD risk, such as trimethylamine (TMA) and short-chain fatty acids (SCFA; Figure 2-2)103. The microbiota present in the lower gastrointestinal tract metabolize indigestible compounds, such as polyphenols, through converting them into more bioavailable metabolites, which makes bacteria likely contributors to conferring the beneficial health effects associated with polyphenol consumption104,105.

Walnuts are found in both vegetarian and Mediterranean diets and may favorably affect the gut microbiome, particularly due to their polyphenol content.

Figure 2-2. Known and proposed interactions between diet and cardiometabolic risk. The Figure illustrates the known risk-reducing properties of Vegetarian and Mediterranean diets on cardiometabolic diseases and the proposed interactions between the Vegetarian and Mediterranean diets with the gut microbiome and gut-derived metabolites that can reduce cardiovascular disease risk. Source: Tindall et al., 2018103

Table 2-5. Effects of walnuts on the gut microbiome (RCTs)1

Reference Study Subjects Treatment Control Duration Results Design Diet(s) Diet Bamberger Crossover Healthy 43 g/day Nut- 8 weeks Walnuts significantly increased Ruminococcaceae et al., 2018 men and walnuts free and Bifidobacteria and decreased Clostridium spp. women diet (n=135) Holscher Crossover, Healthy 42 g/day Nut- 3 weeks No significant differences in diversity et al., 2018 feeding adults walnuts free Walnuts significantly increased Firmicutes and (n=18) diet decreased Actinobacteria Walnuts significantly increased Faecalibacterium, Clostridium, Roseburia, and Dialister No significant changes in primary bile acids Walnuts significantly reduced secondary bile acids

1Search terms used in PubMed: (("juglans"[MeSH Terms] OR "juglans"[All Fields] OR "walnuts"[All Fields]) AND ("microbiota"[MeSH Terms] OR "microbiota"[All Fields] OR "microbiome"[All Fields])) AND (Clinical Trial[ptyp] AND "humans"[MeSH Terms]). Studies that did not include gut microbiome-related endpoints were excluded. Holscher et al. was added to the table although it did not appear in the initial search.

2.3 Rationale for current research

There is substantial evidence that a diet containing PUFA-rich walnuts reduces CVD risk, but the mechanisms remain incompletely understood. The proposed mechanisms by which walnut consumption improves the lipid/lipoprotein profile largely focus on ALA; however, the bioactive profile of walnuts may provide added benefits. The bioactives in walnuts have also been implicated in the improvement of brachial blood pressure and arterial health. It is important to determine whether the individual or combined effect of the bioactives and ALA content of walnuts play a critical role in CVD risk reduction to better understand the underlying mechanisms and provide evidence for strengthening dietary recommendations.

Given the importance of the microbiome in nutrition and health, and its emerging role as a CVD risk factor, it is critical to understand how differing fatty acid and bioactive profiles may modulate the gut microbial community. The phenolic compounds in walnuts may favorably affect the presence or absence of gut bacteria that could affect the production of gut-derived metabolites. It is important to determine if the walnut bioactives, beyond ALA, promote favorable changes in the gut microbiome.

The majority of studies that have examined the effect of walnut consumption on cardiovascular risk are in free-living individuals and subject to confounding variables. To our knowledge, the current study is the first controlled-feeding study to examine the effects of walnuts on lipids, lipoproteins, arterial stiffness, central blood pressure, and the gut microbiome. Examining these risk factors collectively will 35

provide important evidence to understand the role of walnut bioactives and ALA in reducing CVD risk factors.

Chapter 3 Replacing saturated fat with walnuts or vegetable oils improves central blood pressure and serum lipids in adults at risk for cardiovascular disease: a randomized, controlled-feeding trial

Background: Walnuts favorably affect cardiovascular risk factors. It is unclear whether the cardioprotective effects are attributable to the fatty acids, including α- linolenic acid (ALA), and/or the bioactives in walnuts.

Methods: A randomized controlled, 3‐period, crossover, feeding trial was conducted in individuals at risk for cardiovascular disease (n=45). Following a two-week

Standard Western Diet run-in (SWD; 12% SFA, 7% PUFA, 12% MUFA), participants consumed three isocaloric, weight maintenance diets for six weeks each: Walnut diet (WD; 7% SFA, 16% PUFA, 3% ALA, 9% MUFA); walnut FA-matched diet

(WFMD; provided the same FAs as the WD, but devoid of walnuts); oleic acid replaced ALA diet (ORAD; 7% SFA, 14% PUFA, 0.5% ALA, 12% MUFA), was devoid of walnuts and replaced the ALA provided by walnuts in the WD with oleic acid. This design enabled evaluating the effects of whole walnuts (FAs and bioactives) with two constituent components (total FAs and ALA). The SphygmoCor

XCEL system was used to assess central blood pressure and arterial stiffness.

Results: There was a treatment effect (P=0.04) for central diastolic blood pressure

(cDBP). There was a reduction in cSBP from following the WD (P=0.04) and greater change versus ORAD (-1.78±1.0 vs. 0.15±0.7 mmHg, P=0.04), but no differences between the WD and WFMD (-0.22±0.8 mmHg, P=0.20) or the WFMD and ORAD

(P=0.74). The WD significantly lowered brachial and central mean arterial pressure, and total cholesterol (TC):high-density lipoprotein-cholesterol (HDL-C). All diets

lowered TC, LDL-C, HDL-C, and non-HDL-C versus baseline. There were no significant changes in arterial stiffness.

Conclusions: Cardiovascular benefits were achieved with all moderate-fat, high- unsaturated fat diets. As part of a low-SFA diet, the greater improvement in cDBP following the WD versus ORAD indicates unique benefits of walnuts as a whole food-replacement for SFA.

Clinical Trial Registration: URL: https://clinicaltrials.gov. Unique Identifier:

NCT02210767

Key Words: Walnuts, blood pressure, MUFA, PUFA, cardiovascular disease

Introduction

Deaths attributable to high BP rose by 37.5% from 2005 to 2015 in the

U.S.106 and elevated BP increases atherosclerotic CVD risk107. Normal BP is defined as less than 120/80 mmHg108 and predicts lower risk of major adverse cardiovascular events (MACE)72. Central BP, also termed aortic pressure, more directly indicates risk of MACE compared to brachial BP, and thus is also a target for reducing CVD risk75,76. A healthy diet is important for prevention and management of elevated BP and simple and effective dietary recommendations are needed.

Walnuts are a rich source of polyunsaturated fatty acids (PUFA) including the n-3 PUFA, ALA, and n-6 PUFA, linoleic acid (LA), and contain a proportionally greater quantity of polyphenolic compounds compared to other tree nuts and vegetable oils109,110. Observational findings show an association between ALA intake and lower brachial BP, and clinical studies have demonstrated that ALA reduces brachial BP8,111–113. Since walnuts are rich in PUFA, particularly n-3 PUFA, replacement of SFA with unsaturated fatty acids from walnuts is a potential strategy that aligns with current dietary guidelines and recommendations from authoritative organizations such as the American Heart Association, the National Lipid

Association, and the American College of Cardiology to reduce CVD risk6,114–116.

Walnut consumption has been shown to consistently lower LDL-C67–69,81,117, but less is known about effects on other CVD risk factors. For instance, a previous controlled feeding trial reported a reduction in brachial diastolic BP following six weeks of walnut consumption (37 g walnuts/day)81. However, it remains unclear how

walnut consumption affects central blood pressure and arterial stiffness. Further, previous clinical trials have not investigated whether the significant cardioprotective effects of walnuts are attributable to the fatty acid profile, including ALA, and/or the bioactive compounds they provide, such as tocopherols and phenolic compounds.

Thus, these findings support the need for controlled feeding studies testing higher walnut consumption in adults at risk for vascular disease due to elevated BP and

LDL-C to differentiate fatty acid effects from those possibly caused by bioactives.

The present controlled feeding study in adults with increased cardiovascular risk tested the effects of macronutrient-matched diets designed to reduce SFA

(according to the 2015-2020 Dietary Guidelines for Americans6) by replacing SFA with unsaturated fatty acids derived from: 1) 57-99 g (2-3.5 oz) whole walnuts; 2) vegetable oils providing an identical fatty acid profile as whole walnuts without walnut bioactives; and 3) monounsaturated fatty acids (MUFA) as oleic acid substituted for the ALA content of walnuts. We hypothesized the diet with whole walnuts would promote greater reductions in central BP and other CVD risk factors due to the combination of walnut fatty acids and bioactives.

Methods

Study Design and Randomization

This was a randomized, crossover, controlled-feeding trial conducted to examine the effect of a diet containing whole walnuts (WD), a diet with the same fatty acid composition as the WD but devoid of walnuts (walnut fatty-acid matched diet; WFMD), and a diet that replaced the amount of ALA that would come from walnuts with oleic acid (oleic acid replaces ALA diet; ORAD). All other nuts were restricted during the run-in and three diet periods. The study design is depicted in

Figure 3-1. Eligible participants completed a two-week run-in on a standard western diet (SWD). After completion of the run-in diet, baseline assessments were completed and participants were randomized using a six-sequence randomization scheme (computer generated using www.randomization.com in blocks of six). In random order, participants consumed each of three diets for six weeks, with compliance breaks after the first and second diet periods (mean=22.8 days; range=1-164 days). Investigators and study personnel involved in data collection and analysis were blinded to treatment allocation. Only the Metabolic Diet Study

Center (MDSC) manager was aware of the treatment allocation in order to assign participants to the correct menus.

The primary endpoint for the study was central systolic blood pressure

(cSBP). Secondary endpoints included: central diastolic blood pressure (cDBP), central and brachial mean arterial pressure (cMAP; bMAP), brachial systolic and diastolic blood pressure (bSBP; bDBP), central and brachial pulse pressure (cPP;

bPP), augmentation pressure (AP), augmentation index (AI), heart rate (HR), pulse transit time (PTT), pulse wave velocity (PWV), total cholesterol (TC), low-density lipoprotein-cholesterol (LDL-C), high-density lipoprotein-cholesterol (HDL-C), non-

HDL, TC:HDL-C, triglycerides (TG), glucose, insulin, and high-sensitivity C-reactive protein (CRP). All procedures involving human subjects were approved by the

Institutional Review Board of The Pennsylvania State University (University Park,

PA). Written informed consent was obtained from all participants prior to enrollment in the study. All study samples were collected and procedures were conducted at the

Penn State Clinical Research Center (CRC). This trial is registered at clinicaltrials.gov as NCT02210767.

Figure 3-1. Study design for this randomized, 3-period, crossover, controlled- feeding trial. ↑ indicate endpoint testing (vascular assessment and fasting blood draw completed on the last 2 days of the run-in diet and study diets). The average compliance break between diet periods was 23 days. All foods and caloric beverages were provided for all diet periods and the run-in diet.

Subjects Men and women with overweight and obesity [body mass index (BMI), 25-40 kg/m2], aged 30 to 65 years, and who had an LDL-C between the 50th and 90th percentile from a nationally representative sample118 (128-177 mg/dL for men and

121-172 for women) and/or with elevated brachial BP (120-159/80-99 mmHg; brachial systolic/brachial diastolic blood pressure) in central Pennsylvania were eligible for this study. Exclusion criteria included: smoking; blood pressure ≥160/100 mmHg; and a history of myocardial infarction, stroke, diabetes mellitus, liver disease, kidney disease, thyroid disease (unless controlled with medication), gastrointestinal diseases, and inflammatory diseases. Individuals taking the following supplements/medications were excluded unless they were willing to discontinue use prior to enrolling in and for the duration of the study: nutritional supplements (eg, whey protein powder), herbs, vitamins or minerals, nonsteroidal anti-inflammatory drugs, blood pressure-lowering medications, cholesterol-lowering medications/supplements (e.g., psyllium, fish oil capsules, soy lecithin, niacin, fiber, flax, phytoestrogens), and stanol/sterol supplemented foods. Women who were lactating, pregnant, or planning to become pregnant were excluded from the study.

Individuals who were following a vegetarian or vegan diet or had nut allergies were also excluded. Individuals who consumed >14 drinks per week prior to screening or who were not willing to maintain their physical activity habits throughout the course of the study were excluded.

Participant Recruitment

Participants were recruited from October 2014 to September 2017 via flyers posted on university and community bulletin boards, local newspaper advertisements, radio advertisements, campus e-mail lists, and a university research website. Potential subjects emailed or called to indicate interest in participating in the study and were then given additional information about the study. If interested, individuals were asked a series of medical history and lifestyle questions via a telephone screen to obtain information about key inclusion and exclusion criteria and determine eligibility. Following the telephone screening, an in-person screening appointment was scheduled at the CRC to confirm eligibility. After a 12-hour fast and avoidance of alcohol for 48-hours, individuals had their height, weight (without shoes and in light clothing), and resting blood pressure measured (validated sphygmomanometer) by trained nurses. The mean of the last two of three blood pressure readings was used to determine eligibility. A fasting blood sample was also collected for a blood chemistry measurement and analysis of lipids and lipoproteins.

Study Diets

Participants were fed isocaloric weight maintenance diets. Energy requirements were calculated using the Harris-Benedict equation119 and adjustments were made as needed throughout the course of the study. Diets were similar, except for the fatty acid profile and presence or absence of walnuts (Table 3-1). Differences in the nutrient profile of the diets are due entirely to the different fatty acids supplied

through the study foods (Table 3-2). The run-in diet consisted of a similar macronutrient profile as the test diets, but the amount of SFA provided was representative of typical SFA intake in the U.S. (12% of calories)4 and the diet was therefore proportionally lower in PUFA and MUFA compared to the study diets. The

WD included 18% daily energy from walnuts delivered as a snack (56.8-99 g/d, containing 5-8.8 g/d ALA) and matched the macronutrient profile of the two other test diets. The walnut fatty acid-matched diet (WFMD) provided the same amount of ALA and PUFA as the WD, but was devoid of walnuts (and walnut bioactives). In the oleic acid replaced ALA diet (ORAD), 83% of ALA was replaced with oleic acid, which is the amount of ALA delivered from walnuts in the WD. The ORAD was also devoid of walnuts and all other fatty acids were held constant.

All diets used the same six-day cycle menu, developed using Food Processor

SQL software, version 10.8 (ESHA Research, Salem, OR; version 10.8), and prepared in the Penn State MDSC. Participants picked up food daily Monday through Friday. On Fridays, participants were provided with a cooler containing

Saturday and Sunday meals and snacks. Compliance was assessed using daily food logs that participants completed and monitoring participant daily weight logs kept at the MDSC and a fatty acid analysis was completed using GCMS on a sub- sample of participants (n=5) to examine the plasma ALA concentration (Table 3-3).

Participants consumed non-caloric beverages ad libitum, but were limited to <1,184 mL/d caffeinated beverages (5 cups/d) and ≤2 alcoholic drinks per week.

Table 3-1. Nutrient composition of the study diets*

Nutrient SWD (run-in)1 WD WFMD ORAD Total fat, % 34 35 35 35 SFA, % 12 7 7 7 MUFA, % 12 9 9 12 PUFA, % 7 16 (2.7% ALA ) 16 (2.6% ALA) 14 (0.4% ALA) Carbohydrate, % 50 48 48 48 Protein, % 16 17 17 17 Fiber, g/day 25 30 26 26 Cholesterol, 202 117 169 163 mg/day

*All diets used the same six-day cycle menu, developed using Food Processor SQL software, version 10.8 (ESHA Research, Salem, OR). Total calories based on a 2100 calorie diet.

Abbreviations: standard western diet, SWD; walnut diet, WD; walnut fatty acid- matched diet, WFMD; oleic acid replaces α-linolenic acid diet, ORAD.

Table 3-2. Sample menu from one day of the six-day cycle menu

Run-in diet WD1 WFMD ORAD Breakfast 2% Milk Nonfat milk Nonfat milk Nonfat milk Granola Wheat English muffin Wheat English muffin Wheat English muffin Light yogurt Egg beaters Egg beaters Egg beaters Wheat bagel Orange Orange Orange Butter Margarine

Lunch Wheat bun Wheat bun Wheat bun Wheat bun Chicken breast Veggie burger Veggie burger Veggie burger Canola American cheese American cheese American cheese mayonnaise Chipolte spread Lettuce Lettuce Lettuce Fruit blend Dijon mustard Dijon mustard Dijon mustard Sun chips® Graham crackers Graham crackers Graham crackers Pear Pear Pear

Italian dressing

Sunflower oil

Dinner Vegetarian chili Chicken breast Chicken breast Chicken breast Thai noodles and Thai noodles and Thai noodles and Cheddar cheese veggies veggies veggies Lettuce Lettuce Lettuce Lettuce Carrots Carrots Carrots Carrots Cherry tomatoes Cherry tomatoes Cherry tomatoes Cherry tomatoes Light Italian Light Italian dressing Light Italian dressing Italian dressing dressing Corn bread muffin White dinner roll Flaxseed oil High oleic safflower oil High linoleic safflower High linoleic safflower oil oil Snack M&M's® Sun chips® Sun chips® Sun chips® Walnuts

1Abbreviations: walnut diet, WD; walnut fatty acid-matched diet, WFMD; oleic acid replaces α-linolenic acid diet, ORAD.

Table 3-3. Sub-sample total plasma fatty acid analysis (n=5)*

Oleic Acid ALA AA EPA DHA

C18:1n9 C18:3n3 C20:4n6 C20:5n3 C22:6n3 Run -in 13.99% 0.52% 7.41% 0.66% 2.36% WD 11.88% 1.04% 6.56% 0.79% 2.39% WFMD 10.09% 1.03% 6.73% 0.85% 2.57% ORAD 11.86% 0.42% 7.44% 0.35% 2.40%

*All area under the curve (AUC) values have been normalized to the internal standard

Abbreviations: Alpha-linolenic acid, ALA; eicosapentaenoic acid, EPA; docosahexaenoic acid, DHA; walnut diet, WD; walnut fatty acid-matched diet, WFMD; oleic acid replaces α-linolenic acid diet, ORAD.

Assessment of outcome measurements

Baseline and endpoint testing was conducted on the final two days of the run- in diet and the study diets. All study procedures were conducted at the Pennsylvania

State University CRC according to standardized procedures. Prior to testing, participants were instructed to fast for 12 hours, only take pre-approved medications, avoid strenuous exercise, and to refrain from consuming caffeine-containing products and alcohol for 48 hours. At each visit, vascular function testing was performed prior to blood sampling. Fasting blood samples were collected by trained nurses for analysis of lipids, lipoproteins, glucose, insulin, and markers of inflammation.

Blood Pressure & Vascular Measures

The SphygmoCor XCEL system (AtCor Medical, West Ryde, Australia) was used to assess cSBP, cDBP, cMAP, bSBP, bDBP, bMAP, cPP, bPP, AP, AIx

(adjusted to a heart rate of 75 beats per minute; bpm), HR, PTT, and PWV. Prior to measurements, participants were fitted with the correct size blood pressure cuff and these cuffs were consistently used for the remainder of the study. All measurements were performed by trained research personnel in a quiet, temperature-controlled room.

Pulse Wave Analysis (PWA)

The SphygmoCor XCEL system uses a brachial cuff-based method to estimate central BP, AP, and AIx via a generalized transfer function that has been validated against the indwelling catheter method77,120. Following a five minute seated rest, measurements were taken three times, following JNC 7 guidelines121, with a one- minute rest period between each measurement. The last two results were averaged and used for analysis. If the brachial BP measurements were inconsistent (i.e. >10 mmHg difference for systolic BP or >5 mmHg difference for diastolic BP) a fourth reading was taken. The AIx was standardized to a heart rate of 75 to correct for the independent inverse effect of heart rate augmentation of the pulse wave form and is reported as AIx.

Pulse Wave Velocity

Carotid to femoral PWV was assessed noninvasively using the SphygmoCor XCEL system, the gold standard for noninvasive measurement of arterial stiffness122.

Briefly, carotid and femoral arterial pressure waveforms were simultaneously captured via applanation tonometry of the carotid artery and a femoral BP cuff according to the manufacturer’s instructions. Three PWV measurements were obtained in the supine position, with the last two measures averaged for analysis.

Blood Sample Collection and Assay Methods

Blood was drawn into anticoagulated tubes containing lithium heparin or

EDTA and immediately centrifuged for 15 minutes. Blood drawn into serum separator tubes was allowed to clot for 30 minutes at room temperature and then centrifuged for 15 minutes. Serum TC, TG, and plasma glucose were directly measured using spectrophotometry at a certified, commercial laboratory (Quest

Diagnostics, Pittsburgh, PA). TC:HDL-C, and non-HDL-C. LDL-C was calculated using the Friedewald equation [LDL-C=TC-(HDL-C+TG/5)]; no participants had TG values above the limit for this equation123. Serum high-sensitivity CRP was measured using nephelometry and insulin (serum) was measured by immunoassay

(Quest Diagnostics, Pittsburgh, PA).

Statistics

All statistical analyses were performed with SAS (version 9.4; SAS Institute).

All randomized individuals were included in the data analyses for each outcome variable consistent with intent-to-treat (ITT) principles. Outcome variables were checked for normality (PROC UNIVARIATE) positively skewed variables (residuals skew >1; TG and hsCRP) were logarithmically transformed. Individuals with acutely elevated CRP (i.e. hsCRP ≥10 mg/L) values at the end of diet periods were excluded from all CRP analyses because this is indicative of an acute infection.

Differences between male and female participants at baseline were assessed by independent two-sample t tests (PROC TTEST). Change scores for end of study diet values were calculated by subtracting baseline measurements after the run-in diet from post-diet period values.

The mixed models procedure (PROC MIXED) was used to assess effects of the study diets on outcome measures at a pre-determined α value of 0.05. The mixed models procedure does not perform listwise deletion and preserves the degrees of freedom, thus allowing for inclusion of subjects with one or more missing data points in the analyses, permitting an ITT approach to be used to analyze data from subjects who withdrew prior to completing all endpoint testing. The primary analyses investigated if there was a difference in the change from baseline between the study diets for each outcome; within-diet change from baseline was also determined. Secondary analyses were conducted for between-diet mean values at the end of each diet period for all outcome variables. Selection of model covariance structures was based on optimizing fit statistics (evaluated as lowest Bayesian 51

Information Criterion). Baseline values were included as a covariate for assessing differences between the mean values of outcome variables. The effect of sex, randomization sequence, and their interaction with diet were added to the model to determine if carryover or sex effects were present. No carryover effects of the study diets were detected for any outcome. The only variable with a significant sex by diet interaction was PTT. Tukey-Kramer corrected P-values were used to correct for multiple comparisons.

Unblinded exploratory analyses were conducted to identify potential predictors (i.e. BMI, CRP) of participants’ responsiveness to the different study diets.

BMI, was categorized into three classifications: overweight (25.0 – 29.9 kg/m2), obese I (30.0-34.9 kg/m2), and obese II (35.0-39.9 kg/m2) and CRP into two: high-

CRP (≥1 mg/L) and low-CRP (<1 mg/L). The categorized variable was added to the mixed model as a main effect and the interaction with diet was also examined.

Power for this study was calculated based on the primary endpoint of cSBP.

A sample size of 32 people was required to provide 90% power to detect a 3 mm Hg

± 5 mmHg difference between the study diets with a two-sided α = 0.0584.

Results

Participants and Baseline Characteristics

A detailed diagram the flow of participants in the study is provided in Figure

3-2. Briefly, 99 participants were consented and screened for eligibility, 45 were

randomized, and 36 completed the trial. The overall dropout rate after randomization of eligible volunteers was 20%. Compliance to the dietary protocol was 85% among participants in the trial (Table 3-4) and was assessed daily. Noncompliance was considered to be any self-reported deviation from the controlled-feeding dietary protocol. No participants reported any challenges in consuming the amount of walnuts. Walnuts were provided as an evening snack, but participants had the option of consuming them throughout the day. Baseline characteristics for participants who completed the trial were not statistically different from participants who did not complete the trial (Table 3-5). A similar number of men and women were randomized to the study (Table 3-6). There were no differences in baseline BMI, TC,

LDL-C, non-HDL, glucose, insulin, brachial BP, or central BP between men and women. There were significant differences in age, HDL-C, TC:HDL-C, cPP, AP, AIx, and PTT between men and women at baseline. Participants had elevated BP (mean

± SD 121 ± 11.2 / 77 ± 7.8 mmHg) and LDL-C (119.2 ± 30.7 mg/dL).

Responded to advertisement n=568 Declined due to time commitment or food restrictions (n=398) Phone screen n=170

Ineligible (n=39) or Declined (n=32)

Clinic screen Ineligible (n=43) n=99 • LDL-C and/or BP too low (n=35) • BMI too low (n=2) • Triglycerides too high (n=1) • Irregular blood chemistry (n=5) Enrolled Eligible, but declined (n=10) n=46

Dropped out during run-in period (n=1)

Randomized n=45 Non-compliant (n=3) Personal reasons (n=2) Relocation (n=3) Gastrointestinal upset (n=1) Completed n=36

Figure 3-2. CONSORT diagram of participant flow through the study. Abbreviations: BMI, body mass index; BP, blood pressure; LDL-C, low-density lipoprotein- cholesterol.

Table 3-4. Average study compliance for each participant, including participants that did not complete the entire study*

Total compliance Total compliance ID # ID # (%) (%) 1 87.9 25 57.1 2 98.5 26 87.5 3 91.4 27 99.3 4 91.4 28 64.3 5 99.3 29 97.1 6 87.9 30 64.3 7 95.7 31 96.4 8 99.3 32 99.3 9 96.9 33 94.3 10 93.6 34 60.0 11 85.7 35 99.3 12 91.4 36 90.7 13 85.7 37 97.9 14 71.4 38 88.6 15 98.6 39 38.8 16 98.6 40 85.7 17 84.3 41 93.6 18 71.4 42 94.3 19 87.1 43 71.4 20 72.9 44 83.6 21 92.1 45 90.8 22 21.4 23 97.9 24 55.0

*Compliance was assessed through self-report. Any deviation from the dietary protocol was considered non-compliant. Each feeding day was evaluated as either compliant or non-compliant to calculate the overall compliance for each participant.

Table 3-5. Baseline characteristics of participants that did not complete the study*

Characteristic1 Total n 9 Age (y) 42±3 BMI (kg/m2) 31.5±1.3 TC (mg/dL) 184±12 HDL-C (mg/dL) 44±3 TC:HDL-C ratio 4.4±0.5 non-HDL-C (mg/dL) 140±14 TG (mg/dL) 109±22 Glucose (mg/dL) 91±3 Insulin (µIU/mL) 7±1 cSBP (mmHg) 113±4 cDBP (mmHg) 81±3 bSBP (mmHg) 123±5 bDBP (mmHg) 80±3 AIx (%) 27±6 PTT (m/s) 69±4 PWV (m/s) 7±0.4

*Data presented as mean±SEM. Baseline measurements were taken on the last 2 days of the run-in diet.

1Abbreviations: BMI, body mass index; TC, total cholesterol; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; TG, triglycerides (TG), bSBP, brachial systolic blood pressure; bDBP, brachial diastolic BP; cSBP, central SBP; cDBP, central DBP; AP, augmentation pressure; AIx, augmentation index; PTT, pulse transit time; PWV, pulse wave velocity.

Table 3-6. Baseline characteristics of participants randomized to study diets*

Characteristic1 Women Men Total P-value2 n 20 25 45 0.75 Age (y) 47±1.9 40±2.1 43±1.5 0.04 BMI (kg/m2) 30.8±1.2 29.9±0.9 30.3±0.7 0.52 TC (mg/dL) 187.5±6.5 190.9±7.3 189.4±4.9 0.74 HDL-C (mg/dL) 52.9±2.6 41.5±1.4 46.6±1.6 0.001 LDL-C (mg/dL) 114.8±6.0 122.6±6.7 119.1±4.6 0.40 TC:HDL-C ratio 3.7±0.2 4.7±0.2 4.2±0.2 0.004 non-HDL-C 134.7±6.8 149.3±7.0 142.8±5.0 0.15 (mg/dL) TG (mg/dL) 98.4±7.0 133.2±12.8 117.7±8.1 0.03 Glucose (mg/dL) 89.6±1.5 91.9±1.6 90.9±1.1 0.30 Insulin (mg/dL) 7.3±0.8 6.4±0.8 6.8±0.5 0.42 bSBP (mmHg) 121±2.3 121±2.4 121±1.7 0.92 bDBP (mmHg) 77±1.7 77±1.6 77±1.2 0.98 cSPB (mmHg) 113±2.0 110±2.0 111±1.5 0.33 cDBP (mmHg) 78±1.7 78±1.6 78±1.1 0.85 AP (mmHg) 11.7±1.1 5.5±0.8 8.2±0.8 <0.0001 AIx (%) 31.7±2.6 14.9±2.6 22.3±2.2 <0.0001 PTT (ms) 64.0±2.0 71.3±1.5 68.1±1.3 0.004 PWV (m/s) 7.1±0.2 6.8±0.2 6.9±0.1 0.38

*Data presented as mean ± SEM. Baseline measurements were taken on the last 2 days of run-in diet.

2Differences between men and women were computed using a student’s t-test (α<0.05).

Blood Pressure & Vascular Measurements

The WD, WFMD, and ORAD did not significantly change cSBP, the primary endpoint, from baseline and there were no significant differences between diets for the magnitudes of change (Figure 3-3) or mean values (Table 3-7). There was a main treatment effect for cDBP (P=0.04) and a significant reduction from baseline following the WD (-1.8 ± 1.0 mmHg, P=0.02). Post-hoc comparisons showed a greater change in cDBP following the WD compared to the ORAD (0.2 ± 0.7 mmHg,

P=0.04), but the changes were not different between the WD and WFMD (-0.2 ± 0.8 mmHg, P=0.2) or the WFMD and ORAD (P=0.7). Post-hoc testing showed mean cDBP was lower after the WD (76.1 ± 1.1 mmHg) compared to the ORAD (78.3 ±

1.1 mmHg, P=0.02), but there were no differences between the WD and WFMD or the WFMD and ORAD. The WD lowered bMAP (-1.44 ± 0.7 mmHg, P=0.04) and cMAP (-1.72 ± 0.8 mmHg, P=0.02) compared to baseline, but there were no differences between the diets. There were no statistically significant changes from baseline, differences in magnitudes of change, or diet effects for bSBP, bDBP, bPP, cPP, AP, AI, HR, PTT, or PWV (Table 3-7).

Figure 3-3. Mean changes from baseline for blood pressure and vascular measures following each of the three 6-week study diets (n=45). Data are presented as unadjusted means±SEM. Change scores were calculated by subtracting the values following the run-in diet from values following each study diet and were compared using the MIXED procedure (SAS version 9.4; SAS Institute Inc., Cary, NC). There was a significant reduction in bMAP, cDBP, and cMAP from baseline following the walnut diet (WD). Post-hoc comparisons showed a significant difference in cDBP between the WD compared to the oleic acid replaces α-linolenic acid (ALA) diet (ORAD) and no differences between the WD and the walnut fatty acid-matched diet (WFMD) or the WFMD and the ORAD. Post-hoc pairwise tests were adjusted for multiple comparisons using the Tukey-Kramer method; statistically significant between-diet effects are denoted with differing letters (P<0.05). Abbreviations: bSBP, brachial systolic blood pressure; bDBP, brachial diastolic BP; bMAP, brachial mean arterial pressure; cSBP, central systolic BP; cDBP, central DBP; cMAP, central MAP; AP, augmentation pressure; AIx, augmentation index; HR, heart rate; PTT, pulse transit time; pulse wave velocity, PWV. Statistically significant between- diet differences are denoted with differing letters (P<0.05). *P<0.05 for the within-diet change from baseline.

Table 3-7. Between-diet comparisons of vascular measures*

Outcome SWD WD WFMD ORAD Diet Variables1 (Run-in) P-value2 bSBP (mmHg) 121±1.7 119±1.6 120±1.4 120±1.5 0.73 bDBP (mmHg) 77±1.2 76±1.1 77±1.0 78±1.1 0.051 bPP (mmHg) 44±1.0 44±1.0 42±1.1 43±0.9 0.24 bMAP (mmHg) 92±1.3 90±1.2 92±1.0 92±1.2 0.15 cSBP (mmHg) 111±1.5 110±1.4 111±1.3 110±1.3 0.71 cDBP (mmHg) 78±1.1 76±1.1a 78±0.9ab 78±1.1b 0.04 cPP (mmHg) 33±0.7 33±0.9 33±0.8 32±0.8 0.24 cMAP (mmHg) 89±1.2 87±1.1 89±1.0 89±1.1 0.12 AP (mmHg) 8.3±0.8 8.5±0.8 9.0±0.9 8.2±0.7 0.80 AI (%) 22.4±2.2 21.9±2.2 23.3±2.3 21.5±2.1 0.59 HR (bpm) 66±1.5 64±1.7 64±1.4 64±1.6 0.76 PTT (ms) 68.1±1.3 68.2±67.5 68.5±1.3 67.4±1.2 0.52 PWV (m/s) 6.98±0.1 7.02±0.1 6.86±0.1 6.96±0.1 0.43

*Data are presented as means±SEM.

1Abbreviations: bDBP, brachial systolic blood pressure; bDBP, brachial diastolic blood pressure; bPP, brachial pulse pressure; cMAP, central mean arterial pressure; cSBP, central systolic blood pressure; cDBP, central diastolic blood pressure; cPP, central pulse pressure; cMAP, central mean arterial pressure; AP, augmentation pressure; AI, augmentation index; HR, heart rate; PTT, pulse transit time; PWV, pulse wave velocity.

2The MIXED procedure was used to determine the effect of diet on each outcome measure adjusted for the baseline value (SAS version 9.4; SAS Institute Inc., Cary, NC) at α<0.05. Post-hoc tests were adjusted for multiple comparisons using the Tukey-Kramer method. Statistically significant between-diet differences are denoted with differing letters (P<0.05).

Lipids & Lipoproteins

The WD (-15.8 ± 3.2 mg/dL; P<0.0001), WFMD (-14.6 ± 2.9 mg/dL; P<0.0001), and

ORAD (-11.2 ± 2.7 mg/dL; P<0.0001) all significantly lowered TC from baseline, but the magnitudes of change from baseline were not significantly different between groups (P=0.3, Figure 3-4). Similarly, the WD (-13.4 ± 2.9 mg/dL; P<0.0001),

WFMD (-11.1 ± 2.3 mg/dL; P<0.0001), and ORAD (-9.2 ± 2.3 mg/dL, P<0.0001) all significantly improved LDL-C from baseline, with no differences between study diets

(P=0.5). The WD, WFMD, and ORAD also lowered HDL-C from baseline (-1.8 ± 0.8 mg/dL, P<0.0001; -2.1 ± 0.9 mg/dL, P<0.0001; -1.1 ± 0.7 mg/dL, P<0.0001, respectively), with no differences between study diets (P=0.4). Non-HDL-C was reduced following the WD, WFMD, and ORAD (-14.0 ± 2.9 mg/dL, P<0.006; -12.5 ±

2.5 mg/dL, P<0.0009; -11.0 ± 2.2, P<0.03, respectively; P=0.5 for differences between diets). The TC:HDL-C was lowered after the WD compared to baseline

(P=0.03), but there was no main effect of diet (P=0.5). TG did not significantly change from baseline after any of the study diets. There were no significant differences between group means for the three study diets for TC, LDL-C, HDL-C, non-HDL-C, TC:HDL-C, or TG (Table 3-8).

Figure 3-4. Mean changes from baseline for lipids and lipoproteins following each of the three 6-week study diets (n=45). Change scores were calculated by subtracting the values following the run-in diet from values following each study diet and were compared using the MIXED procedure (SAS version 9.4; SAS Institute Inc., Cary, NC) at α<0.05. Post-hoc tests were adjusted for multiple comparisons using the Tukey-Kramer method. Statistically significant between-diet differences are denoted with differing letters (P<0.05).

Abbreviations: TC, total cholesterol; LDL-C, low-density lipoprotein-cholesterol; non- HDL-C, non-high-density lipoprotein-cholesterol; TG, triglycerides. There was a significant reduction in TC, LDL-C, non-HDL-C, and HDL-C after all diets compared to baseline. TC:HDL-C was significantly lower after the walnut diet compared to baseline. *P<0.05 for the within-diet change from baseline.

Table 3-8. Between-diet differences in blood measures and weight*

Outcome SWD Diet P- WD WFMD ORAD Variables1 (Run-in) value2 TC (mg/dL) 189.4±4.9 176.1±5.0 176.6±4.7 181.1±4.2 0.11 LDL-C (mg/dL) 119.2±4.6 107.3±4.3 108.6±4.3 112.3±3.8 0.08 HDL-C (mg/dL) 46.6±1.6 45.4±1.8 44.8±1.8 45.3±1.7 0.41 Non-HDL-C 142.8±4.9 130.7±4.9 131.9±4.8 135.0±4.21 0.41 (mg/dL) TC:HDL-C ratio 4.3±0.2 4.1±0.2 4.2±0.2 4.2±0.2 0.45 TG (mg/dL) 117.7±8.1 116.5±8.5 117.4±8.1 118.1±8.3 0.70 Glucose (mg/dL) 90.9±1.1 93.16±1.0 92.57±1.2 91.68±1.1 0.33 Insulin (µIU/mL) 6.8±0.6 6.02±0.5 6.56±0.6 6.72±0.7 0.85 CRP (mg/L) 2.1±0.3 2.4±0.4 2.40±0.5 2.22±0.4 0.18 Weight (kg) 92.5 ± 2.8 92.0 ± 3.1 91.5 ± 3.0 91.2 ± 3.2 0.34

*Data are presented as means±SEM.

1Total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), non-HDL-C TC:HDL-C, triglycerides (TG), glucose, insulin, and C-reactive protein (CRP).

2The MIXED procedure was used to determine the effect of diet on each outcome measure adjusted for the baseline value (SAS version 9.4; SAS Institute Inc., Cary, NC). Post-hoc tests were adjusted for multiple comparisons using the Tukey-Kramer method; statistically significant between-diet differences are denoted with differing letters (P<0.05). TG and CRP were log transformed due to right skew.

Glucose, Insulin, CRP, and Weight

There were no significant changes in glucose or insulin from baseline or between diets for any of the three study diets. CRP was decreased from baseline (-

0.95 ± 0.5 mg/L, P=0.049) following the WFMD, but there were no significant differences between study diets as end of treatment means or magnitudes of change

(Table 3-8). Weight was decreased from baseline following the WD (-1.2 ± 0.4 kg,

P=0.003), WFMD (-0.8 ± 0.4 kg, P=0.04), and ORAD (-1.1 ± 0.4 kg, P=0.005) but there were no significant differences between study diets as end of treatment means

(P=0.37) or magnitudes of change (P=0.33).

Exploratory BMI Subgroup Analysis of Response to Study Diets

There was an interaction between BMI category and cSBP (P=0.04), but no overall diet effect. Individuals with obesity (BMI 25-29.9 kg/m2; n=11) had a significant reduction in cSBP following the WD (-6 ± 2 mmHg; P=0.001) and WFMD

(-5 ± 2 mmHg; P=0.01) compared to baseline (Table 3-9). Further, individuals with obesity had a greater response to the WD compared to individuals with morbid obesity (-6 ± 2 vs. 3 ± 2; P=0.04; n=10). There were no differences in the effect on cSBP between individuals in the overweight (n=24) or morbidly obese categories.

Similarly, there was an interaction between BMI category and cMAP (P=0.01), but no overall diet effect. Individuals with obesity had a reduction in cMAP following the

WD (-6±1 mmHg; P<0.0001) and the WFMD (-3±1 mmHg; P=0.04) compared to

baseline, and individuals with obesity had a greater reduction in cMAP following the

WD compared to individuals with morbid obesity (3±2; P=0.002). There were no differences in the effect on cMAP between the overweight or morbidly obese groups.

No significant BMI subgroup differences were observed for any other outcome measures.

Table 3-9. Comparisons between diets by BMI classifications for cSBP and cMAP*

P value diet P value diet x cSBP x BMI cMAP BMI classification classification1 Overweight Obese M Obese Overweight Obese M Obese

WD -0.1±1ab -6±2*a 3±2b -0.8±1.0ab -6±2*a 3±2b 0.04 0.01 WFMD -0.01±1 -5±2* 3±2 -0.3±1.0 -4±2* 3±2

ORAD -0.5±1 -3±2 1±2 -0.4±1.0 -2±2 1±2

*Data presented as least squared means ± SEM.

1The MIXED procedure was used to determine the effect of the interaction between diet and BMI classification on the change from baseline in central systolic blood pressure (cSBP) and central mean arterial pressure (cMAP; SAS version 9.4; SAS Institute Inc., Cary, NC). Post-hoc tests were adjusted for multiple comparisons using the Tukey-Kramer method; statistically significant between-BMI class differences are denoted with differing letters and *indicates significant from baseline (P<0.05).

Exploratory CRP, BMI, and Lipid Interaction Analysis

CRP and BMI were positively correlated at baseline (R=0.61, P<0.0001) and in CRP was significantly different between BMI classifications (described in the previous section, P<0.0001). Individuals in the overweight group had significantly lower CRP (1.0±0.4 mg/L) compared to the obese group (3.2±0.6 mg/L, P=0.01) and the morbidly obese group (5.5±0.7 mg/L, P<0.0001). There were also significant differences between the obese and morbidly obese group (P=0.04).

Participants with lower-CRP (<1 mg/L; n=17) had significantly greater reduction in TC (-19.3 mg/L), LDL-C (-15.9 mg/L), and non-HDL-C (-17.2 mg/L) compared to baseline (P<0.05) and individuals with higher-CRP (≥1 mg/L; n=28; -

10.5 mg/L, -8.3 mg/L, and -9.6 mg/L, respectively; P=0.01 for all; Figure 3-5). There were no differences between the higher- and lower-CRP groups for HDL-C,

TC:HDL-C, or TG. There was no interaction between CRP and study diet (Diet*CRP) detected for any outcome variables.

-5 * -10 * High CRP -15 a a* *a Low CRP -20 * b * * b

-25 b Change frombaseline (mg/dL)

Figure 3-5. Comparisons between high and low CRP groups for lipids and lipoproteins. Diets were pooled and participants were dichotomized to high (≥1 mg/L) and low (<1 mg/L) CRP groups. Groups were compared using the MIXED procedure (SAS version 9.4; SAS Institute Inc., Cary, NC) at α<0.05. Post-hoc tests were adjusted for multiple comparisons using the Tukey-Kramer method. Statistically significant between-diet differences are denoted with differing letters (P<0.05).

Abbreviations: TC, total cholesterol; LDL-C, low-density lipoprotein-cholesterol; non- HDL-C, non-high-density lipoprotein-cholesterol; TG, triglycerides

Discussion

This randomized, crossover, controlled-feeding study is the first, to our knowledge, to evaluate the effects of walnut consumption on central BP. The diet containing walnuts as a replacement for SFA significantly lowered cDBP from baseline, and to a significantly greater extent than the ORAD. Our findings corroborate previous research that reported a reduction in brachial BP with walnut consumption81 and may have implications for dietary recommendations for individuals with elevated BP. The WD in our study incorporated approximately 57-99 g/d (2-3.5 oz/d) of walnuts as a snack into a healthy dietary pattern low in SFA. The favorable effects we observed for all three study diets supports the cardiovascular benefits of replacing SFA with unsaturated fats in the diet. Moreover, the greater improvement in cDBP after the WD compared to the ORAD provides further support for the unique benefits of walnuts as part of a low-SFA diet.

Previous studies have reported inconsistent effects of walnut consumption on brachial BP, but only one other trial evaluated effects on central BP. However, in that study74, there was no significant change in brachial or central BP in healthy male

participants after four weeks of consuming 15 g walnuts/d under free-living conditions74. This amount of walnuts is substantially less than the 57-99 g/d we provided in our six week controlled feeding study. Thus, the lack of significant change in these outcomes may be due to inadequate dosing, duration of walnut supplementation, and/or the healthy participants studied. A recent meta-analysis on walnuts and CVD risk factors also did not report a significant effect of walnut consumption on BP; however, fewer than one third of the studies analyzed reported results for brachial BP and only the study discussed above reported central BP results69,74. Notably, two clinical trials reported reductions in brachial BP following consumption of walnuts81,124. Wu et al.124 reported a reduction in brachial systolic (-

8.2 mmHg; 95% CI: -10.to -5.8; P<0.05) and diastolic (-4.2 mmHg; 95% CI: -5.7 to -

2.7; P<0.05) BP following supplementation with 30 g walnuts/d for 12 weeks, and

West et al.81 reported a reduction in brachial diastolic BP (-3±5 mmHg; P=0.0002) after six weeks on a diet containing 37 g walnuts/d as part of a controlled feeding- study. Collectively, these findings support the need for controlled feeding studies testing higher amounts of walnut consumption in adults at risk for vascular disease due to elevated BP and LDL-C.

To our knowledge, this is the first study to report that walnut intake reduced cDBP. Although brachial DBP, a target of the current blood pressure guidelines108, was not significantly reduced, it approached significance (P=0.051) and mirrored the trends in cDBP. The statistically significant reductions observed in diastolic, but not systolic central pressures following the WD could be related to the metabolic health

(e.g., LDL-C, BP, BMI, etc.) status of our participants. Our results suggest replacing

SFA with walnuts in the form of a snack improves central BP; however, further research is needed to confirm these findings and gauge their clinical significance.

The clinical implications for reducing cDBP are not completely understood since there is not a large evidence base on the cardiovascular effects of this blood pressure measurement.

The significantly greater reduction in cDBP following the WD compared to the

ORAD diet may have been due to the lower PUFA and higher MUFA content of the

ORAD compared to the WD. The ORAD was also devoid of walnut bioactives.

Clinical trials have reported benefits with oleic acid consumption as a replacement for SFA, but PUFA are still considered more effective than MUFA for lowering CVD risk108,115. High-oleic oils are being introduced into the food supply as a replacement for PUFA-rich oils and concerns have recently been raised that an adequate intake of linoleic acid and ALA will be difficult to obtain125. Our findings provide further support for the beneficial effects of ALA and walnut bioactives in the diet. Based on our results, providing walnuts is a preferred strategy for not only providing essential fatty acids in the diet but also walnut bioactives that contribute to BP benefits.

The effect of the WD on cDBP appears to be due to both the unique fatty acid profile, as well as their bioactive profile (e.g., phenolic compounds) of walnuts.

Walnuts are one of the richest plant sources of ALA126 and also contain phenolic acids, stillbenes, tocopherols, flavonoids, and melatonin15,38,127. These bioactives in walnuts have in-vitro antioxidant activity and have also been reported to reduce

oxidative damage and endothelial dysfunction84,128. Although we did not see an effect on arterial stiffness, we did observe reductions in cDBP and central and brachial MAP following the WD that may be due to different bioactive mechanisms including angiotensin-converting enzyme (ACE) inhibition, antioxidant, or other vasodilatory effects129,130.

Recent evidence has demonstrated that the potential beneficial effects of walnut bioactives on BP may be related to changes in the gut microbiome11,102. The composition and diversity of the gut microbiome has been associated with various chronic diseases, including CVD131,132. Walnuts are not completely digested in the upper gastrointestinal tract, which provides substrate to the gut microbiota133,134 and may promote the production of short chain fatty acids (SCFA)11. SCFA such as butyrate have been associated with maintenance of normal BP135. Specific types of bacteria, such as Akkermansia munciniphila, have been reported to increase with butyrate supplementation136 and are also associated with decreases in TNFα mRNA, systemic inflammation, and diastolic BP. Microbial production of butyrate through fermentation in the gut may also elicit the beneficial effects observed with butyrate supplementation, and consequently may have a BP-lowering effect. However, further research is needed to investigate the interaction between walnut consumption, the gut microbiome, and BP.

There is extensive research on the effects of walnut consumption on lipids and lipoproteins. A recent meta-analysis of 26 studies investigating the effects of walnut consumption on cardiovascular risk found a significant reduction in TC

(3.25% greater reduction compared to controls), LDL-C (3.73%), TG (5.25%), and apolipoprotein B (-3.74 mg/dL compared to control Western diets) following walnut consumption (15 to 108 g/day for 4 weeks to 1 year)69. This is consistent with our finding of significant reductions in TC, LDL-C, non-HDL-C, HDL-C and TC:HDL-C following the WD versus the standard Western run-in diet. The similarity in PUFA content of the three diets (WD: 16% PUFA; WFMD: 16%; ORAD: 14%) may explain why we did not observe a significantly greater effect of the WD compared to the other two study diets, which also produced significant beneficial effects on TC, LDL-

C, non-HDL-C, HDL-C and TC:HDL-C compared to baseline. However, when we compared the observed lipid changes in the current study with the predicted changes calculated using the Katan equation137, we observed greater reductions in

TC (observed: -15.8 versus expected: -13.5) and LDL-C (-13.4 versus -11.6 mg/dL) following the WD, whereas the observed reductions following the WFMD were closer to the predicted changes for TC (-14.6 versus -13.5 mg/dL) and LDL-C (-11.1 versus

-11.6 mg/dL) as shown in Figure 3-6. Differences in predicted versus observed LDL-

C lowering effects are similar to other study findings reported for tree nuts138 and specifically walnuts67,139. This suggests there may be an additive effect of consuming whole walnuts as a SFA replacement versus vegetable oils that provide the same fatty acid profile.

Obesity and hypertension are interrelated and can influence the response to both diet and drug treatments140–142. Adipocytes from visceral fat secrete inflammatory markers and angiotensinogen, an activator of the renin–angiotensin–

aldosterone system, which may result in hypertension143. Metabolic changes that occur due to the expansion of adipose tissue may affect the potency of dietary interventions; thus, the wide range in baseline BMI in the present study (25-38 kg/m2) may explain the lack of effect on cSBP and cMAP following the study diets.

Although there was not a diet effect for our primary endpoint, cSBP, secondary analyses indicated that BMI played a role in participants’ systolic BP response to the study diets. We expected to see the greatest improvement in central BP in either the overweight group (n=24), where obesity would not blunt the effects of the intervention, or in the obese (n=11) and morbidly obese (n=10) categories, with the highest baseline cSBP (overweight: 108.9±2.2 mmHg; obesity: 113.8±2.2 mmHg; morbid obesity: 113.8±2.5 mmHg). However, we found that only individuals with obesity (30.0-34.9 kg/m2) had a significant reduction in cSBP and cMAP from baseline following the WD and WFMD, with no significant change from baseline in the overweight or morbidly obese group. Importantly, there were no significant differences in weight change between groups (P=0.20) that could account for this.

Further, we observed a greater reduction in cSBP and cMAP in individuals with obesity compared to morbid obesity following the walnut diet. This suggests that changes in central BP following the substitution of SFA with PUFA may be related to the degree of metabolic abnormalities associated with overweight and obesity.

Relative to the obese group, individuals in the overweight group may have fewer metabolic disturbances and individuals in the morbidly obese group may have greater metabolic disturbances, including inflammatory status, which possibly blunted the response to dietary treatment. There was a significant positive 74

correlation between BMI and CRP; therefore, we are unable to understand the individual effects of BMI compared to CRP. Given the rising rates of obesity in the

US and worldwide, further research investigating the role of BMI in the response to dietary changes is warranted.

A strength of this study was the use of a crossover design that allowed participants to act as their own controls and minimized the influence of between- subject variability when analyzing treatment effects. This study represents a realistic food-based approach to replacing SFA and demonstrates that relatively small dietary changes can reduce cardiovascular risk. In addition, the controlled-feeding study design with high rates of compliance, as assessed via daily questionnaire and weight status, were additional strengths. The six week feeding periods also provided a better understanding of the longer-term effects of different treatment diets compared to a SWD. However, the effect size was over-estimated (0.6) for the primary endpoint, and thus the study was underpowered to detect between- treatment effects. To detect the observed effect size (0.14), approximately 516 participants would have been required. Although the run-in diet was designed as a

Western-style diet with a SFA content representing average American consumption, this diet likely differed from participants’ habitual diet, particularly in SFA. This explains the improvements we observed in LDL-C (-6.1±2.2 mg/dL; P=0.009) from the time participants were screened to when they were randomized (baseline). It is possible the LDL-C lowering from the run-in diets may have blunted the effects of the study diets.

In summary, our findings demonstrate that the replacement of SFA with

PUFA from walnuts and vegetable oils and MUFA from vegetable oils significantly reduces cardiovascular risk factors, including: TC, LDL-C, and non-HDL-C.

Incorporation of walnuts into a healthy dietary pattern as a SFA-replacement for six weeks did not significantly alter brachial or central systolic BP, but did result in a greater lowering of central diastolic BP compared to a diet with a similar fatty acid profile from vegetable oils but lower in ALA and higher in oleic acid. The greater reduction in central diastolic BP is attributable to both the unique combination of fatty acids, as well as the bioactive compounds provided by whole walnuts. All three diets replaced SFA with PUFA and reduced cardiovascular risk, with the greatest reduction in central diastolic BP following the WD. These findings provide new evidence that replacing SFA with PUFA from whole walnuts versus vegetable oils confers additional health benefits.

a) b)

c) d)

Figure 3-6. Observed changes in total cholesterol (TC) and low-density lipoprotein- cholesterol (LDL-C) compared to predicted changes in TC and LDL-C using the Katan equation (represented with the black horizontal line). Panels a) and c) represent TC and LDL-C changes from baseline following the walnut diet (WD) andwalnut fatty acid matched diet (WFMD) compared to the predicted changes (black horizontal lines). Panels b) and d) show TC and LDL-C changes from baseline following the oleic acid replaces ALA diet (ORAD) compared to the predicted changes (black horizontal lines). 77

Chapter 4: Walnuts and vegetable oils differentially affect the gut microbiome and associations with cardiovascular risk factors

Background: Walnut consumption favorably affects the gut microbiome. It is unclear whether the effects are attributable to the fatty acids, including α-linolenic acid (ALA), and/or the bioactives in walnuts. Further, it is not well understood how gut bacteria are related to cardiovascular risk factors.

Methods: 42 adults at cardiovascular risk were included in this secondary analysis of a randomized, crossover, controlled-feeding study, which reported beneficial reductions in CVD risk factors. Only significantly changed risk factors were correlated with significantly enriched gut bacteria. Sample collections occurred following a 2-week run-in diet and three 6-week diet periods. Linear discriminant analysis Effect Size plots were used to determine enriched taxa.

Results: Following the WD, there was the greatest abundance of Roseburia

(Relative abundance=4.2%, LDA=4, P=0.0008), Eubacterium eligensgroup (1.4%, 4,

0.05), Lachnospiraceae UCG001 (1.2%, 3.2, 0.03), Lachnospiraceae UCG004

(1.0%, 3, 0.04) and Leuconostocaceae (0.03%, 2.8, 0.05) relative to taxa in the

SWD. The WD was also enriched in Gordonibacter relative to the WFMD. The

WFMD had the greatest abundance of Roseburia (3.6%, LDA=4, P=0.02) and

Eubacterium eligensgroup (1.5%, 3.4, 0.02) and following the ORAD

Clostridialesvadin BB60 group (0.3%, LDA=2, P=0.04) and gutmetagenome (0.2%,

2, 0.005) were most abundant relative to the SWD. There were significant associations between the enriched bacteria following the WD with CVD risk factors.

Eubacterium eligens was associated with brachial mean arterial pressure (MAP, R=- 78

0.5, P=0.0009), central diastolic BP (-0.5, 0.0006), and central MAP (-0.5, 0.002).

Lachnospiraceae was associated with brachial (R=-0.4, P=0.02) and central MAP (-

0.4, 0.02), central diastolic BP (-0.3, 0.04), TC (-0.4, 0.03), and non-HDL-C (-0.4,

0.02). Leuconostocaceae was associated with brachial (R=0.3, P=0.03) and central

MAP (0.3, 0.03).

Conclusions: Similarities between enrichment of favorable bacteria following the

WD and WFMD suggest ALA and linoleic acid affect the microbiome. The unique enrichment of Gordonibacter, a bacterium important for ellagitannin metabolism, following the WD suggests the gut microbiome is involved in walnut bioactive metabolism. Further, Lachnospiraceae and associations with improved CVD risk factors suggests the microbiome contributes to the beneficial health effects of walnut consumption.

Clinical Trial Registration: URL: https://clinicaltrials.gov. Unique Identifier:

NCT02210767

Key Words: Walnuts, microbiome, butyrate, bioactives, PUFA, cardiovascular disease

Introduction

Cardiovascular disease (CVD) is the leading cause of morbidity and mortality globally and poor diet is the primary etiological risk factor underlying the development of CVD3,144. Recent evidence suggests the gut microbiome also plays an important role in cardiovascular risk103,145–148. The gut microbiome is a complex ecosystem of organisms that are critical for human health. The billions of microbes in the gastrointestinal tract protect against pathogens, metabolize undigested dietary components, and synthesize vitamins, among other functions. The microbes present in the lower gastrointestinal tract are involved in the extraction and metabolism of nutrients not fully digested in the small intestine. Diet largely affects the composition and functionality of bacteria present in the large intestine100.

There is evidence that the gut microbiome may be a mediator of the CVD benefits with walnut consumption. It has been demonstrated that the metabolizable energy from walnuts is overestimated by 21% by the predicted Atwater factors, suggesting that walnut-derived nutrients are accessible to the gut microbiome133.

Walnuts are uniquely rich in PUFA, including alpha-linolenic acid (ALA), an omega-3 fatty acid that has cardiometabolic benefits81,149. However, walnuts also contain bioactives, such as hydrolysable tannins, that can be metabolized by gut bacteria and may confer additional benefits beyond the fatty acid profile of walnuts150.

Collectively, these walnut components may contribute to the cardiometabolic benefits of walnuts that have been reported through different mechanisms of action.

The objective of the present secondary analysis of a randomized, crossover, controlled-feeding trial was to examine the differences in gut microbial composition in individuals at risk for CVD following a diet containing walnuts as a SFA replacement compared to two diets that replace SFA with a walnut fatty-acid matched profile (without walnuts) and a lower-ALA diet that replaces oleic acid for

ALA (without walnuts). Exploratory analyses were also conducted to examine the relationship between CVD risk factors and changes in bacterial abundance following each experimental diet. We hypothesized that following a diet containing walnuts, individuals would have a significantly different gut microbiome due to the bioactive content and whole food complex compared to the two other experimental diets and the baseline run-in diet.

Methods and Statistics

Study Design

Details of the study design and results of the CVD-related endpoints, including blood pressure, vascular health, and lipids and lipoproteins, are reported elsewhere151. Data from secondary analyses, including analysis of fecal bacteria diversity, enrichment, predictive functional analysis, and taxonomic co-occurrence, are reported here. Briefly, a randomized, crossover, controlled-feeding study with a two-week run-in diet on a standard Western diet (SWD) followed by three six-week diet periods on a walnut diet (WD), walnut fatty acid-matched diet (WFMD), and oleic acid replaces alpha-linolenic acid diet (ORAD) was conducted at The Pennsylvania 81

State University. The fully controlled, weight-maintenance dietary feeding intervention provided similar macronutrient profiles (Table 4-1) in all three study diets (48% carbohydrate, 17% protein, 35% fat, 7% SFA) and the run-in diet (50%,

17%, 33%, 12%). However, the presence or absence of walnuts and the fatty acid profile of each of the diets differed: WD (contained walnuts; 2.7% ALA), WFMD (did not contain walnuts; 2.6% ALA); ORAD (did not contain walnuts; 0.4% ALA).

Participants’ compliance was examined through daily questionnaires for diet, medications, exercise, and general wellness. Participants were asked to refrain from medication use, including antibiotics, 48 hours prior to sample collections. This trial is registered at clinicaltrials.gov as NCT02210767.

Table 4-1. Nutrient Profiles of the run-in diet and study diets Nutrient SWD (run-in) WD WFMD ORAD Total fat* 34 35 35 35 SFA* 12 7 7 7 MUFA* 12 9 9 12 PUFA* 7 16 (2.7*) 16 (2.6*) 14 (0.4*) Carbohydrate* 50 48 48 48 Protein* 16 17 17 17

All diets used the same six-day cycle menu, developed using Food Processor SQL software, version 10.8 (ESHA Research, Salem, OR). Abbreviations: standard western diet, SWD; walnut diet, WD; walnut fatty acid-matched diet, WFMD; oleic acid replaces α-linolenic acid diet, ORAD.

*Represents % total calories based on a 2100 calorie diet.

Participants

Study participants were recruited from the State College, PA area. Interested volunteers who were likely eligible after a telephone screen were scheduled for a clinic screening to confirm eligibility based on inclusion criteria. Eligible participants were men and women 30-65 years with overweight or obesity (BMI: 25.0 – 39.9 kg/m2) who had elevated blood pressure (120-159/80-99 mmHg) and/or increased low-density lipoprotein cholesterol (LDL-C; 128-177 mg/dL for men and 121-172 for women). Written informed consent was obtained from all participants prior to enrollment in the study. All study samples were collected and procedures were conducted at The Pennsylvania State University Clinical Research Center (CRC).

CVD Risk Factors

Blood pressure and vascular measures were assessed using the

SphygmoCor XCEL system (AtCor Medical, West Ryde, Australia) at the end of each diet period and following the run-in diet. Fasting serum lipids and lipoproteins were measured by a commercial laboratory (Quest Diagnostics, Pittsburgh, PA) on two consecutive days. Additional details are published elsewhere (Tindall et al.151).

Fecal Sample Collection

Participants provided a ~30 g fecal sample from a single defecation at the end of the run-in diet and each of the three diet periods using a collection kit [containing a stool collection hat, long-handled spoon, medical gloves, and Para-Pak clean vial

(Meridian Bioscience, Inc., Cincinnati, OH)] provided by investigators. Samples were stored in a freezer (-18°C) until the following day at which time they were given to research staff who stored the samples in a -80°C at the CRC until analysis.

DNA Extraction and Quantification

Nucleic acid extractions were performed on each sample using a Qiagen

Powersoil DNA Isolation kit following the manufacturer's instructions (Qiagen

Germantown, MD) at Juniata College (Huntingdon, PA). The vortex/lysing step was performed using the Disruptor Genie cell disruptor (Scientific Industries, Bohemia,

NY) for 5 minutes. The resulting genomic DNA (gDNA) was eluted in 50 μl of 10 mM

Tris. The Qubit 4.0 Fluorometer (Invitrogen, Waltham, MA) was utilized to quantify gDNA according to the dsDNA High Sensitivity assay.

PCR Amplification

Illumina iTag Polymerase Chain Reactions (PCR) were performed at Juniata

College (Huntingdon, PA) at a total volume of 25 μL for each sample and contained final concentrations of 1X Ex Taq PCR buffer, 0.8 mM Ex Taq dNTP's, 0.625 U Ex

Taq Polymerase (Takara Bio, Mountain View, CA, 0.2 μM 515F forward barcoded primer, 0.2 μM Illumina 806R primer152 and ~10 ng of template DNA per reaction.

PCR was carried out on a T100 thermocycler (Bio-Rad, Hercules, CA) using the following cycling conditions: 98 °C for 3 min; 35 cycles of 98 °C for 1 min, 55 °C for

40 s, and 72 °C for 1 min; 72 °C for 10 min; and kept at 4 °C. PCR products were

visualized on a 2% Agarose E-gel (ThermoFisher Scientific, Waltham, MA) stained with ethidium bromide153.

Library Purification, Verification and Sequencing

Equimolar concentrations of PCR amplicons were pooled together and were subsequently gel purified using the Qiagen Gel Purification Kit (Qiagen,

Germantown, MD). Clean PCR products were quantified using the Qubit 4.0

Fluorometer (Life Technologies, Carlsbad, CA), and libraries were combined in scaled equimolar amounts with other libraries to provide sufficient material for sequencing and balanced coverage. Prior to submission for sequencing, libraries were quality checked using the 2100 Bioanalyzer DNA 1000 chip (Agilent

Technologies, Santa Clara, CA). Pooled libraries were stored at -20 °C until they were shipped on dry ice to Laragen, Inc. (Culver City, CA) for sequencing.

Library pools were size verified using the Fragment Analyzer CE (Advanced

Analytical Technologies Inc., Ames IA) and quantified using the Qubit High

Sensitivity dsDNA kit (Life Technologies, Carlsbad, CA). After dilution to a final concentration of 1 nM and a 10% spike of PhiX V3 library (Illumina, San Diego

CA), pools were denatured for 5 minutes in an equal volume of 0.1 N NaOH then further diluted to 12 pM in Illumina’s HT1 buffer. The denatured and PhiX-spiked 12 pM pool was loaded on an Illumina MiSeq V2 500 cycle kit cassette with 16S rRNA library sequencing primers and set for 250 base, paired-end reads.

Quality Filtering and ASV Picking

Qiime2 version 2018.8.0 was utilized to implement dada2154 processing of sequence data. Initially, data were quality filtered to 240 base pair on each read and reads with expected errors >0.5% were discarded. Sequences were then dereplicated, checked for sequence variants, merged, and finally chimera checked by dada2. A total of 3,008,191 sequences were obtained after quality filtering and processing. Representative sequences were then assigned taxonomy using the

Silva 132 database. The resulting table and taxonomy artifacts were then exported as a .biom table and text file, respectively, for subsequent analyses following addition of taxa data to the biom-formatted ASV table.

Alpha and Beta Diversity Analysis

Alpha diversity plots were generated using a rarefied biom formatted ASV table with all samples that had greater than 10,000 sequences. Rarefaction was conducted on sequences across all samples to a maximum depth of 10,000 sequences with step size 392 and 20 iterations. Alpha diversities were then collated and plotted using observed species richness metric (α=0.05).

Principal coordinates analyses (PCoA) plots and ANOSIM tests for significance were generated from a weighted UniFrac distance matrix that was produced from a cumulative sum scaling (CSS) normalized biom-formatted ASV table within QIIME 1.9.0 (Paulson, Stine, Bravo, & Pop, 2013). LEfSe tests for

significance were also used to identify significantly enriched taxa within a diet when compared with the other diets (α=0.05).

Taxonomic Comparisons

Taxonomy were assigned for ASVs against the Silva 132 database (as described above) and were added into the biom-formatted ASV table. Relative abundance (RA) facet grid bar plots were generated within Rstudio using the

Phyloseq package (McMurdie & Holmes, 2013) to visualize the abundance of taxa across all samples, grouped by individual and divided by diet.

The RA of the most prominent phyla (Bacteroidetes, Firmicutes, and

Proteobacteria) following each of the study diets (WD, WFMD, and ORAD) were compared. The RA of the Gordonibacter genus, a urolithin producer155, following each of the study diets was also tested. The RA of Bacteroidetes, Firmicutes, and

Proteobacteria by body mass index (BMI) status were evaluated between classifications (25.0-29.9 kg/m2: overweight, 30.0-34.9 kg/m2: obese, 35.0-39.9 kg/m2: morbidly obese). Differences were tested using the mixed models procedure

(PROC MIXED) in SAS (version 9.4; SAS Institute) at a pre-determined α value of

0.05. Tukey-Kramer corrected P-values were used to correct for multiple comparisons.

RA of taxa were subject to linear discriminant analysis effect size (LEfSe)156 analysis to identify enriched taxa during pairwise comparisons of the diets. For each

comparison, each participant’s data for a respective diet was included. LDA scores of 0 were used to identify taxa that were significantly enriched with respect to the other diet in the comparison.

Co-Occurrence Network (CoNet) Analysis

Biom formatted ASV tables were utilized to produce Co-occurrence networks of ASVs present across 50% samples within a given diet using the CoNet tool157.

Pearson and Spearman correlations were utilized to determine the relationships

(positive or negative) between the ASVs (α=0.05) denoted by green edges (positive correlations) or red edges (negative correlations). Nodes are colored by phylum and size is relative to the abundance of the taxa.

Predictive Functional Analysis using Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt)

Filtered sequences were used to predict the functional profile of the microbial communities within each sample via closed reference picking against Greengenes

13_5. Predicted KEGG Orthology (KO) genes were then grouped by functional category and results were then compared pairwise within LefSe (LDA = 0) to determine enriched predicted functional pathways. Additionally, individual KO terms were analyzed to determine individual genes that may be differentially present (LDA

= 0.5)

Correlations with Cardiovascular Risk Factors

Primary analyses from this study showed significant reductions in brachial and central mean arterial pressure (MAP), central diastolic blood pressure (BP), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), non-high-density lipoprotein cholesterol (non-HDL-C), and the ratio of TC to HDL-C following the WD compared to baseline. Additionally, TC, LDL-C, non-HDL-C, and HDL-C were significantly reduced following the WFMD and ORAD compared to baseline. The bacteria that were significantly enriched (LefSe plots) were correlated with the respective cardiovascular risk factors that were significantly changed following each of the study diets. Statistical analyses were performed with SAS (version 9.4; SAS

Institute; Cary, NC). The correlation procedure (PROC CORR) was used with

Spearman’s correlation coefficient to assess association between the cardiovascular risk factors that were significantly changed from baseline with bacteria that were significantly enriched at a pre-determined α value of 0.05.

Multivariate Association with Linear Models (MaAsLin)158 was also used to examine how the entire bacterial community, rather than only the significantly enriched bacteria, was associated with CVD risk factors. This exploratory analysis uses a novel multivariate algorithm. All of the bacteria present in the sample were mapped against the magnitudes of change for significantly changed CVD risk factors for each study diet and mean values for the same CVD risk factors were mapped against the run-in diet bacteria. A multivariate linear model associating metadata with each bacterium independently was boosted, and any metadata selected in at

least 1% of these iterations was finally tested for significance (α < 0.05) in a standard generalized linear model. Since this approach compared every bacteria present in the sample against each individual CVD risk factors, multiple comparisons were adjusted using a Bonferonni correction; multiple hypothesis tests over all bacteria and metadata were adjusted to produce a final Benjamini and Hochberg false discovery rate (Q-value). Statistical analyses were performed in R using the glm package.

Results

Participants

A detailed flow diagram of the phases of this trial is available from Tindall et al.151. Briefly, a total of 46 participants were enrolled in the study, 45 randomized, and 36 completed the study. Of the 45 participants who were randomized, fecal samples were available from 42 participants. There were no significant differences between participants that were included in the analysis compared to individuals not included (Table 1). Analysis was performed on all 42 participants, including participants who did not complete the trial but contributed partial data. Participants included in the gut microbiome analysis had an elevated BMI (30.4 ± 0.7 kg/m2), BP

(121 ± 1.7/77 ± 1.2 mmHg), and LDL-C (120.3 ± 4.8 mg/dL) as shown in Table 4-2.

Table 4-2. Baseline Characteristics Fecal Sample Fecal Sample Not Characteristic P-vaule Available Available n 42 3 - Males 23 2 0.76 Age (y) 43 ± 1.5 40 ± 7.4 0.72 BMI (kg/m2) 30.4 ± 0.7 30.1 ± 1.6 0.89 TG (mg/dL) 118.8 ± 8.6 103.7 ± 15.2 0.44 TC (mg/dL) 190.8 ± 5.1 170 ± 19.2 0.39 LDL-C (mg/dL) 120.3 ± 4.8 103.0 ± 13.6 0.33 HDL-C (mg/dL) 46.7 ± 1.7 46.2 ± 4.9 0.93 TC:HDL-C (mg/dL) 4.3 ± 0.2 3.7 ± 0.3 0.16 non-HDL-C (mg/dL) 144.2 ± 5.2 124.3 ± 16.1 0.34 Glucose (mg/dL) 91 ±1.1 89.3 ± 7.2 0.84 Insulin (µIU/mL) 6.8 ± 0.6 6.2 ± 0.5 0.40 bSBP (mmHg) 121 ± 1.7 130 ± 3.1 0.07 bDBP (mmHg) 77 ± 1.2 84 ± 5.0 0.57

Data presented as mean ± SEM. Baseline measurements were taken on the last 2 days of run-in diet. Differences between individuals included in the microbiome analysis and individuals not included in the analysis were computed in SAS (version 9.4; SAS Institute; Cary, NC) using a student’s t-test (α<0.05).

Abbreviations: BMI, body mass index; TC, total cholesterol; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; TG, triglycerides (TG), bSBP, brachial systolic blood pressure; bDBP, brachial diastolic BP.

DNA Extraction and Quantification, PCR Amplification, Library Purification, Verification and Sequencing, and Quality Filtering and ASV Picking

16S rRNA gene PCR amplification was successful for all collected samples.

High quality sequence data was obtained for 144 out of 149 fecal samples. Overall, sequencing depth of dada2 processed data ranged from 1,955 to 95,549 sequences per sample. A total of 3,008,191 sequences were obtained after quality filtering, merging, and chimera checking. 142 samples were able to be incorporated into community analyses and a CSS normalized ASV table, as sequencing depth exceeded 1000 sequences for each sample.

Alpha- and Beta-Diversity

Alpha diversity rarefaction curves and beta diversity analyses did not reveal distinct differences in microbial community species richness between samples or diets and each diet had similar average observed species richness, ranging from 99 to 120 ASVs (Figure 4-1 and Figure 4-2). A range of 18-246 total unique bacterial

ASVs were observed within each sample when evaluating raw DADA2 processed

ASVs.

The composition of the bacterial communities was similar across each diet.

There was no distinct shaping or clustering observed between the samples by diet

(ANOSIM P=0.86). Similar to the above analyses, no distinct trends were evident in the bacterial community structure of the gut communities with respect to diet and the composition of the communities. Across all of the study diets Bacteroidetes (average

RA=23%), Firmicutes (73%) and Proteobacteria (3%) were the most abundant

bacteria phyla (Figure 4-3) and similar abundances were observed in the run-in diet

(Bacteroidetes=24%, Firmicutes=71%, Proteobacteria=3%).

c. 93

Figure 4-1. Alpha diversity curves and boxplots were generated from a biom- formatted ASV table in QIIME 1.9.0 to compare richness between samples and diets. Samples with less than 10,000 DADA2 processed sequences were removed prior to rarefying the table. Individual samples (A) and diets (B) were then rarefied up to 10,000 sequences. As rarefactions approach 10,000 sequences, fewer observed species are identified and the curves become flat and smooth indicating sequencing depth sufficiently captured the taxa in this community. Boxplots (C) recapitulate the diet curves (B) by stratifying samples into respective diets to demonstrate the range of observed species in each diet and indicated similar alpha diversity trends between the diets

Figure 4-2. Beta diversity analysis of each sample with >1000 sequences was performed using a weighted unifrac distance matrix produced from a CSS normalization of a biom-formatted ASV table in QIIME 1.9.0. Three-dimensional principal-coordinate analysis plots were generated to visualize community structure with EMPeror with each sample colored by their respective diet. No distinct shaping or clustering was observed between the samples in the diets (ANOSIM P=0.86).

0% 0% 0% 0% 0% 0% 0% 3% 1% Bacteroidetes Firmicutes 23% Proteobacteria Actinobacteria Cyanobacteria Elusimicrobia Epsilonbacteraeota Fusobacteria Lentisphaerae 73% Tenericutes Verrucomicrobia

Figure 4-3. Average relative abundance (RA) of phyla in the three study diets.

Taxonomic Comparisons

LefSe plot (LDA 0) comparisons of diets to the SWD showed significantly enriched ASVs (Figure 4-4, Table 4-3). There were nine significantly enriched taxa following the WD relative to the SWD; the most enriched were Roseburia (RA=4.2%,

LDA=4.2, P=0.0008), Eubacterium eligensgroup (RA=1.4%, LDA=3.6, P=0.05),

Lachnospiraceae UCG001 (RA=1.2%, LDA=3.2, P=0.03), and Lachnospiraceae

UCG004 (RA=1.0%, LDA=3.1, P=0.03). There were four enriched taxa following the

WFMD; Roseburia (RA=3.6%, LDA=3.8, P=0.02) and Eubacterium eligensgroup

(RA=1.5%, LDA=3.4, P=0.02) showed the greatest enrichment. Three taxa were enriched following the ORAD; Clostridialesvadin BB60 group (RA=0.3%, LDA=2.5,

P=0.04) had the greatest relative abundance of the enriched ASVs. There were also

significant differences between study diets. There were three enriched taxa following the WD relative to eight enriched taxa following the ORAD; Roseburia (average

RA=4.2%, LDA=4.1, P=0.02) was most abundant and showed greatest enrichment following the WD. Ruminiclostridium (RA=3.4%, LDA=1.4, P=0.04) and

Clostridialesviadin BB60 group (RA=0.3% vs 0.1%, LDA=2.4, P=0.03) showed greatest enrichment following the ORAD. There was one enriched taxa following the

WD [Gordonibacter (RA=0.04%, LDA=3.2, P=0.03)] relative to two enriched taxa following the WFMD [Faecalibacterium (RA=4.4%, LDA=4.3, P=0.03) and

Angelakisella (RA=0.1%, LDA=2.8, P=0.04)]. There were no enriched taxa when comparing the WFMD to the ORAD diet.

There were no significant differences in the RA of Bacteroidetes (P=0.7),

Firmicutes (P=0.7), or Proteobacteria (P=0.5) among the WD, WFMD, or ORAD

(Figure 4-5). There were differences in RA of Gordonibacter (P=0.02) following the

WD (RA= 0.0004), WFMD (RA=0), and ORAD (RA= 0.00003). The RA

Gordonibacter was greater following the WD compared to the WFMD (P=0.04) and the ORAD (P=0.04). After pooling the diets and categorizing by BMI, there were not differences in the RA of Bacteroidetes (P=0.8), Firmicutes (P=0.9), or Proteobacteria

(P=0.5) among overweight (n=23), obesity (n=9), and morbid obesity (n=10) groups

(Figure 4-6).

Figure 4-4. Linear discriminant analysis effect size (LefSe) plots for between-diet and study diet-run-in diet comparisons of enriched taxa. Panel a. shows significant taxonomic features between oleic acid replaces ALA diet (ORAD) and the standard Western, run-in diet. Panel b. shows significant taxonomic features between the walnut diet (WD) and the run-in diet. Panel c. shows significant taxonomic features between the walnut fatty acid matched diet (WFMD) and the run-in diet. Panel d. represents significant taxonomic features between the ORAD and the WD and panel e. shows significant taxonomic features between the WD and the WFMD.

Table 4-3. Linear discriminant analysis effect size (LefSe) analysis between study diets and versus the run-in diet

Comparison Diet RA LDA Score P-value WD vs. Run-in D_5__Roseburia WD 0.042 4.17 0.0008 D_5__Eubacterium_eligensgroup WD 0.014 3.62 0.05 D_5__LachnospiraceaeUCG_001 WD 0.012 3.19 0.03 D_5__LachnospiraceaeUCG_004 WD 0.01 3.07 0.04 D_4__Leuconostocaceae WD 0.0003 2.8 0.05 D_5__RuminococcaceaeUCG_003 WD 0.005 2.75 0.03 D_5__uncultured WD 2.66 0.05 0.002 D_5__uncultured WD 2.48 0.05 D_5__Butyricicoccus WD 0.003 2.37 0.01 WFMD vs. Run-in D_5__Roseburia WFMD 0.036 3.77 0.02 D_5__Eubacterium_eligensgroup WFMD 0.015 3.44 0.02 D_5__gutmetagenome WFMD 0.003 3.12 0.01 D_5__Butyricicoccus WFMD 0.003 2.95 0.02 D_4__Streptococcaceae Run-in 0.006 3.33 0.05 D_5__Streptococcus Run-in 0.006 3.35 0.02 ORAD vs. Run-in D_5__Streptococcus Run-in 0.006 2.83 0.03 D_5__gutmetagenome ORAD 0.002 2.44 0.005 D_5__unculturedorganism ORAD 0.001 2.45 0.04 D_4__ClostridiaesvadinBB60group ORAD 0.003 2.48 0.04 WD vs. WFMD D_5__Faecalibacterium WFMD 0.044 4.27 0.03 D_5__Angelakisella WFMD 0.001 2.78 0.04 D_5__Gordonibacter WD 0.0004 3.22 0.03 WD vs. ORAD D_5__Roseburia WD 0.042 4.1 0.02 D_5__DefluviitaleaceaeUCG_011 WD 0.0007 1.63 0.01 D_4__Defluviitaleaceae WD 0.0007 1.62 0.01 D_5__Ruminiclostridium1 ORAD 0.034 1.43 0.04 D_0__Bacteria ORAD unknown 1.58 0.02 D_0__Bacteria ORAD unknown 1.61 0.02 D_0__Bacteria ORAD unknown 1.66 0.02 D_0__Bacteria ORAD unknown 1.69 0.02 99

D_0__Bacteria ORAD unknown 1.72 0.02 D_5__uncultured ORAD 0.003 1.79 0.05 D_4__ClostridialesvainBB60group ORAD 0.003 2.35 0.03 Linear discriminant analysis effect size (LefSe) plots for between-diet and study diet- run-in diet comparisons of enriched taxa. Abbreviations: relative abundance, RA; linear discriminant analysis, LDA; walnut diet, WD; walnut fatty acid matched diet, WFMD; oleic acid replaces alpha-linolenic acid diet, ORAD.

0.5

Bacteroidetes

Relative Relative Abundance 0 a. WFMD ORAD WD

1 Firmicutes

0.5

Relative Relative Abundance 0 WFMD ORAD WD b.

0.5 Proteobacteria

Relative Relative Abundance 0 WFMD ORAD WD c.

Figure 4-5. Relative abundance of the most prominent phyla following the three study diets. The MIXED procedure was used to determine the effect of diet on each outcome measure adjusted for the baseline value (SAS version 9.4; SAS Institute Inc., Cary, NC). Post-hoc tests were adjusted for multiple comparisons using the Tukey-Kramer method. Statistically significant between-diet differences are denoted with differing letters (P<0.05). Abbreviations: WFMD, walnut-fatty acid matched diet; ORAD, oleic acid replaced alpha-linolenic acid (ALA) diet; WD, walnut diet.

100

1 Bacteroidetes

0.5

Relative Relative Abundance 0 MObese Obese Overweight a.

1 Firmicutes

0.5

Relative Relative Abundance 0 MObese Obese Overweight b.

0.5 Proteobacteria

Relative Relative Abundance 0 MObese Obese Overweight c.

Figure 4-6. Relative abundance of the most prominent phyla in participants with overweight, obesity, and morbid obesity. The MIXED procedure was used to determine the effect of diet on each outcome measure adjusted for the baseline value (SAS version 9.4; SAS Institute Inc., Cary, NC). Post-hoc tests were adjusted for multiple comparisons using the Tukey-Kramer method. Statistically significant between-diet differences are denoted with differing letters (P<0.05). Abbreviations: MObese, morbidly obesity.

CoNet Analysis

Following the three study diets (WD, WFMD, and ORAD) and the run-in,

SWD, the ASV co-occurrence was quantitatively different (Figure 4-7). The WD network was comprised of 30 nodes; 26 were Firmicutes and four were

Bacteroidetes. The WD was largely dominated by Firmicutes including: 101

Agathobacter, Faecalibacterium, Blautia, Lachnospiraceae, and Roseburia and

Bacteroidetes including: Alistipes, Bacteroides-uniformis, and Parabacteroides.

There were six negative and 31 positive interactions between nodes. The nodes that showed the most connectivity were Lachnospiraceae NK4A136 (6 interactions) and

Subdoligranulum (6).

The WFMD network contained 32 nodes; 27 Firmicutes, three Bacteroidetes, and two Proteobacteria. The WFMD network contained a large number of Firmicutes including: Faecalibacterium, Lachnospiraceae, Blautia, Anaerostipes, and

Subdoligranulum and was abundant in three Bacteroidetes: Alistipes, Bacteroides- thetalotaomicron, and Odoibacter and two Proteobacteria: Bilophila and

Parasutterella. Three negative interactions were observed in the WFMD network and

34 positive interactions were among nodes. The nodes with the highest connectivity were Ruminiclostridium (6 interactions), Ruminococcaceae UCG002 (6), and

Roseburia (5).

The ORAD network was made up of 40 nodes; 36 Firmicutes and four

Bacteroides. The ORAD did not have nodes as large as the WD, WFMD, or the

SWD, which indicates it did not have a large relative abundance of any present taxa.

The ORAD network contained Bacteroidetes, including Bacteroides, and Firmicutes, including Faecalibacterium. There were 13 negative and 44 positive interactions between nodes in the ORAD network. The nodes that were most connected were

Subdoligranulum (9 interactions) and Lachnospiraceae NK4A136 (6).

102

The SWD network contained 32 nodes; 30 Firmicutes and two Bacteroidetes.

The network was dominated by Firmicutes: Faecalibacterium, Blautia, Agathobacter,

Lachnospiraceae, and Anaerostipes. The SWD network was also rich in two

Bacteroidetes: Alistipes and Odoribacter. The SWD network had two negative interactions and 58 positive among the nodes. The nodes with the highest connectivity included Subdoligranulum (12 interactions), Faecalibacterium (9), and

Lachnospiraceae (8).

b. 103

Figure 4-7. Co-occurrence network (CoNet) analysis for the walnut diet (a.), walnut fatty acid matched diet (b.), oleic acid replaces ALA diet (c.) and the run-in, standard Western diet (d.). Nodes (colored circles) were colored by phylum (all members of the same phylum are identical colors) and labeled by the genus or lowest available assigned taxonomic level. Blue nodes represent Bacteroidetes, red nodes represent Firmicutes, and green nodes represent Proteobacteria. Additionally, node size is relative to the abundance of the taxa (larger circles indicate larger abundance while smaller indicate a lower abundance). Edges (lines between nodes) are colored according to relationship where green indicates positive relationship and red indicates a negative relationship between the taxa.

104

Correlations with Cardiovascular Risk Factors

There were significant associations between the percentages of enriched bacteria following the WD with CVD risk factors (Table 4-4). Eubacterium eligens was associated with brachial MAP (R=-0.50; P=0.0009), central diastolic BP (-0.52;

0.0006), and central MAP (-0.47, 0.002). Lachnospiraceae was associated with brachial MAP (R=-0.37, P=0.02), central diastolic BP (-0.32; 0.04), central MAP (-

0.35; 0.02), TC (-0.35; 0.03); non-HDL-C (-0.37; 0.02). Leuconostocaceae was associated with brachial MAP (R=0.34; P=0.03) and central MAP (0.34; 0.03). There were no significant correlations between enriched bacteria following the WFMD or

ORAD with CVD risk factors (Table 4-5 Table 4-6).

105

Table 4-4. Correlations between enriched taxa following the walnut diet (WD) and cardiovascular risk factors that were significantly changed from baseline non- TC:HDL Bacteria bMAP cDBP cMAP TC LDL-C HDL-C -C Roseburia -0.03 -0.12 -0.05 -0.23 -0.31 -0.25 -0.17 Eubacterium_ -0.50** -0.52** -0.47** -0.26 -0.31 -0.20 -0.12 eligensgroup Lachnospiraceae -0.37* -0.32* -0.35* -0.06 -0.10 -0.05 0.07 UCG_001 Lachnospiraceae -0.10 -0.17 -0.09 -0.35* -0.27 -0.37* -0.15 UCG_004 Leuconostocaceae 0.34* 0.31 0.34* 0.24 0.17 0.29 0.24 Ruminococcaceae -0.13 -0.11 -0.06 -0.16 -0.26 -0.13 0.01 uncultured 0.07 0.06 0.06 -0.002 -0.03 -0.01 -0.16 Butyricicoccus -0.01 0.03 -0.02 0.02 -0.06 0.02 0.10 Gordonibacter 0.25 0.27 0.26 0.16 0.15 0.15 0.29 DefluviitaleaceaeU 0.04 -0.11 -0.02 0.17 0.11 0.14 -0.03 CG_011 Defluviitaleaceae 0.04 -0.11 -0.02 0.17 0.11 0.14 -0.03

Statistical analyses were performed with SAS (version 9.4; SAS Institute; Cary, NC). The correlation procedure (PROC CORR) was used with Spearman’s correlation coefficient to assess association between the cardiovascular risk factors that were significantly changed with bacteria that were significantly enriched a pre-determined α value of 0.05

Abbreviations: bMAP, brachial mean arterial pressure; cMAP, central mean arterial pressure; cDBP, central diastolic blood pressure; TC, total cholesterol, LDL-C low- density lipoprotein cholesterol, non-HDL-C, non-high-density lipoprotein cholesterol. * indicates P<0.05; ** indicates P<0.01

106

Table 4-5. Correlations between enriched taxa following the walnut fatty acid matched diet (WFMD) and cardiovascular risk factors that were significantly changed from baseline non-HDL- Bacteria TC LDL-C HDL-C C Roseburia 0.04 -0.05 0.13 -0.03 Eubacterium_ 0.13 0.10 0.06 0.11 eligensgroup gutmetagenome -0.24 -0.21 -0.24 -0.16 Butyricicoccus -0.04 -0.12 -0.009 -0.03 Faecalibacterium 0.09 0.14 0.17 -0.15 Angelakisella 0.08 0.11 0.10 -0.05

Abbreviations: TC, total cholesterol, LDL-C low-density lipoprotein cholesterol, non- HDL-C, non-high-density lipoprotein cholesterol.

107

Table 4-6. Correlations between enriched taxa following the oleic acid replaces ALA diet (ORAD) and cardiovascular risk factors that were significantly changed from baseline

Bacteria TC LDL-C non-HDL-C HDL-C gutmetagenome 0.28 0.23 0.23 0.25 uncultured -0.14 -0.008 -0.14 0.20 ClostridialesvadinBB60 -0.16 0.03 -0.16 0.17 Ruminiclostridium1 0.10 0.13 0.09 0.21

Abbreviations: TC, total cholesterol, LDL-C low-density lipoprotein cholesterol, non- HDL-C, non-high-density lipoprotein cholesterol.

The MaAsLin analysis showed when comparing every individual bacteria

present in the sample to the magnitudes of change of CVD risk factors, there were

significant associations between gut bacteria and CVD risk factors following the

WFMD and ORAD; however, the false discovery rate (FDR, indicated by the Q-

value) for the majority of the significant relations was 1, which indicates there is

100% likelihood that these associations occurred by chance. There were not

significant associations between the bacteria present following the WD and CVD risk

factors. The run-in diet bacteria, Hungatella was associated with CVD risk factors,

cMAP (P=-0.006, Q=0.004) and cDBP (0.004, Q=0.08) and Coprococcus was

associated with CRP (P=0.01, Q=0.03).

108

PICRUSt

There were several pathways that were predicted to be enriched in pairwise comparisons (Figure 4-8, Table 4-7). Genes related to bacterial invasion of epithelial cells were enriched following the SWD (LDA=0.12, P=0.049, 406 genes) compared to the WFMD (212 genes) and butirosin and neomycin biosynthesis genes were enriched following the SWD (LDA=1.8, P=0.04, 502999 genes) compared to the WD (464480 genes). Basal transcription factor genes were enriched following the WFMD (LDA=1.5, P=0.03, 7962 genes) relative to the WD

(3268 genes). Beta alanine metabolism genes were enriched following the WD

(LDA=1.6, P=0.03, 1215388 genes) compared to the WFMD (946353 genes). There were no differential pathways between the SWD versus ORAD, WD versus ORAD, and WFMD versus ORAD.

109

Figure 4-8. Linear discriminant analysis effect size (LEfSe) plots were utilized to visualize phylogenetic investigation of communities by reconstruction of unobserved states (PICRUSt) predicted functional pathways of pairwise comparisons between diets. PICRUSt was performed on DADA2 processed sequences and clustered via closed reference clustering to Greengenes 13_5. Subsequently, predicted KO terms were then collapsed into KEGG pathways and compared pairwise between the diets. Four pathways were differentially abundant from three comparisons. The run-in diet compared to the walnut diet (WD) indicated Butirosin and neomycin biosynthesis enriched in the run-in diet (a.). The run-in compared to walnut fatty acid matched diet (WFMD) predicted that Bacterial Invasion of Epithelial cells was weakly enriched in the run-in diet (b.). The WD compared to the WFMD indicated that Beta-Alanine metabolism and Basal Transcription Factors were enriched, respectively (c.).

110

Table 4-7. Linear discriminant analysis effect size (LefSe) analysis of phylogenetic investigation of communities by reconstruction of unobserved states (PICRUSt) between study diets and versus the run-in diet Gene LDA Comparison Diet P-Value Count Score WD vs. Run-in Butirosinandneomycinbiosynthesis Run-in 502999 1.77 0.041 WFMD vs. Run-in Bacterialinvasionofepithelialcells Run-in 406 0.118 0.049 WD vs. WFMD Basaltranscriptionfactors WFMD 7962 1.53 0.031 beta_Alaninemetabolism WD 1215388 1.56 0.033

Linear discriminant analysis effect size (LEfSe) plots were utilized to visualize phylogenetic investigation of communities by reconstruction of unobserved states (PICRUSt) predicted functional pathways of pairwise comparisons between diets. PICRUSt was performed on DADA2 processed sequences and clustered via closed reference clustering to Greengenes 13_5. Subsequently, predicted KO terms were then collapsed into KEGG pathways and compared pairwise between the diets. Four pathways were differentially abundant from three comparisons.

Discussion

To our knowledge, this is the first study to examine the effects of diets

containing walnuts and vegetable oils compared to a diet higher in SFA on the

human gut microbiome composition, predictive functional analysis, taxonomic co-

occurrence, and associations with cardiovascular risk. Our findings together with

previous research on walnuts and the gut microbiome assist in understanding the

role of the gut microbiome in disease risk and prevention and how diet affects this.

Relative to the current study, we have shown that a diet containing whole walnuts

(including bioactives), a diet with a matched fatty acid profile to a walnut-containing

111

diet (devoid of whole walnuts and walnut bioactives), and a diet matched for SFA but includes oleic acid for the content of ALA found in walnuts (devoid of walnuts and walnut bioactives) differentially affected the gut microbiome. Our results demonstrate enrichment of eubiotic bacteria, including butyrate producers such as Roseburia, following the WD and WFMD. The bioactives in walnuts may also play an important role in altering the gut environment. Walnuts contain ellagitannins that are metabolized by gut bacteria to form urolithins, which may provide cardiovascular benefits. Gordonibacter is a bacteria involved in ellagitannins metabolism, which was enriched following the WD. Further, this study showed possible microbiome-CVD interactions between enriched bacteria following the WD and CVD risk factors. Thus, consuming walnuts as a replacement for SFA elicits beneficial changes in gut bacteria which may have important health implications.

The present controlled-feeding study showed differentially enriched bacteria among all three study diets relative to the Western-style, run-in diet, which suggests replacing SFA with unsaturated fatty acids in the presence or absence of walnuts affects the gut microbial environment. Both the WFMD and WD led to significant enrichment in Roseburia, Eubacteria eligens, and Butyricicoccus. The enrichment of

Roseburia following the WD is a similar finding to Holscher and colleagues134, who reported the relative abundance of Roseburia was negatively associated with secondary bile acids. Holscher et al. also reported significant differences in β- diversity. Of note, our study and the study conducted by Holscher et al. were controlled-feeding studies that provide a more robust assessment of diet –

112

microbiome interactions due to the tightly controlled study design implemented. In contrast, Bamberger et al.102 conducted a free living study and, consistent with our data and Holscher et al.’s134 findings, reported enrichment of Ruminococcaceae and

Lachnospiraceae after walnut consumption compared to a walnut-free diet.

Bamberger and colleagues also reported significantly different β-diversity between diets. Although we did not see differences in β-diversity among diets, this may be due to our extremely similarity treatment diets; diets only differed in the walnut bioactive profile or ALA content. Otherwise, our results are similar to the two previous studies102,134, however, these studies were conducted in healthy individuals and were not able to identify which component of walnuts may be inducing beneficial changes to the gut microbiome.

The significantly enriched bacteria and predictive functional analysis suggest walnut consumption and a diet that contains an equal amount of fatty acids promotes a favorable environment in the gastrointestinal tract and may also be important in gut epithelial maintenance. Increased gut permeability or “leaky gut” is associated with a variety of diseases such as non-alcoholic fatty liver disease and obesity159,160. SCFA, butyrate in particular, are a main energy source of enterocytes and are important for the maintenance of the intestinal epithelium161,162. Increased

SCFA availability decreases intestinal permeability163,164. Roseburia and

Eubacterium were enriched following both the WD and WFMD and

Leuconostocaceae was enriched following the WD; all of these bacteria and are capable of SCFA production165,166. Ruminococcaceae was also enriched following

113

the WD and may also be involved in intestinal epithelium maintenance as it is inversely correlated with intestinal permeability167–169. Further, genes involved in bacterial invasion of epithelial cells were enriched following the Western-style, run-in diet relative to the WFMD. Unfortunately when we examined the contributions of bacteria within the enriched PICRUSt pathway (data not shown), the bacteria species had not been identified, but they possess genes related to the enriched pathway. Identification of SJTU_G_09_34, SHZO627, and aab28d05 could provide more information on the enrichment of invasion of epithelial cells. This pathway has been connected to increased gastrointestinal epithelial permeability170 and suggests a higher-SFA, Western-style diet, which participants followed previous to the study diets, may increase gut permeability. The MaAsLin analysis also showed a positive association between gut bacteria and systemic inflammation following the run-in diet, which could be linked to increased gut bacteria infiltration. Although we did not analyze gut-derived metabolites, the enrichment of SCFA-producing bacteria following the WD and WFMD suggests the fatty acid composition, including ALA, may be protective against gut permeability.

The co-occurrence networks constructed to assess the bacterial community following each of the study diets illustrates potentially protective clustering following the WD and WFMD. Baldassano and Bassett171 reported healthy individuals had a high degree of bacterial modularity compared to individuals with inflammatory bowel disease (IBD), who had less distinctive community structures. The WD and WFMD networks had more modular bacterial clusters such as the Blautia, Agathobacter,

114

Lachnospiraceae, Fusicatenibacter cluster in the WD co-occurrence network and the

Lachnospiraceceae, Eubacterium eligens, Lachnospira cluster in the WFMD network. The ORAD network did not indicate any clustering and there was limited clustering in the SWD network. The interactions between bacterial nodes were primarily positive in the WD, WFMD, and SWD networks, with the greatest number of negative interactions following the ORAD. The large numbers of positive bacterial interactions correspond to a stable gut environment. A common trend in the co- occurrence network analyses is the negative interactions among several unfavorable bacteria with potentially beneficial bacteria, Lachnospiraceae indicating this bacterium may have a protective effect within the bacterial community. The observed positive interactions and modular bacterial clustering in the WD and

WFMD networks could provide protection against pathogenic bacteria to maintain host health.

The enrichment of Gordonibacter following the WD relative to the WFMD suggests the walnuts may be altering the gut microenvironment to better metabolize the bioactives present in walnuts. Gordonibacter is known to metabolize ellagitannins, which are uniquely high in walnuts, to urolithins, which is the primary form in which ellagitannins are absorbed into the body155,172. Urolithin A has been negatively correlated with cardiovascular risk factors, such as plasma glucose.

Further, when we examined the RA of Gordonibacter, the RA was significantly greater following the WD compared to both the WFMD and ORAD. The increase in ellagitannins-metabolizing bacteria following walnut consumption suggests the gut

115

microbiome is involved in the underlying mechanisms associated with the cardiovascular benefits of walnut consumption.

The WD bacterial community may also be involved in unique metabolic pathways, such as beta-alanine metabolism. Relative to the WFMD the WD showed an upregulation in genes involved in beta alanine metabolism. This is a surprising finding since beta-alanine is predominantly found in animal products and is considered the rate-limiting step in muscle carnosine synthesis. Beta-alanine is also used as a supplement for athletic performance173, but to understand how walnut consumption could be involved in beta-alanine synthesis a metatranscriptomics analysis is necessary.

In the absence of clearly defined underlying mechanisms for the association between the gut microbiome and cardiovascular risk, we performed correlation analyses between enriched bacteria and CVD risk factors from the primary analysis of this study. We demonstrated that members of the Lachnospiraceae family are inversely correlated with cardiovascular risk factors following a diet containing walnuts. Increases in Lachnospiraceae were correlated with decreases in brachial

MAP, central MAP, central diastolic BP, TC, and non-HDL-C. Increased abundances of Eubacterium eligens and Leuconostocaceae were also correlated with reductions in brachial MAP, central MAP, and central diastolic BP. Although further research examining bacterial functionality and secondary metabolites is needed to provide more mechanistic insight, this exploratory analysis suggests that several bacteria

116

are involved in the underlying mechanisms for walnut consumption and reduced cardiovascular risk.

Previous reports of interactions between BMI and the gut microbiome prompted us to perform an exploratory analysis on phylogenic taxonomy by BMI in the present study. Although we did not observe any differences between individuals with overweight, obesity, or morbid obesity in the prominent phyla, Firmicutes,

Bacteroidetes, and Proteobacteria, there could be differences at lower taxonomic levels or functional differences that were not detected. We did observe possible interactions between bacteria that may affect BMI. The co-occurrence network of the

WD showed Lachnospiraceae could be supressing Lachnoclostridium, a bacteria that may be associated with higher BMI174. The enrichment of Lachnospiraceae following the WD could play a role in weight status as it may be suppressing bacteria associated with higher BMIs. This warrants further exploration of interactions between weight status and gut bacteria given the increasing rates of obesity globally.

The strength of this study is the use of a controlled-feeding, randomized, controlled crossover trial with a run-in period to examine the differences in gut microbial composition and correlations with cardiovascular risk markers in individuals at risk for CVD following a diet containing walnuts as a SFA replacement.

Study limitations include the lack of secondary metabolites, such as SCFA, and metatrancriptomics analyses to assess the functional capacity of gut bacteria.

117

In conclusion, this study, which was designed to investigate which component(s) of walnuts underlie the beneficial cardiovascular effects, suggests that whole walnuts (fatty acids and bioactives), and the fatty acid profile may differentially affect the gut microbiota relative to a Western-style diet higher in saturated fats.

Similarities between enrichment of SCFA-producing bacteria, including Roseburia and Eubacterium, following the WD and WFMD illustrate the effect high unsaturated fat content, including ALA, may have on gut bacteria. The unique enrichment of

Gordonibacter following the WD suggests walnut bioactives modulate gut microbiota. The associations between Lachnospiraceae and improved cardiovascular risk factors suggest the gastrointestinal microbiota may contribute to the underlying mechanisms of the beneficial health effects of walnut consumption.

118

Chapter 5: Research summary and future directions

The overarching aim of this dissertation was to investigate the effects of walnut consumption on traditional and emerging CVD risk factors and the contributions of ALA versus bioactives in adults at risk for CVD. The present randomized, crossover, controlled-feeding study showed incorporation of walnuts into a healthy dietary pattern as a SFA-replacement for six weeks resulted in a greater lowering of cDBP compared to a diet devoid of walnuts and lower in ALA, but otherwise isocaloric, macronutrient-matched profile from vegetable oils. This is the first time walnut consumption has shown improvements in central blood pressure.

The greater reduction in cDBP may be attributable to the unique combination of fatty acids and bioactive compounds provided by whole walnuts. All three diets replaced

SFA with unsaturated fats, primarily with PUFA, and reduced TC, LDL-C, and non-

HDL-C, therefore, reducing cardiovascular risk. Analysis of the gut microbiome showed all study diets differentially affected the gut microbiome relative to the run-in,

Western-style diet. Similarities between enrichment of SCFA-producing bacteria following a diet containing walnuts and a diet matched for ALA suggests ALA may affect the gut environment. The unique enrichment of Gordonibacter, a bacteria involved in metabolism of walnut-derived bioactive, ellagitannins, following the consumption of walnuts suggests walnut bioactives may favorably alter the gut microbiome. Further, correlations with improved cardiovascular risk factors following the WD demonstrate the gut microbiome may play an important role in the cardiometabolic benefits of walnut consumption.

119

These two studies together provide evidence that substituting SFA with whole walnuts versus vegetable oils confers additional health benefits. The favorable blood pressure and gut bacteria changes observed following the walnut diet are important additions to the current literature. The AHA/ACC’s new blood pressure recommendations lowered the definition of hypertension to 130/80 mmHg and elevated blood pressure is classified as 120-129/≥80 mmHg. This means an estimated 103 million Americans will be categorized as hypertensive. Although we did not see significant changes in brachial blood pressure, there was a -2/1

(systolic/diastolic) mmHg reduction following the walnut diet. Other trials, such as the

Dietary Approaches to Stop Hypertension (DASH) reported larger effects, but also had a sample size of 459. If we performed the present study in a larger sample, it is possible we would have detected between-diet effects. Further, if we only performed this study in participants that had elevated blood pressure, rather than also including individuals with elevated LDL-C with normal blood pressure, we may have observed a greater effect on blood pressure. Our analysis of the gut environment is an important contribution to further our understanding of the gut environment as an emerging risk factor for CVD and other chronic diseases. Although we did not measure secondary metabolites, such as butyrate, secondary bile acids, or trimethylamine (TMA), the favorable community structure and eubiotic bacteria enriched following a diet containing walnuts helps to make inferences on the functionality of gut bacteria. A meta-transcriptomics analysis would provide information on the activities of the present gut bacteria, such as polyphenolic metabolism or SCFA production. The clear relation between diet, the gut 120

microbiome, and CVD warrants further investigation. Improvements in lipid/lipoproteins is consistent with the literature that replacement of SFA with PUFA lowers TC and LDL-C69. The combined favorable change in CVD risk factors following consumption of whole walnuts provides mechanistic evidence that both the fatty acid profile and bioactives play a role in reducing CVD risk.

The greatest total improvements in novel and established risk factors following a diet that provided 20% energy from walnuts compared to two diets with the same SFA content suggests a whole-food based approach to replacing SFA provides the greatest reduction in CVD risk. Although walnuts have a favorable fatty acid and bioactive profile, they cannot be ‘simply’ added to any diet and reduce the risk of CVD. It is important to consider the background diet walnuts were added to: a healthy diet that was high in fruits and vegetables and palatable to consumers for six weeks. In order for individuals to obtain walnut-derived benefits, it is critical that they are consuming an overall healthy dietary pattern. Further, the walnut dose we provided is greater than the current 2015-2020 Dietary Guidelines for Americans

(DGA) recommendations (142-198 g of nuts/week based on a 2000 calorie diet) and the long-term health effects require further study. In addition, the dose we provided may not be sustainable for a longer time. In order to provide dietary recommendations for walnut consumption, the context of the entire diet needs to be considered.

These results provide evidence for strengthening the current dietary recommendations to substitute saturated fats with unsaturated fats. The current

121

2015-2020 DGA provide recommendations, including limiting SFA intake to less than

10%, for making changes to consume a healthier dietary pattern. However, these recommendations have not been implemented well by Americans. Although the

DGA provide recommendations for consuming dietary patterns, including a healthy

American pattern, vegetarian pattern, and Mediterranean pattern, these diets are still based on nutrients, rather than whole foods. There needs to be a greater shift towards using whole foods in dietary guidance to make changes more achievable.

Although beyond the scope of this dissertation, factors that affect behavioral changes to shift towards a healthier dietary pattern need to be examined. No single food or nutrient is capable of preventing CVD. In order to reduce CVD risk, sustainable, effective dietary changes need to be implemented. It is clear from the

2015-2020 DGA report that Americans are not meeting dietary recommendations and just increasing walnut consumption will not resolve all of the problems with the

US diet. Future research should investigate effective ways to aid Americans in making maintainable dietary changes to consume a healthier dietary pattern.

In conclusion, this dissertation research addresses numerous nutrition and

CVD related topics, including how a healthy dietary pattern affects intermediate clinical biomarkers and gut bacterial communities. The information presented herein, we believe, will inform future dietary recommendations for CVD prevention.

The novel findings we report begin to describe the underlying mechanisms of walnut consumption on physiological and bacterial responses in humans. Further research

122

is needed to clarify our understanding of the role of walnuts and walnut constituents in modulating vascular health and gut bacterial functionality in humans.

123

Appendix

Informed Consent Form for Clinical Research Study The Pennsylvania State University

Title of Project: Effects of Walnuts on Central Blood Pressure, Arterial Stiffness Indices, Lipoproteins, and other CVD Risk Factors

Principal Investigators: Penny Kris-Etherton, PhD, RD

Department of Nutritional Sciences

110 Chandlee Lab, University Park, PA 16802

814-863-2923; Email: [email protected] Study Personnel: Alyssa Tindall, RDN, Project Coordinator

Email: [email protected]

Phone: 814-863-8056

(Please print your name) ______so that the person in charge of the research, Dr. Penny Kris-Etherton, would know that you have had a chance to read the information below. This form may contain words you do not understand. Please ask the study personnel to explain any words or information you do not clearly understand.

PLEASE READ EVERY PAGE CAREFULLY AND INITIAL THE BOTTOM OF EACH PAGE WHEN YOU HAVE HAD ALL OF YOUR QUESTIONS ANSWERED TO YOUR SATISFACTION.

Purpose of the Study You have been invited to participate in a clinical research study to evaluate the effects of walnut-delivered polyunsaturated fatty acids (PUFA) and bioactives on multiple cardiovascular disease risk factors, including blood lipid (fats), central blood pressure, and measures of blood vessel stiffness.

124

General Overview of the Study

Treatment Design If you agree to participate in this controlled feeding study, your participation will last for ~24 weeks total. Your time commitment would consist of a two week baseline diet and three, 6 week long treatment diets. You will be provided with all your meals and snacks during each treatment period. The order of the 3 treatment diets will be randomly assigned. This assignment is done in a way similar to flipping a coin – we use a computer program to assign the order of the diets that you will receive. An approximate 2-week break will be inserted between diet periods to reduce the monotony of participating in a controlled feeding study.

Diets for the three treatment groups are heart healthy cholesterol lowering diets. The diets include: 1) a high PUFA walnut diet (WD; providing ~2.0 oz of walnuts per day); 2) a matched walnut control diet (WCD) that will have the same fatty acid profile as the walnut diet, but will not contain walnuts (and their bioactives); and 3) a diet low in alpha-linolenic acid (a fatty acid found in walnuts; LAD), which will have a similar macronutrient composition as the walnut and walnut control diet with the exception of the fatty acid profile. Prior to starting the intervention diets you will be provided a Standard Western Diet (SWD) for two weeks to establish your baseline values. Calorie levels will be estimated for weight maintenance. This is not a weight loss study; therefore, calories will be adjusted as needed to ensure that you do not lose or gain weight over the course of the study. All intervention diets contain foods that are commonly found at a grocery store.

125

Study Schematic

Walnut Study Design 2 wk break 2 wk break

WD WCD LAD Run-in SWD

Screening 2 wk WCD LAD WD

LAD WD WCD

6 wk period 6 wk period 6 wk period RM

RM: randomization; SWD: Standard Western Diet; WD: Walnut Diet; LAD: Low ALA Diet; WCD: Walnut Control Diet

Indicates endpoint testing

Procedures to be Followed

Screening Tests If you decide to participate in the study and are considered eligible after the telephone screening, you will be further screened to determine your eligibility during a visit to the Clinical Research Center (CRC) at Penn State. The visit will consist of filling out forms (informed consent, medical history, personal information); measuring height and weight so your BMI can be calculated; and measuring blood pressure (BP). If after these measurements it is determined you are still eligible, a blood sample will be taken from your arm, or hand, and a complete blood count, health panel including liver and kidney function, and a blood fat panel will be performed (approximately 15 mls of blood or ~1 tablespoon will be taken). You will feel a small pinch or discomfort when the needle is inserted. If the initial blood draw is unsuccessful it may need to be repeated, with your permission. In addition, if you take thyroid medication you must provide a current (within 6 months) lab test. If you do not have one, an extra 3.5 ml (0.2 Tbsp) of blood will be taken to conduct a thyroid test. If you are a female of child bearing potential, you will be given a urine pregnancy test. You will be contacted within 3-5 days with the results of the screening blood sample. A clinician at the CRC will review all of the screening data and if you are still eligible for the study, you will be contacted to schedule your start date and baseline data collection appointments. There will be no charge for the screening blood work or measurements, and you will receive these results. If you 126

agree to participate in this study, you will agree to check with the study staff before participating in any other research studies; the study coordinator will let you know if it is alright to participate.

Feeding Study If you agree to participate in the study you will agree to eat only those foods (3 meals and a snack every day) and beverages provided to you (some non-caloric beverages are allowed) during the feeding periods of the study. You will come to one of the diet centers on campus Monday through Friday anytime between 6:30 am and 6:00 pm (you choose what time fits your schedule best), where you may eat your meal in the dining area or have it packed out for you. Your other two meals and snack will be packed and ready for you to take and eat at a place of convenience. On each Friday, you will be given a cooler that contains your Friday, Saturday and Sunday meals and snacks. You will be required to appropriately refrigerate and store all foods provided to you for take-out.

You are to eat only the foods given to you and nothing else. You must consume all of the food given to you. If for some reason you fail to do this, it is important that you tell the study staff that you did not follow protocol so they can make a note of it in your records. The information you provide to the study coordinators will be collected on two separate forms; one to be completed daily and one to be completed weekly. It should take less than 5 minutes to complete these forms each day. You will be weighed regularly when you pick up your foods so that your calorie intake may be adjusted over the course of the study in order to maintain your baseline body weight. This is not a weight-loss study, the diets are designed to meet your calorie needs and keep your body weight constant. Calorie intake will be adjusted up or down as necessary to maintain your weight. Also, you must keep your exercise level constant throughout the whole study.

Baseline and Endpoint Testing

Blood sampling: You cannot consume any food or drinks except water for 12 hours, and cannot drink alcohol during the 48 hours prior to having your blood taken. You also cannot engage in vigorous physical activity 12 hours prior to having your blood taken. In addition to the blood taken at screening, blood samples also will be taken on two days at baseline (following the 2 week SWD run-in) and the end of each test period (for a total of 8 times). After a twelve hour fast (consumption of no food or drinks except water), a blood sample will be taken from your arm or hand. If the initial blood draw is unsuccessful it may need to be repeated, with your permission. We also will measure your weight and perform pulse wave analysis (PWA) and pulse wave velocity (PWV) testing (described further below). Approximately 60 ml (~2 oz)

127

of blood will be collected at baseline and each endpoint across two consecutive days. Therefore, over the 24-week study, blood will be taken 8 times with a total amount of ~240 mls of total blood taken. A typical American Red Cross blood donation schedule is 1 pint (550 ml) per 8 weeks. Blood samples will be frozen and analyzed at the end of the study (when all participants have completed the study). The results of the study will only be available at the end of the entire study (which may take up to a year). Your blood may be tested for the following: blood fats (total cholesterol, LDL-cholesterol, HDL-cholesterol, triglycerides, LDL particle size and apolipoproteins), blood sugar (glucose, insulin), markers of inflammation, and vascular health. NO personal information will be kept with any sample – only ID# will be assigned and only the Principal Investigators and Study Coordinators will have access to the ID# assignments with the study files. If you are female of child bearing potential, you will be asked to provide a urine sample at baseline and the end of each treatment period. If you become pregnant during this time, you will be asked to discontinue your participation in the study.

Pulse Wave Analysis (PWA) and Pulse Wave Velocity (PWV): In addition to collecting blood samples, you will undergo a test that measures your blood pressure and pulse wave forms. The PWA measurement is very similar to a routine blood pressure measurement. Prior to the measurement, you will be asked to sit with your feet flat on the ground and rest for at least 5 minutes. A blood pressure cuff will be placed on your upper arm against your skin. The cuff will inflate, then deflate for 5 seconds, and then partially reinflate. It is important that you remain still during this measurement. After 1 minute of rest, the procedure will be repeated twice, for a total of 3 measurements at each time point this test is performed. Repeated measurements are used to increase accuracy. For the PWV measurement, we will ask you to lay flat on a hospital bed without a pillow. A thin cuff will be placed on your upper leg. We will gently place a hand held probe against an artery in your neck. This probe will measure the pressure waves of the blood in your artery. Once a good waveform is obtained, the blood pressure cuff on your leg will inflate to measure the pressure waveforms in that artery. Having these simultaneous measurements allows the device to calculate the speed at which blood is traveling through your arteries. The PWV test will be performed three times with at least one minute rest between measurements at each of the time points this test is performed.

Ambulatory Blood Pressure: You will arrive at the CRC in the morning to be instrumented with a SpaceLabs Healthcare ambulatory blood pressure device (model 90207) and a standard blood pressure cuff. The device will be worn for 24-hours and programmed to take 3 readings per hour during waking hours and 2 readings per hour during sleep. You should relax your arm when blood pressures are taken, and you have the option of stopping an individual BP reading if it interferes with activities such as driving. Ambulatory blood pressure will be measured at baseline and at the end of each diet period. 128

Urine Collection: At baseline and at the end of each diet period, you will be asked to collect a 24-hour urine sample. We will provide you with instructions and a collection jug that you will use to collect all of your urine over the 24 hours (e.g. 7AM to 7AM). The jug should be returned to the lab the next day when you come in for your second day of endpoint testing.

Fecal Sample Collection: At baseline and at the end of each diet period you will be asked to collect a stool sample (10-20 g) in the week prior to attending your visit to the CRC. You will be provided with a stool sample collection kit and detailed instructions for collection of a clean sample. You will be asked to freeze the sample immediately and keep them frozen until your scheduled endpoint visit.

Compliance with Study Protocol *** Please note: Successful completion of this study depends on the total cooperation of the participants. If during the study, you cannot comply with study procedures (such as consuming the diet or attending clinic visits); you will be asked to leave the study. Every effort will be made to give you a chance to comply with the study requirements, but if you do not follow the above study protocol you may be dropped from the study.

In addition, please advise us of any medical events (such as illness, injury, surgery etc) that arise during the course of the study. Depending on the event, we may require you to obtain a medical clearance before continuing with the study. Some medications may also interfere with our study outcomes so please inform us of any medication changes. ***

Time Commitment for the Study Total time for study visits to the CRC, after the initial screening, is approximately 30 hours. Times may vary and females will require an additional 5 minutes for a urine pregnancy test at screening, baseline and the end of each diet treatment period. The following is an estimate of the amount of time you will spend in study activities:

129

Day 1: Forms, BP, weight, height, blood draw – 45- 60 min (Pregnancy testing: Females only - 5 min)

Day 1: Blood draw, weight, PWA/PWV – 60 min (Pregnancy testing: Females only - 5 min) Day 2: Blood draw: 15 min 24 ambulatory blood pressure (at home)

24 hr urine collection (at home)

Fecal sample collection (at home)

-3: Day 1: Blood draw, weight, PWA/PWV – 60 min (Pregnancy testing: Females only - 5 min) Day 2: Blood draw: 15 min 24 ambulatory blood pressure (at home)

24 hr urine collection (at home)

Fecal sample collection (at home)

food ~ 15 min / 5 days per week for 20 weeks=1500 min or about 25 hrs

At home:

Total time commitment (with at home urine, fecal and blood pressure collections) is ~128 hrs.

Discomforts and Risks

Feeding Study

130

The diets used in this study are nutritionally adequate, whole-food diets. All foods will be prepared daily according to accepted standards of sanitation and provisions are made to ensure the safety of foods provided for off-site consumption. However, it is possible that incorrect food handling during shipping, storage or preparation, if not detected, could result in food-borne illness. Every effort will be made to safeguard against this possibility. To date, no food related contamination or illnesses have occurred. Feeding studies that require on-site eating of meals and strict adherence to the diets provided may interfere with social activities centered around eating such as dining in restaurants. While the menus will provide some variety in the diets, the number of food items will be more limited than that available in an average grocery store. The limited variety may become boring over the course of the study. In addition, some participants experience GI (stomach) upset from the change of diet; symptoms may include, but are not limited to, any of the following: constipation/diarrhea, nausea, and bloating. If this occurs, it will likely subside once you become accustomed to the new diet. Should you experience any type of food related allergic response please inform study personnel immediately and seek medical attention as needed.

Blood Sampling Blood draws often cause mild pain, swelling or bleeding. There may be some bruising (blood under the surface of the skin), which will be minimized by pressing on the site after the needle is removed. There is also a slight chance of infection, dizziness or fainting. These risks will be minimized, and most likely eliminated, by having trained medical staff draw the blood in a clinical setting using sterile supplies. If dizziness or fainting occurs, the symptoms will be alleviated by having the participant lie flat with their feet raised. The medical staff will also ask that the participant remain at the clinic until their blood pressure has been checked and the participant is cleared from any further risk.

Pulse Wave Analysis (PWA) and Pulse Wave Velocity (PWV)

There are no known risks associated with these measurements. The sensation of pressure from the blood pressure cuff or hand-held probe may be uncomfortable. There is a possibility for red blotching or mild bruising (petechiae) appearing on the skin above and below the location of the blood pressure cuff. Studies indicate that petechiae are rare (occurring in less than ½ of 1% of patients) and it is typically not uncomfortable and does not require treatment.

131

Ambulatory Blood Pressure: You will be instrumented with an ambulatory blood pressure device and a standard blood pressure cuff to be worn for 24-hours. The risks involved with wearing this device include minimal interference with daily tasks (such as sleeping and bathing), possible discomfort with inflation of the cuff, and inconvenience due to interference with tasks and having to stop activities to relax your arm when the cuff inflates.

Urine Collection Collecting urine over a 24-hour period while at work or at home can be a disturbance to your normal schedule. Making reminders to yourself to use the provided orange jug, especially during sleep hours, is helpful. In addition, some people may be uncomfortable collecting their own urine at first, but if you are careful it is sanitary and safe.

Fecal collection

You may experience some level of embarrassment or discomfort from being asked to collect stool samples. However, you will be provided with detailed instructions on how to collect the samples within the comfort of your own home, and at your convenience, to help reduce any concerns you may have.

Benefits to You You will have a chance to learn the principles of good nutrition practices. You also will receive the results of your screening blood work and information about how your blood cholesterol changed in response to the experimental diets. At the end of the study you will receive information about the effect of each of the diets on your blood cholesterol, blood pressure, inflammation, and vascular health. The final results of the study will not be available until all of the analysis is completed, which may take up to three years. However, no benefit from participation in this study is guaranteed.

Potential Benefits to Society We believe that walnuts will favorably affect many CVD risk factors due to the bioactive components they provide. Our diet design will allow us to ascertain the specific effects that walnuts and their bioactive components (including and beyond alpha-linolenic acid) may have on CVD risk factors and artery health. Our findings also will be used to inform evidence-based dietary recommendation updates.

Study Funding Source Information The funding for this study is provided by The California Walnut Commission. However the funding source will not be involved in data analysis. They will have the right to review all publications before submission however there are no contractual

132

agreements that allow them to have influence on, or restrict, the publication of results. The PI has no affiliation with The California Walnut Commission.

Statement of Confidentiality Your participation in this research is confidential. All records are coded with a unique ID number and no names are used. Records containing names or other identifying information are kept under lock at the PI’s research office. All records associated with your participation in the study will be subject to the usual confidentiality standards applicable to medical records. In the event of publication of this research, no personal identifying information will be disclosed. Your blood specimens will be coded with your unique ID number and will be maintained until three years after the date from when the study is published, and then destroyed unless (see end of document) you give permission for use to keep your blood samples for future use. At the end of the study (after all subjects have completed the study), you will be given your laboratory results without cost, and informed of the study results.

A description of this clinical trial will be available on http://www.clinicaltrials.gov, as required by US Law. This website will not include information that can identify you. At most, the website will include a summary of the results. You can search this website at any time.

The following may review records related to this research: The Office of Human Research Protections in the U.S. Dept. of Health and Human Services; the Food and Drug Administration (FDA), Penn State University’s Institutional Review Board and Office for Research Protections.

Right to Ask Questions Please contact Dr. Kris-Etherton at (863-2923 or 863-8056) with any questions, complaints or concerns about the research. You can also call this number if you feel this study has harmed you. If you have any questions, concerns, problems about your rights as a research participant or would like to offer input, please contact Penn State University’s Office for Research Protections (ORP) at (814) 865-1775. The ORP cannot answer questions about research procedures. Questions about research procedures can be answered by the research team.

If the principal investigator or study staff becomes aware of new information or research findings that might affect your willingness to participate in this study, you will be given that information. You will be given the opportunity to ask any questions you might have and to decide if you want to continue to participate in the study. 133

Compensation For your time and participation in the study you will receive monetary compensation of $450.00, prorated as follows and paid at the completion of your participation in the study. If you drop out of the study for any reason before its completion the following compensation will be provided: $125 for completion of period 1 $125 for completion of period 2 $200 for completion of period 3 (Total =$450)

Total payments within one calendar year that exceed $600 will require the University to annually report these payments to the IRS. This may require you to claim the compensation that you receive for participation in this study as taxable income.

Injury Statement In the unlikely event you become injured as a result of your participation in this study, medical care is available. It is the policy of this institution to provide neither financial compensation nor free medical treatment for research-related injury. By signing this document, you are not waiving any rights that you have against The Pennsylvania State University for injury resulting from negligence of the University or its investigators.

Voluntary Participation Your participation in this study is voluntary. You may decline to answer any questions during the screening process or during the study. Please be aware that if you refuse to answer a number of the questions, the researchers will not be able to obtain enough needed data and it may become impossible to keep you in the study. You may withdraw from this study at any time by notifying the investigators or other study personnel. You should be aware that once withdrawn, any data collected up to the point of withdrawal must remain part of the study database and cannot be removed. Refusal to take part in or withdrawing from this study will involve no penalty or loss of benefits you would otherwise receive. You may be asked to leave the study at any time if you do not comply with the study protocol.

In the event that abnormal lab test results are obtained during initial screening or subsequently throughout this study, you will be informed as quickly as possible of these results and instructed to contact your private physician for further assessment. The lab test results will be made available to your private physician at your request.

This is to certify that you consent to and give your permission for your participation as a volunteer in the study entitled “Effects of Walnuts on Central Blood 134

Pressure, Arterial Stiffness Indices, Lipoproteins, and other CVD Risk Factors”. You must be 18 years of age or older. You will receive a signed copy of this consent form.

______Signature of Volunteer Date

______Printed Name of Volunteer

______Signature of Investigator Date

In addition to the main part of the research study, there is an optional part of the research. You can participate in the main part of the research without agreeing to take part in this optional part.

Storage of Leftover Specimens (blood, urine and stool) for Future Research Studies

As part of this study, we are obtaining blood, urine and fecal samples from you. If you agree, the research team would like to store these leftover samples so that your samples may be studied in the future after this study is over. These future studies may provide additional information that will be helpful in understanding gut health and cardiovascular disease, but it is unlikely that these studies will have a direct benefit to you. Neither your doctor nor you will receive results of these future research tests, nor will the results be put in your health record. If you have any questions, you should contact Dr. Kris-Etherton at 814-863-2923.

Your leftover samples will be labeled with a code number and stored in Dr. Kris- Etherton’s locked laboratory. If you consent to the storage of leftover samples for future research, the period for the use of the samples is unknown. If you agree to allow your samples to be kept for future research, you will be free to change your mind at any time. You should contact Dr. Kris-Etherton at 814-863-2923 and let her know you wish to withdraw your permission for your samples to be used for future research. Should you choose not to allow for future testing of your samples they will be destroyed 3 years after publication of study results.

135

You should initial below to indicate your preferences regarding the optional storage of your leftover samples for future research studies. a. Your samples may be stored and used for future research studies to learn about, prevent, treat or cure GI disorders, cardiovascular disease and other health problems. ______Yes _____ No b. Your samples may be shared with other investigator/groups without any identifying information. ______Yes _____ No

Participant: By signing below, you indicate that you are voluntarily choosing to take part in this optional part of the research.

______Signature of Participant Date Time Printed Name

Do we have permission to keep your personal information and contact you about your interest in participating in future studies for Dr. Kris-Etherton and her collaborators?

______Yes _____ No ______Initials

Person Explaining the Research: Your signature below means that you have explained the optional part of the research to the participant/participant representative and have answered any questions he/she has about the research.

______Signature of person who explained Date Time Printed Name this optional research study.

136

Six-day study diet menus

Day 1 Run-in diet WD WFMD ORAD Item g Item g Item g Item g Breakfast Canadian Wheat Wheat 28 Bluberries 70 26 26 bacon bagel bagel Wheat Plain bagel 67 26 Nonfat milk 2% Milk 115 bagel Nonfat 2% Milk 244 227 Margarine Margarine 7 milk American 32 Margarine 5 Shake: Shake: cheese Light yogurt 170 Oatmeal 28 Oatmeal 28 Oatmeal 28 Blueberries 70 Blueberries 70

2% Milk 100 2% Milk 120

Applesauce 113 Applesauce 113

High High linoleic linoleic 20 10 safflower oil safflower oil Sunflower 12 oil Lunch Wheat Wheat 55 43 Wheat bun 43 Wheat bun 43 bread bun Beef 50 Meatballs 85 Meatballs 100 Meatballs 100 lunchmeat Sliced Marinara Marinara Marinara 60 61 61 61 tomatoes sauce sauce sauce Sun chips® 30 Lettuce 20 Lettuce 45 Broccoli 36 Light Baby 15 Broccoli 36 Broccoli 36 57 mayonnaise carrots Baby Baby Ranch Apple 100 57 57 15 carrots carrots dressing Canola Ranch Ranch 23 23 eggless 14 dressing dressing mayonnaise Pretzels 28 Pretzels 28 Pretzels 28

High High oleic 7 7 linoleic safflower oil 137

safflower oil

Dinner Beef strip Beef strip Beef strip Spaghetti 140 85 90 87 steak steak steak Ground Fajita Fajita 65 235 235 Fajita sauce 235 beef sauce sauce Marinara Red Red Red 115 75 75 75 sauce peppers peppers peppers Flour Olive oil 5 84 Flour tortilla 84 Flour tortilla 84 tortilla Parmesan Nonfat sour Nonfat sour 7.5 Lettuce 20 24 24 cheese cream cream Green High linoleic 80 Flaxseed oil 9.5 6 beans safflower oil Garlic bread 50 Lettuce 40

Chocolate 99 pudding Snack Chocolate Chocolate M&M's® 35 Apple 100 110 110 pudding pudding Walnuts 57

Chocolate 110 pudding

138

Day 2 Run-in diet WD WFMD ORAD Item g Item g Item g Item g Breakfast 2% Milk 244 Nonfat milk 227 Nonfat milk 235 Nonfat milk 235 Cheerios 40 Granola 75 Granola 75 Granola 75 cereal Orange Orange Orange Butter 9 249 124 165 juice juice juice Wheat 50 Banana 116 Shake: Shake: bagel Orange Orange Banana 116 125 100 juice juice Banana 116 Banana 116

Raspberries 140 Raspberries 140

High- High- linoleic 13 linoleic 16

safflower oil safflower oil Sunflower Flax oil 12 10 oil Lunch Turkey Chicken Chicken Chicken 56 78 82 80 lunchmeat breast breast breast Provolone Green Green Green 28 6 6 6 cheese onion onion onion Wheat Nonfat sour Nonfat sour Nonfat sour 70 12 16 16 bread cream cream cream Canola Sliced Light Light 60 20 15 eggless 14 tomatoes mayonnaise mayonnaise mayonnaise Canola 15 Lemon juice 6 Lemon juice 6 Lemon juice 6 mayonnaise Broccoli 40 Red grapes 45 Red grapes 45 Red grapes 45 Carrots 70 Wheat pita 90 Wheat pita 90 Wheat pita 90 Ranch 20 Cheez-it® 15 Cheez-it® 15 Cheez-it® 15 dressing Pretzels 32

Dinner Turkey taco Chicken Chicken Chicken 160 56 80 80 mixture breast breast breast

139

Cheddar Mashed Mashed Mashed 28 105 105 105 cheese potatoes potatoes potatoes Chicken Chicken Chicken Salsa 64 marsala 60 marsala 30 marsala 32 sauce sauce sauce Green Green Green Lettuce 80 70 70 70 beans beans beans Flour tortilla 70 Margarine 10 Margarine 7 Margarine 7 High- High-oleic Sour cream 23 linolenic 14 5 safflower oil safflower oil Sunflower Corn 75 4 oil Snack Cheddar Cheddar Cheddar Light yogurt 170 22 22 25 cheese cheese cheese Graham Graham Graham 45 45 45 crackers crackers crackers Walnuts 57

140

Day 3 Run-in diet WD WFMD ORAD Item g Item g Item g Item g Breakfast Wheat Wheat Wheat Wheat English English English English muffin 57 muffin 65 muffin 67 muffin 67 Scrambled Scrambled Oatmeal 40 Egg beaters 75 eggs 70 eggs 70 2% Milk 244 Nonfat milk 227 Nonfat milk 235 Nonfat milk 230 Blueberries 80 Orange 130 Orange 130 Orange 130 Canadian Canadian Canadian Butter 9 bacon 25 bacon 28 bacon 28 American cheese 12 Lunch Ham Turkey Turkey Turkey lunchmeat 82 lunchmeat 84 lunchmeat 95 lunchmeat 95 Reduced- fat Swiss provolone Provolone Provolone cheese 40 cheese 20 cheese 20 cheese 20 Wheat Wheat Wheat Rye bread 55 bread 67 bread 72 bread 72 Canola Canola Canola Canola mayonnaise 15 mayonnaise 10 mayonnaise 18 mayonnaise 14 Canned pineapple 110 Applesauce 226 Lettuce 30 Lettuce 30 Fig Potato Potato Potato Newtons 35 chips 25 chips 30 chips 30 Light mayonnaise 10 Lettuce 30 Dinner Pork chop 98 Chili 325 Chili 335 Chili 330 Macaroni Cheddar Cheddar Cheddar and cheese 145 cheese 12 cheese 15 cheese 12 Wheat Wheat Wheat Lettuce 65 saltines 40 saltines 35 saltines 35

141

Cherry Green Green Green tomatoes 70 beans 87 beans 87 beans 87 Grated carrots 40 Margarine 7 Margarine 7 Margarine 7 Wheat High linoleic High linoleic dinner roll 55 safflower oil 15 safflower oil 16 Butter 10 Italian dressing 28 Snack Graham crackers 40 Yogurt 150 Shake: Shake: Walnuts 57 2% Milk 30 2% Milk 35 Yogurt 150 Yogurt 150 Raspberries 40 Raspberries 40 Banana 116 Banana 116 Sunflower Flax oil 7 oil 12 High linoleic High linoleic safflower oil 12 safflower oil 7

142

Day 4 Run-in diet WD WFMD ORAD Item g Item g Item g Item g Breakfast 2% Milk 224 Nonfat milk 227 Nonfat milk 127 Nonfat milk 127 Raisin bran 50 Oatmeal 56 Oatmeal 56 Oatmeal 56 cereal Wheat Canned 52 200 Shake: Shake: bagel peaches Cream 28 2% Milk 100 2% Milk 100 cheese Blackberry 21 Peaches 150 Peaches 140 jam Banana 58 Banana 58

Raspberries 100 Raspberries 80

Sunflower Flax oil 11 15 oil High linoleic High oleic 17 12 safflower oil safflower oil Lunch Wheat Wheat Wheat Wheat pita 62 75 75 75 bread bread bread Canned Canned Canned Canned 56 70 85 85 tuna tuna tuna tuna Canola Canola Canola Canola 10 15 eggless 24 eggless 23 mayonnaise mayonnaise mayonnaise mayonnaise Diced Diced Diced Diced 7 45 45 45 celery celery celery celery Pickle relish 7 Lettuce 20 Lettuce 20 Lettuce 20 Baby Baby Baby Baby 80 70 70 70 carrots carrots carrots carrots Blue cheese 20 Hummus 40 Hummus 40 Hummus 40 dressing Chocolate Chocolate Chocolate Sun chips® 30 110 110 110 pudding pudding pudding Dinner Turkey Ground Ground Ground 56 40 40 40 breast beef beef beef

143

Mashed 195 Spaghetti 145 Spaghetti 155 Spaghetti 155 potatoes Turkey Marinara Marinara Marinara 70 128 128 128 gravy sauce sauce sauce Corn 75 Lettuce 70 Lettuce 70 Lettuce 70 Grated Grated Grated Biscuit 35 30 30 30 carrots carrots carrots Cherry Cherry Cherry Butter 15 30 30 30 tomatoes tomatoes tomatoes Light ranch Light ranch Light ranch Lettuce 52 25 25 25 dressing dressing dressing Cherry White White White 50 50 50 50 tomatoes dinner roll dinner roll dinner roll Balsamic High linoleic High linoleic 30 10 16 vinaigrette safflower oil safflower oil Snack Cheddar Cheddar Light yogurt 160 Walnuts 57 32 20 cheese chese Wheat Wheat Cheddar M&M's® 33 20 saltine 20 saltine 20 cheese crackers crackers Wheat saltine 20

crackers

144

Day 5 Run-in diet WD WFMD ORAD Item g Item g Item g Item g Breakfast 2% Milk 244 Nonfat milk 235 Nonfat milk 260 Nonfat milk 260 Pancakes 90 Granola 75 Granola 65 Granola 65 Orange Orange Orange Light syrup 50 188 75 75 juice juice juice Butter 10 Banana 116

Blueberries 75

Lunch Turkey Chicken Chicken Chicken 71 90 100 100 lunchmeat breast breast breast Lettuce 20 Spinach 115 Spinach 115 Spinach 115 Wheat 70 Red onion 10 Red onion 10 Red onion 10 bread Canola Feta Feta Feta 16 45 50 40 mayonnaise cheese cheese cheese American 28 Red grapes 55 Red grapes 50 Red grapes 50 cheese Baby High linoleic High linoleic 40 Canola oil 12 25 25 carrots safflower oil safflower oil High Balsamic Balsamic Apple 100 linoleic 7 7 15 vinegar vinegar safflower oil Balsamic Unsalted Unsalted Pretzels 32 8 15 15 vinegar pretzels pretzels Yogurt 150

Unsalted 15 pretzels Dinner Chicken Chicken Chicken Chicken 78 90 100 100 breast breast breast breast Spiral pasta 135 Jambalaya 145 Jambalaya 145 Jambalaya 145 Green Green Green Broccoli 62 87 87 87 beans beans beans Cherry 40 Margarine 10 Margarine 10 tomatoes High oleic Carrots 40 safflower oil 145

Green onion 8

Red onion 10

Italian 28 dressing Olive oil 7

Parmesan 10 cheese Wheat 55 dinner roll Butter 16

Snack Fig Newtons 30 Apple 100 Apple 100 Apple 100 Granola bar 24 Granola bar 24 Granola bar 24

Walnuts 57 Shake Shake

Banana 116 Banana 116

Orange Orange 100 100 juice juice Blueberries 24 Blueberries 25

Yogurt 150 Yogurt 150

Sunflower Flax oil 8 17 oil High linoleic 12 safflower oil

146

Day 6 Run-in diet WD WFMD ORAD Item g Item g Item g Item g Breakfast 2% Milk 240 Nonfat milk 220 Nonfat milk 230 Nonfat milk 230 Wheat Wheat Wheat Granola English English English 45 muffin 65 muffin 70 muffin 70 Light yogurt 170 Egg beaters 60 Egg beaters 70 Egg beaters 70 Wheat bagel 50 Orange 130 Orange 130 Orange 130 Butter Margarine 15 7 Lunch Wheat bun 63 Wheat bun 65 Wheat bun 75 Wheat bun 75 Chicken Veggie Veggie Veggie breast 83 burger 60 burger 70 burger 70 Canola American American American mayonnaise 14 cheese 25 cheese 33 cheese 33 Chipolte Lettuce Lettuce Lettuce spread 3 40 40 65 Dijon Dijon Dijon Fruit blend 250 mustard 15 mustard 15 mustard 15 Graham Graham Graham Sun chips® 35 crackers 40 crackers 40 crackers 40 Pear Pear Pear 165 165 165 Italian dressing 20 Sunflower oil 12 Dinner Vegetarian Chicken Chicken Chicken chili 350 breast 50 breast 50 breast 50 Cheddar Spaghetti Spaghetti Spaghetti cheese 26 160 185 185 Thai Thai Thai Lettuce noodles and noodles and noodles and 65 veggies 125 veggies 120 veggies 120 Cherry Lettuce Lettuce Lettuce tomatoes 60 70 70 70 Carrots 50 Carrots 30 Carrots 30 Carrots 30

147

Light Italian Cherry Cherry Cherry dressing 28 tomatoes 30 tomatoes 30 tomatoes 30 Corn bread Light Italian Light Italian Italian muffin 90 dressing 28 dressing 28 dressing 20 White dinner High oleic Flaxseed oil roll 30 9.5 safflower oil 6 High linoleic High linoleic safflower oil 25 safflower oil 13 Snack M&M's® 36 Sun chips® 28 Sun chips® 28 Sun chips® 28 Walnuts 57

148

References

1. Tindall AM, Petersen KS, Lamendella R, Shearer GC, Murray-Kolb LE, Proctor DN, Kris-Etherton PM. Tree Nut Consumption and Adipose Tissue Mass: Mechanisms of Action. Curr Dev Nutr. 2018. doi:10.1093/cdn/nzy069. 2. World Health Organization. Cardiovascular diseases (CVDs). Available at https://www.who.int/news-room/fact-sheets/detail/cardiovascular-diseases- (cvds). Accessed December 7, 2018. 3. Mozaffarian D, Wilson PWF, Kannel WB. Beyond Established and Novel Risk Factors. Circulation. 2008;117:3031–3038. 4. U.S. Department of Health and Human Services and U.S. Department of Agriculture. 2015-2020 Dietary Guidelines for Americans. ; 2015. 5. Micha R, Peñalvo JL, Cudhea F, Imamura F, Rehm CD, Mozaffarian D, G Y, S W. Association Between Dietary Factors and Mortality From Heart Disease, Stroke, and Type 2 Diabetes in the United States. JAMA. 2017;317:912. 6. US Department of Agriculture and US Department of Health and Human Services. 2015-2020 Dietary Guidelines for Americans. 8th Edition. Available at https://health.gov/dietaryguidelines/2015/guidelines/table-of-contents/. Accessed May 9, 2017. 7. Pan A, Chen M, Chowdhury R, Wu JHY, Sun Q, Campos H, Mozaffarian D, Hu FB. α-Linolenic acid and risk of cardiovascular disease: a systematic review and meta-analysis. Am J Clin Nutr. 2012;96:1262–73. 8. Paschos GK, Magkos F, Panagiotakos DB, Votteas V, Zampelas A. Dietary supplementation with flaxseed oil lowers blood pressure in dyslipidaemic patients. Eur J Clin Nutr. 2007;61:1201–6. 9. Kaplan HM, Kuyucu Y, Polat S, Pazarci P, Yegani AA, Şingirik E, Ertuğ P. Molecular basis of vascular damage caused by cigarette smoke exposure and a new approach to the treatment: Alpha-linolenic acid. Biomed Pharmacother. 2018;102:458–463. 10. Byerley LO, Samuelson D, Blanchard EI, Luo M, Lorenzen BN, Banks S, Ponder MA, Welsh DA, Taylor CM, Lauri BO, Derrick S, Eu-gene BI, Meng L, Brittany LN, Shelia B, Monica PA, David WA, Christopher TM. Changes in the Gut Microbial Communities Following Addition of Walnuts to the Diet. J Nutr Biochem. 2017. doi:10.1016/j.jnutbio.2017.07.001. 11. Schlörmann W, Birringer M, Lochner A, Lorkowski S, Richter I, Rohrer C, Glei M. In vitro fermentation of nuts results in the formation of butyrate and c9,t11 conjugated linoleic acid as chemopreventive metabolites. Eur J Nutr. 2016;55:2063–2073. 149

12. Lamuela-Raventos RM, St Onge M-P. Prebiotic Nut Compounds and Human Microbiota. Crit Rev Food Sci Nutr. 2016;:00–00. 13. Jenkins DJ, Kendall CW, Marchie A, Jenkins AL, Augustin LS, Ludwig DS, Barnard ND, Anderson JW. Type 2 diabetes and the vegetarian diet. Am J Clin Nutr. 2003;78:610S–616S. 14. Brufau G, Boatella J, Rafecas M. Nuts: source of energy and macronutrients. Br J Nutr. 2006;96 Suppl 2:S24-8. 15. Bolling BW, Chen C-YO, Mckay DL, Blumberg JB. Tree nut phytochemicals: composition, antioxidant capacity, bioactivity, impact factors. A systematic review of almonds, Brazils, cashews, hazelnuts, macadamias, pecans, pine nuts, pistachios and walnuts. Nutr Res Rev. 2011;24:244–275. 16. Mele L, Mena P, Piemontese A, Marino V, López-Gutiérrez N, Bernini F, Brighenti F, Zanotti I, Del Rio D. Antiatherogenic effects of ellagic acid and urolithins in vitro. Arch Biochem Biophys. 2016;599:42–50. 17. Sánchez-González C, Ciudad C, Noé V, Izquierdo-Pulido M. Health benefits of walnut polyphenols: An exploration beyond their lipid profile Health benefits of walnut polyphenols: An exploration beyond their lipid profile. Crit Rev Food Sci Nutr. 2015;:Crit. Rev. Food Sci. Nutr. 18. Garcia-Muñoz C, Vaillant F. Metabolic Fate of Ellagitannins: Implications for Health, and Research Perspectives for Innovative Functional Foods. Crit Rev Food Sci Nutr. 2014;54:1584–1598. 19. Kang I, Buckner T, Shay NF, Gu L, Chung S. Improvements in Metabolic Health with Consumption of Ellagic Acid and Subsequent Conversion into Urolithins: Evidence and Mechanisms. Adv Nutr. 2016;7:961–972. 20. Mora-Cubillos X, Tulipani S, Garcia-Aloy M, Bulló M, Tinahones FJ, Andres- Lacueva C. Plasma metabolomic biomarkers of mixed nuts exposure inversely correlate with severity of metabolic syndrome. Mol Nutr Food Res. 2015;59:2480–2490. 21. González-Sarrías A, García-Villalba R, Romo-Vaquero M, Alasalvar C, Örem A, Zafrilla P, Tomás-Barberán FA, Selma M V., Espín JC. Clustering according to urolithin metabotype explains the interindividual variability in the improvement of cardiovascular risk biomarkers in overweight-obese individuals consuming pomegranate: A randomized clinical trial. Mol Nutr Food Res. 2017;61:1600830. 22. Spigoni V, Mena P, Cito M, Fantuzzi F, Bonadonna R, Brighenti F, Dei Cas A, Del Rio D, Spigoni V, Mena P, Cito M, Fantuzzi F, Bonadonna RC, Brighenti F, Dei Cas A, Del Rio D. Effects on Nitric Oxide Production of Urolithins, Gut- Derived Ellagitannin Metabolites, in Human Aortic Endothelial Cells. 150

Molecules. 2016;21:1009. 23. Tomás-Barberán FA, García-Villalba R, González-Sarrías A, Selma M V., Espín JC. Ellagic Acid Metabolism by Human Gut Microbiota: Consistent Observation of Three Urolithin Phenotypes in Intervention Trials, Independent of Food Source, Age, and Health Status. J Agric Food Chem. 2014;62:6535– 6538. 24. Selma M V, Romo-Vaquero M, García-Villalba R, González-Sarrías A, Tomás- Barberán FA, Espín JC. The human gut microbial ecology associated with overweight and obesity determines ellagic acid metabolism. Food Funct. 2016;7:1769–74. 25. Jiang Q, Christen S, Shigenaga MK, Ames BN. γ-Tocopherol, the major form of vitamin E in the US diet, deserves more attention. Am J Clin Nutr. 2001;74:714–722. 26. U.S. Department of Agriculture. USDA National Nutrient Database for Standard Reference, Release 28. Available at https://ndb.nal.usda.gov/ndb/. Accessed September 13, 2017. 27. Robbins KS, Shin E-C, Shewfelt RL, Eitenmiller RR, Pegg RB. Update on the Healthful Lipid Constituents of Commercially Important Tree Nuts. J Agric Food Chem. 2011;59:12083–12092. 28. Landrier J-F, Gouranton E, Reboul E, Cardinault N, El Yazidi C, Malezet- Desmoulins C, André M, Nowicki M, Souidi M, Borel P. Vitamin E decreases endogenous cholesterol synthesis and apo-AI-mediated cholesterol secretion in Caco-2 cells. J Nutr Biochem. 2010;21:1207–1213. 29. Mah E, Noh SK, Ballard KD, Park HJ, Volek JS, Bruno RS. Supplementation of a γ-tocopherol-rich mixture of tocopherols in healthy men protects against vascular endothelial dysfunction induced by postprandial hyperglycemia. J Nutr Biochem. 2013;24:196–203. 30. Perkins JM, Joy NG, Tate DB, Davis SN. Acute effects of hyperinsulinemia and hyperglycemia on vascular inflammatory biomarkers and endothelial function in overweight and obese humans. Am J Physiol Endocrinol Metab. 2015;309:E168-76. 31. Ward NC, Wu JHY, Clarke MW, Puddey IB, Burke V, Croft KD, Hodgson JM. The effect of vitamin E on blood pressure in individuals with type 2 diabetes: a randomized, double-blind, placebo-controlled trial. Lippincott Williams & Wilkins; 2007: 1-227-234. 32. Singh I, Turner AH, Sinclair AJ, Li D, Hawley JA. Effects of gamma-tocopherol supplementation on thrombotic risk factors. ; 2007: 1-422-428.

151

33. Jambazian P, Haddad E, Rajaram S, Tanzman J, Sabate J. Almonds in the diet simultaneously improve plasma α-tocopherol concentrations and reduce plasma lipids. J Am Diet Assoc. 2005;105:449–454. 34. Ros E. A Walnut Diet Improves Endothelial Function in Hypercholesterolemic Subjects: A Randomized Crossover Trial. Circulation. 2004;109:1609–1614. 35. Kay CD, Gebauer SK, West SG, Kris-Etherton PM. Pistachios Increase Serum Antioxidants and Lower Serum Oxidized-LDL in Hypercholesterolemic Adults. J Nutr. 2010;140:1093–1098. 36. Haddad EH, Gaban-Chong N, Oda K, Sabaté J. Effect of a walnut meal on postprandial oxidative stress and antioxidants in healthy individuals. . 37. Konturek P, Brzozowski T, Konturek S. Gut Clock: Implications of Circadian Rhythms in the Gastrointestinal Tract. J Physiol Pharmacol. 2011;62:139–159. 38. Reiter RJ, Manchester LC, Tan D-X. Melatonin in walnuts: Influence on levels of melatonin and total antioxidant capacity of blood. Nutr J. 2005;21:920–924. 39. Garcia CP, Lamarque AL, Comba A, Berra A, Silva RA, Labuckas DO, Das UN, Eynard AR, Pasqualini ME. Basic nutritional investigation Synergistic anti- tumor effects of melatonin and PUFAs from walnuts in a murine mammary adenocarcinoma model. Nutr J. 2015;31:570–577. 40. Oladi E, Mohamadi M, Shamspur T, Mostafavi A. Spectrofluorimetric determination of melatonin in kernels of four different Pistacia varieties after ultrasound-assisted solid-liquid extraction. Spectrochim Acta A Mol Biomol Spectrosc. 2014;132:326–329. 41. Hornedo-Ortega R, Cerezo AB, Troncoso AM, Garcia-Parrilla MC, Mas A. Melatonin and Other Tryptophan Metabolites Produced by Yeasts: Implications in Cardiovascular and Neurodegenerative Diseases. Front Microbiol. 2015;6:1565. 42. Szewczyk-Golec K, Woźniak A, Reiter RJ. Inter-relationships of the chronobiotic, melatonin, with leptin and adiponectin: implications for obesity. J Pineal Res. 2015;59:277–91. 43. Emet M, Ozcan H, Ozel L, Yayla M, Halici Z, Hacimuftuoglu A. A Review of Melatonin, Its Receptors and Drugs. Eurasian J Med. 2016;48:135–41. 44. Auld F, Maschauer EL, Morrison I, Skene DJ, Riha RL. Evidence for the efficacy of melatonin in the treatment of primary adult sleep disorders. Sleep Med Rev. 2017;34:10–22. 45. Amstrup AK, Sikjaer T, Pedersen SB, Heickendorff L, Mosekilde L, Rejnmark L. Reduced fat mass and increased lean mass in response to 1 year of

152

melatonin treatment in postmenopausal women: A randomized placebo- controlled trial. Clin Endocrinol. 2016;84:342–347. 46. Cipolla-Neto J, Amaral FG, Afeche SC, Tan DX, Reiter RJ. Melatonin, energy metabolism, and obesity: a review. J Pineal Res. 2014;56:371–381. 47. Pechanova O, Paulis L, Simko F. Peripheral and Central Effects of Melatonin on Blood Pressure Regulation. Int J Mol Sci. 2014;15:17920–17937. 48. Grossman E, Laudon M, Yalcin R, Zengil H, Peleg E, Sharabi Y, Kamari Y, Shen-Orr Z, Zisapel N. Melatonin Reduces Night Blood Pressure in Patients with Nocturnal Hypertension. Am J Med. 2006;119:898–902. 49. Pechanova O, Paulis L, Simko F, Pechanova O, Paulis L, Simko F. Peripheral and Central Effects of Melatonin on Blood Pressure Regulation. Int J Mol Sci. 2014;15:17920–17937. 50. Turner JR, Viera AJ, Shimbo D. Ambulatory blood pressure monitoring in clinical practice: a review. Am J Med. 2015;128:14–20. 51. Iriti M, Varoni EM, Vitalini S. Melatonin in traditional Mediterranean diets. J Pineal Res. 2010;49:101–105. 52. Fraser GE, Sabate J, Beeson WL, Strahan TM. A Possible Protective Effect of Nut Consumption on Risk of Coronary Heart Disease. Arch Intern Med. 1992;152:1416. 53. Fraser GE, Lindsted KD, Beeson WL. Effect of risk factor values on lifetime risk of and age at first coronary event. The Adventist Health Study. Am J Epidemiol. 1995;142:746–58. 54. Hu FB, Stampfer MJ, Manson JE, Rimm EB, Colditz GA, Rosner BA, Speizer FE, Hennekens CH, Willett WC. Frequent nut consumption and risk of coronary heart disease in women: prospective cohort study. BMJ. 1998;317:1341–1345. 55. Albert CM, Gaziano JM, Willett WC, Manson JE. Nut Consumption and Decreased Risk of Sudden Cardiac Death in the Physicians’ Health Study. Arch Intern Med. 2002;162:1382. 56. Ellsworth JL, Kushi LH, Folsom AR. Frequent nut intake and risk of death from coronary heart disease and all causes in postmenopausal women: the Iowa Women’s Health Study. Nutr Metab Cardiovasc Dis. 2001;11:372–7. 57. Kris-Etherton PM, Hu FB, Ros E, Sabaté J. The Role of Tree Nuts and Peanuts in the Prevention of Coronary Heart Disease: Multiple Potential Mechanisms. J Nutr. 2008;138:1746S–1751S. 58. Kelly JH, Sabaté J. Nuts and coronary heart disease: an epidemiological 153

perspective. Br J Nutr. 2006;96 Suppl 2:S61-7. 59. Relja A, Miljković A, Gelemanović A, Bošković M, Hayward C, Polašek O, Kolčić I. Nut Consumption and Cardiovascular Risk Factors: A Cross-Sectional Study in a Mediterranean Population. Nutrients. 2017;9. doi:10.3390/nu9121296. 60. Eslamparast T, Sharafkhah M, Poustchi H, Hashemian M, Dawsey SM, Freedman ND, Boffetta P, Abnet CC, Etemadi A, Pourshams A, Malekshah AF, Islami F, Kamangar F, Merat S, Brennan P, Hekmatdoost A, Malekzadeh R. Nut consumption and total and cause-specific mortality: results from the Golestan Cohort Study. Int J Epidemiol. 2017;46:75–85. 61. Chen G-C, Zhang R, Martínez-González MA, Zhang Z-L, Bonaccio M, van Dam RM, Qin L-Q. Nut consumption in relation to all-cause and cause-specific mortality: a meta-analysis 18 prospective studies. Food Funct. 2017;8:3893– 3905. 62. Guasch-Ferré M, Liu X, Malik VS, Sun Q, Willett WC, Manson JE, Rexrode KM, Li Y, Hu FB, Bhupathiraju SN. Nut Consumption and Risk of Cardiovascular Disease. J Am Coll Cardiol. 2017;70:2519–2532. 63. Clarke R, Frost C, Collins R, Appleby P, Peto R. Dietary lipids and blood cholesterol: quantitative meta-analysis of metabolic ward studies. BMJ. 1997;314:112–7. 64. Hooper L, Martin N, Abdelhamid A, Davey Smith G. Reduction in saturated fat intake for cardiovascular disease. Cochrane Database Syst Rev. 2015;:CD011737. 65. Gardner CD, Kraemer HC. Monounsaturated versus polyunsaturated dietary fat and serum lipids. A meta-analysis. Arterioscler Thromb Vasc Biol. 1995;15:1917–27. 66. Jackson RL, Kashyap ML, Barnhart RL, Allen C, Hogg E, Glueck CJ. Influence of polyunsaturated and saturated fats on plasma lipids and lipoproteins in man. Am J Clin Nutr. 1984;39:589–597. 67. Sabate J, Fraser GE, Burke K, Knutsen SF, Bennett H, Lindsted KD. Effects of Walnuts on Serum Lipid Levels and Blood Pressure in Normal Men. N Engl J Med. 1993;328:603–607. 68. Banel DK, Hu FB. Effects of walnut consumption on blood lipids and other cardiovascular risk factors: a meta-analysis and systematic review. Am J Clin Nutr. 2009;90:56–63. 69. Guasch-Ferré M, Li J, Hu FB, Salas-Salvadó J, Tobias DK. Effects of walnut consumption on blood lipids and other cardiovascular risk factors: an updated 154

meta-analysis and systematic review of controlled trials. Am J Clin Nutr. 2018;108:174–187. 70. Zhang J, Grieger JA, Kris-Etherton PM, Thompson JT, Gillies PJ, Fleming JA, Vanden Heuvel JP. Walnut oil increases cholesterol efflux through inhibition of stearoyl CoA desaturase 1 in THP-1 macrophage-derived foam cells. Nutr Metab (Lond). 2011;8:61. 71. Fukumitsu S, Villareal MO, Onaga S, Aida K, Han J, Isoda H. α-Linolenic acid suppresses cholesterol and triacylglycerol biosynthesis pathway by suppressing SREBP-2, SREBP-1a and -1c expression. Cytotechnology. 2013;65:899–907. 72. Huang Y, Cai X, Li Y, Su L, Mai W, Wang S, Hu Y, Wu Y, Xu D. Prehypertension and the risk of stroke: a meta-analysis. Neurology. 2014;82:1153–61. 73. Mohammadifard N, Salehi-Abargouei A, Salas-Salvadó J, Guasch-Ferré M, Humphries K, Sarrafzadegan N. The effect of tree nut, peanut, and soy nut consumption on blood pressure: a systematic review and meta-analysis of randomized controlled clinical trials. Am J Clin Nutr. 2015;101:966–982. 74. Din JN, Aftab SM, Jubb AW, Carnegy FH, Lyall K, Sarma J, Newby DE, Flapan AD. Effect of moderate walnut consumption on lipid profile, arterial stiffness and platelet activation in humans. Eur J Clin Nutr. 2011;65:234–9. 75. Roman MJ, Devereux RB, Kizer JR, Lee ET, Galloway JM, Ali T, Umans JG, Howard B V. Central Pressure More Strongly Relates to Vascular Disease and Outcome Than Does Brachial Pressure: The Strong Heart Study. Hypertension. 2007;50:197–203. 76. Kollias A, Lagou S, Zeniodi ME, Boubouchairopoulou N, Stergiou GS. Association of Central Versus Brachial Blood Pressure With Target-Organ Damage: Systematic Review and Meta-Analysis. Hypertension. 2016;67:183– 90. 77. Ding F-H, Fan W-X, Zhang R-Y, Zhang Q, Li Y, Wang J-G. Validation of the Noninvasive Assessment of Central Blood Pressure by the SphygmoCor and Omron Devices Against the Invasive Catheter Measurement. Am J Hypertens. 2011;24:1306–1311. 78. Weber T, Wassertheurer S, Hametner B, Parragh S, Eber B. Noninvasive methods to assess pulse wave velocity: comparison with the invasive gold standard and relationship with organ damage. 2015. doi:10.1097/HJH.0000000000000518. 79. Inaba Y, Chen JA, Bergmann SR. Prediction of future cardiovascular outcomes by flow-mediated vasodilatation of brachial artery: a meta-analysis. 155

Int J Cardiovasc Imaging. 2010;26:631–640. 80. Xiao Y, Huang W, Peng C, Zhang J, Wong C, Kim JH, Yeoh E, Su X. Effect of nut consumption on vascular endothelial function: A systematic review and meta-analysis of randomized controlled trials. Clin Nutr. 2018;37:831–839. 81. West SG, Krick AL, Klein LC, Zhao G, Wojtowicz TF, McGuiness M, Bagshaw DM, Wagner P, Ceballos RM, Holub BJ, Kris-Etherton PM. Effects of diets high in walnuts and flax oil on hemodynamic responses to stress and vascular endothelial function. J Am Coll Nutr. 2010;29:595–603. 82. Ma Y, Njike VY, Millet J, Dutta S, Doughty K, Treu JA, Katz DL. Effects of walnut consumption on endothelial function in type 2 diabetic subjects: a randomized controlled crossover trial. Diabetes Care. 2010;33:227–32. 83. Njike VY, Ayettey R, Petraro P, Treu JA, Katz DL. Walnut ingestion in adults at risk for diabetes: effects on body composition, diet quality, and cardiac risk measures. BMJ open diabetes Res care. 2015;3:e000115. 84. West SG, Krick AL, Klein LC, Zhao G, Wojtowicz TF, Mcguiness M, Bagshaw DM, Wagner P, Ceballos RM, Holub BJ, Kris-Etherton PM, Sciences N. Effects of Diets High in Walnuts and Flax Oil on Hemodynamic Responses to Stress and Vascular Endothelial Function. . 85. Allen LH. Priority Areas for Research on the Intake, Composition, and Health Effects of Tree Nuts and Peanuts. J Nutr. 2008;138:1763S–1765S. 86. Kass L, Weekes J, Carpenter L. Effect of magnesium supplementation on blood pressure: a meta-analysis. Eur J Clin Nutr. 2012;66:411–418. 87. Rock CL, Flatt SW, Barkai H-S, Pakiz B, Heath DD. Walnut consumption in a weight reduction intervention: effects on body weight, biological measures, blood pressure and satiety. Nutr J. 2017;16:76. 88. Cani PD, Delzenne NM. Interplay between obesity and associated metabolic disorders: new insights into the gut microbiota. Curr Opin Pharmacol. 2009;9:737–43. 89. Lamendella R, VerBerkmoes N, Jansson JK. “Omics” of the mammalian gut-- new insights into function. Curr Opin Biotechnol. 2012;23:491–500. 90. Lozupone C, Stombaugh J, Gordon J. Diversity, stability and resilience of the human gut microbiota. Nature. 2012;489:220–230. 91. Rawls JF. Special issue: gut microbial communities in health and disease. Gut Microbes. 2012;3:277–8. 92. Kinross JM, Darzi AW, Nicholson JK. Gut microbiome-host interactions in health and disease. Genome Med. 2011;3:14. 156

93. Wang Z, Klipfell E, Bennett BJ, Koeth R, Levison BS, Dugar B, Feldstein AE, Britt EB, Fu X, Chung Y-M, Wu Y, Schauer P, Smith JD, Allayee H, Tang WHW, DiDonato JA, Lusis AJ, Hazen SL. Gut flora metabolism of phosphatidylcholine promotes cardiovascular disease. Nature. 2011;472:57– 63. 94. Bergeron N, Williams PT, Lamendella R, Faghihnia N, Grube A, Li X, Wang Z, Knight R, Jansson JK, Hazen SL, Krauss RM. Diets high in resistant starch increase plasma levels of trimethylamine-N-oxide, a gut microbiome metabolite associated with CVD risk. Br J Nutr. 2016;116:2020–2029. 95. Singh RK, Chang H-W, Yan D, Lee KM, Ucmak D, Wong K, Abrouk M, Farahnik B, Nakamura M, Zhu TH, Bhutani T, Liao W. Influence of diet on the gut microbiome and implications for human health. J Transl Med. 2017;15:73. 96. Desai MS, Seekatz AM, Koropatkin NM, Stappenbeck TS, Nú Ñ Ez G, Martens EC, Kamada N, Hickey CA, Wolter M, Pudlo NA, Kitamoto S, Terrapon N, Muller A, Young VB, Henrissat B, Wilmes P. A Dietary Fiber- Deprived Gut Microbiota Degrades the Colonic Mucus Barrier and Enhances Pathogen Susceptibility. Cell. 2016;167:1339–1353. 97. Haro C, Montes-Borrego M, Rangel-Zúñiga OA, Alcalá-Díaz JF, Gómez- Delgado F, Pérez-Martínez P, Delgado-Lista J, Quintana-Navarro GM, Tinahones FJ, Landa BB, López-Miranda J, Camargo A, Pérez-Jiménez F. Two Healthy Diets Modulate Gut Microbial Community Improving Insulin Sensitivity in a Human Obese Population. J Clin Endocrinol Metab. 2016;101:233–242. 98. Fava F, Gitau R, Griffin BA, Gibson GR, Tuohy KM, Lovegrove JA. The type and quantity of dietary fat and carbohydrate alter faecal microbiome and short- chain fatty acid excretion in a metabolic syndrome ‘at-risk’ population. Int J Obes. 2013;37:216–223. 99. De Filippis F, Pellegrini N, Vannini L, Jeffery IB, La Storia A, Laghi L, Serrazanetti DI, Di Cagno R, Ferrocino I, Lazzi C, Turroni S, Cocolin L, Brigidi P, Neviani E, Gobbetti M, O’Toole PW, Ercolini D. High-level adherence to a Mediterranean diet beneficially impacts the gut microbiota and associated metabolome. Gut. 2016;65:1812–1821. 100. David LA, Maurice CF, Carmody RN, Gootenberg DB, Button JE, Wolfe BE, Ling A V, Devlin AS, Varma Y, Fischbach MA, Biddinger SB, Dutton RJ, Turnbaugh PJ. Diet rapidly and reproducibly alters the human gut microbiome. Nature. 2014;505:559–63. 101. Guetterman HM, Swanson KS, Novotny JA, Baer DJ, Holscher HD. Walnut Consumption Influences the Human Gut Microbiome. FASEB J. 2016;30:406.2-.

157

102. Bamberger C, Rossmeier A, Lechner K, Wu L, Waldmann E, Fischer S, Stark RG, Altenhofer J, Henze K, Parhofer KG. A Walnut-Enriched Diet Affects Gut Microbiome in Healthy Caucasian Subjects: A Randomized, Controlled Trial. Nutrients. 2018;10:244. 103. Tindall AM, Petersen KS, Kris-Etherton PM. Dietary Patterns Affect the Gut Microbiome—The Link to Risk of Cardiometabolic Diseases. J Nutr. 2018;148:1402–1407. 104. Kemperman RA, Bolca S, Roger LC, Vaughan EE. Novel approaches for analysing gut microbes and dietary polyphenols: challenges and opportunities. Microbiology. 2010;156:3224–31. 105. Rastmanesh R. High polyphenol, low probiotic diet for weight loss because of intestinal microbiota interaction. Chem Biol Interact. 2011;189:1–8. 106. Benjamin EJ, Virani SS, Callaway CW, Chamberlain AM, Chang AR, Cheng S, Chiuve SE, Cushman M, Delling FN, Deo R, de Ferranti SD, Ferguson JF, Fornage M, Gillespie C, Isasi CR, Jiménez MC, Jordan LC, Judd SE, Lackland D, Lichtman JH, Lisabeth L, Liu S, Longenecker CT, Lutsey PL, Mackey JS, Matchar DB, Matsushita K, Mussolino ME, Nasir K, O’Flaherty M, Palaniappan LP, Pandey A, Pandey DK, Reeves MJ, Ritchey MD, Rodriguez CJ, Roth GA, Rosamond WD, Sampson UKA, Satou GM, Shah SH, Spartano NL, Tirschwell DL, Tsao CW, Voeks JH, Willey JZ, Wilkins JT, Wu JH, Alger HM, Wong SS, Muntner P, American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2018 Update: A Report From the American Heart Association. Circulation. 2018;137:e67–e492. 107. Blood Pressure Lowering Treatment Trialists’ Collaboration BPLTT, Ninomiya T, Perkovic V, Turnbull F, Neal B, Barzi F, Cass A, Baigent C, Chalmers J, Li N, Woodward M, MacMahon S. Blood pressure lowering and major cardiovascular events in people with and without chronic kidney disease: meta-analysis of randomised controlled trials. BMJ. 2013;347:f5680. 108. Whelton PK, Carey RM, Aronow WS, Casey DE, Collins KJ, Dennison Himmelfarb C, DePalma SM, Gidding S, Jamerson KA, Jones DW, MacLaughlin EJ, Muntner P, Ovbiagele B, Smith SC, Spencer CC, Stafford RS, Taler SJ, Thomas RJ, Williams KA, Williamson JD, Wright JT. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. 2018;71:e127–e248. 109. Orsavova J, Misurcova L, Ambrozova JV, Vicha R, Mlcek J. Fatty Acids Composition of Vegetable Oils and Its Contribution to Dietary Energy Intake 158

and Dependence of Cardiovascular Mortality on Dietary Intake of Fatty Acids. Int J Mol Sci. 2015;16:12871–90. 110. Souza RGM, Gomes AC, Naves MM V., Mota JF. Nuts and legume seeds for cardiovascular risk reduction: scientific evidence and mechanisms of action. Nutr Rev. 2015;73:335–347. 111. Tsukamoto I, Sugawara S. Low levels of linoleic acid and α-linolenic acid and high levels of arachidonic acid in plasma phospholipids are associated with hypertension. Biomed Rep. 2018;8:69–76. 112. Bemelmans WJ, Muskiet FA, Feskens EJ, de Vries JH, Broer J, May JF, Jong BM. Associations of alpha-linolenic acid and linoleic acid with risk factors for coronary heart disease. Eur J Clin Nutr. 2000;54:865–71. 113. Ueshima H, Stamler J, Elliott P, Chan Q, Brown IJ, Carnethon MR, Daviglus ML, He K, Moag-Stahlberg A, Rodriguez BL, Steffen LM, Van Horn L, Yarnell J, Zhou B. Food Omega-3 Fatty Acid Intake of Individuals (Total, Linolenic Acid, Long-Chain) and Their Blood Pressure: INTERMAP Study. Hypertension. 2007;50:313–319. 114. Jacobson TA, Maki KC, Orringer CE, Jones PH, Kris-Etherton P, Sikand G, La Forge R, Daniels SR, Wilson DP, Morris PB, Wild RA, Grundy SM, Daviglus M, Ferdinand KC, Vijayaraghavan K, Deedwania PC, Aberg JA, Liao KP, McKenney JM, Ross JL, Braun LT, Ito MK, Bays HE, Brown WV. National Lipid Association Recommendations for Patient-Centered Management of Dyslipidemia: Part 2. J Clin Lipidol. 2015;9:S1–S122.e1. 115. Miller M, Sorkin JD, Mastella L, Sutherland A, Rhyne J, Donnelly P, Simpson K, Goldberg AP. Poly is more effective than monounsaturated fat for dietary management in the metabolic syndrome: The muffin study. J Clin Lipidol. 2016;10:996–1003. 116. Lichtenstein AH, Appel LJ, Brands M, Carnethon M, Daniels S, Franch HA, Franklin B, Kris-Etherton P, Harris WS, Howard B, Karanja N, Lefevre M, Rudel L, Sacks F, Van Horn L, Winston M, Wylie-Rosett J, Wylie-Rosett J. Diet and Lifestyle Recommendations Revision 2006. Circulation. 2006;114:82–96. 117. Katz DL, Davidhi A, Ma Y, Kavak Y, Bifulco L, Njike VY. Effects of walnuts on endothelial function in overweight adults with visceral obesity: a randomized, controlled, crossover trial. J Am Coll Nutr. 2012;31:415–23. 118. National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report. 159

Circulation. 2002;106:3143–421. 119. Harris JA, Benedict FG. A Biometric Study of Human Basal Metabolism. Proc Natl Acad Sci U S A. 1918;4:370–3. 120. Shoji T, Nakagomi A, Okada S, Ohno Y, Kobayashi Y. Invasive validation of a novel brachial cuff-based oscillometric device (SphygmoCor XCEL) for measuring central blood pressure. doi:10.1097/HJH.0000000000001135. 121. Seventh Report of the Joint National Committee on the Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. ; 2003. 122. Laurent S, Cockcroft J, Van Bortel L, Boutouyrie P, Giannattasio C, Hayoz D, Pannier B, Vlachopoulos C, Wilkinson I, Struijker-Boudier H. Expert consensus document on arterial stiffness: methodological issues and clinical applications. Eur Heart J. 2006;27:2588–2605. 123. Friedewald WT, Levy RI, Fredrickson DS. Estimationof the Concentration of Low-Density LipoproteinCholesterol in Plasma,Without Useof the PreparativeUltracentrifuge. Clin Chem. 1972;18:499. 124. Wu H, Pan A, Yu Z, Qi Q, Lu L, Zhang G, Yu D, Zong G, Zhou Y, Chen X, Tang L, Feng Y, Zhou H, Chen X, Li H, Demark-Wahnefried W, Hu FB, Lin X. Lifestyle counseling and supplementation with flaxseed or walnuts influence the management of metabolic syndrome. J Nutr. 2010;140:1937–42. 125. Raatz SK, Conrad Z, Jahns L, Belury MA, Picklo MJ. Modeled replacement of traditional soybean and canola oil with high-oleic varieties increases monounsaturated fatty acid and reduces both saturated fatty acid and polyunsaturated fatty acid intake in the US adult population. Am J Clin Nutr. 2018. doi:10.1093/ajcn/nqy127. 126. United States Department of Agriculture. Basic Report: 12155, Nuts, walnuts, english. National Nutrient Database for Standard Reference Release 28. Available at https://ndb.nal.usda.gov/ndb/foods/show/3690. Accessed July 20, 2017. 127. Hayes D, Angove MJ, Tucci J, Dennis C. Walnuts (Juglans regia) Chemical Composition and Research in Human Health Walnuts (Juglans regia) Chemical Composition and Research in Human Health. Crit Rev Food Sci Nutr. 2016;568. doi:10.1080/10408398.2012.760516. 128. Alasalvar C, Bolling BW. Review of nut phytochemicals, fat-soluble bioactives, antioxidant components and health effects. doi:10.1017/S0007114514003729. 129. Huang W-Y, Davidge ST, Wu J. Bioactive Natural Constituents from Food Sources—Potential Use in Hypertension Prevention and Treatment. Crit Rev Food Sci Nutr. 2013;53:615–630. 160

130. Caligiuri SPB, Rodriguez-Leyva D, Aukema HM, Ravandi A, Weighell W, Guzman R, Pierce GN. Dietary Flaxseed Reduces Central Aortic Blood Pressure Without Cardiac Involvement but Through Changes in Plasma Oxylipins. Hypertension. 2016;68:1031–1038. 131. Wang J-Z, Du W-T, Xu Y-L, Cheng S-Z, Liu Z-J. Gut microbiome-based medical methodologies for early-stage disease prevention. Microb Pathog. 2017;105:122–130. 132. Yan Q, Gu Y, Li X, Yang W, Jia L, Chen C, Han X, Huang Y, Zhao L, Li P, Fang Z, Zhou J, Guan X, Ding Y, Wang S, Khan M, Xin Y, Li S, Ma Y. Alterations of the Gut Microbiome in Hypertension. Front Cell Infect Microbiol. 2017;7:381. 133. Baer DJ, Gebauer SK, Novotny JA. Walnuts Consumed by Healthy Adults Provide Less Available Energy than Predicted by the Atwater Factors. J Nutr. 2016;146:9–13. 134. Holscher H, Guetterman H, Swanson K, An R, Matthan N, Lichtenstein A, Novotny J, Baer D. Walnut consumption alters the gastrointestinal microbiota , microbial-derived secondary bile acids , and health markers in healthy adults : a randomized controlled trial. J Nutr. 2018;148:1–7. 135. Gomez-Arango LF, Barrett HL, McIntyre HD, Callaway LK, Morrison M, Dekker Nitert M. Increased Systolic and Diastolic Blood Pressure Is Associated With Altered Gut Microbiota Composition and Butyrate Production in Early PregnancyNovelty and Significance. Hypertension. 2016;68:974–981. 136. Roshanravan N, Mahdavi R, Alizadeh E, Ghavami A, Saadat YR, Alamdari NM, Alipour S, Dastouri MR, Ostadrahimi A. The effects of sodium butyrate and inulin supplementation on angiotensin signaling pathway via promotion of Akkermansia muciniphila abundance in type 2 diabetes; A randomized, double-blind, placebo-controlled trial. J Cardiovasc Thorac Res. 2017;9:183– 190. 137. Katan MB, Tilburg TV LM. Katan Calculator: Predicted effect of diet on blood lipids and lipoproteins. Available at http://www.katancalculator.nl/. Accessed July 26, 2018. 138. Griel AE, Kris-Etherton PM. Tree nuts and the lipid profile: a review of clinical studies. Br J Nutr. 2006;96 Suppl 2:S68-78. 139. Ros E, Núñez I, Pérez-Heras A, Serra M, Gilabert R, Casals E, Deulofeu R. A Walnut Diet Improves Endothelial Function in Hypercholesterolemic Subjects. Circulation. 2004;109:1609–1614. 140. Matone A, O’Grada CM, Dillon ET, Morris C, Ryan MF, Walsh M, Gibney ER, Brennan L, Gibney MJ, Morine MJ, Roche HM. Body mass index mediates 161

inflammatory response to acute dietary challenges. Mol Nutr Food Res. 2015;59:2279–2292. 141. Ottaviani S, Gardette A, Tubach F, Roy C, Palazzo E, Gill G, Meyer O, Dieudé P. Body mass index and response to infliximab in rheumatoid arthritis. Clin Exp Rheumatol.;33:478–83. 142. Lee M-R, Lim Y-H, Hong Y-C. Causal association of body mass index with hypertension using a Mendelian randomization design. Medicine (Baltimore). 2018;97:e11252. 143. Kotsis V, Nilsson P, Grassi G, Mancia G, Redon J, Luft F, Schmieder R, Engeli S, Stabouli S, Antza C, Pall D, Schlaich M, Jordan J, WG on Obesity, Diabetes, the High Risk Patient, European Society of Hypertension. New developments in the pathogenesis of obesity-induced hypertension. J Hypertens. 2015;33:1499–508. 144. US Department of Agriculture and US Department of Health and Human Services. Scientific Report of the 2015 Dietary Guidelines Advisory Committee. Washington (DC); 2015. 145. Li J, Zhao F, Wang Y, Chen J, Tao J, Tian G, Wu S, Liu W, Cui Q, Geng B, Zhang W, Weldon R, Auguste K, Yang L, Liu X, Chen L, Yang X, Zhu B, Cai J. Gut microbiota dysbiosis contributes to the development of hypertension. Microbiome. 2017;5:14. 146. Brown JM, Hazen SL. Microbial modulation of cardiovascular disease. Nat Rev Microbiol. 2018;16:171–181. 147. Kelly TN, Bazzano LA, Ajami NJ, He H, Zhao J, Petrosino JF, Correa A, He J. Gut Microbiome Associates With Lifetime Cardiovascular Disease Risk Profile Among Bogalusa Heart Study Participants. Circ Res. 2016;119:956–64. 148. Koeth RA, Wang Z, Levison BS, Buffa JA, Org E, Sheehy BT, Britt EB, Fu X, Wu Y, Li L, Smith JD, DiDonato JA, Chen J, Li H, Wu GD, Lewis JD, Warrier M, Brown JM, Krauss RM, Tang WHW, Bushman FD, Lusis AJ, Hazen SL. Intestinal microbiota metabolism of L-carnitine, a nutrient in red meat, promotes atherosclerosis. Nat Med. 2013;19:576–85. 149. Rajaram S. Health benefits of plant-derived α-linolenic acid. Am J Clin Nutr. 2014;100:443S–448S. 150. Selma M V., González-Sarrías A, Salas-Salvadó J, Andrés-Lacueva C, Alasalvar C, Örem A, Tomás-Barberán FA, Espín JC. The gut microbiota metabolism of pomegranate or walnut ellagitannins yields two urolithin- metabotypes that correlate with cardiometabolic risk biomarkers: Comparison between normoweight, overweight-obesity and metabolic syndrome. Clin Nutr. 2018;37:897–905. 162

151. Tindall AM, Petersen KS, Skulas-Ray A, Richter C, Proctor DN, Kris-Etherton PM. Aging and Chronic Disease. Curr Dev Nutr. 2018;2. doi:10.1093/cdn/nzy028. 152. Walters W, Hyde ER, Berg-Lyons D, Ackermann G, Humphrey G, Parada A, Gilbert JA, Jansson JK, Caporaso JG, Fuhrman JA, Apprill A, Knight R. Improved Bacterial 16S rRNA Gene (V4 and V4-5) and Fungal Internal Transcribed Spacer Marker Gene Primers for Microbial Community Surveys. mSystems. 2015;1. 153. Caporaso J, Lauber C. Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample. Proc Natl Acad Sci U S A. 2011;108:4516– 4522. 154. Callahan B, McMurdie P, Rose M, Han A, Johnson AJ, Holmes S. DADA2: High resolution sample inference from Illumina amplicon data. Nat Methods. 2016;13:581–583. 155. Selma M V, Beltrán D, Luna MC, Romo-Vaquero M, García-Villalba R, Mira A, Espín JC, Tomás-Barberán FA. Isolation of Human Intestinal Bacteria Capable of Producing the Bioactive Metabolite Isourolithin A from Ellagic Acid. Front Microbiol. 2017;8:1521. 156. Segata N, Izard J, Waldron L, Gevers D, Miropolsky L, Garrett WS, Huttenhower C. Metagenomic biomarker discovery and explanation. Genome Biol. 2011;12:R60. 157. Faust K, Raes J. CoNet app: inference of biological association networks using Cytoscape. F1000Research. 2016;5:1519. 158. Morgan XC, Tickle TL, Sokol H, Gevers D, Devaney KL, Ward D V, Reyes JA, Shah SA, LeLeiko N, Snapper SB, Bousvaros A, Korzenik J, Sands BE, Xavier RJ, Huttenhower C. Dysfunction of the intestinal microbiome in inflammatory bowel disease and treatment. Genome Biol. 2012;13:R79. 159. Damms-Machado A, Louis S, Schnitzer A, Volynets V, Rings A, Basrai M, Bischoff SC. Gut permeability is related to body weight, fatty liver disease, and insulin resistance in obese individuals undergoing weight reduction. Am J Clin Nutr. 2017;105:127–135. 160. Soderborg TK, Friedman JE. Imbalance in gut microbes from babies born to obese mothers increases gut permeability and myeloid cell adaptations that provoke obesity and NAFLD. Microb Cell. 2019;6:102–104. 161. Scheppach W, Müller JG, Boxberger F, Dusel G, Richter F, Bartram HP, Christl SU, Dempfle CE, Kasper H. Histological changes in the colonic mucosa following irrigation with short-chain fatty acids. Eur J Gastroenterol Hepatol. 1997;9:163–8. 163

162. Homann H-H, Kemen M, Fuessenich C, Senkal M, Zumtobel V. Reduction in Diarrhea Incidence by Soluble Fiber in Patients Receiving Total or Supplemental Enteral Nutrition. J Parenter Enter Nutr. 1994;18:486–490. 163. Butzner JD, Parmar R, Bell CJ, Dalal V. Butyrate enema therapy stimulates mucosal repair in experimental colitis in the rat. Gut. 1996;38:568–73. 164. Yan H, Ajuwon KM. Butyrate modifies intestinal barrier function in IPEC-J2 cells through a selective upregulation of tight junction proteins and activation of the Akt signaling pathway. PLoS One. 2017;12:e0179586. 165. Meehan CJ, Beiko RG. A phylogenomic view of ecological specialization in the Lachnospiraceae, a family of digestive tract-associated bacteria. Genome Biol Evol. 2014;6:703–13. 166. Nieminen TT, Säde E, Endo A, Johansson P, Björkroth J. The Family Leuconostocaceae. In: Rosenberg E, DeLong EF, Lory S, Stackebrandt E, Thompson F, eds. The Prokaryotes: Firmicutes and Tenericutes. Berlin, Heidelberg: Springer Berlin Heidelberg; 2014: 215–240. 167. Mörkl S, Lackner S, Meinitzer A, Mangge H, Lehofer M, Halwachs B, Gorkiewicz G, Kashofer K, Painold A, Holl AK, Bengesser SA, Müller W, Holzer P, Holasek SJ. Gut microbiota, dietary intakes and intestinal permeability reflected by serum zonulin in women. Eur J Nutr. 2018;57:2985– 2997. 168. Kang C, Wang B, Kaliannan K, Wang X, Lang H, Hui S, Huang L, Zhang Y, Zhou M, Chen M, Mi M. Gut Microbiota Mediates the Protective Effects of Dietary Capsaicin against Chronic Low-Grade Inflammation and Associated Obesity Induced by High-Fat Diet. MBio. 2017;8:1–15. 169. Leclercq S, Matamoros S, Cani PD, Neyrinck AM, Jamar F, Stärkel P, Windey K, Tremaroli V, Bäckhed F, Verbeke K, de Timary P, Delzenne NM. Intestinal permeability, gut-bacterial dysbiosis, and behavioral markers of alcohol- dependence severity. Proc Natl Acad Sci. 2014;111:E4485–E4493. 170. Yu LC-H, Wang J-T, Wei S-C, Ni Y-H. Host-microbial interactions and regulation of intestinal epithelial barrier function: From physiology to pathology. World J Gastrointest Pathophysiol. 2012;3:27–43. 171. Baldassano SN, Bassett DS. Topological distortion and reorganized modular structure of gut microbial co-occurrence networks in inflammatory bowel disease. Sci Rep. 2016;6:26087. 172. Tulipani S, Urpi-Sarda M, García-Villalba R, Rabassa M, Loez-Uriarte P, Bullo N, Jaúregui O, Toma-Barbera F, Salas-Salvado J, Carlos Espín J, Andre- Lacueva C. Urolithins Are the Main Urinary Microbial-Derived Phenolic Metabolites Discriminating a Moderate Consumption of Nuts in Free- Living 164

Subjects with Diagnosed Metabolic Syndrome. J Agric Food Chem. 2012;60:8930–8940. 173. Furst T, Massaro A, Miller C, Williams BT, LaMacchia ZM, Horvath PJ. β- Alanine supplementation increased physical performance and improved executive function following endurance exercise in middle aged individuals. J Int Soc Sports Nutr. 2018;15:32. 174. Amadou T, Hosny M, La Scola B, Cassir N. Lachnoclostridium bouchesdurhonense; a new bacterial species isolated from human gut microbiota. New microbes new Infect. 2016;13:69–70.

165

VITA Alyssa Marie Tindall

EDUCATION The Pennsylvania State University, Dual-Title Ph.D., Nutritional Sciences and Clinical and Translational Science 2019 The Pennsylvania State University, Post-Baccalaureate Certificate, Dietetic Internship (PCT) 2014 The Pennsylvania State University, B.S., Nutritional Sciences 2013

PUBLICATIONS Tindall AM, Johnston EJ, Kris-Etherton PM, Petersen KS. The effect of nuts on markers of glycemic control: a systematic review and meta-analysis of randomized controlled trials. American Journal of Clinical Nutrition. 2019; 109(2):297-314.

Tindall AM, Petersen KS, Kris-Etherton PM. Dietary patterns affect the gut microbiome – the link to cardiometabolic diseases. Journal of Nutrition. 2018; 148(9):1402-1407.

Tindall AM, Petersen KS, Lamendella R, Shearer GC, Murray-Kolb LE, Proctor DN, Kris-Etherton PM. The microbiome as a potential mediator between tree nut consumption and adipose tissue mass. Current Developments in Nutrition. 2018; 2(11) nzy069.

AWARDS Nutritional Science Graduate Educational Enhancement Award, Penn State, 2019 Ruth L. Pike Nutritional Sciences Graduate Fellowship, Penn State, 2018 Golden Spoon Award for “Most likely to lend a hand”, Penn State, 2018 1st place in the 33rd Annual Graduate Exhibition (Health & Life Sciences), Penn State, 2018 John A. Milner Endowment Award, Penn State, 2017 Department of Kinesiology Travel Grant, Penn State, 2017 Pre-Doctoral Fellowship, Penn State Clinical and Translational Science Institute, 2016-2017 Excellence in Graduate Recruitment Award, Penn State 2015-2016