Β-Carotene Absorption and Metabolism

β-Carotene Absorption and Metabolism

Dissertation

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

By:

Matthew K. Fleshman

The Ohio State University Nutrition Graduate Program

The Ohio State University

2011

Dissertation Committee:

Earl H. Harrison, Ph.D., Advisor

Mark Failla, Ph.D.

Steven J. Schwartz, Ph.D.

Ouliana Ziouzenkova, Ph.D.

Matthew K. Fleshman

2011

Abstract

β-Carotene is the most potent provitamin A carotenoid and has potential antioxidant properties. Vitamin A is an essential nutrient that plays a critical role in many biological systems. In many parts of the world provitamin A carotenoids such as

β-carotene are the primary source of vitamin A. A better understanding of the bioaccessibility, absorption, and metabolism of β-carotene may help alleviate symptoms and diseases related to vitamin A deficiency. To better understand β-carotene absorption, we have studied β-carotene in fruits and vegetables, its bioaccessibility from food sources through in vitro digestion, and absorption by humans in a controlled feeding study. We also studied the metabolism of β-carotene by measuring the presence of β-apocarotenoids in foods, mice, and humans, as well as the conversion of newly absorbed β-carotene to its retinoid metabolites.

The first study was conducted to determine the β-carotene content and its bioaccessibility/bioavailability in orange-fleshed melons. Orangedew melons have less consumer risk for food borne illness and have significantly more β-carotene than cantaloupes grown under the same conditions. Micellerization of β-carotene during simulated digestion of orange-fleshed melons was approximately 3.2%. We also detected and quantified β-apocarotenoids in the melons. ii

The second study was conducted to investigate in humans the variability in β- carotene absorption and its conversion to vitamin A and to compare the efficiency of absorption of β-carotene with that of cholesterol. Ten men consumed a 5 mg dose of deuterium labeled β-carotene (d8-βC), with 6 subjects repeating the dose 2 months later.

For this study, we developed a method that provides easier sample preparation than previous β-carotene feeding studies, and allows us to detect and quantify newly absorbed d8-β-carotene, as well as its d4-retinyl ester metabolites, using high performance liquid chromatography-mass spectrometry (HPLC-MS). We employed a simple sample extraction for both retinoids and β-carotene that included minimal sample handling. D8-β-carotene and its d4-retinyl ester metabolites were analyzed using the same liquid chromatography system and solvents for separation of the compounds.

The method allowed us to accurately measure d8-β-carotene absorption, and its extent of conversion to d4-retinyl esters. There was marked inter-individual variability in β- carotene absorption and conversion to retinyl-esters. In contrast, intra-individual variability in β-carotene absorption and its conversion to retinyl esters was low. β-

Carotene and cholesterol may share specific intestinal transporters, but there was no correlation between an individual’s efficiency of absorption of β-carotene and his efficiency of absorption of cholesterol (r=-0.09; p=0.81).

The third study was conducted to determine the β-apocarotenoid levels in commonly consumed products containing β-carotene, such as fruits and vegetables, as well as in murine serum and liver, and human plasma. β-Apocarotenoids are present in our diets but their absorption, metabolism, and biological roles are largely unknown.

iii

Using HPLC-MS, we were able to detect and quantify β-apo-13-carotenone, β-apo-14’- carotenal, β-apo-12’-carotenal, β-apo-10’-carotenal, β-apo-8’-carotenal and β-carotene in several fruits and vegetables. In vitro digestion studies have shown that β- apocarotenoids are micellerized to a greater extent than β-carotene. We have shown here that they are present in considerable amounts in the diet and that they may be absorbed in addition to being generated in vivo. We have also shown that these β- apocarotenoids are present in the liver and serum of β-carotene fed mice and in human plasma. β-Carotene utilization involves many complex processes both at the level of absorption and metabolism. We have uncovered new aspects of β-carotene absorption and metabolism that should be considered in future β-carotene studies, such as conversion of β-carotene to specific retinyl esters, low intra-individual variability of β- carotene absorption, and the presence of β-apocarotenoids in all β-carotene containing samples.

Acknowledgements

This research could not have been accomplished without the generous help and support of many people. First and foremost, I would like to express my sincerest appreciation to my mentor, Dr. Earl H. Harrison, for his support and guidance, and for sharing his vast knowledge of my dissertation topic over the past five years. I would like to thank Dr.

Steven J. Schwartz for allowing me to collaborate with his laboratory on the majority of my studies, for his expertise and for serving on my committee. I would like to thank Dr.

Mark Failla for challenging my scientific mind, his expertise, and for serving on my committee. I would also like to thank Dr. Ouliana Ziouzenkova for her valuable expertise and for serving on my committee. I am grateful to my colleagues and friends

Dr. Rebekah S. Marsh, Dr. Ken Riedl, Dr. Sagar Thakar, Dr. Julie

Chitchumroonchockchai, Vanessa Reed, Rachel E. Kopec, and Abdulkerim Eroglu. I feel very fortunate to have worked with, and learned from, each of you. I also thank other Harrison laboratory members, GSNS members and the multitude of other graduate students that have helped me throughout the years. I am especially grateful to my parents, Jude and Naomi Fleshman for whom none of this would be possible, and my siblings; Gene, Dan, Jimmy, David, Theresa, Stephen, Nancy, Mark, Andrew, Joann,

Michele, Sharon, and their spouses.

Vita

2003 ...... B.S. (Nutrition), The Ohio State University

2006-2007 ...... Graduate Teaching Associate, The Ohio State University

2007-2011 ...... Graduate Research Associate, The Ohio State University

Publications

Igor Shmarakov, Matthew K. Fleshman, Diana N. D’Ambrosio, Roseann Piantedosi, Ken M. Riedl, Steve J. Schwartz, Robert W. Curley, Jr., Johannes von Lintig, Lewis P. Rubin, Earl H. Harrison, and William S. Blaner. Hepatic Stellate Cells are and Important Cellular Site for β-Carotene Conversion to Retinoid. Arch. Biochem. Biophys. (2010), doi:10.1016/j.abb.2010.05.010.

Matthew K. Fleshman, Gene E. Lester, Ken M. Riedl, Rachel E. Kopec, Sureshbabu Narayansamy, Robert W. Curley, Jr., Steven J. Schwartz, and Earl H. Harrison. (2011) Carotene and Novel Apocarotenoid Concentrations in Orange-fleshed Cucumis melo Melons: Determinations of β-Carotene Bioaccessibility and Bioavailability. J Agric. Food Chem. (2011), doi: 10.1021/jf200416a.

Field of Study

Major Field: The Ohio State University Nutrition Graduate Program

Table of Contents

Contents Page

Abstract ...... ii Acknowledgement ...... v Vita ...... vi List of Tables...... ix List of Figures ...... x List of Abbreviations ...... xii

Chapter 1 Literature Review 1.1 Introduction ...... 1 1.2 Carotenoids...... 3 1.3 Health Benefits of Carotenoids ...... 4 1.4 Health Benefits of Vitamin A ...... 4 1.5 Sources of β-Carotene ...... 9 1.6 Absorption of β-Carotene...... 9 1.7 Bioavailability of β-Carotene ...... 12 1.8 Human Studies...... 13 1.9 β-Apocarotenoids ...... 16 1.10 Aims ...... 17

Chapter 2 Carotene and Novel Apocarotenoid Concentrations in Orange-fleshed Cucumis melo Melons: Determinations of β-Carotene Bioaccessibility and Bioavailability 2.1 Abstract ...... 21 2.2 Introduction ...... 22 2.3 Materials and Methods ...... 24 2.4 Results and Discussion ...... 31 2.5 Acknowledgements ...... 36 2.6 Figures and Tables ...... 37

vii

Chapter 3 An HPLC/MS Method for the Detection of d8-β-Carotene and Individual Fatty Acyl Esters of d4-Retinol: Application to the Study of the Intestinal Absorption of β-Carotene and its Conversion to Vitamin A in Humans 3.1 Abstract ...... 44 3.2 Introduction ...... 44 3.3 Materials and Methods ...... 46 3.4 Results ...... 50 3.5 Discussion ...... 52 3.6 Acknowledgements ...... 55 3.7 Figures and Tables ...... 55

Chapter 4 Efficiency of Intestinal Absorption of d8-β-Carotene in Humans: Variability, Rate of Conversion, and Relationship to Cholesterol Absorption 4.1 Abstract ...... 65 4.2 Introduction ...... 66 4.3 Materials and Methods ...... 68 4.4 Results ...... 74 4.5 Discussion ...... 76 4.6 Acknowledgements ...... 78 4.7 Figures and Tables ...... 79

Chapter 5 The Presence of β-Apocarotenoids in Biological Samples 5.1 Abstract ...... 87 5.2 Introduction ...... 88 5.3 Materials and Methods ...... 89 5.4 Results ...... 91 5.5 Discussion ...... 94 5.6 Acknowledgements ...... 96 5.7 Figures and Tables ...... 97

Chapter 6 Epilogue ...... 103

List of References ...... 108

viii

List of Tables

2.1. Dry Weight Percent and β-Carotene Concentration of Glasshouse Grown Frozen and Field-Grown Fresh Orange-Fleshed Honeydew and Cantaloupe Melon Edible

Mesocarp Tissues...... 38

2.2. β -Carotene Concentration from Orange-Fleshed Melon in Whole Edible Tissue,

Digesta (Digestive Stability) and Aqueous Fractions (Micellerization or Bioaccessibility)

Following In Vitro Digestion (N=5) ...... 40

3.1. Standards and Internal Standards ...... 56

3.2. Area Under the Concentration-Time Curves for d8-β-Carotene and d4-Retinyl-

Esters in Whole Plasma and Chylomicrons ...... 63

4.1. Nutrient contents of interest as average intake per day ...... 79

4.2. Subject Characteristics...... 80

4.3. β-Carotene and Cholesterol Absorption ...... 80

5.1. Serum and Liver Levels of β-carotene and β-apocarotenals for β-carotene Fed Wild

Type and BCO1-Deficient Mice ...... 100

5.2. β-Apocarotenoid profiles of mouse feed and β-C beadlet ...... 101

5.3. Bioavailability of β-apocarotenoids ...... 101

List of Figures

1.1. Functions of retinoids ...... 5

1.2. Prevalence of vitamin A deficiency ...... 6

1.3. β-Carotene central and eccentric cleavage ...... 8

2.1. Chromatogram ...... 37

2.2. Chromoplast ...... 39

2.3. β-Apocarotenoids and β-carotene in the ‘OrangeDew’ melons ...... 41

2.4. β-Apocarotenoid levels in two different orange-fleshed melons ...... 42

3.1. Standard curves...... 55

3.2. Labeled and native β-carotene and retinol ...... 57

3.3. Chromatograms of labeled and native β-carotene ...... 58

3.4. Chromatograms of labeled and native retinyl-esters ...... 59

3.5. Plot of the individual d4 retinyl esters and βC-d8 in the CM-rich fraction ...... 60

3.6. Whole plasma v. CM-rich fraction ...... 61

3.7. D4-REs and native REs overlaid ...... 62

4.1. Correlation of the total βC-d8 AUC (µM*hr) versus the percent cholesterol absorption ...... 81

4.2. Correlation of the total βC-d8 AUC (µM*hr) versus the percent conversion to

x vitamin A ...... 82

4.3. Cholesterol absorption of the six subjects who repeated the experiment ...... 83

4.4. β-Carotene absorption of the six subjects who repeated the experiment ...... 84

4.5. β-Carotene conversion to retinyl esters of the six subjects who repeated the experiment ...... 85

5.1. Structure of β-carotene and its β-apocarotenoid eccentric cleavage products ..... 97

5.2. Chromatograms of β-apocarotenoid standards and in cantaloupe ...... 98

5.3. β-Apocarotenoid profiles of commonly consumed β-carotene containing foods 99

5.4. Inverse relationship of chain length to micellerization ...... 102

List of Abbreviations

APcI atmospheric pressure chemical ionization

AUC area under the curve

BCO1 β-carotene oxygenase 1

BCO2 β-carotene oxygenase 2

BHT butylated hydroxytoluene

CAC cholesterol absorption coefficient

CCD carotenoid cleavage dioxygenases

CD36 cluster determinant 36

CEH cholesterol ester hydrolase

CM chylomicron

DMEM Dulbecco’s modified Eagle’s medium

2 d8-βC [ H8] β-carotene

2 d4-RE [ H4] retinyl-ester

FA fatty acid

FSR fractional synthesis rate

HEAT Hexane/Ethanol/Acetone/Toluene

HPLC high performance liquid chromatography

xii

IS Internal standards

KOH potassium-hydroxide

LCMS Liquid chromatography mass spectrometry

LDL low density lipoprotein

LOD limit of detection

LOQ limit of quantification

LRAT lecithin retinol acyltransferase

MeOH Methanol

MTBE methyl-t-butyl-ether

NHANES national health and nutrition survey

NPC1L1 Niemann-Pick type C1 Like 1

PBS phosphate-buffered saline

PPAR peroxisome proliferator activated receptors

RAR retinoic acid receptor

RARE retinoic acid response element

RBC Red blood cell

ROS reactive oxygen species

RXR retinoid X receptor

SDS sodium dodecyl sulfate

SR-B1 scavenger receptor class B, type I

TEM transmission electron microscopy

THF tetrahydrofuran

VLDL very low density lipoprotein

xiii

CHAPTER 1

Literature Review

1.1 Introduction

Vitamin A deficiency is one of the most prevalent vitamin deficiencies throughout the world, affecting over 100 million people and occurring in regions where people obtain most of their vitamin A as provitamin A carotenoids (During et al., 1996).

A better understanding of the mechanisms of absorption and metabolism of provitamin

A carotenoids could help alleviate nutritional problems associated with vitamin A deficiency. Vitamin A was the first discovered essential vitamin, hence the designation

A. Vitamin A is also toxic at high levels. The toxicity of vitamin A has limited its use as a supplement but most western diets are rich in vitamin A fortified products and natural sources of preformed vitamin A such as meat. Plant sources also contain high levels of vitamin A from pro-vitamin A carotenoids such as β-carotene, α-carotene, and β- cryptozanthin with β-carotene being the most potent as a vitamin A source. Natural vitamin A sources are all essentially derived from plant sources, as herbivores, such as ungulants, obtain almost all of their vitamin A from vegetation. However, other

1 retinoid-like compounds exist in pro-vitamin A containing fruits and vegetables that may significantly contribute to vitamin A activity. β-Apocarotenoids and α- apocarotenoids are likely present in all foods containing their respective parent carotenoids and may have important biological roles as well. Some evidence also exists to show that β-apocarotenals may serve as a source of vitamin A in deprivation (X. D.

Wang et al., 1996) but β-apo-13-carotenone appears to act as a retinoic acid antagonist

(Eroglu et al., 2010). The recent finding of pro-vitamin A sources as a source of β- apocarotenoids, up to 4% of carotenoid levels, may be particularly important in regions where people obtain the majority of their vitamin A from plant sources such as sub-

Saharan Africa, India and Asia.

β-Carotene is the best pro-vitamin source of vitamin A because it is a symmetric molecule that stoichimetrically produces two moles of vitamin A for every one mole of

β-carotene, whereas the other two pro-vitamin A carotenoids only give rise to one mole of vitamin A. However, carotenoids are poorly absorbed in humans, which may reduce their vitamin A potency by 12-fold.

As a hydrophobic nutrient, carotenoids are absorbed in the same fashion as other lipids. It was originally thought that carotenoids are taken up by passive diffusion but recent studies suggest that specific transporters exist that facilitate the carotenoid uptake by intestinal mucosal cells, particularly transporters involved in cholesterol transport.

Carotenoid intake also may effect the expression of specific transporters and the differentiation of intestinal cells and proteins that may affect uptake of the carotenoids themselves.

Once taken up by the intestinal cells β-carotene can be metabolized by β- carotene oxygenase 1 (BCO1) and β-carotene oxygenase 2 (BCO2) resulting in retinal and β-apocarotenoids respectively. These metabolites along with intact β-carotene are packaged into chylomicrons and transported to the plasma where they can be taken up by tissues, primarily the liver where they can be further metabolized or utilized.

The bioaccessibility/bioavailability of β-carotene varies across different food sources of the carotenoid and the absorption varies within individual humans and is not well understood. Therefore, my research has been directed towards better understanding

β-carotene bioaccessibility/bioavailability and the absorption and metabolism of β- carotene.

1.2 Carotenoids

Over 700 natural carotenoids have been identified (Britton et al., 1995). There are two types of carotenoids: xanthophylls and carotenes (Britton et al., 2004).

Xanthophylls contain oxygen and carotenes are pure hydrocarbons. Animals do not have the ability to synthesize carotenoids and must obtain them from plant sources.

Carotenoids are synthesized in plants and serve to collect blue light in photosynthesis and to protect the plant from reactive oxygen species (ROS). Carotenoids are highly hydrophobic compounds and are thus associated with other lipids. Within the plant cell, carotenoids are located in organelles such as chloroplasts and chromoplasts. In fruits carotenoids are more concentrated near the rind than near the core.

1.3 Health Benefits of Carotenoids

Epidemiological and biological studies have implicated carotenoids as dietary anticarcinogens and antioxidants (Krinsky, 2001; Prakash et al., 2001). The antioxidant properties of carotenoids come from their ability to quench singlet oxygen due to their conjugated double bonds. For example, lycopene has eleven conjugated double bonds and is a powerful antioxidant (Stahl & Sies, 1996). In the eye lutein and zeaxanthin protect the macula from harmful blue light and ultraviolet light. Much of the beneficial effects of carotenoids are from their vitamin A activity. There is a subclass of carotenoids that can be cleaved to produce retinal. This subclass is known as provitamin

A carotenoids and include β-carotene, α-carotene, and β-cryptoxanthin.

1.4 Health Benefits of Vitamin A

Figure 1.1. Functions of retinoids and their common structures (During & Harrison,

2006).

Vitamin A activity is important in vision, cell differentiation, reproduction, and immunity (Figure 1.1). Vitamin A can be toxic in high amounts. β-carotene can be converted to vitamin A but it cannot cause vitamin A toxicity. However, some studies have shown negative effects of high β-carotene dosing (Albanes et al., 1995; Omenn et al., 1996). Although preformed vitamin A is the best dietary source of vitamin A, it is not readily available in developing countries (Figure 1.2). People in these regions do not have access to red meat or foods fortified with preformed vitamin A and thus obtain

5 most of their vitamin A through provitamin A carotenoids in fruits and vegetables. As will be discussed later, the bioavailability and bioaccessibility of provitamin A carotenoids varies and is affected by many factors. Increasing the consumption, bioavailability and bioaccessibility of provitamin A rich foods may help to alleviate the vitamin A deficiencies that plague these regions.

Figure 1.2. Map of the global prevalence of vitamin A deficiency (World Health

Organization, 2009).

To have provitamin A activity, the carotenoid must have at least one unsubstituted β-ionone ring and the correct number and position of methyl groups.

These carotenoids can be cleaved in mammals by β-carotene oxygenase 1 (BCO1) at the 15,15 double bond to give retinal which can be subsequently be reduced to retinol by retinal hydrogenase or oxidized to retinoic acid by retinal dehydrogenase. Retinal can be reduced to retinol and the retinol can then be esterified by lecithin retinol acyltransferase (LRAT) for storage or can be converted to retinoic acid by retinal dehydrogenase. Retinoic acid regulates the transcription of hundreds of genes through binding to nuclear receptors that act as transcription factors. Of the provitamin A carotenoids, β-carotene is the most abundant and the most potent. Its central cleavage gives two molecules of retinal (Figure 1.3).

Figure 1.3. β-Carotene metabolism. β-Carotene can be enzymatically cleaved by

BCO1 at the central 15, 15‟ double bond or by BCO2 at eccentric double bonds.

β-Carotene

1.5 Sources of β-Carotene

β-Carotene is a red-orange pigment that is present in many fruits and vegetables.

Major sources of β-carotene include carrots, squash, dark green leafy vegetables, pumpkin, sweet potato, and cantaloupe. Melons are a good source of β-carotene (Lester

& Eischen, 1996). β-Carotene, α-carotene, lycopene, lutein, and β-cryptoxanthin are the five most abundant carotenoids present in the human body (During & Harrison, 2006) and are the principal dietary carotenoids. According to NHANES (2000), the major contributors to the intake of provitamin A carotenoids in the United States are carrots, cantaloupes, sweet potatoes, and spinach (During & Harrison, 2006). β-Carotene also may be found in dietary supplements, usually as synthetic all trans β-carotene.

1.6 Absorption of β-Carotene

The first step in the absorption of β-carotene involves its release from the food matrix. Each food matrix is different and the ability of β-carotene to be released from the matrix varies for each food which subsequently affects it absorption. Food processing, such as cooking and cutting, increases the bioavailability of β-carotene by disrupting the cell walls and exposing β-carotene to digestive enzymes and bile salts; the latter are needed to incorporate β-carotene into mixed lipid micelles. It has been shown that the co-ingestion of dietary fats increases the micellerization of β-carotene by stimulating the release of pancreatic enzymes and bile salts necessary for

9 micellerization (Garrett et al., 2000; Hedren et al., 2002; Huo et al., 2007; Mills et al.,

2009).

Once β-carotene is incorporated into the mixed lipid micelles, it can be taken up by intestinal mucosal cells. In vitro studies using Caco2 cells show 10-20% uptake of the available micellerized β-carotene under standard conditions. The uptake was originally thought to occur solely by passive diffusion but recent studies suggest facilitated uptake by transport proteins such as scavenger receptor class B, type I (SR-

B1), cluster determinant 36 (CD36), and Niemann-Pick type C1 Like 1 (NPC1L1)

(During & Harrison, 2006). Once the β-carotene is taken up by the intestinal mucosal cell it is either cleaved by BCO1 or BCO2 or it is absorbed intact. The β-carotene and its metabolites are incorporated into chylomicrons along with cholesterol esters, triglycerides, apolipoprotein B, and phospholipids. The chylomicrons are then secreted in the lymph. In the plasma the chylomicrons are degraded and chylomicrons remnants are taken up by the liver and other tissues where the β-carotene is stored, further metabolized, or used as an antioxidant. Most of the absorbed β-carotene is taken up by the liver where there is the highest expression of BCO1 and BCO2 in humans

(Lindqvist & Andersson, 2004; Lindqvist et al., 2005). Intact β-carotene can be released from the liver into the circulation with VLDL. Other tissues that accumulate high levels of β-carotene include kidney, lungs, prostate, muscle, and adipose, all of which express the BCO enzymes.

As previously mentioned the uptake of β-carotene by intestinal epithelial cells is not well defined. β-Carotene, like other dietary lipids, is solubilized in the mixed lipid

10 micelles which diffuse through the aqueous phase of the intestinal lumen to the brush border membrane. Several recent studies have suggested that uptake of dietary lipids including β-carotene and other carotenoids are protein-mediated. The concentration dependence (saturation) of β-carotene uptake and secretion in CMs, the discrimination between β-carotene isomers for their cellular uptake, and the differential absorption of different carotenoids and their interactions observed during transport through Caco-2 cells all suggest that the intestinal transport of carotenoids might be facilitated by the participation of a specific epithelial transporter (During & Harrison, 2006).

Ezetimibe has been shown to effectively inhibit intestinal cholesterol absorption in humans (Garcia-Calvo et al., 2005; Ge et al., 2008) and has been shown to inhibit carotenoid transport and down-regulate expression of lipid transporters such as SR-BI

(During et al., 2005) and can bind to several brush border membrane proteins (Kramer et al., 2003). Several studies have shown that SR-BI is involved in intestinal cholesterol absorption (Altmann et al., 2002; Levy et al., 2004) and carotenoid uptake (During et al.,

2008; Kiefer et al, 2002), specifically β-carotene absorption (van Bennekum et al.,

2005). However, SR-BI is not essential in cholesterol absorption. SR-BI -/- mice did not show any difference in the absorption of cholesterol when compared to SR-BI +/+ mice

(van Bennekum et al., 2005). It is suggested that the lack of SR-BI in the SR-BI -/- is compensated by other transport proteins such as CD36 and NPC1L1 (Altmann et al.,

2004).

More recently NPC1L1 has been shown to be the critical intestinal sterol transporter which influences whole body cholesterol homeostasis (Davis et al., 2008).

Further localization of lipid transport proteins in human intestines is needed to examine their roles in β-carotene absorption. SR-BI is found throughout the human intestine but is found only in the duodenum of mouse intestine (Lobo et al., 2001; van Bennekum et al., 2005). β-Carotene absorption predominately occurs in the proximal region of the small intestine (Furr & Clark, 1997), thus β-carotene transport proteins would have to be localized within the proximal region. NPC1L1 was found to be highly expressed in the jejunum and not expressed in other tissues in the mouse (Davis et al., 2004; Davis et al., 2008). Like cholesterol (Kramer et al., 2003), the intestinal absorption of β-carotene may not be facilitated by a single transporter protein but by a complex process involving multiple proteins (Hui et al., 2008) and to some extent passive diffusion.

1.7 Bioavailability of β-Carotene

An in vitro digestion model can be used to estimate the bioavailability of β- carotene (Garrett et al., 1999). The model simulates gastric and intestinal phases of digestion and β-carotene is micellerized. The extent of micellerization represents the bioaccessibility of β-carotene and the percent micellerization varies greatly among different food sources and preparations. The in vitro digestion model is a good estimate of in vivo bioavailability (Reboul et al., 2006) when coupled with Caco-2 cell uptake.

The human Caco-2 cells differentiate into cells that exhibit characteristics of enterocytes including cellular polarization and brush border membranes. These cells also express previously mentioned transporter proteins SR-BI and NPC1L1 (During et al., 2005).

When β-carotene is taken up by the intestinal mucosal cell, it is not necessarily secreted into the lymph. Not all of the β-carotene that is taken up by the intestinal cells is immediately packaged into chylomicrons and secreted following the initial meal, and may be package into chylomicrons following later meals containing dietary fats

(Canene-Adams & Erdman, 2009). β-carotene awaiting packaging into chylomicrons may also be lost due to the regular turnover of intestinal cells.

There have been many animal models used to investigate the absorption and metabolism of β-carotene. A recent review by Biehler and Bohn (2010) gives a good overview of the animal models used to study carotenoid absorption and their strengths and limitations. Because rodents have a high efficiency of cleaving β-carotene in the intestines, they do not absorb much intact β-carotene but BCO1–deficient (BCO1-/-) mice have been developed (Hessel et al., 2007), which may better represent the human.

The BCO1-/- mouse is also useful in examining the expression and activity of BCO2 and the possible β-apocarotenoid metabolites (Figure 1.3). Although there are a number of available animal models, human studies are the most appropriate for studying the absorption and metabolism of β-carotene in humans.

1.8 Human Studies

Many human studies have been conducted on the intestinal absorption of carotenoids most of which focus on β-carotene. Although human studies are the best model, there are still limitations to what can be learned from each study. Human studies that use stable isotopes are the best way to study the absorption and metabolism of β-

13 carotene. Liquid chromatography mass spectrometry (LCMS) enables the researchers to distinguish native circulating β-carotene and its metabolites from the newly absorbed dose. The use of stable isotopes as internal standards further optimizes the quantification by adjusting for matrix effects in the samples (Wang et al., 2007).

A plethora of stable isotope human studies have been conducted, most of which have been reviewed (Burri & Clifford, 2004; Furr et al., 2005; Novotny et al., 2005;

Tang et al., 2005; Tang, 2010; van Lieshout et al., 2003) and many have included isotopically labeled β-carotene. Each has examined different aspects of absorption and metabolism and they have different strengths and weaknesses. Stable isotope human studies are costly and complex and samples sizes are inherently low. In most cases, a single pharmacological dose of labeled β-carotene is administered and tracked through the plasma over time. The absorption of β-carotene reported in these studies is highly variable and other studies using unlabeled β-carotene have reported variability in β- carotene absorption (Borel et al., 1998). Peak absorption of β-carotene and its metabolites differs across studies, due to sampling in some cases, but ranges from 3.5-8 hours post-prandially (Burri & Clifford, 2004; Dueker et al., 1994; Dueker et al., 2000;

Pawlosky et al., 2000). Often, the plasma samples are extracted, saponified, and further purified in complex preparations for analysis. Saponification of the samples converts all of the retinyl esters to retinol and does not allow the examination of the individual retinoids. Although retinyl-palmitate is the most abundant retinyl ester, other retinyl esters such as retinyl-linoleate, -oleate and -stearate can contribute to up to 50% of the total retinyl esters. Large variation exists in the conversion of β-carotene to its

14 metabolites where 35-75% of the absorbed β-carotene is converted to retinyl esters in the intestinal cells (During & Harrison, 2006; Edwards et al., 2002; Goodman et al.,

1966; van Vliet et al., 1995). This variability may be due to the different levels of expression of BCO enzymes in different subjects. Leung et al (2009) identified two common single nucleotide polymorphisms in the gene encoding BCO1 that could alter

β-carotene metabolism. Many other factors affect the expression of the BCO enzymes such as vitamin A status because there is a retinoic acid response element (RARE) in the promoter of the BCO gene that may lead to retinoic acid down-regulating the expression at the transcriptional level. The amount of BCO1 expression may be related to the variability of β-carotene absorption.

13 2 Most of the stable isotope human studies use C-β-carotene or [ H8] β-carotene.

Samples are often saponified to eliminate interference of other lipids on the analysis.

Dueker et al. (1994) administered a 73 µmol d8-β-carotene dose to a single human subject and extracted β-carotene and retinol from plasma. They saponified the lipid extract, which includes potassium-hydroxide (KOH) and heat over time, and then it is re-extracted, dried and subjected to a solid phase extraction (SPE). Because the samples were saponified, d4-retinol levels include those hydrolyzed from d4-retinyl-esters. They also further purified β-carotene for MS/MS. Other d8-β-carotene human studies also saponified the plasma samples and detected d8-β-carotene and d4-retinol (Burri & Park,

1998; Novotny et al., 1995; Novotny et al., 2010; Z. Wang et al., 2003). D8-β-carotene human studies that did not saponify the plasma samples still involved rigorous sample preparation and detected only d8-β-carotene (Pawlosky et al., 2000) or d4-retinol (Tang

15 et al., 2000; Tang et al., 2003). It would be useful to have a d8-β-carotene human study with minimal handling of the plasma samples without saponification so it is possible to distinguish between d4-retinol and the d4-retinyl esters. It is also important to have only one sample preparation method for both the retinoid derivatives and β-carotene.

1.9 β-Apocarotenoids

As previously mentioned β-carotene can be eccentrically cleaved by BCO2 at double bonds other than the central 15, 15‟-double bond to give eccentric cleavage products that may have biological activity. β-Apocarotenals have been detected in biological samples by several groups (Barua & Olson, 2000; Ho et al, 2007; Hu et al.,

2006) and they can be formed non-enzymatically from auto-oxidation (Handelman et al.,

1991). The primary product of BCO2 is thought to be β-apo-10‟-carotenal (Kiefer et al.,

2001) but the enzymatic cleavage of β-carotene in vivo needs further confirmation. Ho et al recently showed the probable occurrence of β-apocarotenals in human plasma after the ingestion of β-carotene (Ho et al., 2007). Our laboratory has shown β-apo-13- carotenone to be an inhibitor of RAR which may play a critical role in many biological systems including the down-regulation of the BCO enzymes themselves. Ziouzenkova et al 2007 showed that β-apo-14‟-carotenal represses peroxisome proliferator activated receptors (PPAR) and RXR activation and biologic responses induced by their respective agonists both in vitro and in vivo. β-Apocarotenals and β-apo-13-carotenone have been detected in a few β-carotene containing foods. β-Apocarotenoids may form as a result of degradation and it has been suggested that they could undergo β-oxidation

16 in mitochondria to shorter chain length β-apocarotenoids and retinoic acid (X. D. Wang et al., 1996). Further research into the abundance of β-apocarotenoids in food and biological samples is needed.

1.10 Aims

Aim 1: To compare the β-carotene content of ‘Orange Dew’ melon and a cantaloupe melon grown under the same conditions, and to determine the bioaccessibility and bioavailability of β-carotene.

a. To determine the β-carotene content and its bioaccessibility and bioavailability

from commercially available field-grown orange-fleshed honey dew melons.

b. To determine the chromoplast structure of Orange Dew melons.

c. To determine the levels of β-apocarotenoids in the Orange Dew melons.

My hypothesis is that the β-carotene content of the Orange Dew is similar to that of the cantaloupe.

Aim 2: To determine the extent of correlation between β-carotene absorption and cholesterol absorption in a group of men on the same defined diet. A secondary goal was to define the consistency of the absorption efficiencies of cholesterol and

β-carotene in the same subjects on repeated measures separated by several months.

a. We tested the hypothesis that β-carotene and cholesterol absorption are

correlated.

b. We determined the intra-individual variability of β-carotene absorption and its

conversion to vitamin A.

Aim 3: To determine the existence of β-apocarotenoids from degradation and eccentric cleavage of β-carotene.

a. We tested the hypothesis that BCO1-/- mice have higher plasma and liver levels

of β-apocarotenals than wild-type mice.

b. We determined the concentration of β-apocarotenoids in the plasma of humans

fed β-carotene.

c. We determined the content of β-apocarotenoids in β-carotene containing foods.

The remaining chapters of the dissertation describe the studies conducted to

address the aims described above. Chapter 2 addresses the content and

bioaccessibility of β-carotene, as well as the β-apocarotenoid content in orange-

fleshed melons, as explained in Aim 1. Chapter 2 is a manuscript that has been

published in the journal of Agriculture and Food Chemistry. Chapter 3 describes

the methods used to assess the β-carotene absorption to address Aim 2. Chapter 3 is

currently a methods manuscript in preparation for peer-reviewed publication.

Chapter 4 addresses the results of a β-carotene feeding study using the methods

described in Chapter 3. Chapter 4 is currently a manuscript in preparation for peer-

reviewed publication. Chapter 5 addresses the β-apocarotenoid content of different

18 biological samples described in Aim 3. Lastly, Chapter 6 is the Epilogue and provides a final summary of the findings described in the dissertation.

CHAPTER 2

Carotene and Novel Apocarotenoid Concentrations in Orange-fleshed Cucumis

melo Melons: Determinations of β-Carotene Bioaccessibility and Bioavailability*

* Matthew K. Fleshman, Gene E. Lester, Ken M. Riedl, Rachel E. Kopec, Sureshbabu Narayansamy, Robert W. Curley Jr., Steven J. Schwartz, and Earl H. Harrison. J Agric Food Chem. 2011 May 11;59(9):4448-54

2.1 Abstract

Muskmelons, both cantaloupe (Cucumis melo Reticulatus Group) and orange- fleshed honey dew (C. melo Inodorus Group), a cross between orange-fleshed cantaloupe and green-fleshed honey dew, are excellent sources of β-carotene. Although

β-carotene from melon is an important dietary antioxidant and precursor of vitamin A, its bioaccessibility/bioavailability is unknown. We compared β-carotene concentrations from previously frozen orange-fleshed honey dew and cantaloupe melons grown under the same glasshouse conditions, and from freshly harvested field-grown, orange-fleshed honey dew melon to determine β-carotene bioaccessibility/bioavailability, concentrations of novel β-apocarotenals, and chromoplast structure of orange-fleshed honey dew melon. β-Carotene and β-apocarotenal concentrations were determined by

HPLC and/or HPLC-MS, β-carotene bioaccessibility/bioavailability was determined by in vitro digestion and Caco-2 cell uptake, and chromoplast structure was determined by electron microscopy. The average β-carotene concentrations (µg/g dry weight) for the orange-fleshed honey dew and cantaloupe were 242.8 and 176.3 respectively. The average dry weights per gram of wet weight of orange-fleshed honey dew and cantaloupe were 0.094g and 0.071g, respectively. The bioaccessibility of field-grown orange-fleshed honey dew melons was 3.2±0.3 percent, bioavailability in Caco-2 cells was about 11%, and chromoplast structure from orange-fleshed honey dew melons was globular (as opposed to crystalline) in nature. We detected β-apo-8‟-, β-apo-10‟, β-apo-

12‟-, β-apo-14‟-carotenals and β-apo-13-cartotenone in orange-fleshed melons (at a level of 1-2% of total β-carotene). Orange-fleshed honey dew melon fruit had higher 21 amounts of β-carotene than cantaloupe. The bioaccessibility/bioavailability of β-

Carotene from orange-fleshed melons was comparable to that from carrot (Dacus carota).

2.2 Introduction

Vitamin A deficiency unfortunately affects over 100 million people throughout the world (During & Harrison, 2007). Fruits and vegetables are often utilized as the only treatment for vitamin A deficiencies in human diets as they contain provitamin A carotenoids e.g. β-carotene (Harrison, 2005). β-Carotene is the most potent precursor of vitamin A (Botella-Pavia & Rodriguez-Conception, 2006). An excellent source of β- carotene is orange-fleshed muskmelons: cantaloupe (Cucmis melo Reticulatus Group) and the novel, little studied orange-fleshed honey dew (Cucumis melo, Inodorus group), a cross between cantaloupe and green-fleshed honey dew (Cucumis melo Inodorus

Group) (Lester, 2008). However nothing is known of the bioaccessibility/bioavailability of β-carotene from orange-fleshed melons or of the fruit‟s chromoplast structure.

β-Carotene is lipid soluble and must be incorporated into micelles to be absorbed and bioaccessibility reflects the efficiency of micellerization. Bioavailability represents the amount of the nutrient that is absorbed by the intestinal epithelium and made available for use by the body. β-Carotene bioavailability is affected by chromoplast structure in mature plant tissues, and naturally occurring chromoplasts are globular, tubular, reticulo-tubular, membranous and crystalline types, with globular 22 types providing the best and crystalline types providing the poorest structure for bioavailability of β-carotene (Sitte et al., 1980). Many previous studies have used in vitro digestion and Caco-2 cell uptake to assess the bioaccessibility/bioavailability of β- carotene (Failla et al., 2008; Failla et al., 2009; Garrett et al., 1999; Garrett et al., 1999)

β-carotene can also undergo asymmetric cleavage to yield β-apocarotenoids that have potential biological roles such as vitamin A activity and transcriptional regulation.

The purpose of this study was to 1) determine β-carotene concentrations in previously frozen orange-fleshed cantaloupe and honey dew melons grown under the same glasshouse conditions, 2) determine the bioaccessibility/bioavailability of β- carotene from aforementioned frozen melon tissues and from fresh, field-grown orange- fleshed honey dew melon tissue; 3) determine the morphological features of chromoplasts in the orange-fleshed melon; and 4) report, newly determined, β-apo-13- carotenone, β-apo-14‟-carotenal, β-apo-12‟-carotenal, β-apo-10‟-carotenal, β-apo-8‟- carotenal levels in orange-fleshed melon tissues.

2.3 Materials and Methods

Chemicals and Supplies.

Unless otherwise stated, all chemicals and supplies were purchased from Sigma-

Aldrich (St. Louis, MO, USA) or Fisher Scientific (Pittsburgh, PA, USA).

Plant Material.

Glasshouse grown orange-fleshed honey dew „Orange Dew‟ and cantaloupe

„Cruiser‟ plants, 20 each were grown in Weslaco, Texas in well fertilized potting soil as previously described (Lester, 2005). Upon harvest, following natural abscission

(indicating fruit maturity), melons were washed in tap water, cut and homogenized according to the protocol of Lester (Lester, 2008). Briefly, the two polar ends of the melons were removed and discarded. The outer epidermis and adjacent hypodermal mesocarp tissues were removed. The remaining edible mesocarp tissue devoid of seed cavity inner meoscarp tissue was homogenized in a Waring® blender. The glasshouse grown melon tissue was stored at -80 oC until shipped on dry ice to Columbus, OH for the study, and used to asses dry weights, β-carotene concentration, and bioaccessibility/bioavailability. Five, fresh, field-grown, orange-fleshed honey dew

„Uncle Sam‟ melons, which is the same cultivar as the „Orange Dew‟ melon, were received from The Turlock Fruit Company in Turlock, California for identification of the chromoplast structure and to assess bioaccessibility/bioavailability. Upon arrival from California melon tissues were prepared as described above and stored at 4˚C until further analysis. We also obtained a locally purchased cantaloupe for the detection of β- apocarotenoids.

Compositional analyses.

β-Carotene.

Melon tissue samples (0.5g) were extracted with 5mL HPLC grade hexane (0.1%

BHT), and 200µL of a saturated NaCl solution was added to facilitate phase separation.

Samples were vortexed for 60s and centrifuged at 5000 × g for 5min to separate phases.

The upper layer was collected and the extraction was repeated two times. The combined hexane layer was dried under a stream of nitrogen and resuspended in 1mL of 2:1 isopropanol dichloromethane and filtered thru a 0.22 micron syringe filter into an HPLC vial. β-Carotene standard was dissolved in hexane, the absorbance measured at 450nm, and a dilution series was used to generate a standard curve. The β-carotene was detected using a Waters 600 HPLC Pump with a Waters 996 Photodiode Array Detector and a

Waters 717plus Autosampler set at 10˚C. The separation and quantification was achieved using a Waters YMC C30 reversed phase column (250mm × 4.6 mm) and a two solvent gradient with methanol and methyl t-butyl ether (MTBE) as described in

Thakkar et al., (2007). Dry weights were determined by adding 1g of melon tissue in triplicate to pre-weighed crucibles, drying in an oven overnight and weighing the air temperature crucible plus contents.

β-Apocarotenoids.

Melon samples were analyzed by liquid chromatography mass spectrometry

(LC/MS) for β-carotene, β-apo-8‟-carotenal, β-apo-10‟-carotenal, β-apo-12‟-carotenal,

β-apo-14‟-carotenal, and β-apo-13-carotenone levels. Three 1g samples of fresh, locally

25 purchased, cantaloupe and three 1g samples from a randomly chosen previously frozen

„OrangeDew‟ melon tissue were extracted using a modification of the method of Kane et al. (2005). The first fraction containing neutral lipids was collected and analyzed.

Briefly, 1 mL of 0.025 M KOH was added to the 1g samples and compounds of interest were extracted three times into 10 mL of hexane. The hexane extracts were combined and dried under N2. The dried extracts were subsequently redissolved in a 1:1 (v/v) mixture of MTBE and methanol and 20 µL was injected onto the HPLC for separation of β-carotene and β-apocarotenoids. Separations were accomplished by reversed phase

HPLC using a 4.6 x 150 mm 5 µm YMC C30 column (Waters Corp, Milford, MA). A ternary solvent system consisting of: solvent A, 0.1% formic acid; solvent B, 100% methanol; and solvent C, 100% MTBE was employed. The initial HPLC solvent consisted of 20% A/80% B/0% C and from the time of sample injection through 12 min the solvent was linearly changed to 0% A/30% B/70% C. This was followed by a 3 min re-equilibration period. The flow rate of the solvent was maintained constant at 1.8 mL/min and the column temperature at 35˚C. The UV–VIS absorbance of the eluent was monitored using a Waters 996 photodiode array detector.

The HPLC eluate was interfaced with a quadrupole/time-of-flight mass spectrometer (Q Tof Premier, Micromass, UK) via an atmospheric pressure chemical ionization (APcI) probe. The HPLC/MS method used to detect apocarotenoids was used as previously described in Shmarakov et al 2010 with one modification. β-Apo-13- carotenone was measured in APcI positive mode which required a second 20µL injection of the same sample. β-Carotene, β-apo-8‟-carotenal, β-apo-10‟-carotenal, β-

26 apo-12‟-carotenal, and β-apo-14‟-carotenal ionized in APcI negative mode and β-apo-

13-carotenone ionized in APcI positive mode and were detected as their respective radical ions, at m/z of 536.438, 416.31, 376.28, 350.26, 310.23, and 259.2. The QTof system allowed for quantitative detection with the confidence of accurate mass

(typically 1ppm). Mass spectra were acquired in V-mode (~8000 resolution) from 100-

1000 m/z with a scan time of 0.5s, peak centroiding, and enhanced duty cycle enabled for the parent m/z. At intervals of 30s, a 0.1s lockspray scan was acquired with leucine enkephalin as the lockspray compound (554.2615 m/z) to correct for minor deviations in calibration due to temperature fluctuations. Prior to analysis, the QTof was fully calibrated from 114 to 1473 m/z using a solution of sodium formate. The resultant MS spectra were acquired and integrated with MassLynx software, V4.1 (Micromass UK,

Manchester, UK). Source parameters included: 30 µA corona current, 500 °C probe,

110 °C source block, 35 V cone, 50 L/hr cone gas (N2), 400 L/hr desolvation gas (N2), collision energy 8 eV (non-fragmenting) with argon CID gas (4.2x10-3 mBar). Standard curves were generated using standards of β-carotene (CAS# 7235-40-7), β-apo-8‟- carotenal (CAS # 1107-26-2) (Fluka), β-apo-12‟-carotenal (CAS # 1638-05-7)

(CaroteNature, Lupsingen, Switzerland), β-apo-14‟-carotenal (CAS # 6985-27-9)

(synthesized for the study), and β-apo-13-carotenone (CAS # 17974-57-1) (synthesized for the study). The purity of β-carotene, β-apo-8‟-carotenal, β-apo-12‟-carotenal, β-apo-

14‟-carotenal, and β-apo-13-carotenone by HPLC is 95%, 96%, 99%, 94%, and 98% respectively. β-Apo-10‟-carotenal was quantified using the standard curve for β-apo-

12‟-carotenal. Recoveries for all compounds were determined to be greater than 98%.

Chromoplast Isolation.

The protocol for cantaloupe chromoplast enrichment and subsequent preparation for microscopy was provided through personal communication with Dr. Wayne W. Fish,

South Central Agriculture Research Laboratory, Agricultural Research Service, U.S.

Department of Agriculture (Fish, 2006). Fresh orange-fleshed honeydew mesocarp tissue (1000g) and 1000 mL water were mixed in a Waring® blender for 4 min and filtered through parachute cloth. After two centrifugations (30000 × g, 1hr, 4oC) the top layer of the pellet containing the chromoplasts was scraped and suspended in 15.4 mL water with added sodium ascorbate 10:1 (w/w) and 0.02% sodium azide. Two mL of chromoplasts solution were centrifuged to remove buffer and 5 mL 0.05M sodium phosphate, pH 7.2 + 0.3% SDS was added and mixed. Solubilized chromoplasts were fixed in gluteraldehyde. Acetonitrile was added to precipitate the chromoplasts (Fish,

2007) which were washed with 5 mL phosphate buffer. After the third wash, 4 mL of

0.05M sodium phosphate buffer pH 7.2 + 0.3% SDS + 0.02% sodium azide were added and mixed. Chromoplasts were prepared for microscopy and identified according to

Vaäsquez-Caicedo et al. (2006) by transmission electron microscopy at The Ohio State

University Microscopy and Imaging Facility.

In Vitro Digestion.

In vitro digestion was performed on five fresh and randomly chosen greenhouse grown melons samples, in triplicate to determine the percent digestive stability and the percent micellerization according to the protocol of Thakkar et al. (2007), without the

“oral phase”. Briefly, this involved a “gastric phase” where the pH of the melon homogenate is adjusted to 2.5±0.1, pepsin is added at 40mg/mL and the mixture is incubated in a shaking water bath at 37˚C for 1hr. In the subsequent “intestinal phase” the pH is adjusted to 6.5±0.1, porcine pancreatic lipase, pancreatin, and bile extract is added and the mixture is incubated in a shaking water bath at 37˚C for 2hr. The micelle fraction is then isolated from the digesta by centrifugation at 5000 × g for 45 min at 4˚C and filtration (0.22 mm pore size) of the collected aqueous (micelle) fraction (Thakkar et al., 2007). The aqueous fractions were then applied to Caco-2 cells as described by

Chitchumroonchokchai et al. (2004).

Caco-2 Cells.

Stock cultures of Caco-2 (HTB-39) cells were obtained from American Type

Culture Collection and were maintained as previously described

(Chitchumroonchokchai et al., 2004). The Caco-2 human cell line exhibits characteristics of mature enterocytes (Ellwood et al., 1993). T75 flasks of Caco-2 cells were grown 10-14 days post confluency. Following in vitro digestion, the aqueous fractions containing the micelles were collected and each diluted 1:4 with Dulbecco„s minimum essential medium (DMEM) and 12.5 mL of the medium was added to each flask. At the end of 4 hours the medium was collected and cells were washed with ice cold phosphate-buffered saline (PBS) with albumin, which was also collected and combined with the medium. The cells were washed twice with ice cold PBS and the wash was discarded. 10mL of ice cold PBS was then added to each flask, the cells were scraped, collected, and process repeated. The collected cells in PBS were centrifuged at 29

2000 × g 4˚C for 5 min and the supernatant was discarded. The cell pellet was resuspended in 2mL PBS and extracted for HPLC analysis. 100µL of each of the resuspended cell pellets was used for a protein assay according to Bradford (Bradford,

1976). Aliquots of the whole digestion, aqueous (micelle) fraction, fresh 1:4 media, spent media, and the cells were extracted with tetrahydroflourene (THF) and hexane.

Briefly, 2mL of THF was added to 2mL of sample and vortexed, then 3mL of hexane was added and vortexed, and centrifuged at 5000 × g for 5min to separate phases. The upper layer was collected and the extraction was repeated two times. Extracts were dried under a stream of nitrogen and redisolved in 2:1 isopropanol dichloromethane and filtered thru a 0.22 micron syringe filter and injected into the HPLC. HPLC analysis was performed with a Waters 1525µ Binary HPLC Pump with a Waters 996 Photodiode

Array Detector and a Waters 717plus Autosampler set at 10˚C. A YMC Carotenoid

5µm particle (4.6 × 150mm) column with a YMC Carotenoid 5µm particle (4.0 × 20mm)

Guard Cartridge was used. Separation was achieved by gradient elution with a binary mobile phase of methanol-0.1%(v/v) formic acid (FA) as Solvent A (80:20) and MTBE- methanol-0.1% FA as Solvent B (78:20:2) at a flow rate of 1.8mL/min. Initial conditions were held at 100% A for 1 min then a linear gradient to 40:60 A:B over 5 min, followed by a linear gradient to 100% B over 9 min, a linear gradient back to 100%

A for 1 min, and held at 100% A for 4 min for a final chromatographic run time of 20 min. Identification and quantification of the compounds of interest was accomplished by comparison with synthetic standards run in a dilution series before and after the samples.

Statistical Analysis

All data is presented as averages and standard deviations. Comparisons of glasshouse grown orange dew and cantaloupe for dry weight and β-carotene content were made using a Student‟s t-test.

2.4 Results and Discussion

Orange-fleshed Honey Dew Melon and Cantaloupe:

Dry Weights and β-carotene Content.

Orange-fleshed honey dew „Orange Dew‟ had significantly higher dry weight (P

<0.001) than hybrid orange-fleshed cantaloupe „Crusier‟ (Table 2.1). The difference in dry weight may suggest orange-fleshed honey dew „Orange Dew‟ melons have a different matrix than the cantaloupe „Cruiser‟. β-Carotene was the only carotenoid detected at 450 nm by HPLC in either orange-fleshed honey dew or cantaloupe, and greater than 98% was the all-trans isomer (Bohm et al., 2002) (Figure 2.1).

Since dry weights were determined to be significantly different for orange- fleshed honey dew and cantaloupe, β-carotene concentrations per gram of fresh (wet) weight were divided by the dry weight for each sample to correct for fresh weight differences. The mean β-carotene content for orange-fleshed honey dew was significantly greater (P <0.001) than cantaloupe (Table 2.1).

The average dry weight of fresh orange-fleshed melons „Uncle Sam‟ was similar to dry weights determined for frozen orange-fleshed melon samples (Table 2.1). The average concentration of all-trans β-carotene of fresh orange-fleshed honey dew melon was similar to that of frozen orange-fleshed honey dew samples (Table 2.1).

Chromoplast Structure.

The chromoplast is a plant compartment in which most of the carotenoids accumulate (Vishnevetsky et al., 1996). Chromoplast structure, globular vs. crystalline, has been reported to affect β-carotene bioaccessibility, with a globular structure providing the superior matrix for β-carotene bioaccessibility (Sitte et al., 1980).

Chromoplasts were isolated from fresh melon tissue and imaged by TEM as described above. The TEM image in Figure 2.2 shows that melon chromoplasts were globular in structure as opposed to the crystalline matrix found in carrots (Sitte et al., 1980; Zhou et al., 1996).

Bioaccessibility and Bioavailability of β-carotene from Orange-fleshed Melons.

Fresh orange-fleshed honey dew melon homogenates were subjected to simulated gastric and intestinal phases of in vitro digestion (Table 2.2). The percent digestive stability is the percent of β-carotene present in the digesta after simulated digestion. The digesta was separated into the aqueous fraction which contained micelles.

The percent micellerization (bioaccessibility) was the percent of β-carotene that was incorporated into the micelles. In vitro digestion was performed in triplicate for each of five fresh melon fruit. The average percent digestive stability was 63.8±3.6% and the

32 average percent micellerization (bioaccessibility) of β-carotene from orange-fleshed edible melon tissue was 3.2±0.3%. Percent micellerization was corrected for loss during digestion, i.e., the percent of β-carotene digesta that was micellerized (Table 2.2).

These percentages are consistent with results from previous in vitro digestions with glasshouse grown melons where the digestive stability and the percent micellerization (β-carotene bioaccessibility) did not differ between orange-fleshed honey dew and cantaloupe melons (results not shown). The values for digestive stability are consistent with previously published results in maize (Zea mays), cassava (Manihot esculenta), drumstick (Moringa oleiflora) leaves, carrots (Daucus carota), green leafy vegetables, mangoes (Mangifera indica), sweet potatoes (Ipomoea batatas), squash

(Cucurbita moschata), and pumpkins (Cucurbita maxima) (Hedren et al., 2002; Kean et al., 2008; Mulokozi et al., 2004; Priyadarshani & Chandrika, 2007; Pullakhandam &

Failla, 2007; Thakkar et al., 2007; Thakkar & Failla, 2008). There are many factors that affect the intestinal absorption of carotenoids such as the type and quantity of dietary fat, competition among co-consumed carotenoids, the plant or food matrix (Lemke et al.,

2003), and cooking. We tested the effects of adding 2.5-3% vegetable oil to the in vitro digestion of the melons but there was no significant difference in the extent of micellerization. Since melons are typically eaten raw we did not test the effects of cooking. The bioaccessibilities of β-carotene reported for fruits and vegetables are mostly for cooked foods and, thus, the „lower‟ bioaccessibility of β-carotene in melons may be only relative. Although the bioaccessibility of β-carotene in melons observed in this study is lower than for other fruits and vegetables the reasons appear to be

“physiological” in the sense that melon is eaten raw and not cooked and melons have very high water content. In comparison with carrot, the bioaccessibility of β-carotene of 3.2% found in our melon samples are similar to bioaccessibilities found in raw carrots (Hedren et al., 2002) which was confirmed in a separate in vitro digestion assay using locally purchased carrots (results not shown). Carrots in general have 82 g/g fresh weight β-carotene (U.S. Department of Agriculture, Agricultural Research Service,

2010) but are on average 21 % dry wt. (Lester et al., 1982). Whereas our orange-fleshed honey dew melons samples averaged 22 g/g fresh weight β-carotene, nearly four-fold less β-carotene than carrot, and averaged 9% dry weight. Bioaccessibility comparison of raw carrots and melon at 3.2% on a dry weight basis yields a total bioaccessibility of 12

g β-carotene/g dry wt. and 8 g β-carotene/g dry wt. for carrot and orange-fleshed melons respectively, which are nearly the same. This suggests that the globular chromoplasts of melons appear to provide a more efficient matrix which leads to a higher bioaccessibility of β-carotene than the crystalline chromoplast matrix of carrot

(Mulokozi et al., 2004).

In a separate experiment, Caco-2 cells were used to assess the bioavailability of melon β-carotene. Fresh orange-fleshed honey dew melon underwent in vitro digestion and the diluted micelle fractions were applied to Caco-2 cells. The average uptake of β- carotene by the Caco-2 cells after 4h of incubation (i.e., bioaccessibility) was 11.6% of the β-carotene applied to the cells (results not shown). This is consistent with the uptake of β-carotene in other studies with other fruits and vegetables (De Jesus Ornelas-Paz et al., 2008; Thakkar et al., 2007). Results obtained with fresh orange-fleshed melon were 34 consistent with values obtained from frozen, glasshouse grown melons where there was no significant difference between -carotene of orange-fleshed honey dew and cantaloupe melons for percent digestive stability, micellerization, and Caco-2 cell bioavailability (results not shown).

β-Apocarotenoids in Melons.

We measured the levels of β-apo-13-carotenone, β-apo-14‟-carotenal, β-apo-

12‟-carotenal, β-apo-10‟-carotenal, β-apo-8‟-carotenal, and β-carotene of a locally purchased cantaloupe and a greenhouse grown OrangeDew (Figure 2.3). The β- apocarotenoids in total are present at approximately 1.5% of the level of β-carotene.

Figure 2.4 shows the relative concentration of the β-apocarotenoids. The β- apocarotenoids profiles of the two different types of melons are approximately the same.

β-apocarotenoids are thus present in foods at low levels and may also be formed enzymatically in mammals (Chitchumroonchokchai et al., 2004) where they may have biological activities mediated in part by their interaction with retinoid receptors (Eroglu et al., 2010). We are currently assessing the β-apocarotenoid levels of other β-carotene rich fruits and vegetables with an improved LC-MS method, in addition to assessing the bioaccessibility/bioavailability. β-Apocarotenoids may be more bioavailable than β- carotene due to shorter chain length and greater hydrophilicity.

The orange-fleshed honeydew melons are known to be safer for the consumer because they lack rough outer netting, unlike cantaloupe fruit, which harbors enteric bacteria, and are sweeter and store longer than the typical cantaloupe melon (Flores,

2007). In our study orange-fleshed honey dew had significantly higher dry weight and

β-carotene concentrations than cantaloupe, but each cultivar had similar β-carotene bioaccessibility. Thus, both appear to be comparable sources of dietary provitamin A for humans and on par with carrots, another major source of provitamin A.

2.5 Acknowledgements

We thank Ms. Vanessa Reed for expert technical assistance, Dr. Wayne Fish for sharing details of the isolation procedure for chromoplasts, and Dr. Mark Failla for assistance with in vitro digestions and Caco-2 cells.

2.6 Figures and Tables

0.30

0.20 AU

0.10

0.00 5.00 10.00 15.00 20.00 25.00 30.00 Minutes

Figure 2.1. HPLC chromatogram (at 450 nm) of lipid extracts of orange-fleshed honey dew melon tissues. Cantaloupe melons had identical chromatograms with a peak at retention time of 26 min identified as all-trans-β-Carotene by comparison with authentic standard.

Table 2.1. Dry Weight Percent and β-Carotene Concentration of Glasshouse

Grown Frozen and Field-Grown Fresh Orange-Fleshed Honeydew and

Cantaloupe Melon Edible Mesocarp Tissues.a

Orange-fleshed Tissue Dry weight β-carotene melon type preparation % µg/g fresh wt. µg/g dry wt. Orange dew Frozen 9.4+0.9 *b 22.8+2.4*b 242.3+25.7*b Cantaloupe Frozen 7.1+0.8 12.5+3.8 176.3+54.0 Orange-fleshed honey Fresh 9.8+0.9 21.0+2.6 213.9+33.5 dew aN=20 for orange dew and cantaloupe and N=5 for orange-fleshed honey dew. Means + standard deviations are given. b*Dry weight and β-carotene content of the frozen melons were compared using Student‟s t test. For these parameters „Orange Dew” were significantly higher than cantaloupe (P≤0.001).

Figure 2.2. TEM image of orangedew chromoplast. The dark spheres shown in the image are the chromoplast that contain β-carotene and are globular in structure.

Magnification of 18500x. The box shows an enlargement of the large chromoplast.

Table 2.2. β-Carotene Concentration from Orange-Fleshed Melon in Whole Edible

Tissue, Digesta (Digestive Stability) and Aqueous Fractions (Micellerization or

Bioaccessibility) Following In Vitro Digestion (N=5)a

β-carotene Stage of Percent of Fraction Digestion µg/g fresh wt. µg/g dry wt. whole tissue Whole Whole food 20.98±2.67a 213.9±33.5 100 tissue Digestive Digesta 13.35±1.51 135.7±15.5 63.8±3.6 stability Aqueous Bioaccessibility 0.43±0.03 4.36±0.43 3.1±0.3 fraction a Values are Shown as means + standard deviation.

Figure 2.3. β-Apocarotenoids and β-carotene in the „OrangeDew‟ melons (A) and standards (B). In panel A the top chromatogram shows the absorbance at 452nm and the other chromatograms show the apocarotenoids at their respective masses. β-Apo-10‟- carotenal standard was not available and was quantified using the standard curve for β- apo-12‟-carotenal.

Cantaloupe 60.00

50.00

40.00 13-one 14'-al

30.00 12'-al /g wet wt. wet /g 10'-al

pmole 20.00 8'-al

10.00

0.00 OrangeDew 60.00

50.00

40.00 13-one 14'-al

/g wet wt.wet /g 30.00 12'-al 10'-al

pnmole 20.00 8'-al

10.00

0.00

Figure 2.4. Average β-apocarotenoid levels in two different orange-fleshed melons.

Values are from triplicate 1g extractions of each melon with error bars showing SD.

CHAPTER 3

An HPLC/MS Method for the Detection of d8-β-Carotene and Individual Fatty

Acyl Esters of d4-Retinol: Application to the Study of the Intestinal Absorption of

β-Carotene and its Conversion to Vitamin A in Humans.

3.1 Abstract

The absorption and metabolism of β-carotene is of vital importance in humans especially in parts of the world that obtain the majority of their vitamin A from provitamin A carotenoids. Recent advances in mass spectrometry have allowed for the better understanding of the absorption of β-carotene, the most potent pro-vitamin A carotenoids, through the use of stable isotopes of β-carotene. Previous studies have estimated the absorption of a dietary dose of labeled β-carotene but have neglected the conversion of newly absorbed β-carotene to individual retinyl esters within the intestinal mucosal cells. We have developed a method that eliminates the need for rigorous sample preparation and allows us to detect and quantify newly absorbed d8-β- carotene as well as its d4-retinyl ester metabolites. We employed a simple sample extraction for both retinoids and β-carotene that included minimal sample handling.

Analysis of d8-β-carotene and its d4-retinyl ester metabolites used the same liquid chromatography system and solvents for separation of the compounds. The method allowed us to give an accurate measure of d8-β-carotene absorption as well as its extent of conversion to d4-retinyl esters.

3.2 Introduction

Vitamin A deficiency is one of the most prevalent vitamin deficiencies throughout the world that affects over 100 million people and occurs in regions where people obtain most of their vitamin A as provitamin A carotenoids. A better 44 understanding of the mechanisms of absorption and metabolism of provitamin A carotenoids could help alleviate nutritional problems associated with vitamin A deficiency (During et al., 1996).

The absorption of β-carotene involves its release from the food matrix, incorporation into mixed lipid micelles, uptake by the intestinal mucosal cells, packaging into chylomicrons, and secretion into the lymph. The intestinal uptake of β- carotene may involve facilitated uptake by transport proteins such as scavenger receptor class B, type I (SR-B1), cluster determinant 36 (CD36), and Niemann-Pick type C1

Like 1 (NPC1L1). Once β-carotene is taken up by the intestinal mucosal cell it is either cleaved by β-carotene oxygenase 1 (BCO1) or it is absorbed intact. The β-carotene and its metabolites are incorporated into chylomicrons (CM) along with cholesterol esters, triglycerides, apolipoprotein B and phospholipids. The chylomicrons are then secreted into the lymphatic system and enter the blood via the thoracic duct. Chylomicron remnants are then taken up by the liver and other tissues where β-carotene and its metabolites are stored and further metabolized.

In most human studies of β-carotene absorption, subjects have been dosed with

13 2 stable isotopes of β-carotene such as C-β-carotene or [ H8] β-carotene (d8-βC) in order to distinguish the dietary dose from endogenous species. Samples are often saponified to eliminate interference of other lipids on the analysis. We have developed an LC/MS method to study the absorption and metabolism of d8-β-carotene in humans.

The method has been applied to the analysis of whole plasma or CM-rich plasma fraction for d4-retinol, individual d4-retinyl esters, and d8-β-carotene, as well as native

45 species. The analytical procedure involved minimal sample handling and avoids saponification allowing us to distinguish between d4-retinol and individual d4-retinyl esters. We also have only one sample preparation method for both the retinoid derivatives and β-carotene with minimal steps and good recovery of all compounds of interest. Our method allowed us to separate and quantify four retinyl esters (retinyl stearate, linoleate, oleate, and palmitate) and we were able to detect both the native esters (unlabeled) and the d4 esters in the chylomicron fraction. The method also allowed us to accurately determine the conversion of β-carotene to vitamin A metabolites for each subject. The method can be used for high throughput quantification of the absorption of β-carotene and conversion of β-carotene to its vitamin A metabolites in human blood samples, and may also be useful for the detection of the retinoids and carotenoids in other biological samples.

3.3 Materials and Methods

Chemicals and Solutions

Unless otherwise stated, all chemicals and supplies were purchased from Sigma-

Aldrich (St. Louis, MO, USA) or Fisher Scientific (Pittsburgh, PA, USA). D8-β- carotene was purchased from Cambridge Isotope (Andover, MA) and had 83% all-trans and 17% cis isomers. Isotopomer distribution of d8-β-carotene was 79.44% d8, 17.59%

13 d7 & 2.96% d6. C40-β-carotene was purchased from Martek (Columbia, MD) and d4- retinyl acetate was purchased from Cambridge Isotope (Andover, MA). Retinyl-

46 linoleate, -oleate, and -stearate were synthesized by Dr. Robert Curley (College of

Pharmacy, Ohio State University).

Subjects and Samples Collection

Subjects were recruited from a human study that involved measurement of their cholesterol absorption to participate in a β-carotene absorption study. All procedures were approved by the Johns Hopkins Institutional Review Board, and subjects gave written informed consent before participating. All subjects were male smokers and were on a defined diet. Subjects were given a 5mg bolus dose of deuterated β-carotene (d8-

βC) in 8mL of corn oil. Subjects were catheterized and 8mL of plasma was drawn every hour for nine hours. Four mL of the plasma was separated into a chylomicron (CM) fraction and a non-CM fraction by ultra-centrifugation (Edwards et al., 2001). All aliquots of plasma fractions were stored at -80˚C until analysis.

Sample Extraction

One mL of the CM fraction was transferred to an 11mL glass tube for extraction.

13 Internal standards (IS; retinyl acetate and C40-βC) were added, followed by 1mL of ethanol with 0.1% (w/v) BHT. Next 5mL of HEAT (10:6:7:7

Hexane/Ethanol/Acetone/Toluene) was added and a saturated NaCl solution (200µL) to facilitate phase separation. Samples were vortexed for 60s and centrifuged at 5000 x g for 5 minutes at 4˚C. The upper organic layer was removed and the extraction repeated twice more with HEAT/NaCl, combining the three organic layers and drying them under a stream of nitrogen. The residue was reconstituted in 300µL of 1:1 MtBE/MeOH

47 with water bath sonication and filtered through 0.45µm nylon filter before HPLC injection. An aliquot of whole plasma (1.8 mL) was extracted using the same protocol.

HEAT was determined to be the most efficient extraction solvent for both retinoids and β-carotene when compared to several published extraction methods for retinoid and carotenoid extraction. In all extraction methods tested we added equal volume of ethanol to the plasma/CM-rich fraction and varied the extraction solvents used (i.e. hexane, 2:1 hexane acetone, chloroform, etc.) and the addition of salt, acid, or no addition to facilitate phase separation. HEAT with the addition of NaCl had the best recovery and consistency of recovery of all compounds of interest. Although the addition of acid provided better recovery for retinoic acid, the addition of NaCl proved to be best for other retinoids, particularly retinol and the retinyl esters. Other extraction solvents showed less matrix suppression around the β-carotene peak but this was easily remedied by the use of a stable isotope as an internal standard.

HPLC

Reversed phase chromatography with a C30 4.6x150mm, 5µm column was used to separate retinoids and carotenoids. We used two separate chromatographic separations, one for retinol and retinyl esters and one for β-carotene. Each run utilized the same binary solvent systems: (A) 80:20 MeOH/0.1% formic acid (FA) and (B)

78:20:2 MtBE/MeOH/0.1%FA at 1.8mL/min and a column temperature of 40˚C. UV- vis spectra of eluent were monitored by photodiode array (Waters Acquity PDA). The retinoids separation involved a 25µL injection with an initial condition of 100 %A, a linear gradient to 100 %B over 6min, held at 100 %B for 2 min then returned to

100 %A for a total run time of 10min. The carotenoids LC method had a 50µL injection starting with 100 %A and a linear gradient to 100 %B over 16min, returning to 100 %A for a total run time of 18min. For a given sample, the two separations were run consecutively.

Mass Spectrometry

The HPLC eluate was interfaced with an atmospheric pressure chemical ionization (APcI) probe of a triple quadrupole mass spectrometer (Quattro Ultima,

Micromass, UK). For the retinoid MS/MS method, retinol, retinyl acetate and retinyl esters ionized in APcI positive mode as the M+H-H2O parent ion (269 m/z) and were fragmented by collision induced dissociation to corresponding product ions 93 and 107 m/z. Co-migration with authentic standards confirmed identification. The d4 retinol and d4 retinyl esters were monitored as MS/MS transitions m/z 273>94, 217. The β- carotene HPLC-MS analysis utilized selected ion reaction (SIR) mode with m/z 536.4

13 for native β-carotene, m/z 544.4+543.4 (d8+d7) for d8-βC, and m/z 576.4 for the C40-

βC (β-carotene IS). It was necessary to have two separate injections from the same extract, one for retinoids and one for β-carotene, to allow the MS to run in separate modes of ionization. It is possible to switch modes during a single run but there was not enough separation between the REs and β-carotene in our initial chromatographic separation to successfully switch modes without sacrificing run time and peak quality.

APcI+ and SIR are the best modes of ionization for retinol/REs and β-carotene respectively

Standard curves were generated by mixing solutions with varying concentrations of authentic standards with fixed amounts of IS (retinyl acetate for retinol and retinyl-

13 esters; C40-βC for β-carotene). Standard curves for d4-retinyl acetate and d8-β- carotene are shown in Figure 3.1 and the internal standards and calibrants used are shown in Table 3.1. D4-retinyl palmitate was calculated from the d4-retinyl acetate standard curve and multiplied by the relative PDA response of retinyl palmitate to retinyl acetate standards. Likewise the other retinyl esters were calculated in the same manner with their response normalized to retinyl palmitate.

3.4 Results

Blood was collected hourly for 9 hours, and enrichment of chylomicrons with d8-β-carotene and d4-retinyl esters (d4-RE) was determined by HPLC-MS. Figure 3.2 shows the structure of the native and labeled compounds and Figures 3.3 and 3.4 are representative chromatograms of stable isotopes of retinyl esters and β-carotene in the

CM-rich fraction at six hours in one subject.

Concentrations were initially determined per mL of CM-rich fraction and the volume of the CM-rich fraction was measured. Total plasma volumes were calculated for each individual and the final concentration is reported in moles per 1mL of plasma.

Figure 3.5 shows concentration of d8-β-carotene and d4-retinyl esters in the CM-rich faction versus time for a single subject.

We repeated the experiment for the same individuals using whole plasma aliquots instead of CM-rich fractions to validate our measurements and to look for d4- retinol. As seen in Figure 3.6 and Table 3.2 the results using the whole plasma were similar to the results using the CM-rich fraction. We had the capability of detecting d4- retinol with our method regardless of source of the sample, but d4-retinol was not detected in the CM fraction for this or any other subjects. However, we were able to detect d4-retinol in whole plasma starting at hour 6 as d4-REs concentration peaked and were cleared by the liver and d4-retinol was retro-transported into circulation. Newly absorbed β-carotene may be cleaved within the enterocyte and free retinol may be absorbed across the bilayer, but we were unable to detect free d4-retinol at the early time points (During & Harrison, 2006). Figure 3.6 shows the whole plasma plots for one individual where it can be seen that d4-retinol levels increase as d4-REs levels decrease. Therefore, to determine the conversion of d8-β-carotene to vitamin A we summed the total d4-REs AUC as a representation of d8-β-carotene that was absorbed as metabolites, excluding the formation of d4-retinol, which was small compared to d4- retinyl esters.

Other studies that saponify samples may over estimate the rate of conversion because of the appearance of circulating d4-retinol at the later time points. Figure 3.7 shows an individual subject whose native REs is overlaid with the newly absorbed d4-

REs; the curves/profiles are almost identical. This was often the case across individuals and at every time point. Individuals who repeated the study showed similar RE profile, whereas different individuals showed different RE profiles.

From these CM fraction plots we were able to calculate the area under the curve

(AUC) to give surrogate levels of absorption of the labeled dose of β-carotene (Table

3.2). The central cleavage of one mole of β-carotene by BCO1 gives two moles of retinal which are reduced to retinol and esterified to give two moles retinyl esters. Thus, an index of the absorption of the d8-β-carotene dose is represented as:

(Total d4-RE AUC/2) + d8-βC AUC = Total d8-βC equivalents AUC

((µmol/L)*hr)

We were also able to estimate the amount of newly absorbed β-carotene that was converted to its retinyl ester metabolites.

(Total d4-RE AUC/2) / Total d8-βC equivalents AUC * 100 = % Conversion

This particular subject had about a 46% conversion of β-carotene to its retinyl ester metabolites which is in the middle of the range of percent conversion.

According to standards we were able to detect d8-β-carotene, d4-REs, and d4- retinol at 8.07E-9 M, 8.70E-8 M, and 7.83E-9 M respectively. The LOD/LOQ varied slightly from day to day for standards and for different individual subjects run within the same days. In several samples we were able to detect analytes in the sub-nM (1E-13 mol/ml whole plasma) concentration range.

3.5 Discussion

There have been a plethora of stable isotope studies of β-carotene absorption that have been reported (Burri & Clifford, 2004; Furr et al., 2005; Novotny et al., 2005;

Tang, 2010; van Lieshout et al., 2003). Each has examined different aspects of absorption and metabolism and they have different strengths and weaknesses. Stable isotope studies with human subjects are costly and complex and samples sizes are inherently low. In most cases a single pharmacological dose of labeled β-carotene is administered and tracked through the plasma over time. The absorption of β-carotene reported in these studies is highly variable and other non-isotope studies have reported on the variability of β-carotene absorption (Borel et al., 1998). Often, the plasma samples are extracted, saponified and further purified in complex preparations for analysis. Saponification of the samples converts all of the retinyl esters to retinol and does not allow the individual retinoids to be examined. Although retinyl palmitate is the most abundant retinyl ester, other retinyl esters such as retinyl-linoleate, oleate and stearate can contribute to up to 50% of the total retinyl esters. There is a large variation in the conversion of β-carotene to its metabolites where 35-75% of the absorbed β- carotene is converted to retinyl esters in the intestinal cells (During & Harrison, 2006;

Edwards et al., 2002; Goodman et al., 1966; van Vliet et al., 1995). Dueker et al (1994) administered 73 µmol d8-β-carotene to a single human subject and extracted β-carotene and retinol from plasma. They saponified the lipid extract, which involves reactions with potassium-hydroxide (KOH) and heat over time, and then it is re-extracted, dried and subjected to a solid phase extraction (SPE). Because the samples were saponified, d4-retinol levels included free d4-retinol and d4-retinyl-esters. They also further purified β-carotene for MS/MS. Other d8-β-carotene human studies also saponified the plasma samples and detected only d8-β-carotene and d4-retinol (Burri & Park, 1998;

Novotny et al., 1995; Novotny et al., 2010; Z. Wang et al., 2003). Human studies using d8-β-carotene that did not saponify the plasma samples still involved rigorous sample preparation and detection of only d8-β-carotene (Pawlosky et al., 2000) or d4-retinol

(Tang et al., 2000; Tang et al., 2003). The current method allows accurate quantitation of the absorption and conversion of stable isotopically labeled β-carotene while avoiding complex sample preparation and provides shorter chromatographic run times.

The method allows for the distinction between d4-retinol and d4-REs. This is useful in understanding the metabolism of newly absorbed β-carotene.

The quantification of β-carotene is often hindered by the presence of other lipids in biological samples, generally referred to as the sample matrix. This matrix issue is often remedied by sample saponification but this process results in the loss of REs and their conversion to retinol. We did not saponify and accounted for the variability in sample matrix with the use of a stable isotope for an internal standard, 13C labeled β- carotene. 13C β-carotene has an additional 40 mass units which is easily differentiated from native and d8-β-carotene and still behaves identically to the other forms of β- carotene on column making it the ideal internal standard. The matrix issue was not a major factor in the quantification of the REs, thus retinyl acetate was a suitable internal standard for all retinoids.

3.6 Acknowledgements

We thank Ms. Rachel E. Kopec with here expert assistance and training in the method development and Dr. Robert W. Curley for the synthesis of retinyl-esters.

3.7 Figures and Tables

Figure 3.1. Standard curves were run the same day as samples and concentrations were quantified as a ratio to the internal standard. Internal standards used were retinyl acetate

13 and C40-β-carotene for retinoids and β-carotene respectively.

Table 3.1. Standards and Internal Standards

Calibrant Internal Standard

Retinol Retinol Retinyl Acetate

Retinyl Esters Retinyl Esters Retinyl Acetate

13 β-Carotene β-Carotene C40-β-Carotene

d4-Retinyl Esters d4-Retinyl Acetate Retinyl Acetate

13 d8-β-Carotene d8-β-Carotene C40-β-Carotene

Figure 3.2.The structure, chemical formula, and exact mass of the labeled and native β- carotene and retinol. Native and labeled retinyl esters were also detected in our samples.

Figure 3.3. The four chromatograms, from top to bottom, show the absorbance at

452nm, 13C-β-carotene stable isotope, deuterium labeled β-carotene, and native β- carotene and the lycopene isomers. These chromatograms are from the CM fraction at hour 6 of one of the subjects.

Figure 3.4. Retinyl linoleate (RL), retinyl oleate (RO), retinyl palmitate (RP), and retinyl stearate (RS). The top chromatogram is the absorbance at 325nm, the middle chromatogram is the native retinyl esters, and the bottom chromatogram is the d4 retinyl esters. In extracts of human plasma from a subject consuming βC-d8.

Figure 3.5. The top panel shows the total d4-retinyl esters and d8-βC in the CM-rich fraction. The bottom panel is the individual d4-retinyl esters and d8-βC for the same subject. Individual retinyl esters followed the same trend as the total retinyl esters and

β-carotene with the peak concentrations of both d4-RE and d8-βC at 5-6 hrs post-dose.

Figure 3.6. The upper panel is from the experiment using whole plasma and the bottom plot is from using the CM-rich fraction of the plasma. There was no d4-retinol detected in the CM-rich fraction. Retinyl ester curves are represented by retinyl palmitate in both panels.

Figure 3.7. A representative chromatogram of d4-REs and native REs overlaid. The curves are normalized to the highest value and are nearly identical traces. The green line is the native REs and the red line are the d4-REs.

Table 3.2. Area Under the Concentration-Time Curves for d8-β-Carotene- and d4-

Retinyl-Esters in Whole Plasma and Chylomicrons

Plasma d4-RE Plasma d8-βC Chylo d4-RE Chylo d8-βC

AUC AUC AUC AUC Subject ID {(µmol/L)*hr} {(µmol/L)*hr} {(µmol/L)*hr} {(µmol/L)*hr} 2035 0.1854 0.3060 0.4731 0.1283 2032 1.305 0.3846 1.442 0.4138 2044 0.5911 0.3417 0.8431 0.3760 2015 0.5295 0.06278 0.2768 0.1579

CHAPTER 4

Efficiency of Intestinal Absorption of d8-β-Carotene in Humans: Variability, Rate of Conversion, and Relationship to Cholesterol Absorption

4.1 Abstract

The intestinal absorption of β-carotene and of cholesterol varies among individuals. Experiments in vitro and in mouse models suggested that β-carotene and cholesterol shared certain molecular mechanisms of absorption. A study was conducted to investigate the variability in β-carotene absorption and its conversion to vitamin A and to compare the efficiency of absorption of β-carotene with that of cholesterol. Ten men consumed a 5 mg dose of deuterium labeled β-carotene (d8βC), with 6 subjects repeating the dose 2 months later. Blood was collected hourly for 9 h, and enrichment of chylomicrons with d8βC and d4-retinyl esters (d4RE) was determined by HPLC-MS.

Cholesterol absorption was measured in the same subjects by the oral/IV dual isotope ratio method using deuterated cholesterol and 13C cholesterol and ranged from 28% to

60%. Chylomicron d8βC AUC ranged from 0.047 to 0.414 μM-hr and the CM d4RE

AUC ranged from 0.102 to 2.59 μM-hr. For subjects who repeated the treatment, the total β-carotene absorbed at the two times was correlated (r=0.99; p<0.0001). The conversion of d8βC to d4RE ranged from 37-77%, and the extent of conversion at the two times was correlated (r=0.77; p=0.08). There is marked inter-individual variability in β-carotene absorption and conversion to RE. In contrast, intra-individual variability in β-carotene absorption and its conversion to RE is low. Importantly, there was no correlation between an individual‟s efficiency of absorption of β-carotene and their efficiency of absorption of cholesterol (r=-0.09; p=0.81).

4.2 Introduction

The absorption of β-carotene involves the release from the food matrix, incorporation into mixed lipid micelles, uptake by the intestinal mucosal cells, package into chylomicrons, and secretion into the lymph. Several recent studies have suggested that uptake of dietary lipids including β-carotene and other carotenoids are protein- mediated (During & Harrison, 2006). Once the β-carotene is taken up by the intestinal mucosal cell it is either cleaved by BCO1 or BCO2 or it is absorbed intact. β-Carotene and its metabolites are incorporated into chylomicrons along with cholesterol esters, triglycerides, apolipoprotein B and phospholipids. The chylomicrons are then secreted into lymph and enter the plasma via the thoracic duct. Similarly, cholesterol is also micellerized in the intestinal lumen along with other lipids for absorption. Cholesterol esters must be hydrolyzed by CEHs prior to absorption, much like retinyl esters, and free cholesterol and retinol are re-esterified upon uptake by the enterocyte prior to being packaged in CMs for secretion.

(During et al., 2005), as well as binding to several brush border membrane proteins

(Kramer et al., 2003). Several studies have shown that SR-BI is involved in intestinal cholesterol absorption (Altmann et al., 2002; Levy et al., 2004) and carotenoid uptake

(During et al., 2008; Kiefer et al., 2002), and β-carotene absorption (van Bennekum et al., 2005). However, SR-BI is not essential in cholesterol absorption. SR-BI -/- mice did not show any difference in the absorption of cholesterol when compared to SR-BI +/+ mice (van Bennekum et al., 2005). It is suggested that the lack of SR-BI in the SR-BI -/- is compensated by other transport proteins such as cluster determinant 36 (CD36) and

NPC1L1 (Altmann et al., 2004). More recently NPC1L1 has been shown to be the critical intestinal sterol transporter which influences whole body cholesterol homeostasis (Davis et al., 2008). Like cholesterol (Kramer et al., 2003), the intestinal absorption of β-carotene may not be facilitated by a single transporter protein but by a complex process involving multiple proteins (Davis et al., 2004) and to some extent passive diffusion.

β-Carotene and cholesterol may share common transporters in absorption and we hypothesize that β-carotene and cholesterol absorption is correlated in humans. A secondary goal was to define the consistency of the absorption efficiencies of cholesterol and β-carotene in the same subject on repeated measures. Therefore, β- carotene and cholesterol absorptions were determined for ten male smokers on a defined diet. We also determined the intra-individual variability of conversion of β-carotene to vitamin A.

4.3 Materials and Methods

Chemicals

Unless otherwise stated, all chemicals and supplies were purchased from Sigma-

13 d7 & 2.96% d6. C40-β-carotene was purchased from Martek (Columbia, MD) and d4- retinyl acetate was purchased from Cambridge Isotope (Andover, MA). Retinyl- linoleate, -oleate, and -stearate were synthesized by Dr. Robert Curley (College of

Pharmacy, Ohio State University).

Subjects, Diet, and Samples Collection

Recruitment, IRB, exclusion/inclusion:

Convenience sampling was used to recruit potential subjects for a cholesterol study. They were recruited by advertisement in the area of the Beltsville Agricultural

Research Center, Beltsville, MD. Inclusion criteria included: male gender, cigarette smokers (smoking at least 1 pack of cigarettes a day for 3 years immediately prior to the study and planning to continue smoking in the immediate future), age 21 to 70 y-of-age, stable body weight, no major health problems including diabetes, heart disease, stroke, or cancer, and not taking vitamin supplements or prescription medications that could interfere with study outcomes. Subjects were required to pass a brief medical

68 evaluation by a physician and to agree to consume all of the food, and only the food, provided by the study. The Committee on Human Research, Johns Hopkins University

Bloomberg School of Public Health approved the study. All volunteers provided written consent for participation and received monetary compensation commensurate with the effort required as a study participant. Ten subjects were recruited from the cholesterol study to participate in a β-carotene absorption study.

Diets & dietary treatments:

Male smokers were fed a controlled diet having 26 percent of energy from fat with a P:M:S fatty acid ratio of 1.0 : 1.0 : 0.8. The diet provided 16 percent of energy from protein, 12.8 g dietary fiber/1000 kcal, and 70 mg cholesterol/1000 kcal (Table

4.1). Seven servings of tea, placebo, or placebo with caffeine were fed each day, 2.25 servings with breakfast, 2 servings with lunch, 2 servings with dinner and 0.75 servings in the evening. Each treatment was fed for 7 weeks. The diets were fed in a 7-day menu rotation and were prepared using typical American foods. On weekdays, subjects consumed breakfast and dinner at the Human Study Facility under observation of a registered dietitian. Weekend meals and lunches were provided to the subjects for consumption away from the facility. Three treatments were fed as part of the controlled diet: Black tea; Placebo beverage matching the tea in color and taste; Placebo beverage with caffeine added to approximate that in the tea.

The objective of the cholesterol study was to investigate the effects of black tea consumption on blood lipid, lipoproteins and markers of oxidative status in male

69 smokers consuming a controlled diet. Three treatments were fed in a randomized crossover design with 7 weeks per treatment. There was a six to eight week washout period between each treatment.

D8-β-Carotene Dose:

Each subject received approx 5mg of d8-β-carotene (d8-βC) in 8mL of corn oil by mouth at breakfast (that also contained fat). On the day of d8-β-carotene dosing, subjects underwent cholesterol infusion prior to being given breakfast and the d8-β- carotene dose. Subjects had a light, very low-fat snack for "lunch" 4 hours after dosing.

Sample collection:

Subjects were catheterized and 8mL of plasma was drawn every hour for nine hours. Four mL of the plasma was separated into a chylomicron (CM) fraction and a non-CM fraction by ultra-centrifugation. All aliquots of plasma fractions were stored at

-80˚C until analysis.

Sample Extractions

One mL of the CM fraction was transferred to an 11mL glass tube for extraction.

13 Internal standards (IS; retinyl acetate and C40-βC) were added, followed by 1mL of ethanol with 0.1% (w/v) BHT. Next 5mL of HEAT (10:6:7:7

70 twice more with HEAT/NaCl, combining the three organic layers and drying them under a stream of nitrogen. The residue was reconstituted in 300µL of 1:1 MtBE/MeOH with water bath sonication and filtered through 0.45µm nylon filter before HPLC injection. An aliquot of whole plasma (1.8 mL) was extracted using the same protocol.

HPLC

Reversed phase chromatography with a C30 4.6x150mm, 5µm column was used to separate retinoids and carotenoids. We used two separate chromatographic separations, one for retinol and retinyl esters and one for β-carotene isotopomers. Each run utilized the same binary solvent systems: (A) 80:20 MeOH/0.1% formic acid (FA) and (B) 78:20:2 MtBE/MeOH/0.1%FA at 1.8mL/min and a column temperature of

40˚C. UV-vis spectra of eluent were monitored by photodiode array (Waters Acquity

PDA). The retinoids separation involved a 25µL injection with an initial condition of

100 %A, a linear gradient to 100 %B over 6min, held at 100 %B for 2 min then returned to 100 %A for a total run time of 10min. The carotenoids LC method had a 50µL injection starting with 100 %A and a linear gradient to 100 %B over 16min, returning to

100 %A for a total run time of 18min. For a given sample, the two separations were run consecutively.

Mass Spectrometry

The HPLC eluate was interfaced with an atmospheric pressure chemical ionization (APcI) probe of a triple quadrupole mass spectrometer (Quattro Ultima,

Micromass, UK). For the retinoid MS/MS method, retinol, retinyl acetate and retinyl

71 esters ionized in APcI positive mode as the M+H-H2O parent ion (269 m/z) and were fragmented by collision induced dissociation to corresponding product ions 93 and 107 m/z. Co-migration with authentic standards confirmed identification. The d4 retinol and d4 retinyl esters were monitored as MS/MS transitions m/z 273>94, 217. The β- carotene HPLC-MS analysis utilized selected ion reaction (SIR) mode with m/z 536.4

13 for native β-carotene, m/z 544.4+543.4 (d8+d7) for d8 βC, and m/z 576.4 for the C40-

βC (β-carotene IS).

Cholesterol Analysis

Cholesterol dosing and blood sampling:

Cholesterol absorption was measured at the end of each dietary phase. Subjects provided a 10 mL baseline blood sample prior to receiving an intravenous injection of

15 mg [25,26,26,26,27,27,27] D7-cholesterol and a 75 mg oral dose of [3,4]13C- cholesterol for cholesterol absorption determination. The D7-cholesterol isotope was prepared for injection by dissolving the isotope in ethanol at a concentration of 5 mg/mL under sterile conditions. The isotope/ethanol mixture was added drop-wise to an intravenous fat emulsion (Intralipid, Baxter Corp.), for a total injectable volume of 9 mL.

Following ingestion and injection of the labeled cholesterol, a 10 ml blood sample was collected at 24 hr, 48 hr and 72 hr. Serum was collected from the blood. The ratio of ingested [3,4]13C-cholesterol to injected [25,26,26,26,27,27,27]

D7-cholesterol enrichment in serum cholesterol after 24, 48, and 72 hrs is taken as an indicator of the fractional cholesterol absorption rate.

Cholesterol absorption was determined from cholesterol extracted from red blood cells. Isotope enrichments were measured by differential isotope ratio mass spectrometry. The average 13C and D enrichments of 48 and 72 hr RBC free cholesterol relative to baseline (t=0) samples was used to calculate the cholesterol absorption coefficient (CAC) using the ratio of orally ingested 13C-cholesterol to intravenously administered D7-cholesterol as described by Bosner et al (Bosner et al.,

1993).

Cholesterol synthesis was determined at the end of each diet period using the deuterium incorporation approach. Seventy-two hours following dosing with 13C- cholesterol and D7-cholesterol, subjects were dosed with 0.7 g/kg of estimated body water, deuterium oxide (D) (99.8% atom percent excess, CDN Isotopes). Body water is estimated to be 60% for calculation of the dose. Deuterium oxide was given per os immediately following the collection of the 72 hr blood collection. A 10 mL blood sample was collected 24 hr following ingestion of the deuterium oxide. Serum was collected from the blood.

Cholesterol synthesis was determined as the rate of incorporation of D from body water into the red blood cell membrane free cholesterol over the period between

72 and 96 hr at the end of each feeding period. Cholesterol fractional synthesis rate

(FSR) represents the red blood cell free cholesterol D enrichment values relative to the

73 corresponding plasma water sample enrichment after correcting for the free cholesterol pool. The FSR represents that fraction of the cholesterol pool that is synthesized in 24 hours as calculated by Jones et al (Jones et al., 1993).

Data analysis

Concentrations of analytes were determined by ratio to the internal standard.

Internal standards controlled for variation in recovery and matrix suppression of MS response. The concentrations of the deuterated compounds were plotted versus time after dosing and the area under curve (AUC) calculated to reflect bioavailability.

Results are expressed in µM*hr. A Pearson correlation was used to compare the data.

4.4 Results

The intestinal absorption of β-carotene and of cholesterol varies among individuals. Experiments in vitro and in mouse models suggested that β-carotene and cholesterol shared certain molecular mechanisms of absorption. A study was conducted to investigate the variability in β-carotene absorption and its conversion to vitamin A and to compare the efficiency of absorption of β-carotene with that of cholesterol. Ten men consumed a 5 mg dose of deuterium labeled β-carotene (d8-βC), with 6 subjects repeating the dose 2 months later. Blood was collected hourly for 9 hours, and enrichment of chylomicrons with d8-βC and d4-retinyl esters (d4-RE) was determined by HPLC-MS.

Cholesterol absorption was measured in the same subjects by the oral/IV dual isotope ratio method using deuterated cholesterol and 13C cholesterol and ranged from

27.8% to 59.6% with a mean of 40.0% ± 9.2%. The subject characteristics are shown in

Table 4.2 and β-carotene and cholesterol absorption data are shown in Table 4.3.

The total d8-βC absorbed is represented as the AUC d8-βC + (AUC sum of d4 retinyl esters)/2. The % conversion is the amount of the total d8-βC that was absorbed as d4-RE. AUC of intact d8-βC is positively correlated with AUC sum d4-RE and AUC total d8-βC equivalents (r=0.908; p<0.000, r=0.946; p<0.000, respectively). AUC sum d4-RE is positively correlated with AUC total d8-βC equivalents and % conversion

(r=0.995; p<0.000, r=0.640; p=0.046, respectively). AUC total d8-βC equivalents is positively correlated with % conversion (r=0.585; p=0.075). Chylomicron d8-βC AUC ranged from 0.047 to 0.414 μM-hr and the CM d4-RE AUC ranged from 0.102 to 2.59

μM-hr. Both the cholesterol absorption and the β-carotene absorption varied among individuals but there was no correlation between an individual‟s efficiency of absorption of β-carotene and their efficiency of absorption of cholesterol (r=-0.090; p=0.806) (Figure 4.1). However, there was a strong correlation between an individuals absorption of β-carotene and their conversion of β-carotene to retinyl esters (r=0.585, p=0.075) (Figure 4.2).

Figures 4.3-4.5 show the intra-individual variability of cholesterol absorption,

β-carotene absorption, and percent conversion in subjects that repeated the experiment three months later. Cholesterol absorption, β-carotene absorption, and percent conversion were similar for the seperate measurements. For subjects who repeated the 75 treatment, the total β-carotene absorbed at the two times was correlated (r=0.99; p<0.0001) (Figure 4.4). The conversion of d8-βC to d4-RE ranged from 37-77%, and the extent of conversion at the two times was correlated (r=0.77; p=0.08) (Figure 4.5).

There is marked inter-individual variability in β-carotene absorption and conversion to

RE. In contrast, intra-individual variability in β-carotene absorption and its conversion to RE is low.

In Summary β-carotene absorption is not correlated with cholesterol absorption.

The conversion of d8-βC to d4-RE ranged from 37-77%. β-Carotene absorption is positively correlated with the extent of its conversion to RE. Intra-individual variability in β-carotene absorption and its conversion to RE is low. Native whole plasma levels of

Retinol, RP, and β-carotene were determined for four of the subjects and were not correlated with levels of absorption for cholesterol or d8-βC.

4.5 Discussion

There have been a plethora of stable isotope human studies most of which have been reviewed (Burri & Clifford, 2004; Furr et al., 2005; Novotny et al., 2005; Tang,

2010; van Lieshout et al., 2003) and many have included isotopically labeled β-carotene.

Each has examined different aspects of absorption and metabolism and they have different strengths and weaknesses. Stable isotope human studies are costly and complex and samples sizes are inherently low. In most cases a single pharmacological dose of labeled β-carotene is administered and tracked through the plasma over time. 76

The absorption of β-carotene reported in these studies is highly variable and other studies using unlabeled β-carotene have reported on the variability of β-carotene absorption (Borel et al., 1998). Peak absorption of β-carotene and its metabolites differs across studies, due to sampling in some cases, but ranges from 3.5-8 hours post- prandially (Burri & Clifford, 2004; Dueker et al., 1994; Dueker et al., 2000; Pawlosky et al., 2000). Often, the plasma samples are extracted, saponified and further purified in complex preparations for analysis. Saponification of the samples converts all of the retinyl esters to retinol and does not allow the individual retinoids to be examined.

Although retinyl palmitate is the most abundant retinyl ester, other retinyl esters such as retinyl-linoleate, oleate and stearate can contribute up to 50% of the total retinyl esters.

There is a large variation in the conversion of β-carotene to its metabolites where 35-75% of the absorbed β-carotene is converted to retinyl esters in the intestinal cells (During &

Harrison, 2006; Edwards et al., 2002; Goodman et al., 1966; van Vliet et al., 1995).

As previously seen the absorption of β-carotene had high inter-individual variability, as did the extent of conversion of new absorbed β-carotene. In contrast, intra-individual variability of β-carotene absorption and conversion was low for subjects who repeated the study. This may suggest transcriptional regulation of β-carotene absorption and perhaps genetic differences. We are currently in the process of isolating

DNA from our subjects for genotyping.

4.6 Acknowledgements

We thank Ms. Rachel E. Kopec with here expert assistance and training in the method development and Dr. Robert W. Curley for the synthesis of retinyl-esters.

4.7 Figures and Tables

Table 4.1. Nutrient contents of interest as average intake per day

Protein %En 16.7

Fat %En 29.8

Carbohydrate (CHO) %En 55.7

Alpha-carotene mg/2000 kcal 1.5

Beta-carotene mg/2000 kcal 4.1

Beta-cryptoxanthin mg/2000 kcal 0.2

Lutein+Zeaxanthin mg/2000 kcal 1.6

Lycopene mg/2000 kcal 3.2

Vitamin A (retinol) RE (mcg)/2000 kcal 1382.8

Alpha Tocopherol (VE) (mcg)/2000 kcal 6.1

Table 4.2. Subject Characteristics

Table 4.3. β-Carotene and Cholesterol Absorption

2.0 1.8 1.6 1.4

1.2 C C AUC β 1.0 0.8

0.6 Total d8 d8 Total 0.4 0.2 0.0 20.0 30.0 40.0 50.0 60.0 70.0

Cholesterol absorption (%)

Figure 4.1. Correlation of the total d8-βC AUC (µM*hr) versus the percent cholesterol absorption (r=-0.090; p=0.806).

2.0 1.8 1.6 1.4

1.2 C C AUC β 1.0 0.8

0.6 Total d8 d8 Total 0.4 0.2 0.0 30.00 40.00 50.00 60.00 70.00 80.00

Conversion (%)

Figure 4.2. Correlation of the total d8-βC AUC (µM*hr) versus the percent conversion to vitamin A (r=0.585, p=0.075).

Cholesterol Absorption

65.0

60.0 2015 55.0 2018 50.0 45.0 2032

40.0 2033 35.0 % CHL absorptionCHL % 2037 30.0 2044 25.0 20.0 1st Dose 2nd Dose

Figure 4.3. Cholesterol absorption of the six subjects who repeated the experiment.

Total d8-βC 2 1.8 2015 1.6 2018 1.4 1.2 2032 1 0.8 2033

AUC µmolar*hr AUC 0.6 2037 0.4 0.2 2044 0

1st Dose 2nd Dose

Figure 4.4. Total d8-βC AUC (µM*hr) β-carotene absorption of the six subjects who repeated the experiment. Subjects who repeated the experiment had similar levels of total β-carotene absorption at both time points. Low absorbers remained low and high absorbers remained high.

Conversion 90

80 2015

2018 70 2032 60 2033

% Conversion % 50 2037 40 2044

1st Dose 2nd Dose

Figure 4.5. β-Carotene conversion to retinyl esters of the six subjects who repeated the experiment. Of the six subjects, two had similar percent conversion at both time points

(2032 and 2033), three had less than 30% change in percent conversion (2015, 2037, and 2044), and one had approximately 57% change in percent conversion (2018), but

2018 was the lowest absorber of total d8- β-Carotene.

CHAPTER 5

The Presence of β-Apocarotenoids in Biological Samples

5.1 Abstract

β-Carotene can undergo asymmetric cleavage to yield β-apocarotenoids that have potential biological roles such as vitamin A activity and transcriptional regulation.

In plants carotenoid cleavage dioxygenases (CCDs) cleave β-carotene to apocarotenoids, thus apocarotenoids are present in β-carotene containing fruits and vegetables. The detection of a few of the longer chain β-apocarotenoids have been previously reported in various plants but have not been quantified. In this study we quantify β- apocarotenoid levels in commonly consumed β-carotene containing fruits and vegetables, murine liver and serum, and human plasma. Using HPLC-MS we were able to detect and quantify β-apo-13-carotenone, β-apo-14‟-carotenal, β-apo-12‟-carotenal,

β-apo-10‟-carotenal, β-apo-8‟-carotenal and β-carotene in all fruits and vegetables.

Plants tested were found to have 0.31-4.0% β-apocarotenoids compared to β-carotene on a molar basis. It is important to consider the level of β-apocarotenoid consumption and absorption as they may have potential biological roles. In vitro digestion studies have shown that β-apocarotenoids are micellerized and to a greater extent than β- carotene. β-Apocarotenoids are present in our diets but their absorption, metabolism, and biological roles are largely unknown. We have shown here that they are present in considerable amounts in the diet and that they may be absorbed. We have also shown that these β-apocarotenoids are present in the liver and serum of β-carotene fed mice as well as human plasma. β-Apocarotenoids are present in our diets and available for absorption and they may also be made in vivo by BCO2.

5.2 Introduction

In plants, the oxidative cleavage of carotenoids leads to the production of apocarotenoids and is catalyzed by a family of carotenoid cleavage dioxygenases

(CCDs) (Auldridge et al., 2006). Many of these apocarotenoids have been indentified in plants including short chain products such as β-ionone (Auldridge et al., 2006) and long chain products (Bauernfeind et al., 1981; Singh & Cama, 1974; Winterstein & Hegedus,

1960). Recently we have described the detection and quantification of all of the long chain β-apocarotenoids including β-apo-8‟-carotenal, β-apo-10‟-carotenal, β-apo-12‟- carotenal, β-apo-14‟-carotenal, and β-apo-13-carotenone in melons (Fleshman et al.,

2011). Kopec et al. (2010) have also described the detection of apolycopenals. They found apolycopenal levels to be present in lycopene containing foods at approximately

0.1% of the amount of lycopene, whereas we have found β-apocarotenals to be approximately 1.5% of the amount of β-carotene in β-carotene containing foods

(Fleshman et al., 2011). We also observed what appeared to be other apocarotenoids in foods rich in lutein/zeazanthin and α-carotene, but were unable to identify and quantify them without the appropriate standards. Thus, long chain apocarotenoids are present and substantial in carotenoid containing foods.

There is limited evidence for the presence of apocarotenoids in vivo but recently

Shmarakov et al. (2010) describe a BCO1-deficient mouse that may be more useful in assessing β-carotene absorption and eccentric cleavage in the mouse model. Eccentric

88 cleavage of β-carotene to β-apocarotenoids has been shown in various in vitro models

(Barua & Olson, 2000; Hu et al., 2006; Kiefer et al., 2001; X. D. Wang et al., 1991) but in vivo evidence is limited in humans. Ho et al. (2007) detected the presence of β-apo-

8‟-carotenal in human plasma following an isotopically labeled dose of 14C-β-carotene.

The β-apo-8‟-carotenal did not appear in the plasma until 3 days after the dose suggesting that β-apocarotenals are formed in the peripheral tissues from dietary β- carotene. They also detected a second peak on day 3 that may be β-apo-8‟-carotenyl ester. Confirmation of eccentric cleavage in humans is needed as well as the identification of other apocarotenoids in vivo. Apocarotenoids are present in the diet and may be absorbed intact as well as formed in vivo from the eccentric cleavage of carotenoids by BCO2. The purpose of this study is to quantify the levels of β- apocarotenoids in β-carotene containing plants, mice, and humans to give a better understanding of their potential biological roles.

5.3 Materials and Methods

Chemicals and Supplies.

Unless otherwise stated, all chemicals and supplies were purchased from Sigma-

Aldrich (St. Louis, MO, USA) or Fisher Scientific (Pittsburgh, PA, USA).

Sample Extraction

Samples were analyzed by ultra high pressure liquid chromatography tandem mass spectrometry (UPLC-MS/MS) for β-carotene, β-apo-8‟-carotenal, β-apo-10‟- carotenal, β-apo-12‟-carotenal, β-apo-14‟-carotenal, and β-apo-13-carotenone levels.

Homogenates of fresh, locally purchased, β-carotene-containing foods were extracted in triplicate from three different plants of the same product (i.e. three melons with three samplings taken from each). We extracted samples using a modification of the method of Kane et al (2005). The first fraction containing neutral lipids was collected and analyzed. Dried extracts were subsequently redissolved in a 1:1 (v/v) mixture of MTBE and methanol and 5 µL was injected onto the HPLC for separation of β-carotene and β- apocarotenoids. Human plasma samples were extracted with the same method except various volumes of plasma were extracted and solvent volumes adjusted accordingly.

HPLC

Separations were accomplished by reversed phase HPLC using a 4.6x50mm

1.8µm Zorbax Eclipse XDB C18 column (Agilent Technologies, Santa Clara, CA). A binary solvent system consisting of (A) 20/80 0.1% formic acid/methanol and (B)

78/20/2 MTBE/methanol/0.1% formic was employed. The initial UPLC mobile phase was 100% A increased linearly through 100%B over 5 min followed by a 2.5 min re- equilibration. The flow rate of the solvent was maintained at 2.33 mL/min with a column temperature of 35 ˚C.

Mass Spectrometry

The HPLC eluent was interfaced with a QTRAP 5500 hybrid quadrupole/ion trap mass spectrometer (AB SCIEX, Foster City, CA) via an atmospheric pressure chemical ionization (APCI) probe operated as a triple quadrupole instrument in positive mode. β-Carotene, β-apo-8‟-carotenal, β-apo-10'-carotenal, β-apo-12‟-carotenal, β-apo-

14‟-carotenal, and β-apo-13-carotenone ionized as their respective pseudo-molecular cations of 537.438, 417.31, 377.28, 351.26, 311.23, and 259.2 m/z. Each parent ion was fragmented with collision-induced dissociation (N2 gas) to its most abundant daughter ions and determined by multiple reaction monitoring (MRM). The resultant

MRM chromatograms were integrated with Analyst 1.5.1 software (AB SCIEX, Foster

City, CA). Source parameters included 70-140V declustering potential; 425 °C probe;

4-7V entrance potential; 15-37eV collision energy; 15ms dwell; and 30 psi curtain gas.

Standard curves were generated using standards of β-carotene, β-apo-8‟- carotenal (Fluka), β-apo-10‟-carotenal (synthesized in-house), β-apo-12‟-carotenal

(CaroteNature, Lupsingen, Switzerland), β-apo-14‟-carotenal (synthesized in-house), and β-apo-13-carotenone (synthesized in-house).

5.4 Results

We have recently described the presence of the β-apocarotenoids, shown in

Figure 5.1, in orange-fleshed melons (Fleshman et al., 2011). Figure 5.2 shows a

91 sample chromatogram of standards and the β-apocarotenoids in cantaloupe melon. All of the β-apocarotenoids of interest were present in the fruit and the total moles of β- apocarotenoids were ~1.5% of the total moles of β-carotene.

Upon refining our LC-MS/MS method we analyzed other β-carotene containing foods for the presence of β-apocarotenoids. We were able to detect and quantify β-apo-

8‟-carotenal, β-apo-10‟-carotenal, β-apo-12‟-carotenal, β-apo-14‟-carotenal, and β-apo-

13-carotenone in cantaloupe, pumpkin, sweet potato, and yellow squash as shown in

Figure 5.3 as well as carrots (data not shown).

The average total moles of β-apocarotenoids, compared to total moles of β- carotene, for cantaloupe, pumpkin, sweet potato, and yellow squash was 0.31%, 2.48%,

0.35%, and 2.96% respectively. We also observed what appeared to be other apocarotenoids in foods rich in lutein/zeaxanthin and α-carotene but were unable to identify and quantify them without the appropriate standards. Thus, long chain apocarotenoids are present and substantial in carotenoid containing foods.

There is limited evidence for the presence of apocarotenoids in vivo but recently

Shmarakov et al. (2010) describe a BCO1-deficient mouse that may be more useful in assessing β-carotene absorption and eccentric cleavage in the mouse model. The BCO1- deficient mice were fed a diet rich in β-carotene and as expected there was significantly more β-carotene in both the serum and liver when compared to wild type mice fed the same diet. We were also able to detect β-apo-12‟-carotenal and β-apo-10‟-carotenal in both the BCO1-deficient and wild type β-carotene fed mice (Table 5.1). There was a

92 consistent trend for β-apo-12'-carotenal and β-apo-10'-carotenal levels to be higher in serum and liver of BCO1-deficient compared to wild type mice, but this was only statistically significant (p < 0.05) for hepatic β-apo-12'-carotenal levels. This may indicate a preferential uptake of β-apo-12‟-carotenal and β-apo-10‟-carotenal or the eccentric cleavage of β-carotene by BCO2 in vivo.

We measured by LC/MS β-carotene, retinal, β-apo-8'-, β-apo-10'-, β-apo-12' and

β-apo-14‟-carotenal concentrations in serum and liver of wild type and BCO1-deficient mice maintained on the β-carotene containing diet. Table 5.1 provides a summary of these measurements. As can be seen from Table 5.1, β-apo-10'- and β-apo-12'- carotenals were present in both tissues, but we were unable to detect either β-apo-8'- or

β-apo-14'-carotenal in serum or liver obtained from either BCO1-deficient or wild type mice.

We analyzed the β-carotene containing mouse feed and found β-apo-8‟- carotenal, β-apo-10‟-carotenal, β-apo-12‟-carotenal, and β-apo-14‟-carotenal in the feed itself as well as in the β-carotene beadlets used in formulating the diet (Table 5.2). We also subjected the β-carotene supplemented mouse feed to in vitro digestion shown in

Table 5.3. There is an inverse relationship between chain length and percent micellerization (Figure 5.4). Thus the shorter chain β-apocarotenoids are more bioavailable than β-carotene. For example, the ratio of absorbed β-apo-14‟-carotenal to

β-carotene may be closer 1:10 rather than the source ratio of 1:100.

We have also been able to detect and quantify β-apo-8‟-carotenal, β-apo-10‟- carotenal, β-apo-12‟-carotenal, β-apo-14‟-carotenal, and β-apo-13-carotenone in human plasma. β-Apocarotenoids in human plasma range from sub-nanomolar to nanomolar concentrations (data not shown).

5.5 Discussion

It is possible that the observed elevation in expression of BCO2 in BCO1 deficient mice is related either to the increased hepatic β-carotene concentration and/or to the increased levels of β-apo-10'- or β-apo-12'-carotenals themselves. Since BCO2 catalyzes the eccentric cleavage of β-carotene that can give rise potentially to a number of β-apocarotenoids and mice deficient in BCO1 show increased levels of BCO2 mRNA in their livers, we wondered whether β-carotene-fed BCO1-deficient mice might have higher levels of β-apocarotenal in their livers and/or serum than wild type mice. β-

Apo-10‟-carotenal and β-apo-12‟-carotenal were the only β-apocarotenals detected in serum and livers from both BCO1-deficient and wild type mice. Concentrations of β- apo-12‟-carotenal and β-apo-10‟-carotenal in both serum and liver tended to be higher for the BCO1-deficient mice but only β-apo-12'-carotenal levels in livers from the mutants proved to be significantly (p < 0.05) greater than those of wild type mice. This finding is consistent with the notion that elevated levels of BCO2 expression in livers of

BCO1-deficient mice gives rise to increased β-apocarotenal formation through eccentric cleavage of β-carotene. β-Apo-14‟-carotenal and β-apo-8‟-carotenal were not detected

94 in any serum or liver sample for either strain but surprisingly, β-apo-8‟-, β-apo-10‟-, β- apo-12‟-, and β-apo-14‟-carotenal were all detected in the diet formulated using β- carotene beadlets and in the beadlets themselves. β-Apo-14‟-carotenal was the most abundant β-apocarotenal form detected in the diet, and its absence in the circulation could suggest that immediate metabolism occurs within the intestine. However, this possibility requires confirmation. As expected based on previously published data

(Fierce et al., 2008; Hessel et al., 2007), there was 40-times more β-carotene in serum and 20-times more β-carotene in liver of BCO1-deficient compared to wild type mice.

Expression of BCO1 clearly influences the amount of β-carotene taken up from the diet and retained by the body. We did not observe a decrease in liver retinal concentrations in BCO1-deficient mice. It was hypothesized since retinal is the sole product of BCO1 catalyzed cleavage of β-carotene, tissue retinal levels might be diminished when BCO1 is ablated. However, this was not observed. Since retinal can be formed through oxidation of retinol, as well as from β-carotene cleavage (Olson, 1989; Vogel,S. et al.,

1999), this finding suggests that either β-carotene cleavage by BCO1 is not required for maintaining hepatic retinal pools and/or that the liver strives to maintain retinal levels within a fairly tight concentration range, regardless of its metabolic precursor. Contrary to a recent report regarding retinal concentrations in serum (Ziouzenkova et al., 2007), we were unable to detect retinal in serum from any of our mice.

β-Apo-10'- and β-apo-12'-carotenals, but not β-apo-8'- or β-apo-14'-carotenals, were detected in serum and liver from wild type and BCO1-deficient mice. Hepatic β- apo-12'-carotenal levels were significantly elevated in the livers of BCO1-deficient

95 compared to wild type mice, possibly arising from the elevated expression of BCO2 in the livers of these mutant mice. These data raise the interesting possibility that the body distinguishes metabolically between different β-apocarotenal species. Further investigations will be needed to establish the processes and factors responsible for this.

β-Apocarotenoids are present in β-carotene containing foods and are more bioaccessible than β-carotene. β-apocarotenoids have been detected in both mice and humans but there is need to further characterize concentration ranges in human plasma.

5.6 Acknowledgements

We thank Dr. Robert W. Curley for the synthesis of β-apocarotenoid standards.

5.7 Figures and Tables

Figure 5.1. Structure of β-carotene and its β-apocarotenoid eccentric cleavage products

Figure 5.2. Chromatograms of β-apocarotenoid standards and β-apocarotenoids in cantaloupe. The left panel shows the standards used to identify the β-apocarotenoids and the right panel shows β-apocarotenoids found in cantaloupe. There appears to be a

β-apo-12‟-carotenal isomer in the cantaloupe but we were unable to identify this with authentic standard.

Figure 5.3. β-Apocarotenoid profiles of commonly consumed β-carotene containing foods. β-Apocarotenoids were present in all of the foods tested

TABLE 5.1: Serum and Liver Levels of β-carotene and β-apocarotenals for β- carotene Fed Wild Type and BCO1-Deficient Mice (Shmarakov et al., 2010)

1Serum and liver concentrations are given as the mean followed in parenthesis by the range of values measured for all samples from that group. Concentrations of compounds in the diet are provided as the means ± S.D. N.D. signifies not detected.

2Significantly different from corresponding wild type, p < 0.01.

3Significantly different from corresponding wild type, p < 0.05.

100

Table 5.2. β-Apocarotenoid profiles of mouse feed and β-C beadlet

Table 5.3. Bioavailability of β-apocarotenoids

101

Figure 5.4. Inverse relationship of chain length to micellerization. Chain length increases from right to left and the apocarotenals also contain oxygen which makes them more polar than β-carotene and likely lends toward better micellerization.

102

CHAPTER 6

Epilogue

β-Carotene has been vastly studied and well characterized but there is still much that we do not understand. The previous chapters touched upon a few aspects of β- carotene absorption and metabolism. Major findings include the detection of eccentric cleavage products in various samples containing β-carotene and new aspects of β- carotene absorption and conversion in humans.

The first study, Chapter 2, focused on the β-carotene content of orange-fleshed melons. The orangedew melons are known to be safer for the consumer because they lack rough outer netting, unlike cantaloupe, fruit which harbors enteric bacteria. The problem lies in the cantaloupes‟ rough outer netting, which can harbor pathogens and makes surface cleansing difficult. When netted melons are cut, any microbes present on the exterior can be transferred to the inner flesh. The orangedew has the smooth skin of a honeydew and the orange flesh of a cantaloupe. The smooth skinned orangedew does not carry the same consumer risk for food borne illness as melons with rough outer netting and the orange flesh would suggest that it is a good source of β-carotene.

We found the orangedew to have significantly more β-carotene than cantaloupe grown under the same conditions. Thus the orangedew has less consumer risk and has 103 better nutritional value. We also studied the bioaccessibility of the β-carotene in the orange-fleshed melons. Approximately 3.2% of the β-carotene in the melons was micellerized and made accessible for absorption. This is quite a bit lower than the bioaccessibility of β-carotene in other food sources; perhaps do to water content and food preparation. However, the low bioaccessibility does allude to the problem of vitamin A deficiency in regions that obtain the majority of their vitamin A through provitamin A carotenoids in fruits and vegetables. Even if people in these regions have adequate access to fruits and vegetables containing provitamin A sources, it may not be enough when considering the bioaccessibility. This raises many questions as to how to increase the bioaccessibility of β-carotene in staple crops in regions afflicted by vitamin

A deficiency.

The orange-fleshed melons studied in Chapter 2 also provided us the opportunity to look for β-apocarotenoids in the fruits, which may have potential biological roles. Some of the β-apocarotenoids had been previously reported in other foods in older studies; however our study was the first to report the finding of all of the long-chain β-apocarotenoids in a food. We were able to detect and quantify β-apo-13- carotenone, -14‟-carotenal, -12‟-carotenal, -10‟-carotenal, and -8‟-carotenal in both orangedew and cantaloupe melons. We continued our exploration into the presences of apocarotenoids as discussed in Chapter 5.

Chapters 3 and 4 focus on a β-carotene feeding study where ten male smokers were given an isotopically labeled dose of β-carotene and their blood was drawn for nine hours post-dose. To analyze the samples, we developed and HPLC/MS method to 104 detect and quantify the labeled dose. Chapter 3 describes the method in detail and

Chapter 4 describes our findings.

For this study, we were interested in the absorption of the stable isotopically labeled β-carotene dose, which includes the dose that was absorbed as β-carotene metabolites. Once the β-carotene is taken up by the enterocyte it is able to be centrally cleaved by BCO1 to form retinal, which is ultimately esterified by LRAT to produce retinyl esters, of which the four primary REs are retinyl-linoleate, -oleate, -palmitate, and -stearate. Our method, unlike previous studies that looked at the absorption of a labeled dose of β-carotene, enabled us to detect and quantify the amount of the individual d4-REs. Previous studies saponified their samples and detected only d4- retinol as a β-carotene metabolite. The method described in chapter 4 allows accurate quantitation of the absorption and conversion of stable isotopically labeled β-carotene, while avoiding complex sample preparation and provides shorter chromatographic run times. The method allows for the distinction between d4-retinol and d4-REs. This is useful in understanding the metabolism of newly absorbed β-carotene.

The d8-β-carotene feeding study was originally designed to determine the correlation of β-carotene and cholesterol absorption because of recent evidence that they may share intestinal mucosal transporters. However, we found the two to be poorly correlated in our subjects. This does not suggest that cholesterol and β-carotene do not share common intestinal transporter proteins but may suggest that β-carotene absorption is just as complex and variable as cholesterol absorption involving many different proteins, transcriptional regulation, and dietary factors. 105

Despite the lack of evidence for the correlation between β-carotene and cholesterol absorption, the study provided great insight into the absorption of β-carotene.

As described in the β-carotene feeding study, positive correlation existed between β-carotene absorption and its conversion to retinyl-esters, but the retinoids are not the sole metabolites of β-carotene. β-Carotene metabolism affects β-carotene absorption and may be as equally complex. In Chapter 5, we looked at the presence of eccentric cleavage products of β-carotene; β-apocarotenoids. Our first study involving

β-apocarotenoids was with BCO1-/- mice and as reported in Shmarakov et al. (2010) and in Chapter 5, we detected more β-apo-12‟-carotenal and β-apo-10‟-carotenal in the

BCO1-deficient mice than in the wild-type mice. This would be expected since there is more β-carotene present to be cleaved by BCO2, along with the up-regulation of BCO2.

Although we detected all of the β-apocarotenoids in the mouse diet and showed that the shorter chain β-apocarotenoids are more bioaccessible, we were only able to detect β-apo-12‟-carotenal and β-apo-10‟-carotenal in the serum and liver of the β- carotene fed mice. This may suggest preferential uptake, metabolism of the other β- apocarotenoids, or simply an inability to detect them with our current method. The aldehyde form of the β-apocarotenoids is highly reactive and may be esterified for 106 storage similar to retinal. Once β-apocarotenyl-esters are available as standards, it may be beneficial to re-examine the BCO1 -/- mice and human plasma to look for the possible storage form of β-apocarotenoids as well as the alcohol form.

Humans and other omnivores and herbivores are consuming β-apocarotenoids on a daily basis but there is relatively little known about their biological importance. It is possible that they are involved in feedback inhibition of β-carotene absorption by retinoic acid. It has also been suggested that β-apocarotenoids are partially responsible for the negative affects seen in the β-carotene supplementation trials, but rather than the production of these β-apocarotenoids in vivo, they are most likely present in substantial concentrations in the supplement and directly absorbed. β-Apocarotenoids may be easily metabolized with a normal intake of β-carotene containing foods and are perhaps beneficial, but with high doses of β-apocarotenoids the body may be overwhelmed and the highly reactive β-apocarotenals may cause detrimental effects.

There is much to be learned about β-apocarotenoids as well as β-carotene absorption and metabolism. β-Carotene supplementation is the safest way to supplement vitamin A in vitamin A deficient populations in terms of the risk of vitamin A toxicity but the toxicity of β-apocarotenoids should be examined before its use. It may be beneficial to develop a pure β-carotene supplement that is void of any β-apocarotenoids if they are found to be harmful at high concentrations. It is therefore important to develop ways to improve β-carotene bioavailability in regions that derive most of their vitamin A from plant sources. “If only everyone enjoyed the taste of liver!”

107

List of References

Albanes, D., Heinonen, O. P., Huttunen, J. K., Taylor, P. R., Virtamo, J., Edwards, B. K., et al. (1995). Effects of alpha-tocopherol and beta-carotene supplements on cancer incidence in the alpha-tocopherol beta-carotene cancer prevention study. The American Journal of Clinical Nutrition, 62(6 Suppl), 1427S-1430S.

Altmann, S. W., Davis, H. R.,Jr, Yao, X., Laverty, M., Compton, D. S., Zhu, L. J., et al. (2002). The identification of intestinal scavenger receptor class B, type I (SR-BI) by expression cloning and its role in cholesterol absorption. Biochimica Et Biophysica Acta, 1580(1), 77-93.

Altmann, S. W., Davis, H. R.,Jr, Zhu, L. J., Yao, X., Hoos, L. M., Tetzloff, G., et al. (2004). Niemann-pick C1 like 1 protein is critical for intestinal cholesterol absorption. Science (New York, N.Y.), 303(5661), 1201-1204.

Auldridge, M. E., McCarty, D. R., & Klee, H. J. (2006). Plant carotenoid cleavage oxygenases and their apocarotenoid products. Current Opinion in Plant Biology, 9(3), 315-321.

Barua, A. B., & Olson, J. A. (2000). Beta-carotene is converted primarily to retinoids in rats in vivo. The Journal of Nutrition, 130(8), 1996-2001.

Bauernfeind, J. C., Adams, C. R., & Marusich, W. L. (1981). Carotenes and other vitamin A precursors in animal feed. In J. C. Bauernfeind (Ed.), Carotenoids as colorants and vitamin A precursors (pp. 563-743). New York: Academic Press.

Biehler, E., & Bohn, T. (2010). Methods for assessing aspects of carotenoid bioavailability. Curr. Nutr. Food Sci., 6(1), 44-69.

Bohm, V., Puspitasari-Nienaber, N. L., Ferruzzi, M. G., & Schwartz, S. J. (2002). Trolox equivalent antioxidant capacity of different geometrical isomers of alpha- carotene, beta-carotene, lycopene, and zeaxanthin. Journal of Agricultural and Food Chemistry, 50(1), 221-226.

Borel, P., Grolier, P., Mekki, N., Boirie, Y., Rochette, Y., Le Roy, B., et al. (1998). Low and high responders to pharmacological doses of beta-carotene: Proportion in the

108

population, mechanisms involved and consequences on beta-carotene metabolism. Journal of Lipid Research, 39(11), 2250-2260.

Bosner, M. S., Ostlund, R. E.,Jr, Osofisan, O., Grosklos, J., Fritschle, C., & Lange, L. G. (1993). Assessment of percent cholesterol absorption in humans with stable isotopes. Journal of Lipid Research, 34(6), 1047-1053.

Botella-Pavia, P., & Rodriguez-Conception, M. (2006). Carotenoid biotechnology in plants for nutritionally improved foods. Physiol. Plant., 126, 369-381.

Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry, 72, 248-254.

Britton, G., Liaaen-Jensen, S., & Pfander, H. (1995). Carotenoids. Basel; Boston: BirkhÃ¤user Verlag.

Britton, G., Liaaen-Jensen, S., & Pfander, H. (2004). Carotenoids handbook. Basel; Boston: BirkhÃ¤user Verlag.

Burri, B. J., & Clifford, A. J. (2004). Carotenoid and retinoid metabolism: Insights from isotope studies. Archives of Biochemistry and Biophysics, 430(1), 110-119.

Burri, B. J., & Park, J. Y. (1998). Compartmental models of vitamin A and beta- carotene metabolism in women. Advances in Experimental Medicine and Biology, 445, 225-237.

Canene-Adams, K., & Erdman, J. W.,Jr. (2009). Absorption, transport, distribution in tissues and bioavailability. In G. Britton, S. Liaaen-Jensen, H. Pfander (Ed.), Carotenoids (5th ed., pp. 115-148). Basel; Boston: BirkhÃ¤user Verlag.

Chitchumroonchokchai, C., Schwartz, S. J., & Failla, M. L. (2004). Assessment of lutein bioavailability from meals and a supplement using simulated digestion and caco-2 human intestinal cells. The Journal of Nutrition, 134(9), 2280-2286.

Davis, H. R.,Jr, Basso, F., Hoos, L. M., Tetzloff, G., Lally, S. M., & Altmann, S. W. (2008). Cholesterol homeostasis by the intestine: Lessons from niemann-pick C1 like 1 [NPC1L1). Atherosclerosis.Supplements, 9(2), 77-81.

Davis, H. R.,Jr, Zhu, L. J., Hoos, L. M., Tetzloff, G., Maguire, M., Liu, J., et al. (2004). Niemann-pick C1 like 1 (NPC1L1) is the intestinal phytosterol and cholesterol transporter and a key modulator of whole-body cholesterol homeostasis. The Journal of Biological Chemistry, 279(32), 33586-33592.

109

De Jesus Ornelas-Paz, J., Failla, M. L., Yahia, E. M., & Gardea-Bejar, A. (2008). Impact of the stage of ripening and dietary fat on in vitro bioaccessibility of beta- carotene in 'ataulfo' mango. Journal of Agricultural and Food Chemistry, 56(4), 1511-1516.

Dueker, S. R., Jones, A. D., Smith, G. M., & Clifford, A. J. (1994a). Stable isotope methods for the study of beta-carotene-d8 metabolism in humans utilizing tandem mass spectrometry and high-performance liquid chromatography. Analytical Chemistry, 66(23), 4177-4185.

Dueker, S. R., Lin, Y., Buchholz, B. A., Schneider, P. D., Lame, M. W., Segall, H. J., et al. (2000). Long-term kinetic study of beta-carotene, using accelerator mass spectrometry in an adult volunteer. Journal of Lipid Research, 41(11), 1790-1800.

During, A., Dawson, H. D., & Harrison, E. H. (2005). Carotenoid transport is decreased and expression of the lipid transporters SR-BI, NPC1L1, and ABCA1 is downregulated in caco-2 cells treated with ezetimibe. The Journal of Nutrition, 135(10), 2305-2312.

During, A., Doraiswamy, S., & Harrison, E. H. (2008). Xanthophylls are preferentially taken up compared with beta-carotene by retinal cells via a SRBI-dependent mechanism. Journal of Lipid Research, 49(8), 1715-1724.

During, A., & Harrison, E. H. (2006). Digestion and intestinal absorption of dietary carotenoids and vitamin A. In L. R. Johnson (Ed.), Physiology of the gastrointestinal tract, fourth edition (4th ed., pp. 1735). New York: Raven Press.

During, A., & Harrison, E. H. (2007). Mechanisms of provitamin A (carotenoid) and vitamin A (retinol) transport into and out of intestinal caco-2 cells. Journal of Lipid Research, 48(10), 2283-2294.

During, A., Nagao, A., Hoshino, C., & Terao, J. (1996). Assay of beta-carotene 15,15'- dioxygenase activity by reverse-phase high-pressure liquid chromatography. Analytical Biochemistry, 241(2), 199-205.

Edwards, A. J., Nguyen, C. H., You, C. S., Swanson, J. E., Emenhiser, C., & Parker, R. S. (2002). Alpha- and beta-carotene from a commercial puree are more bioavailable to humans than from boiled-mashed carrots, as determined using an extrinsic stable isotope reference method. The Journal of Nutrition, 132(2), 159-167.

Edwards, A. J., You, C. S., Swanson, J. E., & Parker, R. S. (2001). A novel extrinsic reference method for assessing the vitamin A value of plant foods. The American Journal of Clinical Nutrition, 74(3), 348-355.

110

Ellwood, K. C., Chatzidakis, C., & Failla, M. L. (1993). Fructose utilization by the human intestinal epithelial cell line, caco-2. Proceedings of the Society for Experimental Biology and Medicine.Society for Experimental Biology and Medicine (New York, N.Y.), 202(4), 440-446.

Eroglu, A., Hruszkewycz, D. P., Curley Jr, R. W., & Harrison, E. H. (2010). The eccentric cleavage product of beta-carotene, beta -apo-13-carotenone, functions as an antagonist of RXRalpha. Archives of Biochemistry and Biophysics,

Failla, M. L., Huo, T., & Thakkar, S. K. (2008). In vitro screening of relative bioaccessibility of carotenoids from foods. Asia Pacific Journal of Clinical Nutrition, 17 Suppl 1, 200-203.

Failla, M. L., Thakkar, S. K., & Kim, J. Y. (2009). In vitro bioaccessibility of beta- carotene in orange fleshed sweet potato (ipomoea batatas, lam.). Journal of Agricultural and Food Chemistry, 57(22), 10922-10927.

Fierce, Y., de Morais Vieira, M., Piantedosi, R., Wyss, A., Blaner, W. S., & Paik, J. (2008). In vitro and in vivo characterization of retinoid synthesis from beta- carotene. Archives of Biochemistry and Biophysics, 472(2), 126-138.

Fish, W. W. (2006). Interaction of sodium dodecyl sulfate with watermelon chromoplasts and examination of the organization of lycopene within the chromoplasts. Journal of Agricultural and Food Chemistry, 54(21), 8294-8300.

Fish, W. W. (2007). Novel procedure for the extraction and concentration of carotenoid- containing chromoplasts from selected plant systems. Journal of Agricultural and Food Chemistry, 55(4), 1486-1490.

Fleshman, M. K., Lester, G. E., Riedl, K. M., Kopec, R. E., Narayanasamy, S., Curley, R. W.,Jr, et al. (2011). Carotene and novel apocarotenoid concentrations in orange- fleshed cucumis melo melons: Determinations of beta-carotene bioaccessibility and bioavailability. Journal of Agricultural and Food Chemistry, 59(9), 4448-4454.

Flores, A. (2007, Orange-fleshed, organic honeydews! Agricultural Research Magazine, Vol. 55, 17.

Furr, H. C., & Clark, R. M. (1997). Intestinal absorption and tissue distribution of carotenoids. Nutr. Biochem., 8, 364-377.

Furr, H. C., Green, M. H., Haskell, M., Mokhtar, N., Nestel, P., Newton, S., et al. (2005). Stable isotope dilution techniques for assessing vitamin A status and bioefficacy of provitamin A carotenoids in humans. Public Health Nutrition, 8(6), 596-607.

111

Garcia-Calvo, M., Lisnock, J., Bull, H. G., Hawes, B. E., Burnett, D. A., Braun, M. P., et al. (2005). The target of ezetimibe is niemann-pick C1-like 1 (NPC1L1). Proceedings of the National Academy of Sciences of the United States of America, 102(23), 8132-8137.

Garrett, D. A., Failla, M. L., & Sarama, R. J. (1999). Development of an in vitro digestion method to assess carotenoid bioavailability from meals. Journal of Agricultural and Food Chemistry, 47(10), 4301-4309.

Garrett, D. A., Failla, M. L., & Sarama, R. J. (2000). Estimation of carotenoid bioavailability from fresh stir-fried vegetables using an in vitro digestion/Caco-2 cell culture model. The Journal of Nutritional Biochemistry, 11(11-12), 574-580.

Garrett, D. A., Failla, M. L., Sarama, R. J., & Craft, N. (1999). Accumulation and retention of micellar beta-carotene and lutein by caco-2 human intestinal cells. The Journal of Nutritional Biochemistry, 10(10), 573-581.

Ge, L., Wang, J., Qi, W., Miao, H. H., Cao, J., Qu, Y. X., et al. (2008). The cholesterol absorption inhibitor ezetimibe acts by blocking the sterol-induced internalization of NPC1L1. Cell Metabolism, 7(6), 508-519.

Goodman, D. S., Blomstrand, R., Werner, B., Huang, H. S., & Shiratori, T. (1966). The intestinal absorption and metabolism of vitamin A and beta-carotene in man. The Journal of Clinical Investigation, 45(10), 1615-1623.

Handelman, G. J., van Kuijk, F. J., Chatterjee, A., & Krinsky, N. I. (1991). Characterization of products formed during the autoxidation of beta-carotene. Free Radical Biology & Medicine, 10(6), 427-437.

Harrison, E. H. (2005). Mechanisms of digestion and absorption of dietary vitamin A. Annual Review of Nutrition, 25, 87-103.

Hedren, E., Diaz, V., & Svanberg, U. (2002). Estimation of carotenoid accessibility from carrots determined by an in vitro digestion method. European Journal of Clinical Nutrition, 56(5), 425-430.

Hedren, E., Mulokozi, G., & Svanberg, U. (2002). In vitro accessibility of carotenes from green leafy vegetables cooked with sunflower oil or red palm oil. International Journal of Food Sciences and Nutrition, 53(6), 445-453.

Hessel, S., Eichinger, A., Isken, A., Amengual, J., Hunzelmann, S., Hoeller, U., et al. (2007). CMO1 deficiency abolishes vitamin A production from beta-carotene and alters lipid metabolism in mice. The Journal of Biological Chemistry, 282(46), 33553-33561.

112

Ho, C. C., de Moura, F. F., Kim, S. H., & Clifford, A. J. (2007). Excentral cleavage of beta-carotene in vivo in a healthy man. The American Journal of Clinical Nutrition, 85(3), 770-777.

Hu, K. Q., Liu, C., Ernst, H., Krinsky, N. I., Russell, R. M., & Wang, X. D. (2006). The biochemical characterization of ferret carotene-9',10'-monooxygenase catalyzing cleavage of carotenoids in vitro and in vivo. The Journal of Biological Chemistry, 281(28), 19327-19338.

Hui, D. Y., Labonte, E. D., & Howles, P. N. (2008). Development and physiological regulation of intestinal lipid absorption. III. intestinal transporters and cholesterol absorption. American Journal of Physiology.Gastrointestinal and Liver Physiology, 294(4), G839-43.

Huo, T., Ferruzzi, M. G., Schwartz, S. J., & Failla, M. L. (2007). Impact of fatty acyl composition and quantity of triglycerides on bioaccessibility of dietary carotenoids. Journal of Agricultural and Food Chemistry, 55(22), 8950-8957.

Jones, P. J., Leitch, C. A., Li, Z. C., & Connor, W. E. (1993). Human cholesterol synthesis measurement using deuterated water. theoretical and procedural considerations. Arteriosclerosis and Thrombosis : A Journal of Vascular Biology / American Heart Association, 13(2), 247-253.

Kane, M. A., Chen, N., Sparks, S., & Napoli, J. L. (2005). Quantification of endogenous retinoic acid in limited biological samples by LC/MS/MS. The Biochemical Journal, 388(Pt 1), 363-369.

Kean, E. G., Hamaker, B. R., & Ferruzzi, M. G. (2008). Carotenoid bioaccessibility from whole grain and degermed maize meal products. Journal of Agricultural and Food Chemistry, 56(21), 9918-9926.

Kiefer, C., Hessel, S., Lampert, J. M., Vogt, K., Lederer, M. O., Breithaupt, D. E., et al. (2001). Identification and characterization of a mammalian enzyme catalyzing the asymmetric oxidative cleavage of provitamin A. The Journal of Biological Chemistry, 276(17), 14110-14116.

Kiefer, C., Sumser, E., Wernet, M. F., & Von Lintig, J. (2002). A class B scavenger receptor mediates the cellular uptake of carotenoids in drosophila. Proceedings of the National Academy of Sciences of the United States of America, 99(16), 10581- 10586.

Kopec, R. E., Riedl, K. M., Harrison, E. H., Curley, R. W.,Jr, Hruszkewycz, D. P., Clinton, S. K., et al. (2010). Identification and quantification of apo-lycopenals in

113

fruits, vegetables, and human plasma. Journal of Agricultural and Food Chemistry, 58(6), 3290-3296.

Kramer, W., Girbig, F., Corsiero, D., Burger, K., Fahrenholz, F., Jung, C., et al. (2003). Intestinal cholesterol absorption: Identification of different binding proteins for cholesterol and cholesterol absorption inhibitors in the enterocyte brush border membrane. Biochimica Et Biophysica Acta, 1633(1), 13-26.

Krinsky, N. I. (2001). Carotenoids as antioxidants. Nutrition (Burbank, Los Angeles County, Calif.), 17(10), 815-817.

Lemke, S. L., Dueker, S. R., Follett, J. R., Lin, Y., Carkeet, C., Buchholz, B. A., et al. (2003). Absorption and retinol equivalence of beta-carotene in humans is influenced by dietary vitamin A intake. Journal of Lipid Research, 44(8), 1591- 1600.

Lester, G. E. (2005). Supplemental foliar potassium applications durin muskmelon fruit development can improve fruit quality, ascorbic acid, and beta-carotene contents. J. AMER. Soc. Hort. Sci., 130(4), 649-653.

Lester, G. E. (2008). Antioxidant, sugar, mineral, and phytonutrient concentrations across edible fruit tissues of orange-fleshed honeydew melon (cucumis melo L.). Journal of Agricultural and Food Chemistry, 56(10), 3694-3698.

Lester, G. E., Baker, L. R., & Kelly, J. F. (1982). Physiology of sugar accumulation in carrot breeding lines and cultivars. J. AMER. Soc. Hort. Sci., 107, 381-387.

Lester, G. E., & Eischen, F. (1996). Beta-carotene content of postharvest orange-fleshed muskmelon fruit: Effect of cultivar, growing location and fruit size. Plant Foods for Human Nutrition (Dordrecht, Netherlands), 49(3), 191-197.

Leung, W. C., Hessel, S., Meplan, C., Flint, J., Oberhauser, V., Tourniaire, F., et al. (2009). Two common single nucleotide polymorphisms in the gene encoding beta- carotene 15,15'-monoxygenase alter beta-carotene metabolism in female volunteers. The FASEB Journal : Official Publication of the Federation of American Societies for Experimental Biology, 23(4), 1041-1053.

Levy, E., Menard, D., Suc, I., Delvin, E., Marcil, V., Brissette, L., et al. (2004). Ontogeny, immunolocalisation, distribution and function of SR-BI in the human intestine. Journal of Cell Science, 117(Pt 2), 327-337.

Lindqvist, A., & Andersson, S. (2004). Cell type-specific expression of beta-carotene 15,15'-mono-oxygenase in human tissues. The Journal of Histochemistry and Cytochemistry : Official Journal of the Histochemistry Society, 52(4), 491-499.

114

Lindqvist, A., He, Y. G., & Andersson, S. (2005). Cell type-specific expression of beta- carotene 9',10'-monooxygenase in human tissues. The Journal of Histochemistry and Cytochemistry : Official Journal of the Histochemistry Society, 53(11), 1403- 1412.

Lobo, M. V., Huerta, L., Ruiz-Velasco, N., Teixeiro, E., de la Cueva, P., Celdran, A., et al. (2001). Localization of the lipid receptors CD36 and CLA-1/SR-BI in the human gastrointestinal tract: Towards the identification of receptors mediating the intestinal absorption of dietary lipids. The Journal of Histochemistry and Cytochemistry : Official Journal of the Histochemistry Society, 49(10), 1253-1260.

Mills, J. P., Tumuhimbise, G. A., Jamil, K. M., Thakkar, S. K., Failla, M. L., & Tanumihardjo, S. A. (2009). Sweet potato beta-carotene bioefficacy is enhanced by dietary fat and not reduced by soluble fiber intake in mongolian gerbils. The Journal of Nutrition, 139(1), 44-50.

Mulokozi, G., Hedren, E., & Svanberg, U. (2004). In vitro accessibility and intake of beta-carotene from cooked green leafy vegetables and their estimated contribution to vitamin A requirements. Plant Foods for Human Nutrition (Dordrecht, Netherlands), 59(1), 1-9.

Novotny, J. A., Dueker, S. R., Zech, L. A., & Clifford, A. J. (1995). Compartmental analysis of the dynamics of beta-carotene metabolism in an adult volunteer. Journal of Lipid Research, 36(8), 1825-1838.

Novotny, J. A., Harrison, D. J., Pawlosky, R., Flanagan, V. P., Harrison, E. H., & Kurilich, A. C. (2010). Beta-carotene conversion to vitamin A decreases as the dietary dose increases in humans. The Journal of Nutrition, 140(5), 915-918.

Novotny, J. A., Kurilich, A. C., Britz, S. J., & Clevidence, B. A. (2005). Plasma appearance of labeled beta-carotene, lutein, and retinol in humans after consumption of isotopically labeled kale. Journal of Lipid Research, 46(9), 1896- 1903.

Olson, J. A. (1989). Provitamin A function of carotenoids: The conversion of beta- carotene into vitamin A. The Journal of Nutrition, 119(1), 105-108.

Omenn, G. S., Goodman, G. E., Thornquist, M. D., Balmes, J., Cullen, M. R., Glass, A., et al. (1996). Risk factors for lung cancer and for intervention effects in CARET, the beta-carotene and retinol efficacy trial. Journal of the National Cancer Institute, 88(21), 1550-1559.

115

Pawlosky, R. J., Flanagan, V. P., & Novotny, J. A. (2000). A sensitive procedure for the study of beta-carotene-d8 metabolism in humans using high performance liquid chromatography-mass spectrometry. Journal of Lipid Research, 41(6), 1027-1031.

Prakash, P., Russell, R. M., & Krinsky, N. I. (2001). In vitro inhibition of proliferation of estrogen-dependent and estrogen-independent human breast cancer cells treated with carotenoids or retinoids. The Journal of Nutrition, 131(5), 1574-1580.

Priyadarshani, A. M., & Chandrika, U. G. (2007). Content and in-vitro accessibility of pro-vitamin A carotenoids from sri lankan cooked non-leafy vegetables and their estimated contribution to vitamin A requirement. International Journal of Food Sciences and Nutrition, 58(8), 659-667.

Pullakhandam, R., & Failla, M. L. (2007). Micellarization and intestinal cell uptake of beta-carotene and lutein from drumstick (moringa oleifera) leaves. Journal of Medicinal Food, 10(2), 252-257.

Reboul, E., Richelle, M., Perrot, E., Desmoulins-Malezet, C., Pirisi, V., & Borel, P. (2006). Bioaccessibility of carotenoids and vitamin E from their main dietary sources. Journal of Agricultural and Food Chemistry, 54(23), 8749-8755.

Shmarakov, I., Fleshman, M. K., D'Ambrosio, D. N., Piantedosi, R., Riedl, K. M., Schwartz, S. J., et al. (2010). Hepatic stellate cells are an important cellular site for beta-carotene conversion to retinoid. Archives of Biochemistry and Biophysics,

Singh, H., & Cama, H. R. (1974). Enzymatic cleavage of carotenoids. Biochimica Et Biophysica Acta, 370(1), 49-61.

Sitte, P., Falk, H., & Liedvogel, B. (1980). Chromoplast. In F. C. Czygan (Ed.), Pigments in plants (2nd ed ed., pp. 1117-1148). Stuttgart, Germany: Gustav Fisher Verlag.

Stahl, W., & Sies, H. (1996). Lycopene: A biologically important carotenoid for humans? Archives of Biochemistry and Biophysics, 336(1), 1-9.

Tang, G. (2010). Bioconversion of dietary provitamin A carotenoids to vitamin A in humans. The American Journal of Clinical Nutrition, 91(5), 1468S-1473S.

Tang, G., Qin, J., Dolnikowski, G. G., & Russell, R. M. (2000). Vitamin A equivalence of beta-carotene in a woman as determined by a stable isotope reference method. European Journal of Nutrition, 39(1), 7-11.

Tang, G., Qin, J., Dolnikowski, G. G., & Russell, R. M. (2003). Short-term (intestinal) and long-term (postintestinal) conversion of beta-carotene to retinol in adults as

116

assessed by a stable-isotope reference method. The American Journal of Clinical Nutrition, 78(2), 259-266.

Tang, G., Qin, J., Dolnikowski, G. G., Russell, R. M., & Grusak, M. A. (2005). Spinach or carrots can supply significant amounts of vitamin A as assessed by feeding with intrinsically deuterated vegetables. The American Journal of Clinical Nutrition, 82(4), 821-828.

Thakkar, S. K., & Failla, M. L. (2008). Bioaccessibility of pro-vitamin A carotenoids is minimally affected by non pro-vitamin A xanthophylls in maize (zea mays sp.). Journal of Agricultural and Food Chemistry,

Thakkar, S. K., Maziya-Dixon, B., Dixon, A. G., & Failla, M. L. (2007). Beta-carotene micellarization during in vitro digestion and uptake by caco-2 cells is directly proportional to beta-carotene content in different genotypes of cassava. The Journal of Nutrition, 137(10), 2229-2233.

U.S. Department of Agriculture, Agricultural Research Service. (2010). USDA national nutrient database for standard reference, release 22, nutrient data laboratory.http://www.ars.usda.gov/nutrientdata van Bennekum, A., Werder, M., Thuahnai, S. T., Han, C. H., Duong, P., Williams, D. L., et al. (2005). Class B scavenger receptor-mediated intestinal absorption of dietary beta-carotene and cholesterol. Biochemistry, 44(11), 4517-4525. van Lieshout, M., West, C. E., & van Breemen, R. B. (2003). Isotopic tracer techniques for studying the bioavailability and bioefficacy of dietary carotenoids, particularly beta-carotene, in humans: A review. The American Journal of Clinical Nutrition, 77(1), 12-28. van Vliet, T., Schreurs, W. H., & van den Berg, H. (1995). Intestinal beta-carotene absorption and cleavage in men: Response of beta-carotene and retinyl esters in the triglyceride-rich lipoprotein fraction after a single oral dose of beta-carotene. The American Journal of Clinical Nutrition, 62(1), 110-116.

Vasquez-Caicedo, A. L., Heller, A., Neidhart, S., & Carle, R. (2006). Chromoplast morphology and beta-carotene accumulation during postharvest ripening of mango cv. 'tommy atkins'. Journal of Agricultural and Food Chemistry, 54(16), 5769-5776.

Vishnevetsky, M., Ovadis, M., Itzhaki, H., Levy, M., Libal-Weksler, Y., Adam, Z., et al. (1996). Molecular cloning of a carotenoid-associated protein from cucumis sativus corollas: Homologous genes involved in carotenoid sequestration in chromoplasts. The Plant Journal : For Cell and Molecular Biology, 10(6), 1111-1118.

117

Vogel,S., Gamble,M.V., Blaner, W.S. (1999). Retinoid uptake, metabolism and transport. In Nau,H., Blaner,W.S. (Ed.), The handbook of experimental pharmacology, the retinoids (pp. 31-96). Heidelberg: Springer Verlag.

Wang, S., Cyronak, M., & Yang, E. (2007). Does a stable isotopically labeled internal standard always correct analyte response? A matrix effect study on a LC/MS/MS method for the determination of carvedilol enantiomers in human plasma. Journal of Pharmaceutical and Biomedical Analysis, 43(2), 701-707.

Wang, X. D., Russell, R. M., Liu, C., Stickel, F., Smith, D. E., & Krinsky, N. I. (1996). Beta-oxidation in rabbit liver in vitro and in the perfused ferret liver contributes to retinoic acid biosynthesis from beta-apocarotenoic acids. The Journal of Biological Chemistry, 271(43), 26490-26498.

Wang, X. D., Tang, G. W., Fox, J. G., Krinsky, N. I., & Russell, R. M. (1991). Enzymatic conversion of beta-carotene into beta-apo-carotenals and retinoids by human, monkey, ferret, and rat tissues. Archives of Biochemistry and Biophysics, 285(1), 8-16.

Wang, Z., Jiao, H., Cao, M., Tang, G., Zhao, X., & Yin, S. (2003). Evaluation on intestinal and whole-body conversion of beta-carotene to vitamin A in chinese adults using a stable isotope reference method. Wei Sheng Yan Jiu = Journal of Hygiene Research, 32(3), 215-221.

Winterstein, A., & Hefedus, B. (1960). On the occurrence of retinene in nature]. Hoppe- Seyler's Zeitschrift Fur Physiologische Chemie, 321, 97-106.

World Health Organization. (2009). Global prevalence of vitamin A deficiency in populations at risk 1995–2005. Geneva, World Health Organization.

Zhou, J. R., Gugger, E. T., & Erdman, J. W.,Jr. (1996). The crystalline form of carotenes and the food matrix in carrot root decrease the relative bioavailability of beta- and alpha-carotene in the ferret model. Journal of the American College of Nutrition, 15(1), 84-91.

Ziouzenkova, O., Orasanu, G., Sharlach, M., Akiyama, T. E., Berger, J. P., Viereck, J., et al. (2007). Retinaldehyde represses adipogenesis and diet-induced obesity. Nature Medicine, 13(6), 695-702.

Ziouzenkova, O., & Plutzky, J. (2008). Retinoid metabolism and nuclear receptor responses: New insights into coordinated regulation of the PPAR-RXR complex. FEBS Letters, 582(1), 32-38.

118