Avenanthramide Supplementation in Young and Older Women

AVENANTHRAMIDE SUPPLEMENTATION IN YOUNG AND OLDER WOMEN:

PROTECTION AGAINST ECCENTRIC EXERCISE-INDUCED INFLAMMATION

AND OXIDATIVE STRESS

RYAN THOMAS KOENIG

A dissertation submitted in partial fulfillment of

the requirements for the degree of

Doctor of Philosophy

(Kinesiology)

UNIVERSITY OF WISCONSIN – MADISON

2012

Date of final oral examination: 05/17/12

The dissertation is approved by the following members of the Final Oral Committee: Li Li Ji Gary Diffee William Schrage Kirk Parkin Mitchell Wise i

ACKNOWLEDGMENTS

In the summer of 2005, I was working on the grounds crew of the Saint Vincent

Cemetery in Latrobe, Pennsylvania. I had just graduated college and been rejected by a dozen or so medical schools. With little idea of what I was getting into, I started investigating graduate schools and came across the name Li Li Ji. We spoke on the phone long distance and found that our interests overlapped, and before long I was on a trip to Madison, Wisconsin—a place I assumed was locked in ice for at least ten months a year. I met Dr. Ji, toured the Ji Lab, landed an apartment, and discovered that Madison was not only ice-free in July but also a place I wanted to be. Just like that, I was on my way to starting the research that now culminates in the following pages.

I need to thank Dr. Ji for giving me this chance. He has provided ceaseless advice and guidance to go with the resources that made this work possible. For the final months of this journey, we have both been estranged from Madison, but he has still been as accessible as when I used to knock on his office door. I am so grateful for his help, especially during this time that I have been away from Madison when it would have been easy to stray from the path.

I also need to express my gratitude to Jon Dickman, who is going for the grand slam of degrees from the UW while at the same time remaining a close friend and incredible lab mate. If

Jon was my right hand in the Ji Lab, Choung-Hun Kang was my left, and I thank them both for the endless hours of assistance they provided.

Dr. Mitchell Wise introduced me to avenanthramide, and none of this would have been possible without his guidance. He also served on my thesis committee along with Drs. Diffee,

Schrage, and Parkin, whose insight and advice were critical in the work presented here. ii

Finally, without the love and support of my wife Natalie, I could never have made it this far. Without her as my foundation, I would have tumbled from these lofty heights I chose to climb. Without her arms to fall into at the end of the day, the long hours would never have paid off. And, of course, without her I wouldn’t have my two boys, Arthur and Felix, who make every single day a great one.

It’s been a long trip from the Saint Vincent Cemetery, and I got a lot of help from some very special people along the way.

iii

CONTENTS

ACKNOWLEDGMENTS ...... i

CONTENTS ...... iiiii

TABLES AND FIGURES ...... iix

ABBREVIATIONS ...... xi

INTRODUCTION ...... 1

PURPOSES AND HYPOTHESES...... 5

REVIEW OF THE LITERATURE ...... 6

1. Reactive Oxygen Species ...... 6

A. Mitochondrial source ...... 6

B. Antioxidant defense...... 6

C. Neutrophil source ...... 7

2. Inflammation ...... 7

A. Role in disease ...... 8

B. Cytokines ...... 8

C. Eccentric Contraction ...... 11

3. Aging...... 15

A. Role of ROS and inflammation ...... 15

B. Estrogens and Aging ...... 17

4. Avenanthramides ...... 18 iv

A. Introduction ...... 18

B. Antioxidant Capacity...... 19

C. Antioxidant Activity in Rats ...... 22

D. Bioavailability in Rodents ...... 23

F. Anti-Inflammatory Action ...... 26

G. Nitric Oxide and SMC Proliferation ...... 27

H. NFκB Signaling ...... 29

I. Colon Cancer Prevention ...... 30

STUDY 1: Effect of AVA Supplementation on Eccentric Exercise-induced Inflammation and

Oxidative Stress in Young Women...... 31

Abstract ...... 32

Introduction ...... 34

Methods...... 35

A. Subjects ...... 35

B. Study Visits ...... 36

C. Dietary Supplementation ...... 38

D. Downhill Running ...... 39

E. Blood sample collection and preparation ...... 39

F. Biological Measurements ...... 40

1. ELISA…………………………………………………………………………………40 v

a. Inflammatory markers…………………………………………………………40

b. NFκB………………………………………………………………………….41

2. HPLC………………………………………………………………………………….41

a. Plasma glutathione…………………………………………………………….41

b. Avenanthramide concentration………………………………………………..42

3. Spectrophotometric assays…………………………………………………………….43

a. Plasma TAC…………………………………………………………………...43

b. Plasma creatine kinase………………………………………………………...43

c. Erythrocyte superoxide dismutase…………………………………………….44

d. Erythrocyte glutathione peroxidase…………………………………………...44

e. Hemoglobin……………………………………………………………………45

4. Neutrophil respiratory burst…………………………………………………………...45

5. Pain and soreness ratings……………………………………………………………...46

G. Statistical Analysis ...... 46

Results ...... 49

A. Participant Data ...... 49

B. Muscle damage caused by (DR)...... 49

C. Inflammatory Markers...... 49

D. NFκB ...... 54

E. Plasma Total Antioxidant Capacity ...... 54

F. Erythrocyte Antioxidant Enzymes ...... 54

G. Glutathione Status ...... 61 vi

H. Pain and Soreness Ratings ...... 61

Discussion ...... 67

A. DR-Induced Muscle Damage ...... 67

B. Anti-Inflammatory Effects of AVA ...... 70

C. Effects of DR and AVA on Antioxidant Defense ...... 74

STUDY 2: Effect of AVA Supplementation on Eccentric Exercise-induced Inflammation and

Oxidative Stress in Postmenopausal Women ...... 79

Abstract ...... 80

Introduction ...... 82

Methods...... 85

A. Subjects ...... 85

B. Study Visits ...... 86

C. Dietary Supplementation ...... 87

D. Downhill Walking ...... 87

E. Blood sample collection and preparation ...... 89

F. Biological Measurements ...... 90

1. ELISA…………………………………………………………………………………90

a. Inflammatory markers…………………………………………………………90

b. NFκB………………………………………………………………………….90

2. HPLC………………………………………………………………………………….91

a. Plasma glutathione…………………………………………………………….91 vii

b. Avenanthramide concentration………………………………………………..91

3. Spectrophotometric assays…………………………………………………………….92

a. Plasma TAC…………………………………………………………………...92

b. Plasma creatine kinase………………………………………………………...93

c. Erythrocyte superoxide dismutase…………………………………………….93

d. Erythrocyte glutathione peroxidase…………………………………………...94

e. Hemoglobin……………………………………………………………………94

4. Neutrophil respiratory burst…………………………………………………………...94

5. Pain and soreness ratings……………………………………………………………...95

G. Statistical Analysis ...... 95

Results ...... 96

A. Participant Data ...... 96

B. Muscle damage caused by DW ...... 96

C. Inflammatory Markers...... 100

D. NFκB binding ...... 100

E. Plasma Total Antioxidant Capacity ...... 107

F. Erythrocyte Antioxidant Enzymes ...... 107

G. Glutathione Status ...... 107

H. Pain and Soreness Ratings ...... 107

Discussion ...... 114

A. DW-Induced Muscle Damage ...... 116 viii

B. Anti-Inflammatory Effects of AVA ...... 118

C. Antioxidant Effects of AVA Supplementation ...... 121

CONCLUSIONS OF BOTH STUDIES ...... 124

APPENDIX…………………………………………………………………………………….126

LITERATURE CITED ...... 133

TABLES AND FIGURES

Fig.1: AVA structures ...... 20

Fig. 2: Timeline of visits for young women ...... 37

Fig. 3: A priori comparisons planned for statistical analysis of Study One...... 48

Table 1: Characteristics of young participants...... 50

Fig. 4: Plasma CK activity in young women...... 51

Fig. 5: Neutrophil respiratory burst activity in young women...... 52

Fig. 6: Plasma IL-1β concentration in young women...... 53

Fig. 7: Plasma IL-6 concentration in young women...... 55

Fig. 8: Plasma TNF-α concentration in young women...... 56

Fig. 9: Plasma CRP concentration in young women...... 57

Fig. 10: Mononuclear cell NFκB binding activity in young women...... 58

Fig. 11: Plasma TAC in young women...... 59

Fig. 12: Erythrocyte SOD activity in young women...... 60

Fig. 13: Erythrocyte GPx activity in young women...... 62

Fig. 14: Plasma GSH concentration in young women...... 63

Fig. 15: Plasma GSSG concentration in young women...... 64

Fig. 16: Plasma GSH:GSSG ratio in young women...... 65

Table 2: Ratings of pain reported by young women...... 66

Table 3: Ratings of muscle soreness reported by young women ...... 66

Fig. 17: Timeline of visits for postmenopausal women ...... 88

Fig. 18: A priori comparisons planned for statistical analysis for Study Two...... 97

Table 4: Characteristics of older participants……………………………………………………98

Fig. 19: Plasma CK activity in postmenopausal women...... 99

Fig. 20: Neutrophil respiratory burst activity in postmenopausal women...... 101

Fig. 21: Plasma IL-β in postmenospausal women...... 102

Fig. 22: Plasma IL-6 concentration in postmenopausal women...... 103

Fig. 23: Plasma TNF-α concentration in postmenopausal women...... 104

Fig. 24: Plasma CRP concentration in postmenopausal women...... 105

Fig. 25: Mononuclear cell NFκB binding in postmenopausal women...... 106

Fig. 26: Plasma TAC in postmenopausal women...... 108

Fig. 27: Erythrocyte SOD activity in postmenopausal women...... 109

Fig. 28: Erythrocyte GPx activity in postmenopausal women...... 110

Fig. 29: Plasma GSH concentration in postmenopausal women...... 111

Fig. 30: Plasma GSSG concentration in postmenopausal women...... 112

Fig. 31: Plasma GSH:GSSG ratio in post-menopausal women...... 113

Table 5: Ratings of pain reported by postmenopausal women...... 115

Table 6: Ratings of muscle soreness reported by postmenopausal women...... 115

ABBREVIATIONS

ABTS 2,2’-Azinobis-3-ethylbenzothiazoline-6-sulfonic acid

ACN Acetonitrile

AEM Avenanthramide enriched mixture

AP-1 Activator protein-1

APS Aluminum potassium sulfate

AVA Avenanthramide

AVAO Avenanthramide enhance oat extract

BMI Body mass index

CDK Cyclin-dependent kinase

CH3-AVA-C Methylated avenanthramide-C

CK Creatine kinase

COX Cyclooxygenase

CRP C-reactive protein

DCFH 2’,7’-Dichlorofluorescein

DMSO Dimethyl sulfoxide

DOMS Delayed onset muscle soreness

DPPH 2,2-Diphenyl-1-picrylhydrazyl

DVL Deep vastus lateralis

ELISA Enzyme-linked immunosorbent assay eNOS Endothelial nitric oxide synthase

ETC Electron transport chain

FBS Fetal bovine serum xii

GCL Glutamate-cysteine ligase

GPx Glutathione peroxidase

GR Glutathione reductase

GSH Glutathione (reduced form)

GSSG Glutathione disulfide

H2O2 Hydrogen peroxide

HAEC Human aortic endothelial cells

HBSS Hank’s balance salt solution

HPLC High-pressure liquid chromatography

HRmax Maximum heart rate

HRT Hormone replacement therapy

HUVEC Human umbilical vein endothelial cells

IκB Inhibitory κB

ICAM Intercellular adhesion molecule

IKK Inhibitory κB kinase

IL Interleukin

IL-1ra Interleukin-1 receptor antagonist

IPAQ International Physical Activity Questionnaire

LDH Lactate dehydrogenase

LDL Low density lipoprotein

Lopt Optimal length

MAPK Mitogen-activated protein kinase

MCP Monocyte chemoattractant protein xiii

MDA Malondialdehyde mtDNA Mitochondrial deoxyribonucleic acid

NADH Nicotinamide adenine dinucleotide

NADPH Nicotinamide adenine dinucleotide phosphate

NF Nuclear factor

NO Nitric oxide

NSAID Non-steroidal anti-inflammatory drug

-∙ O2 Superoxide radical

OH∙ Hydroxyl radical

ORAC Oxygen radical absorption capacity

PBMC Peripheral blood mononuclear cell

PBS Phosphate buffered saline

PEP Phosphoenol pyruvate

PG Prostaglandin

PK Pyruvate kinase

PMA Phorbol myristate acetate pRb Retinoblastosis protein

RONS Reactive oxygen and nitrogen species

ROS Reactive oxygen species

SMC Smooth muscle cell

SOD Superoxide dismutase

TAC Total antioxidant capacity sTNF-R Soluble tumor necrosis factor receptor xiv

TNF Tumor necrosis factor

VCAM Vascular adhesion molecule 1

INTRODUCTION

Aging is the decrease in function and increase in disease associated with the latter portion of lifespan. The free radical theory of aging posits that these phenomena are the result of accumulated oxidative damage to DNA, resulting from a lifetime of aerobic free radical generation. Although antioxidant activity has been noted to increase with aging, reactive oxygen species production outpaces this defense resulting in increased oxidative damage.

The skeletal muscle tissue of the aged is subject to sarcopenia, which may be linked to an increase in inflammation (Peake et al., 2010). The correlation of inflammation and aging has been well-studied (Ferrucci et al., 2005) and numerous diseases have been linked to both age and inflammation (Toth et al., 2006; Bal et al., 2007; Cesari et al., 2005). Indeed the link between inflammation and aging is so strong and commonly observed that the term “inflammaging” has been coined (Franceschi et al., 2007). Therefore, strategies and treatments to reduce inflammation in the aging population could be useful in promoting the health of the elderly.

On the other hand, inflammation is regarded as an important part of the repair process following muscle damage; thus, reducing inflammation could result in reduced muscle repair.

However, in aging humans muscle growth and repair appears to be inhibited despite increased inflammation (Gomez et al., 2007), and increased systemic inflammation seen with aging may hinder the local inflammatory process required for tissue regeneration (Peake et al., 2010).

Therefore, as the aging population grows it becomes more important to understand the complex links between muscle damage, inflammation, and sarcopenia in order to find ways to increase and maintain the health of older adults.

The increase in inflammation during aging has been linked to increased nuclear factor

(NF) κB binding to DNA (Hinojosa et al., 2009). NFκB is sensitive to oxidative stress and a

2 variety of other stimuli and is responsible for the regulation of the transcription of a variety of gene targets, including pro-inflammatory cytokines interleukin (IL)-6 and tumor necrosis factor

(TNF)-α (Schreck et al., 1992). Aged rats and mice displayed increased nuclear NFκB with no increase in cytoplasmic NFκB in a variety of tissues, suggesting an upregulation of activity with age (Helenius et al., 1996).

In women the phenomenon of menopause results in a lack of production of estrogen, which adds complexity to the aging milieu. Estrogens are thought to function as antioxidants, and their absence in postmenopausal women could contribute to an increased susceptibility to oxidative stress. In addition estrogens may function to stabilize cell membranes and to regulate cell signaling through the binding to estrogen receptors (Enns and Tiidus, 2010). These mechanisms are thought to provide protection from muscle damage to women following a bout of unaccustomed exercise, meaning that perhaps postmenopausal women are left susceptible to such damage. Indeed, studies have shown that postmenopausal women experience increased serum creatine kinase (CK) and lactate dehydrogenase (LDH) as well as increased skeletal muscle mRNA expression of pro-inflammatory cytokines following strenuous eccentric exercise compared to postmenopausal women treated with hormone therapy (Dieli-Conwright et al.,

2009).

Postmenopausal women, then, represent a group at risk for sarcopenia as inflammation, oxidative damage, and NFκB activation increase with age while the protective effects of estrogrens are lost.

Antioxidants are the body’s defense against over-production of reactive oxygen species

(ROS) and the resulting damage. Nature offers an abundance of sources of antioxidants known as phytochemicals, most of which are present in plants: fruits, vegetables, and grains (Hertog,

1996). While animals are equipped with both enzymatic and non-enzymatic antioxidants, plants rely more on the antioxidant phytochemicals to protect them from auto-oxidation of polyunsaturated fatty acids by natural irradiation and airborne oxidants. Tocols (such as tocopherols and tocotrienols), flavonoids (such as soy isoflavone, tea catechins and anthocyanidines), carotenoids (β-carotene and other pigments), monophenolic acids (such as caffeic acid and ferulic acids), and polyphenolic acids (such as avenanthramides) are the most common phytochemicals. While much recognition has been made to fruits and vegetables as sources of phytochemicals, grains have been largely ignored despite the fact that they are a staple dietary component for most of the world’s population (Peterson, 2001).

Oat (Avena sativa), although consumed in considerably lower quantities worldwide than wheat and rice, has a highly edible quality, and contains high nutritional value compared to other minor grains (Peterson, 2001). Moreover, it is often consumed as a whole-grain cereal with intact bran which is rich in antioxidants.

Other than tocopherols, tocotrienols, and flavonoids, oat contains a unique group of approximately 40 different types of polyphenolic compounds called avenanthramides (AVA) that consist of an anthranilic acid derivative and a hydroxycinnamic acid derivative linked by nitrogen in a bond similar to the peptide bond (Collins, 1989). These compounds may play the role of intrinsic antioxidants in oats, and they share structural similarity to the pharmaceutical antioxidant Tranilast (Isaji et al., 1998). Of all the AVA that have been identified, three stand out due to their abundance and have been labeled as AVA-A, -B, and -C, which differ by a single moiety on the hydroxycinnamic acid ring.

In vitro, all three AVA of interest showed antioxidant activity with AVA-C being the most potent and AVA-A the least (Peterson et al., 2002). Additional studies performed in vitro

4 have shown that AVA have the anti-inflammatory and antiatherogenic effects of decreasing monocyte adhesion to human aortic endothelial cells (HAEC), as well as their expression of adhesion molecules and proinflammatory cytokines (Liu et al., 2004). AVA-C displayed further antiatherogenic potential by inhibiting vascular smooth muscle cell (SMC) proliferation and enhancing nitric oxide production in both SMC and HAEC in parallel with the up-regulation of mRNA expression of endothelial nitric oxide synthase (Nie et al., 2006a). These effects were shown to be derived from decreased NFκB activity (Nie et al., 2006b).

The antioxidant, anti-inflammatory, and NFκB inhibitory properties of AVA make it a candidate for supplementation in the cause of decreasing inflammation and muscle damage in post-menopausal women.

PURPOSES AND HYPOTHESES

The first purpose of this research was to investigate the effects of high-AVA oat supplementation on (1) oxidant-antioxidant balance, (2) inflammation, and (3) NFκB activity in both young, healthy women. The following hypotheses were tested:

Hypothesis 1: High-AVA oat supplementation will significantly decrease exercise-induced inflammation. This will be measured by plasma IL-1β, IL-6, TNF-a, and C-reactive protein

(CRP) concentrations and neutrophil respiratory burst activity.

Hypothesis 2: High-AVA oat supplementation will significantly inhibit NFκB activity. This will be measured by DNA binding levels in mononuclear cell nuclear extracts.

Hypothesis 3: High-AVA oat supplementation will significantly improve oxidant-antioxidant balance. This will be measured by erythrocyte superoxide dismutase and glutathione peroxidase activity and plasma total antioxidant capacity and glutathione concentration.

The second purpose of this research was to investigate the effects of high-AVA oat supplementation on (1) oxidant-antioxidant balance, (2) inflammation, and (3) NFκB activity in both post-menopausal women. The following hypotheses were tested:

Hypothesis 4: High-AVA oat supplementation will significantly decrease exercise-induced inflammation. This will be measured by plasma IL-1β, IL-6, TNF-a, and C-reactive protein

(CRP) concentrations and neutrophil respiratory burst activity.

Hypothesis 5: High-AVA oat supplementation will significantly inhibit NFκB activity. This will be measured by DNA binding levels in mononuclear cell nuclear extracts.

Hypothesis 6: High-AVA oat supplementation will significantly improve oxidant-antioxidant balance. This will be measured by erythrocyte superoxide dismutase and glutathione peroxidase activity and plasma total antioxidant capacity and glutathione concentration.

REVIEW OF THE LITERATURE

1. Reactive Oxygen Species

A. Mitochondrial source

ROS are formed by the incomplete reduction of oxygen in the electron transport chain

(ETC). Complete reduction to H2O requires 4 electrons; addition of 1, 2, or 3 electrons results in

-∙ ∙ the formation of superoxide radical (O2 ), hydrogen peroxide (H2O2), or hydroxyl radical (OH ).

Metal ions can lead to the production of ROS via Fenton reactions (Kasprzak, 2002). While OH∙ is the most reactive and therefore most destructive of the ROS, H2O2, which is not a free radical, can be problematic because of its relative stability that allows it move into other cellular compartments.

B. Antioxidant defense

The antioxidant enzymes superoxide dismutase (SOD), glutathione peroxidase (GPx), glutathione reductase (GR), and catalase are designed to limit the production of OH∙. Two types of SOD are found in cells: MnSOD is the mitochondrial form while CuZnSOD is found in the

-∙ cytosol. It converts O2 to H2O2 (Fridovich, 1978). Catalase and GPx convert H2O2 to H2O and

O2. GPx does so using the ubiquitous tripeptide antioxidant glutathione. GR reduces oxidized glutathione disulfide (GSSG) to reduced glutathione (GSH) using NADPH as a reducing agent.

Uncoupling proteins can also be classified as antioxidant enzymes (Skulachev, 1996).

They reduce the proton gradient across the mitochondrial inner membrane, thereby reducing the

-∙ pressure on the ETC that can lead to electron leakage and O2 formation (Vidal-Puig et al.,

2000).

The limiting of ROS is important because of the threat of oxidative damage. ROS react with protein, lipid, and nucleic acid components of the cell. Protein oxidation leads to carbonyl

7 formation, cleavage, and dysfunction (Stadtman, 1992). Lipid peroxidation is a major step in the initiation of atherosclerosis (Steinberg, 2002). DNA damage, mainly by OH∙, can cause strand breaks and has been implicated in aging (Harman, 1956).

C. Neutrophil source

Another source of ROS besides the ETC is the enzyme NADPH oxidase of the neutrophil

-∙ (Chanock et al., 1994). NADPH oxidase produces O2 in the process of phagocytosis. The reaction and release of ROS is referred to as the respiratory burst. Though phagocytosis is a beneficial process necessary to breakdown foreign material, such as during infection, and damaged cellular material, such as following eccentric exercise, the release of ROS is non- specific and can damage healthy tissue (Pyne, 1994).

Respiratory burst occurs during the inflammatory response and is one link between ROS and inflammation; they are also linked by cell signaling pathways. All aerobic organisms produce ROS, and it appears that they have evolved strategies for using them as biological stimuli (Allen and Tresini, 2000). The mitogen-activated protein kinase (MAPK) and NFκB pathways, among others, are regulated by changes in redox status. Inhibitory κB (IκB) is bound to the NFκB dimer (p50 and p65) in the cytosol. ROS trigger the release of IκB (Schreck et al.,

1991). The promoter regions of the genes of antioxidant enzymes such MnSOD (Meyrick and

Magnusen, 1994), inducible nitric oxide synthase (Adcock et al., 1994), and γ-glutamylcysteine synthetase (Iwanga et al., 1998) contain κB domains, making them redox sensitive. Therefore,

ROS serve a dual role in that they can damage cellular components but also activate genes required for cellular adaptation (Ji et al., 2006).

2. Inflammation

A. Role in disease

ROS and inflammation are also connected via chronic inflammatory diseases, including chronic obstructive pulmonary disease, asthma, rheumatoid arthritis, and atherosclerosis. Such diseases can have multiple routes of initiation and progression, but ROS and oxidative stress are thought to contribute to them by promoting cell proliferation, cytokine production, and adhesion molecule expression (Rahman, 2005). It is important to note that ROS play a factor in disease not merely because of their direct damage but also through their role as cell signals. A variety of protein kinases are redox sensitive, and a wide array of gene targets are activated downstream

(Rahman, 2005).

Interestingly, the pro-inflammatory cytokines are activated by ROS signaling, and they also serve to promote ROS generation (Ma, 2009). Therefore, oxidative stress and inflammation can create a vicious cycle.

B. Cytokines

The inflammatory response is orchestrated by extracellular messengers, the cytokines.

The cytokine response to exercise is similar to the acute phase response to infection (Pedersen and Hoffman-Goetz, 2000). During and after exercise, plasma levels of pro-inflammatory cytokines IL-1, IL-6, and TNF-α increase. However, the plasma levels of anti-inflammatory cytokines IL-1 receptor agonist (IL-1ra), IL-10, and soluble TNF-receptor (sTNF-R) also increase. While the former cytokines advance the inflammatory response to exercise, the latter are thought to limit it. The mode (e.g. running versus cycling, eccentric versus concentric), intensity, and duration of the exercise as well as the training status of the exerciser all affect the cytokine response to exercise; therefore, experimental results are often inconsistent with one another.

The injection of IL-1, IL-6, and TNF-α into rodents and humans induces inflammation and produces most components of the acute phase response, including leukocytosis, increased body temperature, and production of acute phase proteins by hepatocytes (Dinarello, 1992;

Richards and Gauldie, 1998).

Following a marathon IL-1β, IL-6, and TNF-α were found to be elevated (Ostrowski et al., 1999). In addition, IL-1ra, sTNF-R, and IL-10 were also increased. IL-1ra binds competitively to IL-1 receptors, decreasing the action of IL-1β. sTNF-R binds TNF-α in the plasma, making it unavailable to bind to receptors on the cell surface. Pleiotropic IL-10 decreases inflammation by countering the effects of pro-inflammatory cytokines through several means including NFκB and JAK/STAT signaling. While IL-6 acts as an exocrine hormone to regulate the acute phase response in hepatocytes (Gauldie et al., 1990), it has also been shown to trigger the production of IL-1ra in a negative feedback process (Tilg et al., 1994). IL-1 also increases the level of IL-1ra (Arend, 2002).

In peripheral blood mononuclear cells (PBMC), exercise modulates expression of a variety of genes including IL-ra (Connolly, 2004). IL-6 gene expression was unchanged following 30 minutes of strenuous eccentric cycling in that study, supporting previous findings of muscle and not PBMC as the source of IL-6. While many of the gene products that were upregulated, especially immediately after exercise, were considered pro-inflammatory, their expression declined rapidly during recovery, giving way to anti-inflammatory mediators.

Furthermore, distance running was found to increase IL-1β, IL-6, and TNF-α with no mediation by PBMC whose mRNA levels of these products did not change (Moldoveanu et al., 2000;

Starkie et al., 2001).

The increase in cytokines following exercise can be attenuated or eradicated by the supplementation of antioxidants, intimating that ROS play a role in the triggering of cytokine release with exercise (Vassilakopoulos et al., 2002). Again this process was found to be independent of PBMC. Instead, ROS stimulated NFκB to increase IL-6 generation in muscle

(Kosmidou et al., 2002), and IL-6 secretion by muscle accounts for the increase in plasma IL-6 observed with exercise (Steensburg et al., 2000). No other source is necessary to achieve the concentrations observed.

However, in the same cells, eccentric exercise in elderly humans led to increased NFκB activation of iNOS, cyclooxygenase (COX)-2, and IL-6 while eccentric training led to decreased activation following a second bout of eccentric exercise (Jimenez-Jimenez et al., 2008).

Interestingly, although NFκB gene expression has been found to be reduced in aging humans

(Ponnappan, 2002), muscle repair and adaptation to training do not differ between young and old

(Jimenez-Jimenez et al., 2008). These discrepancies point to the differences in mode, duration, and intensity of exercise in research and the effects that such differences can have.

There may be a correlation between IL-6 levels and muscle damage, but there is no clear consensus. It has been observed that IL-6 release is related to muscle damage (Bruunsgaard et al., 1997), but this has been refuted (Croisier et al., 1999). Furthermore, plasma IL-6 concentration has been shown to be reduced by glucocorticoids and correlated with catecholamines (Papanicolaou et al., 1996).

TNF-α is a key regulator of the inflammatory response (Bradley, 2008). It works via the

NFκB signaling pathway in a complex manner; in fact, 221 associations of TNF-α/NFκB have been observed (Bouwmeester et al., 2004).

TNF-α is produced by contracting muscle following prolonged eccentric contraction, likely via NFκB activation (Liao et al., 2009). The main source of TNF-α, however, appears to be activated macrophages and T lymphocytes although many other cell types have also been reported to secrete it (Bradley, 2007). TNF-α acts on two distinct receptors, TNFR-1 and TNFR-

2, to trigger inflammation and cell death as well as tissue repair and angiogenesis.

C. Eccentric Contraction

Downhill running presents a unique challenge to skeletal muscle distinct from those of conventional endurance exercise such as uphill running. Eccentric exercise is associated with muscle damage (Peake et al., 2005; Fielding et al., 1993) and delayed-onset muscle soreness

(DOMS; Proske and Allen, 2005). The damage and resultant soreness are attributed to lengthening contractions wherein myofibrils are stretched while contracting. Proske and Allen

(2005) posit that “some sarcomeres resist the stretch more than others, perhaps because their myofilament overlap is closer to their optimum value or because the cross-sectional area of the myofibril is slightly greater at that point,” causing weaker sarcomeres to be more stretched. Due to the relationship between length and tension, stretched sarcomeres will be weakened until myofilaments no longer overlap, a phenomenon called overstretching. Eccentric exercise consists of a series of lengthening contractions that cause a progressive increase in overstretched sarcomeres that will include progressively stronger ones. During relaxation, the myofilaments of non-overstretched sarcomeres will remain interdigitated. Overstretched sarcomeres may fail to do so resulting in disrupted sarcomeres “scattered at random along the length of the myofibril”

(Proske and Allen, 2005).

Overstretching and disruption of sarcomeres constitute the initial event in the eccentric exercise-induced damage cascade. The next step is membrane damage, which results from

“structural distortions produced by overstretched sarcomeres (Proske and Allen, 2005). Since the membranes damaged include those of the sarcoplasmic reticulum, transverse tubules, and sarcolemma, Ca2+ will enter the sarcoplasm in an uncontrolled manner.

Increased sarcoplasmic [Ca2+] is associated with a host of muscular dysfunctions including a shift in optimum length, a fall in active tension, a rise in passive tension, and DOMS.

Decreased active tension or force generation is associated with eccentric exercise. Obviously fatigue can decrease force generation independently of muscle damage; however, force may remain decreased for a week after an eccentric bout whereas it is recovered in under two hours following a concentric bout, indicating that chronic force reductions are likely due to damage and not fatigue (Walsh et al., 2004).

The optimum length for peak active tension (Lopt) is shifted to longer lengths following eccentric contraction (Whitehead et al., 2003, Jones et al., 1997). Jones et al. (1997) described a dose-response relationship wherein greater stretches during eccentric contraction were associated with greater shifts in Lopt. This shift may be explained by the overstretched sarcomere paradigm wherein overstretched sarcomeres no longer contract and cannot be further stretched during passive stretching of the muscle. Therefore, tension rises at a steeper rate because fewer sarcomeres in series are available to be stretched (Whitehead et al., 2003).

Passive tension is elevated following eccentric contractions, especially at or around Lopt

(Whitehead et al., 2003). This may be explained by the increase in sarcoplasmic [Ca2+] caused by membrane damage. Ca2+ triggers contraction leading to the development of an injury contracture

(Proske and Allen, 2005). In effect the muscle will be in a state of contraction without active excitation leading to increased passive force generation. Proske and Allen (2005) attribute the sensation of stiffness that may accompany eccentric exercise to this phenomenon.

Another consequence of increased sarcoplasmic [Ca2+] is the possibility of triggering proteolysis and the breakdown of damaged fibers (Proske and Allen, 2005). These processes are accompanied by an inflammatory response which may lead to soreness and swelling; thus

DOMS may be attributable to an inflammatory process (Smith, 1991).

The inflammatory process begins with the attraction of leukocytes to the site of injury, the muscle bed damaged by eccentric contractions (Peake et al., 2005). The infiltration of neutrophils into the damaged muscle begins within hours (Beaton et al., 2002) and can last 24 hours (MacIntyre et al., 1996). Macrophages may remain in muscles for two weeks (Round et al.,

1987). The role of these invading cells is to breakdown damaged tissue similar to the breakdown of foreign material during an immune response. The mechanism of degradation is the production of RONS by NADPH oxidase and the respiratory burst (Nguyen and Tidball, 2003).

Furthermore, they may produce and release pro-inflammatory cytokines (Cannon and St. Pierre,

1998) that can attract other leukocytes in addition to a variety of other effects. Since the respiratory burst is not selective for damaged tissue—indeed it is a general release of free radicals into the cellular environment—healthy tissue may also be degraded. Therefore, although inflammation is a necessary process for the clean-up and recovery of muscles damaged by eccentric exercise, it may also be deleterious in itself, and sufferers of such damage may be aided by a reduction in inflammation.

Whereas leukocyte infiltration, cytokine production, and phagocytosis constitute local inflammatory processes, much has been done to assess systemic inflammatory responses following eccentric exercise. An increase in circulating cytokines is one marker of such systemic inflammation. For instance, IL-6 and IL-1β concentrations were shown to increase following downhill running (Petersen et al., 2001). IL-1β binding to IL-1R initiates an intracellular

14 signaling cascade (Qiang and Engelhardt, 2006). This cascade results in the production of several transcription factors including NFκB, which promotes the production of pro-inflammatory cytokines and the progression of inflammation (Qiang and Engelhardt, 2006). Interestingly, ROS can also lead to NFκB activation (Kabe et al., 2005). Thus the etiologies of inflammation and oxidative stress share common elements.

One interesting phenomenon associated with the effects of eccentric contractions is the repeated bouts effect in which the muscles incur damage from an initial eccentric bout but less damage following subsequent bouts (McHugh, 2003; Nosaka et al, 2001). Nosaka et al. (2001) found that the repeated bouts effect resulted in expedited strength recovery, reduced swelling and soreness, and decreased sarcolemmal damage as indicated by decreased plasma CK concentrations and fewer abnormalities noted on magnetic resonance images when eccentric bouts were separated by 6 months, but such protective effects were not found after a 12-month separation. Several studies demonstrate a time-dependent decrease in the protective effects of a bout of eccentric exercise with maximal protection occurring in the first 2 weeks and all protection being lost between 9 and 12 months following a bout (Clarkson and Tremblay, 1988;

Ebbeling and Clarkson, 1989; Mair et al., 1995, Nosaka et al., 2001).

Stupka et al. (2001) found that differences between the response to a first bout of eccentric leg exercise and a second bout performed 5 to 6 weeks later included differences in proteolytic pathways. Specifically, ubiquitination was increased following the second bout but not the first, possibly indicating a change in the regulation of proteolysis toward the more selective and perhaps more efficient pathway. That is, more proteolysis via the ubiquitin- proteasome pathway could mean less degradation via neutrophils (DeMartino and Ordway,

1998). These researchers also investigated gender differences in the response to eccentric

15 exercise and found that neutrophil infiltration was higher in women following the second bout whereas men had no differences after either bout and women had no change after the first bout.

They also found that CK activity returned to baseline more quickly after the first bout (within 7 days versus more than 7 days for men).

One theory for the cause of the repeated bouts effect is that sarcomeres are added to the muscle fibers in series following an unaccustomed eccentric bout, thereby shortening the average length of the sarcomeres and making them less susceptible to damage during eccentric contractions (Faulkner et al., 1993; Morgan and Allen, 1999). Another theory is a decline in force-generating capacity due to the weakened condition of the muscle during the second bout

(Golden et al., 1992).

3. Aging A. Role of ROS and inflammation

Universality, intrinsicality, progressiveness, and deleteriousness are the hallmarks of aging (Strehler, 1977). All older members of a species inevitably decline in function independent of the environment.

Debate and research continues concerning the mechanisms of aging. The free radical theory of aging attributes the process to the interaction of normally-produced free radicals and cellular constituents (Harman, 1954). In effect, the life-long production of free radicals, a necessary provision of aerobic metabolism, leads to irreversible, accumulated damage to protein, lipid, and DNA resulting in progressive decline in health and increased probability of disease

(Harman, 2006). In particular, mitochondrial DNA (mtDNA) deletions due to free radical attack in a variety of tissues have been put forth as a plausible mechanism for aging (Troen, 2003). In human skeletal muscle, an increase in mtDNA damage was associated with increased age (Melov

16 et al., 1995). As mtDNA codes for elements of the electron transport chain, this damage leads to less efficient cellular respiration, thereby increasing free radical production and starting a vicious cycle. Compounding this issue may be the proximity of the site of free radical production in mitochondria and mtDNA (McKenzie et al., 2002).

As free radical generation is inevitable, aerobic organisms have developed a variety of antioxidant defenses to guard against oxidative damage and maintain a balance between the beneficial effects of free radical production (e.g., signal transduction) and the harmful effects. As an organism ages and free radical generation increases, antioxidant defense also increases; however, the increase in defense cannot keep pace and the balance is tipped in favor of oxidative stress (Ji et al., 1990).

Other theories of aging attribute inadequate repair systems and/or evolutionary forces as root causes of aging (Goto, 2008). Both of these factors play a role in “inflammaging,” which, by its name, implies a close relationship between inflammation and aging. The theory of inflammaging holds that the immune system responds throughout the lifespan to a variety of stimuli to maintain homeostasis by progressively increasing pro-inflammatory elements (e.g. activated lymphocytes and macrophages) in the system (Franceschi et al., 2000). Like free radicals, which are produced continuously and serve a beneficial role in cell signaling (Allen and

Tresini, 2000), inflammation plays a key role in protecting an organism from infection and in growth and development. However, its effects become detrimental later in life, a stage perhaps

“unforeseen” by evolution (Franceschi et al., 1999), as the cumulative effects can lead to chronic, systemic inflammation that plays a role in a variety of diseases (Giunta, 2006).

The skeletal muscles of aged humans and animals are slower to recover from injury than younger counterparts (Peake et al, 2010). One day following a bout of resistance exercise, older

17 women showed a 24% reduction in concentric force generation compared to only 6% in younger women (Ploutz-Snyder et al., 2001). Aged rats displayed greater muscle fibrosis than young rats

3 weeks after a contusion injury, indicating an increase in inflammation and a delay in repair

(Ghaly and Marsh, 2010). Delayed repair can lead to disuse and atrophy, resulting in further injury—a cycle that may be at the root of sarcopenia (Faulkner et al., 1995).

Whether as effects or causes, both free radical production and chronic inflammation are important aspects of aging.

B. Estrogens and Aging

Estrogens are a group of corticosteroids that play a major role in the development and maintenance of sexual and reproductive function (Heldring et al., 2007) and that also have effects in the cardiovascular, musculoskeletal, immune, and nervous systems (Enns and Tiidus,

2010). Estrogens have also been shown to act as antioxidants by scavenging free radicals, and their structure is similar to that of vitamin E (Sugioka et al., 1987). In addition, estrogens have been demonstrated to increase antioxidant enzyme activity (Strehlow et al., 2003). As free radical production and oxidative damage are linked to aging, the antioxidant effects of estrogens may play a role in limiting this damage, which may help to explain the increased longevity observed in women compared to men (Vina et al., 2006).

Another important characteristic of estrogens is their ability to integrate into membrane lipids, similar to cholesterol, and to offer added stability to the membrane (Whiting et al., 2000).

This characteristic may help to explain the protective effect of estrogens in reducing muscle damage following exercise in female rats compared to both males and ovariectomized females

(Bar et al., 1988). Human studies have been less conclusive as some researchers report

18 differences between the sexes (Carter et al., 2001; Sewright et al., 2008) and others have found none (Kerksick et al., 2008; Hubal et al., 2008).

Studies that evaluate muscle damage solely by plasma CK may not be able to accurately measure damage as studies that use more direct methods (Hyatt and Clarkson, 1998). For instance, male rats had significantly greater structural damage to their myofibers and losses of sub-membrane proteins after downhill running compared to females (Komulainen, 1999). Thus, the membrane-stabilizing effect of estrogens may provide protection against muscle damage, especially following eccentric exercise.

Post-menopausal women, then, face the threat of increased damage from both aging and the loss of estrogens’ protective effects. One strategy to remedy this situation is hormone replacement therapy (HRT). HRT has been shown to decrease oxidative damage, although without affecting cardiovascular disease outcomes, and also to protect against osteoporosis and colorectal cancer (Subbiah, 2002). Further, HRT alleviated post-exercise muscle damage and inflammation (Dieli-Conwright et al., 2009). On the other hand, HRT has been linked to increased risk of stroke, cancer, gall bladder disease, and heart disease (Van der Mooren and

Kenemans, 2004). Therefore, postmenopausal women could benefit from treatment that offers protective effects without the harmful side effects.

4. Avenanthramides A. Introduction

AVA is a group of polyphenolic, N-cinnamoylanthranilate alkaloids found exclusively in oats. They contain an anthranilic acid group connected to a hydroxycinnamic acid group by a pseudo-peptide bond (Fig. 1). Variations in functional groups bound to the two rings account for the approximately twenty-five individual species of AVA that have been classified. Three

19 species, however, are the most studied due to their abundance in oats. They are classified as

AVA-A (N-[4’-hydroxy-(E)-cinnamoyl]-5-hydroxyanthranilic acid), -B (N-[4’-hydroxy-3’- methoxy-(E)-cinnamoyl]-5-hydroxyanthranilic acid), and -C (N-[3’,4’-dihydroxy-(E)- cinnamoyl]-5-hydroxyanthranilic acid). All three contain 5-hydroxyanthranilic acid while the hydroxycinnamic acid is p-coumaric acid for AVA-A, ferulic acid for AVA-B, and caffeic acid for AVA-C.

Collins (1989) separated AVA from methanolic extracts of oat groats and hulls by ion- exchange chromatography and characterized them. They are crystalline, yellow to yellow-green substances with high melting points. They dissolve readily in organic solvents but are insoluble in water. However, at elevated pH levels they can be dissolved in even cold water, resulting in green or yellow solutions. They are subject to auto-oxidation resulting in orange, red, brown, or black resin formation.

B. Antioxidant Capacity

The aleurone, the outermost layer of the oat groat and the principal constituent of oat bran, has been found to have the highest concentration of antioxidants (Dimberg et al., 1993).

Compounds extracted with methanol from this layer displayed a high degree of antioxidant capacity, indicating that polar substances of the aleurone confer much of the antioxidant activity

Fig.1: AVA structures

21 associated with oats (Handelman et al., 1999). Although the antioxidants present in these extracts were not identified, AVA were considered candidate compounds.

Peterson et al. (2002) determined that AVA do indeed have antioxidant activity.

Synthetic AVA-A, -B, and -C were assayed for activity in two in vitro systems: β-carotene bleaching assay and 2,2-diphenyl-1-picrylhydrazyl (DPPH) assay (Peterson et al., 2002). The β- carotene bleaching assay tests the ability of substance to inhibit the auto-oxidation of β-carotene and linoleic acid in a hydrophobic environment. All three AVA showed antioxidant activity in a dose-dependent fashion in this system with the order of effectiveness being C > B > A. AVA-C was much more effective as an antioxidant than the others; however, even at the highest concentration measured, no AVA was able to completely inhibit bleaching. The DPPH assay measures the ability of a substance to reduce DPPH radicals in an aqueous environment. Due to the insolubility of AVA in water, this assay was less sensitive than the β-carotene bleaching assay; however, the same order of activity was observed (i.e. C > B > A). Both AVA-C and

AVA-B were more effective in terms of the concentration required to achieve 50% reduction of

DPPH than the synthetic antioxidant Trolox. AVA-C also showed a steeper slope than the other

AVA when the amount of DPPH reduced was plotted against the amount of antioxidant introduced to the system. This difference may be explained by the stoichiometry of the functional groups available on the AVA to interact with DPPH. AVA-C contains three hydroxyl groups, allowing it to react with more DPPH radicals than AVA-A, which has two hydroxyl groups and the lowest slope. AVA-B has a methoxy group and two hydroxyl groups; its slope was intermediate compared to AVA-C and -A. Interestingly, the order of reactivity observed for these assays matched the order of reactivity of the hydroxycinnamic acids that make up the AVA

(i.e. caffeic > ferulic > p-coumaric). These results, coupled with the average concentrations of

AVA in oat groats (C > B > A; Emmons and Peterson, 2001), indicate that AVA-C contributes the most out of the AVA species to the antioxidant activity of oats (Peterson et al., 2002).

These results were verified in a similar study again using the DPPH and linoleic acid systems to measure antioxidant activity (Bratt et al., 2003). Here the initial rate of the reactions was measured, and the order of reactivity remained the same in both systems (C > B > A).

Constituent hydroxycinnamic acids were also tested, and the order of activity in both assays also matched the results of the previous study (caffeic > ferulic > p-coumaric). Interestingly, the hydroxycinnamic acids displayed higher antioxidant activity than the AVA compounds of which they are a part. This was unexpected because theoretically the AVA would have more antioxidant activity due to resonance of the amide bond as well as their higher number of functional groups available for interaction with radicals. Bratt et al. posit that “new compounds were formed that were capable of quenching radicals” during these in vitro analyses to account for the unexpected values. The putative steps in this process include 8,8-coupling of phenoxyl radicals and the formation of a dilactone. It is unclear whether such a mechanism would be expected in an in vivo reaction.

C. Antioxidant Activity in Rats

In vivo examinations of AVA antioxidant activity have been performed in rats, hamsters, and humans. Ji et al. (2003) supplemented the diet of Sprague-Dawley rats with 0.1 g of synthetic AVA-C per kg of food for fifty days. This level of AVA supplementation would require a diet of 50-100% oats to be obtained without synthetic AVA-C (Ji et al., 2003). Control rats received a standard rat chow diet. Both AVA-fed and control rats were divided into exercise and rest groups after the diet period. Exercise consisted of a one hour treadmill run at approximately 75% of VO2max. Tissue oxidant generation was assayed using 2’,7’-

23 dichlorofluorescein (DCFH), a fluorescent dye sensitive to reactive oxygen and nitrogen species

(RONS). Exercise-induced RONS generation was attenuated in the heart and soleus muscle of

AVA-supplemented rats, but no difference was observed between supplemented and control animals in the liver, kidney, and deep vastus lateralis (DVL) muscle. Antioxidant enzyme activities were also measured in the various tissues in order to assess changes in the inherent cellular defenses against oxidative stress. SOD activity was elevated in the liver, kidney, DVL, and soleus of AVA-supplemented rats both at rest and following exercise. In the heart, SOD activity was significantly reduced following exercise in AVA-supplemented rats while resting values in the same animals were not different from controls. Heart tissue also showed elevated

GPx activity in AVA-supplemented rats while no other tissue’s GPx activity was affected by supplementation. Malondialdehyde (MDA) concentration was also measured in these tissues in order to assess the extent of lipid peroxidation. Controls showed elevated MDA levels in the liver, heart, and DVL following exercise. AVA supplementation attenuated this elevation in heart but exacerbated it in DVL. There was no effect of AVA supplementation on exercise- induced MDA level in the liver. These results suggest that chronic supplementation of AVA-C can provide protection from oxidative stress in a tissue-specific manner. They also indicate a possible genomic interaction leading to the upregulation of antioxidant enzymes or the activation of existent enzymes in addition to the free radical scavenging activity noted in in vitro studies.

D. Bioavailability in Rodents

To assess the bioavailability and antioxidant effects of AVA after an acute dose, Chen et al. (2004) fed a slurry containing 250 mg oat bran phenol-rich powder containing 40 μmol of phenolics (including AVA-A, -B, and -C) to BioF1B strain Golden Syrian Hamsters by stomach gavage. This dose was based on the estimated daily phenolic intake for a 70-kg person and

24 adjusted for the metabolism of the animals. Control hamsters received an equal volume of saline.

Blood samples were collected at 0, 20, 40, 60, 80, and 120 minutes post gavage, and plasma was analyzed by high-pressure liquid chromatography (HPLC) for the presence of phenolics. In addition to several other phenolic compounds in the mixture, AVA-A and -B were found to be bioavailable. Both detected AVA species reached a peak concentration in the plasma at 40 minutes post gavage and returned to baseline concentration by 120 minutes.

In addition to assaying for bioavailability, Chen et al. examined the ability of the absorbed phenolic mixture to protect low-density lipoprotein (LDL) from Cu+-induced oxidation.

No protective effect was measured when the 40- and 60-minute plasma samples were each added to the reaction mixture. However, when ascorbic acid was added along with the 60-minute plasma sample, a decrease in LDL oxidation that was greater than the effect of ascorbic acid alone was observed, suggesting a synergistic effect. Another antioxidant assay, the oxygen radical absorbance capacity (ORAC) assay, revealed no antioxidant effect of the absorbed phenolics. When the oat phenolic mixture (what was fed to the hamsters, not what was in the plasma) was used in the LDL oxidation assay, a protective effect with a dose-dependent relationship was observed. The synergy with ascorbic acid was also observed.

AVA-C, as well as -A and -B, were found to be bioavailable in Sprague-Dawley rats

(Koenig et al., 2011). AVA was administered to the rats via oral gavage, the rats were killed at 1,

2, 4, and 12 hours post-gavage, and the plasma was analyzed for both free and conjugated AVA by HPLC. Plasma AVA concentration peaked at 1 hour. Furthermore, AVA was found in the liver, heart, and skeletal muscle of these rats with concentrations remaining elevated above baseline up to 12 hours after ingestion. In addition, most AVA detected had undergone

25 conjugation with glucuronide and/or sulfate groups, likely in the liver. Finally, liver glutathione dynamics remained unchanged following ingestion, intimating that AVA is not toxic.

E. Bioavailability and Antioxidant Capacity in Humans

Human subjects were given an AVA-enriched mixture (AEM) that was derived from oat extracts and contained considerably more AVA than the slurry fed to the hamsters above (Chen et al., 2007). Six subjects were enrolled in the study, which had a cross-over design so that each subject served as his or her own control. Two doses of AEM, 0.5 g and 1.0 g, were administered in skim milk with the milk alone serving as a placebo control. Blood samples were collected 15,

30, and 45 minutes and 1, 2, 3, 5, and 10 hours after AEM ingestion and analyzed by HPLC for

AVA concentrations. All three AVA in the AEM (AVA-A, -B, and -C) were detected in the plasma for both doses with peak AVA concentration occurring between 1.5 and 2.3 hours, whereas in hamsters the peak occurred at 40 minutes (Chen et al., 2004) and in rats it occurred at

1 hour (Koenig et al., 2011). Bioavailability in humans was 18- and 5-fold greater than in hamsters for AVA-A and -B, respectively, suggesting there are species-specific differences in the pharmacokinetics of AVA.

This study also examined plasma glutathione concentrations, MDA levels, GPx activity, and LDL resistance to Cu+-induced oxidation as markers of AVA’s antioxidant capacity. The high dose of AEM caused a 21% increase in GSH concentration at 1 hour with a subsequent decrease to baseline. However, the ratio of GSH to GSSG did not change over the entire 10-hour time course. Plasma GPx activity increased similarly regardless of the condition. MDA concentrations also lacked significant differences between conditions. This is to be expected since the subjects did not exercise or do anything else that would be expected to affect lipid peroxidation. There was also no effect of AEM ingestion on LDL oxidation. Overall the

26 antioxidant effects of AEM in this study were modest to null with the only significant effect being a small change in GSH concentration.

The lack of observed antioxidant effects, however, does not necessarily reflect a lack of

AVA activity. Only blood samples were analyzed, so there is no clear picture of what is occurring at the tissue level. Also, the study examined the effects of an acute dose with no induction of oxidative stress via exercise or other model. Therefore, the conditions of the experiment were much different than those of Ji et al. (2003) in which rats were supplemented chronically and exercised. In fact, AVA-C did not show significant effects on tissues in resting animals; its effects were unmasked by acute exercise (Ji et al., 2003). AVA ingestion may confer antioxidant protection to exercising humans, but no study to date has been conducted to test this.

F. Anti-Inflammatory Action

In addition to antioxidant effects, anti-inflammatory effects have also been found for

AVA. All of this work has been conducted in vitro, but evidence suggests that there could be in vivo effects because of the health benefits associated with oat consumption (Anderson, 1995;

Katz et al., 2001).

Liu et al. (2004) investigated the effect of AEM on markers of atherosclerosis, including monocyte adhesion, adhesion molecule expression, and pro-inflammatory cytokine production, in HAEC. Monocyte adhesion to endothelial cells is an early event in atherogenesis. Adhesion molecules mediate the formation of atheromatous plaques during atherogenesis.

Proinflammatory cytokines such as IL-8, IL-6, and monocyte chemoattractant protein (MCP) -1 are produced by endothelial cells and orchestrate the progression of atherosclerosis. HAEC monolayers were incubated with various concentrations (4, 20, and 40 μg/ml) of AEM for 24 hours. None of these concentrations resulted in cytotoxicity.

All concentrations of AEM resulted in reduced IL-1β-stimulated U937 cell (human monocyte line) adhesion. The reductions in U937 cell adhesion displayed a dose-response relationship. AEM also significantly inhibited IL-1β-stimulated expression of intracellular adhesion molecule-1 (ICAM-1), vascular adhesion molecule-1 (VCAM-1), and E-selectin on

HAEC in a dose-dependent fashion. IL-1β stimulated the production of IL-8, IL-6, and MCP-1 in the HAEC. AEM treatment reduced these productions in a dose-dependent fashion.

These results suggest that AVA can reduce the inflammatory response of endothelial cells to an atherogenic stimulus, IL-1β in this case. The inhibitory effects of microgram levels of AVA used in this study were comparable to those of 17 μg/ml of vitamin E, which was used as a positive control and is equivalent to a daily dose of 200 IU of vitamin E supplementation. These findings may be able to help explain the anti-atherosclerotic effects of oat consumption (Liu et al., 2004).

This research was expanded upon by Nie et al. (2006a), who examined the effects of synthetic AVA-C on HAEC as well as human aortic smooth muscle cells (SMC) and rat embryonic smooth muscle (A10) cells, which play an important role in atherogenesis. When blood vessel walls are damaged and atherosclerotic lesions begin to develop, SMC become activated, proliferate, migrate to the intimal layer of the wall, accumulate lipids, and participate in plaque formation. Pretreatment with AVA-C inhibited SMC proliferation induced by fetal bovine serum (FBS) in a dose-dependent manner as measured by 3H-thymidine incorporation, a measure of DNA synthesis, and verified by cell number counts. The effect was reversible, as proliferation resumed when AVA-C was removed. No cytotoxic effects were observed.

G. Nitric Oxide and SMC Proliferation

In addition to its anti-proliferative effect, AVA-C demonstrated an ability to increase nitric oxide (NO) production in SMC and HAEC. NO is a vasodilator as well as an antiatherosclerotic agent as it can prevent platelet aggregation, leukocyte adhesion, SMC proliferation, and atherogenic gene expression (Nie et al., 2006a). AVA-C treatment increased

NO concentrations in both HAEC and SMC in a dose-dependent manner as measured by 4,5- diaminofluorescein assay. Quantitative real-time polymerase chain reaction analysis revealed that AVA-C treatment increased endothelial NO synthase (eNOS) mRNA in a fashion that mirrored the increase in NO production. These results indicate that AVA-C may upregulate the transcription and translation of eNOS, which then increases NO production, and that increased

NO bioavailability may be responsible for the inhibition of SMC proliferation.

More insight into the connection between AVA-C and anti-proliferation comes from further work with cell cultures. A10 cells were treated with AVA-C and examined with flow cytometry, a method which allows for the distinction of cell cycle phases G1, S, and G2-M by analysis of DNA content (Nie et al., 2006b). Treatment with AVA-C for 48 hours caused a shift in the distribution of the cell cycle from S phase to G1, indicating that the cell cycle had been arrested at G1 phase.

The cause of this cell cycle arrest was elucidated by examining several regulatory molecules involved in the progression from G1 to S with immunoblotting (Nie et al., 2006b).

The product of the retinoblastoma tumor suppressor gene, retinoblastosis protein (pRb), a signaling molecule between the cell cycle clock and the gene transcription machinery, regulates the progression of cells from G1 to S phase and is controlled by phosphorylation with hyperphosphorylation to ppRb allowing entry into S phase and hypophosphorylation keeping in

G1 phase (Weinberg, 1995). Cyclin-dependent kinase (CDK)-4 and CDK-6 phosphorylate pRb

29 and are regulated by cyclin D (Sherr, 1994). AVA-C treatment increased the level of ppRb dose- dependently without affecting the total amount of pRb in A10 cells. Furthermore, AVA-C treatment attenuated FBS-induced increases in cyclin D1, also dose-dependently (Nie et al.,

2006b).

In addition to these effects, AVA-C treatment caused a dose-dependent increase in p53 protein levels by stabilizing p53 and decreasing its half-life. The tumor suppressor p53 is a transcription factor for p21cip1, a CDK inhibitor, whose expression was upregulated with AVA-

C treatment (Nie et al., 2006b). These results suggest that AVA-C arrests SMC cell cycle progression by increasing the level of p53, which upregulates p21cip1 expression, leading to inhibition of CDK and decreased phosphorylation of pRb. Zhou et al. (2004) have shown that

NO can increase p53 levels; therefore, the increases in NO associated with AVA-C may cause the increase in p53 which could lead to SMC anti-proliferation

H. NFκB Signaling

A mechanism for the anti-inflammatory action of AVA was proposed by Guo et al.

(2008). They examined the effects of an AVA-enriched oat extract (AVAO), synthetic AVA-C, and the methyl ester of AVA-C (CH3-AVA-C) on NFκB signaling in HAEC. CH3-AVA had been noted previously to be ten times more effective than AVA-C in some assays of antiatherogenic potential likely because the methyl group allows for increased interaction with lipophilic biomolecules, thus its inclusion in this study (Guo et al., 2008).

The expression of pro-inflammatory cytokines is regulated in part by NFκB. This transcription factor is found in the cytosol bound to IκB, which releases NFκB when phosphorylated by its kinase (IKK), which is activated by phosphorylation. Phosphorylated IκB

30 is then ubiquitinated and degraded by the proteasome while the p50 and p65 subunits of NFκB translocate to the nucleus to regulate gene expression.

AVAO, AVA-C, and CH3-AVA-C each dose-dependently reduced the IL-1β-induced

DNA binding of p50 and p65 in HAEC (Guo et al., 2008). However, DNA binding was not affected by any of the agents when measured in HAEC nuclear extracts. This indicates that AVA do not directly inhibit NFκB binding to DNA (Guo et al., 2008). The indirect inhibitory effect of

AVA treatment appears to come via IκB phosphorylation. IL-1β treatment stimulated the phosphorylation of IKK and IκB and resulted in a decrease in IκB protein level in human umbilical vein endothelial cells (HUVEC). Pretreatment with CH3-AVA-C stopped this phosphorylation and maintained IκB level dose-dependently (Guo et al., 2008). Furthermore,

CH3-AVA-C reduced proteasome activity in an endothelial cell line and increased ubiquitin levels in HUVEC.

I. Colon Cancer Prevention

AVA species, particularly CH3-AVA-C, were found to inhibit proliferation of human colon cancer cell lines Caco-2, LS174T, HT29, and HCT116 (Guo et al., 2010). However, the effect was independent of COX-2 expression or prostaglandin (PG) E2 production, suggesting that the cell-cycle arrest effect of AVA may be responsible. In addition, AVA reduced COX enzyme activity and PGE2 production in mouse peritoneal macrophages, cells which are known play a role in the etiology of colon cancer. Taken together, these results point to AVA, in addition to the known effects of the high soluble fiber content of whole grain oat, as a contributor to the cancer-preventive characteristics of oat consumption (Guo et al., 2010).

STUDY 1: Effect of AVA Supplementation on Eccentric Exercise-induced Inflammation

and Oxidative Stress in Young Women

Abstract

Inflammation and reactive oxygen species (ROS) production play important physiologic roles. However, they are both implicated in a wide variety of diseases. Avenanthramide (AVA) is a group of oat phenolics that have shown anti-inflammatory and antioxidant capability. Thus,

AVA supplementation may be an effective means of reducing inflammation and providing antioxidant protection. PURPOSE: To investigate the effectiveness of chronic high-AVA oat supplementation to reduce inflammation and oxidative stress following a bout of downhill running (DR) in young, healthy women. METHODS: Young women (age 18-30 years, N=16) were randomly divided into two dietary groups, receiving two cookies made of oat flour (30 g wet wt each) per day made from oat containing high- (190 mg/kg) or low-AVA (8 mg/kg) for 8 weeks. Before and after the dietary regimen each group of subjects ran downhill on a treadmill (-

9% grade) for 4 bouts of 15 minutes at a speed set to elicit 75% of the maximal heart rate. Blood samples were collected at rest, immediately post-DR, and 24 h post-DR pre- and post- supplementation. Plasma creatine kinase (CK) activity was measured spectrophotometrically, and inflammatory markers interleukin (IL)-1β, IL-6, tumor necrosis factor (TNF)-α, and C- reactive protein (CRP) were measured by enzyme-linked immunosorbent assay (ELISA).

Inflammation was measured in neutrophils isolated from whole blood by quantifying luminol- enhance, phorbol ester-induced respiratory burst activity. Nuclear factor (NF) κB binding to

DNA in mononuclear cells isolated from whole blood was measured by ELISA. Total antioxidant capacity (TAC) and glutathione status were measured in plasma, and superoxide dismutase (SOD) and glutathione peroxidase (GPx) activity were measured in erythrocytes.

RESULTS: While the initial DR session resulted in increased plasma CK activity, there was no increase in plasma CK after supplementation with either level of AVA. High-AVA

33 supplementation abolished the post-DR increase in neutrophil respiratory burst activity. Both concentrations of AVA lowered plasma TNF-α concentration post-DR and elevated plasma

TAC, but only high-AVA reduced plasma IL-6 concentration and erythrocyte GPx activity.

Glutathione status was improved with either concentration of AVA as GSH:GSSG ratio increased 24 h post-DR. CRP, IL-1β, and SOD activity did not change significantly during the study. CONCLUSIONS: While chronic oral supplementation of oats provides some general benefits in raising blood antioxidant capacity, high levels of AVA was effective in reducing DR- induced inflammation and oxidative processes in young, healthy women.

Introduction

Downhill running (DR) consists of repeated eccentric contractions, have led to the overstretching of sarcomeres, the compromising of cellular membranes, and ultimately muscle damage (Proske and Allen, 2005). The skeletal muscle responds to damage by producing pro- inflammatory cytokines which induce an inflammatory response. Local inflammation is associated with swelling, tenderness, and delayed-onset muscle soreness as well as production of reactive oxygen species (ROS) by neutrophils and other macrophages.

While local inflammation is thought to play a role in the repair and regeneration of damaged muscle, systemic inflammation is associated with a variety of disease states. Moreover, while

ROS production at low levels plays a physiologic role, oxidative stress results from ROS production that outpaces antioxidant defense.

Antioxidants are the body’s defense against over-production of reactive oxygen species and the resulting damage. Oat (Avena sativa), although consumed in considerably lower quantities worldwide than wheat and rice, has a highly edible quality and contains high nutritional value compared to other minor grains (Peterson, 2001). Moreover, it is often consumed as a whole- grain cereal with intact bran which is rich in antioxidants.

35 due to their abundance and have been labeled as AVA-A, -B, and -C, which differ by a single moiety on the hydroxycinnamic acid ring.

In vitro, all three AVA of interest showed antioxidant activity with AVA-C being the most potent and AVA-A the least (Peterson et al., 2002). Additional studies performed in vitro have shown that AVA have the anti-inflammatory and antiatherogenic effects of decreasing monocyte adhesion to human aortic endothelial cells (HAEC), as well as their expression of adhesion molecules and proinflammatory cytokines (Liu et al., 2004). AVA-C displayed further antiatherogenic potential by inhibiting vascular smooth muscle cell (SMC) proliferation and enhancing nitric oxide production in both SMC and HAEC in parallel with the up-regulation of mRNA expression of endothelial nitric oxide synthase (Nie et al., 2006a). These effects were shown to be derived from decreased NFκB activity (Nie et al., 2006b).

This study was designed to test the anti-inflammatory and antioxidant capability of AVA in young, healthy women. High-AVA oat supplementation was expected to (1) significantly decrease DR-induced inflammation (2) significantly improve oxidant-antioxidant balance and (3) inhibit NFκB biding.

Methods A. Subjects

Young women aged 18-30 years were recruited from the Madison, WI, community through flyers posted in various campus locations and through direct recruitment from exercise physiology classes. Study personnel were granted permission and class time by instructors to explain the purpose of the research and to recruit interested participants. Neither students’ grades nor class status were affected by their participation or lack thereof.

They were randomly assigned to one of two groups (N = 8 per group). Control groups were designated to receive cookies made with low-AVA oat flour while experimental groups were designated to receive cookies made with high-AVA oat flour. Other than this difference, the groups were treated exactly the same in a double-blind fashion.

All participants gave informed consent before enrolling in the study. They also completed a Health History Survey to ensure that they were eligible for the study and healthy enough to exercise. Criteria for rejection from the study were smoking or other tobacco use, drinking alcohol in excess of 5 drinks per wk, use of nutraceuticals (e.g. St. John’s Wort), use of blood pressure medication, use of non-steroidal anti-inflammatory drugs (NSAIDs) except for occasional use defined as no more than 800 mg ibuprofen or equivalent per wk, vitamin supplementation other than 1 multi-vitamin per day, and use of anticoagulants or antidiabetic or hypoglycemic drugs. Pregnant subjects, as determined by self-report, were not enrolled.

All procedures were approved by the University of Wisconsin-Madison Health Sciences

Institutional Review Board.

B. Study Visits

A total of four visits were required for each subject following recruitment and consenting

(Fig. 2). There were two pre-supplementation visits and two post-supplementation visits identical to the pre-supplementation visits. The first visit of each pair consisted of completion of the

Fig. 2: Timeline of visits for young women

International Physical Activity Questionnaire (IPAQ) and health history questionnaire (see

Appendix), pain and soreness ratings, downhill running (DR), and two blood draws (one before and one after the downhill running). The second visit of each pair occurred 24 hours after the first and consisted of pain and soreness ratings and a single blood draw. The paired visits were separated by 8 wk of supplementation. In order to control for the effect of hormone levels, DR was performed between days 7 and 10 of the participant’s menstrual cycle, determined by self- report.

C. Dietary Supplementation

Subjects in the high-AVA group received cookies made with high-AVA flour (190 mg/kg, Ceapro, Inc., Edmonton, AB, Canada), and groups in the control group received cookies made with low-AVA flour (8 mg/kg). The recipe for each type of cookie was identical except in the type of flour used. Each cookie contains 30 g flour (high- or low-AVA), 5.91 ml unsweetened apple sauce (Surefine), 12.32 ml artificial sweetener (Natrataste Gold), 0.616 ml baking soda, and 0.0308 ml salt. They were baked at a low temperature (250° F) for 15 minutes to ensure that AVA was not broken down or produced by the oat during the process. AVA concentration in the high-AVA cookies was 4.6 mg/cookie (9.2 mg/day), and it was 0.2 mg/cookie (0.4 mg/day) in the low-AVA cookies. Each cookie provided 125 kilocalories (250 kcal/day) regardless of group.

Supplementation began on the evening of the second study visit (following the blood draw) and ended on the evening of the third study visit (following the second blood draw).

Subjects were furnished with cookies and instructed to consume 2 of them per day: one in the morning with breakfast and one in the evening with dinner. They were also instructed to keep

39 uneaten cookies frozen, to thaw them in the refrigerator, and never to put them in a microwave oven.

D. Downhill Running

DR was performed on a treadmill in the UW Biodynamics Laboratory. All sessions were monitored by lab personnel or student volunteers who were trained in first aid and exercise physiology and who had immediate access to the treadmill’s control panel at all times allowing for modulation of the treadmill’s speed, including the ability to immediately stop it given any signs of being unable to continue from the participant, including apparent shortness of breath, visual flushing or paleness, and gait changes or instability. Participants had access to a large red button that would trigger an emergency stop mechanism if pressed and were instructed to stop exercising if they experienced unusual shortness of breath, dizziness, or light-headedness.

Participants were allowed to drink water ad libidum throughout the sessions.

Each of the 2 DR sessions consisted of 4 bouts of 15 minutes of treadmill running separated by 3 sessions of 5 minutes of quiet rest. Each subject wore a heart rate monitor and the treadmill speed was manually adjusted at 5-minute intervals to keep the heart rate at 75% of its maximum as determined by the formula HRmax = 220 – (age in years). The treadmill grade was set at -9%. Heart rate and treadmill speed were recorded every 5 minutes. Both downhill running sessions were conducted in the same manner; therefore, treadmill speed could differ between sessions but heart rate was kept the same.

E. Blood sample collection and preparation

40 ethanol prior to the draw and wrapped in sterile gauze immediately afterward. When necessary, a heating pad was applied to this inner arm to allow for better visualization and palpation of veins.

All personnel drawing blood completed 20 hours of training at the outpatient laboratory of the

University of Wisconsin Hospital.

Whole blood (7 ml per tube; 3 tubes) was placed on ice and then immediately centrifuged at 500 x g at 4 degrees C for use in the glutathione assay (see below) or gently pipetted over two layers (3 ml of each) of density gradient (Histopaque and Ficoll-Paque) for isolation of blood cells. After centrifugation at 500 x g for 30 minutes at 20° C, plasma was removed by aspiration and frozen at -80° C.

A band of monocytes was then removed by aspiration, washed with phosphate buffered saline (PBS), and frozen at -80° C.

Next, the remaining fluid (not packed erythrocytes) was removed and washed with ice cold PBS to attain neutrophils. Any erythrocytes contaminating the sample were lysed with the addition of nanopure water. After gentle inversion, tonicity was restored by the addition of 3%

NaCl. After centrifugation at 900 x g for 5 minutes at 4° C, the neutrophil pellet was resuspended in Hank’s balanced salt solution (HBSS) and the cells counted by microscope and hemacytometer and diluted to 1.5 x 106 cells/ml for immediate analysis of respiratory burst (see below).

Packed erythrocytes were removed and stored immediately at -80° C.

F. Biological Measurements

1. ELISA

Enzyme-linked immunosorbent assay (ELISA) kits (eBioscience, Read-Set-Go! ELISA,

San Diego, CA) were used to test for the plasma concentrations of IL-1β, IL-6, TNF-α, and CRP

41 per manufacturer’s instruction. The principle of the assay is to bind the compound of interest to a capture antibody that is pre-coated on the floor of a well. Detection antibody is then attached to all sample that is bound to capture antibody and the enzyme horseradish peroxidase attached to it. When the enzyme substrate is added to the wells, a color change occurs that is proportional to the amount of enzyme, which is directly correlated to the amount of compound present in the sample. A stop solution is added before measuring the final absorbance of the wells in order to attain the endpoint reading of the assay. All samples were measured in duplicate using 96-well plates coated with capture antibody. Following sample addition, detection antibody, avidin horse radish peroxidase, and enzyme substrate were added in succession with each step separated by room-temperature incubation and thorough washing with a PBS-Tween 20 wash buffer.

Absorption at 525 nm was measured on a plate reader (Spectra MAX 340, Molecular Devices) and used to determine plasma concentration from a standard curve generated using recombinant standards provided by the manufacturer.

NFκB binding to DNA was measured by ELISA in nuclear extracts of mononuclear cells.

The assay principle is as above; however, only p65 bound to DNA was detected. Nuclear extraction was conducted according to manufacturer’s instructions (Millipore Nuclear Extraction

Kit). Manufacturer’s instructions were followed for the ELISA process, which utilized an antibody against p65 (eBioscience InstantOne ELISA). Samples were scanned using a luminometer (Turner Biosystems #2030-000).

2. HPLC a. Plasma Glutathione

Glutathione concentrations were measured by HPLC based on the method of Reed et al.

(1980) and modified by Ji and Fu (1992). Both GSH and GSSG were detected, and the ratio of

GSH:GSSG calculated. This assay was performed on plasma separated from a blood sample that was kept on ice and centrifuged at 4° C immediately upon being drawn. 250 µl of plasma was transferred to a tube containing 10 µl of 0.4 mM iodoacetic acid and excess sodium bicarbonate.

After incubation at room temperature for 1 hour, 2 µl of 2,4-dinitrofluorobenzene (Sanger’s reagent, Sigma Chemical, St. Louis, MO) was added, and the samples were kept in the dark for

28 hours before the HPLC detection. Following the method of Reed et al. (1980), concentrations of GSH and GSSG were determined using a Shimadzu UV-VIS detector at 365 nm wavelength and quantified with standard curves generated using GSH and GSSG standards. b. Avenanthramide Concentration

Cookie AVA concentrations were measured using HPLC. The method of Chen et al.

(2004) was modified for use on homogenized cookies. To 200 µl of sample, 20 µl of vitamin C-

EDTA was added. Then 500 µl of 100% acetonitrile (ACN) was added to the tubes. After 5 minutes, the samples were centrifuged at 15000 × g for 5 min. The supernatant, which contained the AVA, was removed, and the solvent was evaporated by motorized vacuum pump (Fisher

Scientific) at a pressure of approximately 200 mm Hg for approximately 5 minutes. The residue was reconstituted in 200 µl of HPLC aqueous solvent. Again the samples were centrifuged at

15000 × g for 5 min. The supernatant was transferred to an HPLC vial with a punch-through disk cap for HPLC injection.

All samples were analyzed for AVA concentration with a procedure based on Milbury

(2001) on a dual pump Shimadzu HPLC system with a UV-VIS spectrophotometric detector, a

Supelco C18 column with inline guard column, and a 23-minute ACN gradient using two solvents: A (5% ACN in H2O, 0.1% formic acid) and B (99.9% ACN, 0.1% formic acid). Total flow rate was held constant at 1 ml/min with 13% B at time 0 and increasing to 60% B at 18 min,

43 the column was kept at room temperature, and an injection volume of 20 µl was used.

Absorption at 330 nm was tracked by Shimadzu EZStart 7.2.1 software, which generated a trace of absorption over time for each sample. The software detected peaks and reported retention time, which were used to identify the compounds, and area under the curve, which were used to calculate concentration using a standard curve generated with synthetic standards graciously donated by Dr. Mitchell Wise.

3. Spectrophotometric Assays a. Plasma Total Antioxidant Capacity

Plasma total antioxidant capacity (TAC) was measured by spectrophotometer by monitoring the attenuation of 2,2’-azinobis-3-ethylbenzothiazoline-6-sulfonic acid (ABTS) oxidation at 734 nm (Re et al., 1999). A solution of 7 mM ABTS and 2.45 mM aluminum potassium sulfate (APS) was made immediately before the conducting of the assay and kept in the dark. An aliquot of 100 µl plasma was added to a final volume of 1 mL with ABTS/APS solution. The cuvette was mixed by inversion and then incubated at 37° C for 5 minutes. The cuvettes were then read against a Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid) standard curve using a spectrophotometer (Shimadzu UV160). b. Plasma Creatine Kinase

Plasma CK activity was measured as a marker of eccentric muscle damage according to the procedure of Tanzer and Gilvarg (1959). The CK reaction was coupled to NADH conversion to NAD by LDH, which, along with pyruvate kinase (PK), phosphoenol pyruvate (PEP), and

NADH, were present in the reaction mixture. The decrease in NADH concentration was tracked using a spectrophotometer (Shimadzu UV160).

An aliquot of 100 µl of plasma was added to 2.2 ml of assay reagent (8.5 mM ATP, 1.22 mM NADH, 2 mM PEP, 7 U/ml PK, 15 U/ml LDH, 28 mM MgSO4 ∙ 7 H2O, 26 mM GSH, pH =

7.4) and 700 µl of buffered creatine (0.4 M glycine, 53.2 mM creatine, 62 mM K2CO3, pH = 8.9) in a 3 ml cuvette at 25° C. Absorption at 340 nm was measured for 5 minutes. The slope of the linear portion of the absorption graph was converted to units of CK using the molar extinction coefficient of NADH. c. Erythrocyte Superoxide Dismutase

Erythrocyte SOD activity was measured by spectrophotometer () by tracking the decrease in auto-oxidation of epinephrine to adrenochrome (Sun and Zigman, 1978). Erythrocytes were lysed in 2 volumes of nanopure water. 5 µl of lysate were added to 985 µl of assay reagent (100 mM NaHCO3, 1 mM EDTA, pH = 10.2) and 10 µl epinephrine (30 mM in 0.1 N HCl) in a 1 ml cuvette, mixed by inversion, and incubated at 30° C for 6 minutes. Absorption at 320 nm was measured for 3 minutes. The slope of the linear portion of the absorption graph was used to determine SOD activity by determining the percent inhibition of epinephrine autoxidation via comparison to the blank (990 µl assay reagent, 10 µl epinephrine). Activity was normalized to hemoglobin concentration (see below). d. Erythrocyte Glutathione Peroxidase

Erythrocyte GPx activity was measured by spectrophotometer () by monitoring the change in NADPH concentration in a system with excess GSH and GR in the presence of H2O2

(Flohe and Gunzler, 1984). 4 µl erythrocyte lysate was added to a 3 ml cuvette containing 1 ml phosphate buffer (100 mM KH2PO4, 100 mM K2HPO4,1 mM EDTA, 2 mM NaN3, pH = 7.0),

200 µl GSH (10 mM), 200 µl NADPH (1.5 mM in 0.1% NaHCO3), and 6 µl GR at 37° C and read for 2 minutes at 340 nm. Then 200 µl H2O2 (1.5 mM) was added and the absorption at 340

45 nm measured for 2 minutes. The molar extinction coefficient of NADPH was used to convert the change in slope between the two states (with and without H2O2) to GPx activity. Activity was normalized to hemoglobin concentration (see below). e. Hemoglobin

Erythrocyte hemoglobin was measured using Drabkin’s reagent (potassium ferricyanide and potassium cyanide in sodium bicarbonate), which binds hemoglobin to cause a shift in maximal absorbance, which can be measured by the spectrophotometer (van Kampen and

Zijlstra, 1961). 20 µl of erythrocyte lysate (as above) was added to 5 ml of Drabkin’s reagent.

The mixture was incubated at room temperature for 15 minutes, and then 20 µl was added to

1800 µl of nanopure water in a 3 ml cuvette. The absorption at 540 nm was measured and used to calculate hemoglobin concentration from a standard curve generated with hemoglobin standards (Sigma).

4. Neutrophil Respiratory Burst

Neutrophils diluted in HBSS to 1.5 x 106 cells/ml were assayed for respiratory burst activity by luminometer using a procedure based on Benbarek et al. (1996) with modifications.

Neutrophils were incubated with luminol (Sigma) for 5 minutes at 37° C in a shaking water bath.

Then they were activated maximally with 160 µM phorbol myristate acetate (PMA), moderately with 16 µM PMA, or minimally with 1.6 µM PMA. A total of 1 x 106 neutrophils were used in each trial. The respiratory burst chemiluminescence was tracked for 30 minutes by luminometer

(Turner Biosystems #2030-000) with 10 measurements in one second of individual samples every 2.5 minutes.

The following control conditions were analyzed in addition to duplicate measures of stimulated cells. Unstimulated cells received equal volume of PMA vehicle (dimethyl sulfoxide

[DMSO]) and luminol. Luminol-free cells received equal volume DMSO and maximal PMA concentration. A cell-free blank containing equal volume of HBSS received luminol and maximal PMA concentration.

The mean of 10 measurements at each time point was calculated and the time course of the respiratory burst plotted. Area under curve was calculated by the trapezoidal rule and used as a measure of total respiratory burst activity.

5. Pain and Soreness Ratings

Ratings of pain and leg muscle soreness were collected using a visual analog scale.

Participants were prompted to place an X along a 10 cm line segment with the left terminus representing no pain or soreness and the right terminus representing the worst possible pain or soreness. The distance from the left terminus to the X was measured in millimeters and recorded.

G. Statistical Analysis

Data were analyzed using the Planned Comparison method. A3-way repeated measures

ANOVA was conducted using R (version 2.14.1) statistical software. The 3 main factors were

(1) pre- or post-supplementation, (2) timing with respect to exercise (A = rest, B = post-DR, C =

24 h post-DR), and (3) low- or high-AVA supplementation. The standard error estimate of the

ANOVA was used to complete a priori planned comparisons. The 14 comparisons made were between (1) pre- and post-supplementation, (2) A and B pre-supplementation, (3) A and C pre- supplementation, (4) A and B post-supplementation, (5) A and C post-supplementation, (6, 7) A pre-supplementation and A post-supplementation within each group, (8-10) low- and high-AVA at each post-supplementation time point, (11, 12) A and B post-supplementation within each group, and (13, 14) A and C post-supplementation within group (see Fig. 3). Significance for

47 each comparison was set at P < 0.00357, which is the quotient of 0.05 divided across the 14 comparisons.

(1)

Pre Post

(3) (5) A (2) B C A (4) B C

low high low high low high low high low high low high

(9) (11 (12) (10 (8) ) )

(13 (14 (6) ) ) (7)

Fig. 3: A priori comparisons planned for statistical analysis. Pre and Post refer to supplementation. A = Rest. B = Post-DR. C = 24 h. Low and high refer to AVA supplementation groups. Dashed lines represent the 14 comparisons.

Results A. Participant Data

The age, height, weight, and body mass index (BMI) of the study participants are displayed in Table 1. There were no significant differences between groups for any characteristic, and body weight and BMI were unchanged following supplementation.

B. Muscle damage caused by (DR)

Plasma CK activity increased sharply (P<0.05) immediately after exercise in both low-

AVA and high- AVA groups before the dietary supplementation regimen (Fig. 4). Activity returned to baseline level in both groups by 24 h after the DR bout. Following the 8-wk supplementation, the DR-induced increase in CK activity was abolished in both dietary groups and there were no significant changes from baseline level in either group. There were no significant differences between low- and high-AVA groups at any point.

C. Inflammatory Markers

Prior to supplementation, both dietary groups showed a significant increase in neutrophil respiratory burst activity 24 h after DR (P<0.05, Fig. 5), but was not significant different from baseline. Following supplementation, respiratory burst activity increased immediately after exercise and remained elevated for 24 hours only in the low-AVA group, but not high-AVA group (P<0.05).

Plasma IL-1β was measured as a pro-inflammatory cytokine marker in young women

(Fig. 6). Neither DR nor dietary AVA supplementation altered IL-1β levels.

Table 1: Characteristics of young participants. Data are mean ± SEM.

Pre-Supplementation Post- Supplementation Age (y) Height Body BMI Body BMI (m) Weight Weight (kg) (kg) Low- 22.125±0 1.58±0.0 59.32±1. 23.89±0. 59.22±1. 23.85±0. AVA .76 16 78 89 55 81 High- 22.375±0 1.59±0.0 59.09±1. 23.29±0. 59.32±1. 23.40±0. AVA .68 14 67 72 84 79

160 * Low-AVA 140 High-AVA

120

100 + 80

CK CK (U/L) Activity 60

0 Rest Post-DR 24 h Rest Post-DR 24 h Pre-Supplementation Post-Supplementation

Fig. 4: Plasma CK activity in young women. Data are mean ± (N=16). * P < 0.05, Post-DR vs. Rest. + P < 0.05 Post- vs. Pre-Supplementation.

2.5 Low-AVA

High-AVA *

2.0

1.5 §

1.0 (Arbitraryunits)

0.5 Relative Relative RespiratoryBurst Luminescence

0.0 Rest Post-DR 24 h Rest Post-DR 24 h Pre-Supplementation Post-Supplementation

Fig. 5: Neutrophil respiratory burst activity in young women. Data are mean ± SEM normalized to Pre-Supplementation Rest value (N=16). * P < 0.05, 24 h vs. Rest. § P < 0.05, High-AVA vs. Low-AVA.

0.12 Low-AVA High-AVA

0.1

0.08

0.06

Concentration(pg/ml)

β 1

- 0.04 IL

0.02

0 Rest Post-DR 24 h Rest Post-DR 24 h Pre-Supplementation Post-Supplementation

Fig. 6: Plasma IL-1β concentration in young women. Data are mean ± SEM (N=16).

Plasma IL-6 concentration was not affected by DR (Fig. 7). However, 8 wk of high-AVA supplementation resulted in a lower IL-6 level compared to low-AVA diet (P<0.05, AVA-DR interaction)

TNF-α concentration did not change significantly following DR, but there was a trend toward higher levels immediately after DR (P=0.056), Fig. 8). Also, there was a significant interaction between DR and AVA supplementation. In both groups, plasma TNF-α concentration was significantly lower (P<0.05) after supplementation compared to those immediately after DR.

Fig. 9 depicts plasma CRP concentration as an inflammatory marker. CRP values did not differ significantly with either DR or AVA supplementation, although it appeared high in 24 h post-DR.

D. NFκB

NFκB binding activity was significantly decreased (P < 0.05) following AVA supplementation in young women (Fig. 10), but there were no differences between high- and low-AVA groups although there was a trend of lower binding in the high-AVA group (P =

0.056). DR did not have a significant effect on NFκB binding.

E. Plasma Total Antioxidant Capacity

TAC was not affected by DR or after recovery in either low or high AVA group before dietary supplementation, but TAC was significantly elevated in both groups following supplementation (P < 0.05; Fig. 11).

F. Erythrocyte Antioxidant Enzymes

Erythrocyte SOD activity was unchanged following DR or following supplementation in young women (Fig. 12). No differences between groups were detected.

2.5 Low-AVA High-AVA

2.0

1.5 §

1.0

6 6 Concentration(pg/ml)

- IL

0.5

0.0 Rest Post-DR 24 h Rest Post-DR 24 h Pre-Supplementation Post-Supplementation

Fig. 7: Plasma IL-6 concentration in young women. Data are mean ± SEM (N=16). § P < 0.05, High-AVA vs. Low-AVA.

7.0 Low-AVA High-AVA 6.0 *

5.0

4.0

3.0

Concentration(pg/ml)

α -

TNF 2.0

1.0

0.0 Rest Post-DR 24 h Rest Post-DR 24 h Pre-Supplementation Post-Supplementation

Fig. 8: Plasma TNF-α concentration in young women. Data are mean ± SEM (N=16). * P < 0.05, Post-Supplementation vs. Pre-supplementation.

132.5

Low-AVA 132.0 High-AVA

131.5

131.0

130.5

130.0 CRPConcentration (mg/l)

129.5

129.0

128.5 Rest Post-DR 24 h Rest Post-DR 24 h Pre-Supplementation Post-Supplementation

Fig. 9: Plasma CRP concentration in young women. Data are mean ± SEM (N=16).

1.4 + 1.2

1.0

0.8 BindingActivity

B B Low-AVA κ

0.6 High-AVA (Arbitraryunits)

Relative Relative NF 0.4

0.2

0.0 Rest Post-DR 24 h Rest Post-DR 24 h Pre-Supplementation Post-Supplementation

Fig. 10: Mononuclear cell NFκB binding activity in young women. Data are mean ± SEM normalized to Pre-Supplementation Rest value (N=16). + P < 0.05, Post- vs. Pre-supplementation.

68 Low AVA + 67 High AVA

TAC TAC Trolox(nmolequivalents/ml) 62

60 Rest Post-DR 24 h Rest Post-DR 24 h Pre-Supplementation Post-Supplementation

Fig. 11: Plasma TAC in young women. Data are mean ±SEM (N=16). + P < 0.05, Post- vs. Pre-supplementation.

400 Low-AVA High-AVA 350

300

250

200

150 SOD SOD Activity (U/mgHb)

100

0 Rest Post-DR 24 h Rest Post-DR 24 h Pre-Supplementation Post-Supplementation

Fig. 12: Erythrocyte SOD activity in young women. Data are mean ± SEM (N=16).

Erythrocyte GPx activity was unchanged following DR before supplementation in both groups (Fig. 13). However, following supplementation GPx activity in the high-AVA group decreased significantly compared to the low-AVA group immediately after exercise. Both at rest and 24 hours after exercise GPx activity was not significantly different between low and high-

AVA or from their pre-supplementation counterparts.

G. Glutathione Status

Plasma GSH concentration was not altered immediately after an acute bout of DR, but significantly elevated 24 h after both before and after dietary supplementation (P<0.05, Fig. 14).

Following supplementation, resting plasma GSH was significantly greater in the high-AVA group compared to the low-AVA group (P < 0.05).

Plasma GSSG concentration was increased above resting levels immediately after DR (P

< 0.05) both before and after supplementation regimen, regardless of dietary AVA level (Fig.

15).

In both dietary groups, GSH:GSSG was significantly elevated 24 hours after DR sessions

(P < 0.05; Fig. 16). There were no differences between High- and Low-AVA groups.

H. Pain and Soreness Ratings

Both pain and muscle soreness increased significantly after DR and remained elevated at

24 h (Tables 2 and 3). There were no differences in pain or soreness between dietary groups.

500

450 Low AVA High AVA

400

350

300 § 250

200

GPx Activity GPx Activity (U/mgHb) 150

100

0 Rest Post-DR 24 h Rest Post-DR 24 h Pre-Supplementation Post-Supplementation

Fig. 13: Erythrocyte GPx activity in young women. Data are mean ± SEM (N=16). § P < 0.05, High-AVA vs. Low-AVA.

Low-AVA 7.0 * High-AVA § 6.0 *

5.0

4.0 mol/L)

3.0 GSH GSH (µ

2.0

1.0

0.0 Rest Post-DR 24 h Rest Post-DR 24 h Pre-Supplementation Post-Supplementation

Fig. 14: Plasma GSH concentration in young women. Data are mean ± SEM (N=16). * P < 0.05, 24 h vs. Rest. § P < 0.05, High-AVA vs. Low-AVA.

0.6 Low-AVA * High-AVA *

0.5

0.4

0.3 GSSG GSSG (µmol/

0.2

0.1

0.0 Rest Post-DR 24 h Rest Post-DR 24 h Pre-Supplementation Post-Supplementation

Fig. 15: Plasma GSSG concentration in young women. Data are mean ± SEM (N=16). * P < 0.05, Post-DR vs. Rest.

16.0 Low-AVA * High-AVA 14.0 *

12.0

10.0

8.0

GSH:GSSG GSH:GSSG Ratio 6.0

4.0

2.0

0.0 Rest Post-DR 24 h Rest Post-DR 24 h Pre-Supplementation Post-Supplementation

Fig. 16: Plasma GSH:GSSG ratio in young women. Data are mean ± SEM (N=16). * P < 0.05, 24 h vs. Rest.

Table 2: Ratings of pain reported by young women. Data are mean ± SEM. + P < 0.05 compared to Rest.

Pre-Supplementation Post-Supplementation Rest Post-DR 24 h Rest Post-DR 24 h Low-AVA 0.0±0.0 3.2±0.7+ 3.9±0.8+ 0.0±0.0 2.5±0.6+ 2.3±0.8+ High-AVA 0.0±0.0 1.1±0.6+ 1.7±0.6+ 0.0±0.0 2.4±0.6+ 2.6±0.7+

Table 3: Ratings of muscle soreness reported by young women. Data are mean ± SEM. + P < 0.05 compared to Rest.

Pre-Supplementation Post-Supplementation Rest Post-DR 24 h Rest Post-DR 24 h Low-AVA 0.0±0.0 2.1±0.3+ 3.2±1.1+ 0.0±0.0 1.3±0.6+ 2.8±0.9+ High-AVA 0.0±0.0 4.1±0.9+ 5.3±1.3+ 0.0±0.0 2.1±0.7+ 4.6±1.8+

Discussion

A. DR-Induced Muscle Damage

Muscle eccentric contraction (also known as lengthening contraction) is one form of unaccustomed muscle use that occurs in many kinds of physical activity such as weight lifting, sprinting, jumping, and rigorous sports. We selected DR as our experimental model as it has been widely adopted to study eccentric contraction in humans and animals. The initial event in the inflammatory response to DR is the damaging of muscle tissue due to the mechanical strain of sarcomeres’ being stretched during contraction. As a result of sarcolemmal damage associated with myofibrillar disruption, cytosolic enzymes (such as CK) and other constituents are released to the extracellular space and eventually appear in the blood. Following this event, neutrophils are rapidly mobilized into the circulation. Within 4 h they infiltrate muscle (Beaton et al.,

2002b) and remain in muscle for up to 24 h (Beaton et al., 2002a; MacIntyre et al., 1996;

MacIntyre et al., 2000; Raastad et al., 2003; Stupka et al., 2001). By 24 h post-DR, macrophage invasion of damaged muscle begins and neutrophil invasion ceases. Macrophages remain present in muscle for up to 2 weeks (Beaton et al., 2002a; Beaton et al. 2002b, Hamada et al., 2005;

Jones et al., 1986; Malm et al., 2000; Peterson et al., 2003; Round et al., 1987) during which damaged tissue is degraded and the muscle is repaired.

Concomitantly, eccentrically contracting muscle produces pro- and anti-inflammatory cytokines. The pro-inflammatory cytokines IL-1β and TNF-α are produced for up to 5 days

(Cannon et al., 1989; Fielding et al., 1993; Hamada et al., 2005) and act upon neutrophils to initiate the breakdown of damaged muscle tissue (Cannon and St Pierre, 1998). IL-6, which has both pro- and anti-inflammatory effects, is also produced in the days after DR (Hamada et al.,

2005; Malm et al., 2004).

While these changes occur within muscle, effects on the systemic circulation are of importance in this work. Plasma CK typically peaks within 24 h of DR (Clarkson and Ebbeling,

1988; Schwane et al., 1983; Byrnes et al., 1985) but may increase later with less intense eccentric exercise. Cytokine levels peak earlier with the systemic response being tightly regulated in healthy individuals (Peake et al., 2005).To capture the most significant events following DR as outlined above, we selected two time points for blood collection in this study: immediately post-

DR and 24 h post-DR. They were chosen to capture the immediate response to DR as well as the early phase of recovery, when plasma CK and neutrophil activity were expected to peak. By analyzing only circulating blood and not muscle samples, the interpretation of the data is limited to the systemic effects of DR, which are presumed to reflect the local effects within muscle as well as body’s immune system in response to the stress initiated in the muscle. While repeated muscle biopsies would have provided more insight into the myocellular milieu, this procedure has been observed to contribute to local inflammation (Malm et al., 2000). Studies on inflammation and NFκB activity have been conducted on peripheral blood mononuclear cells with results closely mirroring those seen in muscle, suggesting that circulating blood can serve as a reliable proxy for intramuscular samples (Jimenez-Jimenez et al., 2008; Garcia-Lopez et al.,

2007). Although the blood is not the site of the damage induced by eccentric exercise or the site of the resulting inflammatory response, venous blood drains this site, the working muscle bed, and thus contains cytokines and activated immune cells that reflect the condition of the damage and inflammation in the muscle. Therefore, peripheral blood collection can serve as an effective, relatively non-invasive means to study the effects of DR.

Plasma CK activity was the marker of muscle damage used in this study. Following the pre-supplementation DR session, CK increased significantly 24 h post-DR, indicating significant

69 damage to the membranes of the skeletal muscle. Following the post-supplementation DR session, no significant elevation in CK was observed in either dietary group.

Despite the lack of change in CK, there is evidence in this experiment that some damage did occur during the second DR session. First, participants experienced the same amount of pain and muscle soreness after each DR session. Second, in the low-AVA group, neutrophil respiratory burst increased to the same degree 24 h post-DR both pre- and post-supplementation, indicating that both sessions resulted in an increase in neutrophil response, which is activated during DR, prior to the typical peak of plasma CK activity. On the other hand, there was a trend toward an increase in CRP 24 h post-DR pre-supplementation although the difference was not significant (P = 0.056). This trend vanished post-supplementation in both groups, arguing for a lack of inflammation that could be the result of a lack of damage or of the anti-inflammatory effect of AVA.

Supplementation with oat flour of either concentration may have conferred a protective effect on skeletal muscle. The damage that occurs during DR is mechanical in nature as sarcomeres are stretched as they contract causing them to “pop” (Peake et al., 2005); however, some of the effects downstream from the initial damage are biochemical in nature. For instance, the leaking of calcium from the sarcoplasmic reticulum leads to increased intracellular calcium which triggers calpain activation and apoptosis. Calcium channels are redox sensitive; therefore,

AVA and other oat constituents could protect muscle cells from damage by reducing ROS interaction with these channels, thereby preserving membranes and plasma CK levels.

Sarcomeres, however, would still remain damaged and in need of repair, leading to neutrophil activation and infiltration as well as other activities of the inflammatory response. Polyphenols have been shown to blunt the increase in plasma CK following strenuous cycling exercise

70 compared to placebo (Morillas-Ruiz et al., 2006). There are a variety of polyphenols, such as flavonoids, as well as other antioxidants, such as tocopherols, found in oats (Peterson, 2001). As these constituents were present in both the high- and low-AVA flour, both groups would have received the protection against increasing plasma CK.

On the other hand, the initial DR session may have resulted in a muscular adaptation to provide a protective effect against damage during a subsequent bout of DR. Such a protective effect of eccentric exercise has been documented; however, its mechanism remains unknown

(Clarkson et al., 1992; Faulkner et al., 1993; Ebbeling and Clarkson, 1988). Here the DR sessions were separated by 8 weeks, and oat supplementation occurred twice daily throughout that period while the initial DR session was never repeated until the end of the period, eliminating the possibility of a training effect. Although, some of the difference in plasma CK may be attributable to this protective effect, it is improbable that a single bout of DR would confer more protection over such a long timeframe compared to an activity that was repeated with such regularity during that timeframe.

CRP is an acute phase protein produced by the liver that is present at low levels in the blood of healthy individual but increases in concentration dramatically upon response to an inflammatory stimulus, and for this reason, it is a common marker of systemic inflammation

(Puglisi and Fernandez, 2008). No significant differences in plasma CRP were detected in this study; however, there was a trend toward increased CRP 24 h post-DR pre-supplementation, indicating an inflammatory response to the muscle damage induced. This trend was absent in both dietary groups post-supplementation, just as plasma CK did not increase, indicating that along with the protection from damage, oat supplementation may have helped to reduce the subsequent inflammatory response as well.

B. Anti-Inflammatory Effects of AVA

AVA supplementation appeared to reduce inflammation after 8 weeks of supplementation in young, healthy women. NFκB is a major signaling pathway that plays a central role in inflammation, and it may be the key to AVA’s effectiveness as an anti-inflammatory. This pathway is responsive to numerous stimuli including ROS, IL-1β, and TNF-α and regulates the expression of a variety of gene products including pro-inflammatory cytokines and adhesion molecules. Dysfunction or improper regulation of signaling plays a role in many inflammatory diseases, including atherosclerosis (Monaco et al., 2004); thus, inhibiting NFκB signaling may serve as an effective means of reducing inflammation.

Previous work in arterial endothelial cells showed that AVA reduced the phosphorylation of IKK and IκB while decreasing proteasome activity, which is necessary to degrade IκB and allow translocation of p50 and p65 to the nucleus (Guo et al., 2008). Also, in cultured keratinocytes NFκB activity was decreased with a concentration as low as 1 µg/L AVA (Sur et al., 2008). In the present study, even the low dose of AVA effectively reduced NFκB signaling in mononuclear cells, as measured by p65 binding to DNA in nuclear extracts.

The concentration of AVA that the mononuclear cells were exposed to is unknown, but feeding studies would tend to have lower concentrations than experiments with cultured cells by their nature. Based on the ratio of total plasma content to oral dose of AVA-A in humans, which was reported as 0.65 ± 0.20% and was the highest AVA tested (Chen et al., 2007), plasma AVA concentration in this study could be speculated to reach at most 12.6 µg/L after the consumption of a single high-AVA cookie and 0.55 µg/L after consumption of a low-AVA cookie. Although the actual values would be expected to be lower considering that a smaller dose would be expected to have a smaller bioavailability and that AVA-B and -C have lower ratios of plasma

72 content to oral dose (Chen et al., 2007), the value calculated for the high-AVA group far exceeds, while the low-AVA group approaches the effective dose of 1 µg/L reported by Sur et al.

(2008). Furthermore, repeated doses throughout the 8 weeks of supplementation may have allowed the AVA concentration to increase beyond these levels. Besides AVA, several compounds found in oat, such as α-tocopherol (Islam et al., 1998) and flavonoids (Hamalainen et al., 2007) have been shown to reduce NFκB activity and they were likely present in both high and low AVA groups.

Previous research on AVA has demonstrated that it reduces vascular smooth muscle cell proliferation by arresting the cell cycle (Nie et al., 2006a-b). In vivo, AVA supplementation in rats was shown to reduce myeloperoxidase activity and ROS production by NADPH oxidase in skeletal muscle following DR (O’Moore et al., 2006). The anti-inflammatory effects of AVA supplementation in humans, however, had not been studied. In the present study, AVA supplementation was shown to decrease the ex vivo respiratory burst activity of neutrophils and the IL-6 concentration of plasma following DR, marking the first evidence of AVA as an anti- inflammatory agent in humans.

IL-6 is considered a pleiotropic cytokine that has been linked to both pro-inflammatory and anti-inflammatory responses, such as suppressing TNF-α and IL-1β secretion in cultured monocytes (Tilg et al., 1994). AVA may reduce IL-6 production by decreasing NFκB signaling.

However, NFκB binding was decreased in both supplementation groups while IL-6 was reduced only in the high-AVA group. This finding suggests that the inhibitory effect of AVA is dose- dependent or may work via a second mechanism to reduce inflammation. If AVA reduces inflammation via a pathway other than NFκB, such as through its antioxidant effects, then IL-6 secretion, which functions in part to turn off the inflammatory response, may be decreased.

Despite the decrease in IL-6, plasma IL-1β level was not altered with AVA supplementation or by DR, suggesting that perhaps in young, healthy women systemic levels of

IL-1β are tightly regulated. Indeed, IL-1β has been found to remain constant following DR in other studies with similar participant demographics (Bruunsgaard et al., 1997; Cannon et al.,

1991; Hellsten et al., 1997).

Another important pro-inflammatory cytokine that plays a role in DR-induced muscle inflammation is TNF-α. We found a trend toward an increase in plasma TNF-α level immediately after DR (P = 0.0631), whereas in both supplementation groups, this trend of elevated TNF-α after DR was absent (Fig. 8). This effect could also be explained by the decrease in NFκB activity as stated previously.

Neutrophils produce ROS and break down muscle tissue, and they are a major component of the inflammatory process that follows eccentric exercise. While the neutrophils collected in this study were found in circulation, the site of neutrophil activity following DR is in muscle.

Following muscle injury, the release of pro-inflammatory cytokines as well as intracellular molecules such as ATP to the circulation leads to the activation of neutrophils (Phillipson and

Kubes, 2011). Interaction with adhesion molecules on the surface of endothelial cells allows for the neutrophils in circulation to infiltrate the muscle bed. The chemotactic gradients of IL-8, interferon-γ, and other chemokines lead to their migration to site of injury. In the current study, although we examined only circulating neutrophils instead of those infiltrated into the muscle and actively engaged in phagocytosis, they are by no means inactive. Previous studies have shown that neutrophils collected from circulation post-DR are activated, as evidenced by increased expression of surface receptors cluster of differentiation (CD) 11b, CD18, and CD64

(Pizza et al., 1995). CD11b and CD18 are components of the integrin receptor that plays a role in

74 neutrophil adhesion and migration from circulation to muscle (Arnaout, 1990). CD64 is used as a marker of inflammation as it is constitutively active on neutrophils but is upregulated following inflammatory stimuli (Rudensky et al., 2008). Thus, our in vitro data of neutrophil activation should be indicative of systemic response to stress and damage inflicted by DR. This is consistent with other studies showing an increased neutrophil ROS production following eccentric exercise, (Pizza et al., 1999; Pyne et al., 2000).

In the present study, high-AVA supplementation resulted in no increase of neutrophil activity from resting level after DR compared to the significant increase observed pre- supplementation. This may be the result of the post-DR decrease in IL-6 because neutrophils can be activated by cytokines such as IL-6 (Borish et al., 1989). Other signals may also be responsible for the decrease in neutrophil activity following AVA supplementation. Factors known to activate neutrophils via surface receptors include IL-8 and granulocyte colony stimulating factor, which were not measured here but are candidates for future research.

While inflammation is an important component of muscle repair and growth, systemic inflammation, as measured by plasma proinflammatory cytokine levels, is associated with many disease states. IL-6, for instance, is strongly linked with instances of rheumatoid arthritis, and pharmacological treatments have focused on neutralizing this cytokine in the circulation (Tanaka et al., 2012). AVA is a candidate for decreasing systemic IL-6 in a natural, non-pharmacological way. Therefore, while further research in various populations is necessary before AVA’s full effects will be known, this initial investigation of AVA as an anti-inflammatory in humans is promising.

C. Effects of DR and AVA on Antioxidant Defense

DR and AVA supplementation have resulted in profound effects on antioxidant defense capacity in both erythrocytes and plasma. These findings may provide some insights into the efficacy and perspective of long-term dietary oat consumption as recommended by many previous studies (Hammond 1983, Berghofer et al. 1998; Katz et al., 2001; Maki et al., 2003).

The plasma antioxidant marker TAC was elevated in both the high- and low-AVA groups in this study, indicating that either a very small amount of dietary AVA may have the same antioxidant effect as a larger dose or that non-AVA components of the oat flour, such as flavonoids, tocopherols, and phytic acid, may be responsible for the antioxidant effect in both groups. The health effects of eating oats, including reducing blood cholesterol, have been reported previously

(Meydani, 2009), and this finding may add to those reports.

Glutathione is a ubiquitous tripeptide antioxidant that exists in a reduced state and an oxidized, disulfide state. The relative concentrations of these forms serve as a marker of redox status in the cell. Previous research has shown that following an eccentric bout whole blood

GSSG to total GSH ratio increased immediately and returned to baseline by 24 hours, but at no point was GSH:GSSG elevated above baseline (Goldfarb et al., 2011). In another study, eccentric exercise was found to increase GSSG and decrease thiol redox status for 24 hours

(Zembron-Lacny et al., 2010).

Here plasma GSH concentration increased significantly 24 h post-DR prior to AVA supplementation, suggesting a response to an oxidative stimulus by increased GSH production and hepatic output. This is supported by the finding of increased plasma GSSG concentration immediately after exercise.

GSH synthesis in the liver is initiated by the addition of glutamate to cysteine by glutamate-cysteine ligase (GCL), which is followed by the addition of glycine to this dipeptide

76 by glutathione synthase. Oxidative stress has been shown to stimulate GCL activity to increase

GSH production (Rahman et al., 1999). GCL is regulated by the activator protein-1 (AP-1) transcription factor pathway and is sensitive to TNF-α (Rahman et al., 1999). As there was a trend toward increased TNF-α (P = 0.0631) following DR, this may explain the increase in GSH.

Specifically, myocyte mitochondrial ROS production was expected to increase during exercise resulting in a more oxidized cellular environment and oxidation of GSH to GSSG, and the GSSG was released from muscle. Increased skeletal muscle output combined with decreased hepatic uptake of GSSG due to decreased blood flow to the viscera during exercise could result in the increased GSSG concentration measured here. Concomitantly, due to muscle damage during repeated eccentric contraction, skeletal muscle increased production of TNF-α, which was also released to plasma. Upon binding its receptor on hepatocytes, TNF-α triggers, via the AP-1 pathway, an increase in hepatic GCL activity and GSH production and secretion.

Plasma resting GSH concentration post-supplementation was significantly higher in high-

AVA compared to low-AVA supplementation. This result could mark a shift in redox status toward a more reduced environment due to decreased ROS production, increased ROS removal, or the sparing of GSH, which is used as a reducing agent in a variety of places. This would be consistent with the finding of increased plasma TAC; however, TAC was increased in both groups while plasma GSH concentration was elevated only in the high-AVA group. Although not statistically significant, there was a trend toward a greater plasma TAC in the high-AVA group compared to the low-AVA group (P = 0.16). If there was in fact a greater antioxidant effect derived from a greater dose of AVA, this could result in a more reduced environment and effect of a greater GSH concentration. The current data, however, do not allow for reaching that

77 conclusion, but a further investigation of AVA supplementation in humans could look to delineate the antioxidant effect in greater detail.

Plasma GSH:GSSG increased 24 hours after DR in both groups both pre- and post- supplementation, reflecting the effect of increased plasma GSH concentration. On the other hand, the ratio did not change immediately after exercise despite the exercise effect on plasma

GSSG concentration. These data point to the DR protocol as causing an oxidative stress leading to the body’s response to increase antioxidant activity and resulting in a more reduced environment.

High-AVA supplementation resulted in a significant decrease in erythrocyte GPx activity immediately post-DR, indicating a possible decrease in oxidative stress. The change was not seen at the post-supplementation baseline but occurred only after exercise, which is expected to increase oxidative stress; however, there appears to be a general decline in GPx activity in the high-AVA group compared to the low-AVA group at all 3 post-supplementation time points although only the post-DR change was significant. Furthermore, as the effect was measured in erythrocytes, which lack nuclei and the machinery to generate new protein, the decrease was not related to its concentration but may be the result of a change in the levels of substrate and/or regulatory elements. GPx appears to be more sensitive to ROS than SOD, which normally has a higher activity level (Garrett and Kirkendall, 1999). This difference may help explain why GPx activity, and not SOD activity, changed significantly in this study.

In conclusion, the present research shows that AVA is a promising candidate as an anti- inflammatory agent that may be useful for those attempting to increase their physical activity without experiencing increased inflammation. However, only young, healthy women have been evaluated here. Further research will be necessary to elucidate the anti-inflammatory and

78 antioxidant effects of dietary AVA supplementation in humans, but its effects, as well as those derived from oat supplementation even with low-AVA concentration, indicate that long-term oat consumption may confer important health benefits to the population.

STUDY 2: Effect of AVA Supplementation on Eccentric Exercise-induced Inflammation

and Oxidative Stress in Postmenopausal Women

Abstract

During aging, chronic systemic inflammation increases in prevalence and antioxidant balance shifts in favor of oxidant generation. The loss of the antioxidant and membrane- protective effect of estrogen may put postmenopausal women at increased risk of inflammation and oxidative stress. Avenanthramide (AVA) is a group of oat phenolics that have shown anti- inflammatory and antioxidant capability. Thus, AVA supplementation may be an effective means of reducing inflammation and providing antioxidant protection. PURPOSE: To investigate the effectiveness of chronic high-AVA oat supplementation to reduce inflammation and oxidative stress following a bout of downhill walking (DW) in postmenopausal women.

METHODS: Postmenopausal women (aged 50-80, N = 16) were randomly divided into two dietary groups, receiving two cookies made of oat flour (30 g wet wt each) per day made from oat containing high- (190 mg/kg) or low-AVA (8 mg/kg) for 8 weeks. Before and after the dietary regimen, each group of subjects walked downhill on a treadmill (-9% grade) for 4 bouts of 15 minutes at a speed of 5.0 km/h. Blood samples were collected at rest, 24 h post-DR, and 48 h post-DR pre- and post-supplementation. Plasma creatine kinase (CK) was measured spectrophotometrically, and inflammatory markers interleukin (IL)-1β, IL-6, tumor necrosis factor (TNF)-α, and C-reactive protein (CRP) were measured by enzyme-linked immunosorbent assay (ELISA). Inflammation was measured in neutrophils isolated from whole blood by quantifying luminol-enhance, phorbol ester-induced respiratory burst activity. Nuclear factor

(NF) κB binding to DNA in mononuclear cells isolated from whole blood was measured by

ELISA. Total antioxidant capacity (TAC) and glutathione status were measured in plasma, and superoxide dismutase (SOD) and glutathione peroxidase (GPx) activity were measured in erythrocytes. RESULTS: Both DW sessions resulted in increased plasma CK activity. High-

AVA supplementation decreased mononuclear cell NFκB binding and plasma IL-1β concentration. Neutrophil respiratory burst activity was significantly lower in the high-AVA group 24 h post-DW, and plasma CRP was lower in this group 48 h post-DW. Regardless of

AVA concentration, oat supplementation improved plasma TAC and glutathione status.

CONCLUSION: While chronic oral supplementation of oats provides some general benefits in raising blood antioxidant capacity, high levels of AVA was effective in reducing DW-induced inflammation and oxidative processes in postmenopausal women.

Introduction

On the other hand, inflammation is regarded as an important part of the repair process following muscle damage; thus, reducing inflammation could result in reduced muscle repair.

The increase in inflammation during aging has been linked to increased nuclear factor

(NF) κB binding to DNA (Hinojosa et al., 2009). NFκB is sensitive to oxidative stress and a

83 variety of other stimuli and is responsible for the regulation of the transcription of a variety of gene targets, including pro-inflammatory cytokines interleukin (IL)-6 and tumor necrosis factor

2009).

Postmenopausal women, then, represent a group at risk for sarcopenia as inflammation, oxidative damage, and NFκB activation increase with age while the protective effects of estrogrens are lost.

Antioxidants are the body’s defense against over-production of reactive oxygen species

(ROS) and the resulting damage. Nature offers an abundance of sources of antioxidants known as phytochemicals, most of which are present in plants: fruits, vegetables, and grains (Hertog,

Oat (Avena sativa), although consumed in considerably lower quantities worldwide than wheat and rice, has a highly edible quality and contains high nutritional value compared to other minor grains (Peterson, 2001). Moreover, it is often consumed as a whole-grain cereal with intact bran which is rich in antioxidants.

In vitro, all three AVA of interest showed antioxidant activity with AVA-C being the most potent and AVA-A the least (Peterson et al., 2002). Additional studies performed in vitro

85 have shown that AVA have the anti-inflammatory and antiatherogenic effects of decreasing monocyte adhesion to human aortic endothelial cells (HAEC), as well as their expression of adhesion molecules and proinflammatory cytokines (Liu et al., 2004). AVA-C displayed further antiatherogenic potential by inhibiting vascular smooth muscle cell (SMC) proliferation and enhancing nitric oxide production in both SMC and HAEC in parallel with the up-regulation of mRNA expression of endothelial nitric oxide synthase (Nie et al., 2006a). These effects were shown to be derived from decreased NFκB activity (Nie et al., 2006b).

The antioxidant, anti-inflammatory, and NFκB inhibitory properties of AVA make it a candidate for supplementation in the cause of decreasing inflammation and muscle damage in post-menopausal women.

This study was designed to test the anti-inflammatory and antioxidant capability of AVA in postmenopausal women. High-AVA oat supplementation was expected to (1) significantly decrease DR-induced inflammation (2) significantly improve oxidant-antioxidant balance and (3) inhibit NFκB biding.

Methods A. Subjects

Older women aged 50-80 years were recruited from the Madison, WI, community through the use of approved flyers and from the UW Morning Exercise Program through direct solicitation. The Morning Exercise Program is a daily group exercise program for senior women conducted on the campus of the University of Wisconsin-Madison. The exercise routine consists of a rotating regimen of stretching and aerobic dance. The group instructor granted study personnel the opportunity to speak to the group on two occasions in order to explain the purpose

86 of the research and recruit participants on a volunteer basis. Group attendees’ status in the program was in no way affected by their participation in the study or lack thereof.

Participants were randomly assigned to form two separate groups (N = 8 per group).

Control groups were designated to receive cookies made with low-AVA oat flour while experimental groups were designated to receive cookies made with high-AVA oat flour. Other than this difference, the groups were treated exactly the same in a double-blind fashion.

All participants gave informed consent before enrolling in the study. They also completed a Health History Survey to ensure that they were eligible for the study and healthy enough to exercise. Criteria for rejection from the study were smoking or other tobacco use, drinking alcohol in excess of 5 drinks per week, use of nutraceuticals (e.g. St. John’s Wort), use of blood pressure medication, use of non-steroidal anti-inflammatory drugs (NSAIDs) except for occasional use defined as no more than 800 mg ibuprofen or equivalent per week, vitamin supplementation other than 1 multi-vitamin per day, and use of anticoagulants or antidiabetic or hypoglycemic drugs. The reaching of menopause as determined by self-report was a required criterion for inclusion.

All procedures were approved by the University of Wisconsin-Madison Health Sciences

Institutional Review Board.

B. Study Visits

A total of six visits were required for each subject following recruitment and consenting

(Fig. 17). There were three pre-supplementation visits and three post-supplementation visits identical to the pre-supplementation visits. The first visit of each trio consisted of completion of the IPAQ and health history questionnaire (see Appendix), pain and soreness ratings, downhill walking (DW), and a blood draw. The second visit of each trio occurred 24 hours after the first

87 and consisted of pain and soreness ratings and a single blood draw. The third visit of each trio occurred 48 hours after the first and was identical to the second. The pre-supplementation and post-supplementation visits were separated by 8 wk.

C. Dietary Supplementation

Supplementation began on the evening of the second study visit (following the blood draw) and ended on the evening of the third study visit (following the second blood draw).

Subjects were furnished with cookies and instructed to consume 2 of them per day: one in the morning with breakfast and one in the evening with dinner. They were also instructed to keep uneaten cookies frozen, to thaw them in the refrigerator, and never to put them in a microwave oven.

D. Downhill Walking

DW was performed on a treadmill in the UW Biodynamics Laboratory. All sessions were monitored by lab personnel or student volunteers who were trained in first aid and exercise physiology and who had immediate access to the treadmill’s control panel at all times allowing for modulation of the treadmill’s speed, including the ability to immediately stop it given any signs of being unable to continue from the participant, including apparent shortness of breath, visual flushing or paleness, and gait changes or instability.

Participants had access to a large red button that would trigger an emergency stop mechanism if pressed and were instructed to stop exercising if they experienced unusual shortness of breath, dizziness, or light-headedness. Participants were allowed to drink water ad libidum throughout the sessions.

Fig. 17: Timeline of visits for postmenopausal women

Each of the 2 DW sessions consisted of 4 bouts of 15 minutes of treadmill walking separated by 3 sessions of 5 minutes of quiet rest. The treadmill grade was set at -9% and the speed to 4.0 km/h. Heart rate was recorded every 5 minutes using a heart rate monitor.

E. Blood sample collection and preparation

Mixed venous blood was drawn from an antecubital vein into 4 EDTA-coated Vacutainer tubes (7 ml each, Fisher Scientific). An elastic band was tied tightly about the upper, arm and standard antiseptic procedure was followed as the skin site was thoroughly disinfected with 70% ethanol prior to the draw and wrapped in sterile gauze immediately afterward. When necessary, a heating pad was applied to this inner arm to allow for better visualization and palpation of veins.

All personnel drawing blood completed 20 hours of training at the outpatient laboratory of the

University of Wisconsin Hospital.

A band of monocytes was then removed by aspiration, washed with phosphate buffered saline (PBS), and frozen at -80° C.

NaCl. After centrifugation at 900 x g for 5 minutes at 4° C, the neutrophil pellet was resuspended in Hank’s balanced salt solution (HBSS) and the cells counted by microscope and

90 hemacytometer and diluted to 1.5 x 106 cells/ml for immediate analysis of respiratory burst (see below).

Packed erythrocytes were removed and stored immediately at -80° C.

F. Biological Measurements

1. ELISA

Enzyme-linked immunosorbent assay (ELISA) kits (eBioscience, Read-Set-Go! ELISA,

San Diego, CA) were used to test for the plasma concentrations of IL-1β, IL-6, TNF-α, and CRP per manufacturer’s instruction. The principle of the assay is to bind the compound of interest to a capture antibody that is pre-coated on the floor of a well. Detection antibody is then attached to all sample that is bound to capture antibody and the enzyme horseradish peroxidase attached to it. When the enzyme substrate is added to the wells, a color change occurs that is proportional to the amount of enzyme, which is directly correlated to the amount of compound present in the sample. A stop solution is added before measuring the final absorbance of the wells in order to attain the endpoint reading of the assay. All samples were measured in duplicate using 96-well plates coated with capture antibody. Following sample addition, detection antibody, avidin horse radish peroxidase, and enzyme substrate were added in succession with each step separated by room-temperature incubation and thorough washing with a PBS-Tween 20 wash buffer.