.5.

fl ..
.. 3.. .L:., .

.. .11: a

 

 

3 HDDARY
I O Micnlydd State
University

 

 

 

This is to certify that the
dissertation entitled

COOKED COMMON BEANS (PHASEOLUS VULGARIS L.):
PHENOLIC COMPOUND COMPOSITION AND INFLUENCE
OF FRACTIONS ON COLON TUMOR DEVELOPMENT IN
OB/OB OBESE MICE

presented by

Kathleen Grace Barrett

has been accepted towards fulfillment
of the requirements for the

 

 

 

 

Doctoral degree in Human Nutrition
‘ Major Professor’s Signature
1 at ,W - 0 <3
Date

MSU is an Affinnative Action/Equal Opportunity Employer

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:lProj/Aoc&Pres/ClRC/DateDm.indd

fi

COOKED COMMON BEANS (PHASEOLUS VULGARIS L.): PHENOLIC
COMPOUND COMPOSITION AND INFLUENCE OF FRACTIONS ON COLON
TUMOR DEVELOPMENT IN OB/OB OBESE MICE

By

Kathleen Grace Barrett

A DISSERTATION
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Human Nutrition

2009

ABSTRACT
COOKED COMMON BEANS (PHASEOL US VULGARIS L.): PHENOLIC
COMPOUND COMPOSITION AND INFLUENCE OF FRACTIONS ON COLON
TUMOR DEVELOPMENT IN OB/OB OBESE MICE
By
Kathleen Grace Barrett
The consumption of dry beans has been correlated with the inhibition of several
chronic diseases such as cardiovascular disease, type H diabetes mellitus, obesity and
cancer. Dry beans contain a variety of phytochemicals (phytic acid, saponins,
oligosaccharides, and phenolic compounds) that may confer these health benefits.
Phenolic compounds have been shown to possess anti-cancer activity and other health
promoting effects. Phenolic compounds in raw common dry beans have been studied,
however, beans must be heat-treated (i.e. cooked) before consumption and there is little
known about the types and amounts of phenolic compounds found in cooked beans.
Hydroxybenzoic acids and hydroxycinnamic acids were identified in the four

cooked dry common beans. The hydroxybenzoic acids found were p-hydroxybenzoic
acid, vanillic acid, and syringic acid and total concentrations ranged fi'om 3 to 12 mg per
100 g bean flour. Protocatechuic acid was only identified in black, pinto and red beans.
The hydroxycinnamic acids identified were p-coumaric acid, caffeic acid, ferulic acid,
and sinapic acid in total concentrations ranging from llto 36 mg per 100 g bean flour.
Acid hydrolysis of the beans liberated greater quantities of hydroxybenzoic acids while
alkaline hydrolysis liberated greater quantities of hydroxycinnamic acids. Only black,

pinto, and red beans contained the flavan-3ol (+)-catechin (0.3 to 3 mg per 100 g bean

flour) and the flavonols quercetin and kaempferol (7 to 67 mg per 100 g bean flour).

Since dry beans may contain other colon cancer inhibiting compounds in addition
to phenolic compounds, cooked navy beans were fractionated into an aqueous-ethanol
soluble fraction (navy bean extract mix, NBEM; concentrated in saponins,
oligosaccharides, and phenolic compounds) and an aqueous-ethanol insoluble fi'action
(navy bean residue, NBR; concentrated in fiber and protein) to narrow the search for
which components inhibit the development of colon cancer. A control diet and diets

containing either cooked navy beans or its fractions were fed to azoxymethane (AOM)
injected mildly diabetic, obese mice (ob/ob; BG.V-Lep0b /J). This mouse model has not

been previously used to evaluate the effect of dietary ingredients on chemically induced

colon carcinogenesis.

The ob/ob mouse did not have the desired sensitivity to AOM to be recommended
for widespread use to test dietary ingredients for colon tumor inhibiting potential. The
low tumor incidence in the control fed animals weakened the statistical power to
differentiate colon cancer inhibition potential between the extract (NBEM) and the
residue (NBR). Nevertheless, compounds contained in the aqueous-ethanol extract of
navy beans (NBEM) reduced tumor incidence by 100% compared to the control diet.
Additionally, the inhibition of chemically induced colon carcinogenesis in ob/ob mice fed
navy bean and navy bean fractions was accompanied by a downregulation in the
expression c-fos in the colonic mucosa. c-Fos is required for normal cell cycle
progression and proliferation but is overexpressed in some cancers. The downregulation
of c-fos expression in the colon of ob/ob mice fed the navy bean and navy bean fractions

may inhibit carcinogenesis by maintaining normal colon crypt cell homeostasis.

DEDICATION

T o my father Michael T. Barrett, thank you for your unconditional love and support.
I love and miss you Papa.

Love always, Toots

iv

ACKNOWLEDGEMENTS

I would like to thank my advisor, Dr. Maurice R. Bennink, for his guidance and
continued support throughout my graduate program. I have learned so much from him
and will always be in his debt. Additionally, thank you to my committee members Dr.
Dale Romsos, Dr. Mike Orth, and Dr. Leslie Bourquin for their guidance, time and
patience. I would like to express my gratitude to my family and extended family,
especially my mom, sister and brother-in-law, for their love and support and extend my
love and thankfulness to my beautiful nieces, Parker Grace and Quinn Michael Elaine,
who bring a smile to my face everyday. Thank you to my colleague and friend Dr. Beth
Rondini. Her assistance with my research will forever be appreciated. I would also like
to thank her for lending an car when I needed to vent and for all the great conversations.
Finally and most of all, I would like to thank Cliff Carinci who has been my companion
and best friend throughout this long journey. Thank you for always believing in me

Cliffy, especially when I did not believe in myself. I love you.

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................... viii
LIST OF FIGURES ........................................................................................................... ix
LIST OF ABBREVIATIONS ........................................................................................... xii
CHAPTER I.
INTRODUCTION ............................................................................................................... 1
CHAPTER H.
REVIEW OF LITERATURE
A. Legumes (i.e. DryBeans) .................................................................................... 4
B. Phenolic Compounds ........................................................................................... 5
1.1. Background ........................................................................................... 5
1.2. Classification of Phenolic Compounds ................................................. 7
1.3. Biosynthesis of Phenolic Compounds in Plants .................................. 13
1.4. Analytical Methods for Extraction, Purification, Separation and
Detection .................................................................................................... 1 5
1.5. Intake, Absorption and Metabolism of Phenolic Compounds ............ 27
1.6. Phenolic Compounds in Grain Legumes ............................................ 35
C. Colorectal Cancer (CRC) .................................................................................. 48
1.1. Colorectal Cancer Statistics ............................................................... 49
1.2. Colorectal Carcinogenesis .................................................................. 52
1.3. Colorectal Cancer and Diet ................................................................ 66
1.4. Colorectal Cancer and Legumes ........................................................ 69
D. Rationale ........................................................................................................... 74
CHAPTER HI.

PHENOLIC COMPOUNDS IDENTIFIED AND QUANTIFIED IN COOKED DRY
COMMON BEANS (PHASEOL US VULGARIS L.)

A. ABSTRACT ..................................................................................................... 79

B. INTRODUCTION ........................................................................................... 81

C. MATERIALS AND METHODS ..................................................................... 83

D. RESULTS ........................................................................................................ 91

E. DISCUSSION ................................................................................................ 113
CHAPTER IV.

INHIBITORY EFFECTS OF COOKED NAVY BEAN (PHASEOL US VULGARIS L.)
ON AZOXYMETHANE (AOM)-INDUCED COLON CANCER IN OB/OB OBESE
MICE
A. ABSTRACT ................................................................................................... 122
B. INTRODUCTION ......................................................................................... 124

vi

C. MATERIALS AND METHODS ................................................................... 130

D. RESULTS ...................................................................................................... 135
E. DISCUSSION ................................................................................................ 139
CHAPTER V.

DIETS CONTAINING NAVY BEAN (PHASEOL US VULGARIS L.) OR NAVY BEAN
FRACTIONS INHIBIT AZOXYMETHANE (AOM)-INDUCED COLON CANCER IN
OB/OB MICE BY MODULATING THE TGF-B SIGNALING PATHWAY

A. ABSTRACT ................................................................................................... 146
B. INTRODUCTION ......................................................................................... 148
C. MATERIALS AND METHODS ................................................................... 150
D. RESULTS ...................................................................................................... 154
E. DISCUSSION ................................................................................................ 158
CHAPTER VI.
CONCLUSION ................................................................................................................ 162
CHAPTER VII.
FUTURE STUDIES ......................................................................................................... 166
APPENDICES ................................................................................................................. 168
LITERATURE CITED .................................................................................................... 193

vii

LIST OF TABLES

Table 1. Classes of Phenolic Compounds ........................................................................... 8
Table 2. Phenolic Compounds Quantified in Grain Legumes .......................................... 39
Table 3. Unconjugated Phenolic Compounds (mg / 100 g of Bean Flour) ...................... 187
Table 4. Unconjugated Phenolic Compounds (mg / 100 g Bean Fraction) ..................... 188
Table 5. Unconjugated Phenolic Compounds + Phenolic Compounds Liberated by
Alkaline Hydrolysis (mg / 100 g of Bean Flour) .............................................................. 189
Table 6. Unconjugated Phenolic Compounds + Phenolic Compounds Liberated by
Alkaline Hydrolysis (mg / 100 g of Bean Fraction) ......................................................... 190
Table 7. Unconjugated Phenolic Compounds + Phenolic Compounds Liberated by Acid
Hydrolysis (mg / 100 g of Bean Flour) ............................................................................. 191
Table 8. Unconjugated Phenolic Compounds + Phenolic Compounds Liberated by Acid
Hydrolysis (mg / 100 g of Bean Fraction) ........................................................................ 192
Table 9 a.—b. Phenolic Compounds Identified by HPLC-ECD in Whole Bean Flours and
Navy Bean Fractions ........................................................................................................ 108
Table 10. Composition of Diets Fed to ob/ob Mice ........................................................ 132

Table 11. Lesion Incidence in Colons of ob/ob Mice Fed Control or Navy Bean Diets 137
Table 12. Tumor Incidence in Colons of ob/ob Mice Fed Control or Navy Bean Diet..138

Table 13. Genes in the Mouse TGF-B-BMP Signaling Pathway PCR Array ................. 155

viii

LIST OF FIGURES

Figure 1. Structures of Phenolic Compounds ..................................................................... 9
Figure 2. Shikimic Acid Pathway ..................................................................................... 14
Figure 3. Absorption and Metabolism of Phenolic Compounds ...................................... 29
Figure 4. Normal Colonic Crypt ....................................................................................... 53

Figure 5. Adenoma-Carcinoma Sequence: A Genetic Model for Colorectal
Carcinogenesis ................................................................................................................... 55

Figure 6. Chromatogram of Standard Mix 1: Protocatechuic acid, (+)-Catechin, Vanillic
acid, Ferulic acid ................................................................................................................ 92

Figure 7. Chromatogram of Standard Mix 2: p-Hydroxybenzoic acid, Chlorogenic acid,
Syringic acid, Sinapic acid ................................................................................................. 93

Figure 8. Chromatogram of Standard Mix 3: Isovanillic acid, p-Coumaric acid,
Kaempferol ........................................................................................................................ 94

Figure 9. Chromatogram of Standard Mix 4: Caffeic acid, o—Coumaric acid, Quercetin.95

Figure 10a.—-c.

a. Chromatogram of Non-hydrolyzed Black Bean ............................................. 169

b. Chromatogram of Alkaline Hydrolyzed Black Bean ...................................... 170

c. Chromatogram of Acid Hydrolyzed Black Bean ............................................ 171
Figure 11a.—c.

a. Chromatogram of Non-hydrolyzed Pinto Bean .............................................. 172

b. Chromatogram of Alkaline Hydrolyzed Pinto Bean ....................................... 173

c. Chromatogram of Acid Hydrolyzed Pinto Bean ............................................. 174
Figure 12a.—c.

a. Chromatogram of Non-hydrolyzed Red Bean ................................................ 175

b. Chromatogram of Alkaline Hydrolyzed Red Bean ......................................... 176

c. Chromatogram of Acid Hydrolyzed Red Bean ............................................... 17 7
Figure 13a.—c.

a. Chromatogram of Non-hydrolyzed Navy Bean .............................................. 17 8

b. Chromatogram of Alkaline Hydrolyzed Navy Bean ...................................... 179

c. Chromatogram of Acid Hydrolyzed Navy Bean ............................................. 180

ix

Figure 14a.—c.

a. Chromatogram of Non-hydrolyzed NBR ........................................................ 181
b. Chromatogram of Alkaline Hydrolyzed NBR ................................................ 182
c. Chromatogram of Acid Hydrolyzed NBR ...................................................... 183
Figure 15a.—c.
a. Chromatogram of Non-hydrolyzed NBEM .................................................... 184
b. Chromatogram of Alkaline Hydrolyzed NBEM ............................................. 185
c. Chromatogram of Acid Hydrolyzed NBEM ................................................... 186
Figure 16a.—d. Hydroxybenzoic acids in Bean Flours
a. Protocatechuic acid ........................................................................................... 97
b. p-Hydroxybenzoic acid ..................................................................................... 97
c. Vanillic acid ...................................................................................................... 98
d. Syringic acid ..................................................................................................... 98
Figure 17a.—d. Hydroxycinnamic acids in Bean Flours
a. Caffeic acid ....................................................................................................... 99
b. p-Coumaric acid ................................................................................................ 99
c. Ferulic acid ...................................................................................................... 100
d. Sinapic acid ..................................................................................................... 100
Figure 18. F1avan-3ols in Bean Flours ((+)-Catechin ...................................................... 101
Figure l9a.—b. Flavonols in Bean F lours
a. Quercetin ......................................................................................................... 102
b. Kaempferol ..................................................................................................... 102
Figure 20a.—c. Hydroxybenzoic acids in Navy Bean Fractions
a. p-Hydroxybenzoic acid ................................................................................... 103
b. Vanillic acid .................................................................................................... 103
c. Syringic acid .................................................................................................... 104
Figure 21a.—d. Hydroxycinnamic acids in Navy Bean Fractions
a. Caffeic acid ..................................................................................................... 105
b. p-Coumaric acid .............................................................................................. 105
c. Ferulic acid ...................................................................................................... 106
d. Sinapic acid ..................................................................................................... 106

Figure 22. Phenolic Compounds Identified in Whole Bean Flours by HPLC-ECD
(mg/ 100 g of Bean Flour) ................................................................................................ 109

Figure 23. Phenolic Compounds Identified in Navy Bean Fractions by HPLC-ECD

(mg/ 100 g of Bean Fraction) ............................................................................................ 110
Figure 24 a.—b. Total Identified Phenolics and Total Phenolic Content ........................ 112
Figure 25. Growth Chart of Age-matched ob/ob Mice .................................................... 136

xi

LIST OF ABBREVIATIONS

(+)-CE ............................................................................................ (+)—Catechin Equivalent
ACF ....................................................................................................... Aberrant Crypt Foci
AICR ...................................................................... American Institute for Cancer Research
AMH ............................................................................................. Anti-Mullerian Hormone
AOM ............................................................................................................. Azoxymethane
APC ............................................................................................ Adenomous Polyposis Coli
APCI ............................................................... Atomospheric Pressure Chemical Ionization
BB .......................................................................... Cooked, Dried and Ground Black Bean
BCAC ................................................................................... B-Catenin Accumulated Crypts
BMP ......................................................................................... Bone Morphogenic Proteins
C ................................................................................................................................. Carbon
Caf ..................................................................................................................... Caffeic Acid
Cat .................................................................................................................... (+)-Catechin
CDK .............................................................................................. Cyclin dependent Kinase
COX ............................................................................................................ Cyclooxygenase
CRC ........................................................................................................... Colorectal Cancer
CVD ................................................................................................. Cardiovascular Disease
DAD ........................................................................................... Photodiode Array Detector
DCC ............................................................................................... Deleted in Colon Cancer
DMH ............................................................................................... 1, 2-Dimethylhydrazine
ECD ............................................................................................. Electrochemical Detection

xii

ECM ...................................................................................................... Extracellular Matrix

ESI .................................................................................................... Electrospray Ionization
FA ..................................................................................................................... Ferulic Acid
FAB ............................................................................................... F ast Atom Bombardment
FAP .................................................................................. Familial Adenomatous Polyposis
GC ....................................................................................................... Gas Chromatography
GSK-3B ...................................................................................... Glycogen Synthase Kinase
HNPCC .......................................................... Hereditary Non-Polyposis Colorectal Cancer
HPLC ............................................................... High Performance Liquid Chromatography
IGFBP .............................................................. Insulin-like Growth Factor Binding Protein
INK .................................................................................................. Jun N-Terminal Kinase
K ......................................................................................................................... Kaempferol
LAP ........................................................................................... Latency Associated Peptide
LOH ................................................................................................. Loss of Heterozygosity
LPH .......................................................................................... Lactose Phlorizin Hydrolase
LTBPs ................................................................................. Latent TGF-B Binding Proteins
LTGF-B ..................................... Latent Transforming Growth Factor beta protein complex
m- ................................................................................................................................. meta-
MAPKs ......................................................................... Mitogen Activated Protein Kinases
MDF ..................................................................................................... Mucin Depleted Foci
MMR .......................................................................................................... Mismatch Repair
MS ........................................................................................................... Mass Spectrometer
MSI ................................................................................................ Microsatellite Instability

xiii

MSS ...................................................................................................... Microsatellite Stable

NB ........................................................................... Cooked, Dried and Ground Navy Bean
NBEM ............................................................................................. Navy Bean Extract Mix
NBR ....................................................................................................... Navy Bean Residue
NHANES ............................................. National Health and Nutrition Examination Survey
NMR ...................................................................................... Nuclear Magnetic Resonance
o- .................................................................................................................................. ortho-
O ................................................................................................................................ Oxygen
ORAC ......................................................................... Oxygen Radical Absorbing Capacity
p- ................................................................................................................................... para-
PAI-l .................................................................................. Plasminogen Activator Inhibitor
PB ............................................................................ Cooked, Dried and Ground Pinto Bean
PC ........................................................................................................ Phenolic Compounds
PC ..................................................................................................... Paper Chromatography
PCA ....................................................................................................... Protocatechuic Acid
PCA ............................................................................................................ p -Coumaric Acid
PD ........................................................................................................... Plasma Desorption
MALDI .......................................................... Matrix-Assisted Laser Desorption Ionization
PHA ....................................................................................................... Phytohemagglutinin
PHBA .............................................................................................. p-Hydroxybenzoic Acid
PI3K ............................................................................................ Phosphoinositide Kinase-3
Q ............................................................................................................................. Quercetin
RB . ............................................................................ Cooked, Dried and Ground Red Bean

xiv

RER ............................................................................................................ Replication Error

RP-HPLC ................................ Reverse Phase- High Performance Liquid Chromatography
SA ..................................................................................................................... Sinapic Acid
SCFA ............................................................................................... Short Chain Fatty Acids
SFE ......................................................................................... Supercritical Fluid Extraction
SGLTl ................................................................ Sodium Dependent Glucose Transporter 1
SI ................................................................................................................... Small Intestine
SPE .................................................................................................... Solid Phase Extraction
Syr ................................................................................................................... Syringic Acid
TGF-B .............................................................................. Transforming Growth Factor Beta
TIP ................................................................................................ Total Identified Phenolics
TLC .......................................................................................... Thin Layer Chromatography
TPC .................................................................................................. Total Phenolic Content
USDA ................................................................... United States Department of Agriculture
UV/vis .................................................................................................... Ultraviolet / Visible
UV ........................................................................................................................ Ultraviolet
v- ..................................................................................................................................... vic-
Van ................................................................................................................... Vanillic Acid
WCRF .................................................................................... World Cancer Research Fund

XV

CHAPTER I. INTRODUCTION

The consumption of diets high in fruits, vegetables, grains and legumes has been
associated with a reduced risk for developing chronic diseases such as cancer and
cardiovascular disease (Steinmetz and Potter, 1993; Fraser, 1999; Genkinger etal., 2004;
WCRF/AICR, 2007). The general recognition that plant foods are important for good
health led researchers to investigate which components in plant foods impart these
beneficial biological properties. Suggestions that oxidative damage to DNA, RNA and
proteins leads to the initiation and progression of certain diseases spurred interest in
antioxidants in foods and beverages. There is a plethora of scientific and non-scientific
publications extolling the virtues of antioxidants found in various foods and beverages.
Sorting fact fiom fiction is often difficult, particularly for consumers.

Initially, nutritionists encouraged consumption of whole grain cereal products due
to the presence of more minerals, vitamins and fiber in whole grains versus products
made from refined grains. Now there is an expanded push for consumption of whole
grains to increase the dietary intake of non-nutrients such as phytates, sterols, and
phenolic compounds. The consumption of dry common beans is typically encouraged
because of the protein, fiber, and specific mineral and vitamin contributions to the diet.
However dry beans are grain legumes and impart health benefits typically associated with
the consumption of whole grains. Beans contain a greater quantity of potentially
beneficial phenolic compounds than cereal grains (Adom and Liu, 2002; Adorn et al.,

2003; Heimler et al., 2005; Xu and Chang; 2007).

Numerous studies indicate that incorporating beans into the diet could aid in the
prevention and/or management of chronic diseases such as diabetes, obesity, and cancer.
A significant portion of the inverse relationship between bean consumption and chronic
disease is related to the low glycemic index of beans and the reduced glycemic load when
beans are consumed. However, it is now becoming apparent that compounds
traditionally considered as anti-nutrients —— phytic acid, plant sterols, phenolic
compounds, enzyme inhibitors, and lectins — are beneficial, particularly in affluent
Western cultures. The potential for phenolic compounds in beans to impart health
benefits has not been extensively studied.

Epidemiologic studies suggest that bean consumption is inversely related to colon
cancer incidence and mortality. Studies with animal models of human colon cancer
consistently demonstrate that chemically induced colon cancer is reduced by more than
50% by bean containing diets (Hughes and Ganthavom, 1997; Hangen and Bennink,
2002; Bennink and Barrett, 2004; Rondini, 2006). The component in beans responsible
for the colon cancer inhibition and the mechanism of colon cancer inhibition when beans
are consumed are not known. Obesity and Type 2 diabetes increase the odds of
developing colon cancer (WCRF/AICR, 2007). To date, only lean animal models of
colon cancer have been fed beans.

One of the most consistent findings from studies investigating mechanisms
leading to colon cancer is the dysregulation of colonic crypt cytokinetics. It is well
documented that mutations of the so-called “gatekeeper” gene (APC gene) lead to colon

cancer. Dietary reduction of colon cancer proceeds through mechanisms that influence

epithelial cytokinetics prior to mutations of the APC gene (Lipkin, 1983; F earon and
Vogelstein, 1990).
The overall goals of this research are to:
1. Identify and quantify the principal phenolic compounds in cooked beans;
2. Determine if the phenolic content of beans is related to the development of colon
cancer in an obese, diabetic animal model; and
3. Determine if there is dysregulation of gene expression in pathways that are

intricately involved with colon crypt cytokinetics.

CHAPTER II. REVIEW OF LITERATURE

A. Legumes (i.e. Dry Beans)

Dry or common beans (Phaseolus vulgaris L.) belong to the family of plants
called legumes. Legumes are recognized by their ability to “fix” nitrogen and are
classified as: oil seeds such as soybeans and peanuts; and grain legumes such as common
beans, lentils, chickpeas, common peas, cowpeas, lirna and fava beans (Geil and
Anderson, 1994). The high nutritive value of dry beans, as well as other legumes, is well
established. Besides being high in protein and fiber, they contain no cholesterol and are
low in fat, sodium, and calories. Additionally, they contribute iron, phosphorous,
magnesium, manganese, potassium, copper, calcium, zinc, and the B vitamins; folate,
thiamin, riboflavin and niacin to the diet. Due to their low cost and high protein content,
they are often employed as a substitute for meat and meat products. Dry beans have long
been a dietary staple in the traditional diets consumed by populations in developing
countries in Asia, Africa, and in Central and South America. Even though the United
States produces over 1/3 of the world’s dry bean supply, the inclusion of dry beans into
the typical western diet of the US. remains low. In fact, less than 1/3 of the US. adult
population eats beans during any 3-day period (FASEB, 1995). Commonly consumed
dry beans in the US. include pinto, navy, kidney, Great Northern, and lima beans. In
1995 the annual kg per person was 1.5, 0.8, 0.3, 0.2, and 0.1 for pinto, navy, kidney,
Great Northern, and lima beans respectively in the US. (USDA, 1997). Compare this to

the annual 20 kg per person consumption of just black beans in Brazil (Ribeiro et al.,

2003).

Legumes, including dry beans, contain compounds often referred to as non-
nutrients such as amylase inhibitors, protease inhibitors (trypsin inhibitors, trypsin and
chymotrypsin inhibitors), lectins, phytates (inositol hexaphosphate), saponins,
oligosaccharides (raffinose, stachyose and verbacose), and phenolic compounds
(Rochfort and Panozzo, 2007). Some of these non-nutrients are referred to as anti-
nutrients because they interfere with protein digestibility (i.e. trypsin inhibitors, trypsin
and chymotrypsin inhibitors and some phenolic compounds), and mineral bioavailability
(i.e. phytates and some phenolic compounds). Additionally, beans are known to induce
discomfort, bloating and flatulence because of oligosaccharides (Champ, 2002; Rochfort
and Panozzo, 2007). Many of these factors however are destroyed or inactivated with
common processing methods such as heat treatment (i.e. boiling, pressure cooking or
autoclaving), hydration (soaking), germination or fermentation (Champ, 2002; Ibrahim et
al., 2002; Shimelis and Rakshit, 2007). Humans do not consume beans without
processing. Processing improves the palatability, aroma, and acceptability of legumes,
while concomitantly improving protein digestibility and mineral bioavailability
(Tharanathan and Mahadevamma, 2003). Processing also induce compositional changes
to dietary fiber (Kutos et al., 2003; Costa et al., 2006).

B. Phenolic Compounds

1.]. Background

Phenolic compounds belong to an array of structurally diverse secondary plant
metabolites, which are distributed throughout the plant kingdom. They are synthesized
during normal plant development and in response to trauma, such as pathological

infection, UV radiation or wounding (Stalikas, 2007). Although the physiological

function of some phenolic compounds are known, i.e. anthocyanins act as flower
pigments and lignins as structural components of the cell wall (Harbome, 1998), the roles
of other classes of phenolic compounds have yet to be fully understood. Research has
suggested phenolic compounds can act as bactericides, fungicides and as deterrence
factors for predators (i.e. insects and animals). They also may play a role in plant growth
regulation and sexual reproduction (Ribéreau-Gayon, 1972; Harbome, 1998;
Balasundram, 2006; Stalikas, 2007).

The biosynthesis, metabolism, physiological fimctions, and chemical structures of
phenolic compounds have been studied in plant biochemistry for decades (Bravo, 1998).
Food scientists have investigated their organoleptic properties in plant foods and
beverages, such as color, flavor, astringency, and hardness (Krygier, 1982; Bravo, 1998;
Robbins, 2003), as well as their effect on fi'uit maturation, role in enzymatic browning
and as a food preservative (Ribéreau-Gayon, 1972; Harbome, 1998; Lopez-Amoros et al.,
2006). Often their ability to bind proteins and metal ions, leading to reduced protein
digestion and mineral absorption, perpetuated the idea that phenolics also acted as anti-
nutrients (Krygier, 1982; Hu et al., 2006). In contrast, scientists in various disciplines
have now become interested in the potentially beneficial properties of phenolics in
humans. Phenolic compounds are most recognized as antioxidants and are capable of
scavenging different radicals generated in in vitro systems, as well as chelating metal ions
that generate reactive hydroxyl radicals (Croft, 1998; Hollman, 2001). Many propose
that their antioxidant properties are responsible for the protection against cardiovascular
disease, as well as cancer. Additionally, plant phenolic compounds possess anti-

inflammatory and antirnutagenic activity (Hollman, 2001).

1.2. Classification of Phenolic Compounds

In nature phenolic compounds are found either linked to sugars as glycosides or
to organic acids as esters, as such they tend to be water-soluble (Harbome, 1998). They
are not found in their parent (termed aglycone) or fiee form in nature, with the exception
of catechin (Ribéreau-Gayon, 1972; Harbome, 1998). Thus phenolic compounds can
possess the same parent/aglycone but be linked to different sugar moieties or organic
acids and these occur in multiple combinations (Harbome, 1998). Additionally, some
phenolic compounds form polymers, as is the case with lignins and tannins (condensed
and hydrolysable) (Ribéreau-Gayon, 1972). Phenolic compounds are not unifome
distributed in the plant tissues and contents can vary depending on plant species, variety,
time of harvest, maturation, germination, and storage conditions (storage time and
temperature) (Tsao, 2004). These factors increase the difficulty and complexity in
identifying and quantifying specific compounds and classes of phenolic compounds in
plant foods.

There are at least 8,000 naturally occurring phenolic compounds (Robbins, 2003;
Balasundram, 2006; Stalikas, 2007), which possess a common aromatic ring bearing one
or more hydroxyl substituents. This shared structure is referred to as a phenol (Harbome,
1998, Stalikas, 2007). They are further categorized into different classes and subclasses
based on the number of phenol groups (Harbome, 1998; Robbins, 2003). Table 1 lists the
classes of phenolic compounds found in plants. The chemical structures of flavonoids

and the other common classes of phenolic compounds are shown in Figure l. The
phenylpropanoids, C6-C3, are structurally the most common and most important, since a

large amount of phenolic compounds possess this basic carbon structure (Ribéreau-

Table 1. Classes of Phenolic Compounds'I

 

 

Class Carbon Structure
Simple Phenols C6
Hydroxybenzoic Acids and Related compounds C6-C1
Acetophenones and Phenylacetic Acids C6-C2
Phenylpropanoids (Hydroxycinnamic Acids, Coumarins, C _ C
Chromones) 6 3
Flavones C6-C3-C6
Flavanones C6-C3-C6
Isoflavanoids C6-C3-C6
Flavonols, Dihydroflavonols, and Related Compounds C6-C3-C6
Anthocyanidins C6-C3-C5
Chalcones, Aurones and Dihydrochalcones C6-C3-C5
Biflavonyls (C6-C3-C6)2
Lignans (C6-C3)2
Lignins (Chg-(33)“
Melanins (C5)n

Condensed Tannins (Oligomers of catechins and

leucoanthocyanidins = Proanthocyanidins) (CG'C3'C6)“
Hydrolysable Tannins ((36.01)n
Benzophenones, Xanthones and Stilbenes C6-C1-C6, C6'C2'C6
Quinones C6, C10, and C14
Betacyanins C18

 

a. Table from the following references Ribéreau-Gayon, 1972 and Harbome, 1998.

A. Flavonols

HO
0 I 3
OH

OH O
R1=R3=H, R2=OH, Kaempferol
R1=R2=OH, R3=H, Quercetin
R1=R2=R3=OH, Myricetin

C. Isoflavones

    

R2 O

R1=R3=OH, R2=H, Daidzein
R1=R2=R3=OH, Genistein

E. Flavan-3ols

OH
R1=R2=OH, R3=H, (+)-Catechin

B. Flavones

 

OHO

R1=R3=H, R2=OH, Apigenin
R1=R2=OH, R3=H, Luteolin

D. Anthocyanidins
R1

R2
HO 0" O
o . R
/ OH
OH
R1=R2=OH, R3=H, Cyanidin
R1=R2=R3=OH, Delphinidin
Rl=OCH3, R2=OH, R3=H, Peonidin

R1=OCH3, R2=R3=OH, Petunidin
R1=R3=OCH3, R2=OH, Malvidin

F. Leucoanthocyanidins (F1avan—3, 4diols)
R1

R2
.. . 0
0 ..
OH

OH OH
R1=R2=OH, R3=H, Leucocyanidin

Figure 1. Structures of Phenolic Compounds (Ribéreau-Gayon, 1972)

G. Hydroxybenzoic Acids
R3
32 o
R1

R1=R2=R3=OH, Gallic Acid

R1=R3=H, R2=OH, p-Hydroxybenzoic Acid
R1=H, R2=R3=OH, Protocatechuic Acid
R1=H, R2=OH, R3=OCH3, Vanillic Acid

R1=H, R2=OCH3, R3=OH, Isovanillic Acid
R1=R3=OCH3, R2=OH, Syringic Acid

H. Hydroxycinnamic Acids

R3 R4
CZC—COH
R2 0 H H 2
R1

R1=R3=R4=H, R2=OH, p-Coumaric Acid
R1=R2=R3=H, R4=OH, o-Coumaric Acid
R1=R4=H, R2=R3=OH, Caffeic Acid

R1=R4=H, R2=OH, R3=OCH3, Ferulic Acid
R1=R3=OCH3, R2=OH, R4=H, Sinapic Acid

1. Basic Carbon Skeleton of Flavonoids

 

Figure l. (cont’d)

10

Gayon, 1972). The flavonoids, C6-C3-C6, represent the largest group of phenolic

compounds (Balasundram, 2006; Stalikas, 2007).

Structurally, phenolic compounds are comprised of two phenol groups linked by a
3-carbon chain, often a heterocyclic ring linked to an oxygen molecule, except in the case
of chalcones that lack the heterocyclic ring, Figure 1, I (Ribéreau-Gayon, 1972;
Harbome, 1998; Balasundram, 2006; Stalikas, 2007). The flavonoids can be divided
further based on the level of oxidation and saturation in the heterocyclic ring, as well as
the substitution pattern (Ribéreau-Gayon, 1972). The flavonoids include flavonols,
flavones, flavanones, isoflavones, anthocyanidins and the less widely distributed
chalcones, dihydrochalcones, and aurones (Ribéreau-Gayon, 1972; Harbome, 1998). The
degree and pattern of hydroxylation, and the alkylation and/or glycosylation of these
hydroxyl groups in the subclasses of flavonoids, contributes to their structural diversity
(Rice-Evans et al., 1996; Stalikas, 2007). The A ring is meta-dihydroxylated (C5 and C7)
(Ribéreau-Gayon, 1972; Croft, 1998). The B ring is either monohydroxylated (C4’),
ortho-dihydroxylated (C3’ and C4’, most common) or vic-trihydroxylated (C3’, C4’ and
C5”) (Ribéreau-Gayon, 1972; Rice-Evans et al., 1996; Croft, 1998).

Flavonols (Figure 1, A), —- kaempferol, quercetin, myricetin — are the
predominant flavonoids distributed in the plant kingdom. They are found as glycosides,
preferentially bound to glucose, but also to galactose, rhamnose, arabinose and xylose
(Rice-Evans et al., 1996, Harbome, 1998). The preferred site of glycosylation on
flavonols is at the C3 in the heterocyclic Cring, and less fiequently at C7 in the A ring

(Rice-Evans et al., 1996).

11

Flavones (Figure 1, B) do not possess the C3 hydroxyl group of flavonols, but are
also found as glycosides. However, unlike flavonols, they are not widely distributed in
the plant kingdom and are predominantly C-glycosylated, versus O-glycosylated, at C7 in
the A ring (Harbome, 1998). Two common flavones are apigenin and luteolin.

Isoflavones such as genistein and daidzein, are found as glycosides and are
limited to select plant types, i.e. soybeans O-Iarbome, 1998). Anthocyanidins (aglycones)
are highly unstable and not found in this free form (Manach et al., 2004).

Anthocyanidins (Figure 1, D) are instead found as glycosides, termed anthocyanins, as
well as esterified to phenolic and organic acids and sometimes in complexes with other
flavonoids (Manach et al., 2004).

F Iavan-3018 (catechins) and leucoanthocyanidins (flavan—3,4diols) both serve as
precursors for condensed tannins (Figure 1, E and F) (Rommel and Wrolstad, 1993;
Harbome, 1998; Klepacka and F omal, 2006). Catechins, unlike most flavonoids, can be
found as monomers in nature, as well as, gallic acid esters (gallocatechin). The
condensed tannins, which are found universally throughout the plant kingdom, are also
called proanthocyanidins. They are oligomers and polymers of catechins and
leucoanthocyanidins (Ribéreau-Gayon, 1972; Harbome, 1998; Manach et al., 2004).
Although they are not structurally related to anthocyanidins, proanthocyanidins will yield
anthocyanidins when heated in acidic solutions (Ribéreau—Gayon, 1972; Rommel and
Wrolstad, 1993; Harbome, 1998).

Hydroxybenzoic acids (Figure 1, G) and their related compounds are usually
found esterified to alcohol insoluble lignins or as alcohol soluble simple glycosides. p-

Hydroxybenzoic acid, protocatechuic acid, vanillic acid, and syringic acid are the most

12

common and universally widespread hydroxybenzoic acids in the plant kingdom
(Harbome, 1998). Individual gallic acid subunits and dimers of gallic acid, referred to as
ellagic acid, are found in a polymeric form and classified as hydrolyzable tannins
(Ribéreau-Gayon, 1972; Harbome, 1998). Hydrolyzable tannins are limited to
dicotyledonous plants i.e. legumes, oaks, sunflowers (Harbome, 1998).

Hydroxycinnamic acids (Figure 1, H), the most widespread phenylpropanoids, are
found conjugated to organic acids such as quinic acid, tartaric acids, malic acids, and
rosmarinic acid via ester bonds, esterified to hemicelluloses (i.e. arabinoxylans,
xyloglucans) in the cell wall and as soluble forms in the cytoplasm (Harbome, 1998;
Faulds and Williamson, 1999). They can only be found in their free form if released
during food processing (Manach et al., 2004). Additionally they can provide the building
blocks for lignins, be found as dimers cross-linked to hemicelluloses in the cell wall; as
amides of amino compounds; esters of lipids; esters of polysaccharides, simple sugars
and sugar alcohols; esters of glycosides of anthocyanins, flavonols and diterpenes; and as
glycosides (Clifford, 1999; Naczk and Shahidi, 2006). Common hydroxycinnamic acids
include ferulic acid, sinapic acid, caffeic acid, and p-coumaric acid (Harbome, 1998).

1.3. Biosynthesis of Phenolic Compounds in Plants

All phenolic compounds, except the flavonoids, originate from the shikimic acid
pathway, Figure 2. Shikimic acid is a required intermediate for the formation of
phenylpropanoids (Goodwin and Mercer, 1983). Furthermore this pathway is responsible
for the synthesis of aromatic amino acids in plants and microorganisms (Ribéreau-Gayon,

1972; Goodwin and Mercer, 1983; Harbome, 1998).

13

Glucose
Pentose Pathway I— ? Glycolysis

D-Erythose-4-Phosphate
(4 carbon sugar) I

 

. . HO 1 OOH
Qurmc =
Acid ‘
OH
O OH
5-Dehydroquinic Acid

II

Enolisation

Phosphoenol Pyruvate (PEP)

OOH
Hydrolysable
HO OH Tannins
OH
Gallic Acid

NADPH + H +
NADP+

 

 

OOH
H+ IOH \
O HOH

5-Dehydroshikimic Acid (keto form)

NADPH NADPH
+ H + + H +
)1
NADP NADP+

 

CH2

OOH
+ 043
HO OH (5001,

OOH
H
‘ HO OH

OH
S-Dehydroshikimic Acid (enol form)

H20 4%
OOH
HO

OH
Protocatechuic Acid

ShikimiSAicid Phosphoenolpyruvic Add (PEP)

00H

Coumarins
CHZ'CO‘COOH N H =C-COOH
Chorismic "'_I _, 6H / |Benzoic
0 Acid
OOH . .
p Phenylpyruvic Phenylalanine Crnnamrc
Acid fad Condensed
CH2 bCOOH: Tannins
CH 2-CO-COOH __ I
OH \<j‘ C COOH Flavonoids
Prephenic Acid -\—:Hydroxycinnamic Acids
OH \Hydroxybenzoic Acids
p-Hydroxyphenylpyruvic Tyrosine p-Coumaric
Acid Acid Xanthones, Phenols

Figure 2. Shikimic Acid Pathway (Ribéreau-Gayon, 1972)

14

The general carbon skeleton and numbering system for flavonoids is shown in
Figure 1, I. The two aromatic phenol groups in flavonoids arise from two different routes
(Ribércau—Gayon, 1972). The 3-carbon (3C) oxygenated heterocyclic ring (C ring) and
one of the benzene rings (B ring, right side of the C ring) is formed via the shikimic acid

pathway, Figure 2. The 2"d benzene ring (A ring, left side of the C ring) arises from the

conjugation of three acetyl molecules via the acetate/malonate pathway (Balasundrarn et
al., 2006; Goodwin and Mercer, 1983; Harbome, 1998; Ribéreau—Gayon, 1972).

1. 4. Analytical Methods for Extraction, Purification, Separation and Detection

Only a small number of specific plants have been analyzed for phenolic
compounds, thus the data on phenolic compounds in plants is far fiom complete
(Robards, 2003). The plant matrix includes a complex mixture of compounds, including
phenolic compounds, which all contribute to the difficulty found when attempting to
extract and quantify the phenolic compounds in plant foods and beverages. Additionally,
the distribution of phenolics is not uniform throughout the plant material and contents
can vary depending on the plant species, variety, time of harvest, maturation, germination
and storage conditions (Tsao and Deng, 2004). This results in a great challenge in
“extracting a large range of compounds, each of which are structurally complex, from a
small amount of material, resulting in microgram amounts” (Harbome, 1998). There are
several steps for accurate analysis including: sampling and sample preparation,
extraction, separation, and detection (Robbins, 2003; Luthria, 2006). Each step

“introduces error, interferents and artifacts which decrease the recovery” or yield

15

of phenolic compounds (Stalikas, 2007). This emphasizes the importance of choosing
techniques and methods with greater selectivity and sensitivity throughout the analytical
process (Stalikas, 2007).

1.4. 1. Sampling and Sample Preparation

As stated above, phenolic compounds are not evenly distributed throughout the
plant tissue and researchers must analyze a representative sample of the entire plant, or
they need to document from which part of the plant the sample was taken fiom, i.e. seeds,
leaves, fi'uit, stems, peels (Luthria, 2006).

As soon as the plant tissue is damaged, compounds are susceptible to enzymatic
oxidation and hydrolysis via activated endogenous enzymes e. g. esterases, glycosidases,
hydrolases, decarboxylases (Harbome, 1998). Thus before and throughout the analytical
process, researchers need to limit the chemical changes resulting fi'om endogenous
enzymes, and/or due to sample preparation or storage conditions that could manipulate
the type and content of phenolic compounds (Robards, 2003; Tsao and Deng, 2004;
Luthria, 2006). Ideally freshly collected plants are the most desirable material for
phenolic analysis, but more commonly the plant material is dried by vacuum, air, or low
heat. Once thoroughly dried it can be stored for long periods before analysis (Harbome,
1998). Frozen material or freeze-dried material can also be used (Robards, 2003). Other
methods of sample preparation prior to extraction include milling, grinding, maceration,
homogenization, centrifugation, or cooking and are dependent on the type of plant food
or beverage to be analyzed. Beverages such as wines and hit juices often require little,
if any preparation, besides filtration or centrifugation followed by separation and

detection (Harbome, 1998; Robbins, 2003; Stalikas, 2007).

16

1.4.2. Extraction

Due to the complexity of the plant food matrix and the diversity of phenolic
compounds, a generalized or standardized method for extraction of all phenolic
compounds, or for that matter separation and detection, is not available (Harbome, 1998;
Robards, 2003). Plant tissues contain a large variety of structurally diverse compounds,
including phenolic compounds, and each individual component possesses physiochenrical
behaviors (Stalikas, 2007). Additionally interactions between phenolic compounds and
other macromolecules can take place and form insoluble complexes within the plant cells
(Naczk and Shahidi, 2006). The solubility of the analytes (i.e. phenolic compounds) of
interest in a particular extraction solvent and other physiocherrrical behaviors of these
analytes will strongly influence the choice of extraction solvent (Robards, 2003; Robbins,
2003; Stalikas, 2007). Common extraction solvents include alcohols (e. g. methanol,
ethanol), acetone, ethyl acetate and dirnethyfonnamide, together with varying amounts of
water, utilized alone or in combination (Harbome, 1998; Naczk and Shahidi, 2006). The
amount of water added is dependent on the water content in the starting plant material
(Harbome, 1998; Robards, 2003). Other measures employed to enhance the interaction
between the plant material and extraction solvent include homogenization, sonication,
vortexing, and increasing the solvent to sample ratio (Robbins, 2003). Extractions are
often repeated two to three times and extraction time can range from 1 to 12 hours (e.g.
Soxhlet extraction) (Robbins, 2003; Stalikas, 2007). Sequential extractions, via
increasing or decreasing polarity of the extraction solvent, can be utilized to broaden the

range of phenolic compounds extracted while limiting the amount of interfering

l7

components such as protein. The final crude extract is often concentrated via rotary
evaporation (Harbome, 1998).

Although there is not a general extraction method to extract all phenolic
compounds fiom plant material, some methods have been optimized for specific phenolic
classes. For example aqueous-methanol mixtures efficiently extract flavan-3ol
monomers and condensed tannins with a low degree of polymerization, whereas aqueous-
acetone mixtures maximize the extraction of oligomers and polymers of
proanthocyanidins. When extracting very polar phenolic compounds i.e. hydroxybenzoic
acids, hydroxycinnamic acids, and some flavonoid glycosides, increasing the percentage
of water in the aqueous-alcohol solvent mixture will enhance their extraction (Robards,
2003). Adding a small amount of acetic or hydrochloric acid to aqueous-alcohol solvents
enhances the extraction of anthocyanins given that they are unstable in neutral and
alkaline solutions (Harbome, 1998).

Other plant components such as carbohydrates, lipid material, and/or non-
phenolic compounds that interfere with the extraction and quantification of phenolic
compounds are often removed prior to or after the extraction process. For example non-
polar plant components i.e. waxes, chlorophyll, oils, sterols, are extracted with non-polar
solvents e. g. hexane, chloroform, prior to the extraction of phenolic compounds
(Harbome, 1998).

Newer extraction techniques being employed by researchers include supercritical
fluid extraction (SFE) and microwave assisted extraction. These techniques attempt to
reduce extraction time, heat, and oxygen exposure in order to lessen the chemical

alterations to phenolic compounds (Robards, 2003; Stalikas, 2007).

18

Often a hydrolysis step is performed to simplify the separation and quantification
of phenolic compounds, as well as to remove unwanted compounds in the plant matrix
that would otherwise interfere with the analysis (Tsao and Deng, 2004; Stalikas, 2007).
Since the majority of phenolic compounds are found conjugated in the plant matrix,
hydrolysis is applied to release the aglycones or free forms and to simplify their
characterization and quantification (Tsao and Deng, 2004; Stalikas, 2007). When
attempting to identify and/ or quantify phenolic compounds in their conjugated or natural
state, hydrolysis would be bypassed, however this depends on the availability of authentic
commercial standards and a researcher’s access to specific equipment and assays.

Acid and alkaline (saponification) hydrolysis are the most common techniques
applied to release the phenolic compound aglycones from their natural conjugated state.
Enzymatic hydrolysis is another possibility, however it is rarely used with crude plant
extracts. The enzymes: B-glucuronidase and/or sulfatase are often applied to biological
samples (i.e. plasma, serum, urine, bile) to identify the parent phenolic compound of
metabolites (Stalikas, 2007).

Acid hydrolysis is carried out on the solid plant food or the plant extract with l to
2N HCL at reflux or above reflux temperatures in aqueous or alcoholic solvents. The
reaction time varies from 30 minutes to 1 hour. Acid hydrolysis is capable of cleaving
glycosidic bonds (-C-O—C-) resulting in the release of sugar moieties; thus it is commonly
utilized to convert flavonoid glycosides to their aglycones. Yet the resistance to acid
hydrolysis for different glycosidic bonds will vary and acid hydrolysis can induce
undesired changes to some phenolic compounds (Ribéreau-Gayon, 1972). The

anthocyanins are more resistant to acid hydrolysis than flavonol and flavone glycosides,

19

with flavonoid C7-glycosides being particularly more stable compared to C3-glycosides.
Ether bonds (R-O-R’) and C-glycosides vs. O-glycosides are very resistant to acid
hydrolysis (Ribéreau-Gayon, 1972). Additionally, acid hydrolysis cleaves esters bonds to
release free phenolic acids (i.e. hydroxybenzoic and hydroxycinnamic acids).
Researchers, however, have observed the loss of hydroxycinnamic acids exposed to hot
acidic conditions because they undergo decarboxylation; thus they are best obtained by
mild alkaline hydrolysis (Krygier et al., 1982; Gao and Mazza, 1994; Harbome, 1998).
Flavanones are converted to chalcones in acid solutions, resulting in an intense color
(Ribéreau—Gayon, 1972). Leucoanthocyanidins are converted to anthocyanidins and form
condensation products in heated acid solutions resulting in a red colored solution. Thus a
researcher must be aware that anthocyanidins in acid hydrolysates may come about fi‘om
anthocyanins or leucoanthocyanidins (Ribéreau-Gayon, 1972). Catechins, naturally
found as monomers, not glycosides, will also form condensation products in acid, but
they are not red in color. So flavans, i.e. flavan-3ols and leucoanthocyanidins should not
be analyzed following acid hydrolysis (Ribéreau-Gayon, 1972). Rommel and Wrolstad
(1993) reported losses of quercetin and kaempferol aglycones following acid hydrolysis
of raspberry juice extracts. The expected quantities of aglycones determined from the
concentrations of quercetin and kaempferol glycosides in the unhydrolyzed raspberry
juice extracts, were estimated to be much higher than the quantities observed.

Alkaline hydrolysis is carried out on the solid plant food or the aqueous plant
extract with 1 to 4 M NaOH at room temperature or at low heat. To inhibit the oxidation
of phenolic compounds, hydrolysis is carried out under an inert gas such as nitrogen gas.

The reaction time varies from 4 hours to overnight. The alkaline hydrolysate is then

20

acidified with HCL before analysis (Ribéreau-Gayon, 1972; Harbome, 1998; Robbins,
2003). Alkaline hydrolysis is capable of cleaving ester bonds and removing acylated
portions of flavonoids, such as acylated anthocyanins (i.e. phenolic acids acylated to
sugar moiety of anthocyanins) (Gao and Mazza, 1994; Harbome, 1998; Robards, 2003;
Stalikas, 2007). As stated previously, hydroxycinnamic acids are more stable in mild
alkaline solutions than aqueous acidic solutions and best obtained by mild alkaline
hydrolysis (Krygier et al., 1982).

Krygier et al. (1982) observed the percent loss caused by alkaline and acid
hydrolysis on authentic hydroxycinnamic acids standards. They reported a 15.1% to
91.7% and 2.7% to 66.7% loss resulting from acid or alkaline treatment, respectively.
The percent lost by either treatment was dependent on the compound. Acid hydrolysis
resulted in almost the complete loss of caffeic acid (87%) and trans-sinapic acid (92%),
whereas the losses of o-coumaric, p-coumaric, trans-isoferulic and trans-ferulic acid were
15%, 73%, 50% and 78%, respectively. The same compounds were reported to be more
stable under the alkaline treatment, except for caffeic acid (67% loss).

Dabrowski and Sosulski (1984) reported the losses of authentic hydroxybenzoic
and hydroxycinnamic acid standards added to rapeseed flour, extracted and exposed to
alkaline conditions. They reported the losses for the hydroxybenzoic acids: vanillic, p-
hydroxybenzoic and syringic acid did not exceed 10%. The percent loss reported for the
hydroxycinnamic acids: p-coumaric acid, ferulic acid and sinapic acid, following alkaline
treatment were 13%, 18% and 22%, respectively. As was reported by Krygier et al.
(1982), caffeic acid was almost completely lost under the alkali conditions with a

reported 84% loss. Both authors reported that high losses of caffeic acid under alkali or

21

acid treatment were the result of the high reactivity of o-dihydroxyphenols (Figure l, H),
which can be rapidly oxidized to quinones. Caffeic acid appears to be unstable under
either hydrolysis treatment.

Common methods used to clean-up/purify crude extracts or the hydrolysates are
liquid-liquid extraction or solid phase extraction (SPE). Both methods can be used to
fractionate phenolic compounds as well as remove unwanted components.

The solvents used for liquid-liquid extraction depend on the desired compounds
and can be based on polarity or acidity. For example when basing the liquid-liquid
extraction on polarity, the solvents: ethyl acetate, ether, or the combination of the two,
would be used to extract phenolic acids, flavonols and flavones. If anthocyanidins are
present the solvents above would be used to separate or remove all other phenolic
compounds from the anthocyanidins remaining in the more polar solution (Ribéreau-
Gayon, 1972). When the liquid-liquid extraction is based on acidity, neutral compounds
(e. g. flavonols) would be separated from the acidic compounds (e. g. phenolic acids) by
adjusting the plant extract sample’s pH to 7.0 and adding ethyl acetate. The neutral
compounds would be more soluble in the ethyl acetate when the pH=7.0. Once the ethyl
acetate is removed, the remaining extract would be adjusted to pH 2.0 and ethyl acetate
would be used to extract the acidic compounds (Harbome, 1998).

SPE is a fast and reproducible procedure to clean-up extracts and hydrolysates

and can result in fairly purified extracts (Stalikas, 2007). The most common sorbent for
analyzing phenolic compounds is reverse phase C13 bonded to silica (Stalikas, 2007).

The extracts and hydrolysates and/or solvents are slightly acidified to prevent ionization

of the phenolics, which could reduce their retention (Stalikas, 2007). Similar to liquid-

22

liquid extraction, solvents of different polarity or pH can be used to remove unwanted
components, to separate larger phenolics from smaller phenolics, or to separate different
classes of phenolics fi'om each other (Harbome, 1998; Robbins, 2003; Stalikas, 2007).

For example, Skrede et al. (2000) separated anthocyanins fiom other phenolic

compounds in concentrated blueberry juice using a C13 SPE cartridge. After applying the

acidified aqueous sample to the activated C13 SPE cartridge, acidified water (0.01%

HCL) was applied to elute unwanted water-soluble compounds i.e. sugars. This was
followed by acidified methanol (0.01% HCL) to elute the anthocyanins and finally ethyl
acetate to elute all other phenolic compounds (Skrede et al., 2000). A limitation of SPE
is column saturation, so small volumes are required. Recoveries are rarely reported, yet
Rommel and Wrolstad (1993) reported low and variable recovery of hydroxybenzoic

acids (less than 10%), especially protocatechuic and gallic acid, but close to 100%
recovery for flavonol aglycones, flavan-BOIS and ellagic acid from C13 cartridges.

1.4.3. Separation and Detection

Thin-layer chromatography (TLC) and paper chromatography (PC) were often
used to separate, characterize, and quantify phenolic compounds and other plant
components (Ribereau-Gayon, 1972; Harbome, 1998; Robards, 2003). Although TLC
and PC are not widely used today, they remain an important tool to screen crude plant
extracts for phenolic compounds (Harbome, 1998; Robards, 2003).

High performance liquid chromatography (HPLC) is currently the predominant
analytical method for the separation and characterization of phenolic compounds
(Robbins, 2003; Robards, 2003). Equal in sensitivity and selectivity to HPLC, gas

chromatography (GC) can be used, however it is not as popular to separate phenolic

23

compounds because of their low volatility and the possible chemical changes induced by
the high temperatures required for GC (Robards, 2003).

Plant constituents, i.e. phenolic compounds, exhibit unique spectral
characteristics, consequently a means for their detection are ultraviolet/visible (UV N is),
photodiode array (DAD) and ultraviolet/ fluorescence detectors (Harbome, 1998;
Robbins, 2003; Robards, 2003; Naczk and Shahidi, 2006; Stalikas, 2007). No single
wavelength is ideal for all phenolic classes because each compound displays absorption
maximas at distinctly different wavelengths (Robards, 2003). Hydroxybenzoic and
hydroxycinnamic acids have a maxirna absorbance in the range of 200 to 290 nm, except
gentisic acid (hydroxybenzoic acid) whose absorbance extends to 355 nm;
hydroxycinnamic acids have a second maxima absorbance in the range of 270 to 360 nm
due to the additional conjugation; flavonoids have two absorption maximas. The first is
in the range of 240 to 285 nm, and a second in the range of 300 to 550 nm. The first
maxima absorbance is due to the A ring, and the second is due to substitutions and the
conjugation of the heterocyclic C ring. Routinely 280 nm is chosen to represent
flavonoids, 254 nm for phenolic acids (hydroxybenzoic and hydroxycinnamic acids) and
520 nm for anthocyanins (Ribereau-Gayon, 1972; Harbome, 1998; Robbins, 2003;
Robards, 2003; Stalikas, 2007). Analytes can be monitored after separation with 1 or
more UV detectors set at different wavelengths to aid in the detection of a range of
phenolic compounds, e. g. 254 nm and 280 nm; 280 nm and 320 nm. DAD detectors are
capable of scanning real time UVN is spectra of the analytes enhancing the ability to
detect a range of phenolic compounds in a complex matrix. Care must be taken in

selecting mobile phases that do not absorb at the set wavelengths. Although all phenolic

24

compounds do not display fluorescence, when used in combination with UV detection it
can distinguish between co-eluting compounds. However, fluorescence is not widely
used (Stalikas, 2007).

Limitations of spectrophotometric methods of detection include poor sensitivity
and the inability to distinguish between co-eluting compounds even if they absorb light at
different wavelengths (Bocchi et al., 1996; Robards, 2003; Stalikas, 2007).
Electrochemical detection (ECD) is becoming an increasingly popular tool for detection
of phenolic compounds for its higher sensitivity and selectivity compared to UV/V is and
DAD detectors. ECD is based on the oxidation or reduction of compounds. Briefly the
HPLC eluent flows over an electrode or series of electrodes set at increasely higher
voltages. Electroactive compounds will either donate electrons (oxidize) or accept
electrons (reduce) resulting in a current that is measured in real time. The current peak
generated is dependent on the concentration of the analyte and the voltage applied. Each
compound has a specific voltage at which they begin to be oxidized (Svendsen, 1993).
There are two types of EC detectors: 1. flat-plate amperometric detectors which oxidize
or reduce approximately 5% of the compound flowing over it; 2. porous flow-through
amperometric detectors which have larger surface areas and can oxidize close to 100% of
the compound flowing through them, termed coulometry (Svendsen, 1993). Serial arrays
of coulometric electrodes (8 to 16 electrodes) are more sensitive and require less
maintenance than amperometric arrays. As the analytes pass through the electrodes of
increasing voltage, they will be oxidized 100% at their oxidation potential and no other

peaks will be visible at subsequent electrodes (Svendsen, 1993). HPLC-ECD mobile

25

phases are acidic buffers, which suppress the dissociation of weakly acidic phenolic
compounds (Svendsen, 1993; Jandera et al., 2005).

For structural identification of phenolic compounds, HPLC and GC can be
coupled to mass spectrometer (MS) detectors. MS is an important tool because it can
provide structural data on small (microgram) amounts of sample, can give accurate
molecular weights and generate complex fragmentation patterns which can be used to
characterize specific compounds (Harbome, 1998). Thus it can identify conjugated
phenolic compounds. Two MS methods widely employed are electrospray ionization
(ESI) and atmospheric pressure chemical ionization (APCI). Other methods include fast
atom bombardment (FAB), plasma desorption (PD) and matrix-assisted laser desorption
ionization (MALDI). To further increase detection sensitivity and selectivity, MS
analyzers can be coupled in series of 2 (MS-MS) or more (MSn ) (Harbome, 1998; Tsao
and Deng, 2004; Stalikas, 2007).

Nuclear magnetic resonance (NMR) spectroscopy, proton-NMR and carbon 13-
NMR, can be coupled to HPLC to identify novel phenolic compounds and known
compounds without great isolation or separation. It is often used to identify structures of
compounds that could not be distinguished by mass spectral data (Robards, 2003;
Stalikas, 2007). NMR, however, can be quite complex and requires highly trained
individuals to run it (Harbome, 1989).

Many researchers utilize colorimetric methods to quantify total phenolic content
(TPC) in complex mixtures. The F olin-Ciocalteu method is a frequent choice. The
reaction involves the oxidation of phenols by phosphomolybdic-phosphotungstic acid

resulting in a reduced molybdenum-tungsten blue complex. The reaction develops over

26

30-90 minutes and the blue complex is detected at an absorbance between 725-765 nm.
An authentic standard phenolic compound (e. g. gallic acid or (+)-catechin) generates a
standard curve to quantify the total phenolics. A limitation is the presence of other
phenol containing compounds and antioxidants leading to overestimations in total
phenolic compound content (Singleton and Rossi, 1965).

1. 5. Intake, Absorption and Metabolism of Phenolic Compounds

1.5.1. Intake

The consumption of diets high in fruits, vegetables, grains and legumes have been
associated with a reduced risk for developing certain chronic diseases, i.e. cancers and
cardiovascular diseases (Steinmetz and Potter, 1993; Fraser, 1999; Genkinger et al.,
2004; WCRF/AICR, 2007). There is speculation that the high concentration of phenolic
compounds found in plant foods and beverages may exert some of these observed health
benefits. In order to investigate the proposed relationship between diets high in phenolic
compounds and reduced disease risk, knowing their total content and distribution within
plant based foods and beverages are crucial (Manach et al., 2004). Currently phenolic
compound profiles and content for many plant foods and beverages are incomplete,
making it difficult to estimate an accurate daily intake of total and/or individual phenolic
compounds consumed by humans. Research has identified specific phenolic compounds
or classes in widely consumed foods that have displayed the greatest biological activity in
the literature (Manach et al., 2004). It must be acknowledged that the identification and
quantification of phenolic compounds in plant foods are fiequently based on the aglycone
form following hydrolysis of the glycosidic or ester bonds (Walle, 2004; Manach et al.,

2004), and not the compounds in their natural state. However as stated in the previous

27

section this is common practice due to the large diversity and number of phenolic
compounds in nature and simplifies the identification and quantification process
(Ribéreau-Gayon, 1972; Harbome, 1998). In 2007, the USDA published the most current
flavonoid database that compiled data from “acceptable” studies and reported the content
for five subclasses of flavonoids (flavonols, flavones, flavanones, flavan-3ols and
anthocyanidins) measured in 385 food items (U SDA/ARS, 2007). To date this is the only
flavonoid database, yet again it only analyzes the most commonly studied flavonoids in
relatively few food items.

Chun et al. (2007) utilized the USDA flavonoid database and 24 hr recall data
from NHANES 1999-2002 surveys to estimate an average daily flavonoid intake of 187.7
mg/day in the US. If one consumes diets high in red, blue, and purple fi'uits, or drinks
moderate amounts of red wine, ingesting over 100 mg of anthocyanins per day could
occur (Manach et al., 2005). Coffee drinkers consume significant amounts of
Chlorogenic acid, an ester of caffeic acid, since one 200 ml cup provides approximatedly
70-200 mg of Chlorogenic acid (Clifford, 1999). Flavan-3ols (catechins) and
proanthocyanirridins are found in high amounts in green tea, red wine and chocolate and
consumption of these products can markedly influence total phenolic intakes. There does
not appear to be any estimates of typical daily intakes of hydroxybenzoic acids.

1.5.2. Absorption in stomach and small intestine

The extent and site of phenolic compound absorption is far from complete.
Figure 3 presents an overview of the absorption of phenolic compounds. After ingestion,

phenolic compounds enter the stomach. Small quantities of phenolic glycosides and

28

Plant food phenolic compounds

I

 

 

STOMACH

 

Dietary Free Phenolic
Compounds, Phenolic

Glycosides & Esters,
Phenolic Polymers

 

duodenum

SMALL
INTESTINE

jejunum

ileum

 

F—P ? Passive Transport

 

 

 

 

v

v
Free, Glycosidesm‘gm LIVER
Esters & Polymers

of Phenolic

 

 

 

Compounds

Glucosidases i.e.

Phase I Enzymes
i.e. Cytochrome P4503

 

LPH

 

Passive Transport

————> Phase II Enzymes

 

 

 

TISSUES

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

COLON

 

Degradation— Esterases, Glucosidases,
Hydrolases, Ring Fission

 

 

 

I

Feces

Figure 3. Absorption and Metabolism of Phenolic Compounds

29

Active Transport i.e.
i.e. SGLT l Glucuronoyltransferases,
_—__' Sulfotransferases,

Metabolized in Methyltransferases
Enterocytes by
Glucosidases, '
Esterases, Phase I w
& Phase II
Enzymes SI Enterocytes

Free, Glycosides, Esters, & Polymers of I

Phenolic Compounds
Colonic Microflora Metabolism &/or Urine

phenolic esters may be freed during food storage and processing Manach et al., 1998)
making them available for absorption in the stomach. Coffee and wheat bran have been
incubated with in vitro gastric juices to determine the effect of the acidic gastric
environment on the soluble hydroxycinnamic acid esters (i.e. Chlorogenic acid in coffee)
and hydroxycinnamic acids esterified to the plant cell wall in wheat bran). No effect was
observed on the caffeic acid derivatives in the coffee, e. g. no hydrolysis of Chlorogenic
acid to caffeic acid and quinic acid; and minimal amounts ferulic acid and ferulic acid
esters (~1.3%) were released fiom wheat bran (Kroon et al., 1997; Rechner et al., 2001).
There is little good evidence that phenolic compounds are absorbed in appreciable
amounts from the stomach.

The majority of phenolic compounds and their derivatives would pass through the
stomach and enter the small intestine (SI). Once in the SI, the glycoside derivatives
could be deglycosylated by B- glucosidases in the lumen or part of the brush border. LPH
(lactose phlorizin hydrolase), a B-glucosidase located on the brush border deglycosylates
flavonol glucosides and diglucosides prior to absorption (Day et al., 2000; Nemeth et al.,
2003). Phenolic aglycones freed prior to entering the SI or as a result from hydrolysis by
glucosidases could enter the enterocyte by passive diffusion (Hollman, 2001; Petri et al.,
2003). Phenolic glycosides may also enter enterocytes by the sodium-dependent glucose
transporter (SGLTl) (Holhnan et al., 1995; 1996; Gee et al., 1998; 2001; Hollman,
2001).

The absorption and metabolism of the flavonols has been extensively studied.
Holhnan et al. (1995, 1996, 1997) argue that the majority of flavonols are absorbed in the

SI and maybe even in the stomach. However, research also suggests that certain flavonol

30

glycosides (i.e. rutin) are passed to the colon (Holhnan et al., 1995; 1997; Olthof et al.,
2000; Hong et al., 2004). Flavan-3ols, such as catechins and epicatechins, are found as
aglycones, gallate esters or polymers (condensed tannins or proanthocyanidins) in plants.
Although some of the consumed flavan-3ols — catechins and epigallocatechin — are
absorbed prior to the colon (Holhnan et al., 1997; Spencer et al., 2006), most flavan-3ols
and proanthocyanidins pass into the colon (Holhnan et al., 1997; Kahle et al., 2007).
Hydroxycinnamic acids found esterified to the cell wall polymers cannot be absorbed in
the SI and would also pass into the colon (Kroon et al., 1997; Rechner et al., 2001; Adam
et al., 2002; Rondini et al., 2002; 2004).

1.5.3. Metabolism by microflora and absorption in the colon

Phenolic compounds not absorbed in the SI would enter the colon. The colon
contains approximately 1012 microorganisms/cm and the microorganisms have the ability

to hydrolyze and metabolize dietary compounds (Spencer et al., 2006). Colon
nricrofloral enzymes include: B-glucosidases, B-rhamnosidases, esterases, xylanases,
hydrolases and oxidases (Kroon et al., 1997 ; Plumb et al., 1999; Andreasen et al., 2001;
2001; Rechner et al., 2004; Jenner et al., 2005). Esterases cleave phenolic compound
esters and phenolic acids esterified to plant cell walls, releasing the free phenolic
compounds (aglycones) (Plumb et al., 1999; Andreasen et al., 2001; 2001). The
metabolites resulting from the microfloral metabolism of phenolic compounds are
phenolic acids and other aromatic acids, mainly derivatives of phenylpropionic,
phenylacetic, benzoic and cinnarnic acid (Gonthier et al., 2003). These microbial
metabolites can then be absorbed in the colon, metabolized in the liver and excreted in

the urine or excreted in the feces.

31

Rechner et al. (2004) specifically examined colonic metabolites formed when
hydroxycinnamic acids (Chlorogenic acid), flavanone glycosides (naringin) and flavonol
glycosides (rutin) are incubated with a mixed culture of human colonic microflora in
vitro. For Chlorogenic acid (5-caffeoyl-quinic acid), the first metabolic product was
caffeic acid resulting from the cleavage of the ester bond between caffeic acid and quinic
acid, followed by the reduction of the caffeic acid double bond forming 3-(3,4-
dihydroxyphenyl)-propionic acid, then dehydroxylation to form 3-(3-hydroxyphenyl)-
propionic acid and finally the 2"d dehydroxylation to form 3-phenylpropionic acid (non-
phenolic acid). In vivo the metabolites formed at each step could be absorbed fiom the
colon. For example, caffeic acid could be absorbed intact or as any of the microbial
metabolites — 3-(3,4-dihydroxyphenyl)-propionic acid, 3-(3-hydroxyphenyl)-propionic
acid or 3—phenylpropionic acid (Rechner et al., 2004).

The first metabolic step for both flavanone glycosides and flavonol glycosides
was deglycosylation. This was followed by ring fission in the heterocyclic C-ring (see
Figure 1, I). For naringen, the C-ring ring fission was between the oxygen and C2 and
between C4 on the C-ring and the A-ring. This resulted in the transient formation of
phloroglucinol derived from the A-ring fragment, and either 3-(4-hydroxyphenyl)—
propionic acid and 3-phenylpropionic acid (via dehydroxylation) derived fiom the B-ring
fragment. These two metabolites can be absorbed. For example, rutin entering the colon
is hydrolyzed by microbial enzymes to quercetin and rutinoside. Fission of the
heterocyclic C-ring produced three different ring fission products: 1) product 1— fission
between the oxygen and C2 and between C3 and C4 on the C-ring; 2) product 2 —

fission between the oxygen and C2 and between C4 on the C-ring and the A-ring; 3)

32

product 3 — fission between oxygen and the A-ring and C2 and C3 in the C-ring. The
A-ring fi'agments were not detected and assumed to be degraded rapidly or destroyed
with the ring fission. The endproducts from the first two ring fissions include 3,4-
dihydroxyphenylacetic acid or 3-(3-hydroxyphenyl)-pr0pionic acid, respectively, both
derived fiom the B-ring. The colonic bacteria can dehydroxylate the 3,4-

dihydroxyphenylacetic acid to hydroxyphenylacetic acid which can be then be absorbed
along with 3-(3-hydroxyphenyl)-propionic acid. The hydroxyphenylacetic acids are
specific metabolites of flavonol glycosides arising from colonic microorganism
metabolism (Rechner et al., 2004).

1.5.4. Metabolism after absorption

A phenolic compound absorbed in the SI enterocyte as a glycoside can enter the
bloodstream as the intact glycoside. However they are more likely to be deglycosylated
by cytoplasmic B—glucosidases. Phenolic esters absorbed from the SI are de-esterified by
cytoplasmic esterases. The resulting aglycones are either released into the portal
circulation or further metabolized by the intestinal cells (Andreasen et al., 2001;
Hollman, 2001; Petri et al., 2003; Spencer et al., 2003). Phase I (cytochrome P4505) and
phase II enzymes are found in the SI enterocytes (Spencer et al., 2006). The phase II
enzymes are responsible for conjugating phenolic compounds to glucuronic acid, sulfates
and methyl groups. Studies show they are conjugated to glucuronic acid via UDP-
glucuronosyltransferases (Petri et al., 2003), sulfated by sulfo-transferases, and 0-
methylated by catechol-O-methyl transferases (COMT) (Olthof et al., 2000; Spencer et
al., 2006). The phase II metabolites can be excreted from the enterocyte into the SI

lumen or into the portal circulation. Phenolic compounds not undergoing phase I and II

33

metabolism in the enterocyte are metabolized in the liver i.e. glucuronidated, sulfated, or
methylated. The liver will excrete the phenolic metabolites into the bile or into the
circulation. Metabolites excreted in the bile enter the 81 where they may be reabsorbed
or they pass to the colon where they can be metabolized by the colonic microflora and
reabsorbed or excreted in the feces (Spencer et al., 2006). Metabolites entering the
circulation from the enterocyte and liver are excreted in the urine.

At normal dietary levels, absorbed phenolic compounds are unlikely to escape
first-pass metabolism, thus the predominant form found in the plasma would be
conjugates (i.e. glucuronates and sulfates) of the phenolic compounds (Kroon et al., 2004;
Manach et al, 2004; Spencer et al., 2006). The form of the phenolic compound and dose
given appear to affect where the compound is metabolized and the type of conjugates
formed. When phenolic compounds are given in pharmacological doses as aglycones or
in their native forms, they may saturate the phase I and phase H enzymes, resulting in
higher amounts of the aglycone or native compound in circulation. For example, “the
sulfate pathway is a high affinity, low capacity pathway, high doses would saturate this
pathway and the compounds would be redirected to the glucuronidate pathway, which is
high capacity” (Spencer et al., 2006). The metabolites formed when a pharmacological
dose of a compound or a mix of compounds in a concentrated extract are given would be
different then when the compounds are provided at normal dietary levels or as part of the
food matrix (Kroon et al., 2004; Manach, 2004; Spencer et al., 2006). At larger doses
metabolism occurs predominantly in the liver, whereas at dietary levels it occurs almost

exclusively in the S1 with the liver playing a secondary role (Scalbert and Williamson,

34

2000; Spencer et al., 2006). Moreover, the administration of single phenolic compounds
may overestimate the absorption from dietary sources (Goldberg et al., 2003).

Bacterial metabolites of flavanoids -— 3-(3-hydroxyphenyl)-propionic acid and 3-
phenylpropionic acid ——absorbed in the colon are converted to 3-hydroxyhippuric acid
and hippuric acid, respectively, via B-oxidation and glycination in the liver and then
excreted in the urine (Rechner et al., 2004).

1.6. Phenolic Compounds in Grain Legumes

The majority of the studies identifying and quantifying phenolic compounds in
legumes has focused on the compounds found in soybeans and other oil seeds (Krygier et
al., 1982). Numerous reviews examining bioactive soybean components and their health
effects have been published (Messina et al., 2006; Larkin et al., 2008; Xiao, 2008).
Studies identifying phenolic compounds in grain legumes, including dry common beans,
are limited and were excuted to examine their organoleptic and anti-nutrient effects.
Phenolic compounds (i.e. condensed tannins and flavonoids) in beans are associated with
adverse tastes and colors in food products, and can reduce protein digestibility and iron
bioavailability. Additionally, consumers have developed specific preferences for the
size, shape and seed coat color of dry common beans and this has established the various
market classes available (Beninger and Hosfield, 2003). Seed coat (i.e. hull) color varies
considerably among dry common beans (Phaseolus vulgaris L.) and is determined by the
concentration and presence of flavonol glycosides, anthocyanins and condensed tannins
(i.e. proanthocyanidins). In fact various seed coat color genotypes have been identified,
as well as the phenolic compounds responsible for the colors (Letenne and Munoz, 2002;

Beninger et al., 1999; 2003; 2005). Thus in many studies the seed coat is separated fi'om

35

the cotyledon (meat or fi'uit of the bean), and only the coats are analyzed for phenolic
compounds.

Investigators have identified flavonol glycosides, hydroxybenzoic acids,
hydroxycinnamic acids and condensed tannins (proanthocyanidins) in the seed coats of
some grain legumes (Krygier et al., 1982; Sosulski and Dabrowski, 1984; Beninger et al.,
1999; 2003; 2005; Madhujith et al., 2004; Aparicio-Femandez et al., 2005; Hu etal.,
2006; Lin and Lai, 2006; Ranilla et al., 2007). Anthocyanins have only been identified in
beans possessing blue and blue-violet colored seed coats (Takeoka et al., 1997; Beninger
et al., 2003; Choung et al., 2003; Oomah et al., 2005; Aparicio-Femandez et al., 2005;
Macz-Pop et al., 2006). Darker seed coat colors, i.e. red, brown or black, are associated
with greater total phenolic content and antioxidant activity measured by various in vitro
assays compared to light seed coats i.e. white, gray, yellow (Madhujith et al., 2004;
Oomah et al., 2005; Lin and Lai, 2006; Xu and Chang, 2007), however this is
controversial (Rocha-Guzman et al., 2007). It is dependent on the legume type, the
classes of compounds being investigated and whether the whole or part of the bean is
analyzed.

Several papers have characterized specific classes or phenolic compounds in
different grain legumes. Beninger et al. (1999) identified quercetin, kaempferol,
unknown flavonol glycosides and proanthocyanidins in raw dark kidney bean seed coats
(Phaseolus vulgaris L. cul. Montcalm). Takeoka et al. (1997) reported the presence of
anthocyanins: delphinidin-, petunidin- and malvidin-3-glycosides, in raw black beans
(Phaseolus vulgaris L.). Aparicio-Femandez et al. (2005) identified anthocyanins in the

seed coats of raw Black Jamapa beans (Phaseolus vulgaris L.), as well as

36

proanthocyanidins and flavonol glycosides of kaempferol, quercetin and myricertin. Hu
et al. (2006) investigated the phenolics in raw white, black, red and pinto bean seed coats
(Phaseon vulgaris L.). The flavonol kaempferol and astragalin (kaempferol-3-0-
glucoside) were identified in the seed coats of the red and pinto beans. Kaempferol was
not identified in the seed coats of black beans, but other unknown flavonols and
anthocyanins were. No flavonoids were detected in the white seed coats. Choung et al.
(2003) investigated the anthocyanins in seed coats of different colored raw kidney beans
(Phaseolus vulgaris L.). They observed five different anthocyanins, whose content and
composition were dependent on seed coat color. In the seed coats from Brazilian and
Peruvian bean cultivars (Phaseolus vulgaris L.), Ranilla et al. (2007) identified
condensed tannins, anthocyanins and kaempferol and quercetin glycosides. They found
the cotyledons to be rich in phenolic acids such as ferulic acid, sinapic acid, Chlorogenic
acid and other hydroxycinnamic acids. Sosulski and Dabrowski (1984) reported the
presence of hydroxycinnamic acids such as ferulic acid, p-coumaric acid, syringic acid in
the seed coats (hulls) and flours (cotyledons) of 10 legume species. They also observed
hydroxybenzoic acids, i.e. p-hydroxybenzoic acid, protocatechuic acid and gallic acid, in
select species.

Others investigating grain legumes quantified the total phenolic content or classes
of phenolic compounds e. g. total flavonoid content, total condensed tannin content.
Heimler et al. (2005) investigated the phenolic compounds in 3 varieties of dry beans
(Phaselous vulgaris L.). They reported the total phenolic content (F olin-Ciocalteu
method) ranged from 117 to 440 mg gallic acid equivalents per 100g of raw bean flour

and total flavonoid content ranged from 22 to 143 mg (+)-catechin equivalents per 100 g

37

of raw bean flour. They claimed the presence of kaempferol, quercetin and caffeic acid
derivatives, but did not quantify them. Xu and Chang (2007) reported the total phenolic
content (F olin—Ciocalteu method) in different grain legumes ranged from 131 to 373 mg
of gallic acid equivalents per 100 g of raw bean flour and total flavonoid content 10 to 98
mg of (+)—catechin equivalents per 100 g of raw bean flour.

Sosulski and Dabrowski (1984) investigated the free and hydrolysable phenolic
acids in the flours (cotyledons) and hulls (seed coats) of 10 legume species. When they
attempted to hydrolyze the remaining insoluble solid residue after the removal soluble
free and phenolic acid esters via alcohol-aqueous extraction, in either the flours
(cotyledons) or hulls (seed coats) of raw navy beans, it failed to yield any further
appreciable amount of phenolic acids. This was observed with the oil legume, rapeseed,
as well. Conversely, cereals yield much higher concentrations of hydroxycinnamic acids
after alkaline treatment on the insoluble solid residue (Krygier et al., 1982; Sosulski et al.,
1982). Since the cell walls of dicotyledons, e. g. legumes, have much smaller amounts of
phenolic acids associated with their cell walls in comparison to monocotyledons e. g.
cereals and grasses (Hartley, 1987), alkaline hydrolysis of the insoluble solid residue may
not be a necessary step when analyzing grain legumes.

Table 2. summarizes the quantitative data reported of specific phenolic
compounds found in different whole grain legumes.

Cai et a1 (2003) identified seven different phenolic acids in 17 varieties of raw
cowpeas (Vigna unguiculata) including protocatechuic acid, p-hydroxybenzoic acid,

caffeic acid, p-coumaric acid, ferulic acid, 2,4-dimethoxybenzoic acid, and cinnarnic

38

I. IP‘I‘II IIII‘ ‘ II.‘ -I I--“'I-'I‘D\ l...“..l~.-~ IN l'~.~.~r.r

 

A EEM~§

 

 

 

 

 

N 2:2 2.2 22.2 N 388E 3.22.2.
. . . . 32 new?
N 2. N 2 N 2 N 2 N $28332 agar
NSN ..2N . . . 2a: NNN 3332. 2.2225
2 N N o N N . . . soc ..2 225N225 2822025, N8
20 882.22 . . 2. N4 22 aaNNNv 95 eases 2223.. N
.. n a an 3 .3 093 28053
-3825 o 2 22 a 22 64,222.? m2220.232 s8 x2 52 2 E . 220
Be: 53
2.N N2 224.22 22.32802 N ..2 “.2332: 2822023» 2.22
-vd 622 8528222 «iguana 53on
2 N N 8N
NSN ..2a . 2222. NNN 3308 2 5m .
2 N E . . . . 2822 22822 .2 EB? 2822025 2‘:
a 2.22.322 . . 2. 2e 22 .aaNNNV 95 E38322 232 a N N a 88 228 N
.. u 0
$5 2 2 N 22 0.2%-? 2:002 $3 3 SN 2 E 2 22
uogxo game
0.8 mm 5.8 22 .2 Sec 8335?: 0202823 0328
e . 2 a 22 NE SN . 282202.292 N:
258—? 89260
. . 222a NNN Ba 822 822.28% 8a
22 - ma» o m e o BEES
..2M222%2N6 we: -uwwe own; 822 ENC 95 oz unmet“ M82 522 as N 2 was canons, E
. -22.: 6.2.5.? NE SN . 8980
8cm A2552” 383m 350
.2“ a 222.2. 2.22 2.22.22 3.22 .8: 2223 92.2 2.5 22220 2.222205 22522 522 N .2 new? bore.
385. NS .23 No.22 No.22 05222-52 m232822.222 as N 222 328.32: «22280 3
$223 oz m220.2222 s8
<m: .3 28 a
235 N > (NE <2
5535...:—
...._2__u...“.....mm...2.2fia .ufifim. 2. as... . 32...
5232:2—

 

359”: 59.0 E 6052:2220 22559—30 ozone-E .N 033—.

39

A!i-in.n~-d\ .N .vhghcn.!.

 

ocom

02.8322 350

 

 

 

.. . . 222222 22522 .
22“ 2o 2.N 22 N 22 - 22a. 22220 292205 2 EN? buss
. . 2222 8N 2:22 222N2 . 22822
82222 N 22 2.N 22 3.202.222-2222 .2383: 38 N 222 388.222 22.255 3
8222.22 oz 222002 32222
22.2 22.2. 2.222 N.N .2 E2232: 2222222222
232322223 .
. . . . A «2.232322 2
v 222m N w v a w o §~omwoao~ MFA
. . . . .2 225N223
2 2 N N 22 222 N 2 2.2228823 E22520
2.NN 2.N 4.: N.N .2 2.23%: 2322322 2232
2.58.32: .
. . . . .2 EN?
2. N N N N 2.2 22 2 38,8232 .2059 222N22
22.22N NN N.N2 N2 .2 E2222? E22202 228
32882.2 .
. . . . .2 EN?
N 222. N N N N 22 N2 388i 22:2
NNN 2.22 N.: Na .2 82222223; 28222.5 N2
-22.22N -2222 22.2 -222. 388232 252222222 see
22eN 22.22 NNN 3 .2 222222222: 28220:?
-SN .3 -N.N2 -3. a288.32 N2 3222.2
. gem 3808
8222.228 N.N2. 22.N N.NN 2. 22 2222: 2.2225 .2822 22822 as: .2 2.2.3223 28222225
8222.2 {2.N -5 -2 2 -2. 2.2 NNN 2223 a: 2N2 25 22.22822»: m 2 Be 2222.23.22.22 N2 220.222
225 22220222222222 . NE SN
222292222.
2222222222
522
a
25222. N 22.2 <02 2 o
2.25.35.52—
.5223 use: E: he a 2x: 2 w:— aecoouo: .25. apnea—22a
v.23. Nuisance—2.N! .252 net—2.33m deuce-«um .U m

 

222.2282 .N «2.2.2.2.

40

 

J «2.222.232

 

 

2.2 2.2 N.N 2.22 22.N N 32882222 222222220
. . . . . i— @2933.»

N 2. 22 2 N 2 N 22 N 2 N 3882222 2222.22.

N2222N .222 388 2.22222
20 825:. EN2 22.N NN 22.2 2N.22 2e22 NNN 222aNNN2 mm232222.222 Soc N .2 22:223.: 28220252 N22
-aosem .2 -2222 -222 -2222 92.902.222.22 2223.. 22822 252 38322.2 522252.

. 2220...: $2222

Be: 2822
N.N-N.2 N.N N22 22828222 N .2 222.23.: 2822.22.52 2.22
-N. 2 -22: BEN—0825.. $229362: 52222on
N SN

88 .222 388 2.2.2222
No £89m mated 0.2m :2 €25va ”Nam—c.2222»: So: 9223 m A @8322: Amounts, 3V
$5 2.2 .3 929322222222 22222... as N 2222N 38332 5222522

. 22002 NON
N528 22:82.3
22225220 0328

. . . . 22 Sec
2. 8 2. N2 N N 2. N moo: $222 22322 as N sacsNes 2232252222
-222 2.22 -222 2.22 .222 max—ougm we 22cm 3%: + 85
22:82.22 9302.223 CV
3930
. . . . A222: Boa 8:25:22 3.5
I Nam 22 z o N2 22 323M223

..2M222%2N6 .2.“me hm. ow“. 02w“; NNN 22% a: 2.NN2 imam M22224 522 an N 2 22.N: 2822022? 22
. 9.902.222.2222 NE SN . 8223220

 

2.2.2.282 .N 0.22.2.2.

41

23:22 53» 223382223 N223 23.2228 20525222 89:8 02.: No 29232 .22 6.532222 2x22 2 n N283 322.2 3822322 .333 Emma? smog Soc

283 938 592.2 ENS Emma? .CE Sou 2222022 .32 .«o w 2: Non 9: 8 “23.5228 0.2025 Nos—22> .o fine—8,5220 Boa 22583829222 Na 330.298
338.2 .225me: NBS—2235-258 .3 22022222522022 2228280 222525 289—. 05 Beomofiom .22 22808022 «on do 26.20.29582 .M 22222882222 .0
$222332 Sufi ”Boa 292:8 .<m ”Eon 22:25.2 .22 ”Bow ogooi {On 6.8a 06.2.2220 @220 ”N28 oggozxofimn J20: 2222 PEEP? .5m 2222

BEEN.» 23> 2222 Eggnoagnfi {mam ”Boa. 02222282Noo2oa {022 $an 28523on3 .225: .332: we emcee N No N22282: one N025» .N

 

42

 

 

 

. . . .2 «.53....
N N 2 N N 22 N 3283222 22222222.v
. . . .2 atom?
22 N N 2 2 N N 383$ 2222.22.
2.22222N
.2... . -. .5252... 2 .2222...“ 2 .22. it... its...
8822... 22 2.2 2. 22 N N N 22 N N N 22 . NNN 95 2 222.22.. . 2 222 22 522352 N 328822.22 502222.222.
9.82%.”2 0222222222 2223. 22032 sNN
Boa 2.2322
. - . . - . . - . 2232822 .2 MEN? 2322222222.2 2.22
N N2 2 2 N N2 N 22 N 2 2. 22 222922823 N 32882222 582522
N 2222N .
88 .222 28252282 23822 so .
. - . . - . . - . . 22 22222222 2 2.2322: 28222222“.2 2.22
20 22222223 N NN 2 N N R 2. 2 N N N 22 9222 2.25.22. Nazca? as N 2222N N .3288: 52252
.5222 0222222222 2229.. 2220022 $2222
NEE
252; 22 0
2222228222»...—
BeE add—eHMW—M—Mw 3!»:— awwucoth—wmaa can Nam—22.2.2»! .— nod—O 56m
2222228222222

 

92.22282 .N 22222."...

acid. The total phenolic content (TPC) determined fi'om the F olin-Ciocalteau method
was highly varied and ranged from 34.6 to 376.6 mg/ 100 g of bean flour. The
predominant hydroxybenzoic acid present following alkaline hydrolysis was
protocatechuic acid (9.3 to 92.7 mg/ 100 g of dry bean flour), whereas ferulic acid (0.4 to
12.4 mg/ 100g of dry bean flour) was the predominant hydroxycinnamic acid present after
hydrolysis.

Luthria and Pastor-Corrales (2006) quantified hydroxycinnamic acids in 15
different varieties of raw dry common beans (Phaseolus vulgaris L.). They ranged from
19.1 to 48.4 mg per 100 g of dry bean flour following alkaline hydrolysis. Cranberry
beans had the lowest content, while navy bean had the greatest, followed by black and
pinto beans. p—Coumaric, ferulic and sinapic acid were found in all varieties analyzed,
but caffeic acid was only identified in black beans. Ferulic acid was the predominant
hydroxycinnamic acid identified after alkaline hydrolysis, ranging from 10.6 mg/ 100 g of
dry bean flour in alubia beans up to 26.6 mg/ 100 g of dry bean flour in navy beans.
prez-Amoros et al. (2006) reported the phenolic compounds in raw dry beans
(Phaseolus vulgaris L. var. La Granja). Hydroxybenzoic acids identified included
protocatechuic, p-hydroxybenzoic and vanillic acid. Total ranged from 0.15 to 0.18
mg/ 100 g of dry bean flour. The only hydroxycinnamic acid identified was ferulic acid,
ranging from 0.34 to 0.37 mg/100g dry bean flour. No hydrolysis was performed on the
bean samples.

Diaz-Batalla et al. (2006) investigated the phenolic compounds in several
varieties of Mexican common beans (Phaseolus vulgaris L.) raw and cooked. This was

one of the few studies investigating phenolic compounds in processed beans. After acid

43

hydrolysis hydroxybenzoic acids identified included p-hydroxybenzoic, vanillic and
syringic acid, totals ranging from 1.0 to 3.1 mg/100g dry bean flour for the raw beans.
This was reduced to 0.4 to 2.1 mg/ 100 g dry bean flour if the beans were cooked,
approximately 46% reduction. p-Coumaric and ferulic acid were the only
hydroxycinnamic acids identified in the raw and cooked Mexican bean varieties. Totals
ranged from 2.0 to 4.3 mg/ 100g dry bean flour to 1.2 to 3.3 mg/100g dry bean flour for
raw and cooked beans, respectively. There was approximately a 30% reduction in
hydroxycinnamic acids after processing (cooking). Ferulic acid was the predominant
phenolic acid regardless of processing. Additionally they identified the flavonol
aglycones, quercetin and kaempferol. Quercetin ranged from 0.7 to 2.3 mg/ 100g dry
bean flour in raw beans and 0.4 to 1.2 mg/ 100g dry bean flour in autoclaved/cooked
beans. Kaempferol was the predominant flavonol, ranging from 1.4 to 20.9 mg/ 100 g dry
bean flour in raw beans and 0.7 to 12.3 mg/ 100 g dry bean flour in autoclaved/cooked
beans. They reported the reduction from cooking the Mexican common beans ranged
from 12 to 65% and 5 to 71% for quercetin and kaempferol, respectively.
Espinosa-Alonso et al. (2006) investigated the phenolic compounds in 62
different varieties and cultivars of Mexican dry common beans. They also looked at the
phenolic compounds in J amapa black beans and pinto beans. They identified
hydroxybenzoic acids, hydroxycinnamic acids and flavonol aglycones following acid
hydrolysis. The hydroxybenzoic acids included: p-hydroxybenzoic, vanillic and syringic
acid. The total hydroxybenzoic acids in raw Mexican common beans ranged fiom 1.0-
7.1 mg/ 100 g dry bean flour. The raw J amapa black and pinto beans containd 4.2 mg and

2.9 mg total hydroxybenzoic acids/ 100 g dry bean flour, respectively. Caffeic, p-

coumaric, ferulic and sinapic acid were identified in all beans analyzed. The total
hydroxycinnamic acid in raw Mexican common beans ranged from 1.3 to 7.7 mg/ 100 g
dry bean flour. For raw Jamapa black bean the total hydroxycinnamic acids was 4.9

mg/ 100 g dry bean flour, and for pinto beans it was 6.1 mg/ 100 g dry bean flour. Ferulic
acid was the predominant hydroxycinnamic acid in all bean samples. Quercetin and
kaempferol ranged from 0.2 to 5.2 and 0.2 to 8.8 mg/ 100 g dry bean flour, respectively,
in the raw Mexican beans. J amapa black beans had 2.1 mg of quercetin and 1.5 mg
kaempferol/100 g dry bean flour. Pinto beans had 0.2 mg of quercetin and 2.1 mg
kaempferol/ 100 g dry bean flour. Espinosa-Alonso et al. (2006) also identified the
anthocyanidins: delphinidin, petunidin, cyanidin, malvidin, pelargonidin and peonidin, in
the Mexican beans possessing black, mottled gray and heterogeneous color mixture seed
coats and the Jamapa black bean. A few of the varieties and/or cultivars were found to
contain the isoflavones: daidzein and coumestrol. Both Diaz-Batalla et al. and Espinosa-
Alonso et al., however used acid hydrolysis to quantify hydroxycinnamic acids, as stated
in a previous section, acidic conditions could increase the loss of hydroxycinnmic acids
and decrease the total amount of hydroxycinnamic acids.

Different processing methods will induce physical and chemical compositional
changes in legumes, altering their nutrient content and these changes are dependent on
the legume type and processing conditions. Legumes must be processed before
consumption to inactivate anti-nutrients e.g. lectins, protease inhibitors, phytates.
Processing methods include hydration (soaking), heat treatment (i.e. boiling, pressure
cooking or autoclaving), germination or fermentation, or a combination of 2 or more

methods (Geil and Anderson, 1994; Messina, 1999; Champ, 2002; Ibrahim et al., 2002;

45

Shimelis and Rakshit, 2007). Hydration or soaking legumes prior to either heat
treatment, germination and/or fermentation, is a typical preliminary step. It softens the
legume’s texture and reduces their cooking time. Generally, the more polar compounds
will be leached into the soaking water, as well as into the cooking water, which are
normally discarded. Thus in studies examining the effect of soaking on grain legumes’
phenolic content and antioxidant activity, a decrease has been observed (Towo et al.,
2003; Lopez-Amoros et al., 2006; Xu and Chang, 2008). It has been estimated that 30%-
40% of the phenolic compounds in legumes are lost when the cooking water is discarded
(Bressani and Elias, 1980).

Very few studies have investigated the effect of processing on phenolic
compounds in legumes. Of those that have, the majority simply measured total phenolic
content or classes of phenolic compounds. Granito et a1. (2008) reported an overall
decrease in the total phenolic content and antioxidant activity after cooking (boiled) or
fermenting black beans (Phaseolus vulgaris L.). However, fermentation did not reduce
antioxidant activity to the same degree as cooking. Towo et a1. (2003) investigated the
affect of germination, cooking or dehulling on total phenolics on several species of grain
legumes. Dehulling significantly reduced the total phenolic content by 52%-67% in all
legumes analyzed and condensed tannins content by 82%-93%. Cooking significantly
reduced the total phenolic content and condensed tannin content by 29%-5 5% and 46%—
69%, respectively. Germination significantly reduced the total phenolic content and
condensed tannins by 25%-44% and 35%-72%, respectively. The authors concluded that
dehulling and heat treatment were the most effective in reducing the total phenolic

content in the analyzed grain legumes. The dramatic loss in condensed tannins following

46

dehulling is not surprising because they are highly concentrated in the seed coats of
legumes (Beninger and Hosfield, 2003; Madhujith et al., 2004). Lopez-Amoros et al.
(2006) also examined the effect of germination on grain legumes. They concluded that
germination alters the phenolic composition of legumes and their antioxidant activity, but
the changes are dependent on the type of legume and the germination conditions such as
length of time and the presence or absence of light (Lopez-Amorés et al., 2006). Rocha-
Guzman et al. (2007) evaluated the effect of pressure cooking (autoclaving) on
antioxidant activity in common beans. Cooking induced a large reduction in total
phenolic content measured in the seed coats and cotyledons of common beans and the
antioxidant activity in the cooked legumes was similar to that in raw legumes.
Interestingly, when the freeze-dried material from the cooking water was included in the
analysis, the antioxidant activity in cooked legumes increased. Xu and Chang (2008)
investigated the effect of soaking, boiling and steaming on total phenolic content and
antioxidant activity of cool season legumes i.e. green and yellow peas (Pisum sativum
L.), chickpeas (Cicer arietinum L.), and lentils (Lens culinaris). They concluded that all
processes resulted in reductions in total phenolic content ranging fiom 40%-68%, but the
amount depended on the legume type and processing conditions. Loss due to soaking
ranged from 2% to 38%. Loss in antioxidant activity depended on the in vitro antioxidant
assay utilized, the legume type and the processing conditions. DDPH free radical
scavenging activity value was significantly reduced by all processes ranging from 6%-
95%. The oxygen radical absorbing capacity (ORAC) values were reduced in all

legumes only after soaking or regular boiling, range of 4%-7 7 %. Conversely, the ORAC

47

values significantly increased after pressure boiling (except in lentils) and pressure
steaming, ranging from 5% up to 175%.

The modifications in phenolic composition via processing methods, may be a
result of the following: phenolic compounds leaching into the soaking or cooking water;
decreases in the amount of extractable phenolics because of the formation of complexes
with macromolecules; degradation and/or formation of compounds, phenolic or
otherwise; endogenous enzyme activation during germination i.e. polyphenoloxidases,
hydrolases, esterases (Towo et al., 2003; Lopez-Amoros et al., 2006; Sangronis, 2006;
Shimelis and Rakshit, 2007).

With the increasing public awareness of phenolic compounds, especially as
antioxidants, and the health benefits associated with fi'uit and vegetable consumption,
identifying and quantifying phenolic compounds in all plant foods is imperative. If a
plant food is processed prior to consumption, it is more nutritionally relevant and just as
vital to analyze the phenolic compounds in that processed plant food. In a published
review on the subject of hydroxycinnamic acids, Clifford (1999) observed that assessing
the dietary intake of hydroxycinnamic acids was not possible. He stated, “With the
exception of some beverages, the lack of data for the composition of commodities as
consumed, i.e. the edible portion after processing, cooking, baking, etc, effectively make
such an assessment impossible”. Since legumes must be exposed to some sort of heat
treatment to be safe to consume, it is important from a dietary standpoint to quantify the
phenolic compounds in heat treated grain legumes i.e. dry common beans.

C. Colorectal Cancer (CRC)

48

1.1. Colorectal Cancer Statistics
Colorectal cancer (CRC) is the 3rd most commonly diagnosed cancer and the 4th
most common cause of cancer death worldwide (WCRF/AICR, 2007). In the United

States alone, CRC is the 4th most commonly diagnosed cancer following lung, female

breast and prostate cancers, and the 2"d leading cause of cancer death (J emal et al, 2008).

It was recently noted that in the US. significant decreases in CRC incidence and death
rates for men and women combined have been observed, 2.1% per year (from 1998 to
2003) and 2.8% per year (from 2001 to 2003), respectively (Ries et al, 2006). These
declines have been attributed to increased public awareness and improvements in the
methods of early detection and treatment (J emal et al., 2007). Colorectal cancer
screenings allow for the detection and removal of pre-cancerous polyps (hyperplastic
polyps or adenomas) and early-stage cancers in asymptomatic individuals, preventing
their progression to more invasive cancer (adenocarcinomas) and allow for detecting
them at the more treatable early stages, leading to a reduction in the number of deaths
fi‘om CRC (Hawk et a1, 2004). In fact, experts believe as many as 50-60% of CRC deaths
could be prevented if everyone age 50 or older were to be screened regularly (Selby et al.,
1992). Nevertheless it is estimated that 148,810 new cases of CRC will be diagnosed this
year in the US, resulting in 49,960 deaths (J emal et a1, 2008).

Approximately 88% to 94% of CRC cases are sporadic, occurring in individuals
who do not have an inherited colon cancer syndrome (Compton, 2003; Weitz et al.,
2005). An individual’s risk of developing colorectal cancer increases with age. Risk
rises sharply fi'om age 40 to 50, doubles with each subsequent decade and peaks at age 70

(Compton, 2003). Other risk factors include male sex, hormone exposure, history of

49

colorectal polyps, history of CRC or cancers of the small bowel, endometrium, breast or
ovary (W eitz et al., 2005). Lifestyle and environmental risk factors include diets rich in
red and processed meats, high in animal fats and sugars, low in folate, calcium and
dietary fiber; low physical activity; obesity; diabetes mellitus; smoking; high alcohol
intake (especially in men); occupational hazards e. g. asbestos exposure; or radiation
exposure (W eitz et al., 2005; WCRF/AICR, 2007). Approximately 20% of sporadic CRC
cancers are due to hereditary factors or familial history of colorectal cancer without
fulfilling the criteria for the known hereditary syndromes (Weitz et al., 2005). Chronic
inflammatory bowel diseases i.e. ulcerative colitis and Crohn’s disease increase the risk
of developing CRC and account for 1%-2% of the CRC cases (Weitz et al., 2005).
Patients with inheritable CRC syndromes are predisposed to cancer and these
syndromes account for 5%-10% of cases of CRC. Two highly studied syndromes are
familial adenomatous polyposis (F AP) and hereditary nonpolyposis colorectal cancer
(HNPCC). FAP is an autosomal dominant disorder, with approximately 80% of affected
FAP patients possessing a gerrnline mutation in one allele of the adenomatous polyposis
coli (APC) gene. Although individuals with F AP are predisposed for CRC, it does not
guarantee they will develop it. Studies of PAP patients and mice with similar mutations
in the murine homolog of APC gene require the loss of the 2nd allele via a somatic
mutation or by chromosomal loss or deletion, the latter being more common (Kinzler and

Vogelstein, 1996). The loss of the 2ml allele is required for the CRC phenotype to

manifest. This loss of heterozygosity (LOH) is required for the development of colon
cancer, and is consistent with the Knudson “two-hit” hypothesis that requires at least two

genetic alterations for the development of cancer. FAP patients will develop 10 to 1000

50

adenomas throughout their colons, 50% by age 15 and 95% by age 35. Although the
adenomas are benign, if left untreated, CRC will develop by the age 50. The most
common treatment for F AP patients is prophylactic colectomy. Additionally patients are
at increased risk for developing extracolonic tumors such as duodenal carcinomas,
desmoid tumors, osteomas and brain tumors (Kinzler and Vogelstein, 1996; Weitz et al.,
2005; Bametson and Dunlop, 2007).

HNPCC or Lynch syndrome is an autosomal dominant syndrome caused by a
germline mutation in the DNA mismatch repair genes (MMR) i.e. hMLHI, hMSHZ. A
typical molecular characteristic of HNPCC tumors is the presence of microsatellite
instability (MSI). Microsatellites are short repeats of non-coded DNA sequences
distributed throughout the genome. M81 is described by small insertions or deletions in
microsatellite regions in the DNA of tumors. Interestingly, MSI is found in
approximately 15% of sporadic CRC not from a germline mutation in MMR genes but
resulting from the epigenetic silencing of MMR genes. The onset of CRC in HNPCC
patients occurs at around 40 yrs of age, and characteristically a larger proportion of
tumors are located within the proximal colon (right sided) compared to sporadic cancers
(left sided, distal colon). Moreover extracolonic tumors fi'equently develop (W eitz et al.,
2005; Bametson and Dunlope, 2007).

Other inheritable autosomal dominant syndromes include Peutz-Jegher’s
syndrome, Juvenile polyposis and Cowden Syndrome. These disorders are more rare
compared to FAP or HNPCC and are characterized by the formation of hamartomas in

the colon (Weitz et al., 2005).

51

1.2. Colorectal Carcinogenesis

Carcinogenesis is a multi-step process wherein a series of genetic events are
necessary to drive the transformation of normal cells to malignant cancers (Hanahan and '
Weinberg, 2000). The process begins with an initial event where a single cell
experiences a genetic mutation resulting in a selective growth advantage over
surrounding cells, followed by clonal expansion of the mutated cell and the acquisition of
further genetic alterations (F earon and Vogelstein, 1990). These mutations occur in
genes responsible for normal cell proliferation, differentiation, adhesion and survival, and
these alterations often advance the growth of the transformed cells.

Carcinogenesis progresses slowly, taking years to fully manifest, approximately
10 to 20 years from the initiation event to malignant transformation (Lipkin, 1999; Dove-
Edwin and Thomas, 2001). In “Western” populations 50% of the population by the age
of 70 will likely develop colorectal tumors, and in about 10% of these individuals the
benign adenoma will progress to the malignant adenocarcinoma or CRC (Kinzler and
Vogelstein, 1996). CRC remains the paradigm utilized to study the molecular,
biochemical, cellular, and genetic alterations occurring in the carcinogenesis process
(F earon and Vogelstein, 1990). The large intestine is made up millions of crypts, which
are around 50 cells deep (see Figure 4., Lipkin, 1999; Heavey et a1, 2004, Humphries and
Wright, 2008). Colonic cells are under constant renewal, with the entire human colonic
epithelium being replaced every 3 to 8 days (Lipkin, 1999). Stem cells located at the
bottom of the crypt replicate resulting in the original stem cell and a new daughter cell.
The daughter cells continue to replicate and push upward toward the top of the crypt.

The cells occasionally divide as they migrate up the crypt column. There are very few

52

 

Int tinal L
it} Apoptotic Cells es umen

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

i133 ——'
CI Differentiating Cells 5:?

I Proliferating Cells [ Ll l j 1 1 1 {3‘ ]
I Stem Cells : : : ‘

Proliferative
Zone
Crypt Crypt
Basal Side

Figure 4. Normal Colonic Crypt

The large intestine is made up of millions of crypts, which are around 50 cells
deep (Lipkin, 1999; Heavey et a1, 2004, Humphries and Wright, 2008). Colonic cells are
under constant renewal, with the entire human colonic epithelium being replaced every 3
to 8 days (Lipkin, 1999). Stem cells located at the bottom of the crypt replicate resulting
in the original stem cell and a new daughter cell. The daughter cells continue to replicate
and push upward toward the top of the crypt. The cells occasionally divide as they
migrate up the crypt column. There are very few dividing cells in the top 1/3 of the crypt
where they begin to differentiate into mature non-proliferative epithelial cells. Once at

the lumenal surface the mature cells undergo apoptosis and are excreted in the feces.

53

dividing cells in the top 1/3 of the crypt where they begin to differentiate into mature
non-proliferative epithelial cells. Once at the lumenal surface the mature cells undergo
apoptosis and are excreted in the feces. The crypt design ensures that damage to any
epithelial cells induced by genotoxic compounds found in the intestinal lumenal contents
will not influence the integrity of the crypt (Heavey et al., 2004). Crypt cell kinetics or
the molecular processes involved in maintaining the balance between new cell production
and cell death are under tight regulation in highly renewing tissues like the colon (Lipkin,
1999). Disruptions in any one of these processes can contribute to the development of
colorectal cancer.

Given the accessibility of the colon, animal models of colon carcinogenesis and
inheritable human CRC syndromes have permitted researchers the opportunity to study
the pathological, histological and molecular alterations occurring during the different
stages of colorectal cancer development, or during the “adenoma-carcinoma sequence”
(Figure 5). This term is used to describe the progression from normal epithelial cells to .
dysplastic cells to benign adenoma to malignant adenocarcinoma to metastasis, and the
accompanying epigenetic and genetic alterations (Fearon and Vogelstein, 1990; Leslie et
al., 2002).

Fearon and Vogelstein (1990) proposed the following: 1) colorectal tumors result
fi‘om the mutational activation of oncogenes coupled to the inactivation of tumor
suppressor genes, 2) at least 5 to 7 genetic alterations are required for normal cells to
progress to carcinoma (malignant tumor), although less are required for benign

tumorigenesis, 3) even though genetic changes commonly occur in a preferred sequence,

54

ASSN 2.? .0 >950: 5&3 2583032» .252 2222.2:th
ammo—20359301332220 .28 .2222: 252.00 < “cones—20m «52259222022822.2352 .m ensur—

 

 

 

 

 

 

 

 

 

 

 

 

55

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2.2232362

222232822

2:55—25 coca—.AN—«ofieam:
(Za—
mEoEBNO H

«228522.22. 2222222222242. 222222832. Ema—QED 2522205222..”—

NEE . 2:2 022.2 N22222222223222 ram 382222252 2822222
.22 N223
NE
2222 N8.— » neEu< VORQ Sui 353:3
GOG «22 22.25 N22 2222:5222 we N234 .222 22223532

it is the total accumulations of mutations, not their order of occurrence in respect to each
other, that determines the tumor’s biological properties, 4) some mutated tumor
suppressor genes appear to exert a phenotypic effect even when present as a heterozygous
state. Two pathways, the gateway pathway and the caretaker pathway, can be used to
explain the CRC process (Figure 5). Mutations in genes that regulate cell grth such as
oncogenes (e.g. K-ras, c-myc, c-erb2, c-src) and tumor suppressor genes (e. g. APC,
DCC, DPC'4/Smad4, p53, nm32) represent the gatekeeper pathway. Whereas the
caretaker pathway involves mutations or epi genetic changes in genes that maintain
genetic stability e. g. mismatch repair genes (Kinzler and Vogelstein, 1996; Leslie et al.,
2002; Weitz et al., 2005). Figure 5 describes the widely accepted and well characterized,
genetic alterations that occur during the adenoma-carcinoma sequence (F earon and
Vogelstein, 1990; Kinzler and Vogelstein, 1996). Nevertheless, the order of genetic
alterations and the possibility of subsets of colorectal tumors proceeding through a
different set of genetic alterations cannot be overlooked (F earon and Vogelstein, 1990;
Yamashita et al., 1995).

Animal models of colon carcinogenesis have provided a great deal of information
about the pathogenesis of colon cancer and the influence of dietary and environmental
factors upon its development. Unlike humans however, animals rarely develop
spontaneous cancer of the colon and rectum (F emia and Caderni, 2008). Therefore colon
cancer must be induced in animals using exogenous chemicals. 1, 2-Dimethylhydrazine
(DMH) and its metabolite azoxymethane (AOM) are two commonly used carcinogens
utilized in studies to induce colon cancer in animals (Femia and Caderni, 2008).

Evidence has shown that chemically induced colon carcinogenesis follows the multi-

56

stage carcingoenesis process, beginning with initiation followed by promotion and
progression (Reddy, 2004; F emia and Caderni, 2008).

Azoxymethane (AOM) is a potent generator of colon carcinomas of the large
intestine in selective strains of rodents and an accepted model used to study the pathology
of human colon carcinogenesis (Druckrey, 1973; Reddy, 2004). Many studies
investigating the effect of dietary/nutritional factors and chemopreventive agents on
colorectal carcinogenesis have utilized AOM (Reddy, 2004). The first tumor is often
visible 15 weeks afier treatment and tumors have a mean latency period of about 20
weeks (5 months). Studies using tumors as the endpoint are usually terminated 7 to 9
months after AOM treatment when tumors are developed (Reddy, 2004). AOM
methylates DNA primarily at the 06 position of guanine, resulting in 06-methylguanine
adducts. If left unrepaired, the adduct will pair with thymine, instead of cytosine,
resulting in G—vA transition during DNA replication (Wali et al., 1999).

In short-term colon carcinogenesis experiments, it is widely accepted to utilize
aberrant crypt foci (ACF), early preneoplastic lesions, as biomarkers to screen for
potential chemopreventive agents as well as carcinogens prior to long-tenn studies
(Roncucci et al., 1991; Femia and Caderni, 2008). ACF have been observed in the
intestines of rodents injected with the colon carcinogens: AOM and dimethylhydrazine
(DMH, an AOM precursor). These morphological precursors can be viewed
microscopically after whole colons are stained with methylene-blue. Single (aberrant
crypts) or clusters (ACF) of morphologically modified colonic crypts stain darker than
surrounding areas, possess altered lumenal openings (oval or slit-like), thickened

epithelia and appear larger than adjacent normal crypts. They are visible by 5 weeks, but

57

commonly examined 12-15 weeks post-carcinogen treatment (Bird et al., 1989).
Additionally ACF have been identified in the colons of humans with FAP or CRC
(Roncucci et al., 1991; Pretlow et al., 1991; Yokota et al., 1997; Schoen et al., 2008).
There is some debate on whether ACF correlate with tumor development and if they are
true precursors of tumors (Davies and Rumsby, 1998; Mori et al., 2005; Gupta and
Schoen, 2008; Schoen et al., 2008). Other preneoplastic lesions identified and suggested
to that correlate better with tumor development include B—catenin accumulated crypts
(BCAC) (Y amada et al., 2001) and mucin depleted foci (MDF) (Caderni et al., 2003).
Tumors remain the best endpoint to evaluate the efficacy of chemopreventive
agents and dietary components against AOM-induced colon cancer. The distribution of
the intestinal tumorsiin AOM injected rodents is similar to that exhibited in humans.
Small intestinal tumors, although rare, are generally found in the duodenum, and large
intestinal tumors are generally found in the distal 2/3 of the colon (corresponding to the
left side or descending and sigmoid colon in humans) (Ward et al., 1973; Zedeck, 1980).
Histolopathological characteristics and genetic alterations identified in sporadic colonic
tumors in humans have also been identified in AOM-induced tumors in rodents
(Druckrey, 1973; Ward et al., 1973; Reddy, 2004). Three types of colonic tumors
identified include: adenomas, adenocarcinomas and mucinous adenocarcinomas (Ward et
al., 1973; Shmnsuddin, 1983). I
There are numerous genetic alterations occurring during the colorectal cancer
process, resulting in perturbations in signaling pathways regulating cellular homeostasis.
Mutations and/or deletions in APC, K-ras, DCC and p53 genes are fiequently found in

human colorectal cancer (Figure 5). Alterations in APC and K-ras genes often occur

58

during the early stages of the carcinogenesis process, whereas DCC and p53 are involved
during the late stages (Kinzler and Vogelstein, 1996; Takahashi and Wakabayashi, 2004).
In AOM-treated rats, mutations in codon 12 of the K-ras oncogene have been detected at
high frequencies in ACF and adenocarcinomas (Erdman et al., 1997; Takahashi and
Wakabayashi, 2004). This mutation results in the continuous activation of the K-ras
oncogene, a component in the RAS-RAF-MAPK signaling pathway involved in cell
growth and survival and has been identified in 50% of colorectal carcinomas (Fearon and
Vogelstein, 1990). Although mutated K-ras has also been detected in AOM-treated mice
models, it does not occur as frequently as that in rats (Guda et al., 2004). In contrast,
mutations in APC, DCC and p5 3 are rarely detected in ACF and adenocarcinomas found
in AOM-treated rodents (Erdman et al., 1997; Takahasi and Wakabayashi, 2004).

One of the earliest genetic events to occur in colorectal carcinogenesis is the loss
of a functioning APC protein. Approximately 80% of affected FAP patients possess a
germline mutation in the APC gene and somatic mutations have been observed in more
than 50% to 80% of sporadic colorectal tumors (Kinzler and Vogelstein, 1996; Cassidy et
al., 2007; F emia and Caderni, 2008). Yet mutations in this gene are rarely observed in
AOM induced ACF, adenomas or adenocarcinomas in rodents. The APC protein is
responsible for multiple cellular fiinctions, but in carcinogenesis, its role as a tumor
suppressor is of great importance. It is responsible for regulating the cellular levels and
location of B-catenin, a cell adhesion protein family member and essential component of
the Wnt signaling pathway, a pathway reportedly involved in the transformation of
normal cells to malignant tumors (Buda and Pignatelli, 2002). The loss of a functioning

APC protein leads to the accumulation of B-catenin in the cytoplasm and nucleus and

59

increased activation of B-catenin target genes i.e. cyclin D and c-myc, both important in
cell proliferation and shown to be up-regulated in colon cancers (Takahashi and
Wakabayashi, 2004). Although APC is not often mutated in AOM induced colon cancer,
researchers have reported altered cellular location of B-catenin in AOM induced
dysplastic ACF, adenomas and adenocarcinomas in rodents and mutations within the ,B-
catenin gene were identified (Guda et al., 2004; Takahashi and Wakabayashi, 2004). In
humans, 4 - 15% of all sporadic colorectal adenomas and carcinomas have mutations in
the fl-catenin gene, yet a functioning APC protein (Morin et al., 1997; Buda and
Pignatelli, 2002). Thus mutations in the ,B-catenin gene can contribute to the nuclear
accumulation of B-catenin protein and increased transcription of target genes, similar to
that observed with mutant ACF protein.

High throughput cDNA microarrays have permitted researchers to analyze the
expression of tens to thousands of genes in a variety of species. Using microarray
technology Rondini (2006) investigated genetic alterations in AOM-induced colon
tumors compared to surrounding normal appearing colonic mucosa in F344 rats. The
expression of genes involved in inflammation, extracellular matrix (ECM) formation, and
cell cycle regulation were upregulated in AOM-induced tumors, whereas anti-inflammory
genes, tumor suppressors genes and genes encoding ion transporters and metabolic
enzymes were downregulated. This is in agreement with evidence showing enhanced
expression of enzymes associated with inflammation observed in human CRC and in
AOM induced colon carcinogenesis. For example, COX-2 and iNOS, important
inflammatory enzymes upregulated in human CRC, are reportedly over expressed in

AOM induced rat colorectal cancers as well (Takahashi and Wakabayashi, 2004). In

60

addition to an increased expression of COX-2 in AOM induced rat colonic adenomas and
adenocarcinomas, Shao et al. (1999) found an overexpression in TGF-Bl in the same
colonic lesions. These authors suspected the overexpression of COX-2 observed during
AOM induced colon carcnogenesis results fi'om the overexpression of TGF-Bl (Shao et
al., 1999). I

The TGF-B signaling pathway controls cell division, differentiation, migration,
adhesion, extracellular matrix deposition and programmed cell death (apoptosis) in many
cell types (Harradine and Akhurst, 2006; Massagué et al., 2006). TGF-B and its family
members can exert dual functions depending on the cell type and time of development,
for example during embryogenesis TGF-B promotes tissue growth and morphogenesis,
whereas in mature adult tissues it maintains homeostasis by activating cytostatic
processes i.e. inhibiting cell growth and multiplication, as well as cell death processes
(Massagué et al., 2006).

Transforming growth factor (TGF-B) is a member of a superfamily of cytokines
that include: the nodals, activins, bone morphogenic proteins (BMP), myostatin and anti-
Muellerian hormone (AMH). Three isofonns of TGF-B are expressed in mammalian
cells, i.e. TGF-Bl , TGF-BZ and TGF-BB. TGF -B1 is the most abundant and ubiquitiously
expressed isoform. Mammalian cells secrete TGF-B into the extracellular matrix as the
inactive latent protein complex, LTGF-B. The LTGF-B complex consists of the mature
TGF-B protein non-covalently associated to the N-terminal TGF-B propeptide called
latency-associated peptide (LAP), which is covalently bound to a latent TGF-B binding
proteins (LTBPs). In order for the TGF-B ligand to associate with its membrane surface

receptors it must be activated by being released from the LTGF-B complex. Although the

61

exact mechanism activating the TGF-B ligand is not fully understood, it involves
proteolytic cleavage of LTBP and the release or a conformational change in LAP.
Activation occurs in response to microenvironmental signals. Once activated the
members of the TGF-B superfamily signal through one of two ligand-receptor-Smad
signaling branches: 1.) TGF-Bs/ activins/ nodals to Smad 2/ 3; 2.) BMPs and AMHs to
Smad 1/ 5/ 8. There are two high affinity TGF-B receptors, type I and type II that are
serine/threonine kinase transmembrane receptors and are ubiquitously expressed. A third
low affinity TGF-B receptor, type III or betaglycan, is a membrane-anchored
proteoglycan responsible for capturing the TGF-B ligand and presenting it to the TGF-B
receptors type I and II. The activated TGF-B ligand binds the type II receptor which
results in the recruitment of the type I receptor forming a heteromeric complex. The type
II receptor phosphorylates the type I receptor and activates its kinase. The activated
ligand-receptor complex attracts and phosphorylates downstream target proteins i.e.
Smad proteins. Humans and mice encode for eight Smad proteins, i.e. Smad 1-5,8 and
Smads 6 and 7 the inhibitor Smads. The type I receptor in the ligand-receptor complex,
phosphorylates the cytoplasmic Smad proteins. These phosphorylated Smads form of
heteromeric complex with Smad 4, an essential component in all Smad signaling
pathways, which allows the complex to be translocated into the nucleus. The Smad
complex can then combine with a variety of DNA-binding proteins or cofactors, i.e.
FOX, HOX, RUNX, E2F, AP-l, CREB/ATF, Zinc-finger. It is this Smad-cofactor
combination, determined by cell-specific factors, that specifies the subset of genes that
will be bound, and which transcriptional co-activators or co-repressors will be recruited.

Smad activation can lead to hundreds of immediate gene activation or repression

62

responses. Although Smad signaling is essential for most TGF-B responses, other
mediators of signaling such as mitogen activated protein kinases (MAPKs)/ERK, Jun N-
terminal kinase (INK), p38, P-I3K kinase, PP2A phosphotases and Rho family members,
can be activated (Derynck et al., 2001; Elliott and Blobe, 2005; J avelaud and Mauviel,
2005; Harradine and Akhurst, 2006; Massagué et al., 2006; Seoane, 2006).

TGF-B inhibits cell proliferation in specific cell types, i.e. epithelial, endothelial
and hemotopoietic cells (Massagué et al., 2006; Seoane, 2006), chiefly by preventing cell
progression through the G1 phase of the cell cycle. In epithelial cells in the normal
colonic crypt, the expression of TGF-B increases from the bottom of the crypt to top, with
the highest amounts expressed at the lumenal surface where the mature, differentiated
cells are located (Shao et al., 1999). Its expression is inversely related to proliferation.
TGF-B induces the transcriptional expression of p21 CIPI and p1 5Ink4b, cyclin-
dependent kinase (CDK) inhibitors, while concurrently repressing the expression of c-
myc and ID], 2 and 3. c-Myc, a proto-oncogene, promotes cell growth and proliferation
and the IDs are nuclear factors capable of inhibiting cell differentiation. c-Myc is
frequently mutated in many types of cancers. Furthermore the increased expression of
p15 leads to the redistribution of another CDK inhibitor, p27Kip1, from the active p27-
cyclin D-Cdk 4/6 complex to cyclin E-Cdk 2, inactivating Cdk 2, a late Gl/S-phase
kinase. Thus in normal epithelial cells TGF-B signaling induces cell cycle arrest and
clearly helps to regulate colon crypt cell kinetics (Massagué et al., 2006).

Abnormal TGF-B signaling has been associated with cancer, and plays a dual role
in the tumorigenesis process by acting as a tumor suppressor during the early stages and a

promoter in the late stages (Seoane, 2006). In fact, a large majority of epithelial derived

63

tumors (285% of all human cancers) no longer respond to the anti-proliferative signals
induced by TGF—B, resulting in a selective growth advantage for these tumor cells (Elliott
and Blobe, 2005). In CRC, resistance to the tumor suppressor effect of TGF-B results
from mutations in components of the TGF—B signaling pathway, i.e. T GF—flRII receptor
and SMAD 4 (Markowitz et al., 1995; Zhou et al., 1998; Grady et al., 1999; Elliott and
Blobe, 2005; Harradine and Akhurst, 2006). Additionally, in vitro studies have shown
that some colon carcinoma cell lines no longer respond to the anti-growth effects of TGF-
8 , while other colon carcinoma cell lines remain sensitive. Sensitivity to the anti-growth
effects of TGF- B was found to be positively associated to the degree of differentiation of
the colon carcinoma cell lines (Hoosein et al., 1989; Hsu et al., 1994). Insensitivity to
anti-growth signals represents one of the hallmarks of cancer governing the
transformation of normal cells to malignant cancers (Hanahan and Weinberg, 2000).

As previously mentioned, altered TGF-B expression has been identified in AOM
induced colon cancer in animal models (Shao et al., 1999). When Guda and coworkers
(2001; 2003) examined the TGF-B signaling pathway in the AOM induced colon
carcinogenesis in the susceptible A/J mouse, they found significantly higher expression
of TGF-Bl protein and reduced expression of TGF-BRII (mRNA and protein) in colon
tumors compared to normal colonic mucosa fiom control animals. In an attempt to
clarify where the disruption in the TGF-B signaling pathway occurred the expression of
several components in the pathway were analyzed. No significant changes were revealed
in T GF-fil, TGF-flRI, Smad 3 or Smad 7 mRNA levels, however, the expression of c—myc
mRNA was significantly upregulated, while p15 was significantly downregulated (Guda

et al., 2001; 2003). p15 and c—myc are direct targets of the TGF-B. In normal cells TGF-

64

B would activate the transcription of p] 5 , a CDK inhibitor, while simultaneously
suppressing the transcription of c-myc, a proto-oncogene that promotes cell growth and
proliferation (Massagué et al., 2006). These two opposing actions are required to block
the progression of cells through the Gl phase of the cell cycle (Massagué et al., 2006).
The data in this study suggest that tumor cells in AOM injected A/J mice are resistance to
TGF-B growth inhibition is a consequence of reduced T GF -,BRII R expression leading to
the downregulation in p15 expression permitting the expression of c-myc which promotes
progression through the cell cycle and continued cell proliferation (Guda et al., 2001).
Although the mRNA levels of TGF -,61 were not altered in AOM induced colonic
adenomas in NJ mice, higher TGF-fil protein expression was reported. Since high TGF-
Bl protein levels are reported in different types of human cancer and TGF-Bl is secreted
in a latent inactive form, the authors wanted to determine the amount of active mature
TGF-Bl protein vs. latent (i.e. inactive bound TGF-Bl) in the tumors. In the tumors 50%
of the TGF-Bl protein was active compared to 80% in the normal colonic mucosa. The
higher amount of latent TGF-Bl protein was suspected to result from the lower plasmin
activity and higher PAI-l mRNA identified in the tumors. Plasmin is a known activator
of TGF 431, whilst PAI-l (plasminogen activator inhibitor) regulates the production of
plasmin from plasminogen and serves as a negative feedback regulator of the TGF-B
signaling pathway. To confirm that a disruption in TGF-Bl signaling was occurring the
expression of 21 previously identified TGF-B- specific target genes were analyzed. It
was determined that in AOM induced tumors expressing high amounts of TGF-Bl protein
the expression of all 21 target genes were downregulated by at least 1.5 fold compared to

the normal mucosa. These target genes were involved in cell matrix interaction (i.e.

65

IGFBP3, fibronectin), cell-to-cell interaction (i.e. Notch2, APC, CD44) and cytoplasmic
regulators and effectors (i.e. rhoB, rho GDP-dissociation inhibitor 1, P21-RAC2). The
authors concluded that a defect in the activation and turnover of the TGF-Bl protein, in
addition to the downregulation in TGF-BRII receptors contributes to the progression of
AOM induced colon carcinogenesis in Al] mice (Guda et al., 2001; 2003).

Raju and colleagues (2002) examined alterations in TGF-Bl and TGF-BZ mRNA
and protein expression in the colons of AOM injected F344 rats fed diets containing
different dietary lipids. Dietary lipids differentially regulate the expression of TGF-B in
vitro in some tissues (Biasi et al., 2007). The dietary intervention was initiated 10 weeks
afier the final AOM inj ection; thus these researchers were investigating the promotional
effects of different dietary lipids on preneoplastic lesions (i.e. ACF) and whether this
involved the TGF-B signaling pathway. The AOM induced tumors from all animals
regardless of diet, exhibited higher levels of TGF-Bl and TGF-BZ mRNA and protein
compared to the surounding normal colonic mucosa. In a minor subset of ACF, TGF-Bl
and TGF-BZ protein levels were also higher than the normal mucosa. The overexpression
of TGF-Bl and TGF-B2 mRNA and protein in AOM induced preneoplastic lesions and
tumors suggests that during the later stages of colon carcinogenesis an overexpression of
TGF-B may promote the progression of tumors to more advanced lesions (Raju et al.
2002)

1.3. Colorectal Cancer and Diet

Population-based studies have demonstrated large geographical variations in CRC
incidence across the world, with lower incidences in Africa and Asia, moderate

incidences in South America and high incidences in North America, Australia/New

66

Zealand and Western Europe (Parkin et a1, 2005; WCRF/AICR, 2007). In 1975
Armstrong and Doll reported a 60-fold variation in global CRC rates among countries,
whereas in 2003, Parkin and colleagues reported a 25-fold variation in CRC rates among
countries. Epidemiological studies examining dietary patterns and disease risk have
observed changes for risk in diseases predominantly found in higher-income,
industrialized nations, beginning to increase in less developed nations as they become
more urbanized or “Westernized”. These countries simultaneously assimilate the dietary
patterns and lifestyle practices of Western nations (Heavey et a1, 2004; WCRF/AICR,
2007). The best evidence indicative of these changes in disease risk is from migrant
studies. Migrant studies observing the changes in cancer rates in populations migrating
from low-risk areas to high-risk areas (or vice versa) and their off-spring, have
demonstrated a strong relationship between lifestyle and environmental factors, i.e.
dietary patterns, cultural and social practices, physical activity, obesity, socioeconomic
status, and cancer incidence (Heavey et a1, 2004). This relationship is found when
migrants move from rural to urban areas within their country of origin or from one
country to another. Researchers have found that as migrants’ dietary patterns change,
disease patterns begin to reflect those of their new region within one to two generations
(W illett, 2001; WCRF/AICR, 2007). “Disease rates that shifi with migration are those
with important environmental causes” (WCRF/AICR, 2007). Therefore migrant studies
validate the causal effect of lifestyle and environmental factors in cancers of most sites,
including CRC (W CRF/AICR, 2007). Although genetics accounts for 5-10% of
colorectal cancer cases, the more predominant modulators in colorectal cancer

development are lifestyle and environmental factors (Weitz et a1, 2005; Willett, 2001).

67

Theoretically, an estimated 80-90% of all cancers could be prevented by altering
environmental and/or lifestyle, suggesting cancer is a preventable disease (Doll, 1998;
WCRF/AICR 2007).

Evidence has been shown that diet is an important lifestyle factor that can be
modified and alter CRC risk. Since food components come in direct contact with the
colonic epithelial cells and colorectal carcinogenesis requires a number of molecular
alterations to develop, it is logical that food components can interfere, inhibit, or reverse
its development (J anne and Mayer, 2000). However, distinguishing which food
components or food groups specifically protect or promote the development of CRC and
the molecular mechanisms by which they exert these biological activities have yet to be
clearly comprehended. Until recently the suggestion of a positive association between
total fat intake and the risk of developing cancers of the breast, prostate, and colon was
commonly accepted, yet in the 2007 WCRF/SICR report reviewing the evidence on food,
nutrition, physical activity and the prevention of cancer, a consensus panel found limited
proof to suggest that high total fat intakes increase an individual’s risk of developing
colon, breast or prostate cancer.

When reviewing the evidence regarding CRC, the WCRF/AICR panel concluded
that there was convincing evidence to support the positive association between higher
intakes of red meat and processed meat, greater abdominal and total body fatness and
substantial consumption of alcoholic drinks, and the risk of developing CRC, whereas
physical activity was found to protect against its development. They concluded that
consuming foods containing dietary fiber, garlic, milk and calcium probably protects

against CRC. Studies examining legume consumption and CRC risk were not analyzed

68

by the panel because they were limited in number and/or they may not have met the
predetermined criteria required to be included in the analysis. However the panel did
report that evidence from a limited amount of studies suggested a protective effect
against stomach and prostate cancer with higher legume consumption (WCRF/AICR,
2007).

1.4. Colorectal Cancer and Legumes

The consumption of legumes including dry beans has been inversely associated
with the development of several chronic diseases such as cardiovascular disease (CVD),
type 2 diabetes mellitus, obesity and colon cancer (Geil and Anderson, 1994; Anderson
and Major, 2002; Guillon and Champ, 2002; Giovarmucci, 2003; 2007; Gunter and
Leitzmann, 2006). Many of the risk factors associated with obesity, type 2 diabetes,
CVD and colorectal cancer overlap and nutrition plays an important role in the etiology,
prevention, treatment, and management of these diseases.

1.4.1. Epidemiological Studies

Although limited, some epidemiological studies have investigated the relationship
between colorectal cancer and legume consumption. Similar to studies examining
legume consumption and risk of developing obesity, diabetes or CVD, the majority of
studies clump legumes in with other high fiber plant foods, rather than analyzing them
individually. Although it is probable that dietary fiber does protect against CRC, the
relationship between fiber and cancer has not been clear. This is due to many factors
including food surveys, differences in the definition of fiber, vast number of plant foods
having high fiber content, and the other components exerting biological activity found in

high fiber foods.

69

In 1981, Correa examined the role of vegetables in cancer risk and found a
significant negative correlation coefficient (-0.68, P<0.05) between dry bean intake and
colon cancer risk (Correa, 1981). The Seventh-day Adventists Health cohort prospective
study in California, examined the relationship between intakes of specific foods or food
components and colon cancer incidence. The Seventh-day Adventists consume a largely
plant-based diet, lower in saturated fat and higher in fiber compared to other US.
populations and their mortality rates from cancer, heart disease, and diabetes are lower
compared to non-Adventists living within the same community. An inverse association
between legume intake and colorectal cancer risk was found (>2x per week vs. <lx per
week, RR=0.53 [95% CI: 0.33-0.86]; p=0.03). Additionally, the authors observed a Z 3-
fold increase in colon cancer risk for individuals having a high BMI, consuming diets
high in red meat, but low in legumes. The authors suggested that the consumption of
legumes counteracts the colon cancer promoting effects induced by red meat (Singh and
Fraser, 1998). Steinmetz and Potter (1993) reported an inverse relationship between
legume intake and colon cancer risk in both men and women enrolled in a case-control
study. With the observed reduction being greater in women consuming the greatest
amount of legumes vs. those consuming none, approximately a 50% reduction (OR=O.53
[95% CI: 0.26-1.07]), whereas a 30% reduction was found for men (OR=0.74 [95% CI:
0.39-1.38]). When examining the association between fruit and vegetable intake and the
prevalence and incidence of tumors in the distal colon and rectum in the Nurses’ Health
Study, Michels and colleagues (2006) reported women who regularly consumed legumes
(>4 serving/week) had a lower incidence of colorectal adenomas (OR= 0.67 [95% CI:

0.51-0.90]; p=0.005). In support of the previous study, a dietary intervention study

70

called the Polyp Prevention Trial (PPT), researchers examined the association of fruit and
vegetable intakes and the recurrence of colorectal adenomas. They found a significant
inverse association between increased dry bean intake (the greatest increase in dry bean
intake vs. the lowest; OR=0.35 [95% CI: 0.18-0.69]; p=0.001) over a 4 yr period and
advanced adenoma recurrence (Lanza et al., 2006).

In a prospective cohort study, published by Lin et al. (2005), no significant
association between fi'uit, vegetable and fiber intake and colorectal cancer was found in
women. Yet the authors did report a significantly inverse relationship between legume
fiber and colorectal cancer risk, when comparing the highest intake relative to the lowest
(RR=0.60 [95% CI= 0.40-0.90], p=0.02).

1.4.2. Experimental Studies

Several researchers have investigated the antirnutagenic and antigenotoxicity
effect of dry beans. Using a modified version of the Ames test, Cardador-Martinez and
colleagues (2002 and 2006) reported that a methanolic phenolic extract of the seed coats
of raw common beans (Phaseolus vulgaris L., Flor de Mayo FM-38 variety) inhibited

50% of mutagenic activity induced by aflatoxin Bl (AF B1) in vitro. In addition the

authors found that the phenolic extract from beans possessed antioxidant activity and was

capable of scavenging free radicals; however this was not the mechanism responsible for

inhibiting the mutagenic activity of AF B1. Their data suggest that the phenolic extract
from dry beans inhibits the mutagenicity induced by AF Bl by reducing its bioavailability
and the amount of the active ultimate mutagen (8,9- APB1 epoxide) formed by P450
enzymes, either by competing with the AFB1 for P450 activity, directly inhibiting P450

activity or by binding the 8,9- AFB1 epoxide (Cardador-Martinez et al., 2002; 2006). In

71

another study examining black beans and antimutagen activity, Azevedo et al. (2003)
reported feeding diets containing 1%, 10% or 20% cooked black bean to mice injected
with cyclophosphamide (CP), a well-known indirect acting mutagen, reduced the DNA
damage.

Bawadi et al. (2005) investigated the anti-angiogenic activities of tannins
extracted fi'om black beans on several human cancer cell lines. The aqueous-acetone
(70% acetone) soluble components in raw black beans (including tannins) inhibited the
grth and angiogenesis of colon, breast, and prostate cancer cells in a dose-dependent
manner yet had no effect on normal human fibroblast lung cells (Bawadi et al., 2005).
Concentrations of condensed tannins (expressed as catechin equivalents) added ranged
from 2 to 24 uM. The concentrations of phenolic compounds used in this study, as well
as in many other cell culture studies, are unlikely to be found in the peripheral circulation
in viva. Human plasma concentrations of intact flavonoids have rarely exceeded 1 uM
after consuming normal dietary levels of polyphenols (Scalbert and Williamson, 2000).
However, there is the possibility of higher concentrations being found in the GI tract. For
example, higher levels (M to mM range) of flavonoids and simple phenolic compounds
would be found in the stomach and small intestine prior to absorption afier ingesting a
meal containing plant foods. The colon could also be exposed to higher levels (uM
range) of simple phenolic compounds; intact flavonoids; metabolites of flavonoids and
other larger phenolic compounds; and microflora metabolites of phenolic compounds, all
of which possess biological activity and possibly protect against colorectal cancer (Gee

and Johnson, 2001; Halliwell, 2007).

72

It is important to note that the aqueous-methanolic and aqueous-acetone extracts
from dry beans used in the previously mentioned studies were crude extracts. They most
likely contained other soluble compounds that exert biological activity besides phenolic
compounds, such as saponins and oligosaccharides.

Using animal models, Hughs et al. (1997) found that feeding pinto beans to
AOM-injected rats significantly reduced tumor incidence compared to a casein-based
control diet. Twenty-four percent of the rats fed the pinto bean diet had colonic tumors
versus 50% tumor incidence in the rats fed the casein based control diet. Recently,
Boateng and coworkers (2007) examined the effect of feeding dry beans: black-eyed
peas, pinto beans, or soybeans, on AOM-induced colonic ACF in rats. Animals fed the
bean diets weighed significantly more, and had significantly lower cecal pH than the
casein control fed animals, yet all animals consumed similar amounts of food and had
similar cecal weights. Compared to the casein control fed rats, the number of colonic
ACF were reduced by 77%, 64% and 56% in the rats fed the black-eyed peas, pinto
beans, and soybeans, respectively. The authors also found the dry bean diets significantly
induced hepatic glutathione—S-transferase activity (GST). GST is a phase H detoxifying
enzyme and its increased activity may help detoxify fi'ee radicals produced by the
metabolism of carcinogens and other compounds (Boateng et al., 2007).

Our laboratory has performed several in vivo studies investigating the
chemopreventive effect of feeding bean-based diets on AOM induced colon
carcinogenesis in a rat model. Hangen'and Bennink (2002) determined that feeding navy
and black beans to AOM injected rats reduced colon tumor (adenoma and

adenocarcinoma) incidence by 59% and 54%, respectively, compared to rats fed a control

73

casein-based diet. Similarly, when Rondini (2006) fed AOM injected rats defatted
soybean flour or black beans, colon tumor incidence was reduced by approximately 60%
with both bean based diets. Based on cDNA microarray technology, feeding beans
inhibits AOM induced colon tumorigenesis by modulating the expression of genes
involved in regulating colon crypt cytokinetics and in reducing inflammation (Rondini,

2006).

D. Rationale

The consumption of dry common beans is inversely associated with the
development of chronic diseases, including CVD (Finley et al., 2007; Winham et al.,
2007), Type 2 diabetes (Villegas et al., 2008), obesity (van Dam and Seidell, 2007) and
cancers of the breast (Thompson et al., 2008; 2009) and colon (Hughes and Ganthavom,
1997; Hangen and Bennink, 2002). Dry beans are a rich source of protein, dietary fiber,
vitamins and minerals, as well as many non-nutrients such as phenolic compounds.

Phenolic compounds have been suggested to contribute to the protective effects
observed when fi'uits, vegetables and grains, including dry beans, are consumed. They
have been shown to act as antioxidants (Croft, 1998), anti-inflammatory agents (Rahman
et al., 2006), and anti-cancer agents (Bode and Dong, 2004), yet the exact mechansism(s)
of action has yet to be elucidated. Increased interest in the health benefits of phenolic
compounds makes identifying and quantifying them in all plant foods of great
importance.

Investigators have identified different types of phenolic compounds in grain

legumes, including flavonol glycosides, hydroxybenzoic acids, hydroxycinnamic acids,

74

anthocyanins and condensed tannins (proanthocyanidins) (Krygier et al., 1982; Sosulski
and Dabrowski, 1984; Beninger et al., 1999; 2003; 2005; Madhujith et al., 2004;
Aparicio-Femandez et al., 2005; Hu et al., 2006; Lin and Lai, 2006; Ranilla et al., 2007).
The majority of these studies analyzed raw grain legumes, however legumes must be
exposed to some sort of heat treatment to be safe to consume. Heat processing induces
physical and chemical compositional changes in legumes, altering their nutrient and
phenolic content. Thus from a dietary standpoint it is important to quantify the phenolic
compounds in heat treated grain legumes i.e. dry common beans, in the form they are
consumed by humans.

Epidemiologic studies suggest that bean consumption is inversely related to colon
cancer incidence (Correa, 1981) and animal studies have found that feeding pinto beans,
black beans, and navy beans inhibit chemically induced colon cancer by more than 50%
(Hughes and Ganthavom, 1997; Hangen and Bennink, 2002; Rondini, 2006). Often the
anti-cancer activity of dry beans is attributed to the high dietary fiber content of beans.
Although an inverse relationship between dietary fiber and CRC risk is probable the
evidence supporting the relastionship has been inconsistent (W CRF IAICR, 2007). Thus
narrowing which specific components in dry beans are responsible for the protection
against the development of CRC are of high priority. Of equal importance is the
molecular identification of the molecular mechanism by which dry beans inhibit cancer.
One approach to narrowing the search for the anti-cancer components of legumes is
partitioning beans into different fractions and determining the anti-cancer activity for the
various fractions. Bennink and Barrett (2004) used this approach with soybeans (an oil

legume). They fed heat-treated, full-fat soybean flour, soybean flour minus lipids,

75

 

defatted soybean flour minus aqueous-ethanol components and a control diet to
determine which fractions protected against development of AOM-induced colon cancer
in rats. Full-fat soybean flour and defatted soybean flour diets significantly reduced
tumor incidence by 45% (P < 0.05) compared to rats fed a casein control diet. Based on
this experiment and other unpublished studies it was determined that soybean fiber,
sterols, fatty acids, protein and isoflavones could not be providing the anti-cancer action
noted with defatted soy flour. However, removing the aqueous-alcohol soluble
components completely eliminated any anti-cancer action (Bennink and Barrett, 2004).
The aqueous-alcohol fraction would contain not just isoflavones, but other phenolic
compounds such as hydroxycinnamic and hydroxybenzoic acids i.e. syringic acid, ferulic
acid, sinapic acid and p-coumaric acid (Seo and Morr, 1984), as well as saponins and
oligosaccharides, which are reportedly bioactive. Subsequent studies have shown
soybean isoflavones do not inhibit colon cancer, which suggests further that other
compounds in legumes, which are soluble in an aqueous-alcohol solvent, may be
responsible for the anti-cancer activity observed in colon cancer prevention studies in
rats.

Consistent with the protection observed in the previous animal study (Bennink
and Barrett, 2004), in vitro studies have shown crude aqueous-alcohol and aqueous-
acetone extracts from dry beans (Phaseolus vulgaris L.) possess anti-mutagenic, anti-
genotoxic, and anti-angiogneic properties (Azevedo et al., 2003; Cardador-Martinez et
al., 2002; 2006; Bawadi et al.,2005). These crude extracts would be concentrated in

phenolic compounds as well as saponins and oligosaccharides.

76

Besides colorectal cancer, dry bean consumption has been inversely associated
with risk of developing obesity and type 2 diabetes mellitus (Geil and Anderson, 1994;
Maskarinec et al., 2000; Guillon and Champ, 2002; Venn and Mann, 2004; Villegas et
al., 2008). Many of these biological perturbations resulting from obesity and diabetes,
have been positively associated with colon cancer (Gunter and Leitzmann, 2006).
However, few studies have evaluated the effect of chemically induced colon cancer in
obese animal models, of those that have, the authors reported a higher numbers of
preneoplastic lesions and tumors in the colons of obese animals compared to lean (Weber
et al., 2000; Hirose et al., 2004; Ealey et al., 2008). To date an obese animal model has
not been utilized to evaluate the chemopreventive affect of feeding dry beans on
chemically induced colon cancer.

Colon carcinogenesis is a multi-step process requiring a number of molecular
alterations to fully manifest, thus it is logical that dietary-induced changes in metabolized
and undigested food components might alter colon cancer development. The mechanism
by which dietary components, including dry beans, inhibit chemically induced colon
cancer in animals has not been fully elucidated. Colon tumorigenesis is associated with
perturbations in colon crypt homeostasis, including changes in cell proliferation,
differentiation and apoptosis (Fearon and Vogelstein, 1990; Chang et al., 1997).
Consistent with the idea that diet can modulate gene expression, Rondini (2006) reported
the downregulation in the expression of genes involved with colon crypt cytokinetics and
in anti-inflammation within the colonic mucosa of AOM-inj ected rats fed black beans

and soybean flour diets.

77

The objectives of my dissertation are as follows:
1. To extract, separate, identify and quantify the principal phenolic compounds in
cooked dry common beans (Phaseolus vulgaris L.) using hi gh-perfonnance liquid
chromatography (HPLC)-electrochemical detection (ECD).
2. To extract a sufficient amount of phenolic compounds fiom cooked beans
using an aqueous-alcohol solvent to allow long-term colon cancer studies to be
conducted with rodents, and to identify and quantify the principal phenolic
compounds in the bean extract (aqueous-alcohol soluble components) and in the
residue (insoluble particulate matter remaining after extraction of alcohol-aqueous
soluble compounds).
3. To evaluate whether the obese, mildly diabetic mouse (ob/ob mouse (B6.V-
LepOb /J)) is an appropiate animal model for studying azoxymethane-induced
colon cancer.
4. To feed the cooked common beans and its fractions to obese, mildly diabetic
mice (ob/ob mice (B6.V-Lep0b /J)) with chemically induced colon cancer to
narrow the search for which components in dry beans are responsible for the
observed reduction in the development of chemically-induced colon cancer.
5. To determine if feeding bean diets and its fractions to ob/ob mice reducing the
development of colon tumorigenesis and if this is associated with normalizing

gene expression in pathways intricately involved with colon crypt cytokinetics.

78

CHAPTER III. PHENOLIC COMPOUNDS IDENTIFIED AND QUANTIFIED IN

COOKED DRY COMMON BEANS (PHASEOLUS VULGARIS L.)

A. ABSTRACT

The consumption of legumes such as dry common beans has been inversely
associated with development of several chronic diseases (Thompson et al., 2008; 2009).
Dry beans contain a variety of phytochemicals that may provide health benefits. Phenolic
compounds in many commonly consumed fi'uits and vegetables, have been identified and
quantified; however, there is little data regarding the phenolic compounds in cooked dry
common beans (Phaseolus vulgaris L.).

The primary objective of this study was to extract, separate, identify, and quantify
the principal phenolic compounds in cm dry common beans using high-performance
liquid chromatography (HPLC)-electrochemical detection (ECD). A secondary objective
was to: a) extract a sufficient amount of phenolic compounds fi'om cooked navy beans to
allow long-term colon cancer studies to be conducted with rodents; and b) to identify and
quantify the principal phenolic compounds in the navy bean extract (concentrated in
phenolic compounds) and in the residue (insoluble particulate matter remaining after
extraction of alcohol-aqueous soluble compounds).

Extracts of four cooked dry common beans (black, pinto, red, and navy beans) as
well as in the navy bean fractions (navy bean residue and navy bean extract mix) were
analyzed for phenolic compounds with and without acid and alkaline hydrolysis.
Hydroxybenzoic acids and hydroxycinnamic acids were identified in the four cooked dry

common beans. The hydroxybenzoic acids found were p-hydroxybenzoic acid, vanillic

79

acid, and syringic acid and total concentrations ranged from 3 to 12 mg per 100 g bean
flour. Protocatechuic acid was only identified in black, pinto and red beans. The
hydroxycinnamic acids identified were p-coumaric acid, caffeic acid, ferulic acid, and
sinapic acid in total concentrations ranging from 11 to 36 mg per 100 g bean flour. In
general, acid hydrolysis liberated greater quantities of hydroxybenzoic acids while
alkaline hydrolysis liberated greater quantities of hydroxycinnamic acids. Only black,
pinto, and red beans contained the flavan-301 (+)-catechin (0.3 to 3 mg per 100 g bean
flour) and the flavonols quercetin and kaempferol (7 to 67 mg per 100 g bean flour). The
large scale aqueous-ethanol extraction was more efficient in extracting hydroxycinnamic
acid conjugates than hydroxybenzoic acid conjugates from cooked navy beans.

In conclusion, this is one of few studies to identify and quantify phenolic
compounds in cooked dry common beans. It is important to know the types and amounts
of phenolic compounds in cooked beans to further investigate their potential health

benefits.

80

B. INTRODUCTION

The consumption of dry beans has been inversely associated with development of
chronic diseases such as cardiovascular disease (CVD), type 2 diabetes mellitus, obesity,
colon cancer (Hughes and Ganthavom, 1997; Hangen and Bennink, 2002), and breast
cancer (Thompson et al., 2008 and 2009). Earlier studies exploring the disease
preventing components in legumes focused on dietary fiber. However, legumes,
including dry beans, contain phenolic compounds that may confer some of these
observed health benefits (Geil and Anderson, 1994; Messina, 1999).

Phenolic compounds belong to an array of structurally diverse secondary plant
metabolites that are distributed throughout the plant kingdom (Stalikas, 2007). Phenolic
compounds are not uniformly distributed in the plant tissues and tissue concentrations can
vary depending on plant species, variety, time of harvest, maturation, germination, and
storage conditions (i.e. time and temperature) (Tsao, 2004). There are at least 8,000
naturally occurring phenolic compounds, all of which possess a common aromatic ring
bearing one or more hydroxyl substituents. This shared structure is referred to as a
phenol (Harbome, 1998, Robbins, 2003; Balasundram, 2006; Stalikas, 2007).

Classes of phenolic compounds previously identified in grain legumes include
hydroxybenzoic acids, hydroxycinnamic acids, flavan-3ols, flavonols, anthocyanins, and
condensed tannins (Krygier et al., 1982; Sosulski and Dabrowski, 1984; Beninger et al.,
1999; 2003; 2005; Madhujith et al., 2004; Aparicio-Femandez etal., 2005; Hu et al.,
2006; Lin and Lai, 2006; Ranilla et al., 2007). However, the cited publications analyzed
raw, mature dry beans, but dry beans must be heat-treated before the bean is safe to

consume.

81

Heat processing induces physical and chemical compositional changes in
legumes, altering their nutrient and phenolic content. The changes are dependent on the
legume type and processing conditions. Legumes are heat-treated before consumption to
inactivate antinutrients such as lectins, protease inhibitors and amylase inhibitors.
Processing methods include hydration (soaking), germination or fermentation, and some
form of heat treatment (i.e. boiling, pressure cooking, canning, roasting, micronization, or
autoclaving) (Champ, 2002; Ibrahim et al., 2002; Shimelis and Rakshit, 2007). Very few
studies have investigated the phenolic compounds in legumes afler processing. Of those
that have, the majority only measured total phenolic content and report an overall
decrease in the total phenolic content in grain legumes after cooking (Towo et al., 2003;
Rocha-Guzman et al., 2007; Granito et al., 2008; Xu and Chang, 2008).

In a published review on the subject of hydroxycinnamic acids, Clifford (1999)
observed that assessing the dietary burden of hydroxycinnamic acids was not possible.
He stated, “With the exception of some beverages, the lack of data for the composition of
commodities as consumed, i.e. the edible portion after processing, cooking, baking, etc,
effectively make such an assessment impossible”. With the ever increasing public
awareness of phenolic compounds as health promoting non-nutrients, especially as
antioxidants, it is critical to identify and quantify phenolic compounds in plant foods that
have been shown to impart health benefits when consumed.

One of the strongest health benefits derived from eating beans is the prevention of
colon cancer. The cancer inhibiting component(s) of beans is(are) unknown. Since the
phenolic compounds in beans could be involved in cancer inhibition, the primary

objective of this study was to extract, separate, identify and quantify the principal

82

phenolic compounds in cooked dry common beans (Phaseolus vulgaris L.) using hi gh-
performance liquid chromatography (HPLC)-electrochemical detection (ECD). A
secondary obj ective was to: a) extract a sufficient amount of phenolic compounds from
cooked navy beans to allow long-term colon cancer studies to be conducted with rodents;
and b) to identify and quantify the principal phenolic compounds in the navy bean extract
(concentrated in phenolic compounds) and in the residue (insoluble particulate matter

remaining after extraction of alcohol-aqueous soluble compounds).

C. MATERIALS AND METHODS
Chemicals

Methanol, 2N Folin-Ciocalteu phenol reagent, ellagic acid, gallic acid, isovanillic
acid, salicylic acid, gentisic acid, Chlorogenic acid, (-)-epicatechin, luteolin, apigenin,
esculetin, protocatechuic acid, p-hydroxybenzoic acid, vanillic acid, isovanillic acid,
syringic acid, Chlorogenic acid, caffeic acid, p-coumaric acid, ferulic acid, sinapic acid, 0-
coumaric acid, (+)-catechin, quercetin, and kaempferol were purchased from Sigma
Chemicals (St. Louis, MO). Acetonitrile — HPLC grade, acetic acid and hydrochloric
acid were obtained from EMD Chemicals Inc. (Gibbstown, NJ). Sodium monobasic
phosphate-monohydrate, sodium dibasic phosphate-anhydrous, phosphoric acid, sodium
hydroxide were from J .T. Baker (Phillipsburg, NJ).

Navy, black, pinto and red beans were kindly provided by the Michigan Bean
Commission and were stored at 4°C in dry storage prior to cooking.

Preparation of Whole Bean Flours

83

Five kg of raw, dry, common beans — black, pinto, red and navy beans —- were
soaked overnight at 4°C in sufficient distilled water to keep the beans covered with water.
The soaked beans and soak water were added to large steam kettles (Model N 30 SP,
Groen M.F.G. Co., Chicago, IL) and cooked for one hour at 98°C. Immediately after
cooking, the beans and cooking water were dried in stainless steel trays in a convection
oven at 75°C :1: 5C overnight (Machine # K 12395, Proctor and Schwartz, Inc.,
Philadelphia, PA). Once dried, the beans were ground to pass through a screen with 1.6
mm holes using a Fitzmill (Model D Comminuting Machine, The W.J. Fitzpatrick Co.,
Chicago, IL).

Preparation of Navy Beans and Fractions

Navy Beans (675 kg raw dry bean weight) were soaked and cooked as described
above. Approximately 65% of the cooked navy beans were immediately ground with the
Fitzmill fitted with a screen with 5 mm holes. Ethanol (95%) was added to the cooked
ground navy beans and was mixed periodically for 18 — 24 hr to extract phenolic
compounds. The final concentration of ethanol was approximately 60%. The liquid
fraction (navy bean extract) was separated from the insoluble particulate matter (navy
bean residue) by screen filters. The residue was placed into burlap bags and pressed at 20
kg/cm2 (Bar N.A., Inc., Seymour, IL) to remove additional liquid in the wet residue. The
liquid collected after pressing was combined with the filtrate. The liquid extract was left
undisturbed for 12 — 24 hr to allow particulate matter passing through the screen filter to
settle out. The liquid fraction was decanted and the solid sediment was combined with

the residue. The navy bean residue (NBR) was dried and ground as described above.

84

The aqueous-alcohol liquid navy bean extract (approximately 3,200 L) was
concentrated (approximately 130 L) using a Marriot Walker Falling Film Evaporator
(Marriot Walker, Birmingham, MI). The concentrated liquid extract contained 27 g of
dry material per 100 ml of liquid concentrate. Since the dry extracted material was very
hygroscopic, the concentrated liquid bean extract was mixed with cellulose, casein, and
cornstarch (27%:15%:19%:39%, wt:wt:wt:wt, on a dry basis, respectively) to facilitate
drying and storage stability. The wet mix was oven dried overnight at 7 5C :1: 5C. The
final dried product (navy bean extract mix, NBEM) was then ground using a Fitzmill
equipped with a screen with 1.6 mm holes.

For each navy bean fraction — whole beans, bean extract mix (NBEM) and navy
bean residue (NBR) —- a portion representing the entire ground fraction was immediately
removed and stored at 4°C for subsequent analyses.

Extraction of Phenolic Compounds

Two g of each flour were extracted with 30 ml of 50% methanol containing 1%
acetic acid (volzvol), sonicated for 30 minutes, centrifuged (3,000 x g for 10 min) and the
supernatant was decanted. The sediment remaining after centrifirgation was extracted
three more times using 70% and 90% methanol containing 1% acetic acid followed by
100% methanol. Supematants from each cycle of the extraction process were combined.
An additional 35 ml of 100% methanol was added to the supematants to obtain a final
concentration of approximately 80% methanol. The combined extracts were stored at
4°C overnight to precipitate macromolecules and then centrifuged at 3,000 x g for 20

min. The supematants were evaporated to dryness by rotary evaporation, using vacuum

85

and low heat (5 45 °C). The residue after evaporation was dissolved in 15 ml of 80%
methanol. Two extractions were performed on different days for each bean sample.
Total Phenolic Content Analysis

Total phenolic content (TPC) was determined using a modified version of the
Folin-Ciocalteu method according to Singleton and Rossi (1965) and similar to Mosca et
al. (2000). Briefly, 200 u] of bean extract and 50 ul of Folin-Ciocalteu reagent were
combined in a test tube and vortexed. After 3 to 8 minutes, 100 pl of 35% sodium
carbonate (wtzvol) was added and tubes were vortexed. Samples were brought to a final
volume of 2.25 ml with distilled water (2.15 ml) and vortexed again. Samples were
incubated in the dark at room temperature for 90 minutes to allow the color reaction to
come to completion. Absorption was determined at 725 nm on a spectrophotometer.
Bean extracts were diluted with 80% methanol as necessary to fall within the standard
curve. Each sample was measured in triplicate on two different days. Calibration curves
were prepared daily by diluting the (+)-catechin stock solution (0.503 mg/ml, diluted in
80% methanol) to a final concentration range of 2 - 34 M (0.67-10.07 ug/ml). Total
phenolic content (TPC) was expressed as mg of (+)-catechin equivalents ((+)-CE) per
100 g bean flour or bean fraction.
Hydrolysis of Conjugated Phenolic Compounds

Three aliquots (1 m1) of each bean extraction were evaporated under nitrogen gas
at 50 °C. Before evaporation, 200 pl of the internal standard o-coumaric acid (10.2 pg)
was added to each aliquot. One ml of distilled water was added to one of the dried
aliquots and served as the control/non-hydrolyzed sample. Then the sample was acidified

to a pH of 2.0 with 5 N HCL.

86

To identify and quantify the principal phenolic compounds present in the bean
extractions, the second and third dried aliquots were hydrolyzed using alkali or acid to
convert conjugated phenolic compounds to their free form (parent or aglycone form).
Alkali hydrolysis cleaves ester bonds, whereas acid hydrolysis cleaves glycosidic bonds.

Alkali hydrolysis was performed as follows (Cai et al., 2003). One ml of 5N
NaOH was added to the dried extracts and vortexed. The mixtures were continually
flushed with nitrogen gas, heated for 4 hours at 50 °C and vortexed periodically. After
hydrolysis, the alkali treated bean extracts were acidified to a pH of 2.0 with 5 N HCL.

Acid hydrolysis was performed as follows (Harbome, 1998). Five N HCL (0.800
ml ) and 1.2 ml of distilled water were added to the dried extracts, (final concentration of
HCL was 2N). Tubes were flushed with nitrogen gas, capped and placed in boiling water
for 30 min. After acid hydrolysis, samples were placed in ice water until cool.

Sample Clean-up and Concentration

Solid-phase extraction (SPE) was used to purify and concentrate the free and
hydrolyzed phenolic compounds. The control/non-hydrolyzed, alkali and acid
hydrolyzates were applied to preconditioned (washed with 15 ml of 80% methanol and
15 ml of distilled water) SPE columns (Alltech Extract-Clean C18 column, (500 mg bed
volume), Alltech, Deerfield, IL). After loading the extract onto the SPE column, the
column was washed with 15 ml of distilled water to remove the unadsorbed, unwanted
water soluble compounds such as sugars. To elute the fi'ee and hydrolyzed phenolic
compounds, 5 ml of 80% methanol was applied to the columns 1 ml at a time. Pressure

was applied to the top of the columns to obtain a drop-wise flow rate. The eluates were

87

collected and evaporated to dryness under nitrogen at 50°C. After drying they were re-
dissolved in 1.5 ml of 80% methanol.

The hydrolysis and purification processes were performed on duplicate bean
aliquots on different days. Duplicate control/non-hydrolyzed bean aliquots were passed
through SPE columns on different days.

Preparation of Standard Solutions and Calibration Curves

Preliminary experiments revealed that the following phenolics were not present in
the four bean market classes of beans used in this research: ellagic acid, gallic acid,
salicylic acid, gentisic acid, (-)-epicatechin, luteolin, apigenin, and esculetin. Therefore,
these phenolics were not included in standards. Stock standard solutions of
protocatechuic acid, p-hydroxybenzoic acid, vanillic acid, isovanillic acid, syringic acid,
caffeic acid, p-coumaric acid, ferulic acid, sinapic acid, o—coumaric acid (internal
standard), and (+)—catechin were prepared by dissolving 20 mg in 20 ml of 80% methanol
[1 mg/ml]. Chlorogenic acid, quercetin, and kaempferol stock solutions were prepared
by dissolving 11 mg in 10 ml [1.1 mg/ml], 40 mg in 30 ml [1.3 mg/ml] and 5 mg in 10 ml
[0.5 mg/ml] -in 80% methanol, respectively. Each standard was injected three times to
establish retention times and oxidation profile. Then phenolic compounds that did not
have similar retention times were combined to form four standard mixes. The mixed
standards were diluted from 1:5 to 122500 with 80% methanol to establish calibration
curves. Standard mix 1 contained: protocatechuic acid, (+)-catechin, vanillic acid, and
ferulic acid; Standard mix 2 contained: p-hydroxybenzoic acid, Chlorogenic acid,

syringic acid, and sinapic acid; Standard mix 3 contained: isovanillic acid, p-coumaric

88

acid, and kaempferol; Standard mix 4 contained: caffeic acid, quercetin, and o-coumaric
acid (internal standard).

Estimation of Phenolic Compound Loss During Sample Clean-up and Hydrolysis
To estimate loss of phenolic compounds due to sample preparation, a 1:10
dilution of the standard mixes were subjected to hydrolysis with alkali and with acid and

SPE clean-up as described for the bean extracts.
High-Performance Liquid Chromatography-Electrochemical Detection

Separation and quantification of the phenolic compounds was achieved by using
high-performance liquid chromatography (HPLC) equipped with an electrochemical
detector (ECD). The HPLC-ECD equipment included a Dionex GSSO gradient pump
(Dionex, Corp., Sunnydale, CA), a Spectra System autosampler injector (ASIOOO,
Thermo Separation Products) and an 8-channel 5600A ESA Coularry® Electrochemical
Detector (ESA Inc., Chehnsford, MA). The detector channels were set at —200 mV and
from 200 to 800 mV, in increments of 100 mV. As a compound passes through the
coulometric array detector cells, it is oxidized in a stepwise fashion until complete
electrochemical conversion occurs resulting in a peak at a particular set current (Aaby et
al., 2004). If the compound has more than one oxidizable moiety, than another peak at
the current at which that moiety is oxidized will also be displayed. The ESA Coularray®
software was used to analyze the chromatograrns.

Separation was performed using a reversed phase Luna C18 Column, (250 x
4.60mm, Sum particle size, Phenonenex, Torrance, CA). A guard column packed with
C18 silica (20 um) was placed before the separation column. Column temperature was

maintained at 37°C. The mobile phases were as follows: solvent A - 2% Acetonitrile in

89

30mM phosphate buffer (pH 2.3); solvent B - 70% Acetonitrile in 30mM phosphate
buffer (pH 2.3); and solvent C - 90% Acetonitrile and 10% water. The gradient applied
was: 0-30 min, 0%-26% B; 30-65 min, 26%-65% B; 65-70 min, O%-100% C; 70-75 min,
0%-100% A. The column was equilibrated with 100% A for 15 min before the next
injection was made. The flow rate was lml/min and the injection volume was 20 pl.

Each reagent used to make up the mobile phases was measured individually using
graduated cylinders, before being combined and mixed. The water used to prepare
mobile phases was reverse osmosis water passed through a Millipore Milli-Q purification
system (Millipore Corp., New Bedford, MA) and then “polished” by passing through a
Millipore C18 SPE column (Millipore Corp., New Bedford, MA). Mobile phases A and
B were filtered through a Millipore nylon membrane filter (47 m) using vacuum
(Millipore Corp., New Bedford, MA). The mobile phases were degassed by sonication
for 5 min. All standards and bean extracts were filtered through a 0.45um nylon filter
(Alltech, Deerfield, IL) prior to HPLC-ECD analysis.
Quantification of Phenolic Compounds

Amounts of phenolic compound standards injected into the HPLC were plotted
versus peak height and linear regression equations (Y = aX + b) were computed to
establish standard curves. The regression equations were then used to quantify the
phenolic compounds in bean extracts and in the standard mixes that had been treated to
estimate losses during sample clean-up and hydrolysis. Phenolic compounds in bean
samples were identified by retentions times, oxidation profile, and co-elution of spiked
standards. All data were corrected for losses that occurred during sample clean-up and

hydrolysis steps and are expressed as mg per 100 g of bean flour or bean fraction.

90

D. RESULTS
Separation, Identification, and Quantification of Phenolic Compounds in Dry
Common Beans by HPLC-ECD

Chromatograrns showing the four phenolic acid standard mixtures are displayed
in Figures 6—9. Varying concentrations of these 4 standard mixes were injected into the
HPLC to generate the standard curves used to quantify the various phenolic compounds
in beans.

The phenolic compounds in plant materials are present conjugated to a variety of
moieties at one or more sites on the aromatic rings. Thus, several compounds could be
the source of one aglycone. To make the identification and quantification of plant
phenolics even more difficult, only the most common phenolic glycosides or phenolic
esters are available commercially for preparing standards. Therefore, to simplify
identification of the various phenolic compounds, researchers frequently hydrolyze plant
extracts so that only phenolic aglycones need to be identified. Representative
chromatograms showing the effect of alkaline and acid hydrolysis on cooked dry
common beans and navy bean fiactions are shown in Figures 10a — 150 in Appendix A.
Table 3, 5, and 7 in Appendix B list the concentrations of unconjugated phenolic
compounds; unconjugated phenolic compounds plus phenolic compounds liberated by
alkaline hydrolysis; and unconj ugated phenolic compounds plus phenolic compounds
liberated by acid hydrolysis, respectively, identified in cooked dry common beans.
Tables 4, 6, and 8 in Appendix B list the concentrations of unconjugated phenolic
compounds; unconjugated phenolic compounds plus phenolic compounds liberated by

alkaline hydrolysis; and unconjugated phenolic compounds plus phenolic compounds

91

Eon oar—eh .an ufi::a>.5__3uaUA+v .Eua omsaoouaueueam "n £2 Enenfiw he iguana—.330 .w 353

98

$3285 08:. 5:551

98

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ea 81

 

$8 83.

 

$8 82

 

”>5 8:

 

98 82.

 

$5 83

 

98 8:

 

Iowa”! 1 Ron 9.8-
.1 W.
1.2-1: I 2 -1», QB

_ (a m.

J A m m

a m 9 W

W. I. o m

I. I. .

W m. m. m.

o W o

I I V

P P m.
p

98 8a

 

 

 

 

.2.

Ga

dm

6*

(vrt) esuodseu

92

Eva 8935 .23 Bantam .29: 3:03.330 .23 umeuaeA—Wecgflum "N 52 FEE—3m he 82338050 .5 9.5»?—

chSEEV oEF cozcoaom

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l 9.8 . 9.8 l a? . 0.8 9.8! .
$8 82 o
l _L SE 83.
LL. - $8 82.2
L $8 8:
L. - .1 98 28
ll 1 _ a.
LL $8 8:. m
l L o ,L ,i ll! Fe 28 a m
S S L w
m. m .._.H . ,.
.m. a. .m x
W. W. m. M ,8
D. D. WV mu
I. L
p m
m.
V .8

 

IO

 

 

 

L

93

Bap—9:05— .23 3.35.503 .33 25.56% "m NE FEE—Sm no Eanueueaouao .m Baum.”—

AmoEEEV 05? 52:23“

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

prov opemnog-d

 

 

 

 

 

98 9.8 99. l 9.8 9.8 .
98 8.2
L $8 83.
LL .. 1 9a 88.
L. LL...) L E...
-i L ,,.LL.L.. J]. LL 95 83.
1 L I- L], L. -- $8 88.
l . L L ‘ LLL L L, 98 88
ll 1'! IJ: .
L L , $8 83
W...
m
d
m... I
m. m
A
m.
.anuJ
v
m.
p

 

 

on

.0?

dm

(vrl) esuodseu

94

5322.0 .Auuauafim 353.5 Eon 8.8830...» .22" £3.30 3. 52 FEE—Sm he Eauweuafieuao .a 95»:—

meSEE oEF cozcouom

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I owe l 9.8 - 9E. obm .. 9.8
98 83:
l ll- ll lLL $8 83.
L: L l L l . 95 83.2
Ll _ LL. ll... Lulllll l: l 98 83.
. lll ..L_ l L
L L L- 9a 83.:
L1l- lll ll l) _. . ll l l l LLLLLL .. l.l Ll ll l l -.
L L LL 98 83.
ill I: ...;..L. 2.. w.LLLL lllllll l ll l l l
L L LL 98 83.8
L, l l l l l l l L I lll
L L 98 83.
L... 3
L m L.
L m L Mm. .ov
L L o
m. L w
o m
m .8
m.
0
W 28
w.

 

 

 

(vrl) asuodsaa

95

liberated by acid hydrolysis, respectively, identified in the navy bean fractions, NBR and
NBEM.

The hydroxybenzoic acids identified in extracts from the cooked dry common
beans were: protocatechuic acid, p-hydroxybenzoic acid, vanillic acid, and syringic acid.
The relative amounts of these acids present as aglycones and after hydrolyses are shown
in Figures l6a-d. Protocatechuic acid was not found in navy beans, but it was the major
hydroxybenzoic acid in black, pinto and red beans. Isovanillic acid was not detected in
any of the bean extracts.

Four hydroxycinnamic acids were identified in the cooked dry common beans —
caffeic acid, p-coumaric acid, ferulic acid, and sinapic acid — and the relative amounts
that were present as unconjugated aglycones and the amounts liberated by hydrolysis are
shown in Figures l7a—d. Ferulic acid was the major hydroxycinnamic acid found in each
of the whole bean extracts. Chlorogenic acid, a quinic acid ester of caffeic acid, was not
identified in any of the bean extracts.

The flavan-3ol — (+)-catechin — and the flavonols — quercetin and kaempferol
— were not present in navy bean, but were identified in black, pinto and red beans
(Figures 18 - 19b).

Figures 20 a—c and 21 a—d show the hydroxybenzoic acids and
hydroxycinnamic acids, respectively, identified in the navy bean fractions, NBEM and
NBR. The same phenolic acids identified in the whole, cooked navy beans were also
found in its fractions. The NBEM (concentrated in phenolic compounds) contained 1.2
to 3.6 fold more hydroxybenzoic acids than the NBR fi'action, and 3.2 to 6.4 fold more

hydroxycinnamic acids.

96

 

LLL L LL?

 

 

 

 

 

 

 

 

 

 

 

 

i
l Protocatechuic Acid
i l
l .. 8 flLiberated by Acid 1
: g 7 L 1” L Hydrolysis .
. I: 6 l l
I a I Liberated by i

I3 5 Alkaline Hydrolysisl
l a 4 i
l 8 3 lUnconjugated
l '2 2
i a
i E 1
l 0

Bean Flour
a.
p- Hydroxybenzoic Acid

'5 3'5 DLiberated by Acid

E9 3 Hydrolysis
‘ 5 2.5 ILiberated by
l (3 2 Alkaline Hydrolysis
. u .
l Unconjugated
l ‘3- 1.5

s 1

a

a 0.5
‘ BB PB RB NB

Bean Flour

'LLLLL L LL _ LL LLL L
b.

Figure l6a—d. Hydroxybenzoic acids in Bean Flours*
*BB, Black beans; PB, Pinto beans; RB, Red beans; NB, Navy bean.

97

 

Vanillic Acid

 

 

 

 

 

—_'_i

 

 

 

 

 

 

 

 

 

5 liberated by Acid 2
g Hydrolysis I
I u I
‘1 g lLiberated by . I
I (3 Alkaline Hydrolysus I
a
g raiUnconjugated I
a I
I e I
‘ BB PB RB NB I
Bean Flour I
c. , L LLL L LLL L L LLLL L LLLLLLL LL LLLL
Syringic Acid I
g 4 , __ ; Liberated by Acid I
E 3.5 I—l‘ ‘ —— Hydrolysis I
§ 3 —] I ‘ W lLiberated by I
In 2'5 ' i I 7 7 ‘ Alkaline Hydrolysis I
a 2 _ A - i I
§ 15 I 7 v~ V‘m —— a 7% lUnconjugated I
. 1 L L ‘ L , w—r L ;
3 ~ I
a 0.5 V a — ~ 7 l
E o A L L I
BB PB RB NB ‘
Bean Flour 1
I
d.
Figure l6a—d. (cont’d)

*BB, Black beans; PB, Pinto beans; RB, Red beans; NB, Navy bean.

98

 

Caffeic Acid

1.4

 

Li Liberated by Acid
Hydrolysis

_ I Liberated by
Alkaline Hydrolysis

b3 Unconjugated

mg per 100 9 Bean Flour

 

 

  

 

BB PB RB

 

 

 

 

 

 

 

 

 

 

 

Bean Flour
1.
I ’——' A — FF *—_“fl
I
I
h :j Liberated by Acid I
3 Hydrolysis I
— I
l; A l Liberated by
3 Alkaline
m Hydrolysis
in [ii Unconjugated ‘
o I
o
‘-
h I
a I
a |
E I
Bean Flour I
I , LLL LL I
b.

Figure l7a—d. Hydroxycinnamic acids in Bean Flours*
*BB, Black beans; PB, Pinto beans; RB, Red beans; NB, Navy bean.

99

 

mg per 100 9 Bean Flour

 

 

 

 

 

l

Liberated by Acid

Hydrolysis

A I Liberated by

Alkaline Hydrolysis

Unconjugated

 

 

 

 

 

I Bean Flour
cl " ’ 7 7 7’ 7 ’
I Sinapic Acid
I . 10
:I 9
5 8
l c 7
I z 6
I m
I 5’ 5
I 5°: 4
I h 3
I 3. 2
I E’ 1
I 0
I
ILLLLLL LLLL L L
d.
Figure 17a—d. (cont’d)

*BB, Black beans; PB, Pinto beans; RB, Red beans; NB, Navy bean.

100

I I Liberated by Acid
Hydrolysis

I Liberated by
Alkaline Hydrolysis

P; Unconjugated

I

I
I

 

(+)-Catechin

Unconjugated

mg per 100 9 Bean Flour

 

BB PB RB
Bean Flour

 

Figure 18. Flavan-30ls in Bean Flours*
‘88, Black beans; PB, Pinto beans; RB, Red beans.

101

 

 

E Liberated by Acid
Hydrolysis

I Liberated by I
Alkaline Hydrolysis I

Unconjugated

 

mg per 100 9 Bean Flour

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a.
I
I
.5 :3 Liberated by Acid .
2 Hydrolysis ‘
I: I Liberated by I
:3? Alkaline Hydrolysis
{-3.3 U ‘
a nconjugated
c
a
1-
h
a
a
E
I Bean Flour
b.

Figure 19a—b. Flavonols in Bean Flours*
*BB, Black beans; PB, Pinto beans; RB, Red beans.

102

 

 

 

 

 

 

 

 

 

 

 

p- Hydroxybenzoic Acid
:0: 5:1 Liberated by Acid
g Hydrolysis
E +I I Liberated by
: Alkaline Hydrolysis
a .
a I Unconjugated
a .
o
o
I-
h
a
a
E
NBEM NBR
Bean Fraction
a.
Vanillic Acid
4 . .
c I] Liberated by ACld
§ 3.5 Hydrolysis
E 3 [Liberated by
: 2_5 Alkaline Hydrolysis
a
a 2 Unconjugated
g 1.5
t 1
a
I g 0.5
0
NBEM NBR
Bean Fraction
b.

Figure 20a—c. Hydroxybenzoic acids in Navy Bean Fractions*
*NBEM, Navy bean extract mix; NBR, Navy bean residue.

103

 

 

 

F L. ,L_

Syringic Acid

 

 

mg per 100 g Bean Fraction

 

 

 

 

 

I Bean Fraction

Figure 20a—c. (cont’d)
*NBEM, Navy bean extract mix; NBR, Navy bean residue.

104

 

I
III Liberated by Acid I
HYdrOIYSIS I

I Liberated by I
Alkaline Hydrolysis I

i

I Unconjugated

I
I
I
I
I

 

Caffeic Acid

I
I Liberated by Acid I
Hydrolysis

 

I Liberated by
Alkaline
Hydrolysis

Unconjugated

 

mg per 100 g Bean Fraction

 

 

 

NBEM NBR
Bean Fraction

 

 

 

a Liberated by Acid ‘
Hydrolysis

I Liberated by
Alkaline I

H rol is
Iubgonilifgated

 

 

 

mg per 100gBean Fraction
0 - N o: 4:- U1 c» \l co co

NBEM NBR I
Bean Fraction

 

Figure Zla—d. Hydroxycinnamic acids in Navy Bean Fractions*
*NBEM, Navy bean extract mix; NBR, Navy bean residue.

105

 

Ferulic Acid

I

ILiberated by Acid

 

 

 

 

 

 

 

    

 

 

f: Hydrolysis I
g I Liberated by
E Alkaline Hydrolysis
5 m Unconjugated
8
a:
o
2
5 I
a. .
°’ I
E
I NBEM NBR
I Bean Fraction I
LLL L L L LL LLL L L LLI
c.
f A ’7
I Sinapic Acid
I
I 12 Liberated by Acid I
I g- 10 f ‘ "I Hydrolysis I
§ lLiberated by I
I I: 8 4 Alkaline HydrolySISI
I g Ea Unconjugated I
. a 6 ___ i d I
u
. o
I2 4* if I
8 .
, a: 2 -L
IE .
I NBEM NBR ‘
I Bean Fraction I
I I
d.

Figure 21 a—d. (cont’d)
*NBEM, Navy bean extract mix; NBR, Navy bean residue.

106

Tables 9a—b show the method (no hydrolysis, alkali hydrolysis or acid
hydrolysis) that resulted in the maximum amount of phenolic aglycones in cooked beans
and navy bean fractions, respectively. Of the 11 phenolics identified in cooked beans,
only (+)-catechin was present entirely in the free form and not conjugated. In general,
alkaline hydrolysis resulted in the greatest concentration of hydroxycinnamic acids. For
some phenolics in certain beans, acid or base hydrolysis led to similar estimates (e. g.,
syringic for red beans and p-coumaric acid for pinto beans).

The data in Tables 9a———b is displayed as a bar graph in Figures 22 and 23, to
visualize the relative amounts of the identified phenolics in cooked beans and navy bean
fractions, respectively. In Figure 22 the first four bars (left to right) for black bean, pinto
bean, and red bean and the first three bars for navy bean show the relative amounts of the
hydroxybenzoic acids. In Figure 23, the first three bars for the navy bean fractions,
NBEM and NBR, show the relative amounts of the hydroxybenzoic acids. The amounts
of p-hydroxybenzoic, vanillic, and syringic acids are relatively low and similar in all four
bean classes and in the navy bean fractions. Protocatechuic acid was the major
hydroxybenzoic acid in the black, pinto and red bean, but was not detected in the navy
bean or its fractions.

The hydroxycinnamic acids are shown in Figure 22, bars 5-8 from the left for
black bean, pinto bean and red bean and the last four bars for the navy bean, and in
Figure 23, bars 4-7 for the navy bean fractions. Ferulic and sinapic acids were present in
much greater quantities than caffeic and p-coumaric acids. F erulic acid was the major
hydroxycinnamic acid in navy bean, the navy bean fractions and black bean. Red bean

had the least amount of ferulic acid. Although (+)-catechin was detected in black, pinto,

107

Table 9a—b. Phenolic Compounds Identified by HPLC-ECDa in Whole Bean
Flours and Navy Bean Fractions

 

 

 

 

 

 

 

 

 

a.
Bean Flour 1mg / 100 g of Bean Flour}
PC BB PB RB NB
PCA
PHBA
Van
S r
Caf
PCA
FA
SA
Cat
Q
K
b.
Navy Bean Fractions 1mg / 100 g Bean Fraction!
PC NBEM NBR

 

PCA ND ND

 

 

a. All values are dry basis. PC, Phenolic Compound; BB, black bean; PB, pinto bean; RB, red

bean; NB, navy bean; NBEM, navy bean extract mix; NBR, navy bean residue; PCA,
protocatechuic acid; pHBA, p—hydroxybenzoic acid; Cat, (+)-catechin; Van, vanillic acid; Caf,
caffeic acid; Syr, syringic acid; pCA, p-coumaric acid; FA, ferulic acid; SA, sinapic acid; Q,
quercetin; K, kaempferol; ND, not detected.

No Hydrolysis

Unconjugated PC plus PC liberated by Alkaline Hydrolysis
Unconjugated PC plus PC liberated by Acid Hydrolysis H

    

I’(" liberated by Alkaline or Acid Il_\(|riil) scs \\ are similar

108

 

 

60 L. 7 7' LL. 7,7 7 ..

 

 

 

 

 

I

Figure 22. Phenolic Compounds Identified in Whole Bean Fiours by HPLC-ECD
(mg/ 100 g of Bean Flour)II

 

 

a. BB, black bean; PB, pinto bean; RB, red bean; NB, navy bean; PCA, protocatechuic
acid; pHBA, p—hydroxybenzoic acid; Cat, (+)-catechin; Van, vanillic acid; Caf, caffeic
acid; Syr, syringic acid; pCA, p—cournaric acid; FA, ferulic acid; SA, sinapic acid; Q,

quercetin; K, kaempferol.

109

 

a pHBA

IVan

ISyr

ICaf

 

 

 

 

 

 

NBEM NBR

Figure 23. Phenolic Compounds Identified in Navy Bean Fractions by HPLC-ECD
(mg/100 g of Bean Fraction)ll

a. NBEM, navy bean extract mix; NBR, navy bean residue; pHBA, p—hydroxybenzoic
acid; Van, vanillic acid; Caf, caffeic acid; Syr, syringic acid; pCA, p-coumaric acid; FA,

ferulic acid; SA, sinapic acid.

110

and red bean, (Figure 22, bar 9 fiom the left) it is present in relatively small amounts
compared to kaempferol, quercetin, ferulic acid, and sinapic acid. Kaempferol (Figure
22, bar 11 from the left) was the major phenolic in red beans and pinto beans. In red
beans, kaempferol was 8 times greater than any other phenolic. Kaempferol was not
present in navy beans or its fractions. Quercetin (Figure 22, bar 10 from the left) content
was 4 to 8 mg per 100 g of flour for the black, pinto and red beans, but it was absent in
navy beans or its fractions.
Total Phenolic Content of Beans and Bean Fractions

The values shown in Table 9a—b were summed and the total identified phenolics
(TIP) are shown in the cooked whole bean flours and navy bean fractions in Figure 24a—
b. Figure 24a—b also shows the total phenolic content (TPC) of beans and navy bean
fi'actions as estimated by the Folin-Ciocalteu method. TPC ranged from 149 to 275 mg
of (+)—catechin equivalents ((+)-CE) per 100 g of bean flour or bean fraction and the
values are listed in Appendix B, Table 3 and Table 4. There is a large difference between
TIP and TPC. It is acknowledged that not all peaks observed by HPLC-ECD were
identified and quantified. Figures 13a — 15c (navy beans, NBEM, NBR) show only 2
unidentified peaks, but there were more unidentified peaks (up to 13) for the other beans.
Based on peak heights of unidentified peaks, it was estimated that unidentified peaks
accounted for approximately 5% (navy beans), 18% (red beans), 20% (black beans), and
30% (pinto beans) of the phenolic compounds in the bean extracts. Thus, there remains a
very large gap between the phenolics identified by HPLC-ECD and TPC and this

difference is most likely due to the non-specific nature of the Folin-Ciocalteu method.

111

 

 

ITlP
arTPC

 

 

ITIP
at TPC

 

NBEM NBR

b.
Figure 24a—b. Total Identified Phenolics. and Total Phenolic Content”

8. TIP, total identified phenolics—mg of imconjugated phenolics plus phenolics liberated by
alkaline or acid hydrolysis (which ever was greatest) per 100 g of Bean Flour or Bean Fraction
quantified by HPLC-ECD. b. TPC, total phenolic content—mg of (+)-catechin equivalents per

100 3 Bean Flour or Bean Fraction as determined by Folin-Ciocalteu method.

112

Cooked navy beans were separated into two fractions, the aqueous-ethanol
soluble components (NBEM) and the aqueous-ethanol insoluble components (NBR) in
order to extract a sufficient amount of phenolic compounds from cooked navy beans to
allow long-term colon cancer studies to be conducted with rodents. Prior to drying the
hygroscopic condensed aqueous-ethanol navy bean extract (concentrated in phenolic
compounds) was mixed with cellulose, casein and cornstarch, the TPC was determined.
Compared to the TPC measured in the cooked navy bean (1487 mg (+)-CE per kg dry
bean flour vs. 671 mg (+)-CE per kg dry bean residue, 45% of total phenolic compounds
was extracted from the cooked navy beans using ~60% ethanol, and 10% was lost or

could not be accounted for after the extraction process.

D. DISCUSSION

This study is the first to identify and quantify the major hydroxybenzoic acids,
hydroxycinnamic acids, flavan-3ols, and flavonols in four market classes of dry common
beans that had been prepared for human consumption. In previous studies raw legumes
or just the seed coats (bulls) were analyzed; however, people do not consume raw beans
or hulls. People eat whole or split beans that have undergone some form of heat
treatment. Therefore, one goal of this research was to identify and quantify the phenolic
compounds in beans that a person would consume.

Bean processing usually includes hydration (soaking) and a heat treatment (i.e.
boiling, pressure cooking, canning, roasting, rnicronization or autoclaving). Some
cultures germinate or ferment beans prior to consumption, but a heat treatment is imposed
at some point prior to consumption. The various processing methods induce physical and

chemical compositional changes that result in the loss of phenolic compounds (Champ,

113

2002; Ibrahim et al., 2002; Shimelis and Rakshit, 2007). For example, Diaz-Batalla et al.
(2006) found a 42% to 48% reduction in both quercetin and kaempferol in autoclaved
(cooked) beans compared to raw beans. Others have reported a decrease in TPC by 29%
to 68% during preparation of beans for consumption (Towo et al., 2003; Xu and Chang,
2008; Granito et al., 2008). These studies underscore the importance of identifying and
quantifying phenolics in foods in forms that are normally consumed.

With the exception of (+)-catechin, phenolic compounds are not found in their
free, aglycone form in nature, but bound as glycosides and esters (Ribereau-Gayon, 1972;
Harbome, 1998). When free soluble aglycones are found in plant foods, they are in all
probability released during storage, processing, and extraction process (Harbome, 1998).
As soon as plant tissue is disrupted it becomes susceptible to enzymatic oxidation and
hydrolysis via endogenous enzymes such as esterases, glycosidases, and decarboxylases
(Harbome, 1998; Cheynier, 2005). Although researchers attempt to limit oxidation and
decarboxylation of phenolic compounds to avoid alterations in chemical structures,
inevitably some will occur.

When identifying and quantifying phenolic compounds in plant foods a hydrolysis
step is commonly performed to simplify identification and quantification. Hydrolysis
also removes unwanted compounds in the complex plant matrix that would otherwise
interfere with the analysis (T sac and Deng, 2004; Stalikas, 2007). Acid and alkaline
(saponification) hydrolysis are the most common techniques applied to release the
phenolic compound aglycones from their natural conjugated state (Stalikas, 2007). Each

step during the extraction process introduces errors and artifacts that will reduce the yield

114

of total phenolic compounds (Stalikas, 2007). These losses are rarely reported and the
data are not corrected for these losses.

In the present study, the percentage of phenolic compounds lost during the clean
up step (solid phase extraction, SPE C13 cartridges) of the extraction process, and due to
alkaline or acid hydrolysis was assessed by applying the same conditions to pure
standards. Losses varied greatly and were dependent on the hydrolysis treatment used
and the phenolic compound, similar to previous studies. Krygier et a1. (1982) observed a
3% to 67% and 15% to 92% loss in hydroxycinnamic acids resulting from alkali or acid
treatment, respectively. Dabrowski and Sosulski ( 1984) reported the losses of
hydroxybenzoic acids and hydroxycinnamic acids in rapeseed flour exposed to alkaline
conditions. Losses in hydroxybenzoic acids did not exceed 10% and losses in
hydroxycinnamic acids ranged from 13%-22%. Both authors reported high losses of
caffeic acid under alkali or acid treatment and this was the result of the high reactivity of
o—dihydroxyphenols, which can be rapidly oxidized to quinones (Ribereau-Gayon, 197 2).
In the present study, 10% to 90% of phenolic compounds were lost due to sample clean
up and hydrolysis by base or acids. The data in the present study were adjusted for these
losses.

The classes of phenolic compounds previously identified in the grain legumes
were flavonols, hydroxybenzoic acids, hydroxycinnamic acids, condensed tannins
(proanthocyanidins), and for a few bean classes, anthocyanins (Krygier et al., 1982;
Sosulski and Dabrowski, 1984; Beninger etal., 1999; 2003; 2005; Madhujith et al., 2004;

Aparicio-Femandez et al., 2005; Hu et al., 2006; Lin and Lai, 2006; Ranilla et al., 2007).

115

Protocatechuic acid was the predominant hydroxybenzoic acid identified in the
black, pinto and red beans, but navy beans did not contain protocatechuic acid. Previous
studies have not reported the presence of protocatechuic acid in beans, although Cai et a1
(2003) reported that protocatechuic acid was in cowpea (Vigna unguiculata). Diaz-
Batalla et al (2006) quantified hydroxybenzoic acids in several varieties of raw wild and
weedy Mexican common beans (Phaseolus vulgaris L.) and Espinosa—Alonso and co-
workers (2006) quantified hydroxybenzoic acids in black J amapa beans and pinto beans.
Both studies reported the presence of p-hydroxybenzoic acid, vanillic acid and syringic
acid in beans. Espinosa—Alonso et al (2006) did not detect syringic acid in pinto beans.
Syringic acid was found in pinto beans when analyzed by HPLC-ECD, but it was found
in the lowest concentrations. The two groups analyzing hydroxybenzoic acids in beans
reported totals ranging from 1.0 to 7.1 mg/ 100 g bean flour. Total hydroxybenzoic acids
determined in this study were slightly higher (2.9 — 11.7), probably due to differences in
methods, corrections for losses, and identification of more acids.

The hydroxycinnamic acids identified in cooked dry common beans were caffeic,
p-coumaric, ferulic and sinapic acids. Total amounts of hydroxycinnamic acids in beans
were 11 (red beans), 20 (pinto beans), 24 (black beans) and 36 (navy beans) mg/ 100 g
bean flour. Navy beans had the largest concentration hydroxycinnamic acids. The
predominant hydroxycinnamic acid was ferulic acid followed by sinapic, p-coumaric and
caffeic acids. Ferulic acid was present in 1.3 — 4.5 times greater than the amount of
sinapic acid. Four other studies have reported data for hydroxycinnamic acids in dry
common beans. The other studies either used raw samples or they did not hydrolyze with

base, which is required to free the majority of the hydroxycinnamic acids.

116

Espinosa-Alonso et al. (2006) and Luthria and Pastor-Corrales (2006) identified
and quantified the same four hydroxycinnamic acids as found in this study. Luthria and
Pastor-Corrales (2006) analyzed 15 varieties of raw dry common beans (Phaseolus
Vulgaris L.) for hydroxycinnamic acid content and reported totals ranging from 19.1 to
48.3 mg/ 100 g bean flour. Since cooking results in a 23%-40% reduction in the total
amounts of hydroxycinnamic acids (Diaz-Batalla et al., 2006), the values found in this
study and those reported by Luthria and Pastor-Corrales (2006) are comparable. Luthria
and Pastor-Corrales (2006) identified caffeic acid only in black beans, whereas Espinosa-
Alonso et al. (2006) reported the presence of caffeic acid in all bean varieties as in this
study. Diaz-Batalla et a1 (2006) quantified hydroxycinnamic acids in several varieties of
raw wild and weedy Mexican common beans (Phaseolus vulgaris L.) after acid
hydrolysis and reported values for ferulic and sinapic acids. Both Diaz-Batalla et al
(2006) and Espinosa—Alonso ct al. (2006) used acid hydrolysis in their measurements of
the hydroxycinnamic acids and so the values they reported are much lower than reported
in this study or by Luthria and Pastor-Corrales (2006).

Cai et al. (2003) quantified hydroxycinnamic acids in raw cowpeas. Following
alkaline hydrolysis, they identified caffeic, p-coumaric, ferulic and cinnamic acids with
totals ranging from 1.8 to 20.4 mg / 100 g bean flour. Ferulic acid (0.4 to 12.4 mg / 100 g
bean flour) was the predominant hydroxycinnamic acid present in cowpeas, similar to
what has been reported for dry common beans. However, the amounts of
hydroxycinnamic acids present in cowpeas appear to be less than what is in dry beans.

Previously (+)-catechin has been identified in the seed coats (hulls) of light brown

Brazilian beans (Phaseolus vulgaris L.) (Ranilla et al., 2007). Table 9a shows that (+)-

117

catechin was in black, pinto and red beans samples, but not in navy beans. Pinto had the
greatest amount, 2.7 mg / 100 g bean flour, followed by red and black (0.3 mg / 100 g bean
flour). (+)-Catechin was not detectable following hydrolysis with alkaline or acid. This
was not unexpected, because (+)-catechins form condensation products in acid (Ribereau-
Gayon, 1972), and are not usually quantified after alkaline hydrolysis.

Flavonols and their glycosides have been identified in common beans previously
(Beninger et al., 1999, 2003, 2005; Aparicio-Femandez etal., 2005; Hu et al., 2006;
Diaz-Batalla et al., 2006; Espinosa-Alonso et al., 2006; Ranilla etal., 2007). All reported
the presence of quercetin and kaempferol and their glycosides, but one study reported the
presence of myricetin glycosides in the seed coat of black J amapa beans (Phaseolus
vulgaris L.) (Aparicio-Femandez et al., 2005).

In the current study, quercetin and kaempferol were identified in cooked black,
pinto and red beans, but not in navy beans. Total flavonols were 7 (black beans), 24
(pinto beans) and 67 (red beans) mg/ 100 g bean flour. Red beans had much more total
flavonols than pinto and black beans. Diaz-Batalla et a1. (2006) and Espinosa-Alonso et
al. (2006) reported total flavonol concentrations of 0.4 to 23.2 mg/ 100 g bean flour in
several varieties of raw wild and weedy Mexican common beans (Phaseolus vulgaris L.).
Espinosa-Alonso et al. (2006) and Ranilla et a1. (2007) reported that pinto beans had
more kaempferol than quercetin, whereas the black bean had more quercetin than
kaempferol. The results shown in Table 9a are consistent with these reports; black beans
had a more quercetin than kaempferol whereas pinto and red beans had more kaempferol

than quercetin.

118

Diaz-Batalla et a1 (2006) additionally quantified flavonols in autoclaved (cooked)
beans and found a 42% to 48% reduction in both quercetin and kaempferol (1.1 to 13.5
mg of total flavonols per 100 g bean flour (cooked beans) compared to 2.1 to 23.2
mg/ 100 g bean flour (raw beans)). Despite analyzing cooked beans, which would be
expected to have reduced total flavonols, the flavonol concentrations shown in Table 9a
(7 — 67) are much greater than previously reported. The greater concentrations most
likely resulted fi'om improved methodology and the analysis of red beans which had high
concentrations of kaempferol.

Pinto beans had the greatest TPC, followed by black, red and navy bean.

Previous studies reporting the total phenolic content (TPC) in raw grain legumes using
the Folin-Ciocalteu method, observed ranges of 35 to 440 mg per 100 g of raw bean flour
(Cai et al., 2003; Heimler et al., 2005; Xu and Chang, 2007). In this study the T PC
(Figure 24a; Appendix B, Table 3) in the cooked dry common bean extracts ranged from
149 to 275 mg per 100 g bean flour. The values reported here for the cooked beans are
consistent with those in the literature, even though cooked beans were analyzed.

Treating the cooked dry common bean extracts with acid resulted in greater
amounts of hydroxybenzoic acids being liberated compared to treatment with alkali. This
suggests that the majority of hydroxybenzoic acids in the cooked common beans
analyzed in this study are simple glycosides. Treatment with alkali liberated larger
quantities of hydroxycinnamic acids than acid hydrolysis, suggesting that
hydroxycinnamic acids in black, pinto, red and navy beans are predominantly found as

esters. Hydroxycinnamic acids, especially caffeic acid, have been shown to be unstable

119

under acidic conditions, thus they are best obtained under mild alkali hydrolysis (Krygier
et al., 1982; Gao and Mazza, 1994; Harbome, 1998).

Cooked navy beans were extracted with aqueous-ethanol (~60% ethanol) to
separate navy beans into two fractions for future animal studies: 1. navy bean extract
(NBEM)—aqueous-ethanol soluble components (concentrated in phenolic compounds);
2. navy bean residue (NBR)—aqueous-ethanol insoluble solid components (i.e. dietary
fiber and protein). The aqueous-ethanol solvent extracted 45% of the total phenolic
compounds quantified in cooked navy beans.

The hydroxybenzoic acids were fairly similar in the NBR and navy beans
suggesting that the aqueous-ethanol solvent extracted only small amounts of the
hydroxybenzoic acids (Figure 20a—c). Although NBEM had approximately 2 times as
much total hydroxybenzoic acids, the NBEM is a concentrate and does not reflect
efficient extraction of the hydroxybenzoic acids.

The aqueous-ethanol was more efficient in extracting hydroxycinnamic acids
(Figure 21a—d) than it was in extracting hydroxybenzoic acids. Fifiy to 74% of the
hydroxycinnamic acids were extracted and identified by HPLC-ECD. Since ferulic acid
was the predominant acid in navy beans and since ferulic acid was efficiently extracted
(74%), the extraction of total hydroxycinnamic acids was 74% and the extraction of all
phenolics was 67%. The NBEM had approximately 1.7 times more total
hydroxycinnamic acids and total identified phenolics than navy beans.

In conclusion, all four cooked dry common beans (Phaseolus vulgaris L.), black,
pinto, red and navy beans contained hydroxybenzoic and hydroxycinnamic acids. The

hydroxycinnamic acids included p—coumaric, caffeic, ferulic and sinapic acids. Ferulic

120

acid was the predominant hydroxycinnamic acid. The hydroxybenzoic acids — p-
hydroxybenzoic, vanillic and syringic acids — were present in all four beans whereas
protocatechuic acid was not in navy beans. Black, pinto and red beans contained, (+)-
catechin, quercetin and kaempferol, but navy beans did not contain these phenolics.
Additionally, a large scale aqueous-ethanol extraction of phenolic compounds in cooked
navy beans was performed. Aqueous-ethanol was more efficient in extracting
hydroxycinnamic acid conjugates than hydroxybenzoic acid conjugates in cooked navy

beans.

121

CHAPTER Iv. INHIBITORY EFFECTS OF COOKED NAVY BEAN
(PHASEOL US VUL GARIS L.) ON AZOXYMETHANE (AOM)-INDUCED

COLON CANCER IN OB/OB OBESE MICE

A. ABSTRACT

Epidemiological and experimental studies have observed an inverse relationship
between dry bean consumption and colorectal cancer (CRC). Additionally, dry bean
consumption has been inversely associated with risk of developing obesity and type 2
diabetes mellitus, both are independent risk factors for CRC.

The overall goals of this study were to l) evaluate whether obese, mildly diabetic
mice (ob/ob mice (B6.V-Lep0b /J)) are good animals for studying azoxymethane (AOM)-

induced colon cancer and, 2) to narrow the search for which component in dry beans
inhibits the development of chemically-induced colon cancer, by feeding a) navy beans
(Phaseolus vulgaris L.); b) an aqueous-ethanol extract of navy beans (NBEM) and c) the
residue remaining after aqueous-ethanol extraction (NBR).

The ob/ob mice fed the control diet had a low tumor incidence, only 10%
adenoma incidence and 10% adenocarcinoma incidence, which seriously compromised
the ability to distinguish dietary induced differences in colon cancer inhibition. Although
increasing the AOM dosage would have increased tumor incidence, higher doses of
AOM would have increased mortality in this animal model because AOM is a

hepatotoxin and ob/ob mice have impaired hepatic repair and regenerative responses to

injury.

122

Feeding the aqueous-ethanol extract of navy beans rich in phenolic compounds
(NBEM) completely inhibited tumor development (tumors include adenomas and
adenocarcinomas, tumor incidence = 0%). Although tumor incidence was inhibited by
69% in mice fed the aqueous-ethanol insoluble components of navy beans (NBR)
compared to the control diet fed mice this was not statistically significant. The low tumor
incidence in the control fed animals weakened the statistical power in this study making it
impossible to partition the cancer inhibiting activities of navy beans clearly.

In conclusion, the ob/ob mouse did not have the desired sensitivity to the colon
carcinogenic properties of AOM to be recommended for widespread use to test dietary
ingredients for colon tumor inhibiting potential. Nevertheless, the compounds contained
in the aqueous-ethanol extract of navy beans (NBEM) reduced tumor incidence by 100%

compared to the control diet.

123

B. INTRODUCTION
Colorectal cancer (CRC) is the 3rd most commonly diagnosed cancer and the 4th
most common cause of cancer death worldwide (WCRF/AICR, 2007). In the United

States alone, CRC is the 4th most commonly diagnosed cancer following lung, female

breast, and prostate cancers, and the 2"d leading cause of cancer death (J emal et al.,

2008). While some individuals are predisposed to developing CRC because of
inheritable factors, environmental and lifestyle factors such as — being a non-smoker,
being physically active, maintaining a healthy body weight, and consuming a healthy diet
— can greatly reduce an individual’s risk (WCRF/AICR, 2007). Approximately 70% of
colorectal cancers may be preventable by a healthful diet and other lifestyle choices
(Platz et al., 2000). Thus, diet is an important factor that can be modified and alter CRC
risk.

Evidence demonstrating the important role environmental and/or lifestyle factors
play in modulating CRC risk was first observed in population-studies. Large
geographical variations in CRC incidence across the world were reported, with lower
incidences in Africa and Asia, moderate incidences in South America and high
incidences in North America, Australia/New Zealand and Western Europe (Parkin et al,
2005; WCRF/AICR, 2007). Dry beans have long been a dietary staple in the traditional
diets consumed by the populations living in the aforementioned low risk regions.
Moreover, epidemiological studies have observed an inverse relationship between dry
bean consumption and CRC risk (Correa, 1981; Steinmetz and Potter, 1993; Michels et
al., 2006). In the Polyp Prevention Trial (PPT, a prospective trial-based cohort), the

association between the consumption of fruits, vegetables, dry beans and other vegetable

124

groups and the recurrence of colorectal adenomas in humans was examined. Researchers
reported a significant 65% reduction in the recurrence of advanced colorectal adenomas
in individuals who increased their dry bean intake (highest quartile) compared to those
with the least change in intake (lowest quartile) over a 4 yr period (p = 0.001). Four-day
food records revealed the five most highly consumed dry beans were baked beans (navy
beans), kidney beans, pinto beans, and lima beans (Lanza et al., 2006).

Besides colorectal cancer, dry bean consumption has been inversely associated
with risk of developing obesity and type 2 diabetes mellitus (Geil and Anderson, 1994;
Maskarinec et al., 2000; Guillon and Champ, 2002; Venn and Mann, 2004; Villegas et
al., 2008). Characteristics of obesity include impaired glucose tolerance,
hyperinsulinemia, dyslipidemia, high blood pressure, elevated circulating
proinflammatory cytokines (i.e. TNF-a, IL-6) (Gunter and Leitzmann, 2006). Many of
these metabolic perturbations ensuing from obesity, have been positively associated with
colon cancer (Gunter and Leitzmann, 2006). Hyperinsulinemia (high circulating levels of
insulin) results from increased synthesis and secretion of insulin from the pancreatic B-
cells. Hyperinsulinemia is a compensatory response that occurs because higher blood
levels of insulin are required to maintain glucose homeostasis. As insulin resistance
persists and worsens, the pancreatic B-cells lose the ability to secrete sufficient insulin to
maintain glucose homeostasis and that leads to hyperglycemia and to the development of
type 2 diabetes mellitus (Sandhu etal., 2002). Both obesity and type 2 diabetes mellitus
are independent risk factors for the development of CRC (Coughlin et al., 2004; Brun et

al., 2007; WCRF/AICR, 2007).

125

Aberrant crypt foci (ACF) are preneoplastic lesions commonly used as an early
biomarker in short term colon carcinogenesis studies (Bird, 1995). Additionally, they
have been identified in the colon of humans at high risk for CRC such as those with
inheritable syndromes such as familial adenomatous polyposis (F AP) or a history of
carcinomas or adenomas (Glebov et al., 2006; Femia and Caderni, 2008). Several studies
utilizing chemically-induced colon cancer in genetically obese animal models have found
an increased number of ACF compared to their lean, wild type littermatcs. Increases in
ACF have been reported in genetically obese, diabetic mouse models (db/db mice and
ob/ob mice) injected with the colon carcinogen azoxymethane (AOM) compared to their
lean counterparts (Hirose et al., 2004; Ealey et al., 2008). Similarly, in obese rats
(Zucker,fa/fa) injected with AOM, the number of colon tumors and aberrant crypts were
greater compared to their lean littermates (Weber et al., 2000).

Several studies have evaluated the potential for dry beans and other legumes to
reduce chemically-induced colon carcinogenesis in lean rodents. Tumor incidence was
reduced ~52 to 60% when cooked dry common beans (Phaseolus vulgaris L.) were fed to
AOM-inj ected rats (Hughes and Ganthavom, 1997; Hangen and Bennink, 2002; Rondini,
2006). In a recent study, a group fed pinto beans, black-eyed peas and soybeans to
AOM-injected rats and found a 64%, 77% and 56% reduction in ACF, respectively
(Boateng et al., 2008). There does not appear to be any study that has evaluated the
potential for dry beans to reduce colon carcinogenesis in an obese animal model.

Dry beans are a great source of dietary fiber, containing 8% to 28% on a dry
weight basis (Guillon and Champ, 2002; USDA/ARS, 2009), and it is probable that the

dietary fiber is protective against CRC, however the evidence has not always been

126

consistent (Fuchs et al., 1999; Schatzkin et al., 2000; McCullough et al., 2003;
WCRF/AICR, 2007). Dry beans contain numerous bioactive constituents that may slow
the development of colon cancer. Potentially bioactive non-nutrients found in dry beans
include phytic acid, saponins, oligosaccharides, and phenolic compounds (Geil and
Anderson, 1994; Messina, 1999; Guillon and Champ, 2002). Phenolic compounds have
been shown to possess anti-cancer activity and numerous studies have been published
examining the mechanism of action (Kwon et al., 2007; Stevenson and Hurst, 2007). In
Chapter III, phenolic compounds were identified and quantified in four dry, cooked
common beans — black, pinto, red and navy beans. All four beans contain
hydroxybenzoic acids and hydroxycinnamic acids. Navy beans were found to contain
approximately 2 to 3 times more total hydroxycinnamic acids than the black, pinto and
red beans. Several phenolic compounds identified in black, pinto and red beans were not
found in navy beans, including protocatechuic acid (hydroxybenzoic acid), (+)—catechin
(flavan—Bol), quercetin (flavonol), and kaempferol (flavonol). Additionally anthocyanins
have been identified in beans possessing blue and blue-violet colored seed coats, i.e.
black beans (Beninger et al., 2003). Yet despite these differences in phenolic compound
composition pinto beans, black beans and navy beans all inhibited chemically induced
colon cancer in a rat model by > 50% (Hughs et al., 1997; Hangen and Bennink, 2002;
Rondini, 2006).

Distinguishing which components found in dry beans protect against the
development of CRC and the molecular mechanism by which they confer protection has
not been delineated. Partitioning beans into different fractions and determining the anti-

cancer activity for the various fractions is one approach to narrowing the search for the

127

anti-cancer components of legumes. Bennink and Barrett (2004) used this approach with
soybeans (an oil legume). They fed heat-treated, full-fat soybean flour, soybean flour
minus lipids, defatted soybean flour minus aqueous-ethanol components and a control
diet to determine which fractions protected against development of AOM-induced colon
cancer. Full-fat soybean flour and defatted soybean flour diets significantly reduced
tumor incidence by 45% (P < 0.05) compared to rats fed a casein control diet. Based on
this experiment and other unpublished studies it was determined that soybean fiber,
sterols, fatty acids, protein and isoflavones could not be providing the anti-cancer action
noted with defatted soy flour. However, removing the aqueous-alcohol soluble
components completely eliminated any anti-cancer action (Beninnk and Barrett, 2004).
The aqueous-alcohol fraction would contain not just isoflavones, but other phenolic
compounds such as syringic acid, ferulic acid, sinapic acid and p-coumaric acid (Seo and
Morr, 1984), as well as saponins and oligosaccharides. Subsequent studies showed that
the isoflavones do not inhibit colon cancer.

In the previous study (Chapter III) the aqueous-alcohol soluble components
(NBEM) in navy beans were separated from the aqueous-alcohol insoluble components
(NBR) in order to extract a sufficient amount of phenolic compounds from cooked navy
beans to allow long-term colon cancer studies to be conducted with rodents. Forty-five
percent of the total phenolic compounds in cooked navy beans were extracted by an
aqueous-ethanol solvent (~60% ethanol) and identified in the NBEM. The aqueous-
ethanol solvent was more efficient in extracting the hydroxycinnamic acid conjugates-

than the hydroxybenzoic acid conjugates from cooked navy beans.

128

Although the bioactive constituents in dry beans responsible for inhibiting the
development of colon cancer have not been fully elucidated, phenolic compounds (i.e.
flavonoids, phenolic acids) have been identified in the human fecal water in
concentrations ranging from 0.56 M and 627.9 M, with the majority being phenolic
acids. These levels are comparable to those found to exert anti-cancer activity in in vitro
and cell culture studies (J ener et al., 2005). Phenolic compounds in dry beans not
absorbed in the small intestine will enter the colon. Some will remain in the solid phase
of the human feces (bound to solid food components) while others will be part of the
aqueous phase or fecal water. The biochemical composition of fecal water is directly
influenced by the diet and varies greatly between individuals (Rafter et al., 1987;
Nordling et al., 2003; Record etal., 2003). The fecal water is able to interact more with
the colonic epithelial cells in comparison to the solid material, and can influence cellular,
molecular and genetic changes associated with the etiology of colon cancer (Nordling et
al., 2003; Karlsson et al., 2005). Thus, phenolic compounds in dry beans may be
responsible for some of the anti-cancer activity observed in previously studies.

Since navy beans have previously been shown to inhibit chemically induced colon
cancer, are the type of bean used to make baked beans, and the dominant type of dry bean
consumed by non-Hispanic patients enrolled in the Polyp Prevention Trial, they were
chosen to conduct the current study.

The overall goals of this study were to evaluate whether obese, mildly diabetic
mice (ob/ob mice (B6.V-Lep0b /J)) are good animals for studying azoxymethane-induced

colon cancer and to narrow the search for which component in dry beans inhibits the

development of chemically-induced colon cancer.

129

The specific objectives of this study were to determine: 1) if greater than 50%
adenoma and greater than 20% adenocarcinoma incidence could be induced when a
control diet was consumed and 2) if feeding a) navy beans (Phaseolus vulgaris L.); b) an
aqueous-ethanol extract of navy beans and c) the residue remaining after aqueous-ethanol

extraction would reduce adenoma and adenocarcinoma incidences.

C. MATERIALS AND METHODS
Chemicals

Azoxymethane (AOM) was purchased from Sigma-Aldrich Chemical Co. (St.

Louis, MO). Ethanol, Paraplast® Tissue embedding media (paraffin pellets), Protocol

Eosin Y (1% Alcohol), Hematoxylin Stain Solution Gill 3 Formulation (pH 2-4) were
purchased fiom Fisher Scientific (Pittsburgh, PA, USA). 37% Formaldehyde and
Xylene were from J .T. Baker (Phillipsburg, NJ, USA.)
Experimental animals

Male ob/ob mice (B6.V-Lep0b /J), n=160, 4 to 5 weeks in age were obtained from
Jackson Laboratory in Bar Harbor, ME. They were housed 4 mice per cage in a
temperature (24°C :1: 2°C) and humidity (40%-60%) controlled room with a 12 h
light/dark cycle. Animals were allowed free access to food and water. The animals were
checked daily for health status and weighed weekly for 9 weeks, then monthly. This
study was approved and conducted under the guidelines of the All University Committee

on Animal Use and Care at Michigan State University.

130

Diet Preparation

Preparation of the navy bean flour and the fiactions derived from it is described in
Chapter HI, C. Material and Methods. The beans were kindly provided by the Michigan
Bean Commission and were stored at 4°C prior to cooking.

Table 10 lists the composition of the four diets fed to the mice; 1) control diet —
modified AIN-93G diet; 2) NB — cooked navy bean diet, 3) NBR —— navy bean residue
diet; and 4) NBEM—navy bean extract mix diet. All four diets were formulated to
provide 17.1% protein, 16.7% fat, 14.4% fiber (wt:wt) and 4 kcal/ g. The vitamin mix
was added according to the American Institute of Nutrition standard AIN—93G for
laboratory rodents (Reeves et al., 1993). The mineral mix did not contain calcium, but all
other mineral concentrations were as described for the AIN—93G mineral mix. Calcium
was added at a lower concentration than the AlN-93G since the high concentration of
calcium in the standard A1N-93G mineral mix may inhibit colon cancer. All diets were
supplemented with 0.3 g of methionine per 100 g of diet and 0.005 g of tryptophan per
100 g of diet was added to the NB and NBR to meet amino acid requirements for
growing rodents (Reeves et al., 1993).

Navy beans contain approximately 19.5% dietary fiber. In order to minimize
dietary fiber effects, all diets were modified to match that found in the navy bean and
navy bean residue diets or 14.4% dietary fiber (wt:wt), compared to the 4% found in
AIN-93G diet (Reeves et al., 1993).

The diets were formulated to provide 36% of the energy fi'om fat, which simulates

the average American diet, but the 16.7% fat content is much higher than 7% fat in the

131

Table 10. Composition of Diets Fed to ob/ob Micea

 

 

 

Diets (g/100g)

Ingredients

Control NB NBR NBEM
Beans 74.0
Residue 74.0
Concentrate 9.0
Casein 18.0 18.0
Methionine 0.3 0.3 0.3 0.3
Tryptophan 0.005 0.005
Corn oil 1.1 1.1
Lard 12.9 12.9 12.9 12.9
Soybean oil 2.6 2.6 2.6 2.6
Cornstarch 44.4 4.0 4.0 35.4
Fiber (cellulose) 14.4 14.4
Sucrose 1.0 1.0 1.0 1.0
Calcium Carbonate 0.3 0.3 0.3 0.3
Mineral mix 3.6 3.6 3.6 3.6
Vitamin mix 1.0 1.0 1.0 1.0
Choline bitartrate 0.3 0.3 0.3 0.3
BHT 0.0014 0.0014 0.0014 0.0014
Total 100.0 100.0 100.0 100.0
TPC b 0 110.6 49.4 109.8

 

a. Control is a modified AIN 93 G (Reeves et al., 1993); NB = whole navy bean diet;
NBR = navy bean residue diet; NBEM = navy bean extract mix diet. b. TPC, total

phenolic content—mg of (+)-catechin equivalents per 100 g of diet as determined by

Folin-Ciocalteu method.

132

standard AlN-93G diet. The fatty acid profile also simulated the typical American diet
(Ernst et al., 1997).

The NBEM contained the same amount of total phenolics as the NB diet, 110.6
and 109.8 mg (+)-catechin equivalents per 100 g of diet, respectively. The NBR diet
contained 49.4 mg (+)—catechin equivalents of phenolic compounds per 100 g of diet.
The NB and NBEM were formulated to be equivalent in total phenolic acid content, and
contained 2.2 times more phenolic compounds than the NBR diet.

Initiation of Colon Cancer

Male ob/ob mice were received when they were 4 to 5 weeks of age. The mice
were given two subcutaneous injections of 7 mg of AOM dissolved in saline per kg of
body weight one week apart to initiate the cancer process. The control diet (Table 10)
was fed from arrival to 1 week following the second injection. The animals (9 to 10
weeks of age) were then randomized by weight to 1 of the 4 dietary treatment groups
(n=40 mice per dietary treatment). The study was terminated 27 to 29 weeks after the
second injection of AOM.

Necropsy and Histology

Animals were sacrificed using carbon dioxide inhalation and exsanguination.
Colons were removed, cut longitudinally, rinsed with lukewarm tap water and flattened
on filter paper and visually inspected for lesions. The colons were then fixed in 10%
neutral buffered formalin (NBF, pH 7.4) for 5-6 hours. Other internal organs were also

examined grossly for lesions and fixed in 10% NBF if an organ appeared abnormal.

133

Mucosa (epithelial cells) was collected fiom the mid colon region of 12-15
animals per dietary treatment by gently scraping the normal appearing tissue with the
edge of a glass slide. The mucosa was immediately snap fiozen on dry ice and then
stored at -80°C until analyzed. -

After fixation in NBF, the colons were re-examined by two researchers,
independently, for lesions. Locations of visible tumors and any slightly raised and
opaque areas were recorded, dissected, and placed in pathology cassettes. The remaining
colon tissue was transected into 4 segments: proximal colon, mid-colon 1 (2 cm), mid-
colon 2 (2 cm) and distal colon and placed cassettes. Any organ tissue suspected of being
a tumor was dissected and place in cassettes. All cassettes were dehydrated and
infiltrated with paraffin using standard histological procedures. Tissue sections (4 micro
thickness) from all lesions and colon segments were processed using standard
histological procedures and subsequently stained with hematoxylin and eosin.

A pathologist examined all tissues, and lesions were categorized as focal
hyperplasia, dysplasia, adenoma and adenocarcinoma based on standard histological
grading criteria (Krutovskikh and Turusov, 1994).

Statistical Analysis

Two-way ANOVA with repeated measures was used to analyze the grth data
and post-hoe analysis was done by the Fisher’s LSD method. The Chi square test was
used to determine statistical differences in lesion incidence. A probability of P < 0.05

was considered statistically significant.

134

D. RESULTS
Growth

The mean weights of the ob/ob mice fed the various dietary treatments are shown
in Figure 25. After feeding the assigned dietary treatments for 3 or 4 weeks, ob/ob mice
fed the NB and NBR weighed more (P<0.05) than mice fed the control diet. Mice fed the
NBEM for 15 weeks weighed less (P<0.05) than mice fed the NB diet.
Lesion Incidence

Lesions include focal hyperplasia, dysplasia, adenomas and adenocarcinomas
(Table 11). All lesions were located in the mid- or distal colon. In mice fed the control
diet, the lesion incidence was 40% (16/40). There was no difference (P>0.05) in lesion
incidence between mice fed the control diet and mice fed the NB diet. Mice fed the NBR
diet had fewer lesions than mice fed the control diet (P<0.05), but the number of lesions
was similar for mice fed the NB diet and the NBR diet (P>0.05). Also, the number of
lesions was similar for mice fed the NBR diet and the NBEM diet (P>0.05). Mice fed the
NBEM diet had the fewest lesions. The lesion incidence for mice fed the NBEM was
less than mice fed the control or the NB diets (P<0.05).
Tumor Incidence

Tumors included adenoma and adenocarcinoma (Table 12). In mice fed the
control diet, the tumor incidence was 18% (7/40) and the number of tumors per tumor
bearing mice was 1.3 per mouse. Mice fed the whole navy bean (NB) diet had a tumor
incidence of 15% (5/33) and the tumor incidences for the mice fed the control and NB

diets were similar (P>0.05). For mice fed the NBR diet, the tumor incidence was 5%

135

on

£05 anom— bfiz .5 5.530 3% new: no}: tuneuazuou< «e 5.5qu A3 unaut—
85 x=2 68cm comm >>mzl0l .20 game”. comm >>mzlll .20 comm >>wzlll .05 _o.=coo+
8.025 c. 02
mm cm 3 or m

 

r ov

 

 

om

 

 

 

8
(B) nuBIOM "new

 

 

on

 

 

on

136

Table 11. Lesion Incidence'I in Colons of ob/ob Mice Fed Control or Navy Bean

 

 

Diets."c
Diets # #:2113111; 111153;?“ % Incidenced % Inhibition
Control 1.6 40 (16140)“
NB 1.0 21 (7/33)ab
NBR 1.0 14 (5/37)be 66% vs. control
NBEM 1.0 5 (2/39)c 87% vs. control & 76%

vs. NB

 

a. Lesions include: focal hyperplasia, dysplasia, adenomas and adenocarcinomas. b.
Control = Control diet; NB = Navy bean diet, NBR = Navy bean residue diet, NBEM =
Navy bean extract mix. c. Values in a column are significantly different if they don’t
share the same superscript (P < 0.05). d. In parenthesis is the number of animals with

lesions per number of mice alive at the end of the experiment.

137

Table 12. Tumor Incidencea in Colons of ob/ob Mice Fed Control or Navy Bean

 

 

moth"
# Tumors /
Diets Tumor bearing % Incidenced % Inhibition
mouse
Control 1.3 13 (7140)a
NB 1.0 15 (5/33)81 13% vs. control
NBR 1.0 5 (213 7)“ 69% vs. control
0

NBEM 0.0 0 (0/39)b 100 /o vs. control &

100% vs. NB

 

a. Tumors include: adenomas and adenocarcinomas. b. Control = Control diet; NB =
Navy bean diet, NBR = Navy bean residue diet, NBEM = Navy bean extract mix. 0.
Values in a column are significantly different if they don’t share the same superscript (P
< 0.05). d. In parenthesis is the number of animals with tumors per number of mice

alive at the end of the experiment.

138

(2/32) and similar to the control and NB diets (P>0.05). No adenomas or
adenocarcinomas were present in the colons of mice fed the NBEM diet and mice fed the
NBEM diet had a significantly (P<0.05) reduced tumor incidence compared to the control

and NB diets.

E. DISCUSSION

The goal of producing an adenoma incidence 2 50% and an adenocarcinoma
incidence 2 20% in obese, mildly diabetic ob/ob mice (B6.V-LepOb/J mice) fed a control

diet was not achieved. The mice fed the control diet had only 10% adenoma incidence
and 10% adenocarcinoma incidence. The low tumor incidence seriously compromises
the ability to distinguish dietary induced differences. Increasing the AOM dosage would
increase tumor incidence, but the trade-off would be increased mortality. A second
approach would be to use earlier end-points such as ACF, B-catenin accumulating crypts
(BCAC), or mucin depleted foci (MDF) that occur prior to dysplastic and hyperplastic
lesions and tumors.

End-points that occur prior to the appearance of tumors have been used in short-
term chemically induced colon carcinogenesis experiments. One widely accepted
biomarker often used to screen for potential chemopreventive agents as well as
carcinogens prior to long-term studies is ACF (Roncucci et al., 1991; Femia and Caderni,
2008). ACF, early preneoplastic lesions, have been observed in the colon of rodents
injected with colon carcinogens as well as in the colons of humans with hereditary forms
(i.e. familial adenomatous polyposis) and spontaneous CRC (Bird, 1989; Roncucci et al.,

1991; Pretlow et al., 1991; Yokota et al., 1997; Schoen et al., 2008). More than 98% of

139

ACF never progress to macroscopically visible lesions and the correlation between ACF
and colorectal tumors in chemically induced animal models and humans have been
inconsistent (Davies and Rumsby, 1998; Wijnands et al., 2004; Mori et al., 2005; Gupta
and Schoen, 2008; Schoen et al., 2008). These two criticisms seriously weaken the
predictive value of ACF for the development of colorectal tumors. However, papers
using ACF as the end-point for assessing chemoprevention continue to be published.
Other preneoplastic lesions have been identified and suggested to correlate better with
tumor development include BCAC (Y amada et al., 2001) and MDF (Caderni et al.,
2003). BCAC are technically very difficult to detect and only a few labs in the world
have successfully utilized this model. The MDF end-point has not been fully evaluated
regarding the potential for predicting tumor incidence.

When evaluating the efficacy of chemopreventive agents and/or dietary
components to prevent colon cancer, the occurrence of adenocarcinomas in the colon is a
good end point. In the AOM model of colon carcinogenesis, it appears that most
adenomas progress to adenocarcinomas with time. Additional considerations are that the
distribution of the intestinal tumors in AOM injected rodents has been found to be similar
to that exhibited in humans, with large intestinal tumors generally found in the distal
colon (corresponding to the lefi side or descending and sigmoid colon in humans) (Ward
et al., 1973; Zedeck, 1980). Also, histopathological characteristics and genetic alterations
identified in sporadic colon tumors in humans have also been identified in AOM-induced
tumors in rodents (Druckrey, 1973; Ward et al., 1973; Reddy, 2004). Therefore, the best

available end-point is adenocarcinoma or adenoma plus adenocarcinoma.

140

Prior to the current study, our lab administered six weekly injections of AOM (15
mg per kg body weight; based on the weights of age matched lean C57BL/6J mice) to
male and female ob/ob mice. This was equivalent to AOM doses ranging from 5.5-6.0
mg and 6.7-8.6 mg per kg of body weight in female and male ob/ob mice, respectively.
This AOM dosing regimen was based on previous research (Guo et al., 2004; Hirose et
al., 2004). The mice were fed the same 4 dietary treatments used in the current study
throughout AOM administration until the end of the experiment. Unexpectedly, 80% of
the male mice fed the control diet died before the experiment ended, the first dying

approximately 4 days after the 3rd AOM injection. When the experiment was terminated,

the livers from the remaining males fed the control diet appeared a great deal smaller than
those from mice in the other dietary groups. Although the female mice fed the control
diet survived until the end of the experiment, when sacrificed roughly 65% had abnormal
appearing livers. They contained nodules throughout, a common characteristic of
cirrhosis. It was concluded that the male ob/ob mice fed the control diet died fiom AOM
induced liver failure. The females most likely survived the toxic effects of AOM because
they received a smaller dose.

The mouse model used in the previous and current experiments (ob/ob mice)
exhibit impaired hepatic repair and regenerative responses to injury (Yang et al., 2001;
Leclercq et al., 2003) and increased sensitivity to endotoxin hepatic injury (Yang et al.,
1997). In order to avoid the hepatotoxic effects of AOM observed in the preliminary
study, the number of AOM injections was reduced to two (7 mg per kg of body weight;

based on the average weight of ob/ob mice in the study). However beginning sixteen

weeks after the 2"d AOM injection until the end of the study (27 to 29 weeks after the 2nd

141

AOM injection), 9 animals spontaneously died. The hepatotoxic effects of AOM have
been demonstrated, for example a single high dose (ranging from 30-200 mg AOM per
kg of body weight) induced fillminant hepatic liver failure (FHF) in C57BL/6J mice, the
background strain of ob/ob mice (B6.V-Lep0b /J mice), in a dose-dependent manner

(Matkowsky et al., 1999). It is probable that ob/ob mice are more susceptible to the
hepatotoxic effects of AOM even when administered at subtoxic doses than C57BL/6J
mice and the consequences of AOM hepatotoxicity manifest before colon tumors can
fully develop.

The statistical power in this study was less than optimal because of the
hepatotoxic effects of AOM in ob/ob mice and because of the low adenoma and
adenocarcinoma incidence in control fed ob/ob mice. For example, if there was one less
lesion in the NB group, then feeding NB would have been considered to have reduced
lesion incidence (P<0.05). In addition, the NBR would have to have zero adenomas +
adenocamcinomas to have significantly less tumors than the controls (i.e., just one
adenoma or adenocarcinoma in the NBR group made it so that the NBR was not
significantly different from the mice fed the control diet (P>0.05)). These points
underscore the importance in having a higher tumor incidence in experiments designed to

show a cancer inhibiting effect of diets.

In the current study, the animals thrived on all four dietary treatments: control
diet, navy bean diet or navy bean fiaction diets (NBR and NBEM). However, the ob/ob
mice fed the control diet gained less weight than those fed the whole NB diet and NBR
diets. The ob/ob mice fed the NBEM diet also gained less weight than mice fed the NB

diet or the NBR diets. Investigators have previously shown energy restriction is

142

associated with increased inhibition of chemically induced colon carcinogenesis. Kumar
et al. (1990) reported feeding rats a 20% and 30% calorie restricted diet (resulting in 23%
and 38% reduction in total weight gain respectively) significantly inhibited AOM-
induced colon tumor incidence in rats by 34% and 39%, respectively. Whereas feeding a
10% calorie restricted diet (16% reduction in total weight gain) did not significantly
reduce tumor incidence (Kumar et al., 1990). It is unlikely that the differences in weight
influenced tumor incidence in the present study for the following reasons. The ob/ob
mice fed the control diet gained 20% and 11% less weight than those fed the whole NB
diet and NBR diets, yet the control mice had the highest lesion (40% vs. 21% or 14%)
and tumor incidence (18% vs. 15% or 5%). Additionally, the ob/ob mice fed the NBEM
diet gained 15% and 5% less weight than those fed the whole NB diet and NBR diet,
respectively. However, these animals had the lowest lesion and tumor incidence. In this
study if a reduction in weight gain affected lesion and tumor incidence, the lesion and
tumor incidences in the control diet fed group and the NBEM diet fed group would have
been similar. Additionally ob/ob mice are overtly obese by 4 to 5 weeks of age compared
to their lean littermates, thus they were grossly obese before the beginning of the study
and all weighed 2 to 3 times more than an average lean mouse at the end of the study.
Thus any differences found in weight are negligible and had no consequence on tumor

outcome.

The second objective of this study was to determine if feeding diets containing a)
navy beans (Phaseolus vulgaris L.); b) an aqueous-ethanol extract of navy beans and c)
the residue remaining after aqueous-ethanol extraction would reduce adenoma and

adenocarcinoma incidences in AOM injected ob/ob mice. Due to the limitations of the

143

 

ob/ob mouse model discussed above, it could not be ascertained if feeding navy beans
inhibited lesion and tumor incidence as reported previously in a rat model (Hangen and
Bennink, 2002).

All three navy bean diets were designed to contain the same amount of dietary
fiber to eliminate its protective affect on chemically induced colon cancer. The NB and
NBEM were formulated to be equal in total phenolic content, 110.6 and 109.8 mg (+)-
catechin equivalents per 100 g of diet, respectively (Table 10). Both diets contained
approximately twice the amount found in the NBR diet. In Chapter IH the ~60% ethanol
extraction of the aqueous-alcohol soluble components in navy beans was more efficient at
extracting hydroxycinnamic acids conjugates than hydroxybenzoic acids conjugates.
Using the data from Chapter HI, the NB and NBEM diets were found to contain similar
amounts of hydroxycinnamic acids, whereas all three navy bean diets had similar
amounts of hydroxybenzoic acids. No flavan-3ols or flavonols were identified in navy
beans or its fi'actions. Besides being rich in phenolic acids, the aqueous-alcohol fiaction
of cooked navy beans (NBEM) would contain oligosaccharides, saponins and phytic
acids which were not quantified in this or the previous study. These non-nutrients have
been shown to possess anti-cancer activity (J enab and Thompson, 2000; Francis et al.,
2002; Guillon and Champ, 2002). The NB diet would contain these non-nutrients as
well. We expected the tumor incidence to be reduced by at least 50% in animals fed both
the NB diet and NBEM diet, similar to that observed previously in a rat model fed a navy
bean diet (Hangen and Bennink, 2002). Tumor incidence was expected to be reduced in
animals fed the NBR diet, but to a lesser extent compared to the other two navy bean

diets.

144

Feeding the aqueous-ethanol extract of navy beans (NBEM) reduced tumor
incidence (tumors include adenomas and adenocarcinomas) by 100% compared to
animals fed the control diet and the navy bean diet (P<0.05). Interestingly tumor
incidence was inhibited by 69% (not statistically significant) in mice fed the aqueous-
ethanol insoluble components of navy beans (NBR) compared to the control diet fed mice
The finding that feeding NBR and NBEM, were more protective against AOM-induced
colon carcinogenesis was unexpected. When bioactive molecules in beans are part of the
whole food matrix, the bioactivity may work together or synergistically, whereas others
may work against one another or antagonistically (Jacobs and Steffen, 2003). Thus the
biological activity of a particular compound(s) could be enhanced if it is separated from
an antagonist. Altemately, because of the weak statistical power in this study, it was not
possible to make a true differentiation in cancer inhibiting potential between the two navy
bean fractions—the aqueous-ethanol soluble components (NBEM) and the aqueous-
ethanol insoluble components (NBR).

In conclusion, the mouse model used in this study did not have the desired
sensitivity to AOM to be recommended for widespread use to test dietary ingredients for
colon cancer inhibiting potential. Nevertheless, the compounds contained in the aqueous-
ethanol extract of navy beans (NBEM) reduced tumor incidence by 100% compared to

the control diet.

145

 

CHAPTER V. DIETS CONTAINING NAVY BEAN (PHASEOLUS VULGARIS L.)
OR NAVY BEAN FRACTIONS INHIBIT AZOXYMETHANE (AOM)-INDUCED
COLON CANCER IN OB/OB MICE BY MODULATING THE TGF-B

SIGNALING PATHWAY

A. ABSTRACT

Understanding the molecular, biochemical, cellular and genetic alterations
occurring at different stages during colon cancer development may lead to
chemoprevention strategies, targets for chemotherapy drugs and early biomarkers
representing cancer risk. Diet is an important lifestyle factor that can be modified to
reduce colorectal cancer risk.

Food components not digested and absorbed in the upper gastrointestinal tract
come in direct contact with the colonic epithelial cells and may modify gene expression.
Previous studies have found that feeding dry beans reduces colon tumor incidence in
AOM injected rats by modulating gene expression involved in regulating colon crypt
cytokinetics and in reducing inflammation. Consistent with those findings, disruptions in
the regulation of crypt cytokinetics (i.e. cell proliferation, migration, differentiation, or
apoptosis) can contribute to the development of colorectal cancer.

Transforming growth factor (TGF-B) is a cytokine that controls cell division,
differentiation, migration, adhesion, extracellular matrix deposition and programmed cell
death (apoptosis) in many cell types. Deregulation in the TGF-[i signaling has been
shown to play a role in human colorectal carcinogenesis and animal models of colorectal

cancer. Since the biological actions and gene expression of TGF-B are consistent with

146

maintaining colon crypt homeostasis and since colon tumorigenesis is associated with
perturbations in colon crypt homeostasis, it was hypothesized that reduced colon
tumorigenesis observed when navy bean diets were fed would be associated with
normalizing expression of genes involved in the TGF-B signaling pathway.

The results in this study suggest that feeding diets containing cooked navy bean
and navy bean fractions — NBR and NBEM — inhibit AOM induced colon
carcinogenesis in obese, mildly diabetic mice (ob/ob mice (B6.V-Lep0b /J)) by modifying
the expression of genes involved in the TGF-B signaling pathway. A downregulation in
the gene expression of c-fos was observed in the colonic mucosa of bean fed mice and
this was related to reductions in colon lesion incidence. c-Fos is required for normal cell
cycle progression. Studies have found c-fos to be upregulated in some human cancers.
Thus feeding navy beans and its fractions to AOM injected ob/ob obese mice inhibits
colon carcinogenesis by maintaining colonic crypt cell homeostasis by modulating

cytostatic processes involved in preventing uncontrolled cell proliferation.

147

B. INTRODUCTION

Carcinogenesis is a multi-step process wherein a series of genetic events are
necessary to drive the transformation of normal cells to malignant cancers (Hanahan and
Weinberg, 2000). Understanding the genetic, molecular, cellular and biochemical
alterations occurring at the different steps in this process may lead to chemoprevention
strategies through diet, the discovery of targets for chemotherapy drugs and early
biomarkers representing cancer risk. Diet is an important lifestyle factor that can be
modified to reduce CRC risk. Diet can alter metabolism and blood constituents (i.e.,
hormones, growth factors) that in turn alter carcinogenesis. In addition, food components
not digested and absorbed in the upper gastrointestinal tract come in direct contact with
the colonic epithelial cells. Because colon carcinogenesis is a multi-step process
requiring a number of molecular alterations to fully manifest, it is logical that dietary-
induced changes in metabolized and undigested food components might alter colon
cancer development.

The large intestine is made up millions of crypts, which are around 50 cells deep
(see Chapter 1, Figure 4) (Lipkin, 1999; Heavey et a1, 2004, Humphries and Wright,
2008). Colonic cells are under constant renewal, with the entire human colonic
epithelium being replaced every 3 to 8 days (Lipkin, 1999). Stem cells located at the
bottom of the crypt replicate resulting in the original stem cell and a new daughter cell.
The daughter cells continue to replicate and push upward toward the top of the crypt.
About two-thirds up the crypt, the proliferative activity ceases and cells differentiate into
mature epithelial cells. Once at the lumenal surface the mature cells undergo apoptosis

and are exfoliated into the fecal stream. The crypt design provides a high probability that

148

damaged or mutated cells will not influence the integrity of the crypt (Heavey et al.,
2004). Crypt cell kinetics or the molecular processes involved in maintaining the balance
between new cell production and cell death are under tight regulation in highly renewing
tissues like the colon (Lipkin, 1999). Disruptions in any one of these processes (cell
proliferation, migration, differentiation, or apoptosis) can contribute to the development
of colorectal cancer.

Rondini (2006) fed defatted soybean flour and black beans (75g per 100 g of diet)
to rats injected with AOM. Feeding soy reduced tumor incidence by 67% and feeding
black bean reduced tumor incidence by 56% compared to rats fed the casein diet. Based
on cDNA microarray technology, the consumption of beans inhibits AOM induced colon
turnori genesis by modulating the expression of genes involved in regulating colon crypt
cytokinetics as well as genes involved in reducing inflammation. The expression of
several genes involved in the cell’s progression through the cell cycle, were significantly
downregulated in the colonic mucosa of bean fed rats (Rondini, 2006). This is consistent
with the idea of food components coming in contact with the colonic epithelial cells,
inducing changes in the expression of genes responsible for maintaining crypt
cytokinetics, and consequently, altering colon cancer development.

Transforming grth factor (TGF-B) is a member of a large superfamily of
cytokines that controls cell division, differentiation, migration, adhesion, extracellular
matrix deposition and programmed cell death (apoptosis) in many cell types (Harradine
and Akhurst, 2006; Massagué et al., 2006). Whether TGF-B inhibits or stimulates cell
proliferation is dependent on cell type and the presence or absence of growth factors.

However it is recognized as a potent inhibitor of proliferation in epithelial, endothelial

149

 

and hematopoietic cells (Hartsough and Mulder, 1997). TGF-B inhibits the proliferation
of cells (Massagué et al., 2006; Seoane, 2006) primarily by preventing cell progression
through the G1 phase of the cell cycle. In the normal colonic crypt, the expression of
TGF-B increases from the bottom of the crypt to top, with the highest amounts expressed
at the lumenal surface where the mature, differentiated cells are located (Shao et al.,
1999). TGF-B expression is inversely related to proliferation. Since the biological
actions and gene expression of TGF-B are consistent with maintaining colon crypt
homeostasis and since colon tumorigenesis is associated with perturbations in colon crypt
homeostasis, it was hypothesized that reduced colon tumorigenesis observed when navy

bean diets were fed would be associated with normalizing expression of genes involved

in the TGF-B signaling pathway.

C. MATERIALS AND METHODS
Experimental animals

Described in Chapter IV, section C. Material and Methods.
Preparation of Navy Beans and Fractions
Preparation of the navy bean flour and the fi'actions derived from it is described in
Chapter III, C. Material and Methods.
Diet Preparation

Preparation of the four diets fed to the ob/ob mice; 1) control diet — modified
AIN-93G diet; 2) NB — cooked navy bean diet, 3) NBR -— navy bean residue diet; and
4) NBEM—navy bean extract mix diet are described in Chapter IV, C. Material and

Methods. The composition of each diet is listed in Table 7 (Chapter IV).

150

 

Initiation of Colon Cancer

Described in Chapter IV, C. Material and Methods.
Necropsy and Histology

Described in Chapter IV, C. Material and Methods.
RNA Isolation

RNA was isolated and purified fi'om frozen (-80°C) colonic mucosa scrapings of
the mid-colon segment from ob/ob mice (n = 3 per dietary treatment). The scraped
section had no macroscopic lesions or tumors.

The RNeasy Mini Kit (Qiagen, Valencia, CA) was used to isolate and purify total
RNA according to the manufacturer’s protocol. Briefly, a portion of colonic mucosa was
added to a 2m] microcentrifuge tube containing 1ml of glass beads (1.0 mm in diameter,
Biospec Products Inc., Bartleville, OK) and 1 ml of Buffer RLT. The tissue was
homogenized by mixing on a Mini-bead beater-l (Biospec Products Inc., Bartleville, OK)
for 20 sec at 46 rpm. After centrifuging at full speed for 3 min, the supernatant (lysate)
was transferred to a new 2ml microcentrifuge tube. One ml of 70% ethanol was added to
the lysate (700111) and mixed. Lysate + 70% ethanol mix (1700111) was transferred to an
RNeasy spin column, placed into a collection tube, centrifuged for 15 s at >8,000 x g and
the flow through was discarded. The previous step was repeated. Then 350111 of Buffer
RWl was added to the RNeasy spin column, centrifuged for 15 s at >8,000 x g and the
flow through was discarded. To prevent DNA contamination, 80111 of RNase free DNase
I incubation mix was added to the RNeasy spin column and incubated on the benchtop
(20-30°C) for 15 min. Buffer RWI (3 50111) was added to RNeasy spin column,

centrifuged for 15 s at >8,000 x g, and the flow through was discarded. The DNase I

151

digest step was performed 2 times. Buffer RPE (500111) was added to the RNeasy spin
column, centrifuged for 15 s at >8,000 x g, and the flow through was discarded. Buffer
RPE (500111) was again added to the RNeasy spin column, this time it was centrifuged for
2 nrin at >8,000 x g, and the flow through was discarded. The RNeasy spin column was
placed into a new collection tube and centrifuged for 1 min at full speed. The RNeasy
spin column was placed into a new 1.5m] collection tube, 25111 of RNase free water was
added and centrifilged for 1 min at >8,000 x g, and the cleaned up RNA was eluted. The
prior step was repeated and RNA eluates were combined.
Quantitative Real Time PCR

First strand cDNA synthesis was performed using the RT2 PCR Array First Strand
Kit (SABiosciences, Frederick, MD). Briefly, 2.511g of RNA plus 1111 of P2 (Primer and
External control mix) was added to a sterile PCR tube, mixed with a pipettor followed by
brief centrifugation, this was termed the annealing mixture. The annealing mixture was
placed into a thermocycler at 70°C for 3 nrin and then chilled on ice for 1 min. While the
annealing mixture was incubating the RT Cocktail (contains the reverse transcriptase)
was prepared according to the manufacturer’s protocol. 10111 of the RT Cocktail was
added to the chilled annealing mixture, mixed well with a pippettor and incubated at 37°C
for 60 min in the thermacycler. The cDNA synthesis reaction (annealing mixture plus
RT Cocktail) was heated to 95°C for 5 min. Then 91111 of ddHZO was added to the PCR
tube, the tube was mixed and placed on ice until Real Time-PCR was performed.

Prior to Real Time-PCR, the Experimental Cocktail was prepared following the
manufacturer’s protocol. Briefly, 1325111 RT2 Real Time" SYBR-Green / F luroescein

PCR master mix (contains the polymerase, SABiosciences, Frederick, MD), 1219111 of

152

ddeO and 106111 first strand cDNA synthesis reaction were added to a 10 ml sterile
conical tube and gently mixed using a pipettor. The Experimental Cocktail was then
carefully poured into a reservoir. Using an eight-channel pipettor, 25111 of the
Experimental Cocktail was added to the 96 well RT2 ProfilerTM PCR Array: Mouse TGF-
13/ BMP Signaling Pathway (SABiosciences, Frederick, MD), 3 plates per dietary
treatment; 1 animal per plate. The RT2 ProfilerTM PCR Array: Mouse TGF-B / BMP

Signaling Pathway profiles the expression of 84 genes related to the TGF-B / BMP-
mediated signal transduction and contains a 96 well primer set. Each plate contains a
genomic DNA contamination control, Reverse Transcription control, Positive PCR

control and Five housekeeping genes. Real-Time PCR was performed using the Bio-Rad
iQm5 Real-Time PCR Detection system (Bio-Rad, Hercules, CA). Real-time conditions

were step as follows: Initial enzyme (Hotstart DNA polymerase) activation step of 10
min at 95°C followed by 40 cycles of denaturation/melting for 15 sec at 95°C followed by
annealing / extension for 1 min at 60°C (Real time PCR step). A melting curve was
generated by decreasing the set point temperature 95°C to 55°C over 1 min, followed by
81 cycles for 30 sec at 55°C.
Statistical Analysis and Biological Significance Criteria

Excel based Data Analysis Template was downloaded from the SABiosciences

Website (http://www.sabiosciences.corrr/pcr/arravanalysis.php). Differences in
normalized gene expression — 2°30, (Ct is threshold cycle; AC1 = Ct(GOI) — Ct (HK);
GOI = gene of interest, HK = housekeeping gene) — due to dietary treatment were

determined using the Student’s 2 tailed t-test. Changes in gene expression were

153

considered statistically significant when P S 0.05. A strong trend for a changes in gene
expression was when P >0.05 but $0.15.

In addition to meeting statistical criteria, changes in gene expression had to reflect
lesion incidence and tumor incidence to be considered biologically significant. Thus,
gene expression important in reducing hyperplasia and dysplasia must be significantly
increased or decreased in mice fed NBEM and NBR and there must be a significant or a
strong trend for an increase or decrease in mice fed NB. To be considered relevant for
reducing tumor incidence, gene expression must be significantly increased or decreased
in mice fed NBEM and there must be a significant or strong trend for a difference in gene

expression for rrrice fed NBR

D. RESULTS

Total RNA was isolated and purified from the colonic mucosa with no grossly
visible lesions from three mice per dietary treatment. The RT2 ProfilerTM PCR Array:

Mouse TGF-B / BMP Signaling Pathway array includes the 84 genes listed in Table 13.
Out of the 84 genes only fos (FBJ osteosarcoma oncogene; D12Rfi l/c-fos) met
the pre-established criteria for statistical and biological significance for reducing
hyperplasia, dysplasia, and tumor development. The changes in gene expression were:
NB -3.10; NBR -2.15; and NBEM -3.97 compared to mice fed the control diet. All of
these changes were statistically significant (P<0.05).
Lefty 1 met the criteria for down regulation of tumor development. The gene

expression for Lefty 1 was significantly down regulated in mice fed NB (-1.47, P = 0.02)

154

 

Table 13. Genes in the Mouse TGF-fl-BMP Signaling Pathway PCR Array

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Symbol Gene Name Gene Description

Acvrl ALK2/ActR-I Activin A receptor, type 1

Acvr2a ActRIIa/Acvr2 Activin receptor HA

Acvrll A1115505/AI427544 Activin A receptor, type II-like l

Amh MIS Anti-Mullerian hormone

Amhr2 MISHR/Misrii Anti-Mullerian hormone type 2 receptor
Bambi 2610003H06Rik BMP “dh?£$§;fl;::‘?:fig)Mb1“”:
Bglap2 OG2/mOC-B Bone gamma-carboxyglutamate protein 2

Bmpl TLD Bone morpho genetic protein 1

Bmp2 A1467 020/Bmp2a Bone morpho genetic protein 2

Bmp3 9130206H07/9530029IO4Rik Bone morphogenetic protein 3

Bmp4 Bmp2b/Bmp2b-1 Bone morphogenetic protein 4

Bmp5 AU023399/se Bone morphogenetic protein 5

Bmp6 D13Wsu1 15e/Vgr-l Bone morpho genetic protein 6

Bmp7 OPl Bone morphogenetic protein 7

Bmper 31 1005 6H04Rik/CV-2 BMP-binding endothelial regulator
Bmprl a 111003 7122Rik/ALK3 Bone morpho genetic protein receptor, type 1A
Bmprlb A13 85617/ALK-6 Bone morpho genetic protein receptor, type 1B

 

Bmpr2 2610024H22Rik/AL117858 3°” m°rph9gem° ”0‘51“ 43°61’10” type H
(senne/threonrne klnase)

 

 

 

 

 

 

 

 

 

 

 

C d79a Ig-alpha/Iga CD79A antigen (migcigbbulin-associated

C dc2 5a D9E rt d393e Cell division cycle.2'5 homolog A (S.
cerevrsrae)

Cdknla CAP20/CDKI Cyclin-dependent kinase inhibitor lA (p21)
Cdkr12b AV083695/INK4b Cyclin-dependent kinase inhibitor ZB (p15)

Chrd Chd Chordin
Collal Colla-l/Cola-l Procollagen, type 1, alpha 1
Colla2 AA960264/AI325291 Procollagen, type 1, alpha 2
Col3al AW550625/Col3a-1 Procollagen, type III, alpha 1

Dlx2 AW121999/Dlx-2 Distal-less homeobox 2

Eng A1528660/CD105 Endoglin
Evil D630039M04Rik/Evi-l Ecotropic viral integration site 1

 

155

Table 13. (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fkbplb AW494148 FK506 binding protein 1b
Fos D12Rfi l/c-fos FBJ osteosarcoma oncogene
Fst FST Follistatin
Gdfl AI385651/Gdf-l Growth differentiation factor 1
Gdf2 Bmp9 Grth differentiation factor 2
Gdf3 C78318/Gdf-3 Growth differentiation factor 3
Gde CDMP-l/bp Grth differentiation factor 5
Gdf6 BMPl3/GDF 16 Growth differentiation factor 6
Gdf7 BMP12 Growth differentiation factor 7
Gsc GSC Goosecoid
Idl AI323524/DZWsul40e Inhibitor of DNA binding 1
Id2 A1255428/C7 8922 Inhibitor of DNA binding 2
Igfl C730016P09Rik/Igf-1 Insulin-like growth factor 1
Igfbp3 AI649005/IGFBP-3 Insulin-like growth factor binding protein 3
116 Il-6 Interleukin 6
Inha AW555078 Inhibin alpha
Inhba INHBA Inhibin beta-A
Inhbb INHBB Inhibin beta-B
Itgb5 AA475909/AI874634 Integrin beta 5
Itgb7 Ly69 Integrin beta 7
Jun AP-l/Junc Jun oncogene
J unb JUNB J un-B oncogene
Leflyl AI450052/Leftb Lefi right determination factor 1
Ltbpl 9430031G15Rik/9830146M04 Latent mmmmigg‘l’f‘l fact“ beta binding
L tbp2 sz 08 642 Latent transforming3 5:331:13 factor beta binding
Ltb p4 2310046A13Rik Latent transformlnigtcém factor beta brndmg
Myc AU016757/Myc2 Myelocytomatosis oncogene
Nbll D 4H1 S 1 73 3E /D AN Neuroblastoma, supprelssion of tumorigenicity
Nodal Tg.413d Nodal
Nog NOG Noggin
1 56

Table 13. (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

N rObl AHX/Ahc Nuclear receptor subfamily 0, group B,
member 1
Pdgfb PDGF-B/Sis Platelet derived growth factor, B polypeptide
Plat AU020998/AW212668 Plasminogen activator, tissue
Plau u-PA/uPA Plasminogen activator, urokinase
Runxl AI462102/AML1 Runt related transcription factor 1
Serpinel P AL] /P All Serine (or cysteigefirgtéfilasie inhibitor, clade
Smadl A1528653/Madh1 MAD homolog 1 (Drosophila)
Smad2 Madh2/Madr2 MAD homolog 2 (Drosophila)
Smad3 AU022421/Madh3 MAD homolog 3 (Drosophila)
Smad4 D18Wsu70e/DPC4 MAD homolog 4 (Dros0phila)
Smad5 11 1005 1M15Rik/AI451355 MAD homolog 5 (Drosophila)
Smurfl 4930431El ORik/mKLAAl625 SMAD specific E3 ubiquitin protein ligase l
Sox4 AA682046/Sox-4 SRY-box containing gene 4
S tatl 201 000 5 J02Rik/ AA 4081 97 Signal transducer and alctivator of transcription
ngfl CRl/cripto Teratocarcinoma-derived growth factor
Tgfbl TGF-betal/Tgfb Transforming growth factor, beta 1
Tgfblil ARA55/Hic5 Transfornung grtoaulgtilfizcttqr beta 1 1nduced
Tsc22dl AA589566/AW105905 TSC22 domain family, member 1
Tgfb2 BB105277/Tgf-beta2 Transfonning grth factor, beta 2
Tgfb3 Tgfb-3 Transforming growth factor, beta 3
Tgfbi 68kDa/A1181842 Transforming grth factor, beta induced
Tgfbrl ALK5/AU017191 Transforming grth factor, beta receptor I
Tgfbr2 1110020H15Rik/AU042018 Transforming growth factor, beta receptor 11
Tgfbr3 1110036H20Rik/AU015626 Transforming grth factor, beta receptor III
Tgfbrapl 3110018K12Rik/AU024090 Transforming growth factor, beta receptor

associated protein

157

 

and NBEM (-l .42, P = 0.01). There was only a trend for down regulation of Lefiy 1 (-

0.24, P = 0.11) in mice fed NBR.

E. DISCUSSION

The biological actions and gene expression of TGF-B, a potent inhibitor of
epithelial cell growth, are consistent with maintaining colon crypt homeostasis. Since
colon tumorigenesis is associated with perturbations in colon crypt homeostasis the
objective of this study was to determine if the inhibition of colon carcinogenesis observed
in AOM-inj ected obese, mildly diabetic mice fed navy bean and navy bean fiactions —
NBR and NBEM —is associated with normalizing the expression of genes in the TGF-B
signaling pathway. Feeding navy beans and its fractions downregulated the expression of
c-fos mRNA (2 to 4 fold) compared to control fed ob/ob mice. These changes in gene
expression were correlated with reductions in colon lesion incidence (see Chapter IV).

c-Fos, a direct target of TGF-B, is a proto-oncogene and member of the Fos
family of transcription factors (in humans: c-Fos, FosB, Fra-l and Fra-2). Normally it is
not highly expressed (Milde-Langosch, 2005), but can be transiently and rapidly induced
by a variety of stimuli including growth factors, cytokines, TPA, and UV irradiation
(Angel and Karin, 1991). Fos proteins heterodimerize with members of the Jun family of
transcription factors (in humans: c-Jun, J unB, erD) to form the protein complex:
activator protein-l (AP-l). This complex can bind TPA-responsive elements (TRE) and
other similar AP-l like binding motifs, within the promoter region of target genes. AP-l
responsive genes are involved in the regulation of cell proliferation, differentiation and

apoptosis; angiogenesis; hypoxia; invasion and metastasis (Busslinger and Bergers, 1994;

158

Milde-Langosch, 2005). The partnering between different J un—J un or J un—Fos protein
homodimers or heterodimers, respectively, in the AP-l complex determines the AP-
ltarget gene response and are cell type specific (Angel and Karin, 1991). Fos proteins
are required for normal cell cycle progression and entry into S phase (Kovary and Bravo,
1991), yet an overexpression in c-fos has been associated with tumorigenesis and
correlated with the expression of several cell cycle regulatory proteins in different cancer

cells (Bamberger et al., 2001; Milde-Langosch et al., 2002; Milde-Langosch, 2005). In

 

cervical carcinoma cells, the c-Fos protein was found to replace Fra-l in AP-l dimers, i
and this shifi in expression was positively associated with malignancy (Soto et al., 2000).
Additionally, the expression of c-fos is required for the progression of mouse skin
papillomas to malignant tumors (Saez et al., 1995). These experimental studies suggest a
role for c-fos in the transformation of normal cells to malignant cancers in some tissues.
Furthermore, in the colons of AOM injected rats, ACF were found to express
higher levels of c-fos (mRNA and protein) than normal crypts (Stopera et al., 1992). The
expression of c-fos mRNA followed the pattern of cell proliferation in colonic crypts,
with expression being greatest in the lower compartment of the crypt where highly
proliferating cells are located (Stopera et al., 1992). Additionally, the advanced lesions,
ACF exhibiting dyplasia, expressed higher levels than early lesions exhibiting
hyperplastic (Stopera et al., 1992). This evidence suggests that an overexpression of c-
fos may contribute to the increased cell proliferation observed in early and advanced
colonic lesions, which may help promote their progression to more invasive, malignant
tumors. Thus c-fos overexpression may represent an early genetic change occurring in

AOM induced colon carcinogenesis.

159

Feeding navy bean and navy bean fractions —— NBR and NBEM was found to
inhibit AOM induced colon carcinogenesis in ob/ob mice and this inhibition was
accompanied by a downregulation in the expression c-fos in the colonic mucosa. In
normal cells c-fos expression is required for cell cycle progression and cell proliferation,
however its overexpression is associated with tumorigenesis. Thus the downregulation of
c-fos observed in the colonic mucosa of navy bean and navy bean fiaction fed AOM

injected ob/ob mice play a role in maintaining normal crypt cell proliferation.

 

Additionally, feeding navy bean and it fractions downregulated the expression of I
Lefty 1 mRNA in the colonic mucosa of AOM injected ob/ob mice compared to the
control fed group. This change in gene expression was only significant in animals fed the
NBEM and navy bean diet, but not rrrice fed the NBR. Lefty negatively regulates
different TGF-B family members i.e. TGF-Bs, BMPs and nodal, and plays a critical role
in implantation and embryo genesis (U lloa and Tabibzadeh, 2001; Tabibzadeh and
Hemmati-Brivanlou, 2006). It blocks TGF-B signaling by antagonizing pathway specific
co-receptors (Cheng et al., 2004) and by inhibiting the phosphorylation and activation of
receptor activated SMAD transcription factors (U lloa and Tabibzadeh, 2001). Many
epithelial derived tumors (285% of all human cancers) and several human colon
carcinoma cell lines no longer respond to the antigrowth signals induced by TGF-B
(Hoosein et al., 1989; Hsu et al., 1994; Elliott and Blobe, 2005) and an overexpression in
Lefty has been observed in certain forms of human cancer, including colon cancer (Ulloa
and Tabibzadeh, 2001). Thus an overexpression in Lefiy 1 may assist initiated cells in
evading growth inhibition, a recognized hallmark of cancer (Hanahan and Weinburg,

2000), and provide them with a selective growth advantage compared to surrounding

160

normal cells. Thus the downregulation in Lefty 1 may preserve epithelial cell sensitivity
to TGF-B antigrowth effects and maintain colon crypt cytokinetics in navy bean and navy
bean fraction fed AOM injected ob/ob mice.

In conclusion, feeding diets containing cooked navy bean and navy bean fractions
— NBR and NBEM — inhibits AOM induced colon carcinogenesis in mildly diabetic,
obese mice (ob/ob) by modifying the expression of genes involved in the TGF-B L
signaling pathway. In navy bean and navy bean fraction fed mice, a significant

downregulation in the expression of c-fos mRNA was found in the colonic mucosa of

 

AOM injected ob/ob mice when compared to the control diet fed mice and these changes
correlated with the observed reductions in colon lesion incidence. This change in gene
expression may prevent colonic epithelial cells fi'om gaining the ability to proliferate
without regulation, an early change found to occur in colon tumorigenesis (Lipkin, 1983)
and recognized hallmark of cancer. Thus feeding navy beans and its fractions to AOM
injected ob/ob obese mice inhibits colon carcinogenesis by maintaining colonic crypt cell
homeostasis by modulating cytostatic processes involved in preventing uncontrolled cell

proliferation.

161

CHAPTER VI. CONCLUSION

The consumption of dry beans has been associated with the inhibition of colon
cancer and phenolic compounds in dry beans could be partly responsible for the observed
protection. Dry beans must undergo some form of heat-treatment before they are safe for
human consumption and heat treatments will induce physical and chemical compositional
changes in beans, altering their nutrient and phenolic content. Thus it is important to
determine the phenolic compounds and content in cooked common beans in order to
further investigate their potential health benefits. In Chapter III, we identified several
classes of phenolic compounds in the aqueous-alcohol extracts of four cooked dry
common beans (Phaseolus vulgaris L.)—black, pinto, red, and navy beans— as well as
in two fiactions of cooked navy beans 1) navy bean extract (NBEM)—aqueous-ethanol
soluble components (concentrated in phenolic compounds) and 2) navy bean residue
(NBRy—aqueous-ethanol insoluble solid components (i.e. dietary fiber and protein),
treated with and without acid and alkaline hydrolysis. Hydroxybenzoic acids and
hydroxycinnamic acids were identified in all four cooked dry common beans, this
included p-hydroxybenzoic acid, vanillic acid, and syringic acid. Protocatechuic acid
was only identified in black, pinto, and red beans. The hydroxycinnamic acids identified
in all beans were p-coumaric acid, caffeic acid, ferulic acid, and sinapic acid. Ferulic
acid was the predominant hydroxycinnamic acid in all four bean types. In general, acid
hydrolysis liberated greater quantities of hydroxybenzoic acids while alkaline hydrolysis
liberated greater quantities of hydroxycinnamic acids. Only black, pinto, and red beans

contained the flavan-3ol (+)-catechin and the flavonols quercetin and kaempferol. The

162

 

partitioning of cooked navy beans into two fractions—NBEM and NBR—using aqueous-
ethanol extraction (~60% ethanol) was more efficient in extracting hydroxycinnamic acid
conjugates than hydroxybenzoic acid conjugates.

Identifying which components in dry common beans protect against the
development of CRC and the molecular mechanism by which they exert these biological
activities are of high priority. Partitioning beans into different fractions and determining L
the anti-cancer activity for the various fractions is one approach to narrowing the search

for the anti-cancer components of beans. In our lab navy beans have been shown to

 

inhibit chemically induced colon cancer in rats (Hangen and Bennink, 2002). Thus in 5
Chapter IV, we fed diets containing cooked navy beans and its fractions—the aqueous-
ethanol soluble components in cooked navy beans (NBEM) and the aqueous-ethanol
insoluble components (NBR), or a control diet to AOM injected mildly diabetic, obese
mouse (ob/ob; B6.V-Lep°b /J) to determine whether one fraction inhibits colon cancer
more than another. The ob/ob obese mouse model has not been previously used to
evaluate the effect of dietary ingredients on chemically induced colon carcinogenesis.
Obesity and type 2 diabetes mellitus are independent risk factors for CRC and dry beans
have been associated with preventing both conditions. We concluded that the ob/ob
mouse does not have the desired sensitivity to the colon carcinogenic properties of AOM
to be recommended for widespread use to test dietary ingredients for colon tumor
inhibiting potential. Ob/ob mice are susceptible to the hepatotoxic effects of AOM even
when administered at subtoxic doses and the consequences of AOM hepatotoxicity
manifest before colon tumors can fully develop. The low tumor (adenoma plus

adenocarcinoma) incidence in the ob/ob mice fed the control diet seriously compromised

163

the ability to distinguish dietary induced differences. Nevertheless, the NBEM fiaction
significantly reduced tumor incidence by 100%. The NBR reduced tumor incidence by
69%, however it was not significant, further demonstrating the weak statistical power in
this experiment.

Experimental studies have shown that feeding bean based diets to rats injected
with AOM can reduce colon tumor incidence by modulating the expression of genes
involved in regulating colon crypt cytokinetics and in reducing inflammation (Rondini,
2006). In Chapter V we investigated whether the reduction in colon tumori genesis
observed when navy bean and its fiactions are fed to AOM-inj ected ob/ob mice was
associated with normalizing the expression of genes involved in the TGF-B signaling
pathway. The biological actions and gene expression of TGF-B are consistent with
maintaining colon crypt homeostasis and disruptions in colon crypt homeostasis can
contribute to the development of colorectal cancer. The results in Chapter V suggest that
feeding diets containing cooked navy bean and navy bean fractions ——NBR and NBEM—
inhibit AOM induced colon carcinogenesis in mildly diabetic, obese mice by modifying
the expression of genes involved in the TGF-B signaling pathway. A downregulation in
the gene expression of c-fos was observed in the colonic mucosa of bean fed mice and
this reflected the reductions in colon lesion incidence. c-Fos is required for normal cell
cycle progression and studies have found it to be upregulated in some human cancers.
Thus feeding navy beans to AOM injected ob/ob obese mice may inhibit colon
tumorigenesis by inhibiting uncontrolled proliferation in colonic epithelial cells.

Collectively, Chapter III was one of the first studies to identify and quantify

phenolic compounds in cooked common beans and to separate cooked navy beans into

164

two fi'actions 1) the aqueous-ethanol soluble components (NBEM; rich in phenolic
compounds) and 2) the aqueous-ethanol insoluble solid components (NBR), to allow
long-term colon cancer studies to be conducted in order to narrow the search for the anti-
cancer components in beans. In Chapter IV it was determined that the mildly diabetic,

obese mouse (ob/ob mice; B6.V-Lep°b /J) did not have the desired sensitivity to AOM to

be recommended for widespread use to test dietary ingredients for colon cancer inhibiting
potential. Nevertheless, feeding cooked navy bean and navy bean fractions — NBR and
NBEM — inhibited AOM induced colon carcinogenesis in mildly diabetic, obese mice
(ob/ob) by maintaining colonic crypt cell homeostasis by modulating cytostatic processes

involved in preventing uncontrolled cell proliferation (Chapter V).

165

CHAPTER VII. FUTURE STUDIES

I would recommend the following studies to firrther investigate and narrow the

search for the anti-cancer components of cooked dry common beans and the possible

mechanisms responsible for the inhibition of colon cancer.

Study 1:

1.

Using more sensitive detection systems, i.e. MS/MS and NMR, in an attempt to
identify unknown phenolic compounds in cooked dry common beans.
Identify and quantify anthocyanins and their aglycones in cooked dry common

beans with blue and blue-violet colored seed coats, e. g. black beans.

. Quantify the other non-nutrients found in cooked navy beans and in the navy bean

fi'actions -— aqueous-alcohol soluble components in cooked navy beans (NBEM)
and in the aqueous-alcohol insoluble components in cooked navy beans (NBR)—
witlr proposed anti-cancer properties such as: oligosaccharides, phytic acid, and

saponins.

Study 2:

1.

Repeat the study performed in Chapter IV using AOM injected rats, an accepted
model of chemically induced colon carcinogenesis. Additionally, a fifth diet
should be added to the study. This diet would consist of adding the NBEM
fraction (concentrated aqueous-alcohol extract mix) back to the NBR fraction in
an attempt to mimic the phenolic content in the cooked whole navy bean diet.
This fifth diet would be another way to help distinguish which compounds in dry

common beans possess anti-cancer properties.

166

 

Study 3:

l.

2.

Determine if c-fos protein expression levels corroborate the changes in c-fos
mRNA expression observed in the colonic mucosa of AOM injected ob/ob mice
fed the navy bean and navy bean fraction diets—NBR and NBEM.

Determine the location of c-fos protein within the colonic crypts and whether their
expression correlates with cell proliferation patterns in the colon crypts.
Determine if these changes in mRN A expression also occur in AOM injected rats
fed the cooked navy bean and navy bean fraction diets—NBR and NBEM— and

the fifth diet discussed above (under Study 2).

167

APPENDICES

168

 

.35883 5505:... me 933 8 teammate". flung: xenomnfioax .v— atop—25 .0 ”Bow £35m .<m ”Eon ”UL—Eon {a ”Bow 05:58
1% Jan ”Eon omwfibmm .bnm axon 20% HO ”Eon 3:18.» .55 HEofluaAi .30 ”Row amoueontmxohiai {mm—a ”Eon 32—08338; JAE
58m Jun—m— eoubeaehateez he EfiweuaEeEU .2: 95w:—

Amflsfifiv 0:5. 55:83—

 

 

 

 

 

 

 

 

 

APPENDIX A

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

owe can 2:. - o8 - QR
- 1.3
.+. -11-,LL11 11 11 11 . MMM 9:. 8%.
H 1, 11-1 1 1 11 1 98 cows;
1 MLO111 ._L_.1 1 FE am.
L..- 2..., L. 1 1. 1 11 L- 1 $8 Sada
.>_,. L1211 1 1.1 .L 1 1 11 111 1 11 1 1.. L.)
L L _ . H>E coo
,L L. 11.11! 1111111 1 .11 : w
1111---. -1111 1 L .11 U
L L 9:. 8a a
L .3. M
L L
E .3

 

 

 

<Uo

 

 

 

 

 

169

.mvczomfiou =Bo§== mo 8.qu S vcommoboo Eon—SE :EoEEouv— .M 6:023: .O ”Boa £92.? n<m ”Eon 0:20.“ .<1._ ”Eon urns—=8
1% {Um L28 oLwESm chm “Eon Bots uao ”Eon 0:253 .53 EEoBaoAi duo ”Bow fiend—onhxeaai {min ”Boa 032088805 .<Um
:aom gun—m Bum—PG»: 255—: «c EauucuanEU £3 PEME

$8255 25.—1 53:33—

o.oo odm o.ov 0.0m CAN
1 1 1 1 i 11. 1 11 111 111 1 ,od

f 1 1111 1 1 1 L1 L H L: L » 45L 19E oomLL
L _,

L L: c L m Lmu N 9191 82L .
L 111 1 11 1 1 1 1 11 a 1 1 L L 11 1 1 1 Low
L M L , ,, 38 8;
1 1 11111 ; , L 1 1 1 . 1 L 1 1 9882,
, 1 L 1 1916 83
1 91:. co;

w 1 L
L: 1 1 1 1 111 2L» .111 1 L 9883

. 5 oo

of

 

(vd) ma

01m

<m

170

 

 

$25888 .5655. mo 9.3a 8 388:3 Eon—Ea: "—0.62808— .M £30226 .0 ”Eon £35m .<m ”Eon 3:on .<...~ ”Eon 23:58
1% .<Ua axon 2mg .Sm axon omobso Mao ”Boa 02:5.» .53 m55§u1+v .80 U28 eBay—oaxohwhsfi 4mm; ”28 03533805 .<Om
53m :3:— gnbeuemm 23.. mo E2u3afic£0 .2: BEE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

$223 as: 83.38
1 0.8 1 ob... . o9. . oven 1 98 1
ed
1K 11/ 11 111-fl?» 1111 U1 +1 14: «813U
11% 1111 11111111m11hy S 2 U71<udUJ91. - 1 1 1- H>E coma
[fl/J \4 1—1 11 3111/14 U \I1!\.1 * _..)..1-.1..1..1.1....- 11 1 11111111U 1111 E
1 1.11) M 10 111111112 1 .U U 111 o .1U- 1U 11,- - 1 1,1 :1, 8&3
U U U an}... U. 1 .U n a U - $8 81.2
7- 1 U 11111....UU 1111111111111 1 U 11111.17)\/.UU-11-..m111111Hw-U11J11U U-1-.,11.j8 o8 U
1111 11111 11111111! 1111; 1 .- , 1111,- 1-11U11 a
T U ._ 11ng2+ . ,1 1 a U. is 8: w
i 11 11 1111111111 11 1 11 1 111l 111111 1 111 111.1 1.111 11 1 11 - . U
U U <08 U . is 82 o v a
U <m §> $3 _ W
_ am U 1
UU n U
U <8 U
U U3
U U
U U
«Go U

 

 

171

£25888 568?: m0 838 8 3032.80 838:: scuomnEoS— J— ncmaoeosa .0 ”Bow 235m .<m "Eon 23.—om JG ”Eon
052.39% .<Ua “Boa aware? ..Sm 66a omobao £5 ”28 05:81» .§> EEoSaoAi .30 66m amongmxegn—i {min ”Eon oaaooaaoouoa {Um
50m 85m Bubcuuhm~1=cz no 83335930 A“: 95$..—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

asfié «HF 8:58
0.8 9.8 1 99 28 9.8
3
1'1 ([11 H 11,21—11|1J_1UUU<1111_11,111--1.UUU_..1s 111 11 AVE com
U8 a 111 E11 w U1 c 11..--.-.121.. 9:. com
TJUU11 1., ...-11..U__1.. 111112 9:.
. ,_ 8v
11,. U1 1111111 ,1 1-- 11 1 1 1 U , . 1 1 1 .3
U. ..r 1 1 - ..UUUUU U ....._.U. . . .-UU. . .UU< H>E 8M-
1|1(.. .U UNI. 111111 11111 1 I111111I11111 11 111 ,- , U /./\/.. ..1/x1 , ..UU,-11.1111.111...U ...1.
.U.., U. 1,. as...
111;...U. [llll 11|111111 11 '11 U 1 .11. ..1 1 . .
11 U UU 11 11 1 .3 1. . 121,111 .. - 119E111
. U ._ UU{ U U U a U. . 8:
U 1U U _ U U _ 171281in
U U U <m <08 m Si U
. UUU U. 80
U U m U. U
U UU E 5> UU
M __ U m U 5.0

(W) esuodsea

172

.mccsonEoo c3053: mo 832. 8 teammate... 83:5: schema—coax J— :fioouozc .0 "Bow 335m .<w ”Eon czar—om Iona ...—:3 3.85:8
1% {0; €an omwg .Sm ”Boa 0:58 ”EU ”28 o==§> .55 ”cicada-At .80 ”Bow oUQuaoabneanfi {mm-E €35 £20338an {on
53: 3:5 coma—EPA: 25% a: 8.233.550 ...: 35H..—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

AmoSEEV 0:51 33:33—
o..om 1 odm . odv . 0.0m o._o~
.1.-111111.00
., -\1-,_./11(111J\11\I(1\} .111.1.r.)()1\../ .1111-1111..11._ .11 11111.1)-..111111 9:.- cow/H.
1K..1U..1,,1-1. -11U1111111.
.1. U6 . <8 9:. 8n
111-111-21.111.11. .1111. 1 1 ....11 1 . SJ
11) UV. .1 1 11 111 11NUU11 111 ; COW rO.N
U 1. ..U ... , U 95 8a
. 1 11) 1-..U..1 11 111..U._ 1 1 11111 11 .1
U, U ,UU 9.: 83
1111 U.. 1. 111 H
a 11 1 a U 9:. 8: m
1 1111-....11-11 111 -1 -1 - . U
m U U 9.: 81 o v a
“a0 U W
U. 5> $3 (
hnm
U:

 

 

 

 

173

 

6258800 8592:: no 8.000 8 2532.50 3385: "—20.3803— .v— ”£000.03 .0 ”E00 030:? .<m ”Eon 023$ .<m
€60 058:00um {Um ”Bus 035% Sam €30 05% ”ED ”20a 3553 .§> EEoofioAi 400 630 0_0N=3>xo€>:§ {min ”22“ 0150030803 .<Um
53m 85.— 30226»: 20< a: 52333020 .0: PEME

 

fioBEEV 0:5. 5:503—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

o.38 98 3 0.3 98 EN
. od
_ 4 4 .33-,,_.33 33-3333. 33 3 3. . 3 3 33 - 3 3933:...
33 .. T3333- ..333333--.33333, 333,44 __3.---_.,3333 9333,... 33333- 9 .
4 2 4.. 4 >08 <8 9.: 82
.33 |333333 4433333333 330 333333 3 333 ... 333334: 4 33,333 333m .33333.333333%... .43 E ooldd
33333333.. 33333333 3333333] 33333333 33 3 33. 33333 3. 333 _ 3.3.33. 333 3 33 ..4._.3-..3-_3,
..4 4.. ,4. m _4 4 4.44 ...44 yr 4.9“ .7 .4. .4... CV8 OOWH
3333333.. .444 3333333333333 . _.. ...4 \3} 3 3333333333; 3333333333333.) .333 .33 _ "333333333 3.33 3.. ... 3333/ 33 f .333 33]
333.3,. ...... 4. 4 : .3 .4 -. 433. .4. 333. 3 3 4.33 33393382.
3|‘333333333l\3\/. gltll 442.4 .....3/\\3333/3\l (Ill- J3 ll|[3333(/3 I \l \ . ..‘l333ll L (3, 3; 3333)! 3133133314.!) 3333/. .llrll33, _ .33 {333’
_ - 3333333 3 333 33,... -33 333 33 3 33:3 39.3333 3.333 . -33... -3333 33 .
_7 _... _ _4 <m EU .. 9>E SE o v
_ _ :4 o m §> 4
4 _4 4 Km SE
_ n. E <0Q 4
4 _.. 4 v 4
_ _ 44‘ 4.4 4 .
4 t 4.4 ~
4
. 4
. x
_F <00
L

 

 

(v31) asuodsea

174

 

 

62:50:50 8505—5. .5 8.000 00 253850 Eon—=0: ”—05.5608— .M ”50000000 .0 ”200 050:? .<m ”200 0:300 in
660 oranooi £00 ”200 2mg Sam “20¢ 20% ”RU ”200 055.3 .§> mcfiofiaii .000 4200 23553.0‘ {man ”300 0330800055 {Um
50m 00m EBA—05.nfléoz ..0 80.33.5550 .02 0.53.”—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

?oBEEV 35. 002.0qu
0.00 0.8 0.0? 3 ofim odN
0.0
......3.4 b.
333 . 4 . 333333333443, 23.2332 - -3 -. _H.>E
. 4 com .
4. 33.. 433 333433. 3 4.
4 N4 4440 <00 9.... 8m.
3,. I 333... -.le3) 33]!) 333|3I 3333)} '33! 3° . N
.4 4 4 <w 4 47 > 98 02..
all, ,4... 4ft).-. .3333’llljr;33\l3\33 3 J (.\J/\. it 4.3/3, 33A73)\3I ..333/333.333 314333331 13333333333333.3313 4444133331333
44 ._ , o . ._ <m .. ____ :32. 82
333333 4433333333333. 333 33333333333333} 3 333.333 3 . ..4 .33... 4.333333. 3333, 3333333...4.4..33r3333 3333 33.333
44 44 . 34 4.3.4.... .
333) J 333333333! 3333 333.. J _ _ 3 ./ ,, .3 33 -...) 4 3.33333 3, 33333 3333
. . 44 3 44 44... 4. 4 . , <33 4 98 84.4
..- _. 3 3333333333 3 33 3 44.44. 4.4» 4 v 3 30 w .
4 .4 _ 4.4 3. a .4 §> #34 3333_>E com”— a
4 44 4 4 <4 <0 944m 4 . . M
4. <95 0 0
44 S v
.4 . h
4 44
4.4 44
.4 444 <00
. dd

 

 

 

175

 

.muqaonfioo E4651:— mo 8:30 8 24008430 €383: ”40.3.508— .v~ ”atop—0:0 .0 ”20¢ 0400:? .<m "Boa 0=Eom JE 4200 0435000
3% £50 ”28 035.54 dam €48 20kg .330 §0a 0===a> 60> E3083A+v .30 6400 20n§n>xo€>a3m “<9? 4200 04240030805 {Um
03m 00m tuna—9...»: 05—5544. H0 EafiSanEU £3 03mg

ABEEEV 0E:- messaged

 

com com 44.9. 08 com
3 3 - 3 33 3 -3 3 3 3 3. - 33 3 ..... - 13 - 3 - 49o
- - - - 34 43 3 4 WEN-4.4
44 N4441400 <0; 4>E 8&4
4. 4 -
3 - - 3 .44” - 3 3 3 3 3 - 3 3 - .. - -3 - - - - - - - 3 E3
3 3 3 43- 3 3 3 3 w .. b - 4, - 3 - -_-> 44444443
.. .44 O . .2: 834 6
3 3 ..- - 3 - 3 33-3 - - 4 - . - - -- 3 33 - 3 7
. . 4 4 95824 1
- - 4.4 - 3 I i .44 - - 3 - - 3 3. w
4.4 . 4. . __ - au- 4 9:. 024 m
- - .4 3 - - a §> - - 958444-43 444
4 .44 $4 44 w
<0n4 - S444 (
\- 5m
o_ o
<34 m .
a .34
Z

 

:4
<
C.)

0

50.50950 =3054§ .40 84.000 3 9.0002000 309:3: ”4000350004 4v— Etoohosc “O 6400 04004.46 8% 6400 04430.0 JG ”300 0:08:00
3% £00 6400 Emit? .3m 4200 30%—00 ago ”Eon 0:445; a§> EEoBaoAi J00 6400 omonconhxeanfi {mu-E 6400 0520800805 .<Om
gen 00% cons—0.6.»: E0< no Banana-H.050 .02 0.5»:—

Amsaifiv 2::- coucoaom
0.9 0.8 0.8
- - - 3 - - - - 336.0

 

- -- 1-344- -1 .-3 - a 3 :- a - - agegm
3 - 3 .3- .4. - - - 3 3 N4- - - -- -. 1 3.444440 3 --«w.434>&oo3m4°.~
4 - - I , - - -:- - 3-3 938$

00
'\O-

438-44..

-3 - --.o.v
.4 4 9583
3 44 -o - 444 440- - - - 4384-4.
- - - .4 - .4 - - 3 - 3 9584-43
4 <m <04

 

(wnmmw

ca
Sm > <20 4

0
In

od

[\
fl.

 

.o.ow

177

 

”Eon 0:530. .<m “200 0:30.“ {m ”Boa 0§o?m £09 ”200 0E5"? ..Sm. ”20¢ 20mg MED €60 0:58». .5> ”200 06N=0§xo€>n§ {mag

 

dun—59:00 6505:... .3 8.03 3 9530.80 £380:

anom— bfiz §»_0uemfl-=oz .8 Baas—2:050 .03 95$...—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Km §>

 

 

 

 

 

 

$035.5 08:. nouasom
o..oo odm ode odm odm
3 3 . 3 3 ad
7 3 3 <7333333Jv3/3 3 3 33 8 com .
T 34 4.3.3-3333? 3.33 98 com 0..
4K - 3 33 3 33-13-.33/x33l./f ..I..)r.\3(33333\l\3 Ljfi 3 \ 33)...3,._/.4..3(3\33 33- I9:— 8*
_.. .,,,3333/ . 3.333333333333333 3,..33/.33333I3\)/ ....3331 -3 3.4 .. 33 3 113/.\3/ 3 333). 33/333 .
< ,4 44 4 144 44.4.4. 4 4 44 ya 8% o N
444 ,3 44 3333 3 44 334.333.3343 3334733 -3 Se 82
. H
4.4. .-3 333 3 33. 333 3 3 333 33 33 4 333345343 333.33 3.3 33 l 3 3 3 3 3-393336U40 m 9
f. 3 33333333 333. .333 33 33 3 r 3.3 3 33%\3 3 3 3 4 m
4 398 8444 a
.30 W

4.06

 

 

178

 

 

2:58:50 55051.: «o 8.000 3 “0000000000 0.69:5:
€60 0600?. .<m €60 0:30.“ 3E €60 00.0800: €00 €60 035.0? ..Sm €60 06b00 W00 €60 0===0> 60> €60 06003.0“8—03i {mm—n
:00m .902 Eon—0.»: 253—3. .6 83303030 £2 0.5m:—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Amoassfiv 08E. 00:033—
odm 3 odm . 9.00 - 9mm . ofim

od

33ng
3333.3

333- E 03..
344333.:

-333 __d-adfl.
33 3 H3 33934030.:

33-433 -,-3-3c3@3%fl
6.0
.o.m
6.0

 

 

 

 

(vrl) esuodsea

179

 

 

02:09:00 0305:: .«0 00—00: 0» 30:09:00
0.60:5: €60 030:6 .<m €60 00—30% ém €60 3.380003% :00: €60 035.30 chm €60 06b00 ..«00 €60 0==:0> .:0> €60 060:3?053i .452:
:00m 0902 cash—0.6.3.— Eo< ..0 8003008030 .03 0.5a:—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

<00

 

 

 

 

 

 

 

 

 

 

 

 

 

03:55 25... 80:33.
98 . 0.8 3 99. 0.8 O38
0.0
333333 3 333 00.0.3307 ->3<33J- Fe 80
33333 33 0 3333-333 Sea...
r'II‘J/ . x33.\|IlI\). 131,3. .3(3i...33,(. x
3.» 3.3300 05.3 - 4U.» 3,. 96 80
33 0.33. - 33333 3 33333.3 ..
[3... ., . (\1 .\./.0 . , 3'3 3\ I33 \. .1/0 .. 0:4 13%....“ 3\3 Ill\l.r\313r\0r3\3(3).. .333 00.3!!! E O8) H0
3333 ,0 3133-30 3333 333333. .. 0 33333 /7...3......r.333-I3-333333333.33133 33333.00
T -333 [(335333 3333333333} 3 33 0 333333- . .
003,8 800
0 0 0 ,3
8...> $00
<0 0:0
<00
0000 .00
<0
3 5.0

(W) asuodsaa

180

 

ode

 

9.8

032E 20:. 8083:

0.?

03:00:50 5005:: mo 860: 00 0:008:00 000060:
€60 200:6 .<m €60 0:30.“ {a €60 0000:5003m JO: €60 0&5”? Ham €60 06b00 .000 €60 00:80.» .:0> €60 030:0nmxougnfi JAE:
0:203— :00m .902 E.0:uml-:0z ..0 8030005050 .0: 0.593.—

9.8

 

 

 

 

 

 

 

 

od

 

95 800

 

 

 

 

 

<m

 

 

 

95 800 3

 

HZ: cow

 

98 8a 00

 

9:. 8030

 

9a 02000

 

- 9:. 80

.o.v

dd

dd

 

 

(vrl) asuodsaa

181

 

.3800800 55058: .00 8.00: 00 0:800:00 00008::
€60 080:6 .<m €60 0:800 .<m €60 060800: $.00 €60 0m??? ..Sm €60 060.000 .000 €60 0==:0> .:0> €60 0_0n:00%xo€3§ {mu-E
0:303— :55 .90 Z BEEPmfl 0.5% 00 80000008080 .0: unawmh

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

03:55 25. 80800
3 9% 0.8 0.9. 09,. 300 .
fl 3 3 0.0
.. 0 » ...-3 3- H2: 08
33 3,0000 03 _ = 5:. 800 3
.0. .. _. 9... ...-0
033% 0,333.3. 3333-3. 3 )_ .33 333. 3.3) 3333 .. . 3-- - .
.00 .0 0 0 0 . .0 0 0 SE 800 o N
00 0 0..
33 333 4 .- 3. 03-33 3 ...3333 3 3-3- 33333.- .
__.0 w. 00 .0. is 800
333, 33 3 . 3 3 3 3. 33 {(33, .33 33 - 3 - .
. 0.7 ,0 # H 98 8: o 0
333 3.3 3 33 3 33 333 333.3 3 33-033- - 33 33.3 3 3
_ 0 9:. 0000
_ am . .
<0 <00 8.5 0 o 0.
_ <:
<oo <0 08
.00

 

6.0

 

 

 

 

(W) 933106898

182

 

635888 8505?: we $.on 3 288:3 E383:

fig own-«Em .<m ”Boa czar—om JG axon organs-k {Om ”Boa omwfibnm ..Sm ”28 £0an £5 £8 05:?» .§> 66a omoaconbfieaafi {mu-E
«528% anom— bfiz cons—cue»: Eo< a: Hausa—2:0 .9: 0.5»:—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Awsasavoéh 8:53
26 OS 99. gm SN
3 b 3 p 3 . 3 _ _ 0.0
.- - 3330.733 9:. 8a
3333333) 0.333,..- .333 )3 3 3) T«333/- 33. 3 E °.F
- .0 _ .7 _> 82
3 3 0 03- 3. 0.- ,. - 33 >- - ,0. 9e 8e

03 ,- , . 0... o ...

3 - 0..., 00 , - . ‘3 3 ., )3 3 333 . 331 3333/0 0...3-3\3-3 33333 3 3 3333.... 0.3 _ we °CW—O.N

333-..-...03333 0 3 333 - 3 .3 $3300.33- 3333 3 - 330 033 .

3 3 ,.3 ...-333333-333 3 3 3333 .-3 33,... 33, 0 . a
- 0 000 3.. - k . f 00 98 code” m
333 3 3 3 3-3 33 333, 3333 3333 .3-3.33L-.33 - ,333 3. u

<m H _ H2: 83 m
<m a am 5> a a 03 M
<0 <: ~
0 0
0 5 0
0 056
0
0 0
00 So
0.

 

 

 

 

183

 

 

.35888 .5655. 90 8.3g 8 2.88:8 £383:
”ES 032% .<w ”Eon ozaom .<m ”Eon 2.38:2 .<Uq ”Ron 2??? .Sm ”Eon 06% ”EU ”Eon o===a> 48> ”Eon gouaonbfieaai {mu-E
52 «ow—«um— :aom .032 cogcucmméez u: EfiufiaEEAU £3 «...-a:—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Amssaav 25 8:83
0.8 98 99. 9.8 P8.
2.
33333 3003.3 33.033 95E
3 J 33/ 333/33..) .2- #133333 (23 )3. 3(3):. .\.}. .333/33 3-.
.0 .0 . - 95 8m
3333/, 3333 3 333.. .333/<33] \30 3.. _. 3 .33 3 ..3333 03>.) \-.<.(I3I3/.). 0,333 3.33.333 3. O.N
0 05 .. .0 < 00 a cow
3333,... 00.333 3. 3 3333 333 33 , 3r 383/ 0... .31:3.3-3.13Kw.-<30023333>3-0 03 3,333.3.- 3-3 a 66m .
_
3 0033333 33 33 333333- 333333 33333.3). .31 K33,-5.3!3]..)3f23../33333) )- 3 3 3 SE cow
,0 , 3 - 3 3-3 33,, 3 3.. 3, .33, 303333, .3 ., 3 - 3 .3
0 .... 33.0 / .00 a .0 - HME ch
.33 3 33 3 33 ,0 - 33 - -3 3333 938%.
0 <m o ._ 0
00 .7 _ 3
0 am <13
§>
0 <0.H
0 Mao
0.0 <n~ 0
0 do
<Uo

 

 

 

 

(W3 “W8

184

 

.mucsonfioo 85050:: .00 9.8: 8 9.83:8 23:5: 608

292i .<m ”Eon 0:950 JE ”Eon £35305. {Um ”Eon Emit? .bmm ”Eon “gobs uaU HEN £553 .53 ”Bow o0§3~nxo€~mni {mag

52 «autum— uaom .0»: Z eons—Ev»: 255—? no Eauuouaficuau £3 PEEK

92:55 2:: 50803
0.9. 0.8 0.8.
- - - - :3 3 3 3- 0.0

3 -3 3 3 19368-0-

5,- -3- 3, 3 3 39E8Q

3 J ..0 - 3 3 - - - - L 3 - 3 .3833 00
_> cola-m

. - , 3 3 95820
-- ,, 00 - - , 3 - aid-ca.
.. - - - j - 9e 8:
95 83.0.8

 

(v31) asuodsaa

<m SQ §>
50 m
.w U 6.2.

 

 

<m

185

 

$25888 850515 mo $33 8 vaonmotoo 33:5:
axon cams—mm .<m ”Eon £80m rod MES“ 05:50: {On— man "mucin ..Sw ”Eon amuse ”EU fies 02:5? .§> ”Boa omonconbsfixnfl {min
5: .2"an 53m .Paz wash—PG»: Eo< .3 5.2335950 .93 charm

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

<00

 

 

$8336 «8:. EVE—80m
0.8 L 0.8 adv omolm [ odm
.l lilac
I I 1/221lelv lz 11.11} a 21313 ).,..\]>3JJI 9:. 31%
III (({fxlllllulc\|)\|: (IK /, ill/.1)! ll. ‘1 ) v
a 9:. Sn
[(EIKII/ .\. [.1] 0 JRJEJEPIJ“! _.\ )(3 , ll\. I 1? 1 I \(ls)(\./\
( N x V HNE DOE
r) . L, ,2) l . «r sisal)? 41/2? ‘ i! - - dd
fl a . 5} \_. . ..x 9:. 8a
I; ..lak \. .. \__« all} K 1 J _,,:2\J_. f\(!u1|.l11;f}f. {Ii} [Illfiptliisxx J, fl {I KW} 131...... ..2 _.-: l\l JR) 1.), 1i: (1 XIII.
q . < e 83
.4 i A Y
I}. l)l)\). ...Tii! \\\\\\\\\ 1 l)..( 1i! lllllztly‘ I? {1. I |\Jl|ll\ J ‘ .4 I 131;). l J \ \: x i
fl \ ...; , _ NE ca
_lzstr .rxlz i 1, q ‘ \ 4| Ill}; ‘45}- iv ,5): L}. a ii~NEo¢fld+
_ fl .
_ M §> St
w.
_ HNU _.o.®

 

 

 

(W) asuodsea

186

APPENDIX B

Table 3. Unconjugated Phenolic Compounds (mg /100 g of Bean Flour)*

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

BB PB RB NB
Hydroxybenzoic Acids
PCA 1.63 d: 1.37 2.61 i 3.13 5.48 i 4.53 nd
pHBA 0.36 i 0.27 0.90 :1: 0.12 2.41 d: 0.68 0.37 i 0.21
Van 0.69 d: 0.21 0.81 :t 0.21 0.55 :t 0.11 0.40 :1: 0.11
Syr 0.93 :t 0.27 0.43 i 0.27 0.53 :1: 0.25 0.49 :1: 0.23
Total 3.61 4.75 8.97 1.26
Hydroxycinnamic Acids
Caf 0.17 i 0.03 0.43 :I: 0.15 0.50 i 0.20 0.16 :1: 0.03
pCA 0.28 a: 0.10 0.49 :I: 0.21 0.44 d: 0.14 0.55 i 0.22
PA 1.50 d: 0.47 3.06 :1: 0.92 1.58 i 0.55 2.93 i 0.88
SA 0.40 :t 0.07 1.06 i 0.18 0.61 :I: 0.19 0.59 :1: 0.17
Total 2.35 5.04 3.13 4.23
Flavan-3ols
Cat 0.25 :t 0.14 2.65 :t 1.41 0.31 :l: 0.13 nd
Total 0.25 2.65 0.31 nd
Flavonols
Q 2.66 i: 0.82 0.86 i 0.55 3.86 :t 1.50 nd
K 0.64 :1: 0.16 6.71 :t 2.02 25.70 :1: 9.99 nd
Total 3.30 7.57 29.56 nd
TIP” 9.50 20.00 41.96 5.49
259.06 :t 274.69 i 230.10 :t 149.45 :1:
TPC!) 9.11 4.87 6.16 7.10

 

‘ All values are dry basis. Values are means of n= 5-8 determinations :t standard deviation adjusted for
loss due to Sep Pak. BB, black bean; PB, pinto bean; RB, red bean; NB, navy bean; PCA, protocatechuic
acid; pHBA, p-hydroxybenzoic acid; Cat, (+)-catechin; Van, vanillic acid; Caf, caffeic acid; Syr, syringic
acid; pCA, p-coumaric acid; FA, ferulic acid; SA, sinapic acid; Q, quercetin; K, kaempferol; nd, not
detected. a. TIP; Total phenolic compounds identified by HPLC-ECD. b. TPC = Total Phenolic Content =
mean mg of (+)-Catechin Equivalent per 100 g of bean flour d: standard deviation (n = 4-6) determined by

Folin-Ciocalteu’s Assay.

187

Table 4. Unconjugated Phenolic Compounds (mg /100 g Bean Fraction)*

 

 

 

 

 

 

 

 

 

NBEM NBR
Hydroxybenzoic Acids
PCA nd nd
pHBA 0.66 :I: 0.37 0.34 :h 0.05
Van 0.86 d: 0.34 0.27 d: 0.13
Syr 0.55 d: 0.21 0.53 :1: 0.35
Total 2.01 1.14
Hydroxycinnamic Acids
Caf 0.27 d: 0.09 0.09 :t 0.05
pCA 1.65 d: 1.18 0.29 d: 0.07
FA 5.90 :h 1.46 1.19 i 0.30
SA 1.25 :1: 0.37 0.27 :t 0.05
Total 9.07 1.84
NP“ 11.15 2.99
TPCb 328.40 :I: 15.44 66.75 :1: 5.96

 

*All values are dry basis. Values are means of n= 5-8 determinations 3: standard

deviation adjusted for loss due to Sep Pak and alkaline hydrolysis. NBEM, navy bean

extract mix; NBR, navy bean residue; PCA, protocatechuic acid; pHBA, p-

hydroxybenzoic acid; Van, vanillic acid; Caf, caffeic acid; Syr, syringic acid; pCA, p-
coumaric acid; FA, ferulic acid; SA, sinapic acid; nd, not detected. a. TIP; Total phenolic

compounds identified by HPLC-ECD. b. TPC = Total Phenolic Content = mean mg of

(+)-Catechin Equivalent per 100 g of bean flour i standard deviation (n = 4-6)

determined by Folin-Ciocalteu’s Assay.

188

 

Table 5. Unconjugated Phenolic Compounds + Phenolic Compounds Liberated by
Alkaline Hydrolysis (mg / 100 g of Bean Flour)*

 

 

 

 

 

 

 

 

 

 

 

BB PB RB NB
Hydroxybenzoic Acids
PCA 2.29 :1: 1.78 3.03 i 1.77 4.32 :1: 2.73 nd
pHBA 0.46 :1: 0.37 1.06 :1: 0.12 2.44 :1: 0.67 0.44 :1: 0.27
Van 1.85 :1: 0.57 1.12 i: 0.41 0.58 i 0.42 1.05 i 0.70
Syr 2.06 :1: 1.63 0.74 :1: 0.63 0.84 d: 0.87 1.22 :1: 1.12
Total 6.66 5.95 8.18 2.71
Hydroxycinnamic Acids
Caf 0.99 :t 0.70 0.92 :1: 0.57 0.92 :L- 0.79 0.58 i 0.48
pCA 2.40 :1: 1.80 0.80 :1: 0.46 0.69 3c 0.43 3.00 d: 1.79
FA 13.68 :1: 4.87 10.62 :1: 5.19 5.12 :1: 2.28 26.49 i: 7.88
SA 6.60 :1: 1.70 7.75 :1: 2.34 3.82 :1: 0.96 5.93 i 1.23
Total 23.67 20.09 10.55 36.00
Flavonols
Q 5.24 :1: 1.87 1.70 :1: 0.28 6.36 i 0.81 nd
K 1.99 :1: 0.51 15.82 d: 2.93 55.65 i 9.63 nd
Total 7.23 17.52 62.01 nd
TIP“ 37.55 43.55 80.76 38.72

 

* All values are dry basis. Values are means of n= 5-8 determinations :h standard

deviation adjusted for loss due to Sep Pak and alkaline hydrolysis. BB, black bean; PB,

pinto bean; RB, red bean; NB, navy bean; PCA, protocatechuic acid; pHBA, p-

hydroxybenzoic acid; Van, vanillic acid; Caf, caffeic acid; Syr, syringic acid; pCA, p-

coumaric acid; FA, ferulic acid; SA, sinapic acid; Q, quercetin; K, kaempferol; nd, not

detected. a. TIP; Total phenolic compounds identified by HPLC-ECD.

189

Table 6. Unconjugated Phenolic Compounds + Phenolic Compounds Liberated by
Alkaline Hydrolysis (mg / 100 g of Bean Fraction)*

 

 

 

 

 

 

 

 

NBEM NBR
Hydroxybenzoic Acids
PCA nd nd
pHBA 0.67 :1: 0.39 0.26 :t 0.22
Van 2.26 i 1.66 0.27 :1: 0.06
Syr 2.79 :t 2.22 1.36 :t 1.07
Total 5.72 1.89
Hydroxycinnamic Acids
Caf 0.88 :t 0.79 0.32 :1: 0.30
pCA 5.83 :1: 5.37 0.83 :1: 0.46
FA 45.45 :1: 14.32 5.41 i 1.95
SA 10.03 i 2.30 2.73 :t 0.71
Total 62.19 9.29
TIP“ 67.90 11.18

 

*All values are dry basis. Values are means of n= 5-8 determinations i standard
deviation adjusted for loss due to Sep Pak and alkaline hydrolysis. NBEM, navy bean
extract mix; NBR, navy bean residue; PCA, protocatechuic acid; pHBA, p-
hydroxybenzoic acid; Van, vanillic acid; Caf, caffeic acid; Syr, syringic acid; pCA, p-
coumaric acid; FA, ferulic acid; SA, sinapic acid; nd, not detected. a. TIP; Total phenolic

compounds identified by HPLC-ECD.

190

Table 7. Unconjugated Phenolic Compounds + Phenolic Compounds Liberated by
Acid Hydrolysis (mg /100 g of Bean Flour)*

 

 

 

 

 

 

 

 

 

 

 

BB PB RB NB
Hydroxybenzoic Acids
PCA 4.00 i: 3.18 5.02 :t 2.49 7.00 d: 4.06 nd
pHBA 0.79 :t 0.56 1.22 i 0.97 2.98 i 2.40 0.34 i 0.23
Van 1.63 :h 0.89 1.28 :t 0.55 0.88 :1: 0.79 1.29 i 1.10
Syr 2.50 i 2.09 1.05 :1: 0.91 0.81 d: 0.81 0.76 i 0.59
Total 8.92 8.57 11.67 2.39
Hydroxycinnamic Acids
Caf 0.41 :1: 0.30 0.51 i 0.45 0.67 :1: 0.69 0.33 :t 0.23
pCA 0.74 i 0.30 0.84 :1: 0.33 0.85 d: 0.43 1.45 :1: 0.30
FA 4.10 :1: 1.60 5.70 :1:1.94 3.31 i 1.18 8.75 d: 3.08
SA 1.09 i 0.38 1.92 d: 0.46 1.22 d: 0.40 1.27 :1: 0.30
Total 6.34 8.97 6.05 11.80
Flavonols
Q 4.45 :h 1.82 4.34 i 1.53 7.88 :t 3.33 nd
58.60 :1:
K 1.57 d: 0.62 19.15 i 8.91 26.07 nd
Total 6.02 23.49 66.48 mi
TIP“ 21.28 41.02 84.20 14.18

 

* All values are dry basis. Values are means of n = 5-8 determinations 1: standard

deviation adjusted for loss due to Sep Pak and alkaline hydrolysis. BB, black bean; PB,

pinto bean; RB, red bean; NB, navy bean; PCA, protocatechuic acid; pHBA, p-

hydroxybenzoic acid; Van, vanillic acid; Caf, caffeic acid; Syr, syringic acid; pCA, p-

coumaric acid; FA, ferulic acid; SA, sinapic acid; Q, quercetin; K, kaempferol; nd, not

detected. a. TIP; Total phenolic compounds identified by HPLC-ECD.

191

 

 

 

Table 8. Unconjugated Phenolic Compounds + Phenolic Compounds Liberated by
Acid Hydrolysis (mg /100 g of Bean Fraction)*

 

 

 

 

 

 

 

 

NBEM NBR
Hydroxybenzoic Acids
PCA nd nd

pHBA 0.43 :1: 0.31 0.56 :1: 0.50
Van 2.13: 1.93 1.62:1: 1.12
Syr 2.59 :l: 2.02 0.30 :h 0.24

Total 5.15 2.48

Hydroxycinnamic Acids

Caf 0.44 :1. 0.43 0.06 i 0.03
pCA 4.36 :t 2.33 0.89 :1: 0.47
FA 12.32 :1: 4.26 3.88 d: 1.38
SA 2.19 :1: 0.58 0.59 :h 0.14

Total 19.31 5.42

TIP“ 24.46 7.90

 

 

*All values are dry basis. Values are means of n= 5-8 determinations i standard
deviation adjusted for loss due to Sep Pak and alkaline hydrolysis. NBEM, navy bean
extract mix; NBR, navy bean residue; PCA, protocatechuic acid; pHBA, p-
hydroxybenzoic acid; Van, vanillic acid; Caf, caffeic acid; Syr, syringic acid; pCA, p-
coumaric acid; FA, ferulic acid; SA, sinapic acid; nd, not detected. a. TIP; Total phenolic

compounds identified by HPLC-ECD.

 

192

LITERATURE CITED

 

193

 

LITERATURE CITED

Adam, A., Crespy, V., Levrat-Verny, M. A., Leenhardt, F., Leuillet, M., Demigne, C., &
Remesy, C. (2002) The bioavailability of ferulic acid is governed primarily by the food
matrix rather than its metabolism in intestine and liver in rats. Journal of Nutrition 132:
1962-1968.

Adorn, K. K., Sorrells, M. E., & Liu, R. H. (2003) Phytochemical profiles and antioxidant
activity of wheat varieties. Journal of Agricultural and Food Chemistry 51: 7825-7834.

Adorn, K. K., & Liu, R. H. (2002) Antioxidant activity of grains. Journal of Agricultural
and Food Chemistry 50: 6182-6187.

(AICR), W. C. R. F. W. A. I. f. C. R. (2007) Food, Nutrition, Physical Activity, and the
Prevention of Cancer: a Global Perspective. WCRF/AICR, Washington DC.

Anderson, J. W., & Major, A. W. (2002) Pulses and lipaemia, short- and long-term effect:
Potential in the prevention of cardiovascular disease. British Journal of Nutrition 88:
$263-$271.

Andreasen, M. F., Kroon, P. A., Williamson, G., & Garcia-Conesa, M. T. (2001) Esterase
activity able to hydrolyze dietary antioxidant hydroxycinnamates is distributed along the
intestine of mammals. Journal of Agricultural and Food Chemistry 49: 5679-5684.

Andreasen, M. F., Kroon, P. A., Williamson, G., & Garcia-Conesa, M. T. (2001)
Intestinal release and uptake of phenolic antioxidant diferulic acids. Free Radical Biology
and Medicine 31 : 304-314.

Angel, P., & Karin, M. (1991) The Role of Jun, Fos and the Ap-l Complex in Cell-
Proliferation and Transformation. Biochimica Et Biophysica Acta 1072: 129-157.

Aparicio-Femandez, X., Yousef, G. G., Loarca-Pina, G., de Mejia, E., & Lila, M. A.
(2005) Characterization of polyphenolics in the seed coat of Black J amapa bean
(Phaseolus vulgaris L.). Journal of Agricultural and Food Chemistry 53: 4615-4622.

Armstrong, B., & Doll, R. (1975) Environmental Factors and Cancer Incidence and
Mortality in Different Countries, with Special Reference to Dietary Practices.
International Journal of Cancer 15: 617-631.

Azevedo, L., Gomes, J. C., Stringheta, P. C., Gontijo, A., Padovani, C. R., Ribeiro, L. R.,

& Salvadori, D. M. F. (2003) Black bean (Phaseolus vulgaris L.) as a protective agent
against DNA damage in mice. Food and Chemical Toxicology 41: 1671-1676.

194

 

Balasundram, N., Sundram, K., & Samman, S. (2006) Phenolic compounds in plants and
agri-industrial by-products: Antioxidant activity, occurrence, and potential uses. Food
Chemistry 99: 191-203.

Bamberger, A. M., Milde-Langosch, K., Rossing, E., Goemann, C., & Loning, T. (2001)
Expression pattern of the AP-l family in endometrial cancer: correlations with cell cycle
regulators. Journal of Cancer Research and Clinical Oncology 127: 545-550.

Bametson, R. A. D., M.G. (2007) Genetic Susceptibility to Colorectal Cancer. In J .J .
Cassidy, P.; Van Cutsem, E., ed. Colorectal Cancer Infonna Healthcare USA, Inc., New
York.

Bawadi, H. A., Bansode, R. R., Trappey, A., Truax, R. E., & Losso, J. N. (2005)
Inhibition of Caco-2 colon, MCF-7 and Hs578T breast, and DU 145 prostatic cancer cell
proliferation by water-soluble black bean condensed tannins. Cancer Letters 218: 153-
162.

Beninger, C. W., Gu, L. W., Prior, R. L., Junk, D. C., Vandenberg, A., & Bett, K. E.
(2005) Changes in polyphenols of the seed coat during the after-darkening process in

pinto beans (Phaseolus vulgaris L.). Journal of Agricultural and Food Chemistry 53:
77 77-7 782.

Beninger, C. W., & Hosfield, G. L. (1999) Flavonol glycosides from Montcalm dark red
kidney bean: Implications for the genetics of seed coat color in Phaseolus vulgaris L.
Journal of Agricultural and Food Chemistry 47: 4079-4082.

Beninger, C. W., & Hosfield, G. L. (2003) Antioxidant activity of extracts, condensed
tannin fractions, and pure flavonoids from Phaseolus vulgaris L. seed coat color
genotypes. Journal of Agricultural and Food Chemistry 51 : 7879-7883.

Beninger, C. W., Hosfield, G. L., & Bassett, M. J. (1999)F1avonoid composition of three
genotypes of dry bean (Phaseolus vulgaris) differing in seedcoat color. Journal of the
American Society for Horticultural Science 124: 514-518.

Beninger, C. W., Hosfield, G. L., & Nair, M. G. (1998) Physical characteristics of dry
beans in relation to seedcoat color genotype. Hortscience 33: 328-329.

Bennink, M. R., & Barrett, K. G. (2004) Inhibition of colon cancer with soy and other
plant products: anticancer constituents. Agra Food Industry Hi-Tech 15: 4-6.

Biasi, F., Mascia, C., & Poli, G. (2007) TGF beta 1 expression in colonic mucosa:
modulation by dietary lipids. Genes and Nutrition 2: 233-243.

Bird, R. P. (1995) Role of Aberrant Crypt Foci in Understanding the Pathogenesis of
Colon-Cancer. Cancer Letters 93: 55-71.

195

 

 

Bird, R. P., McLellan, E. A., & Bruce, W. R. (1989) Aberrant Crypts, Putative
Precancerous Lesions, in the Study of the Role of Diet in the Etiology of Colon Cancer.
Cancer Surveys 8: 189-200.

Boateng, J. A., Verghese, M., Walker, L. T., Shackelford, L. A., & Chawan, C. B. (2007)
Inhibitory effects of selected dry beans (Phaseolus spp L) on azoxymethane-induced
formation of aberrant crypt foci in Fisher 344 male rats. Nutrition Research 27: 640-646.

Bobe, G., Barrett, K. G., Mentor-Marcel, R. A., Safflotti, U., Young, M. R., Colbum, N.
H., Albert, P. S., Bennink, M. R., & Lanza, E. (2008) Dietary cooked navy beans and
their fractions attenuate colon carcinogenesis in Azoxymethane-Induced Ob/Ob mice.
Nutrition and Cancer-an International Journal 60: 373-381.

Bocchi, C., Careri, M., Groppi, F., Mangia, A., Manini, P., & Mori, G. (1996)
Comparative investigation of UV, electrochemical and particle beam mass spectrometric
detection for the hi gh-performance liquid chromatographic determination of benzoic and
cinnamic acids and of their corresponding phenolic acids. Journal of Chromatography A
753: 157-170.

Bode, A. M., & Dong, Z. (2004) Targeting signal transduction pathways by
chemopreventive agents. Mutation Research-Fundamental and Molecular Mechanisms of
Mutagenesis 555: 33-51.

Bravo, L. (1998) Polyphenols: Chemistry, dietary sources, metabolism, and nutritional
significance. Nutrition Reviews 56: 317-333.

Bressani, R. E., LG. (1980) The nutritional role of polyphenols. In J .H. Hulse, ed.
Polyphenols in Cereals and Legumes International Development Research Centre,
Ottawa.

Brun, P., Castagliuolo, 1., Di Leo, V., Buda, A., Pinzani, M., Palu, G., & Martines, D.
(2007) Increased intestinal permeability in obese mice: new evidence in the pathogenesis

of nonalcoholic steatohepatitis. American Journal of Physiology-Gastrointestinal and
Liver Physiology 292: G518-G525.

Buda, A., & Pignatelli, M. (2002) Genetics - cellular basis. British Medical Bulletin 64:
45-58.

Busslinger, M. B., & Bergers, G. (1994) Identification of AP-l regulated genes. In P.E.H.
Angel, P., ed. The Fos and Jun Families of Transcription Factors CRC Press Inc., Boca
Raton, FL.

Caderni, G., Femia, A. R, Giannini, A., Favuzza, A., Luceri, C., Salvadori, M., & Dolara,
P. (2003) Identification of mucin-depleted foci in the unsectioned colon of

azoxymethane-treated rats: Correlation with carcinogenesis. Cancer Research 63: 2388-
2392.

196

 

 

Cai, R., Hettiarachchy, N. S., & J alaluddin, A. (2003) High-performance liquid
chromatography determination of phenolic constituents in 17 varieties of cowpeas.
Journal of Agricultural and Food Chemistry 51: 1623-1627.

Cardador-Martinez, A., Albores, A., Bah, M., Calderon-Salinas, V., Castano-Tostado, E.,
Guevara-Gonzalez, R., Shimada-Miyasaka, A., & Loarca-Pina, G. (2006) Relationship
among antirnutagenic, antioxidant and enzymatic activities of methanolic extract from
common beans (Phaseolus vulgaris L.). Plant Foods for Human Nutrition 61: 161-168.

Cardador-Martinez, A., Loarca-Pina, G., & Oomah, B. D. (2002) Antioxidant activity in
common beans (Phaseolus vulgaris L.). Journal of Agricultural and Food Chemistry 50:
6975-6980.

Cassidy, P, Steele, R. J. C. (2007) Colorectal Cancer Screening. In J .J . Cassidy, P.; Van
Cutsem, E., ed. Colorectal Cancer Informa Healthcare USA, Inc., New York.

Champ, M. M. J. (2002)Non-nut1ientbioactive substances of pulses. British Journal of
Nutrition 88: 8307-83 1 9.

Cheng, S. K., Olale, F., Brivanlou, A. H., & Schier, A. F. (2004) Lefty blocks a subset of
TGF beta signals by antagonizing EGF-CFC coreceptors. PLoS Biology 2: 215-226.

Cheynier, V. (2005) Polyphenols in foods are more complex than often thought.
American Journal of Clinical Nutrition 81: 2238-2298.

Choung, M. G., Choi, B. R., An, Y. N., Chu, Y. H., & Cho, Y. S. (2003) Anthocyanin
profile of Korean cultivated kidney bean (Phaseolus vulgaris L.). Journal of Agricultural
and Food Chemistry 51: 7040-7043.

Chun, O. K., Chung, S. J ., & Song, W. O. (2007) Estimated dietary flavonoid intake and
major food sources of US adults. Journal of Nutrition 137: 1244-1252.

Clifford, M. N. (1999) Chlorogenic acids and other cinnamates - nature, occurrence and
dietary burden. Journal of the Science of Food and Agriculture 79: 362-372.

Compton, C. C. (2003) Colorectal carcinoma: Diagnostic, prognostic, and molecular
features. Modern Pathology 16: 376-388.

Correa, P. (1981) Epidemiological Correlations between Diet and Cancer Frequency.
Cancer Research 41: 3685-3689.

Costa, G. E. D., Queiroz-Monici, K. D. S., Reis, S., & de Oliveira, A. C. (2006) Chemical

composition, dietary fibre and resistant starch contents of raw and cooked pea, common
bean, chickpea and lentil legumes. Food Chemistry 94: 327-330.

197

 

Coughlin, S. S., Calle, E. E., Teras, L. R., Petrelli, J ., & Thun, M. J. (2004) Diabetes
mellitus as a predictor of cancer mortality in a large cohort of US adults. American
Journal of Epidemiology 159: 1 160-1167.

Croft, K. D. (1998) The chemistry and biological effects of flavonoids and phenolic
acids. Towards Prolongation of the Healthy Life Span - Practical Approaches to
Intervention. Annals of the New York Academy of Sciences 854: 435-442.

Dabrowski, K. J ., & Sosulski, F. W. (1984) Quantitation of Free and Hydrolyzable
Phenolic-Acids in Seeds by Capillary Gas-Liquid-Chromatography. Journal of
Agricultural and Food Chemistry 32: 123-127.

Davies, M. J ., & Rumsby, P. C. (1998) Long-term analysis of colonic aberrant crypt
formation after treatment of Sprague-Dawley rats with azoxymethane. T eratogenesis
Carcinogenesis and Mutagenesis 18: 183-197.

Day, A. J ., Canada, F. J ., Diaz, J. C., Kroon, P. A., McLauchlan, R., Faulds, C. B.,
Plumb, G. W., Morgan, M. R. A., & Williamson, G. (2000) Dietary flavonoid and
isoflavone glycosides are hydrolysed by the lactase site of lactase phlorizin hydrolase.
FEBS Letters 468: 166-170.

Derynck, R., Akhurst, R. J ., & Balmain, A. (2001) TGF-beta signaling in tumor
suppression and cancer progression. Nature Genetics 29: 117-129.

Diaz-Batalla, L., Widholm, J. M., Fahey, G. C., Castano-Tostado, E., & Paredes-Lopez,
O. (2006) Chemical components with health implications in wild and cultivated Mexican
common bean seeds (Phaseolus vulgaris L.). Journal of Agricultural and Food
Chemistry 54: 2045-2052.

Doll, R. (1998) Epidemiological evidence of the effects of behaviour and the
environment on the risk of human cancer. Recent Results in Cancer Research 154: 3-21.

Dove-Edwin, 1., & Thomas, H. J. W. (2001) Review article: The prevention of colorectal
cancer. Alimentary Pharmacology & Therapeutics 15: 323-336.

Druckrey, H. (1973) Specific Carcinogenic and Teratogenic Effects of Indirect
Alkylating Methyl and Ethyl Compounds, and Their Dependency on Stages of Ontogenic
Developments. Xenobiotica 3: 271-303.

Ealey, K. N., Li, S. Y., & Archer, M. C. (2008) Development of aberrant crypt foci in the
colons of ob/ob and db/db mice: Evidence that leptin is not a promoter. Molecular
Carcinogenesis 47: 667-677.

Elliott, R. L., & Blobe, G. C. (2005) Role of transforming growth factor beta in human
cancer. Journal of Clinical Oncology 23: 2078-2093.

198

 

Erdman, S. H., Wu, H. D., Hixson, L. J ., Ahnen, D. J ., & Gemer, E. W. (1997)
Assessment of mutations in Ki-ras and p53 in colon cancers from azoxymethane- and
dimethylhydrazine—treated rats. Molecular Carcinogenesis 19: 137-144.

Espinosa-Alonso, L. G., Lygin, A., Widholm, J. M., Valverde, M. E., & Paredes-Lopez,
O. (2006) Polyphenols in wild and weedy Mexican common beans (Phaseolus vulgaris
L.). Journal of Agricultural and Food Chemistry 54: 4436-4444.

(FASEB), L. S. R. O. L. F. o. A. S. f. E. B. (1995) Third report on nutrition monitoring in
the United States 1. US. Goverrnnent Printing Office, Washington, DC.

 

F aulds, C. B., & Williamson, G. (1999) The role of hydroxycinnamates in the plant cell
wall. Journal of the Science of Food and Agriculture 79: 393-395.

Fearon, E. R., & Vogelstein, B. (1990) A Genetic Model for Colorectal Tumorigenesis.
Cell 61: 759-767.

Femia, A. P., & Caderni, G. (2008) Rodent Models of Colon Carcinogenesis for the
Study of Chemopreventive Activity of Natural Products. Planta Medica 74: 1602-1607.

Finley, J. W., Burrell, J. B., & Reeves, P. G. (2007) Pinto bean consumption changes
SCFA profiles in fecal fermentations, bacterial populations of the lower bowel, and lipid
profiles in blood of humans. Journal of Nutrition 137: 2391-2398.

Francis, G., Kerem, Z., Makkar, H.P.S., & Becker K. (2002) The biological action of
saponins in animal systems: a review. British Journal of Nutrition 88: 587-605.

Fraser, G. E. (1999) Associations between diet and cancer, ischemic heart disease, and
all-cause mortality in non-Hispanic white California Seventh-day Adventists. American
Journal of Clinical Nutrition 70: 5328-5388.

Fuchs, C. S., Giovarmucci, E. L., Colditz, G. A., Hunter, D. J ., Stampfer, M. J ., Rosner,
B., Speizer, F. E., & Willett, W. C. (1999) Dietary fiber and the risk of colorectal cancer
and adenoma in women. New England Journal of Medicine 340: 169-176.

Gao, L., & Mazza, G. (1994) Quantitation and Distribution of Simple and Acylated
Anthocyanins and Other Phenolics in Blueberries. Journal of Food Science 59: 1057-
1059.

Gee, J. M., DuPont, M. S., Rhodes, M. J. C., & Johnson, I. T. (1998) Quercetin
glucosides interact with the intestinal glucose transport pathway. Free Radical Biology
and Medicine 25: 19-25.

Gee, J. M., & Johnson, I. T. (2001) Polyphenolic compounds: Interactions with the gut
and implications for human health. Current Medicinal Chemistry 8: 1245-1255.

 

199

Geil, P. B., & Anderson, J. W. (1994) Nutrition and Health Implications of Dry Beans - a
Review. Journal of the American College of Nutrition 13: 549-558.

Genkinger, J. M., Platz, E. A., Hoffinan, S. C., Comstock, G. W., & Helzlsouer, K. J.
(2004) Fruit, vegetable, and antioxidant intake and all-cause, cancer, and cardiovascular
disease mortality in a community-dwelling population in Washington County, Maryland.
American Journal of Epidemiology 160: 1223-1233.

Giovarmucci, E. (2003) Diet, body weight, and colorectal cancer: A summary of the
epidemiologic evidence. Journal of Womens Health & Gender-Based Medicine 12: 173-
1 82.

Giovannucci, E. (2007) Metabolic syndrome, hyperinsulinemia, and colon cancer: a
review. American Journal of Clinical Nutrition 86: 8368-8428.

Glebov, O. K., Rodriguez, L. M., Soballe, P., DeNobile, J ., Cliatt, J., Nakahara, K., &
Kirsch, I. R. (2006) Gene expression patterns distinguish colonoscopically isolated

human aberrrant crypt foci fi'om normal colonic mucosa. Cancer Epidemiology
Biomarkers & Prevention 15: 2253-2262. '

Gonthier, M. P., Verny, M. A., Besson, C., Remesy, C., & Scalbert, A. (2003)
Chlorogenic acid bioavailability largely depends on its metabolism by the gut microflora
in rats. Journal of Nutrition 133: 1853-1859.

Goodwin, E. I. & Mercer, T.W. (1983) Introduction to plant biochemistry. Pergamon
Press, Oxford; New York.

Grady, W. M., Myeroff, L. L., Swinler, S. E., Rajput, A., Thiagalingam, S., Lutterbaugh,
J. D., Neumann, A., Brattain, M. G., Chang, J ., Kim, S. J ., Kinzler, K. W., Vogelstein, B.,
Willson, J. K. V., & Markowitz, S. (1999) Mutational inactivation of transforming

growth factor beta receptor type II in microsatellite stable colon cancers. Cancer
Research 59: 320-324.

Granito, M., Paolini, M., & Perez, S. (2008) Polyphenols and antioxidant capacity of

Phaseolus vulgaris stored under extreme conditions and processed. L WT-Food Science
and Technology 41 : 994-999.

Guda, K., Claffey, K. P., Dong, M., Nambiar, P. R., & Rosenberg, D. W. (2003)
Defective processing of the transforming growth factor-beta 1 in azoxyrnethane-induced
mouse colon tumors. Molecular Carcinogenesis 37: 51-59.

Guda, K., Giardina, C., Nambiar, P., Cui, H. Y., & Rosenberg, D. W. (2001) Aberrant

transforming growth factor-beta signaling in azoxymethane-induced mouse colon tumors.
Molecular Carcinogenesis 31: 204-213.

200

 

Guda, K., Upender, M. B., Belinsky, G., Flynn, C., Nakanishi, M., Marino, J. N., Ried,
T., & Rosenberg, D. W. (2004) Carcinogen—induced colon tumors in mice are
chromosomally stable and are characterized by low-level microsatellite instability.
Oncogene 23: 3813-3821.

Guillon, F., & Champ, M. M. J. (2002) Carbohydrate fi'actions of legumes: uses in human
nutrition and potential for health. British Journal of Nutrition 88: $293-$306.

Gunter, M. J ., & Leitzmann, M. F. (2006) Obesity and colorectal cancer: epidemiology,
mechanisms and candidate genes. Journal of Nutritional Biochemistry 17: 145-156.

Guo, J. Y., Li, X. S., Browning, J. D., Rottinghaus, G. E., Lubahn, D. B., Constantinou,
A., Bennink, M., & MacDonald, R. S. (2004) Dietary soy isoflavones and estrone protect

ovariectomized ER alpha K0 and wild-type mice from carcinogen-induced colon cancer.
Journal of Nutrition 134: 179-182.

Gupta, A. K., & Schoen, R. E. (2009) Aberrant crypt foci: are they intermediate
endpoints of colon carcinogenesis in humans? Current Opinion in Gastroenterology 25:
59-65.

Halliwell, B. (2007) Dietary polyphenols: Good, bad, or indifferent for your health?
Cardiovascular Research 73: 341 -347.

Hanahan, D., & Weinberg, R. A. (2000) The hallmarks of cancer. Cell 100: 57-70.

Hangen, L., & Bennink, M. R. (2002) Consumption of black beans and, navy beans
(Phaseolus vulgaris) reduced azoxymethane-induced colon cancer in rats. Nutrition and
Cancer-an International Journal 44: 60-65.

Harbome, J. B. (1998) Phytochemical Methods: A Guide to Modern Techniques of Plant
Analysis. Chapman and Hall, London; New York.

Harradine, K. A., & Akhurst, R. J. (2006) Mutations of TGF beta signaling molecules in
human disease. Annals of Medicine 38: 403-414.

Hartsough, M. T., & Mulder, K. M. (1997) Transforming growth factor-beta signaling in
epithelial cells. Pharmacology & Therapeutics 75: 21-41.

Hawk, E. T., Umar, A., & Viner, J. L. (2004) Colorectal cancer chemoprevention - An
overview of the science. Gastroenterology 126: 1423-1447.

Heavey, P. M., McKenna, D., & Rowland, 1. R. (2004) Colorectal cancer and the

relationship between genes and the environment. Nutrition and Cancer-an International
Journal 48: 124-141.

201

Heimler, D., Vignolini, P., Dini, M. G., & Romani, A. (2005) Rapid tests to assess the
antioxidant activity of Phaseolus vulgaris L. dry beans. Journal of Agricultural and Food
Chemistry 53: 3053-3056.

Hirose, Y., Hata, K., Kuno, T., Yoshida, K., Sakata, K., Yamada, Y., Tanaka, T., Reddy,
B. S., & Mori, H. (2004) Enhancement of development of azoxymethane-induced colonic
premalignant lesions in C57BL/KsJ-db/db mice. Carcinogenesis 25: 821-825.

Hollman, P. C. H. (2001) Evidence for health benefits of plant phenols: local or systemic
effects? Journal of the Science of Food and Agriculture 81: 842-852.

Hollman, P. C. H., Devries, J. H. M., Vanleeuwen, S. D., Mengelers, M. J. B., & Katan,
M. B. (1995) Absorption of Dietary Quercetin Glycosides and Quercetin in Healthy
Ileostomy Volunteers. American Journal of Clinical Nutrition 62: 1276-1282.

Hollman, P. C. H., vanderGaag, M., Mengelers, M. J. B., vanTrijp, J. M. P., deVries, J.
H., & Katan, M. B. (1996) Absorption and disposition kinetics of the dietary antioxidant
quercetin in man. Free Radical Biology and Medicine 21: 703-707.

Hollman, P. C. H., vanTrijp, J. M. P., Buysman, M., VanderGaag, M. S., Mengelers, M.
J. B., deVries, J. H. M., & Katan, M. B. (1997) Relative bioavailability of the antioxidant
flavonoid quercetin from various foods in man. FEBS Letters 418: 152-156.

Hong, Y., Ho, K. 8., Eu, K. W., & Cheah, P. Y. (2007) A susceptibility gene set for early
onset colorectal cancer that integrates diverse signaling pathways: Implication for
tumorigenesis. Clinical Cancer Research 13: 1107-11 14.

Hoosein, N. M., McKnight, M. K., Levine, A. E., Mulder, K. M., Childress, K. E.,
Brattain, D. E., & Brattain, M. G. (1989) Differential Sensitivity of Subclasses of
Human-Colon Carcinoma Cell-Lines to the Growth Inhibitory Effects of Transforming
Growth Factor-Beta-l. Experimental Cell Research 181: 442-453.

Hsu, S., Huang, F., Hafez, M., Winawer, S., & Friedman, E. (1994) Colon-Carcinoma
Cells Switch Their Response to Transforming Growth-F actor Beta(1) with Tumor
Progression. Cell Growth & Difl‘erentiation 5: 267-275.

Hu, Y., Cheng, Z. Q., Heller, L. I., Krasnoff, S. B., Glahn, R. P., & Welch, R. M. (2006)
Kaempferol in red and pinto bean seed (Phaseolus vulgaris L.) coats inhibits iron
bioavailability using an in vitro digestion/human Caco-2 cell model. Journal of
Agricultural and Food Chemistry 54: 9254-9261.

Hughes, J. S., & Ganthavom, C. (1997) Inhibition of AOM-induced colon cancer in rats
by dry beans (Phaseolus vulgaris). FASEB Journal 1 1: 2137-2137.

Humphries, A., & Wright, N. A. (2008) Colonic crypt organization and tumorigenesis.
Nature Reviews Cancer 8: 415-424. '

202

 

 

 

Ibrahim, S. S., Habiba, R. A., Shatta, A. A., & Embaby, H. E. (2002) Effect of soaking,
germination, cooking and fermentation on antinutritional factors in cowpeas. Nahrung-
Food 46: 92—95.

Jacobs, D. R., & Steffen, L. M. (2003) Nutrients, foods, and dietary patterns as exposures

in research: a framework for food synergy. American Journal of Clinical Nutrition 78:
508 S-5 13$.

Jandera, P., Skerikova, V., Rehova, L., Hajek, T., Baldrianova, L., Skopova, G., Kellner,
V., & Homa, A. (2005) RP-HPLC analysis of phenolic compounds and flavonoids in

beverages and plant extracts using a CoulArray detector. Journal of Separation Science
28: 1005-1022.

Janne, P. A., & Mayer, R. J. (2000) Primary care: Chemoprevention of colorectal cancer.
New England Journal of Medicine 342: 1960-1968. 1,

J avelaud, D., & Mauviel, A. (2005) Crosstalk mechanisms between the mitogen-activated
protein kinase pathways and Smad signaling downstream of TGF-beta: implications for
carcinogenesis. Oncogene 24: 5742-5750.

Jemal, A., Siegel, R., Ward, R., Hao, Y. P., Xu, J. Q., Murray, T., & Thun, M. J. (2008)
Cancer statistics, 2008. CA-A Cancer Journal for Clinicians 58: 71-96.

Jemal, A., Siegel, R., Ward, E., Murray, T., Xu, J. Q., & Thun, M. J. (2007) Cancer
statistics, 2007. CA-A Cancer Journal for Clinicians 57: 43-66.

Jenab, M. & Thompson, L.U. (2000) Phytic acid in wheat bran affects colon morphology,
cell differentiation and apoptosis. Carcinogenesis 21(8): 1547-1552.

Jenner, A. M., Rafter, J ., & Halliwell, B. (2005) Human fecal water content of phenolics:
The extent of colonic exposure to aromatic compounds. Free Radical Research 39: S63-
S63.

Karlsson, P. C., Huss, U., Jenner, A., Halliwell, B., Bohlin, L., & Rafier, J. J. (2005)
Human fecal water inhibits COX-2 in colonic HT—29 cells: Role of phenolic compounds.
Journal of Nutrition 135: 2343-2349.

Kinzler, K. W., & Vogelstein, B. (1996) Lessons from hereditary colorectal cancer. Cell
87: 159-170.

Klepacka, J ., & F omal, L. (2006) F erulic acid and its position among the phenolic
compounds of wheat. Critical Reviews in Food Science and Nutrition 46: 639-647.

Kroon, P. A., Clifford, M. N., Crozier, A., Day, A. J., Donovan, J. L., Manach, C., &
Williamson, G. (2004) How should we assess the effects of exposure to dietary
polyphenols in vitro? American Journal of Clinical Nutrition 80: 15-21.

203

 

 

Kroon, P. A., Faulds, C. B., Ryden, P., Robertson, J. A., & Williamson, G. (1997)
Release of covalently bound ferulic acid from fiber in the human colon. Journal of
Agricultural and Food Chemistry 45: 661-667.

Krygier, K., Sosulski, F ., & Hogge, L. ( 1982) Free, Esterified, and Insoluble-Bound
Phenolic-Acids .1. Extraction and Purification Procedure. Journal of Agricultural and
Food Chemistry 30: 330-334.

Kumar, S. R, Roy, S. J ., Tokumo, K., & Reddy, B. S. (1990) Effect of Different Levels
of Calorie Restriction on Azoxymethane-Induced Colon Carcinogenesis in Male F344
Rats. Cancer Research 50: 5761-5766.

Kutos, T., Golob, T., Kac, M., & Plestenjak, A. (2003) Dietary fibre content of dry and
processed beans. Food Chemistry 80: 231-235.

Kwon, K. H., Barve, A., Yu, S., Huang, M. T., & Kong, A. N. T. (2007) Cancer
chemoprevention by phytochemicals: potential molecular targets, biomarkers and animal
models. Acta Pharmacologica Sinica 28: 1409-1421.

Lanza, E., Hartman, T. J., Albert, P. S., Shields, R., Slattery, M., Caan, B., Paskett, E.,
Iber, F ., Kikendall, J. W., Lance, P., Daston, C., & Schatzkin, A. (2006) High dry bean
intake and reduced risk of advanced colorectal adenoma recurrence among participants in
the polyp prevention trial. Journal of Nutrition 136: 1896-1903.

Larkin, T., Price, W. E., & Astheirner, L. (2008) The key importance of soy isoflavone
bioavailability to understanding health benefits. Critical Reviews in Food Science and
Nutrition 48: 538-552.

Leclercq, I. A., Field, J ., & Farrell, G. C. (2003) Leptin-specific mechanisms for impaired
liver regeneration in ob/ob mice afier toxic injury. Gastroenterology 124: 1451-1464.

Leslie, A., Carey, F. A., Pratt, N. R., & Steele, R. J. C. (2002) The colorectal adenoma-
carcinoma sequence. British Journal of Surgery 89: 845-860.

Leterme, P. (2002) Recommendations by health organizations for pulse consumption.
British Journal of Nutrition 88: $239-$242.

Lin, J ., Zhang, S. M. M., Cook, N. R., Rexrode, K. M., Liu, S. M., Manson, J. E., Lee, I.
M., & Buring, J. E. (2005) Dietary intakes of fruit, vegetables, and fiber, and risk of

colorectal cancer in a prospective cohort of women (United States). Cancer Causes &
Control 16: 225-233.

Lin, P. Y., & Lai, H. M. (2006) Bioactive compounds in legumes and their germinated
products. Journal of Agricultural and Food Chemistry 54: 3807-3814.

204

Lipkin, M. (1983) Cell Proliferation in Colon Carcinogenesis. In H.W. Autrup, G. M., ed.
Experimental Colon Carcinogenesis CRC Press, Boca Raton, FL.

Lipkin, M. (1999) Preclinical and early human studies of calcium and colon cancer
prevention. Cancer Prevention: Novel Nutrient and Pharmaceutical Developments.
Annals of The New York Academy of Sciences 889: 120-127.

Lopez-Amoros, M. L., Hernandez, T., & Estrella, I. (2006) Effect of germination on
legume phenolic compounds and their antioxidant activity. Journal of Food Composition
and Analysis 19: 277-283.

Luthria, D. L. (2006) Significance of sample preparation in developing analytical
methodologies for accurate estimation of bioactive compounds in functional foods.
Journal of the Science of Food and Agriculture 86: 2266-2272.

Luthria, D. L., & Pastor-Corrales, M. A. (2006) Phenolic acids content of fifteen dry
edible bean (Phaseolus vulgaris L.) varieties. Journal of Food Composition and Analysis
19: 205-21 1.

Macz-Pop, G. A., Rivas-Gonzalo, J. C., Perez-Alonso, J. J ., & Gonzalez-Paramas, A. M.
(2006) Natural occurrence of free anthocyanin aglycones in beans (Phaseolus vulgaris
L.). Food Chemistry 94: 448-456.

Madhujith, T., Naczk, M., & Shahidi, F. (2004) Antioxidant activity of common beans
(Phaseolus vulgaris L.). Journal of Food Lipids 11: 220-233.

Manach, C., & Donovan, J. L. (2004) Pharmacokinetics and metabolism of dietary
flavonoids in humans. Free Radical Research 38: 771-785.

Manach, C., Scalbert, A., Morand, C., Remesy, C., & Jimenez, L. (2004) Polyphenols:
food sources and bioavailability. American Journal of Clinical Nutrition 79: 727-747.

Manach, C., Williamson, G., Morand, C., Scalbert, A., & Remesy, C. (2005)
Bioavailability and bioefficacy of polyphenols in humans. 1. Review of 97 bioavailability
studies. American Journal of Clinical Nutrition 81: 2308-2428.

Markowitz, S., Wang, J ., Myeroff, L., Parsons, R., Sun, L. 2., Lutterbaugh, J ., Fan, R. S.,
Zborowska, E., Kinzler, K. W., Vogelstein, B., Brattain, M., & Willson, J. K. V. (1995)
Inactivation of the Type-Ii Tgf-Beta Receptor in Colon-Cancer Cells with Microsatellite
Instability. Science 268: 1336-1338.

Massagué, J ., & Gomis, R. R. (2006) The logic of TGF beta signaling. FEBS Letters 580:
2811-2820.

Maskarinec, G., Novotny, R., & Tasaki, K. (2000) Dietary patterns are associated with
body mass index in multiethnic women. Journal of Nutrition 130: 3068-3072.

205

Matkowskyj, K. A., Marrero, J. A., Carroll, R. E., Danilkovich, A. V., Green, R. M., &
Benya, R. V. (1999) Azoxymethane-induced fulrninant hepatic failure in C57BL/6J mice:
characterization of a new animal model. American Journal of Physiology-
Gastrointestinal and Liver Physiology 277: G455-G462.

McCullough, M. L., Robertson, A. S., Chao, A., Jacobs, E. J., Stampfer, M. J ., Jacobs, D.
R., Diver, W. R., Calle, E. E., & Thun, M. J. (2003) A prospective study of whole grains,
fruits, vegetables and colon cancer risk. Cancer Causes & Control 14: 959-970.

Mentor-Marcel, R. A., Bobe, G., Barrett, K. G., Young, M. R., Albert, P. S., Bennink, M.
R, Lanza, E., & Colbum, N. H. (2009) Inflammation-Associated Serum and Colon
Markers as Indicators of Dietary Attenuation of Colon Carcinogenesis in ob/ob Mice.
Cancer Prevention Research 2: 60—69.

Messina, M., McCaskill-Stevens, W., & Larnpe, J. W. (2006) Addressing the soy and
breast cancer relationship: Review, commentary, and workshop proceedings. Journal of
the National Cancer Institute 98: 1275-1284.

Messina, M. J. (1999) Legumes and soybeans: overview of their nutritional profiles and
health effects. American Journal of Clinical Nutrition 70: 4398-4508.

Michels, K. B., Giovarmucci, E., Chan, A. T., Singhania, R., Fuchs, C. S., & Willett, W.
C. (2006) Fruit and vegetable consumption and colorectal adenomas in the nurses' health
study. Cancer Research 66: 3942-3953.

Milde-Langosch, K (2005) The Fos family of transcription factors and their role in
tumourigenesis. European Journal of Cancer 41: 2449-2461.

Milde-Langosch, K., Riethdorf, S., Kraus-Poppinghaus, A., Riethdorf, L., & Loning, T.
(2001) Expression of cyclin-dependent kinase inhibitors p16(MTS1), p21(WAF1), and
p27(KIP1) in HPV-positive and HPV-negative cervical adenocarcinomas. Virchows
Archiv 439: 55-61.

Mori, H., Hata, K., Yamada, Y., Kuno, T., & Hara, A. (2005) Significance and role of
early-lesions in experimental colorectal carcinogenesis. Chemico-Biological Interactions
155: 1-9.

Morin, P. J ., Sparks, A. B., Korinek, V., Barker, N., Clevers, H., Vogelstein, B., &
Kinzler, K. W. (1997) Activation of beta-catenin-ch signaling in colon cancer by
mutations in beta-catenin or APC. Science 275: 1787-1790.

Mosca, L., De Marco, C., Visioli, F., & Carmella, C. (2000) Enzymatic assay for the

determination of olive oil polyphenol content: Assay conditions and validation of the
method. Journal of Agricultural and Food Chemistry 48: 297-301.

206

Naczk, M., & Shahidi, F. (2006) Phenolics in cereals, fruits and vegetables: Occurrence,
extraction and analysis. Journal of Pharmaceutical and Biomedical Analysis 41: 1523-
1542.

Nemeth, K., Plumb, G. W., Berrin, J. G., Juge, N., Jacob, R., Naim, H. Y., Williamson,
G., Swallow, D. M., & Kroon, P. A. (2003) Deglycosylation by small intestinal epithelial
cell beta-glucosidases is a critical step in the absorption and metabolism of dietary
flavonoid glycosides in humans. European Journal of Nutrition 42: 29-42.

Nordling, M. M., Glinghammar, B., Karlsson, P. C., de Kok, T., & Rafter, J. J. (2003)
Effects on cell proliferation, activator protein-1 and genotoxicity by fecal water from

patients with colorectal adenomas. Scandinavian Journal ofGastroenterology 38: 549-
555.

Olthof, M. R., Hollman, P. C. H., Buijsman, M., van Amelsvoort, J. M. M., & Katan, M.
B. (2003) Chlorogenic acid, quercetin-3-rutinoside and black tea phenols are extensively
metabolized in humans. Journal of Nutrition 133: 1806-1814.

Olthof, M. R., Holhnan, P. C. H., & Katan, M. B. (2001) Chlorogenic acid and caffeic
acid are absorbed in humans. Journal of Nutrition 131: 66-71.

Olthof, M. R., Holhnan, P. C. H., Vree, T. B., & Katan, M. B. (2000) Bioavailabilities of
quercetin-3-glucoside and quercetin-4 '-g1ucoside do not differ in humans. Journal of
Nutrition 130: 1200-1203.

Oomah, B. D., Cardador-Martinez, A., & Loarca-Pina, G. (2005) Phenolics and
antioxidative activities in common beans (Phaseolus vulgaris L). Journal of the Science
of Food and Agriculture 85: 935-942.

Parkin, D. M., Bray, F., Ferlay, J ., & Pisani, P. (2005) Global cancer statistics, 2002. CA-
A Cancer Journal for Clinicians 55: 74-108.

Petri, N., Tannergren, C., Holst, B., Mellon, F. A., Bao, Y. P., Plumb, G. W., Bacon, J .,
O'Leary, K. A., Kroon, P. A., Knutson, L., Forsell, P., Eriksson, T., Lennernas, H., &
Williamson, G. (2003) Absorption/metabolism of sulforaphane and quercetin, and
regulation of phase H enzymes, in human jejunum in vivo. Drug Metabolism and
Disposition 31: 805-813.

Plumb, G. W., Garcia-Conesa, M. T., Kroon, P. A., Rhodes, M., Ridley, S., &
Williamson, G. (1999) Metabolism of Chlorogenic acid by human plasma, liver, intestine
and gut microflora. Journal of the Science of Food and Agriculture 79: 390-392.

Pretlow, T. P., Barrow, B. J ., Ashton, W. S., Oriordan, M. A., Pretlow, T. G., Jurcisek, J.

A., & Stellato, T. A. (1991) Aberrant Crypts - Putative Preneoplastic Foci in Human
Colonic Mucosa. Cancer Research 51: 1564-1567.

207

Rafter, J. J ., Child, R, Anderson, A. M., Alder, R., Eng, V., & Bruce, W. R. (1987)
Cellular Toxicity of Fecal Water Depends on Diet. American Journal of Clinical
Nutrition 45: 559-563.

Rahman, I., Biswas, S. K., & Kirkham, P. A. (2006) Regulation of inflammation and
redox signaling by dietary polyphenols. Biochemical Pharmacology 72: 1439-1452.

Raju, J ., McCarthy, 3., & Bird, R. P. (2002) Steady state levels of transforming growth
factor-beta 1 and -beta 2 mRNA and protein expression are elevated in colonic tumors in

vivo irrespective of dietary lipids intervention. International Journal of Cancer 100: 635-
641.

Ranilla, L. G., Genovese, M. I., & Lajolo, F. M. (2007) Polyphenols and antioxidant
capacity of seed coat and cotyledon from Brazilian and Peruvian bean cultivars
(Phaseolus vulgaris L.). Journal of Agricultural and Food Chemistry 55: 90-98.

Rechner, A. R., Smith, M. A., Kuhnle, G., Gibson, G. R., Debnam, E. S., Srai, S. K. S.,
Moore, K. P., & Rice-Evans, C. A. (2004) Colonic metabolism of dietary polyphenols:

Influence of structure on microbial fermentation products. Free Radical Biology and
Medicine 36: 212-225.

Rechner, A. R., Spencer, J. P. E., Kuhnle, G., Hahn, U., & Rice-Evans, C. A. (2001)
Novel biomarkers of the metabolism of caffeic acid derivatives in vivo. Free Radical
Biology and Medicine 30: 1213-1222.

Record, 1. R., McInemey, J. K., Noakes, M., & Bird, A. R. (2003) Chocolate
consumption, fecal water antioxidant activity, and hydroxyl radical production. Nutrition
and Cancer-An International Journal 47: 131-135.

Reddy, B. S. (2004) Studies with the azoxymethane-rat preclinical model for assessing
colon tumor development and chemoprevention. Environmental and Molecular
Mutagenesis 44: 26-35.

Reeves, P. G., Nielsen, F. H., & F ahey, G. C. (1993) Ain-93 Purified Diets for
Laboratory Rodents - Final Report of the American Institute of Nutrition Ad Hoc Writing
Committee on the Reforrnulation of the AIN-76a Rodent Diet. Journal of Nutrition 123:
1939-1951.

Ribeiro, L. R., & Salvadori, D. M. F. (2003) Dietary components may prevent mutation-
related diseases in humans. Mutation Research-Reviews in Mutation Research 544: 195-
201.

Ribéreau—Gayon, P. (1972) Plant Phenolics. Oliver and Boyd, Edinburgh.

208

 

 

Rice-Evans, C. A., Miller, N. J ., & Paganga, G. (1996) Structure-antioxidant activity
relationships of flavonoids and phenolic acids. Free Radical Biology and Medicine 20:
933-956.

Robards, K. (2003) Strategies for the determination of bioactive phenols in plants, fruit
and vegetables. Journal of Chromatography A 1000: 657-691.

Robbins, R. J. (2003) Phenolic acids in foods: An overview of analytical methodology.
Journal of Agricultural and Food Chemistry 51 : 2866-2887.

Rocha-Guzman, N. E., Gonzalez-Laredo, R. F., Ibarra-Perez, F. J ., Nava-Berumen, C. A.,
& Gallegos-Infante, J. A. (2007) Effect of pressure cooking on the antioxidant activity of

extracts fi'om three common bean (Phaseolus vulgaris L.) cultivars. Food Chemistry 100:
31-35.

Rochfort, S., & Panozzo, J. (2007) Phytochemicals for health, the role of pulses. Journal
of Agricultural and Food Chemistry 55: 7981-7994.

Rommel, A., & Wrolstad, R. E. (1993) Influence of Acid and Base Hydrolysis on the
Phenolic Composition of Red Raspberry Juice. Journal of Agricultural and Food
Chemistry 41: 1237-1241.

Roncucci, L., Medline, A., & Bruce, W. R. ( 1991) Classification of Aberrant Crypt Foci
and Microadenomas in Human Colon. Cancer Epidemiology Biomarkers & Prevention 1:
57-60.

Rondini, E. A. (2006) Black bean and soy flour inhibit experimental colon carcinogenesis
by reducing inflammation and enhancing colonocyte differentiation. Department of Food
Science and Human Nutrition. Michigan State University, East Lansing.

Rondini, L., Peyrat-Maillard, M. N., Marsset-Baglieri, A., & Berset, C. (2002) Sulfated
ferulic acid is the main in vivo metabolite found after short-term ingestion of free ferulic
acid in rats. Journal of Agricultural and Food Chemistry 50: 3037-3041.

Rondini, L., Peyrat-Maillard, M. N., Marsset-Baglieri, A., Fromentin, G., Durand, P.,
Tome, D., Prost, M., & Berset, C. (2004) Bound ferulic acid fi'om bran is more
bioavailable than the free compound in rat. Journal of Agricultural and Food Chemistry
52: 4338-4343.

Saez, E., Rutberg, S. E., Mueller, E., Oppenheim, H., Smoluk, J ., Yuspa, S. H., &
Spiegelman, B. M. (1995) C-Fos Is Required for Malignant Progression of Skin Trunors.
Cell 82: 721-732.

Sandhu, M. S., Dunger, D. B., & Giovarmucci, E. L. (2002) Insulin, insulin-like growth

factor-I (IGF-I), IGF binding proteins, their biologic interactions, and colorectal cancer.
Journal of the National Cancer Institute 94: 972-980.

209

Sangronis, E., Rodriguez, M., Cava, R., & Torres, A. (2006) Protein quality of
germinated Phaseolus vulgaris. European Food Research and Technology 222: 144-148.

Santos, M. R., Rodriguez-Gomez, M. J., Justino, G. C., Charro, N., Florencio, M. H., &
Mira, L. (2008) Influence of the metabolic profile on the in vivo antioxidant activity of

quercetin under a low dosage oral regimen in rats. British Journal of Pharmacology 153:
1 750-1 761 .

Scalbert, A., & Williamson, G. (2000) Dietary intake and bioavailability of polyphenols.
Journal of Nutrition 130: 20738-20858.

Schoen, R. E., Mutch, M., Rall, C., Dry, S. M., Seligson, D., Umar, A., & Pinsky, P.
(2008) The natural history of aberrant crypt foci. Gastrointestinal Endoscopy 67: 1097-
1 102.

Selby, J. V., Friedman, G. D., Quesenberry, C. P., & Weiss, N. S. (1992) A Case Control
Study of Screening Sigrnoidoscopy and Mortality fi'om Colorectal-Cancer. New England
Journal of Medicine 326: 653-657.

Seo, A., & Morr, C. V. (1984) Improved High-Performance Liquid-Chromatographic
Analysis of Phenolic-Acids and Isoflavonoids from Soybean Protein Products. Journal of
Agricultural and Food Chemistry 32: 530-533.

Seoane, J. (2006) Escaping from the TGF beta anti-proliferative control. Carcinogenesis
27: 2148-2156.

Shao, J. Y., Sheng, H. M., Aramandla, R., Pereira, M. A., Lubet, R. A., Hawk, E.,
Grogan, L., Kirsch, I. R., Washington, M. K., Beauchamp, R. D., & DuBois, R. N. (1999)
Coordinate regulation of cyclooxygenase-2 and TGF-beta 1 in replication error-positive
colon cancer and azoxymethane-induced rat colonic tumors. Carcinogenesis 20: 185-191.

Shimelis, E. A., & Rakshit, S. K. (2007) Effect of processing on antinutrients and in vitro
protein digestibility of kidney bean (Phaseolus vulgaris L.) varieties grown in East
Afiica. Food Chemistry 103: 161-172.

Singh, P. N., & Fraser, G. E. (1998) Dietary risk factors for colon cancer in a low-risk
population. American Journal of Epidemiology 148: 761-774.

Singleton VL, R. J. (1965) Colorimetry of total phenolics with phosphomolybdic-
phosphotungstic acid reagents. American Journal of Enology Viticulture 16: 144-158.

Skrede, G., Wrolstad, R. E., & Durst, R. W. (2000) Changes in anthocyanins and

polyphenolics during juice processing of highbush blueberries (Vaccinium corymbosum
L.). Journal of Food Science 65: 357-364.

210

Sosulski, F ., Krygier, K., & Hogge, L. (1982) Free, Esterified, and Insoluble-Bound
Phenolic-Acids .3. Composition of Phenolic-Acids in Cereal and Potato Flours. Journal
of Agricultural and Food Chemistry 30: 337-340.

Sosulski, F., & Youngs, C. G. (1979) Yield and Functional-Properties of Air-Classified
Protein and Starch Fractions fiom 8 Legume Flours. Journal of the American Oil
Chemists Society 56: 292-294.

Sosulski, F. W., & Dabrowski, K. J. (1984) Composition of Free and Hydrolyzable
Phenolic-Acids in the Flours and Hulls of 10 Legume Species. Journal of Agricultural
and Food Chemistry 32: 131-133.

Soto, U., Denk, C., Finzer, P., Hutter, K. J ., zur Hausen, H., & Rosl, F. (2000) Genetic
complementation to non-tumorigenicity in cervical-carcinoma cells correlates with
alterations in AP-l composition. International Journal of Cancer 86: 811-817.

Spencer, J. P. E., Abd E1 Mohsen, M., & Minihane, A. M. (2006) Metabolism of dietary
phytochemicals: A review of the metabolic forms identified in humans. Current Topics in
Nutraceutical Research 4: 187-203.

Stalikas, C. D. (2007) Extraction, separation, and detection methods for phenolic acids
and flavonoids. Journal of Separation Science 30: 3268-3295.

Steinmetz, K. A., & Potter, J. D. (1993) Food-Group Consumption and Colon Cancer in
the Adelaide Case-Control Study .1. Vegetables and Fruit. International Journal of
Cancer 53: 711-719.

Stevenson, D. E., & Hurst, R. D. (2007) Polyphenolic phytochemicals - just antioxidants
or much more? Cellular and Molecular Life Sciences 64: 2900-2916.

Stopera, S. A., Davie, J. R., & Bird, R. P. (1992) Colonic Aberrant Crypt Foci Are
Associated with Increased Expression of C-Fos - the Possible Role of Modified C-Fos
Expression in Preneoplastic Lesions in Colon Cancer. Carcinogenesis 13: 573-578.

Svendsen, C. N. (1993) Multielectrode Array Detectors in Hi gh-Performance Liquid-
Chromatography - a New Dimension in Electrochemical Analysis. Analyst 118: 123-129.

Tabibzadeh, S., & Hemmati-Brivanlou, A. (2006) Lefty at the crossroads of "stemness"
and differentiative events. Stem Cells 24: 1998-2006.

Takahashi, M., & Wakabayashi, K. (2004) Gene mutations and altered gene expression in
azoxymethane-induced colon carcinogenesis in rodents. Cancer Science 95: 475-480.

Takeoka, G. R., Dao, L. T., Full, G. H., Wong, R. Y., Harden, L. A., Edwards, R. H., &
Berrios, J. D. (1997) Characterization of black bean (Phaseolus vulgaris L.)
anthocyanins. Journal of Agricultural and Food Chemistry 45: 3395-3400.

211

Tharanathan, R. N., & Mahadevamma, S. (2003) Grain legumes - a boon to human
nutrition. Trends in Food Science & Technology 14: 507-518.

Thompson, M. D., Brick, M. A., McGinley, J. N., & Thompson, H. J. (2009) Chemical
Composition and Mammary Cancer Inhibitory Activity of Dry Bean. Crop Science 49:
179-186. .

Thompson, M. D., Thompson, H. J ., Brick, M. A., McGinley, J. N., Jiang, W. Q., Zhu, Z.
J ., & Wolfe, P. (2008) Mechanisms Associated with Dose-Dependent Inhibition of Rat

Mammary Carcinogenesis by Dry Bean (Phaseolus vulgaris L.). Journal of Nutrition
138: 2091-2097.

Towo, E. E., Svanberg, U., & Ndossi, G. D. (2003) Effect of grain pre-treatment on
different extractable phenolic groups in cereals and legumes commonly consumed in
Tanzania. Journal of the Science of Food and Agriculture 83: 980-986.

Tsao, R., & Deng, Z. Y. (2004) Separation procedures for naturally occuning antioxidant
phytochemicals. Journal of Chromatography B-Analytical Technologies in the
Biomedical and Life Sciences 812: 85-99.

U. S. D. A. / A. R. S. (2007) USDA Database for the Flavonoid Content of Selected
Foods: Release 2.1. USDA/ARS, Beltsville.

U. S. D. A. / A. R. S. (2009) USDA National Nutrient Database for Standard Reference,
Release 22. Nutrient Data Laboratory Home Page, http://www.ars.usda.gov/ba/bhnrc/ndl.

Ulloa, L., & Tabibzadeh, S. (2001) Lefty inhibits recepter-regulated smad
phosphorylation induced by the activated transforming growth factor-beta receptor.
Journal of Biological Chemistry 276: 21397-21404.

(USDA), C. A. D. C. E. R. S. E. U. S. D. o. A. (1997) Dry edible beans: U.S. capita use
for selected classes, 1970-1997. US. Government Printing Office, Washington, DC.

van Dam, R., & Seidell, J. C. (2007) Carbohydrate intake and obesity. European Journal
of Clinical Nutrition 61: S75-S99.

Venn, B. J ., & Mann, J. I. (2004) Cereal grains, legumes and diabetes. European Journal
of Clinical Nutrition 58: 1443-1461.

Villegas, R., Gao, Y. T., Yang, G., Li, H. L., Elasy, T. A., Zheng, W., & Shu, X. 0.

(2008) Legume and soy food intake and the incidence of type 2 diabetes in the Shanghai
Women's Health Study. American Journal of Clinical Nutrition 87: 162-167.

212

Wali, R. K., Skarosi, 8., Hart, J ., Zhang, Y. C., Dolan, M. E., Moschel, R. C., Nguyen, L.,
Mustafi, R., Brasitus, T. A., & Bissonnette, M. (1999) Inhibition of O-6-methylguanine-
DNA methyltransferase increases azoxymethane-induced colonic tumors in rats.
Carcinogenesis 20: 2355-2360.

Walle, T. (2004) Absorption and metabolism of flavonoids. Free Radical Biology and
Medicine 36: 829-837.

Wang, J., Sun, L. 2., Myeroff, L., Wang, X. F., Gentry, L. E., Yan, J. H., Liang, J. R.,
Zborowska, E., Markowitz, S., Willson, J. K. V., & Brattain, M. G. (1995)
Demonstration That Mutation of the Type-II Transforming Growth-Factor-Beta Receptor
Inactivates Its Tumor-Suppressor Activity in Replication Error-Positive Colon-
Carcinoma Cells. Journal of Biological Chemistry 270: 22044-22049.

Ward, J. M., Yamamoto, R. S., & Brown, C. A. (1973) Pathology of Intestinal
Neoplasms and Other Lesions in Rats Exposed to Azoxymethane. Journal of the National
Cancer Institute 51 : 1029-1039.

Weber, R. V., Stein, D. E., Scholes, J ., & Kral, J. G. (2000) Obesity potentiates AOM-
induced colon cancer. Digestive Diseases and Sciences 45: 890-895.

Weitz, J ., Koch, M., Debus, J ., Hohler, T., Galle, P. R., & Buchler, M. W. (2005)
Colorectal cancer. Lancet 365: 153-165.

Wijnands, M. V. W., van Erk, M. J ., Doombos, R. P., Krul, C. A. M., & Woutersen, R.
A. (2004) Do aberrant crypt foci have predictive value for the occurrence of colorectal

tumours? Potential of gene expression profiling in tumours. Food and Chemical
Toxicology 42: 1629-1639.

Willett, W. C. (2001) Diet and cancer: One view at the start of the millennium. Cancer
Epidemiology Biomarkers & Prevention 10: 3-8.

Winham, D. M., Hutchins, A. M., & Johnston, C. S. (2007) Pinto bean consumption
reduces biomarkers for heart disease risk. Journal of the American College of Nutrition
26: 243-249.

Xiao, C. W. (2008) Health effects of soy protein and isoflavones in humans. Journal of
Nutrition 138: 12448-1249S.

Xu, B. J ., & Chang, S. K. C. (2007) A comparative study on phenolic profiles and

antioxidant activities of legumes as affected by extraction solvents. Journal of Food
Science 72: $159-$166.

Xu, B. J ., & Chang, S. K. C. (2008) Effect of soaking, boiling, and steaming on total

phenolic content and antioxidant activities of cool season food legumes. Food Chemistry
110: 1-13.

213

Yamada, Y., Yoshimi, N., Hirose, Y., Matsunaga, K., Katayama, M., Sakata, K.,
Shimizu, M., Kuno, T., & Mori, H. (2001) Sequential analysis of morphological and
biological properties of beta-catenin-accumulated crypts, provable premalignant lesions

independent of aberrant crypt foci in rat colon carcinogenesis. Cancer Research 61:
1 874-1 878.

Yamashita, N., Minamoto, T., Ochiai, A., Onda, M., & Esumi, H. (1995) Frequent and
Characteristic K-Ras Activation and Absence of P53 Protein Accumulation in Aberrant
Crypt Foci of the Colon. Gastroenterology 108: 434-440.

Yang, S. Q., Lin, H. 2., Lane, M. D., Clemens, M., & Diehl, A. M. (1997) Obesity
increases sensitivity to endotoxin liver injury: Implications for the pathogenesis of

steatohepatitis. Proceedings of the National Academy of Sciences of the United States of
America 94: 2557-2562.

Yang, S. Q., Lin, H. 2., Manda], A. K., Huang, J. W., & Diehl, A. M. (2001) Disrupted
signaling and inhibited regeneration in obese mice with fatty livers: Implications for
nonalcoholic fatty liver disease pathophysiology. Hepatology 34: 694-706.

Yokota, T., Sugano, K., Kondo, H., Saito, D., Sugihara, K., Fukayama, N., Ohkura, H.,
Ochiai, A., & Yoshida, S. (1997) Detection of aberrant crypt foci by magnifying
colonoscopy. Gastrointestinal Endoscopy 46: 61-65.

Zedeck, M. S. (1980) Colon Carcinogenesis and the Role of Dehydrogenase Enzyme-
Activity - Inhibition of Tumori genesis by Pyrazole. Preventive Medicine 9: 346-351.

Zhou, S. B., Buckhaults, P., Zawel, L., Bunz, F ., Riggins, G., Dai, J. L., Kern, S. E.,
Kinzler, K. W., & Vogelstein, B. (1998) Targeted deletion of Smad4 shows it is required
for transforming growth factor beta and activin signaling in colorectal cancer cells.
Proceedings of the National Academy of Sciences of the United States of America 95:
2412-2416.

Zhu, Y. A., Richardson, J. A., Parada, L. F., & Graff, J. M. (1998) Smad3 mutant mice
develop metastatic colorectal cancer. Cell 94: 703-714.

214

11111”

11

IIIWI

1 VI
3 030

3 129

H
Ill
"
I"
H"