 

LIBRARY
Michigan State
University

 

 

 

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

 

AUG 2 8 2004

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJClRC/DateDuepGS-p. 1 5

INHIBITION OF INTESTINAL TUMORIGENESIS IN APCMIN MICE BY TART
CHERRY ANTHOCYANINS AND SULINDAC

BY

Soo-Young Kang

A DISSERTATION

Submitted to
Michigan State University
In partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Food Science and Human Nutrition
Institute of Environmental Toxicology

2002

ABSTRACT

INHIBITION OF INTESTINAL TUMORIGENESIS 1N ABCMIN MICE BY TART
CHERRY ANTHOCYANINS AND SULINDAC

By

Soo-Young Kang

Anthocyanins, a member of the ﬂavonoid family of phytochemicals,
previously have been found to inhibit the activities of cyclooxygenase (COX)
enzymes. We hypothesized that tart cherry anthocyanins suppress intestinal
tumorigenesis in ApcMin mice because compounds that inhibit COX potentially are
protective against colon cancer development. A series of experiments were
conducted to assess the inﬂuence of anthocyanins and a nonsteroidal anti-
inﬂammatory drug, sulindac, on intestinal tumor development in ApcMin mice. First,
we examined time frame of intestinal adenoma development in ApcMi" mice. The
numbers and size of adenomas in small intestine increased with age, whereas tumor
number in the cecum and colon did not increase signiﬁcantly with age. The volume
of colonic tumors increased signiﬁcantly aﬁer 75 days of age. These results
characterized the early and rapid development and progression of intestinal tumors in
ApcMi“ mice. We then conducted experiments to test the potential of anthocyanins to
inhibit intestinal tumor development in ApcMin mice and growth of human colon
cancer cell lines. Mice consuming anthocyanins and cyanidin in drinking water and
a 20% tart cherry diet had fewer and smaller tumors in the cecum, whereas mice

consuming sulindac in drinking water had more cecal adenomas. The mean diameter

of adenomas in the small intestine was smaller in mice consuming sulindac, and

larger in mice consuming the tart cherry diet, compared to mice consuming the
control diet. Anthocyanins and cyanidin also reduced the growth of human colon
cancer cell lines HT 29 and HCT 116. Based on these results, a second feeding
study was conducted to determine effects of dietary anthocyanins and sulindac on 1)
Min

intestinal tumorigenesis in Apc mice, 2) expression and cellular localization of B-

catenin, and 3) expression of target genes of the Wnt/B-catenin signaling pathway
and global gene expression proﬁles. APCMin mice consuming dietary anthocyanins
had signiﬁcantly fewer adenomas in all three intestinal sections compared to mice
consuming the control diet. Small intestinal adenomas in mice consuming sulindac
were fewer in number and smaller in diameter compared to mice consuming
anthocyanins, but mice consuming sulindac had more cecal adenomas compared to
mice consuming anthocyanins. Sulindac did not inﬂuence the expression of B-
catenin protein in intestinal epithelium, whereas anthocyanins increased B-catenin
expression in the small intestine and cecum. Expression of mRN A of Apc, B-catenin
and most of the target genes of Wnt/B-catenin signaling was not altered by treatment.
cDNA microarray analysis identiﬁed several genes that were differentially expressed
in response to treatment with anthocyanins or sulindac. Based on cellular expression
of B-catenin and cDNA microarray analysis, we concluded that tumor inhibition in
ApcMi" mice by anthocyanins is unlikely to involve COX inhibition or inhibition of
the Wnt/B-catenin signaling pathway. Taken together, these results suggest that tart

cherry anthocyanins and their aglycone, cyanidin, may reduce the risk of colon

cancer by mechanisms that are unidentiﬁed at this time.

T 0 my family

iv

ACKNOWLEDGEMENTS

I would like to thank Dr. Les Bourquin for his guidance throughout the
program as a major advisor. I appreciate his time and special effort put towards my
research. I would also like to extend my gratitude to Dr. Maurie Bennink for his
precious advice and kindness. Special thanks extend to Drs. Dale Romsos and Norm
Hord for serving as the guidance committee members. And I am truly grateful for
Dr. Steve Bursian for his support. He was always with me for the highs and lows
and encouraged me to move on especially in many tough times of discouragement
and depression.

I thank Drs. Muralee Nair and Navindra Seeram for providing tart chenies
and compounds for the experiments and their valuable suggestions during the lab
meetings. Dr. John LaPres deserves sincere thanks for allowing me to use his lab
and expertise for the microarray experiments. He has provided great help and
support when I needed his advice.

I would like to express my gratitude to Bing Wang, Elizabeth Rondini,
Elizabeth Gignac and Crystal Ybarra for helping me with mice care and sacriﬁce.
My sincere appreciation is extended to friends from the department. Drs. Mridvika,
Anahita Mistry, Miyoung Jang, Sang-J in Chung and Joowon Lee have been source
of cheer and advice throughout the program. I appreciate their friendship and advice.
Thanks to Megan Waynar for her friendship and all those good times during the
stressful days of exams and experiments.

There were times that I thought I was walking in the tunnel of darkness and

there were always my parents, Cheonun Kang and Gunja Moon, who did listen and

pray for me to step forward. I thank them for their pride and faith in me. I also want
to thank my late father-in-law for watching us up in heaven and mother-in-law for
taking good care of Brandon and Chloe during my comprehensive and defense. My
sisters-in-law, especially J inhee and Sunhee were great help on my busy days and I
am thankful for their love on my two kids and compassion inherited from my
mother-in-law. For my loving husband, J ae-ug, no words can easily describe how
much I appreciate him for his love, affection and trust on me. I couldn’t have been
through it without his limitless support, compliments and humor. At last, my two
kids, Brandon and Chloe were always making me feel truly blessed and strong

enough to shelter them and to face myself with no shame and fear.

vi

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................................. ix
LIST OF FIGURES ........................................................................................................... xi
CHAPTER I. INTRODUCTION ........................................................................................ 1
CHAPTER H. REVIEW OF LITERATURE ...................................................................... 6
l. Colon cancer .................................................................................................................... 7
2. The multi-step carcinogenic process ............................................................................... 8
3. Phytochemicals as anti-carcinogenic factors .................................................................. 9
4. ApcMm mouse model ..................................................................................................... 17
5. Nonsteroidal anti-inﬂammatory drugs (N SAIDs) and colorectal cancer ...................... 20
6. Genetics of intestinal neoplasia: APC mutation and its consequences ......................... 23
7. Hypotheses and research objectives .............................................................................. 33

CHAPTER III. ONTOGENY OF INTESTINAL ADENOMA DEVELOPMENT IN

APCMIN MICE ................................................................................................ 36
1. Abstract ......................................................................................................................... 37
2. Introduction ................................................................................................................... 38
3. Materials and methods .................................................................................................. 40
4. Results ........................................................................................................................... 42
5. Discussion ..................................................................................................................... 5 1

CHAPTER IV. TART CHERRY ANTHOCYANINS INHIBIT TUMOR
DEVELOPMENT IN ARCMIN MICE AND REDUCE GROWTH OF HUMAN

COLON CANCER CELLS .............................................................................................. 54
1. Abstract ......................................................................................................................... 55
2. Introduction ................................................................................................................... 56
3. Materials and methods .................................................................................................. 58
4. Results ........................................................................................................................... 63
5. Discussion ..................................................................................................................... 73

CHAPTER V. TART CHERRY ANTHOCYANIN S REDUCE INTESTINAL
TUMORIGENESIS AND ALTER INTESTINAL EPITHELIAL CELL GENE

vii

EXPRESSION, BUT DO NOT INFLUENCE INTRACELLULAR DISTRIBUTION

OF B-CATENIN IN INTESTINE OF APCMIN MICE ...................................................... 77
1. Abstract ..................................................................................................................... 78
2. Introduction ............................................................................................................... 8O
3. Materials and methods .............................................................................................. 83
4. Results ....................................................................................................................... 93
5 . Discussion .............................................................................................................. 13 1
CHAPTER VI. OVERALL SUMMARY AND CONCLUSIONS ................................ 147
APPENDICES ............................................................................................................... 154
LITERATURE CITED ................................................................................................... 179

viii

TABLE
TABLE
TABLE
TABLE

TABLE

TABLE

TABLE
TABLE

TABLE

TABLE

TABLE

TABLE
TABLE

TABLE
TABLE

TABLE

TABLE
TABLE
TABLE

TABLE
TABLE
TABLE

LIST OF TABLES

1: Lengths of small intestine and colon of all mice ........................................... 44
2: Composition of the experimental diets ......................................................... 59
3: Average intensity scores of B-catenin expression in intestinal sections ..... 104

4: Expression of B- catenin and target genes of Wnt/B- catenin signaling 1n
intestinal sections of ApcMm mice ................................................................ 109

5: Expression of 8 genes inﬂuenced by treatment both in small intestine
and colon ...................................................................................................... 111

6: Expression of 12 genes inﬂuenced by treatment both in small intestine
and cecum .................................................................................................... 1 12

7: Expression of 20 genes inﬂuenced by treatment both in cecum and colon 1 l3

8: List of genes up-regulated by anthocyanins vs control and sulindac in
cecum ........................................................................................................... 1 15

9: List of genes down-regulated by anthocyanins vs control and sulindac in

cecum ........................................................................................................... 118
10: List of genes up- -regulated by anthocyanins vs control and sulindac 1n

colon ............................................................................................................. 120
11: List of genes down-regulated by anthocyanins vs control and sulindac

in colon ......................................................................................................... 122
12: List of genes up-regulated by sulindac vs control in small intestine ........ 124
13: List of genes up-regulated by sulindac and anthocyanins in small

intestine ........................................................................................................ 126
14: List of genes down-regulated by sulindac vs control in small intestine 127
15: List of genes down-regulated by sulindac and anthocyanins in small

intestine ........................................................................................................ 129
A1: Carrier status, body weights, and tissue lengths of mice from

experiment I ................................................................................................. 155
A2: Numbers and sizes of adenomas in ApcMin mice from experiment I ........ 158
A3: Weekly body weights (g) of ApcMin mice from the experiment 11 ........... 160
A4: Tissue lengths and numbers and sizes of adenomas in ApcMin mice

from experiment II ....................................................................................... 162
A5: Sex, carrier status, and issue lengths of mice from experiment III .......... 164
A6: Weekly body weights (g) of mice from experiment III ........................... 168
A7: Numbers and sizes of adenomas in ApcMin mice from experiment III ..... 172

ix

TABLE A8: Intensity scores of B-catenin expression in small intestine of mice from
experiment III ............................................................................................... 175

TABLE A9: Intensity scores of B-catenin expression in colon of mice from
experiment HI ............................................................................................... 177

TABLE A10: Intensity scores of B-catenin expression in cecum of mice from
experiment 1]] ............................................................................................... 178

FIGURE
FIGURE
FIGURE
FIGURE
FIGURE
FIGURE
FIGURE

FIGURE

FIGURE

FIGURE
FIGURE
FIGURE

FIGURE

FIGURE

FIGURE

FIGURE

FIGURE

FIGURE

FIGURE
FIGURE

FIGURE

LIST OF FIGURES

1: The chemical structures of ﬂavonoids. ................................................... 12-13
2: Anthocyanins ﬁom tart cherries .................................................................. 14
3: A genetic model for colorectal tumorigenesis .............................................. 25
4: Schematic of the APC protein primary structure .......................................... 26
5: APC-and NSAID-mediated suppression of intestinal tumorigenesis ........... 32
6: Inﬂuence of age on body weights ................................................................. 43
7: Inﬂuence of age on adenoma numbers in proximal (white), middle

(gray) and distal (hatched) segments of the small intestine and total
(black) numbers in the small intestine ........................................................... 45

8: Inﬂuence of age on the average diameter of small intestinal adenomas ...... 46

9: Inﬂuence of age on the average adenoma numbers in the cecum (gray)

and colon (black) ............................................................................................ 48
10: Inﬂuence of age on adenoma volume (mm3) in the cecum ....................... 49
11: Inﬂuence of age on adenoma volume (mm3) in the colon ......................... 50
12: Inﬂuence of anthocyanins, cyanidin, tart cherries and sulindac on body

weights (g) of mice throughout the experimental period .............................. 65
13: Inﬂuence of anthocyanins, cyanidin, tart cherries and sulindac on

adenoma numbers in cecum (gray) and colon (black) ................................... 66
14: Inﬂuence of anthocyanins, cyanidin, tart cherries and sulindac on

adenoma volumes (mm3) in cecum (gray) and colon (black) ....................... 67
15: Inﬂuence of anthocyanins, cyanidin, tart cherries and sulindac on

adenoma numbers in small intestine .............................................................. 68
16: Inﬂuence of anthocyanins, cyanidin, tart cherries and sulindac on

average diameters (mm) of adenomas in small intestine ............................... 69
17: Inﬂuence of anthocyanins on growth of human cancer cells, HCT 116

(gray) and HT 29 (black) ............................................................................... 71
18: Inﬂuence of cyanidin on grth of human cancer cells, HCT 116

(gray) and HT 29 (black) ............................................................................... 72
19: Aminoallyl labeling loop of RNA for microarray analysis ........................ 90

20: Inﬂuence of treatment on body weights (g) of mice throughout the
experimental period ....................................................................................... 95

21: Inﬂuence of anthocyanins and sulindac on adenoma number in cecum
(gray) and colon (black) ........ ' ......................................................................... 96

xi

FIGURE 22: Inﬂuence of anthocyanins and sulindac on total adenoma volumes

(mm3) in cecum (gray) and colon (black) ...................................................... 97
FIGURE 23: Inﬂuence of anthocyanins and sulindac on adenoma number in small
' intestine. ......................................................................................................... 98
FIGURE 24: Inﬂuence of anthocyanins and sulindac on mean diameters (mm) of
adenomas in small intestine ........................................................................... 99
FIGURE 25: Expression of B-catenin in mice consuming the control diet ................... 101
FIGURE 26: Expression of B—catenin in mice consuming sulindac .............................. 102
FIGURE 27: Expression of B-catenin in mice consuming anthocyanins ..................... 103
FIGURE 28: Numbers of genes inﬂuenced by any treatment (P<0.05) in intestinal
sections of APCM‘" mice ............................... ’ ............................................... l 10

xii

CHAPTER I

INTRODUCTION

Colon cancer is the second most common cause of cancer mortality and the
fourth most common in incidence in the United States (American Cancer Society,
2000). It has been estimated that dietary factors may account for 29 - 41% of
cancers at all sites and that 66 - 75% of colon cancer is preventable by diet
(WCRF/AICR, 1998). Results from epidemiological studies have shown that
consuming fruits and vegetables is associated with reduced incidences of various
cancers including colon cancer. The inverse association between cancer and fruit
and vegetable intake may be due to number of components, including micronutrients
and phytochemicals, contained in fruits and vegetables. Phytochemicals are deﬁned
as the naturally occurring, non-nutritive substances present in plant foods (Caragay,
1992). Studies have found that approximately 14 classes of dietary phytochemicals
are present in common foods. These include sulﬁdes, phytates, ﬂavonoids,
glucarates, carotenoids, coumarins, mono-terpenes, tri-terpenes, lignans, phenolic
acids, indoles, isothiocyanates, phthalides, and polyacetylenes. Food sources that are
rich in ﬂavonoids are often found to be protective against colon cancer in animal
models and humans.

Anthocyanins, members of the ﬂavonoid family of phytochemicals, are
widely distributed in plants including beans, fruits, and vegetables, suggesting that
plant-based diets can provide considerable amounts of anthocyanins (Harbome and
Grayer, 1988). Anthocyanins primarily occur in plants as glycosides. C yanidin is
the major anthocyanin aglycone in cherries. Montmorency and BalatonTM cherries

contain 0.40 and 0.80 mg/g, respectively, of anthocyanins (Wang et al., 1999a).

These anthocyanins were later found to be antioxidants and inhibit cyclooxygenase
(COX) enzymes (Wang et al., 1999b). Cyanidin was intermediate in efﬁcacy for
inhibition of COX-1 and COX-2 when compared with aspirin and other non-steroidal
anti-inﬂammatory drugs (N SAIDs) such as ﬂurbiprofen and ibuprofen. Compounds
that inhibit COX potentially are protective against colon cancer development since it
has been demonstrated that NSAIDs inhibit the growth of colon tumors in animal
models and reduce the risk of colon cancer in humans (Giovannucci et al., 1995;
Labayle et al., 1991). Since anthocyanins have COX-2 inhibiting activity, this
suggests they may have anticarcinogenic effects in the intestine.

The ApcMin mouse strain is a mutant mouse lineage predisposed to multiple
intestinal neoplasia (Min). This phenotype is the consequence of a mutation in the
murine homolog of the adenomatous polyposis coli (APC) gene (Su et al., 1992).
The APC gene is also mutated in humans who develop sporadic colon cancer as well
as persons with familial adenomatous polyposis (FAP), an autosomal dominantly
inherited disease that predisposes to colorectal cancer. The primary phenotype of
mice carrying this mutation is the development of multiple intestinal adenomas,
which progress to adenocarcinomas of the intestine in older mice. Min is transmitted
by affected mice to 50% of the progeny with an unbiased sex distribution, as is
characteristic of a fully penetrant autosomal dominant trait (Moser, 1990).

Research on the effects of NSAIDs sulindac and piroxicam on tumor
development in ApcMin mice showed that they reduced the incidence of intestinal

tumors (Boobol et al., 1996; Jacoby et al., 1996). Studies have shown that NSAIDs

possessing COX-2 inhibitory activity inhibited small intestinal tumori genesis, but
not necessarily colonic tumorigenesis, in ApcMin mice (McEntee et al., 1999).

In most cases, colon carcinogenesis depends on mutation of the APC gene
and this gene is considered a gatekeeper in the carcinogenic process (Kinzler and
Vogelstein, 1996). A major function of the APC protein is to bind cellular B-catenin
and facilitate its degradation (Rubinfeld et al., 1993). Most mutations in the human
APC gene have been found to occur between codons 1286 and 1513, which is within
a region encoding B-catenin binding sites (Nakamura, 1993). These mutations
generally result in the production of a truncated APC protein that cannot facilitate the
degradation of B-catenin, suggesting that mutation of APC is essential to the process

of intestinal tumor development by causing a loss of regulation of B-catenin

expression. B-catenin functions as a component of the E-cadherin adhesion complex
and also as a transcriptional regulator in a signaling pathway by binding T cell factor
(Tet) transcription factor (Morin et al., 1997). Aberrant accumulation of B-catenin
can inﬂuence trancriptional activation and expression of target genes in this signaling
pathway. A few target genes in this pathway such as peroxisome proliferator
activated receptor 8 (PPARO), c-MYC, cyclin D1 (CLNDl) and matrilysin (matrix
metalloproteinase 7) recently have been identiﬁed (He et al., 1999; He et al, 1998;
Tetsu and McCormick, 1999; Crawford et al., 1999).

Cherries, anthocyanins and sulindac may inﬂuence tumorigenesis in the
intestinal tract by modulating expression of genes involved in the carcinogenesis.

The purpose of this research is to determine the effects of tart cherry anthocyanins

and the NSAID sulindac on the development of intestinal adenomas and on gene

expression in intestine, cecum and colon of ApcM‘n mice.

CHAPTER II

LITERATURE REVIEW

Coloréfancer

Colon cancer is the third leading cause of cancer mortality (after lung and
stomach cancers) worldwide, with mortality estimated at 492,000 in 2001 (IARC,
2001). In the United States, colon cancer is the second most common cause of
cancer mortality (after lung cancer) and is the fourth most common form of cancer in
terms of incidence (after breast, prostate, and lung cancers) (American Cancer
Society, 2000). Five-year survival following the diagnosis of colon cancer is
approximately 55% in the United States and the rest of the developed world (Potter
et al., 1993).

Numerous studies have characterized the pathogenesis of colorectal cancer as
a multistage process that begins with hyperproliferation of colonic mucosal epithelial
cells. The intestinal epithelium has a complex architecture and undergoes
continuous turnover. Maintenance of crypt structures requires control mechanisms
that balance the rates of cell proliferation and cell death. Failure of the control
mechanisms results in disorganization of the tissue and abnormal tissue growth.
Normally, epithelial proliferation takes place at the base of mucosal glands, called
the crypts of Lieberkuhn (Wright and Irvin, 1982). Quiescent (G0) cells then migrate
to the superﬁcial or luminal zone of the crypt, where they undergo apoptosis
(programmed cell death) (Potten, 1996; 1997).

Cell kinetic studies of rectal mucosal biopsies determined that patients at
increased risk of developing colorectal cancer have an expansion of the proliferative

compartment and distribution of S-phase cells toward the lumenal zone of the crypt

(Terpstra et al., 1987). Hyperproliferation may contribute to mutations that lead to
the formation of adenomas, the precursor lesions to colonic carcinoma. To prevent
development of colon cancer, it has been suggested that the multistage
carcinogenesis process must be delayed or pre-empted. This could be achieved by
modifying the diet to include more cancer preventive foods, since it has been
estimated that 66 - 75% of colon cancer is preventable by adequate diets
(WCRF/AICR, 1998). Indeed, results from epidemiological studies have shown that
consuming diets rich in fruits and vegetables is associated with reduced incidences of

various cancers including colon cancer (Willet, 2001).

The Multi-Step Carcinogenic Process

Carcinogenesis is a multistep process. Moolgavkar and Knudson (1981)
suggested two distinct stages, preneoplastic and malignant steps. Miller and Miller
(1981) summarized the carcinogenisis process as follows: “Initiation” is the cellular
change forming the ﬁrst preneoplastic (initiated) cell in a tissue. The chemically
induced initial step from a normal cell to a premalignant cell has been shown to be
irreversible in mouse skin (Berenblum and Shubik, 1947). “Promotion” can be
deﬁned as the action of any compound that increases the tumor size by expanding
the pool of initiated cells. Treatment with promoting substances shortens the long
latency of tumor development and the action of these substances is reversible (Miller
and Miller, 1981). The irreversible processes of visible tumor formation thereafter
can be termed “progression”. Progression can be considered the terminal stage of

neoplastic development (Pitot, 1989).

Phytochemicals as Anti-Carcinogenic Factors

It has been estimated that dietary factors may account for 29 - 41% of cancers
at all sites and that 66 - 75% of colon cancer is preventable by diet (WCRF/AICR,
1998). Several human cohort and case-control studies for a number of common
cancer sites have shown that increased intakes of fruits and vegetables are associated
with decreased cancer incidence and mortality rates (Doll, 1990; Ocke et al, 1997).
The inverse association between cancer risk and fruit and vegetable intake may be
due to a number of components, including micronutrients and phytochemicals,
contained in fruits and vegetables.

Phytochemicals are deﬁned as the naturally occurring non—nutritive
substances present in plant foods (Caragay, 1992). Studies have found that
approximately 14 classes of dietary phytochemicals are present in common foods.
These include sulﬁdes, phytates, ﬂavonoids, glucarates, carotenoids, coumarins,
mono-terpenes, tri-terpenes, lignans, phenolic acids, indoles, isothiocyanates,
phthalides, and polyacetylenes. Some of these components may potentially block
metabolic pathways leading to carcinogenesis and have been found to decrease
cancer risk in experimental animals. The effect of the substances may depend on
animal species, structure of the carcinogen, dose regimen, timing of dosages, and
many other experimental parameters. The effective doses for some anticarcinogens
may exceed what humans ingest from plant sources. On the other hand, the
combination of several anticarcinogens at sub-effective levels has been shown to

result in antitumor effects in some cases (Renner, 1990).

Food sources that are rich in ﬂavonoids are often found to be protective
against colon cancer. Approximately 4,000 different ﬂavonoids have been identiﬁed
in many plant families (Harbome, 1988). The ﬂavonoid family includes a wide
variety of chemical structures and are the largest group of plant phenolics (King and
Young, 1999). The ﬂavonoids (Fig. l) are polyphenolic compounds possessing 15
carbon atoms; two benzene rings joined by a linear three carbon chain. These low
molecular weight compounds are usually glycosylated on one or more of their
hydroxyl groups. Many ﬂavonoids are easily recognized as ﬂower pigments in most
angiosperm families (ﬂowering plants). However, their occurence is not restricted to
ﬂowers and in fact can be found in all parts of the plant. Major classes of ﬂavonoids

include ﬂavanones, C-glycosylﬂavonoids, isoﬂavonoids, and anthocyanins (Fig. 1).

Anthocyanins are widely distributed in plants including beans, fruits, and

, vegetables (Harbome and Grayer, 1988). There are more than 240 known structures
for anthocyanins based on 6 major aglycones-pelargonidin, cyanidin, peonidin,
delphinidin, petunidin, and malvidin. Anthocyanins are universal plant colorants and
largely responsible for colors of ﬂower petals and fruits of higher plants (Harbome,
1988). Another function of anthocyanins includes light absorbance, thereby
protecting the photosynthetic system from excess light (Timberlake, 1980).
Cyanidin based anthocyanins have been found in a wide variety of berries and
cherries as well as sweet potatoes, red onions and currants. Characterization of tart
cherries (Wang et al., 1997) showed that the predominant anthocyanins in Balaton
and Montmorency cherries were anthocyanin 1, anthocyanin 2 and anthocyanin 3

(Figure 2). Wang et al. (1999a) reported that 100 g of fresh and pitted cherries

10

contain 12.5 - 25.0 mg of anthocyanins. Anthocyanins 1 — 3 and the parent aglycone,
cyanidin, were found to possess antioxidant and anti-inﬂammatory activities (Wang
et al., 1999b). The antioxidant activities of anthocyanins were superior to that of or-
tocopherol and comparable to that of the commercial antioxidants butylated
hydroxyanisole (BHA) and butylated hydroxytoluene (BHT). C yanidin inhibited
cyclooxygenase (COX) 1 and 2 enzymes with ICSO values of 90 and 60 11M,
respectively (Wang et al., 1999b).

Cyanidin was found to be four times more powerful as an antioxidant than
Trolox, a vitamin E analogue (Rice-Evans, 1995). Tsuda et al. (1996) reported
inhibition of lipid peroxidation and the active oxygen radical scavenging effect of
anthocyanin pigments and cyanidin 3-O-beta-D-glucoside (C3G). Dietary cyanidin
C3G also resulted in a signiﬁcant decrease in generation of thiobarbituric acid
reactive substances and showed a significantly lower susceptibility to lipid
peroxidation in rats (Tsuda, 1988). Cyanidin has been found to function as a potent
antioxidant in vivo in another recent study (Tsuda, 2000). C3G showed a potency to
scavenge active oxygen species in rats fed a diet containing C3G (2 g/kg diet) for 14
days, and then subjected to hepatic ischemia-reperfusion (I/R) as an oxidative stress
model (Tsuda, 2000). Ramirez-Tortosa et al. (2001) have shown that an
anthocyanin-rich extract decreased lipid peroxidationand DNA damage in vitamin
E-depleted rats. Seeram et a1 (2001) reported cyclooxygenase inhibitory and
antioxidant effects of cyanidin glycosides from cherries and berries. The antioxidant

activity of anthocyanins was comparable to commercial antioxidants and superior to

11

a) The skeleton structure of ﬂavonoids

 

b) Chalcones

OH

HO 0
c) Flavones
OH
HO I 0
H0 0
d) Flavonols
OH

HO 0

Figure 1. The chemical structures of ﬂavonoids.

12

e) Flavanone

OH
HO 0
HO 0
f) Isoﬂavones
HO o
0
Ho OH
g) Anthocyanins
OH
HO 0
OH
110

Figure 1 continued. The chemical structures of ﬂavonoids.

OR2
OH

 

Figure 2. Anthocyanins from tart cherries

Anthocyanin l. R1=glucose, R2: rhamnose,
[3-cyanin 2’ ’-O-B-D-glucopyranosyl-6’ ’-O-or-L-rhamnopyranosyl-B-D-
glucopyranoside]
Anthocyanin 2. R1=H, R2= rhamnose,
[3-cyanin 6’ ’-O-or-L-rhamnopyranosyl-B-D-gluc0pyranoside]
Anthocyanin 3. R1=R2=H,
[3-cyanin O-B-D—glucopyranoside]

l4

vitamin E whereas the COX inhibitory activities were comparable to those of
ibuprofen and naproxen at 10 uM concentrations (Seeram et al, 2001). In general,
anthocyanins are stable in acidic conditions (below pH 7) and unstable in alkaline
conditions. Anthocyanins can be hydrated at the C-2 position and degraded to
Chalcones in alkaline conditions of pH 7 or higher. A mechanistic study done by
Tsuda et a1. (1996) showed that C3G had strong antioxidative activity even at neutral
pH. These results indicate that the ﬂavylium cation form, which predominates at
acidic pH, may not be essential for the antioxidative activity of anthocyanins. The
study also showed that C3G reacted with azo-compounds to generate alkylperoxyl
radicals. The reaction products were identiﬁed as 4,6-dihydroxy-2-0-b-D-glucosyl-
3-oxo-2,3-dihydrobenzofuran and protocatechuic acid (Tsuda et al., 1996). Thus,
degradation of anthocyanin pigments could produce additional radical scavengers. A
subsequent study on absorption and metabolism of C3G (Tsuda et al., 1999) showed
that C3G was rapidly found in the plasma in rats ingesting 0.9 mmol/kg body weight
of C3G by direct stomach intubation, indicating that the glycoside can be absorbed
intact. Protocatechuic acid was found in the plasma at a concentration 8-fold higher
than that of C3G, whereas cyanidin was not detected in plasma (Tsuda et al., 1999).
Miyazawa et a1. (1999) also found that oral supplementation of C3G and cyanine-3,
5-diglucode incorporated into plasma and liver of rats and human plasma by UV-
HPLC.

Little research has been conducted for the effect of anthocyanins and
cyanidin on colon carcinogenesis. Koide et a1. (1996) ﬁrst reported that feeding red

rice prolonged survival of Balb/C mice inoculated i.p. with Meth/A lymphoma cells

15

and that hydrolyzed anthocyanins decreased the growth of HCT-1 5 cancer cells.
Later, they found that sugar-bonded bioﬂavonoids extracted from red soybeans and
red beans prolonged survival of Balb/C mice bearing tumor Meth/A cells and that
anthocyanin fractions extracted ﬁom red soybeans inhibited the growth of HCT-1 5
cells (Koide et al., 1997). Pool-Zobol et al. (1999) have shown that anthocyanins
reduced HzOz-induced DNA strand breaks but did not prevent endogenous
generation of oxidized DNA bases in HT 29 clone 19A human colon cancer cells.
Recently, Meiers et a1. (2001) investigated the inﬂuence of anthocyanins on
the growth of human tumor cells. They reported that the aglycones of anthocyanins,
cyanidin and delphinidin inhibited the growth of human vulva carcinoma cell line

A431 and their ICso values were 42 and 18 uM, respectively. Cyanidin (0.8 i 0.2

uM ) and delphinidin (1.3 i 0.2 1.1M) were potent inhibitors of the tyrosine kinase
activity of the epitherrnal growth-factor receptor (EGFR). Treatment with an
irreversible inhibitor of the EGFR-kinase, EKI-569, in combination with sulindac
showed remarkable protection from tumor development (95% reduction in polyp
number) in ApcMin mice (Torrance et al., 2000). Recently, it has been reported that
purple corn anthocyanins suppressed the promotion of colon tumors caused by 2-
amino-l-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) in rats initiated with
dimethyl hydrazine. However, these researchers did not test the potential of purple

corn color anthocyanins to directly suppress tumor development (Hagiwara et al.,

2001).

16

ApcMi" Mouse Model

The ApcMin (mutiple intestinal neoplasia) mouse has been proposed to be a
useﬁrl model for the study of human colorectal cancer. Moser et a1. (1990) reported
a mutation in the mouse genome that resulted in the development of multiple
intestinal tumors in all carriers of the mutation. The mutation was identiﬁed in a
pedigree established during a mutagenesis project in which C57BL/6J (B6) males
were treated with ethylnitrosourea and then mated to AKR/J (AKR) females. The
primary phenotype of ApcMin mice is the development of multiple adenomas that
progress to adenocarcinomas of the intestine in older mice. Min is transmitted by
affected mice to 50% of progeny with an unbiased sex distribution, as is
characteristic of a fully penetrant autosomal dominant trait (Moser, 1990). The
mutation is currently maintained by crossing Min/+ males with B6 females. Tumors
usually are present throughout both the small and large intestine and the cecum, but
are not present in the stomach.
ApcMin mice are heterozygous for a nonsense mutation (ApcMin ) at the Apc
locus, the murine homologue of the adenomatous polyposis coli (APC) gene (Su et
al., 1992). Two candidate tumor suppressor genes, APC and MCC (mutated colon
cancer) on human chromosome 5q21 frequently are mutated in sporadic colorectal
cancers (Kinzler et al., 1991). The APC gene is also mutated in humans with familial
adenomaous polyposis (F AP), an autosomal dominantly inherited disease that

predisposes to colorectal cancer. The frequency of somatic APC mutations is the

same in adenomas and carcinomas, suggesting that mutation of APC is an early

17

event in sporadic colorectal cancer (Levy et al., 1994). Using a site-speciﬁc
quantitative polymerase chain reaction assay, Luongo et a1. (1994) found extensive
loss of the remaining wild-type Apc (Apc+) allele at the Apc locus in spontaneously-
occurring intestinal adenomas from mice heterozygous for the ApcMin nonsense
mutation.

Examination of human FAP kindreds demonstrates that family members who
inherit the same APC mutation may differ dramatically in tumor burden (Leppert et
al., 1990). Although environmental factors may be partially responsible, modiﬁer
genes have been proposed to account for some of this variability (Paul et al., 1993;
Giardiello et al., 1994). Further study of Min-induced neoplasia by manipulation of
the mouse gerrnline identiﬁed genes modifying the Min phenotype: Moml(modiﬁer
of Min 1), genes for the 5-cytosine DNA methyltransferase, and DNA mismatch
repair factor Msh 2 (Bilger et al., 1996). Quantitative trait loci mapping identiﬁed
the Mom 1 locus in the distal region of mouse chromosome 4 (McPhee et al., 1995).
Mom 1 also resides in a region of synteny with human chromosome 1p35, a region
frequently associated with loss of heterozygosity in neuroblastoma and colon cancer
(Martinsson et al., 1989; Leister et al., 1990). The gene for secretory type II
phospholipase A2 (sPLA2) maps to the same region that contains Mom 1 (McPhee et
al., 1995). The sPLA2 are extracellular enzymes that, upon activation, cleave acyl
ester bonds in the sn-2 position in glycerophospholipid. These enzymes play a
crucial role in the generation of arachidonates and eicosanoids, proinﬂammatory

mediators.

18

Therefore, the Min mutation exerts a decisive effect in a pathway leading to
intestinal tumor formation. However, the fact that all cells of a ApcMin mouse carry
this predisposing gerrnline alteration, but only a limited number of tumors develop,
indicates that somatic events are also necessary for tumor formation. Min-dependent
tumorigenesis also occurs with less frequency in mammary glands, the pancreas, and
the body wall (Bilger et al., 1996). Mouse models such as the Min mutant should
provide a major resource for studying the anticarcinogenic potential of dietary
factors, pathways of tumorigenesis, and the molecular identiﬁcation of genes that can
control particular neoplastic processes.

There has been great interest in studying the effects of dietary or other
interventions on colon carcinogenesis. However, most research has been conducted
using animal models of carcinogenesis with tumors initiated by injecting large doses
of carcinogens and using different experimental conditions (i.e. different carcinogens,
injection doses, etc). This makes it difﬁcult to compare results from studies
conducted in different laboratories. The ApcMin mouse has been used as a model to
study effects of dietary factors on human colorectal cancer.

Several studies have been conducted to study the potential of diet to inhibit
the formation of intestinal tumors in ApcMin mice. Bowman-Birk protease inhibitor
concentrate (0.5%) from soybeans reduced colonic tumors and small intestinal
tumors in Min mice by approximately 40% (Kennedy et al., 1996). Paulsen et a1
(1997) showed that eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA)-
enriched ﬁsh oil suppressed tumor number and burdon in ApcMin mice. Rye-bran

Min

diet also has been shown to prevent the formation of intestinal polyps in Apc mice

19

(Mutanen et al., 2000). Mahmoud et a1. (2000) reported that plant phenolics,
including caffeic acid phenethyl ester and curcumin, decreased intestinal tumor
formation by 63% in ApcMin mice. Green tea extract and sulindac acted

Mm mice

synergistically to inhibit development of intestinal tumors in Apc
(Suganuma et al., 2001). Treatment with both green tea extract and sulindac
signiﬁcantly reduced the number of tumors from 72 to 32 tumors whereas green tea
extract or sulindac alone reduced tumor number to 57 or 49, respectively (Suganuma
et al, 2001).

Results from studies using ApcMin mice have not always supported the
ﬁndings of epidemiological studies and studies with carcinogen-injected animal
models. Some dietary factors do not inﬂuence tumor development when
administered to ApcMin mice. Pierre et a1. (1997) reported that addition of short
chain fructooligosaccharides to the diet reduced colonic tumor incidence but had no
effect on small intestinal tumors in ApcMin mice. Van Kranen et a1. (1998) found no
effect of dietary fat or a fruit and vegetable mixture on tumor formation in ApcMin
mice. Sorenson et a1. (1998) reported that soy isoﬂavones had no effect on tumor
Min

formation in Apc mice.

N onsteroidal Anti-Infammatory Drugs (NSAIDsland Colorectal Cancer
Several lines of investigation have demonstrated that nonsteroidal anti-
inﬂammatory drugs (N SAIDs) inhibit the growth of colon tumors in animal models

and reduce the risk of colon cancer in humans. Giovannucci et a1. (1995) found that

four to six tablets of aspirin per week sustained over years substantially reduced the

20

risk of colorectal cancer. Other NSAIDs such as sulindac and ibuprofen have been
shown to reduce the relative risk of colorectal cancer aﬁer 2 or more years of
continuous use (Muscat et al., 1994).

The mechanism of the anticarcinogenic effect of these drugs has not been
fully established. One potential explanation for this protective effect is inhibition of
intestinal epithelial growth. It has been demonstrated that NSAIDs, including
indomethacin and sulindac, inhibit the rate of growth of carcinogen-induced colon
tumors in animal models (Olsson et al., 1984; Skinner et al., 1991). Levi et al.
(1990) showed that indomethacin inhibits the gastric epithelial cell
hyperproliferation that normally restores mucosal integrity aﬁer injury in rodent
models. Studies of sulindac in FAP patients provide further evidence that NSAIDs
inhibit epithelial cell proliferation in the gastrointestinal tract. Regression of
colorectal polyps was observed after 6 months of treatment with oral sulindac
(Labayle et al., 1991; Rigau et a1.; 1991). Polyps returned aﬁer treatment was
discontinued. Reductions of mucosal prostaglandin E2 levels paralleled polyp
regression (Labayle et al., 1991).

Most hypotheses for explaining the anticarcinogenic effects of NSAIDs have
focused on the common property of NSAIDs to inhibit COX enzymes and thereby
cause a subsequent reduction in levels of prostaglandins in tissue. Inducible COX-2
is overexpressed in human colon cancers. Sulindac inhibited tumor formation and
reduced overexpression of COX-2 level in Min mice (Boolbol et al., 1996). Jacoby
et al. (1996) demonstrated that dietary piroxicam signiﬁcantly reduced intestinal

tumor multiplicity in Min mice to 12% of that observed in control animals. Plasma

21

thromboxane B2 levels were reduced in parallel with tumor inhibition (Jacoby et al.,
1996). Ritland and Gendler (1999) conﬁrmed chemopreventive effects of proxicam
Min

on intestinal adenomas in Apc mice. Short-term administration of proxicam (200

Mi" mice with the

ppm) reduced tumor multiplicity by 90%. Treatment of Apc
NSAID nabumetone resulted in a dose dependent suppression of intestinal
tumorigenesis (Roy et al., 2001). Nabumetone induced apoptosis by

Min

downregulating the anti-apoptotic protein, Bcl-2, in the intestine of Apc mice.

McCracken et al. (1996) investigated the effect of COX-inhibiting (S-form)
and non-COX-inhibiting (R-form) enantiomers of ﬂurbiprofen on rat colonocyte
proliferation. Sprague Dawley rats were treated orally with R- or S-ﬂurbiprofen or
vehicle. Both R and S forms signiﬁcantly reduced colonocyte labeling index, by 34%
and 23%, respectively, compared with vehicle. R-ﬂurbiprofen caused minimal ulcer
formation (4.48 mmz) compared with S-ﬂurbiprofen (94.4 mmz). These ﬁndings
suggest that R-ﬂurbiprofen-mediated control of colonocyte proliferation is
independent of prostaglandin biosynthesis. Shiff et al. (1995) showed a possible
alternate mechanism by which NSAIDs inhibit gastrointestinal cell proliferation.
They demonstrated that both sulindac and sulindac sulﬁde caused cell cycle
quiescence and induced apoptosis by inhibition of the p33 cdk prOtein, a key
regulator of the cell cycle. Sulindac was effective despite its relative weak COX-
inhibitory activity compared to sulindac sulﬁde. Notably, disruption of the mouse
sPLA2 locus has been shown to increase tumor number and restoring sPLA2

expression through transgenic constructs decreases tumor number in ApcM'n mice,

22

despite the prediction that sPLA2 should decrease prostaglandin production
(Kennedy et al., 1995; Cormier et al., 1997).

The clinical and experimental evidence that NSAIDs are anticarcinogenic is
strong, and several studies have suggested COX-independent mechanisms for this
action. The exact relationship of prostaglandin synthase inhibition by NSAIDs to

tumor prevention remains uncertain.

Genetics of Intestinal Neoplasia: APC Mutation and its consequences

Different forms of human colon cancer share mutations in APC and may also
share other genetic features (Fearon and Vogelstein, 1990). Certain mutations are
common in colon cancer, but the sequence of events may vary. Temporal series of
somatic changes including mutation of APC, K-RAS, DCC and P53 and
hypomethylation of DNA have been demonstrated by molecular analysis of human
intestinal cancer (Fig. 3). This observation shows that a particular molecular change
is correlated with a particular stage in the carcinogenesis. It generates the hypothesis
that certain gene mutations are necessary for particular carcinogenic steps in colon
carcinogenesis.
In most cases, colon carcinogenesis depends on mutation of the APC gene and it is
considered a gatekeeper in the carcinogenic process (Kinzler and Vogelstein, 1996).
APC encodes a cytoplasmic polypeptide of 2843 residues (Fig. 4). The N-terminus
contains oligomerization sites indicating that the molecule acts as a polymer (Su et
al., 1993a). The central part of the molecule contains sites for 20 amino acid [3-

catenin binding repeats and also contains heptad repeats and armadillo (Arm) repeats

23

(Su et al., 1993b). The heptad repeats lead to supercoiling of helix and arm repeats
composed of three alpha helices mediate protein-protein interactions. The
truncations in the midpoint cluster region, which eliminate all the axin-binding and
most of the B-catenin-binding repeats, account for more than 60% of oncogenic
mutations in APC (Miyoshi et al., 1992). The C terminal region contains tubulin
binding sites (Munemitsu et al., 1994; Smith et al., 1994). Tubulin is the protein that
polymerizes into long chains or ﬁlaments that form microtubules, hollow ﬁbers
which serve as a skeletal system for processes such as cell division and the
movement of materials within cells. The region also contains binding sites for the

human homologue of the Dlg (Drosophila large discs) tumor supressor protein.

24

 

 

 

 

 

 

Chromosome Sq 12p 18q 17p
Alteration _ Loss Activation Loss Loss
Gene APC k-ras DCC p53
Hypomethylation
V V V
Normal Hyperproliferative + Adenoma
Epithelium epithelium Early—-> lntennediate —> Late

 

 

 

 

 

 

 

 

 

 

Y

 

 

Carcinoma

 

Figure 3. A genetic model for colorectal tumorigenesis (modiﬁed from Vogelstein

and Kinzler, 1993)

25

 

Heptad repeats Three lS-amino acid repeats DLG binding sites
¢—>

 

 

 

 

 

I | | I | l <———> ‘ ’
N D-D-Doooo-ooo C
/ |||||| , ,
Oligomerization ‘—> ¢ >
Domain Arm repeats Seven 20-amino acid Tubulin

binding sites
B-catenin binding repeats

<———>
Midpoint Cluster Region

Figure 4. Schematic of the APC protein primary structure (modiﬁed from Spink et al.,

2001)

26

It has been shown that intestinal adenomas in ApcMin mice and human colon
tumor cell lines have lost the wild type allele or express only short N-temrinal
fragments of the Apc polypeptide (Luongo et al., 1994). The Min allele contains a
nonsense codon at position 850 (Moser et al., 1990) and other targeted mutant Apc
alleles have been developed that result in Apc polypeptide truncation of codons 716,
1638 and 474 by Oshima et a1. (1995), Fodde et a1. (1994), and Sasai et al. (2000),
respectively. Ape“6 mice are predisposed to develop 200 - 500 adenomas mainly in

the small intestine, whereas Apc1638

are predisposed to develop 5 - 6 adenomas and
adenocarcinomas along the intestinal tract and a number of extra-intestinal lesions
including mammary tumors (Oshima et al., 1995; Fodde et al., 1994). Apc474 mice

develop around 100 intestinal polyps and 18% of them manifested mammary tumors

(Sasai et al., 2000).

The APC cytoplasmic polypeptide plays a regulatory part in the cell and
participates in several protein assemblies. Most human APC mutations are found
between codons 1286 and 1513 and this midpoint mutation cluster region occurs
near the B-catenin binding sites (Nakamura, 1993). Mutations result in the
production of a truncated protein which cannot facilitate the degradation of B-catenin,
suggesting that mutated APC is essential in the process of tumor development by
causing a loss of regulation of B-catenin.

B-catenin performs different functions within the cell. Cytoplasmic and
nuclear B—catenin accumulate only in response to activation of the Wnt/Wingless

growth factor signaling pathway whereas a membrane associated pool is bound to E-

27

cadherins and or—catenin in the adherins junction (Yap etal., 1997). E-cadherin is
one of a large family of genes which code for calcium-dependent cell adhesion
molecules (Takeichi et al., 1988). Any free B-catenin not bound to E-cadherens is
targeted for degradation in the absence of a Wnt signaling. The APC gene product is
required in the B-catenin degradation process as an essential part of a mutiprotein
complex. A complex of glycogen synthase kinase (GSK)-3B and conductin/axin
phosphorylates the APC/B-catenin complex, which induces ubiquitination and
degradation of B-catenin by proteosomes (Easwaran et al., 1999). The ubiquitination
and degradation of B-catenin is a highly regulated process and requires a multi-
molecular complex with APC, conductin/axin, and GSK-3B.

APC mutation allows increased levels of cytoplasmic B-catenin, which may
then trigger a cascade of events resulting in the initiation of adenomas. The
cytoplasmic B-catenin is translocated to the nucleus and functions as a transcription
factor in intracellular signaling. B-catenin binds to T cell factor(ch)/lymphoid
enhancer factor (Let) transcription factors in the nucleus (Ilyas and Tomlinson,
1997). Thus, APC mutation causes aberrant accumulation of B-catenin, which then
translocates to the nucleus and binds ch/Lef , causing increased transcriptional
activation of some genes. These events ultimately result in abnormal cell growth
since normal APC function is essential for maintaining normal cell architecture and
growth.

Genes that might be targets for transcriptional activation by B-catenin/ch

transactivation have been identiﬁed. Crawford et al. (1999) suggested that matrilysin

28

is a target gene for B-catenin/ch transactivation. Matrilysin is a matrix
metalloproteinase 7 (MMP~ 7) expressed in approximately 90% of human colonic
adenocarcinomas and in many human colonic tumor cell lines (N ewell et al., 1994).
Cyclin D1 (CLNDl) has also been identiﬁed as a target gene of B-catenin
transactivation (Tetsu and McCormick, 1999). CLNDl is a regulator of the cell
cycle and is highly expressed in adenomatous polyps and colorectal
adenocarcinomas (Zhang et al., 1997). Overexpression of CLND 1 may derange the
cell cycle machinery and contribute inappropriate neoplastic cell growth (Hartwell
and Kastan, 1994). He et al. (1998) identiﬁed c-MYC as a target gene in the
signaling pathway of APC mutation. C-MYC expression mediated through ch—4
binding sites in the promoter region is repressed by APC and activated by B-catenin.
He et al. (1999) also reported that peroxisome proliferator activated receptor 6
(PPARO) is a target of the APC tumor suppressor pathway and is an APC-regulated
target of NSAID drugs. They found that APC repressed PPAR5 expression and that
PPARS activity was regulated by the APC/B-catenin/ch-4 pathway when APC was
introduced in HT-29 cell lines. Interestingly, it has been suggested that PPARs may
be a target of NSAIDs in suppressing tumorigenesis since NSAIDs can perturb
eicosanoid metabolism and mediators involved in eicosanoid metabolism are known
ligands for PPARs (Prescott and White, 1996). Sulindac sulﬁde resulted in a dose-
dependent repression of PPARS activity in HCT116 and SW480 cell lines by
disrupting the DNA binding ability of PPARO/RXROL (retinoid X receptor or)

heterodimers (He et al., 1999). Recently, the role of PPARO in colorectal

29

tumorigenesis was evaluated by Park et al. (2001). They created PPARO null cell
lines (PPAR5 -/-) by deleting the PPARB gene in a human colon cancer cell line
using targeted homologous recombination. When inoculated as xenograﬁs in nude
mice, PPARO -/- exhibited a decreased ability to form tumors compared with PPAR5
+/— and wild-type controls, suggesting that suppression of PPARO expression

contributes to the growth-inhibitory effects of the APC tumor suppressor (Park et al.,
2001)

Recent immunohistochemical analysis of B-catenin showed that B-catenin
staining was mainly localized to the cell membrane in normal epithelial cells of
stomach, duodenum, appendix, and rectum, whereas B-catenin accumulated in the
cytoplasm and/or nucleus in 57 cases (79.2 %) out of 72 cases of human
gastrointestinal carcinoid tumors (Fujimori et al., 2001). McEntee et al. (1999)
reported that sulindac regressed small intestinal adenomas and reduced B-catenin
expression in adenomas of ApcMm mice. Mutanen et al. (2000) also reported that rye
bran diet prevented tumor formation in ApcMin by decreasing the cytosolic B-catenin
level. Sheng et al. (1998) found nuclear localization of B-catenin in hereditary and
carcinogen-induced intestinal adenomas. B-catenin was expressed in the cell
membrane and cytoplasm of morphologically normal intestinal cells whereas it was
found in the nucleus of adenomas from ApcMin mice and Apc 716 mice. Cellular
distribution of B-catenin was also modiﬁed by sphingolipid consumption (Schmelz et
al., 2001). Sphingolipid feeding reduced adenoma numbers in all regions of intestine
Min

in Apc mice and sphingosine reduced cytosolic and nuclear B-catenin and

30

inhibited growth of human colon cancer cell lines SW 480 and T84 (Schmelz et al,
2001). He et a1 (1999) showed that sulindac sulﬁde repressed PPARO activity in
HCT116 and SW480 cell lines in a dose-dependent manner.

A proposed model for APC- and NSAID-mediated suppression of intestinal
tumorigenesis (Fig. 5) shows several features of NSAID-mediated chemopreventive
action. First, numerous studies have shown that NSAIDs inhibit COX activity and
subsequently decrease levels of prostaglandins. However, apoptosis-inducing
activities of NSAIDs are not entirely related to the inhibition of COX, suggesting

additional cellular targets. Second, NSAIDs inhibit PPARS activity (He et al., 1999).

The pro-apoptotic effect of NSAIDs may be explained by direct inhibition of PPARO.
Third, NSAID-induced regression of intestinal adenomas is associated with reduced
levels of B-catenin (McEntee et al., 1999). McEntee et a1. (1999) observed a 50%

M'" mice

decrease in B-catenin as well as Bcl-2 in small intestinal tumors from Apc
treated with sulindac. Rosin-Arbesfeld et al. (2000) identiﬁed a nuclear export
function of the APC gene. They showed that APC contains highly conserved nuclear
export signals 3’ and adjacent to the mutation cluster region that enable it to export

B-catenin from the nucleus. When this ability is lost in APC mutant cancer, the result

is accumulation of B-catenin in the nucleus.

31

   

 

 
   
 
      
  

Sulindac
NSAIDs

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

    
 
 

Lipids
sPLA2 l '—
TCF-4/B-catenin
Arachidonic Acid
LOX / \ COX l__
‘A
Leukotrienes Prostaglandins
PPAR5 CLND]

c-MYC MMP 7

Altered Eicosanroid Metaboliirg

 

 

Figure 5. APC-and NSAID-mediated suppression of intestinal tumorigenesis

(Modiﬁed from He et al., 1999)

32

Hypotheses and Research Objectives

Tart cherry anthocyanins may suppress intestinal tumorigenesis because they
inhibit the activities of COX enzymes. However, the potential of the anthocyanins
and cyanidin to inhibit colon cancer development has not been fully established. Few
studies have examined the chemopreventive potential of anthocyanins and cyanidin.
Cyanidin inhibited epidermal growth factor receptor (EGF R) kinase activity in
cancer cells (Meiers, 2001). Treatment with an irreversible inhibitor of the EGFR-
kinase reduced tumor development in ApcMin mice (Torrance et al., 2000). Hagiwara
et al. (2001) demonstrated that anthocyanins in purple corn color reduced the
promotion of colon tumors caused by PhIP in rats initiated with dimethylhydrazine.
However, they did not test the potential of purple corn color anthocyanins to directly
suppress tumor development (Hagiwara et al, 2001).

The overall objective of the study is to determine the effects of tart cherry
anthocyanins on the development of intestinal adenomas and on global gene

Min

expression in intestine, cecum and colon of Apc mice.

The hypotheses to be tested and speciﬁc aims are:
l. Tart cherry anthocyanins inhibit intestinal tumori genesis in ApcMin mice.

Speciﬁc aims:

a. Examine ontogeny of intestinal adenoma development in mouse model (Apc
mouse) to obtain baseline data on tumor development.

b. Examine effects of tart cherry anthocyanins, their aglycone, cyanidin, and their
degradation products on growth of human colon cancer cell lines HT 29 and
HCT 116.

c. Examine effects of tart cherry diets, anthocyanins and their aglycone, cyanidin
on multiplicity and size of intestinal adenomas in ApcM'" mice.

Min

33

2. The inhibiton of intestinal tumorigenesis caused by tart cherry anthocyanins is

M'" mice.

related to changes in gene and protein expression in Apc
Speciﬁc aims:

a. Examine the exaression and cellular distribution of [Pi-catenin in intestinal
sections of Apc '" mice fed tart cherry anthocyanins in comparison to Apc
mice fed control or sulindac-supplemented diets.

b. Examine the expression of cancer-related genes in intestinal epithelial cells of
ApcMm mice fed tart cherry anthocyanins in comparison to ApcMm mice fed
control or sulindac-supplemented diets. I will speciﬁcally determine the
expression of the predicted target genes of the Wnt/B-catenin signaling
pathway such as PPARS, C-Myc, CLND] and MMP7.

Min

The ApcMin mouse strain is an excellent model for studying colon cancer
development and the anticarcinogenic potential of dietary factors. The time frame of
tumor development in ApcMin mice will inﬂuence experimental protocols. First, we
will examine the timing of small intestinal and colonic tumor development in ApcM'n
mice and determine the inﬂuence of sex on small intestinal and colonic adenoma
Min

multiplicity and size in Apc mice. We will then determine the effects of tart cherry

anthocyanins, cyanidin and sulindac on intestinal adenoma multiplicity and size in
the small intestine, cecum and colon in ApcMin mice. The ability of tart cherry
anthocyanins to directly inhibit the growth of human colon cancer cell lines HT 29
and HCT 116 also will be determined.

Expression of thousands of genes can be examined simultaneously using a
novel technique called cDNA microarray analysis. cDNA microarrays are capable
of proﬁling gene expression patterns of tens of thousands of genes in a single

experiment (Duggan et al., 1999). Microarrays consist of a collection of immobilized

genes which can be simultaneously analyzed to ascertain their expression in tissue.

34

DNA targets, in the form of 3’ expressed sequence tags (ESTs), are arrayed onto
glass slides and probed with ﬂuorescent-labeled cDNAs. Premade microarrays of
gene-speciﬁc cDNA fragments immobilized on glass slides for the mouse will be
used to analyze 15,000 genes that survey major biological pathways (Tanaka et al.,
2000)

Screening changes in expression of such target genes as well as oncogenes
and tumor suppressor genes in the different sections of intestine may yield insights
about the mechanisms responsible for tumor modulation by anthocyanins and
sulindac. We speculate anthocyanins may inhibit expression of PPARB, c-MYC,
CLNDl or matrilysin and inﬂuence B-catenin expression and distribution in the cell

membrane, cytoplasm and nucleus in the intestinal cells.

35

CHAPTER III

ONTOGENY OF INTESTINAL ADENOMA DEVELOPMENT IN
APCM'N MICE.

36

ABSTRACT

ApcMin mice (C57BL/6J Apcmmh) develop small intestinal and colonic
adenomas spontaneously. This experiment was conducted to determine the inﬂuence
of age on the numbers and sizes of small intestinal, cecal and colonic tumors in
ApcMin mice. ApcMin mice were weaned at 28 days of age and fed Teklad 8640
rodent diet throughout the experiment. Mice were sacriﬁced at 30, 45, 60, 75, 90
and 120 days of age. The entire small intestine, cecum, and colon were removed and
the small intestine was separated into proximal, middle and distal segments.
Intestinal sections were ﬁxed in 10% neutral buffered formalin and stained with
0.2% methylene blue. Intestinal tumors were counted and measured. Body weights
and the lengths of the small intestine and the colon increased with age. The numbers
and sizes of adenomas in the small intestine also increased with age (p<0.05). The
numbers of adenomas in the cecum and colon did not increase signiﬁcantly with age,
but the volume of cecal tumors was not inﬂuenced by age. Colonic tumors increased
signiﬁcantly after 75 days of age. Sex did not inﬂuence the numbers of small
intestinal or colonic tumors. Mouse body weights, colon length, and colon tumor
Min

volume were inﬂuenced by sex. Male Apc mice had greater body weights, colon

lengths and colon tumor volumes than female ApcMin mice. Tumors were already
present in 30-day-old mice. These results indicate that intestinal tumors develop

early in life and progress in size with increasing age in ApcMin mice. Experiments
examining the effects of dietary or other interventions in ApcMin mice should take

into account their early and rapid tumor development.

37

INTRODUCTION

Colon cancer is the second most common cause of cancer mortality and the
fourth most common in incidence in the United States (American Cancer Society,
2000). It has been estimated that 66-75% of colon cancer could be preventable by
proper diets (WICR/AICR, 1999). Hence, there has been great interest in studying
the effects of dietary or other interventions on colon carcinogenesis. Most research
on colon cancer prevention by diet has been conducted using animal models of
carcinogenesis. Most of these studies rely on tumors initiated by injecting large doses
of carcinogens and oﬁen have used different experimental conditions (i.e. different
carcinogens, injection doses, etc). This makes it difﬁcult to compare results from
studies conducted in different laboratories.

The ApcMin mouse has been proposed to be a model for the study of human
colorectal cancer (Moser et al., 1990). ApcMin rrrice are a mutant mouse lineage
predisposed to multiple intestinal neoplasia (Min) resulting from a mutation in the
murine homolog of the adenomatous polyposis coli (APC) gene (Su et al., 1992).
The APC gene also is frequently mutated in humans who develop sporadic colon
cancer and is the mutation responsible for causing familial adenomatous polyposis
(FAP), an autosomal dominantly inherited disease that predisposes persons to
colorectal cancer. The primary phenotype of ApcMin mice is the development of
multiple adenomas which progress to adenocarcinomas of the intestine in older mice.
Min is transmitted by affected mice to 50% of progeny with an unbiased sex

distribution, as is characteristic of a fully penetrant autosomal dominant trait (Moser

et al., 1990).

38

The Min mouse strain may be an excellent model for studying colon cancer

development and the anticarcinogenic potential of dietary factors. However, the time

frame of tumor development in ApcMi" needs to be more thoroughly investigated to

allow optimization of experimental protocols. Therefore, the purpose of this study

was to examine the timing of small intestinal and colonic tumor development in

Min

Apc mice and to determine the inﬂuence of sex on small intestinal and colonic

Min

adenoma multiplicity and size in Apc mice.

39

MATERIALS AND METHODS
Animals and Diets. All research was conducted with approval of the MSU
All University Committee on Animal Use and Care. Mice were housed in MSU

University Laboratory Animal Resources maintained facilities. A colony of ApcMin

mice was maintained by crossing carrier (C57BL/6J ApcMm/T

) male mice with

normal adult female C57BL/6J mice. Mice were housed in a temperature and
humidity-controlled room (20-22 °C, 50%) with a 12-h light/dark cycle. Mice were
weaned at 28 days of age and fed Teklad 8640 rodent diet throughout the experiment.
Mice were not genotyped before being randomly assigned to study groups (ages).

We anticipated that 50% of the mice allotted to each group would carry the ApcM'"

mutation and 50% would be normal with respect to Apc gene status. Body weights
were measured when mice were sacriﬁced at 30, 45, 60, 75, 90 and 120 days of age.

Tumor Number and Size. Upon sacriﬁce, a section of liver was removed
and frozen immediately for Apc genotyping using a PCR-based procedure (Su et al.,
1992) followed by gel electrophoresis. The intestine was removed and tissues were
separated into the following sections: proximal one-third of small intestine, middle
one-third of small intestine, distal one-third of small intestine, cecum, and colon. All
intestinal sections were opened longitudinally, rinsed thoroughly with water, ﬁxed in
neutral-buffered formalin, and then stained with 0.2% methylene blue. Tumor
numbers and sizes were determined in each intestinal segment by direct counting
with the aid of a dissecting microscope. Because small intestinal tumors are

typically sessile and cecal and colonic tumors are typically polypoid, the tumor sizes

were determined by measuring the mean diameter of ﬂat tumors in the small

40

intestine and by measuring volume of three-dimensional tumors in the cecum and
colon. Spheric volumes were calculated for cecal and colonic tumors using the
formula V = 0.523(width x length x height). Tumor numbers in each small intestinal
segment were summed to obtain a total small intestinal tumor burden for each mouse.
Body weights, tissue lengths, tumor numbers and tumor diameters in the small
intestine were analyzed by two-way analysis of variance to detect the effects and
potential interactions of age and sex. For tumor numbers and volumes in the cecum
and colon, data were transformed to ranks and then analyzed by two—way analysis of
variance (AN OVA). When signiﬁcant treatment effects were detected (P<0.05),

means were compared using the Least Signiﬁcant Difference (LSD) method.

41

RESULTS

Body weights of mice were inﬂuenced by age, sex, and Apc status (Min vs.
normal). Body weights of all mice increased with increasing age (p<0.05) (Fig. 6a).
Male mice weighed approximately 3 grams heavier than female mice for ages 45
through 120 days. Body weights of ApcMin mice (Fig. 6b) averaged about 1 gram
less than those of normal Apcm mice at all ages. Small intestine and colon length
increased with age (p<0.05) (Table 1). ApcMin mice had small intestine lengths about
9 mm less than normal Apcm mice. Small intestine length was not inﬂuenced by sex.
However, colons were approximately 4 mm longer in males than females. Carrier
status did not inﬂuence colon length.
Min

Intestinal tumors were only detected in Apc mice. Tumor numbers (Fig.

7) in the small intestine of ApcMin mice also increased with age (p<0.05). Mice
averaged 13.6 small intestinal tumors at 30 days of age. Small intestinal tumor
numbers increased steadily to a maximum of 70-80 tumors per mouse by 90 days of
age. As mice aged, the numbers of adenomas in the distal small intestine increased
to a greater magnitude compared to those in the proximal and middle small intestine.
The proportion of small intestinal adenomas in proximal, middle, and distal segments
was 28, 34, and 38%, respectively, in 30-d old mice and 10, 34, and 56%,
respectively, in 120-d old mice. The average diameter of the small intestinal
adenomas (Fig. 8) was relatively constant from 30-75 days of age, averaging 0.94
mm per tumor. This diameter increased signiﬁcantly in 90-d old mice (1.2 mm) and

increased dramatically in 120—d old mice to 1.8 mm per tumor. There was no sex

effect on small intestinal tumor numbers or diameters.

42

 

 

a) -o—All females b) —0— Min females

 

 

 

 

 

 

 

 

   

 

 

 

 

 

 

+All males -+—Min males
30 30
25 25
m 20 20
(I)
E 15 E 15 --
o (’5
10 - 10 .
5 l- 5 ..
0 1 1 1 1 0 1 4 I r
30 45 60 75 90120 30 45 60 75 90 120
Age (Days) Age (Days)

Figure 6. Inﬂuence of age on body weights a) Inﬂuence of age on body weights of
all mice. Each point represents the mean of the age groups (30 days; n=22, 45 days;
n=l7; 60 days; n=l7; 75 days; n=22; 90 days, n=23,l20 days; n=23). Body weight
increased with age at all time points except between 60 and 75 days. Error bars
indicate SEM. Males were signiﬁcantly larger (p<0.05) than females after 30 days
of age. b) Inﬂuence of age on body weights of ApcMin mice. Each point represents
the mean of the age groups (30 days; n=8, 45 days; n=10; 60 days; n=10; 75 days;
n=9; 90 days, n=9,120 days; n=8). Error bars indicate SEM. Body weights increased

with age at all time points except between 60 and 75 days. Body weights were

signiﬁcantly different between males and females after 30 days of age.

43

Table 1. Lengths of small intestine and colon of all mice (30 days; n=22, 45 days;

n=l7, 60 days; n=l7, 75 days; n=22, 90 days; n=23,120 days; n=23) (Normal; n=54,

Min,

Apc , n=70) (Female; n=56, male; n=68). Each value represents mean i SEM.

Means in the same column within a category not sharing a common superscript are

signiﬁcantly different (p<0.05).

 

 

 

 

 

 

 

 

Age (days) Small intestine (mm) Colon (mm)
30 308 i 42‘ 68 i 2 a

45 319 i 5 a 76 :1: 2 b

60 339:4b 79:2bc
75 349i4b 80:2bc
90 349 a: 4 b 78 i 2 b

120 378:4c 84:2c
Apc Status Small intestine (mm) Colon (mm)
Normal 345 i 2 b 78 i 1

Min 336 i 2 a 78 i 1

Sex Small intestine (mm) Colon (mm)
Female 338 i 2 75 1r 13
Male 343 i 2 80 i 1 b

 

44

 

LC] Proximal I Middle Distal I Total i

 

 

100

90 T” d cd
80 c
70 -
60 ~
50 ~~

   

o

 
     
   

Number

 

 

4o 4 7

b/

30 . g

20 - {13%

1o -- 1?

0 _ I1?
90 120

 

Figure 7. Inﬂuence of age on adenoma numbers in proximal (white), middle (gray)
and distal (hatched) segments of the small intestine and total (black) numbers in the
small intestine (30 days; n=8, 45 days; n=10; 60 days; n=10; 75 days; n=9; 90 days,
n=9,120 days; n=8). Error bars indicate SEM. Columns of the same shade not

sharing a common superscript are signiﬁcantly different (p<0.05).

45

 

1.8 *-
1.6 4
1.4 a
1.2 ‘—

T

Millimeters

0.8 *-
0.6 4
0.4 ~
0.2 ~—

I

 

 

   

30 45 . 60 75 90 120
Age (Days)

Figure 8. Inﬂuence of age on the average diameter of small intestinal adenomas (30
days; n=8, 45 days; n=10; 60 days; n=10; 75 days; n=9; 90 days, n=9, 120 days; n=8).
Error bars indicate SEM. Columns not sharing a common superscript are

signiﬁcantly different (p<0.05).

46

The numbers (Fig. 9) of adenomas in the cecum were not signiﬁcantly
inﬂuenced by age. The mean volume of cecal adenomas was larger in older mice
(Fig. 10), but this effect did not reach statistical signiﬁcance. There was no sex
effect on cecal tumor number or volume. The number of adenomas in the colon (Fi g.
9) was inﬂuenced by age, but not in a consistent manner. Mice of 90-d of age had
signiﬁcantly more colonic tumors than mice at 45 or 60 days. The total volume of
colonic tumors was signiﬁcantly inﬂuenced by age (Fig. 11). Colonic tumors were
small in size through 75 days of age, and then increased in volume dramatically after
this time. There was also a signiﬁcant sex effect on colon tumor volume. Colon
adenoma volume averaged 28.9 mm3 larger in males (43.8 mm3) than in females

Min

(14.9 mm3) at 120 days of age. This sex effect was not detected in the Apc mice

from the subsequent studies.

47

 

ll Cecum I Colon :

 

 

Number

 

 

 

3O 45 60 75 90 120
Age (Days)

Figure 9. Inﬂuence of age on the average numbers in the cecum (gray) and colon
(black) (30 days; n=8, 45 days; n=10; 60 days; n=10; 75 days; n=9; 90 days, n=9,
120 days; n=8). Error bars indicate SEM. Columns of the same shade not sharing a

common superscript are signiﬁcantly different (p<0.05).

48

 

2.5 _-

 

0.5 —~

Cubic Millimeters/mouse

 

 

           

30 45 60 75 90 120
Age (Days)

Figure 10. Inﬂuence of age on adenoma volume (mm3) in the cecum (30 days; n=8,
45 days; n=10; 60 days; n=10; 75 days; n=9; 90 days, n=9, 120 days; n=8). Error

bars indicate SEM.

49

 

Cubic Millimeters/mouse

 

 

 

30 45 60 75 90 1 20
Age (Days)

Figure 11. Inﬂuence of age on adenoma volume (mm3) in the colon (30 days; n=8,
45 days; n=10; 60 days; n=10; 75 days; n=9; 90 days, n=9, 120 days; n=8). Error
bars indicate SEM. Columns not sharing a common superscript are signiﬁcantly

different (p<0.05).

50

DISCUSSION
Mouse models such as the Min mutant should provide a major resource for
studying the anticarcinogenic potential of dietary factors, pathways of tumorigenesis,
and the molecular identiﬁcation of genes that can control particular neoplastic
processes. The Min mouse has been and will continue to be used as a model to study

M' -
'" mice

effects of dietary factors on human colorectal cancer. Effective use of Apc
in such studies requires a thorough assessment of their normal process of tumor
development.

Min

A few concerns have been identiﬁed concerning the use of Apc mice as a

model for research of colorectal cancer. Unlike human intestinal cancers, which are
Min

predominantly colorectal, Apc mice develop the majority of their tumors in the

small intestine. The utility of ApcMin mice may also be limited by their lack of large
numbers of colonic adenomas and aberrant crypt foci, the putative precursors of
large-bowel polyps and cancers. These potential limitations also would pertain to
various other Apc knockout mouse strains such as Apc‘m, ApcA716 and Apc474
(Fodde et al., 1994; Oshima et al., 1995; Sasai et al., 2000). Accelerated polyposis
models have been deveIOped to increase the observed numbers of tumors and
aberrant crypts. For example, additional carcinogen injections (Shoemaker et al.,
1995; Steffensen et al., 1997) or genetic mutation in mismatch repair genes have
been added to ApcMin (Baker et al., 1998; Lal et al., 2001). However, the majority
Min

still uses the standard Apc mouse model for the study of colorectal cancer

development.

51

Realtively little research has been conducted to determine the tumor

development characteristics of ApcMin mice. Shoemaker et al. (1995) have shown

age-related effects on the formation of adenomas in ApcMin mice after additional
treatment with ethylnitrosourea (ENU). Their results demonstrate that ENU-induced
Min

intestinal tumors in Apc mice are most likely to be initiated during the ﬁrst 2

weeks of life.
In the present study, the time frame of tumor development in Min mice was

determined. Our results provide baseline data for the inﬂuence of age and sex on

Min

body weight and intestinal tissue length of Apc mice in comparison with those of

normal Ape”+ mice. These results also describe the inﬂuence of age and sex on
adenoma development of ApcMin mice. Small intestinal, cecal and colonic adenomas
were already present in 30 day-old mice. Tumor volumes remained relatively small
throughout the intestine for the ﬁrst 75 days of age. After this time, tumor volumes
increased rapidly up to 120 days of age.

We chose 120 days as the ﬁnal time point in this study because it has been
observed previously that ApcMin mice suffer considerable morbidity and mortality
beyond this age (Moser, 1990). Morbidity and mortality in ApcMin mice is typically
due to anemia caused by hemorrhraging associated with the small intestinal tumors
after they progress beyond a certain size. We have observed that there is little

Mi“ mice that have small intestinal tumors less than 1.5 mm in

morbidity in Apc
diameter. However, small intestinal tumors greater than 1.5-2.0 mm in diameter tend

to perfolate and hemorrhrage. Hence, the choice of an end point (age) for feeding

studies conducted using ApcMin mice must be fully considered. Tumor progression

52

occurs predominantly in the age range of 75-120 days, with signiﬁcant morbidity
observed thereafter. These results suggest that experiments examining the effects of
dietary or other interventions in ApcMin mice should take into account their early and

rapid tumor development.

53

CHAPTER IV

TART CHERRY ANTHOCYANINs INHIBIT CECAL TUMOR
DEVELOPMENT IN APCMIN MICE AND REDUCE GROWTH OF HUMAN

COLON CANCER CELLS

54

ABSTRACT

Anthocyanins, which are bioactive phytochemicals, are widely distributed in
plants and especially enriched in tart cherries. Based on previous observations that
tart cherry anthocyanins and their respective aglycone, cyanidin, can inhibit
cyclooxygenase enzymes, we conducted experiments to test the potential of
anthocyanins to inhibit intestinal tumor development in ApcMin mice and growth of
human colon cancer cell lines. Mice consuming the cherry diet, anthocyanins, or
cyanidin had signiﬁcantly fewer and smaller cecal adenomas than mice consuming
the control diet or sulindac. Colonic tumor numbers and volume were not
signiﬁcantly inﬂuenced by treatment. Anthocyanins and cyanidin also reduced cell
grth of human colon cancer cell lines HT 29 and HCT 116. The IC50 of

anthocyanins and cyanidin was 780 11M and 63 uM for HT 29 cells, respectively and

285 1.1M and 85 uM for HCT 116 cells, respectively. These results suggest that tart

cherry anthocyanins and cyanidin may reduce the risk of colon cancer.

55

INTRODUCTION

Tart cherries contain substantial quantities of anthocyanins in addition to
other bioﬂavonoids (Wang et al., 1997). Anthocyanins, a member of the bioactive
phytochemicals, are widely distributed in ﬁ'uits, vegetables and beans, suggesting
that plant-based diets can provide considerable amounts of anthocyanins (Wang et al.,
1999; Seeram et al., 2001). Like the vast majority of ﬂavonoids, anthocyanins
primarily occur in plants as glycosides. Cyanidin is the major anthocyanin aglycone
in tart cherries. Montmorency and Balaton;rM tart cherries contain 0.40 to 0.80 mg/ g,
respectively, of anthocyanins (Wang et al., 1997). These anthocyanins were found to
function as antioxidants and cyanidin was shown to inhibit the activities of COX
enzymes in vitro (Wang et al., 1999a; Wang et al., 1999b). Several studies have
demonstrated that nonsteroidal anti-inﬂammatory drugs (N SAIDs) inhibit the growth
of colon tumors in animal models and reduce the risk of colon cancer in humans
(Giovannucci et al., 1995; Labayle et al., 1991). In most cases, colon
carcinogenesis depends on mutation of the APC gene, which is considered a
gatekeeper in the carcinogenic process (Kinzler and Vogelstein, 1996). Human APC
gene germline mutations cause familial adenomatous polyposis, an autosomal
dominantly inherited disease that predisposes affected individuals to develop
numerous adenomatous polyps and, ultimately, colorectal cancer (Su et al., 1992).
APC gene mutations also are a frequent and early event in sporadic colon cancer.
ApcMin mice are a mutant mouse lineage predisposed to multiple intestinal neoplasia

(Min) due to a mutation in the murine homolog of the adenomatous polyposis coli

(APC) gene (Moser et al., 1990). The primary phenotype of ApcMin mice is the

56

development of multiple intestinal adenomas that progress to adenocarcinomas of the
intestine in older mice.

The objectives of this research were to determine the potential of tart cherry
anthocyanins and cyanidin to inhibit intestinal tumor development in ApcMin mice
and to determine the potential of anthocyanins and cyanidin to directly inhibit the

grth of human colon cancer cells.

57

MATERIALS AND METHODS

Animals and Diets. This research was conducted with approval of the
Michigan State University All-University Committee on Animal Use and Care.
Mice were housed in a temperature and humidity-controlled room (20-22 °C, 50%)
with a 12-h light/dark cycle. ApcMin progeny were identiﬁed by a PCR-based assay
(Su et al, 1992) and were randomly assigned to ﬁve treatment groups (n = 10 per
group; equal numbers of males and females) at 4 - 5 weeks of age and fed treatment
diets for 10 weeks. Deionized water and diets was provided ad libitum. The
treatments (Table 2) were; 1) control diet (modiﬁed American Institute of Nutrition
93G diet containing 220 g/kg protein, 150 g/kg soybean oil and 50 g/kg cellulose), 2)
control diet + 800 mg/L anthocyanins in the drinking water, 3) control diet + 200
mg/L cyanidin in the drinking water, 4) control diet + 200 mg/L sulindac (an
NSAID) in the drinking water, and 5) modiﬁed control diet containing 200 gfkg
freeze—dried pitted tart cherries. Cherries were frozen, freeze-dried and ground
before they were incorporated into the diet. Anthocyanins were isolated from tart
cherries and were a mixture of 3-cyanidin 2"-0-,B-D-glucopyranosyl-6"-0-a-L-
rhamnopyransyl-ﬂ-D-glucopyranoside and 3-cyanidin 6"-0-a—L-rhamnopyranosyl-
,B—D-glucopyranoside (Wang et al, 1999) at 65 and 35%, respectively. The aglycone,
cyanidin, was prepared from the anthocyanins (Wang et al, 1997). Ascorbic acid (50
mg/L) was added to the drinking water of all mice to enhance the stability of

anthocyanins and cyanidin in solution by lowering the pH.

58

Table 2. Composition of the experimental diets (Percentage of diet)

 

 

Ingredient Modiﬁed AIN-93G diet 20% Cherries diet
Casein 22.12 22.12
Soybean Oil 15.00 15.00
Corn Starch 31.72 24.22
Dyetrose 10.57 8.07
Sucrose 10.00 0.00
AIN-93G-MX 3.87 3.87
AIN-93G-VX 1.11 1.11
L-Cystine 0.33 0.33
Choline Bitartrate 0.28 0.28
Tert-Butylhydroquinone 0.003 0.003
Cellulose 5.00 5.00
Dried Cherries 0.00 20.00

 

59

Tumor Number and Size. The mice were sacriﬁced after 10 weeks of treatment
and the numbers and sizes of adenomas in the intestinal sections were measured.
The entire small intestine, cecum, and colon were removed from each mouse to
determine the number and size of adenomas. Intestinal sections were opened
longitudinally, rinsed thoroughly with water, ﬁxed overnight in 10% neutral-buffered
formalin, and then stained with 0.2% methylene blue. Tumor numbers and
dimensions for each intestinal segment were determined by direct counting with the
aid of a dissecting microscope and measuring grid. The tumor sizes were determined
by measuring the spherical (three dimensional) volume of adenomas in the cecum
and colon and the average diameter of tumors in the small intestine. Tumors in the
cecum and colon of ApcMin mice typically are polypoid in appearance, whereas the
small intestinal tumors are sessile. Spherical volumes of cecal and colonic tumors
were calculated by the formula: Volume = 0.532 x (width x length x height of tumor).
Tumor numbers and diameters in the small intestine were analyzed by two-
way analysis of variance (treatment, sex). For tumor numbers and volumes in cecum
and colon, data were transformed to ranks and then ranks were analyzed by two-way
analysis of variance. When signiﬁcant treatment effects were detected (P<0.05),
treatment means were compared using the Least Signiﬁcant Difference method. Six
mice (one from the control group, two from the anthocyanin group, and three from
the cyanidin group) were excluded from the ﬁnal statistical analysis because it was

determined at the end of the experiment that these animals did not carry the APCMin

gene mutation, and therefore did not develop intestinal tumors.

60

Cell Culture and Growth Assays. The human colorectal cancer cell lines
HCT 116 and HT 29 (American Type Culture Collection) were cultivated in
McCoy’s 5A media supplemented with 10% fetal bovine serum. Cells were
harvested for growth assays when they had reached 50-80% conﬂuence by
trypsianDTA treatment and counted using a hemacytometer. Cells were then seeded
at 15 ,000 cells/well in 24-well tissue culture plates. Plates were incubated overnight
at 37°C and 5% CO; to allow cells to attach and begin proliferating.

At the beginning of treatments, the media was gently aspirated from each of
the wells, which were then rinsed with PBS. One ml of treatment media was added
to each well (n = 8-12 wells per treatment level per cell line) and plates were
incubated for 72 h. Treatment media was McCoy’s 5A media supplemented with
10% fetal bovine serum and containing the respective concentrations of anthocyanins
(0-1000 11M) or cyanidin (0-250 uM). Anthocyanins were dissolved in distilled
water before addition to the treatment media, whereas cyanidin was solubilized in
DMSO before addition. When used, the DMSO concentration was equalized for all
treatment media and never exceeded 0.1% (v/v) of the ﬁnal treatment media.

Total cell numbers in each well were quantiﬁed after 72 h of incubation in
treatment media. Cell numbers were calculated based on total DNA content in each
well using a procedure that quantiﬁes DNA based on ﬂuorescence of bound Hoescht
33258 (Labarca et a1, 1980). Fluorescence was measured by a Cytoﬂuor II
ﬂuorimeter (Applied Biosystems; Foster City, CA) using an excitation wavelength of
360 nm and an emission wavelength of 460 nm. Fluorescence readings were

converted to DNA by comparison to standard solutions of Salmon testis DNA

61

(Sigma Chemical Company; St. Louis, MO). The cell numbers in each well were
calculated by converting the quantity of DNA in each well by the amount of DNA
present in each cell (determined experimentally) for HCT 116 and HT 29 cells.

Cell numbers observed in each well after 72 h of growth were corrected for
initial cell number (determined at the time treatment media was added). These data
were then subjected to multiple regression analysis to develop least-squares
polynomial equations describing the inﬂuence of anthocyanins or cyanidin
concentration on cell number. These equations were then used to iteratively calculate
the concentration of anthocyanins or cyanidin required to cause a 50% reduction

(ICso) in grth (cell number) for each cell line.

62

RESULTS

Anthocyanins, Cyanidin and Tart Cherries Inﬂuence Intestinal Tumor
Development. Final body weights of mice were signiﬁcantly inﬂuenced by
treatment and averaged 22.8, 24.1, 21.3, 19.7, and 25.5 grams for mice consuming
control diet, anthocyanins, cyanidin, tart cherries, and sulindac, respectively (Fig. 12).
Final body weights for mice consuming anthocyanins and sulindac were greater
(P<0.05) than for mice consuming tart cherries. Averaged across all treatments,
male mice weighed signiﬁcantly greater (P<0.05) than female mice both at the
beginning (18.3 versus 15.6 grams) and the end (24.6 versus 20.8 grams) of the
experiment. Male mice weighed approximately 4 grams heavier than female mice.

Treatments had differential effects on tumor incidence and burden in the
various sections of the intestinal tract. Mice consuming anthocyanins, cyanidin, or
tart cherries had fewer (P<0.05) adenomas in the cecum than mice consuming the
control diet or sulindac (F ig. 13). The average number of adenomas in the cecum of
mice that consumed anthocyanins, cyanidin and tart cherries was 0.6, 0.6, and 0.6,
respectively, whereas mice that consumed the control diet or sulindac had an average
of 2.4 and 3.8 cecal tumors, respectively. Colonic adenoma number was not
inﬂuenced by treatment (Fig. 13). Average adenoma numbers in the colon were 3.6,
3.1, 3.6, 3.3, and 5.2 for mice that consumed the control diet, anthocyanins, cyanidin,
tart cherries, and sulindac, respectively. The total burden (volume) of cecal
adenomas was less (P<0.05) in mice consuming anthocyanins, cyanidin or tart
cherries when compared to mice consuming the control diet or sulindac (Fig. 14).

Adenoma volumes in the cecum of mice that consumed anthocyanins, cyanidin, or

63

tart cherries were 0.7, 0.6, and 1.8 mm3 respectively, whereas those in mice
consumed the control diet and sulindac were 3.0 and 4.0 mm3, respectively (Fig. 13).
Although mice that consumed tart cherries had the greatest adenoma burden in the
colon (8.4 m3), this was not statistically greater than that observed in mice
consuming the other treatments (Fig. 14).

Tumor multiplicity in the small intestine was not signiﬁcantly inﬂuenced by
treatment and averaged 48 tumors per mouse (Fig. 15). Mice that consumed sulindac '
had the smallest number of small intestinal adenomas (28 per mouse), but this was
not statistically different than small intestinal tumor numbers observed for the other
treatments. The average size of small intestinal adenomas (Fig. 16) was increased
(P<0.05) by feeding tart cherries (1.7 mm) and reduced (P<0.05) by sulindac (0.9
mm) relative to that observed in mice consuming the control diet (1.4 mm),

anthocyanins (1.3 mm), or cyanidin (1.3 mm).

 

l

l _._Contr0| —o—Anthocyanins ' ‘ " ' Cyanidin ii

i- + — Sulindac —II— Cherry

 

30

 

 

 

 

 

Figure 12. Inﬂuence of anthocyanins, cyanidin, tart cherries and sulindac on body

weights (g) of mice throughout the experimental period. Error bars indicate SEM.

65

 

{I Cecum I Colonj

 

 

Adenoma Number

 

 

 

E33

Anthocyanins Cyanidin Cherries Sulindac

 

Figure 13. Inﬂuence of anthocyanins, cyanidin, tart cherries and sulindac on
adenoma numbers in cecum (gray) and colon (black). Error bars indicate SEM.
Columns of the same shade not sharing a common superscript are signiﬁcantly

different (P<0.05).

66

 

1E Cecum I Colon ]

 

 

12.0

10.0 ~—

8.0 +

Cubic millimeters

 

 

 

. 2.241 at
Control Anthocyanins Cyanidin Cherries Sulindac

 

Figure 14. Inﬂuence of anthocyanins, cyanidin, tart cherries and sulindac on
adenoma volumes (mm3) in cecum (gray) and cOlon (black). Error bars indicate SEM.
Columns of the same shade not sharing a common superscript are signiﬁcantly

different (P<0.05).

67

too
90 -
80 -
70 ~
so 4
50 .
4o -
30 -
20 -
1o 4

O -

 

Adenoma Number

 

 

 

Control Anthocyanins Cyanidin Cherries Sulindac

Figure 15. Inﬂuence of anthocyanins, cyanidin, tart cherries and sulindac on

adenoma numbers in small intestine. Error bars indicate SEM.

68

 

-L

m N
1

T

1.61

ﬂ

I

T

Adenoma Diameter (mm)

 

 

.0 .O .0 _o 7" I"
O N -§ 0) m —* N A
1 1 1 % 1 H 1 1

   

 

 

 

Control Anthocyanins Cyanidin Cherries Sulindac

Figure 16. Inﬂuence of anthocyanins, cyanidin, tart cherries and sulindac on average
diameters (mm) of adenomas in small intestine. Error bars indicate SEM. Columns

not sharing a common superscript are signiﬁcantly different (P<0.05).

69

Anthocyanins and Cyanidin Inhibit Human Colon Cancer Cell Growth.
Treatment with anthocyanins (Fig. 17) or cyanidin (Fig. 18) caused a dose-dependent
reduction in cell numbers for both HCT 116 and HT 29 cells. Neither anthocyanins
nor cyanidin caused cytotoxicity even at the highest concentrations tested, as
indicated by little or no dead cells. Cyanidin was far more effective in inhibiting the
grth of these cancer cell lines than anthocyanins. The observed IC50 values for

cyanidin were 85 and 63 1.1M for HCT 116 and HT 29 cells, respectively, whereas

those for anthocyanins were 260 and 585 11M for HCT 116 and HT 29 cells,

respectively.

70

 

EJHCT116 IHT29'

 

120
100
80 ,
60 I
4o .
20 ~

Cell Number (% of control)

    

 

 

 

0 31.3 62.5 125 250 500 1000

Anthocyanin Concentration (um)

Figure 17. Inﬂuence of Enthocyanins on growth of human cancer cells, HCT 116

(gray) and HT 29 (black). Error bars indicate SEM.

71

cctccc Co .ch cchzz :60

 

F rgurt

and H'

IHCT116 IHT29

 

140
120

Cell Number (% of control)

20

100 -
60
40 ~ ‘

 

 

 

0 7.8 15.6 31.3 62.5 125 250
Cyanidin Concentration (um)

Figure 18. Inﬂuence of cyanidin on growth of human cancer cells, HCT 116 (gray)

and HT 29 (black). Error bars indicate SEM.

72

DISCUSSION
Our interest in testing the potential of tart cherry anthocyanins and cyanidin

Mi“ mice stemmed from the observation that

to inhibit tumor development in Apc
these compounds inhibit the activities of COX enzymes (Wang et al., 1999). Other
studies have demonstrated that sulindac (and other NSAIDs) reduce small intestinal
tumor multiplicity and size in ApcMin mice (Boolbol et al., 1996; Chiu et al., 1997;
Mahmoud et al., 2000). In this study, we found that anthocyanins, cyanidin, and tart
cherries (presumably as a source of anthocyanins) all signiﬁcantly reduced the
number and burden of tumors in the cecum of ApcMin mice. Conversely, sulindac
did not inﬂuence tumor development in the cecum. None of the treatments tested
inﬂuenced the numbers of tumors in the small intestine or the numbers or burden of
tumors in the colon. Our inability to detect a signiﬁcant reduction in small intestinal
adenoma number by sulindac was likely due to a number of factors, including the
relatively small numbers of mice per treatment group and large variations among
individual mice in adenoma development. We also have observed that the ability of
sulindac to inhibit small intestinal tumor development decreases as ApcMin mice age
(Bourquin, L. D., unpublished research). The lack of effect of anthocyanins or
cyanidin on colonic tumor development may be a consequence of their metabolism
by intestinal bacteria.

The average size of small intestinal tumors was smaller in mice consuming
sulindac and larger in mice consuming tart cherries when compared to that in mice
consuming the control diet, anthocyanins or cyanidin. Progression in the size of

Min

small intestinal tumors is highly correlated with morbidity in Apc mice. We have

73

observed that signiﬁcant morbidity and weight loss occur when small intestinal
tumors reach an average size of 1.5 mm diameter. At this stage, the tumors tend to
hemorrhage and may perforate the small intestine. The observed differences in ﬁnal
body weight in this experiment likely are a consequence of these differences in small
intestinal tumor promotion. It is well documented that sulindac and other NSAIDs
consistently reduce the size of small intestinal tumors in ApcMin mice (Boolbol et al.,
1996; Chiu et al., 1997; Mahmoud et al., 2000). The mechanism for the larger small
intestinal tumor size in mice consuming rat cherries is not known.

The effects of the treatments on tumor development were not consistent
throughout the intestinal tract. Sulindac and other NSAIDs typically reduce small
intestinal tumor development in ApcMin mice, but often have little effect on tumor
development in the cecum and colon (Barnes et al., 1998; also Bourquin, L. D.,
unpublished research). Mice consuming anthocyanins, cyanidin and tart cherries had
fewer and smaller tumors in the cecum, but these compounds did not afford any
protection to the small intestine. These results indicate that although anthocyanins
and cyanidin inhibit the activities of COX enzymes in vitro (Wang et al., 1999;
Seeram et al., 2001), they likely do not inﬂuence intestinal tumor development in
ApcMin mice through a pathway involving COX inhibition.

Both anthocyanins and cyanidin inhibited the growth of the colon cancer cell
lines HT 29 and HCT 116, although cyanidin was much more effective. We have
identiﬁed three degradation products from anthocyanins and cyanidin in cell culture

medium. These were protocatechuic acid, 2,4-dihydroxybenzoic acid, and 2,4,6-

trihydroxybenzoic acids (Seeram et al., 2001). In addition, we have detected trace

74

 

 

d1

(11

CC

bu

qu

for

sin

V11:

C01

C01

1'65

P36

quantities of cyanidin—3-glucoside and cyanidin in culture medium after 72 h of cell
grth in anthocyanin treatments. We have evaluated the potential of these
degradation products to inhibit the growth of HCT 116 and HT 29 cells. None of the
degradation compounds assayed demonstrated any inhibition of cell growth at
concentrations ranging up to 250 M (Seeram et al., 2001).

Anthocyanins and cyanidin are unstable at pH 7.0 and spontaneously degrade
to chalcone and benzoic acid derivatives. The red cyanidin cation is stable at pH < 3,
but deprotonates and produces ketoquinonoidal bases and ﬁnally an ionized
quinonoid base at pH > 7 (Seeram et al., 2001). At pH 3—6, the cyanidin cation
forms a carbinol pseudobase or chalcone pseudobase.

Anthocyanins are highly water-soluble and considered to be structurally
similar to a number of strong DNA intercalators (Mas et al., 2000). Both DNA and
RNA act as strong copigments for anthocyanins (Mistry et al., 1997). Also,
anthocyanins protect DNA against oxidative damage (Sarrna et al., 1999). Under in
vivo and cell culture conditions, both anthocyanins and cyanidin potentially form
corresponding pseudobases due to pH variations. These pseudobases are transition
compounds and may be stable in vivo as protein bound complexes. Given these
results, we predict that the anthocyanins, the aglycone cyanidin, or its varying
pseudobases directly suppress cell growth and subsequent tumor development.

We believe that our results are the ﬁrst to demonstrate that anthocyanins and
cyanidin have the potential to directly interfere with intestinal tumor development.
Hagiwara et al. (Hagiwara et al., 2001) demonstrated that anthocyanins in purple

corn color reduced the promotion of colon tumors caused by 2-amino-1-methyl-6-

75

phenylimidazo[4,5-b]pyridine (PhIP) in rats initiated with dimethyl hydrazine.
However, they did not test the potential of purple corn color anthocyanins to directly
suppress tumor development (Hagiwara et al., 2001).

In summary, we have demonstrated that tart cherry anthocyanins and their
aglycone cyanidin signiﬁcantly reduced tumor development in the cecum of ApcMin
mice. These compounds also directly inhibited the growth of human colon cancer
cells in vitro, with the aglycone cyanidin being far more effective than the
anthocyanin glycosides. Benzoic acid derivatives yielded from the degradation of
anthocyanins and cyanidin had no inﬂuence on colon cancer cell growth. Taken
together, these results suggest that cyanidin or its corresponding pseudobase is
directly inhibiting tumor development in the cecum of ApcMin mice. Anthocyanins
also are effective presumably due to their deglycosylation to cyanidin by cecal
bacteria. The lack of a clear suppression of tumor development in the colon

probably is due to further degradation of the cyanidin molecule by elevated pH in the

intestinal lumen and bacterial metabolism.

76

CHAPTER V

TART CHERRY ANTHOCYANINS REDUCE INTESTINAL ,
TUMORIGENESIS AND ALTER INTESTINAL EPITHELIAL CELL GENE
EXPRESSION, BUT Do NOT INFLUENCE INTRACELLULAR

DISTRIBUTION OF B-CATENIN IN INTESTINE OF APCMIN MICE.

77

ABSTRACT

Anthocyanins are ﬂavonoid compounds that impart color to fruits and
vegetables including cherries. Previous research in our laboratory has demonstrated
the potential of tart cherry anthocyanins to inhibit tumor development in ApcMin mice
and reduce proliferation of human colon cancer cell lines. The objective of this
research was to determine the extent to which dietary anthocyanins inhibit intestinal
tumorigenesis, alter B-catenin expression and distribution, and inﬂuence global gene
Min

expression proﬁles in intestinal epithelial cells of Apc mice. Mice were randomly

assigned to one of three treatment groups at 4 weeks of age (11 = 25 normal + 25
ApcMin mice per treatment). The treatments were 1) AIN 93G control diet, 2) the
control diet + 200 mg/kg sulindac, and 3) the control diet + 750 mg/kg anthocyanins.

The number of adenomas in the small intestine was reduced (P<0.05) by diets
containing either sulindac (:17 = 9) or anthocyanins (X = 17 ) relative to the controls

(X = 27). The number and volume of cecal adenomas were reduced (P<0.05) in
mice that consumed anthocyanins (0.6, 0.7 m3) compared to mice consuming
sulindac (2.1, 4.7 m3) or the control diet (1.0, 1.5 mm3). Mice consuming
anthocyanins had signiﬁcantly fewer colonic adenomas (0.8) than those consuming
sulindac (1.6) or the control diet (2.2). Colonic adenoma burden was not inﬂunced
by treatment. Dietary anthocyanins reduced tumor development throughout the

Min

intestinal tract of Apc mice, whereas tumor inhibition by sulindac was speciﬁc to

the small intestine. Anthocyanins increased (P<0.05) B-catenin mRNA expression in
middle small intestine and cecum. Intracellular distribution of B-catenin mRN A was

primarily in the epithelial cell membrane for mice on all dietary treatments. cDNA

78

microarray analysis showed that expression of mRNA of B-catenin was not altered
by anthocyanins or sulindac. Anthocyanins reduced the expression of cyclin D1
mRNA cOmpared to sulindac in cecum, but expression of mRNA for other predicted
target genes of the Wnt/B-catenin signaling pathway - proxisome proliferator
activated receptor 6, C-Myc and matrix metalloproteinase 7 — were not altered by
anthocyanins or sulindac. These observations suggest that the tumor inhibition
caused by feeding anthocyanins or sulindac was not a consequence of any reduction
in the cytoplasmic level and translocation of B-catenin to the nucleus and subsequent
inﬂuences on target gene expression. cDNA microarray analysis identiﬁed a total of
1,492 genes that were inﬂuenced by any treatment in the small intestine, colon or
ceum. Genes inﬂuenced by anthocyanins and sulindac are involved in a wide variety
of cellular and molecular processes including DNA synthesis and repair,
transcription, mRNA processing, translation, protein interactions and turnover, cell
proliferation and differentiation, cell cycle control, and programmed cell death. The
patterns of gene expression changes suggest that alterations of cellular and molecular

M'" mice.

processes by anthocyanins and sulindac likely are different in Apc
Additional conﬁrmatory analysis of these gene expression results will be necessary
to further establish potential mechanisms whereby anthocyanins and sulindac

inﬂuence tumor development.

79

INTRODUCTION

Tart cherries contain substantial quantities of anthocyanins in addition to
other bioﬂavonoids (Wang et al., 1997). Anthocyanins, a member of the bioactive
phytochemicals, are widely distributed in fruits, vegetables and beans, suggesting
that plant-based diets can provide considerable amounts of anthocyanins (Wang et al.,
1999; Seeram et al., 2001). Like the vast majority of ﬂavonoids, anthocyanins
primarily occur in plants as glycosides. Cyanidin is the major anthocyanin aglycone
in tart cherries. Montmorency and BalatonTM tart cherries contain 0.40 to 0.80 mg/ g,
respectively, of anthocyanins (Wang et al., 1997). These anthocyanins were found to
function as antioxidants and cyanidin was shown to inhibit the activities of COX
enzymes in vitro (Wang et al., 1999a; Wang eta1., 1999b). Several studies have
demonstrated that NSAIDs inhibit the growth of colon tumors in animal models and
reduce the risk of colon cancer in humans (Giovannucci et al., 1995; Labayle et al.,
1991). In previous research, we demonstrated that anthocyanins and cyanidin, when
administered in drinking water, reduced the numbers and total burden of tumors in
the cecum of ApcMin mice (Chapter IV). Cyanidin and anthocyanins also inhibited
the growth of human colon cancer cell lines HT 29 and HCT 116 in vitro (Chapter
IV).

Min

Apc mice are a mutant mouse lineage predisposed to multiple intestinal

neoplasia (Min) due to a mutation in the murine homolo g of the adenomatous

M' -
'" mrce

polyposis coli (APC) gene (Su et al., 1992). The primary phenotype of Apc
is the development of multiple intestinal adenomas that progress to adenocarcinomas

of the intestine in older mice. In most cases, colon carcinogenesis depends on

80

mutation of APC gene and this gene is considered a gatekeeper in the carcinogenic
process (Kinzler and Vogelstein, 1996). Most mutations in the human APC gene are
found to occur between codons 1286 and 1513, which is within a region encoding B-
catenin binding sites (Nakamura, 1993). These mutations result in the production of
a truncated APC protein product which cannot facilitate the degradation of B-catenin.
Recently, a nuclear export function of the APC gene has been identiﬁed (Rosin-
Arbesfeld et al., 2000). APC contains highly conserved nuclear export signals 3’
adjacent to the mutation cluster region that enable it to bind and export [3-catenin
from the nucleus. When this ability is lost due to APC mutation, B-catenin
accumulates in the nucleus.

B-catenin functions as a component of the E-cadherin adhesion complex and
also as a transcriptional regulator in a signaling pathway by binding to T cell factor
(ch) transcription factor in the nucleus. Aberrant accumulation and nuclear
localization of B-catenin can enhance trancriptional activation and expression of
target genes in this signaling pathway. A few target genes in this pathway —
peroxisome proliferator activated receptor 8 (PPARS), c-MYC, cyclin D 1 (CLNDI)
and matrilysin (matrix metalloproteinase 7) — have been recently identiﬁed (He et al.,
1999; He at al., 1998; Tetsu and McCormick, 1999; Crawford et al., 1999).
Screening changes in expression of these target genes as well as full genome in the
different sections of intestine should yield insights into the potential mechanisms

responsible for tumor modulation by anthocyanins and sulindac.

81

Expression of thousands of genes can be examined simultaneously using a
novel technique called cDNA microarray. Microarrays consist of a collection of
gene-speciﬁc cDNA fragments immobilized on glass slides which can be
simultaneously analyzed to ascertain their expression in tissue. Anthocyanins and
sulindac likely inﬂuence tumorigenesis in the intestinal tract by modulating the
expression of genes involved in carcinogenesis. The objectives of this study were to
determine the extent to which dietary anthocyanins and sulindac 1) inhibit intestinal
tumorigenesis, 2) inﬂuence the expression and celluar localization of B-catenin, and
Min

3) inﬂuence global gene expression proﬁles in intestine of Apc mice.

82

MATERIALS AND METHODS
Animals and Diets. This research was conducted with approval of the

Michigan State University All-University Committee on Animal Use and Care.

Min +/+

Apc mice were produced by mating normal C57BL/6J (Apc ) female mice with
Min C5 7BL/6J (ApcMiM) male mice. One-hundred-ﬁfty mice were grouped by sex,
weight and litter and then randomly assigned to three treatment groups (n= 50 per
group) at 4 weeks of age. Mice were not genotyped to determine Apc status before
the experiment, but it was expected that approximately 25 Min mice and 25 normal
mice would be randomly allotted to each treatment. Deionized water and diets were
provided ad libitum. Body weights were measured weekly. The dietary treatments
were; 1) control diet (modiﬁed American Institute of Nutrition 93G diet containing
220 g/kg protein, 150 g/kg soybean oil and 50 g/kg cellulose), 2) control diet + 200
ppm sulindac, and 3) control diet + 750 ppm anthocyanins. Sulindac and
anthocyanins were incorporated into the appropriate diets upon formulation. The
concentrations of sulindac (200 ppm) and anthocyanins (750 ppm) were based on
efﬁcacious concentrations used in a previous experiment conducted in our laboratory
(Chapter IV). Anthocyanins were isolated from tart cherries and were a mixture of
3-cyanidin 2"-0—,B-D-glucopyranosyl-6"-0-a—L-rhamnopyransyl-ﬂ-D-
glucopyranoside and 3-cyanidin 6"-0-a—L-rhamnopyranosyl-,6-D—glucopyranoside
(Wang et al., 1997) at 65 and 35%, respectively.

Tumor Number and Size. The mice were sacriﬁced after 8 weeks of dietary

treatment. A sample of liver tissue was obtained from each mouse for conﬁrmation

of Apc gene status by a PCR-based procedure (Su et al., 1992). The entire small

83

intestine, cecum, and colon were removed from each mouse. Intestinal sections were
opened longitudinally and rinsed thoroughly with water. Sixteen ApcMin mice from
each treatment were used for cDNA microarray analysis (8 males and 8 females).
The middle and distal two-thirds of the small intestine were ﬁxed overnight in 10%
neutral buffered formalin (NBF) and then stained with 0.2% methylene blue. Tumor
numbers and dimensions for the cecum and colon were determined immediately on
unﬁxed tissue sections — these sections and proximal small intestine were then used
for microarray analysis. For the remaining mice, all tissues were ﬁxed in 10% NBF
prior to tumor measurement and embedded for immunohistochemistry. Tumor
numbers and dimensions for each intestinal segment were determined by direct
counting with the aid of a dissecting microscope and measuring grid. The tumor
sizes were determined by measuring the spherical (three dimensional) volume of
adenomas in the cecum and colon and the average diameter of tumors in the small
intestine. Tumors in the cecum and colon of ApcMin mice typically are polypoid in
appearance, whereas the small intestinal tumors are seSSile. Spherical volumes of
cecal and colonic tumors were calculated by the formula: Volume = 0.523 x (width x
length x height of tumor).

Weekly body weights were statistically analyzed by repeated-measures
analysis of variance as a split-plot design. Main plot terms included treatment, sex,
Apc status and their interactions. Sub-plot terms included time and its interactions
with main plot effects. Tissue lengths were analyzed by two-way analysis of

variance (AN OVA) to detect the effects and potential interactions of treatment and

sex. Tumor numbers and diameters in the small intestine were analyzed by two-way

84

ANOVA (treatment, sex). For tumor numbers and volumes in cecum and colon, data
were transformed to ranks and then ranks were analyzed by two-way AN OVA.
When signiﬁcant effects were detected (P<0.05), appropriate means were compared
using the Least Signiﬁcant Difference method. Mice which did not carry the APCMin
gene mutation, and therefore did not develop intestinal tumors, were excluded from
the ﬁnal statistical analyses of tumor number and size.

Microarray Analysis. RNA Isolation and Extraction Sixteen (8 males and 8
Min

females) Apc mice from each treatment were used for microarray analysis. Total

RNA was extracted from pooled samples (4 mice with same sex for each pooled
sample) of the proximal small intestine, cecum and colon of these 16 ApcMin mice for
each treatment. Tumors in the cecum and colon were quantiﬁed under a dissecting
microscope and excised prior to epithelial cell isolation. Colon and cecum were
scraped with a glass slide to obtain epithelial cells. RNAlaterTM (200 to 1000 111,
Ambion, Austin, TX) was added and cells were stored at —80 °C. Approximately
800 mg of cells were obtained from the proximal small intestine section of each
mouse, and approximately 200 mg and 250 mg of cells were obtained from the
cecum and colon of each mouse, respectively. Equal amounts of cells from each
mouse were pooled within same treatment, sex and tissue and used for RNA
extraction and microarray analysis.

The improved single step RNA isolation method (Chomczynski and Sacchi,
1987) was used to isolate total RNA from epithelial cells. RNA was prepared with

Trizol LS reagent (Gibco BRL, Grand Island, NY) via manufacturer’s instructions.

Brieﬂy, cell pellets were resuspended in Trizol and incubated 5 minutes at room

85

temperature. Chloroform was added to the suspension to a ﬁnal concentration of
16 % (V/V) and phases were separated by centrifugation for 5 minutes (10,000 x g).
The aqueous phase was removed and RNA precipitated with isopropyl alcohol. The
total RNA pellet was washed with 75% ethanol and concentration determined by
spectrophotometry. RNA integrity was monitored by agarose gel electrophoresis.
Fluorescence Labeling of Microarray Probes. F our pooled RNA samples for each
tissue site (2 males and 2 females) were prepared from 8 male and 8 female ApcM’n
mice for each treatment. Methods for ﬂuorescence labeling followed the protocol
developed by Hegde et al. (2000). cDNA was prepared by reverse transcription of
total RNA samples (15 to 40 rig) in the presence of aminoallyl modiﬁed dUTP (aa-
dUTP). Reverse transcription was performed with Superscript II (Gibco) via
manufacturer’s instructions. Brieﬂy, the secondary structure of RNA was relaxed in
the presence of anchored oligo dT (T21VN). cDNA was produced upon addition of
dNTPs (aa-dUTP: dTTP 2:1) and Superscript II (20 units) and incubation at 42 °C
for 2 hours. Unincorporated dNTPs were removed using commercially available kit
(Qiagen PCR, Qiagen Inc., Valencia, CA) and probes were evaporated in a speed
vaccum for 1 hour.

Dye coupling was done by incubation of cDNA with the monoreactive dyes
cyanine 3 (Cy3) or cyanin 5 (Cy5) (Amersham, Buckinghamshire, UK) for 1 hour at
room temperature in the dark. Cy3 and Cy5 labeled samples for a given slide were
mixed following addition of hydroxylamine at a ﬁnal concentration of 2M.

Unincorporated dyes were removed using the Qiagen cleanup kit (Quiagen Inc.).

Each probe (Cy3 and Cy5 labeled pair) was dried in a speed vacuum and

86

resuspended in buffer (50% forrnamide, 5X SSC, 0.1% SDS + 20 pg COT-l DNA
and 20 ug polyA DNA). Each probe mixture then was applied to microarray slides
in a loop design (Fig. 19). After adding the probes, the slides were coverslipped and
placed in a hybridization oven at 42 °C for 20 hr. The slides then were removed and
washed with 1X SSC /0.2 % SDS until the cover slips were released. The slides
were again washed two times with 0.1X SSC/0.2% SDS for 5 min and 0.1XSSC for
1 min to remove residual SDS.

Preparation of the microarray slides. Microarray slides were printed with 15,360
genes received from the National Institute on Aging’s (NIA) mouse 15,264-gene
(15K) set. The 15K cDNA clones were rearrayed from 52,374 ESTs (expressed
sequence tags) derived from various embryonic and neonatal tissues. Full details on
development on the 15K gene set are described by Tanaka et al. (2000). The 15K
gene set was supplemented with an additional 96 known genes to allow symmetrical
printing of genes in the slides. The cDNA library contains sequences coding for
genes with known ﬁrnction, no hit genes (genes with no sequence information),
unknown genes (genes with sequence information only), and genes coding for
hypothetical protein (genes encoding a hypothetical protein based on high BLAST
[basic local alignment search tool] score).

Printing of this array was carried out using a superaldehyde substrate
according to the manufacturer’s protocol (Telechem). Brieﬂy, PCR ampliﬁcation of
inserts from each plasmid was performed using primers that contain a C6 amine
group coupled to the 5’ end. The amplicons were precipitated and resuspended in

spotting solution (3XSSC). The solutions were robotically spotted (Omnigrid robot,

87

GeneMachines, CA) onto superaldehyde coated glass slides at the Plant Research
Laboratory at Michigan State University. The unbound DNA was removed and the
bound probes denatured at 100°C. Free aldehydes were reduced with sodium
borohydride and the slides washed extensively with 0.1X SSC/0.2% SDS and
0.1XSSC.
Scanning, Immage Ouantitation, Normalization and Statistical Analysis. The slides
were scanned in an array scanner (Affymetrix 428TM tray, Santa Clara, CA) and
array image ﬁles were generated using the Jaguar 2.0 program TM (Affymetrix).
Both Cy3 and Cy5 channels for each of the 15,360 genes (spots) on each slide were
recognized, quantitated, and normalized against overall and peripheral background
intensity level using the GenImage program developed by Dr. Sorin Draghici’s
Bioinformatics lab at Wayne State University. These data were again normalized for
dye variation using the GP3 R Script developed by Dr. Timothy Zachrawski’s lab at
Michigan State University.

A total of 36 pooled RNA samples were used in this analysis (2 male and 2

Mi" mice for each of the three treatments for each tissue

female samples for Apc
[small intestine, cecum, colon]). The full dataset, based on Cy3 and Cy5 channels in
72 slides, includes four independent estimates of expression for each of the 15,360
genes for each pooled RNA sample. Outlier observations were excluded by deleting
any of the four independent observations of expression for each gene (for each RNA
sample) that differed from the mean by more than one standard deviation. The

remaining observations were averaged to obtain a single expression value for each

gene for each RNA sample. These data were then statistically analyzed using SAS

88

(version 8.1) by tissue site and by gene using a 2-way analysis of variance (treatment,
sex). When signiﬁcant treatment effects were detected (P<0.05), mean values were

compared using the Least Signiﬁcant Difference method.

89

SI Colon Cecum

//\\ /‘/’\\ //\\

A23 } A_—'S A—ES

Figure 19. Aminoallyl labeling 100p of RNA for microarray analysis

C, A and S represent three different dietary treatments, control, anthocyanins and
sulindac, respectively. Each arrow represents a single microarray slide. In each
slide the sample was labeled with Cy3 (tail of arrow) and also with Cy5 (head of
arrow). The entire scheme was repeated four times, corresponding to the four pooled
RNA samples (2 male, 2 female) obtained from animals on each treatment at each

tissue site. (SI; small intestine)

90

B-catenin Immunohistochemistry Analysis. Expression of B-catenin in
intestinal sections of normal and ApcMin mice exposed to the various dietary
treatments was conducted. Four micrometer parafﬁn-embedded sections of small
intestine, colon and cecum were deparafﬁnized and stained for B-catenin by indirect
peroxidase biotin-streptavidin immunohistochemistry.. Sections were pretreated by
an antigen retrieval procedure using 10 mM citrate buffer (pH 6) for 20 minutes at 95
°C and preblocked with 3% H202 in PBS followed by normal goat serum. After
incubation of sections for 14 hr at 4 °C with anti-B-catenin monoclonal antibody
(Transduction Laboratories, Lexington, KY) at a dilution of 1:250, biotinylated anti-
mouse immunoglobulin (DAKO Laboratories, Carpinteria, CA) was applied to the
sections at room temperature for 40 minutes. Sections then were treated with
streptavidin-peroxidase (DAKO) at room temperature for 40 minutes. Linked
peroxidase was Visualized by staining with aminoethylcarbazole (DAKO) for 30
minutes. The B-catenin immunostaining results for different locations within the
tissue (top and bottom for small intestine and top, middle, and bottom for colon and
cecum) and distribution within the cells (membrane and cytoplasm) were evaluated
by visually comparing the staining intensities. The intensity of B-catenin
immunostaining was scored using a scale ranging from absent (0) to very high (4).
The scores for intensity of B-catenin immunostaining were transformed to

ranks and then ranks were analyzed by four-way analysis of variance (treatment, Apc

status, region within the tissue, and distribution within the cells). When signiﬁcant

91

effects were detected (P<0.05), means were compared using the Least Signiﬁcant

Difference method. Images in this dissertation are presented in color.

92

RESULTS

Anthocyanins Inhibit Intestinal Tumor Development. Averaged across all
treatments, sexes and time points, mice consuming control (21.8 g) and anthocyanins
(21.3 g) weighed more than mice consuming sulindac (20.3 g) (P<0.05). Male mice
weighed signiﬁcantly more (P<0.01) than female mice both at the beginning (14.0
versus 13.0 grams) and the end (26.4 versus 22.3 grams) of the experiment (Fig. 20).
Final body weights of ApcMin mice consuming control diet, anthocyanins, and
sulindac averaged 24.8, 23.5, and 24.9 grams , respectively. The lengths of the small
intestine, cecum and colon were not inﬂuenced by dietary treatment (data not shown).

Mice consuming sulindac had more (P<0.05) adenomas in the cecum than
mice consuming control diet or anthocyanins (Fi g. 21). Cecal adenoma number for
mice consuming control diet also was greater (P<0.05) than in mice consuming
anthocyanins. The average number of adenomas in the cecum of mice that
consumed anthocyanins, sulindac and control diet was 0.6, 2.1, and 1.0, respectively.
Colonic adenoma number also was inﬂuenced by treatment (Fig. 21), with mice
consuming control or sulindac having more tumors (P<0.05) than mice consuming
anthocyanins. Average adenoma numbers in the colon were 0.8, 1.6, and 2.2 for
mice that consumed anthocyanins, sulindac and control diet, respectively. The total

burden (volume) of cecal adenomas was smallest (P<0.01) in mice consuming

anthocyanins when compared to mice consuming control diet or sulindac (Fig. 22).

3

9

Adenoma volume in the cecum of mice that consumed anthocyanins was 0.7 mm
whereas adenoma volumes in mice consuming control diet and sulindac were 1.5 and

4.7 m3, respectively (Fig. 22). The total burden of colonic adenomas was not

93

statistically signiﬁcant at the P<0.05 level, but there was a trend that anthocyanins
and sulindac reduced the total burden of colonic adenomas relative to mice

consuming the control diet (P=0.08). Adenoma volumes in the colon of mice that

3

9

consumed anthocyanins, control, and sulindac were 2.6, 5.6, and 3.0 mm
respectively (Fig. 22).

Tumor multiplicity in the small intestine was signiﬁcantly inﬂuenced by
treatment (P<0.05). Adenoma number of the small intestine was obtained by
combing adenoma numbers of the middle and distal small intestine segments. Mice
that consumed sulindac had the smallest number of small intestinal adenomas (9 per
mouse) whereas mice that consumed the control diet had the greatest number of ,
small intestinal adenomas (27 per mouse) (F i g. 23). Adenoma number of small
intestine was intermediate in mice consuming anthocyanins (17 per mouse) (Fig. 23).
Mice consuming sulindac had signiﬁcantly fewer adenomas in both middle and distal
small intestine segments than mice consuming control diet, and signiﬁcantly fewer
adenomas in the distal small intestine than mice consuming anthocyanins (P<0.05)
(Fig. 23). The average size of small intestinal adenomas (Fig. 24) was reduced
(P<0.05) by feeding sulindac (1.1 mm) relative to that observed in mice consuming
the control diet (1.3 mm). The average size of small intestinal adenomas was

intermediate in mice consuming anthocyanins (1.2 mm) (Fig. 24).

94

a) Male normal mice b) Female normal mice

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

—*—C +3 ---A---A —x—C —-——S man-A
35 3o
304 .5" 251
3.421-
25“ IBM 20
m 1,. .
E 20~ 4‘55 E”
(U / (015-
6 15:1 6
10¢ 10"
5« 5:
0 I I ' ' I ' I 0 I r T T I r f
o 1 2 3 4 5 6 7 8 o 1 2 3 4 s 6 7 8
VVeek VVeek
c) Male ApcMm mice (1) Female ApcMm mice
—x—C —I—S ---a---A
25 j
20’ ﬁaﬂdﬂ
Earn”
a aw
(U 15 (U {I
L— L.-
o 1 om
10«
5« 5‘
0 1' T Y T Y I U 0 I U I I I I I
o 1 2 3 4 s 6 7 8 o 1 2 3 4 5 5 7 8
Week Week

Figure 20. Inﬂuence of treatment on body weights (g) of mice throughout the
experimental period. C; control, S; sulindac, A; anthocyanins. Error bars indicate

SEM. a) Body weights of male normal mice b) Body weights of female normal

mice c) Body weights of male ApcMin mice (1) Body weights of female ApcMin mice

95

 

 

Cecum I Colon

 

 

 

I"
more:

Adenoma Number
8

0.5

 

 

 

Control Anthocyanins Sulindac

Figure 21. Inﬂuence of anthocyanins and sulindac on adenoma number in cecum
(gray) and colon (black). Error bars indicate SEM. Columns of the same shade not

sharing a common superscript are signiﬁcantly different (P<0.05).

96

 

 

 

Cecum I Colon

 

\)

 

Adenoma Volume (mma)
N w «h Uu ON

ﬂ

 

 

 

0

Control Anthocyanins Sulindac

Figure 22. Inﬂuence of anthocyanins and sulindac on total adenoma volumes (mm3)
in cecum (gray) and colon (black). Error bars indicate SEM.

8’ b’ cSigniﬁcant treatment effect (P<0.05). Columns not sharing a common
superscript are signiﬁcantly different.

d' cTreatment effect trend (P=0.08).

97

 

EMSI DSI ITotal

 

 

 

 

Adenoma Number

 

Control Anthocyanins Sulindac

 

Figure 23. Inﬂuence of anthocyanins and sulindac on adenoma number in small
intestine. MSI; middle small intestine, DSI: distal small intestine. Error bars
indicate SEM. Columns of the same shade not sharing a common superscript are

signiﬁcantly different (P<0.05).

98

 

 

.3

b
1
CT

Adenoma Mean Diameter (mm)
o
oo

 

 

 

 

Control Anthocyanins Sulindac

Figure 24. Inﬂuence of anthocyanins and sulindac on mean diameters (mm) of
adenomas in small intestine. Error bars indicate SEM. Columns not sharing a

common superscript are signiﬁcantly different (P<0.05).

99

 

Anthocyanins Do Not inﬂuence Intracellular Distribution of B-catenin.
When the score of intensity of B-catenin immunostaining was compared, the

expression of [3-catenin protein was consistently higher at the top of villi in the small
intestine and top one third of crypts in cecum and colon of all carrier (ApcMin) and
normal (Apc*’+) mice (P<0.05) (Fig. 25, 26 and 27). Within epithelial cells,
expression of B-catenin protein was higher in cell membrane than in cytoplasm in all
tissue sections (P<0.05) (Fig. 25, 26 and 27). Nuclear expression of B-catenin

protein was negligible in all tissue sections. The expression of B-Catenin protein was

Min

greater (P<0.05) in intestinal tissues of Apc mice than those of normal mice in all

sites (Fig. 25, 26 and 27).

Mice consuming anthocyanins had increased expression of B-catenin protein
in the cecum relative to mice consuming control diet or sulindac (P<0.05) (Table 4).
Signiﬁcant interactions between treatment and Apc status were found in the proximal
and middle small intestinal sections. The expression of B-catenin protein was
increased in the proximal small intestine as well as in the middle small intestine of
Min

Apc mice consuming anthocyanins compared with those consuming control diet

(P<0.05) (Table 4).

100

(a) Small Intestine

    
    

N (Apc+’*) C (Ape+ '“)
(b) Colon

 

N (Apc*’*) C (Ape+ i")

(c) Cecum

N (Apc+ +) C (AchMm)
Figure 25. Expression of B-catenin in mice consuming the control diet.

101

(a) Small Intestine

 

wah qmﬂ"
(b) Colon

 

NMWD CNW”D
(c) Cecum

 

N (Apc+ 4») C (AchMin)
Figure 26. Expression of [El-catenin in mice consuming sulindac

102

(a) Small Intestine

 

wipe“) " I V C(Ap")
(b) Colon

 

N (APCH) C (Apc*

(c) Cecum

    

’ N (Ape+ ‘) C (Ape+ " I.)
Figure 27. Expression of B-catenin in mice consuming anthocyanins

103

Table 3. Average intensity scores of B-catenin expression in intestinal sectionsl’

A. Membrane staining in normal mice

2

 

 

 

 

 

 

 

 

 

 

 

Treatment PSI3 MSI3 DSI Colon Cecum4
Control 1.8 i- 0.3abc 1.5 i- 0.2b 1.5 i 0.4 1.7: 0.4 1.2 i- 0.221
Sulindac 1.5 i 0.3ab 1.3 3. 0.2“ 2.3 i 0.3 1.2 i 0.3 1.3 i 0.38
. Anthocyanins 1.5 i 0.33 2.3 i 0.2c 2.0 i 0.3 1.8 i 0.3 2.0 i 0.2b
B. CytoSol staining in normal mice
Treatment PSI MSI DSI Colon Cecum
Control 0.7 i 0.3“bc 0.3 i 0.2b 0.5 i 0.4 0.7 i 0.4 0.7 i 0.23
Sulindac 0.5 :1: 0.3ab 0.3 i 0.2“b 0.8 i 0.3 0.7 i 0.3 0.3 i 0.3a
Anthocyanins 0.3 i 0.3a 0.8 :r. 0.2c 0.5 i 0.3 0.7 i 0.3 0.8 i 0.2b
C. Membrane staining in ApcMin mice
Treatment PSI MSI DSI Colon Cecum
Control 1.5 i 0.2“ 0.8 i 0.2“ 1.8 3. 0.2 1.8 i 0.2 1.6 i 0.28
Sulindac 1.7 i 0.2be 1.5 i 0.2b 2.3 i 0.2 2.1 i 0.2 1.4 i 0.23
Anthocyanins 2.0 i 0.2c 2.3 i 02° 2.3 i 0.2 2.0 i- 0.2 2.1 i 0.1b
D. Cytosol staining in ApcMin mice
Treatment PSI MSI DSI Colon Cecum
Control 0.3 i 02“" 0.0 i‘ 0.23 0.5 i- 0.2 0.9 i 0.2 0.6 i 0.23
Sulindac 0.8 _+. 0.2be 0.5 i 0.21) 0.7 :1: 0.2 0.9 i 0.2 0.3 i 0.2a
Anthocyanins 1.0 i 02° 1.3 i 02° 0.8 :1: 0.2 1.4 i 0.2 0.8 i 0.2b

 

IPSI; proximal small intestine, MSI; middle small intestine, DSI; distal small

intestine.

2The intensity of B-catenin immunostaining was scored using a scale ranging from

absent (0) to very high (4).

3Treatment x carrier status interaction (P<0.05). Means in the same column not
sharing a common superscript are signiﬁcantly different (p<0.05).
4Treatment effect (P<0.05). Means in the same column within each category not
sharing a common superscript are signiﬁcantly different (p<0.05).

Anthocyanins and Sulindac Do Not Inﬂuence Expression of mRNA of
APC, B-catenin and Most of the Predicted Target Genes of Wnt/B-catenin

Signaling. Expression of B-catenin and the predicted target genes of the Wnt
signaling pathway was determined in small intestine, colon and cecum of ApcMin
mice using cDNA microarray analysis (Table 4). Anthocyanins and sulindac did not
alter the expression of mRNA of APC and B-catenin in all three intestinal sections of
ApcMin mice. For the predicted target genes of the Wnt/B-catenin signaling,
anthocyanins decreased the expression of cyclin D1 compared to sulindac in cecum
(0.59 vs 0.79). Treatment did not inﬂuence the expression of mRNA for the other
predicted target genes — MMP7, PPARB and C-Myc - in all three tissue sections of
ApcMin mice.
cDNA Microarray Analysis Identified Several Clusters of Differentially
Expressed Genes in Small intestine, Colon and Cecum of ApcMin Mice. All gene
identiﬁcations were based on BLAST scores greater than 100 as of May 1, 2002.
There were 497 genes inﬂuenced by any treatment (F value for treatment effect <
0.05) in SI, 534 in Colon, and 501 in cecum (Fig. 28). None of the genes was
inﬂuenced by treatment in all three intestinal sections. There were 8 genes
inﬂuenced by treatment both in small intestine and colon (Table 5). Relative to
controls, sulindac down-regulated 3 of these 8 genes in small intestine and four
genes in colon. Sulindac up-regulated 2 genes in colon compared with controls.
Relative to controls, anthocyanins up-regulated 2 Of the 8 genes and down-regulated

three genes in small intestine. Anthocyanins down-regulated three of the 8 genes in

colon compared with controls. There was little overlap in the effects of sulindac and

105

anthOcyanins on expression of these eight genes. In small intestine, both compounds
down-regulated the expression of a single gene (unknown function). Both
compounds down-regulated the expression of placental lacto gen 2 in colon, but
expression of this gene was up-regulated by anthocyanins and not inﬂuenced by
sulindac in small intestine.

There were 12 genes inﬂuenced by treatment both in small intestine and
cecum (Table 6). These include 6 genes coding for known proteins (voltage-
dependent calcium channel beta-3 subunit, grth arrest speciﬁc protein 5, tubulin
beta 3, collagen IV, kangai), 5 unknown and 1 hypothetical protein gene. Of these
12 genes, sulindac reduced expression of 3 genes in small intestine and none of the
genes in the colon when compared to controls. None of these 12 genes were up-

_ regulated by sulindac relative to control. In small intestine relative to controls,
anthocyanins reduced the expression of calcium channel beta 3 subunit, growth
arrest protein 5, tubulin beta 3, collagen IV and one unknown gene and increased the
expression of alternative splicing factor, 1 unknown gene, and 1 hypothetical protein.
Expression of these genes was similary affected by anthocyanins in cecum except for
an increased expression of calcium channel subunit and decreased expression of
kangai 1, (Table 6).

Twenty genes were differentially inﬂuenced by treatment in both cecum and
colon. Among those genes changed both in colon and cecum, anthocyanins up-
regulated 9 genes and down-regulated 3 genes compared to both controls and
sulindac in cecum (Table 7). The up-regulated genes were nuclear antigen SplOO,

serine/threonine-protein kinase 4, acid ceramidase 1, CD36 antigen-like 2, signal

106

transducing adaptor molecule 2, UMP-CMP kinase, and 3 unknown genes. The
down- regulated genes relative to both control and sulindac were zinc ﬁnger protein
RIN ZF, 1 no hit, and 1 hypothetical protein gene. In colon, sulindac inﬂuenced the
expression of most of these 20 genes, whereas anthocyanins did not, compared to
control. Relative to controls, sulindac up-regulated genes for zinc ﬁnger protein LIN
ZF, MORF-related gene X, nuclear antigen SplOO, serine/threonin-protein kinase 4,
acid ceramidasel, 1 no hit, and 7 unknown genes and down-regulated CD36 anti gen-
like 2, signal transducing adaptor molecule 2, UMP-CMP kinase, 1 hypothetical
protein, and 1 no hit gene (Table 7).

Differentially Expressed Genes in Cecum of ApcMin Mice. Cecal -
epithelium from anthocyanin-treated mice had signiﬁcantly increased expression of
79 genes compared to mice fed the control and sulindac diets (Table 8).
Anthocyanins increased the expression of the hemachromatosis (HFE) gene, ras-
related C3 botulinum toxin substrate 1, and Ly6/Neurotoxin 1 protein genes more
than 2-fold compared to both control and sulindac.

Anthocyanins also signiﬁcantly decreased the expression of 44 genes
compared to both control and sulindac (Table 9). The decrease in expression of 2
unknown genes by anthocyanins was more than 2-fold compared to control or
sulindac.

Differentially Expressed Genes in Colon of ApcM’" Mice. Anthocyanins
signiﬁcantly increased the expression of 33 genes in colonic epithelium compared to
both control and sulindac treated mice (Table 10). Anthocyanins signiﬁcantly

decreased the expression of 33 genes compared to both control and sulindac treated

107

mice (Table 11). The larger fold changes were an increase of expression for
cytochrome b5, adolase C5 and 1 unknown gene (Table 10) and decrease for serum
protein MSE55 (Table 11).

Differentially Expressed Genes in Small Intestine of ApcMin Mice. In
' small intestine, 53 and 244 genes were up-regulated by sulindac (Table 12 and 13)
and anthocyanins (Table 13 and data not shown), respectively, when compared to
controls. Among these genes, 19 genes were up-regulated by sulindac and
anthocyanins (Table 13). The genes that were increased in their expression in small
intestine both by sulindac and anthocyanins include 9 known genes. Among genes
inﬂuenced speciﬁcally by sulindac (Table 12), the expression of ATPase subunit 4
gene was increased by 3.36-fold in mice consuming sulindac compared to mice
consuming the control diet. The expression of Tax interacting protein gene was
increased most (1.5-fold) by sulindac and LIM domain binding 1 gene (1.79-fold) by
anthocyanins (Table 13).

Sulindac and anthocyanins also down-regulated a large group of genes in the
small intestine. 108 genes were down-regulated by sulindac (Table 14 and 15) and
47 genes were down-regulated both by sulindac and anthocyanins (Table 15) when
compared to control. Genes that were down-regulated speciﬁcally by sulindac more
than 1.5-fold were 1 no hit and 1 unknown gene. Adipocyte-speciﬁc protein 5 and
novel cell death-regulatory protein GRIM19 were down-regulated by 1.49-fold and
1.44-fold, respectively (Table 14). Genes that were down-regulated more that 1.5-
fold in their expression by both sulindac and anthocyanins were fatty acid binding

protein 3, glutathione S-transferase and 2 unknown genes (Table 15).

108

Table 4. Expression of B-cate'nin and target genes of Wnt/B-catenin signaling in

 

 

intestinal sections of ApcMin mice" 2
Plate ID3 Gene Name C S A SEM4
Small Intestine
H3080G10 APC 0.75 0.69 0.77 0.08
H3080F09 APC 0.62 0.71 0.63 0.05
H3031E05 B-catenin 1.26 1.59 1.25 0.23
H3084D05 Cyclin D1 0.67 0.64 0.57 0.04
H3093H12 MMP7 0.69 0.71 0.80 0.04
H3155B12 PPARdelta 1.48 1.13 1.45 0.24
H3 076D1 0 C-myc 1.26 1.27 1.17 0.12
H3089H11 C-myc 0.87 0.80 0.70 0.09
Colon
H3080G10 APC 0.64 0.76 0.72 0.07
H3080F09 APC 0.73 0.78 0.79 0.04
H3031E05 B-catenin 1.45 1.54 1.67 0.20
H3084D05 Cyclin D1 0.66 0.61 0.62 0.06
H3093H12 MMP7 0.84 0.83 0.79 0.05
H3155B12 PPARdelta 1.16 1.08 1.18 0.14 '
H3076D10 C-myc 1.24 0.97 1.00 0.1 1
H3089H11 C-myc 0.63 0.71 0.67 0.06
Cecum
H3080G10 APC 0.68 0.73 0.82 0.06
H3080F09 APC 0.72 0.69 0.68 0.03
H3031E05 B-catenin 1.48 1.70 1.67 0.31
H3084D05 Cyclin D1 0.69ab 0.76b 0.59a 0.16
H3093H12 MMP7 0.88 0.85 0.70 0.12
H3155812 PPARdelta 1.18 1.09 1.16 0.27
H3076D10 C-myc 1.01 0.93 0.90 0.14
H3089H11 C-myc 0.64 0.65 0.71 0.06

 

IData represent mean intensity values of mRNA expression for the corresponding

gene in the treatment groups, C(control), S(sulindac) and A(anthocyanins).
Microarray slides have two spots for a few identical genes. APC and C-myc have

two spots each and 2 intensity values from each spot were obtained.

3 Plate ID: sequence number for NIA 15K mouse cDNA library set.

4SEM; standard error of mean.

a""cMeans in the same row not sharing a common superscript are signiﬁcantly

different (P<0.05).

109

Small Intestine

 

Figure 28. Numbers of genes inﬂuenced by any treatment (P<0.05) in
intestinal sections of APCMin mice. Numbers overlapping two sections were

inﬂuenced by treatment in both tissues.

110

Table 5. Expression of 8 genes inﬂuenced by treatment both in small intestine and
colon1

 

 

Plate ID2 Gene Name C S A SEM3
Small Intestine
H3046BO4 placental lactogen2 0.66a 0.753 0.94b 0.05
H3085E07 unknown 0.72a 0.85ab 0.91b 0.04
H3022F05 3-phosphoglycerate dehydrogenase (Phgdh) 0.84ab 0.7821 1.06b 0.06
H3084BO6 F-box protein FBX15 . 0.80b 0.65a 0.88b 0.03
H3045C02 ubiquitin-conjugating enzyme 0.69b 0.54a 0.71b 0.02
H3003D111ymphoid-speciﬁc transcription factor NFATc3 1.70b 1.34ab 0.71a 0.20
H3090D11unknown 0.89b 0.708 0.71“ 0.04
H3015G01no hit found 1.611) 1.27““’1.16a 0.10
Colon
H3046B04placentallactogen2 0.80c 0.72b 0.64a 0.01
H3085E07 unknown 0.66a 0.93b 0.63“ 0.06
H3022F05 3-phosphoglycerate dehydrogenase (Phgdh) 0.75a 0.88b 0.82ab 0.03
H3084B06 F-box protein FBX15 0.84b 0.67b 0.5921 0.06
H3045C02 ubiquitin-conjugating enzyme 0.73b 0.75b 0.67a 0.02
H3003D111ymphoid-speciﬁc transcription factor NFATo31.92b 1.32a 1.71b 0.09
H3090D11unknown 0.92b 0.71a 0.92"~0.04
H3015G01no hit found 1.48b 1.2521 1.59b 0.05

 

rData represent mean intensity values of mRNA expression for the corresponding
gene in the treatment groups, C(control), S(sulindac) and A(anthocyanins).

Plate ID: sequence number for NIA 15K mouse cDNA library set
3SEM; standard error of mean.

2|“Means in the same row not sharing a common superscript are signiﬁcantly
different (P<0.05).

Ill

Table 6. Expression of 12 genes inﬂuenced by treatment both in small intestine and

l

 

 

cecum
Plate ID? Gene Name C S A SEM3
Small Intestine
voltage-dependent
H3117A06 calcium channel beta-3 subunit 0.75“ 0.59“ 0.64“ 0.02
H3113A12 growth arrest speciﬁc protein5 1.07“ 0.88““ 0.79“ 0.06
H3057B12 tubulinbeta3 1.44“ 1.27““ 1.00“ 0.08
H3067F05 collagenIV 0.96“ 0.72“ 0.66“ 0.06
H3051G02 unknown 1.68“ 1.46“b 1.07“ 0.13
kangai 1
H3154D02 (suppression of tumorigenicity 6) 0.82ab 0.61a 0.91b 0.06
H3060C12 unknown 0.84b 0.66“ 0.78“ 0.03
H3119E07 hypothetical 0.75“ 0.73“ 0.86b 0.03
H3104H02 alternative splicing factor 0.79a 0.813 0.98b 0.02
H3033B06 unknown 0.74““ 0.65“ 0.94“ 0.05
H3042A04 unknown 0.74“ 0.69“ 0.87“ 0.03
H3064D02 unknown 0.73““ 0.63“ 0.80“ 0.03
Cecum
voltage-dependent
H3117A06 calcium channel beta-3 subunit 0.62“ 0.61“ 0.75“ 0.03
H3113A12 growth arrest speciﬁc protein5 1.78“ 1.55““ 1.38“ 0.08
H3057B12 tubulin beta3 3.16“ 2.40““ 1.94“ 0.26
H3067F05 collagen 1v 0.88“ 0.91“ 0.66“ 0.05
H3051G02 unknown 1.35“ 1.16““ 0.81“ 0.12
kangail
H3154D02 (suppression oftumorigenicity 6) 0.84“ 0.70““ 0.63“ 0.04
H3060C12 unknown 0.62“ 0.63“ 0.73“ 0.02
H3119E07 hypothetical 0.63“ 0.62“ 0.98“ 0.04
H3104H02 alternative splicing factor 1.55“ 1.29“ 1.04“ 0.08
H3033B06 unknown 0.68“ 0.67“ 0.91“ 0.05
H3042A04 unknown 0.77“ 0.76b 0.62“ 0.04
H3064D02 unknown 0.61“ 0.59“ 0.73“ 0.02

 

TData represent mean intensity values of mRNA expression for the corresponding

gene in the treatment groups, C(control), S(sulindac) and A(anthocyanins).
Plate ID: sequence number for NIA 15K mouse cDNA library set

3SEM; standard error of mean.

a'b'cMeans in the same row not sharing a common superscript are signiﬁcantly

different (P<0.05).

112

Table 7. Expression of 20 genes inﬂuenced by treatment both in cecum and colon1

 

 

Plate ID2 Gene Name C S A SEM3
Cecum
H3112C06no hit found 3l.02“30.29“7.06“ 4.99
histocompatibility complex region
H3073B01containingtheQregion ofclassI 0.76“ 0.69““ 0.59“ 0.04
H3061A052inc ﬁnger proteinRINZF (RINZF) 0.82b 0.85b 0.663' 0.04
H3114G07no hit found 1.25“ 1.04“ 1.01“ 0.05
H3117D10hypothetical 0.93“ 0.87b 0.78“ 0.03
H3088E03 unknown 0.85““088“ 0.77“ 0.03
H3132A08unknown 0.64““ 0.70“ 0.59“ 0.02
H3148F01 unknown 0.69“ 0.60“ 0.66““ 0.02
H3068E08 unknown 0.66“ 0.81“ 0.63“ 0.03
H3030E03 unknown 0.66“ 0.77“ 0.71““ 0.02
H3078H05MORF-related geneX 0.70““065“ 0.83“ 0.04
H3052B09nuclear antigen Sp100 (Sp100) 0.61" 0.612’ 0.74b 0.03
H3056H07unknown 0.59“ 0.63“ 0.72“ 0.02
SERINE/THREONINE-PROTEIN KINASE 4
H3138A02(P21-ACTIVATED KINASE 4) 0.58“ 0.61“ 0.74“ 0.02
H3092C04 acid ceramidase(Asah1) 0.57“ 0.64“ 0.78“ 0.04
H3156B09 unknown 0.63“ 0.63“ 0.90“ 0.03
CD36 antigen (collagen type I receptor,
H3111D12thrombospondin receptor)-like2(Cd3612) 1.74a 1.858 2.53b 0.19
signal transducing adaptor molecule
H3007H10(SH3 domain and ITAM motif)2 1.76“ 1.51“ 2.79“ 0.28
H3144A09unknown 0.83“ 0.82“ 1.35“ 0.13
H3007F10 UMP-CMP kinase (UMP-CMPK) 1.51“ 1.62“ 2.69“ 0.16
Colon
H3112C06no hit found 25.10“7.04“ 22.14“3.29
histocompatibility complex region
H3073BOl containing theQregion ofclassI 0.69““ 0.79“ 0.64“ 0.03
H3061A052inc ﬁnger proteinRINZF (RINZF) 0.69a 0.90b 0.68‘1 0.05
H3114G07no hit found 1.08“ 1.61“ 1.20“ 0.06
H3117D10hypothetical 0.97“ 0.75“ 0.79“ 0.04
H3088E03 unknown 0.71“ 0.85“ 0.75“ 0.02
H3132A08unknown 0.62“ 0.79“ 0.68“ 0.03
H3148F01 unknown 0.62“ 0.72“ 0.68““ 0.02
H3068E08 unknown 0.59“ 0.72“ 0.65“ 0.02

 

113

 

H3030E03 unknown 0.69“ 0.94“ 0.62“ 0.04

H3078H05MORF-related geneX 0.70“ 0.88“ 0.68“ 0.05
H3052B09nuclear antigen Sp100(Sp100) 0.59“ 0.73“ 0.63““ 0.03
H3056H07unknown 0.66““ 0.72“ 0.58“ 0.03
SERINE/THREONINE-PROTEN KINASE 4
H3138A02(P21-ACTIVATED KINASE 4) 0.65“ 0.85“ 0.66“ 0.05
H3092C04 acid ceramidase(Asah1) 0.63“ 0.82“ 0.62“ 0.04
H3156B09unknown 0.81“ 0.89“ 0.81“ 0.02

CD36 antigen (collagen type I receptor, .
H3111D12thrombospondin receptor)-like 2 (Cd3612) 2.01b 1.40a 1.87b 0.13
signal transducing adaptor molecule

H3007H10(SH3 domain and ITAM motif)2 2.67“ 1.79“ 2.51“ 0.19
H3144A09unknown 0.67“ 0.82“ 0.85“ 0.03
H3007F10 UMP-CMP kinase (UMP-CMPK) 2.92“ 1.76“ 2.86“ 0.27

 

IData represent mean intensity values of mRNA expression for the corresponding
gene in the treatment groups, C(control), S(sulindac) and A(anthocyanins).

Plate ID: sequence number for NIA 15K mouse cDNA library set
3SEM; standard error of mean.

a’b‘cMeans in the same row not sharing a common superscript are signiﬁcantly
different (P<0.05).

114

Table 8. List of genes up—regulated by anthocyanins vs control and sulindac in cecum

 

 

Plate IDI Gene Name2 A/C3 A/S3
H3144FO8 hemochromatosis (HFE) gene 2.86 3.26
ras-related C3 botulinum toxin substrate 1
H3018C09 (rho family, small GTP binding protein Rac 1) 2.64 2.79
H3131F08 Ly6/neurotoxin 1 protein 2.33 3.38
H3004AO9 unknown 1.79 1.61
H3042E10 unknown 1.75 1.60
H3001H08 adrenomedullin precursor 1.71 1.79
H3101B09 unknown 1.61 1.67
H3016E09 unknown 1.59 1.42
H3144C09 tetratricopeptide repeat domain 1.58 1.49
H3119E07 hypothetical 1.55 1.57
H3004B09 mitochondrial ribosomal protein S7 1.55 1.39
H3078D06 unknown 1.49 1.48
H3004F11 unknown 1.47 1.52
H3014HO2 hypothetical 1.45 1.38
H3023DO9 unknown 1.45 1.42
H3045D09 NEUROENDOCRINE DIFFERENTIATION FACTOR 1.45 1.35
H3051H08 initiate factor 5 1.44 1.55
H3061E10 unknown 1.44 1.35
H3104E01 unknown 1.44 1.46
H3146B05 unknown 1.44 1.25
H3035612 unknown 1.44 1.45
H3012B09 nemo-like kinase 1.43 1.33
H3006D05 no hit found 1.41 1.31
H3023C01 hypothetical 1.41 1.39
H3014GO7 astrocytic phosphoprotein 1.40 1.36
H3008G10 bladder cancer associated protein 1.39 1.44
H3008G05 translocase of inner mitochondrial membrane 8 1.39 1.36
H3001A02 secretin 1.37 1.41
H3094G12 unknown 1.35 1.26
H3082F05 unknown 1.34 1.36
H3144GO9 unknown 1.33 1.26
H3047H09 unknown 1.32 1.17
H3033B06 no hit found 1.32 1.35
H3067GO9 unknown 1.32 1.27
H3067GlO unknown 1.29 1.25
H3098E01 no hit found 1.29 1.25
H3052C306 p1astin2 1.29 1.25
H3064F09 chloride channel regulator Icln pseudogene 1.28 1.26
H3147FO6 malate NADP oxidoreductase 1.28 1.23

 

115

 

H3081EO7
H3082A04
H3100F05
H3 047C03
H3134C08
H3098802
H3’06OC06
H3035E03
H3039A12
H3134A09
H3OSSBO7
H3027AO7
H3060D05
H3080E02
H3 087C08
H3022DO9
H3 O37AO4
H3134E02
H3052C08
H3014B04
H3082F09
H31 17AO6
H3065G09
H3064D02
H3157BO9
H3044E02
H3045C06
H3060C12
H3084CO3
H3134H09
H3065E12
H3074E09
H3032A02
H3133HO4
H3055A04
H3093G12
H3126A12
H3057E08
H3027D1 1
H3054F10

Sjogren syndrome antigen A2
unknown

unknown

early onset breast cancer protein 2
unknown

unknown

unknown

hypothetical

unknown

unknown

unknown

translation repressor NATl
beta-defensin 8

unknown

unknown

beta-glucoronidase

no hit found

solitary transfer RNA-Asp
coproporphyrinogen oxidase
unknown

unknown

voltage-dependent calcium channel beta-3 subunit
unknown

unknown

unknown

ribonucleotide reductase M2
unknown

unknown

hippocalcin—like 1

unknown

unknown

unknown

unknown

unknown

unknown

unknown

DNA POLYMERASE EPSILON P12 SUBUNIT
unknown

unknown

unknown

1.28
1.28
1.27
1.27
1.26
1.26
1.26
1.25
1.25
1.25
1.24
1.24
1.23
1.23
1.23
1.22
1.21
1.21
1.21
1.21
1.21
1.20
1.20
1.19
1.19
1.19
1.18
1.18
1.18
1.17
1.16
1.16
1.15
1.15
1.13
1.12
1.11
1.11
1.10
1.08

1.29
1.28
1.13
1.16
1.13
1.22
1.22
1.20
1.26
1.23
1.24
1.16
1.17
1.28
1.22
1.21
1.17
1.12
1.17
1.21
1.23
1.23
1.13
1.24
1.12
1.23
1.13
1.16
1.19
1.15
1.14
1.12
1.20
1.17
1.18
1.15
1.14
1.17
1.12
1.19

rPlate identiﬁcation number: sequence number for NIA 15K mouse cDNA library set

2This list of genes includes those that were signiﬁcantly up-regulated by

anthocyanins compared to control and sulindac (P<0.05). Names written in capital

116

letters are for the respective homolog of the genes. The total number of genes that
were up-regulated by anthocyanins is 79, which includes 44 unknown, 4 no hit found
and 4 hypothetical protein genes.

3 Ratio of intensity score: A/C; anthocyanins versus control, A/S; anthocyanins versus
sulindac. Ratio represents fold change in gene expression in tissue of mice
consuming anthocyanins when compared that of mice consuming control or sulindac.

117

Table 9. List of genes down-regulated by anthocyanins vs control and sulindac in

 

 

cecum

Plate ID1 Gene Name2 C/A3 S/A3
H3119A02 unknown 2.27 1.77
H3146GO9 unknown 2.01 2.35
H3065C02 GTP-binding protein NGB 1.78 1.42
H3119C02 olfactory receptor gene cluster 1.72 1.71
H3003GO4 cell adhesion molecule nectin-3 alpha 1.63 1.55
H3104D01 beta-1,4-galactosyltransferase 1.57 1.45
H3039F01 LEUKOTRIENE B4 lZ-HYDROXYDEHYDROGENASE 1.55 1.51
H3117H05 OXYSTEROL BINDING PROTEIN-RELATED PROTEIN 3 1.55 1.43
H3072C07 unknown 1.45 1.28
H3157E05 zinc transporter 4 1.42 1.35
H3071EO4 unknown 1.37 1.44
H307 5A02 omithine decarboxylase 1.36 1.33
H3134D02 intermediate ﬁlament proteins family 1.36 1.32
H3 090H01 chromatin structural protein 1.34 1.33
H3094A05 hypothetical 1.34 1.28
H3035A05 no hit found 1.34 1.35
H3157D05 unknown 1.32 1.40
H3067F05 unknown 1.32 1.38
H3032FO4 unknown 1.31 1.26
H3082FO4 unknown 1.29 1.30
H3105F07 hypothetical 1.28 1.30
H3090D06 unknown 1.28 1.21
H3142G03 unknown 1.25 1.37
H3131EO6 AS oncogene family 1.25 1.25
H3042A04 unknown 1.24 1.23
H3044D08 unknown 1.24 1.25
H3131E05 unknown 1.24 1.20
H3074C05 unknown 1.24 1.16
H3081D02 Bc1-2-related ovarian killer protein, apoptosis activator 1.23 1.26
H3096C02 no hit found 1.22 1.29
H3033AO9 no hit found 1.21 1.33
H3067G06 no hit found 1.21 1.26
H3033C10unknown 1.19 1.19
H3044D03 unknown 1.19 1.16
H3072C05 unknown 1.19 1.12
H3150H07no hit found 1.18 1.14
H3105G09 aryl hydrocarbon receptor nuclear translocator-like 1.18 1.26
H3033D05 unknown 1.17 1.20

H3037C05 unknown

1.15

1.17

 

118

 

H3080607 unknown 1.15 1.14

H3117H06unknown 1.14 1.19
H3084A06spindlin 1.14 1.18
H3126D02unknown 1.14 1.12
H3136G08nohit found 1.12 1.11

 

IPlate identiﬁcation number: sequence number for NIA 15K mouse cDNA library set
2This list of genes includes those that were signiﬁcantly down-regulated by
anthocyanins compared to control and sulindac (P<0.05). The total number of genes
that were down-regulated by anthocyanins is 44, which includes 24 unknown, 6 no
hit found and 2 hypothetical protein genes.

3Ratio of intensity score: C/A; control versus anthocyanins, S/A; sulindac versus
anthocyanins. Ratio represents fold change in gene expression in tissue of mice
consuming anthocyanins when compared that of mice consuming control or sulindac.

119

Table 10. List of genes up-regulated by anthocyanins vs control and sulindac in

 

 

colon

Plate IDl Gene Name2 A/C3 A/S3
H3029A08 cytochrome b5 1.73 1.36
H3063C01 unknown 1.53 1.54
H3051C03 aldolase C 1.52 1.48
H3009C07 heat shock protein 1.49 1.32
H3071C08 E(y)2 PROTEIN, CG10347 PROTEIN 1.48 1.37
H3069H09 unknown 1.44 1.31
H3008E10 no hit found 1.41 1.50
H3095E02 unknown 1.40 1.22
H3008C06 unknown 1.38 1.34
H3003G03 458 pre rRNA 1.37 1.36
H3075E12 TRANSPORTlN-SR 1.36 1.32
H3125GO9 unknown 1.34 1.25
H3036GO7 epithelial cell transmembrane protein antigen precursor 1.33 1.43

TRANSLOCON-ASSOCIATED PROTEIN,

H3131A08 SIGNAL SEQUENCE RECEPTOR BETA SUBUNIT) 1.32 1.18
H3074A08 TRANSFORMING PROTEIN RHOA (RhoA GTPase) 1.32 1.35
H3051E03 unknown 1.31 1.48
H3042EO8 interleukin 1 receptor-associated kinase(111rak) 1.30 1.35
H3003DO6 inhibitor of growth family member 1 (INGl) 1.29 1.27
H3094GO6 unknown 1.27 1.37
H3108A08BIMP1 (apoptosis regulator) 1.27 1.34
H3136F07 guanine monphosphate synthetase (GMPS) 1.25 1.13
H3153H04 CALCINEURIN B 1.25 1.28
H3058C10 coagulation factor H receptor-like 3 (F2RL3) 1.23 1.34
H3010D11unknown 1.23 1.55
H3113G05unknown 1.21 1.16
H3122A12 S-adenosylmethionine decarboxylase 1.21 1.10
H3032A06butyrophilin-like 2 1.20 1.28
H3041G12 CELL CYCLE PROGRESSION RESTORATION GENE 3 1.15 1.11
H3044FO3 unknown 1.14 1.14
H3050C12 unknown 1.14 1.10
H3008810 ATPase, H+ transporting, lysosomal I (Atp6i) 1.14 1.26
H3008G04 GATA binding protein 2 1.13 1.23
H3006F08 unknown 1.13 1.34

 

TPlate identiﬁcation number: sequence number for NIA 15K mouse cDNA library set

2This list of genes includes those that were signiﬁcantly up-regulated by

anthocyanins compared to control and sulindac (P<0.05). The total number of genes
that were up-regulated by anthocyanins is 33, which includes 10 unknown genes and

1 no hit gene.

120

3Ratio of intensity score: A/C; anthocyanins versus control, A/S; anthocyanins versus
sulindac. Ratio represents fold change in gene expression in tissue of mice
consuming anthocyanins when compared that of mice consuming control or sulindac.

121

Table 11. List of genes down-regulated by anthocyanins vs control and sulindac in
colon

(I

 

 

Plate 11)1 Gene Name2 C/A3 S/A3
H3130F06 SERUM PROTEIN MSE55 1.59 1.43
H3086B08 unknown 1.39 1.24
H3073E04 initiation factor 5 1.36 1.34
H3089F08 unknown 1.33 1.24
H3074H01 unknown 1.31 1.28
H304ODO3 PSD-95/SAP90-associated protein-4 1.30 1.35
H3123D04 unknown 1.29 1.29
H3089D05 hypothetical 1.28 1.44
H3026H05 unknown 1.27 1.36
H3053A06 inositol 1,4,5-trisphosphate 3-kinase B (ITPKB) 1.26 1.19
H3028H05 KLAAO938 protein (KIAAO938) 1.26 1.39
H3046B04 placental lactogen 2 1.25 1,14

PHOSPHOLYSINE PHOSPHOHISTIDINE
H3105D11 INORGANIC PYROPHOSPHATE PHOSPHATASE 1.25 1.18

H3086H05 Deleted in colorectal cancer(Dcc) 1.24 1.34
H3026D05 N-myristoyltransferase 1 1.22 1.21
H3033FO6 neuronal apoptosis inhibitory protein 7 (Naip7) 1.21 1.15
H3017AO7 A kinase (PRKA) anchor protein (yotiao) 9 (AKAP9) 1.21 1.20
H3 O68H04 ribonuclease/angiogenin inhibitor 2 (Rnh2) 1 .21 1 . l 9
H3106BO7 orphan nuclear receptor (NR2E1) 1.20 1.20
H3090HO8 KIAAO676 PROTEIN 1.20 1.19
H3128F 02 hypothetical ' 1.19 1.26
H3045H05 unknown 1.19 1.30
H3068F08 protein tyrosine phosphatase, non-receptor type 21 1.19 1.25
H3096F11 HNK-l sulfotransferase 1.18 1.22
H3078H10 ADP-ribosylation factor 1 1.18 1.27
H3112F03 degenerative sperrnatocyte 1.18 1.21
H3056D05 no hit found 1.15 1.14
H3100D11no hit found 1.14 1.09
H3134CO9 Rhesus blood group protein (Rh) 1.14 1.33
H3068BOZ hypothetical 1.13 1.18
FYVE-FINGER-CONTAINING
H3074F09 RABS EFFECTOR PROTEIN RABENOSYN-S 1.13 1.14
H3128811 putative oncogene (HLC14-06-P) 1.13 1.07
H3045C02 ubiquitin-conj ugating enzyme 1.09 1.1 1

 

IPlate identiﬁcation number: sequence number for NIA 15K mouse cDNA library set
2This list of genes includes those that were signiﬁcantly down-regulated by
anthocyanins compared to control and sulindac (P<0.05). The total number of genes

122

that were down-regulated by anthocyanins is 33, which includes 6 unknown, 2 no hit
and 3 hypothetical protein genes.

2Ratio of intensity score: C/A; control versus anthocyanins, S/A; sulindac versus
anthocyanins. Ratio represents fold change in gene expression in tissue of mice
consuming anthocyanins when compared that of mice consuming control or sulindac.

123

Table 12. List of genes up-regulated by sulindac vs control in small intestine

 

 

Plate IDl Gene Name1r S/C3
H3111B05 no hit found 3.53
H3139D09 ATPase subunit 6 3.36
H3119B08 no hit found 2.43
H3016H08 no hit found 2.15
H3112H02 no hit found 2.06
H3020A01 no hit found 1.95
H3155HO7 unknown 1.85
H3117HO8 actin related protein 2/3 complex, subunit 4 1.77
H3 009D08 unknown 1 .72
H3109B12 putative nuclear protein 1.57
H3 069D02 hypothetical 1 .56
H3117C01 KRUPPEL—RELATED ZINC FINGER PROTEIN 1.55
H3114G09 no hit found 154
H3118A01 no hit found 1.52
H3 05 8EO7 microtubule-associated protein 4 1.42
H3139D12 cytoplasmic dynein intermediate chain 2 1.40
H3 076E07 unknown 1 .39
H3158GO6 erythroid differentiation regulator 1.37
H3085E11 unknown 1.33
H3130A07 unknown 1.32
H3132GO6 hypothetical 1 .28
H3068C02 unknown 1.27
H3032A01 translocase of inner mitochondrial membrane 1.26
H3085C07 zinc ﬁnger protein 1.26
H3117E10 ~ unknown 1.23
H3 150H1 1 hypothetical 1 .22
H3139C06 unknown 1.22
H3 047G07 unknown 1 .22
H3138CO7 unknown 1.21
H3041A07 no hit found 1.19
H3155A07 unknown 1.18
H3 046A02 unknown 1 .1 8
H3040C 12 unknown 1.15
H3001D07 unknown 1 . 12

 

TPlate identiﬁcation number: sequence number for NIA 15K mouse cDNA library set
2The list of genes includes those that were signiﬁcantly up-regulated by sulindac
compared to control (P<0.05). Names written capital letters are for the respective
homolog of the genes. The total number of genes that were up-regulated by sulindac
is 53, which includes 19 co-regulated with anthocyanins (Table 13), 16 unknown, 6
no hit and 3 hypothetical protein genes.

124

3Ratio of intensity score; S/C: sulindac versus control. Ratio represents fold change
in gene expression in tissue of mice consuming sulidac when compared that of mice
consuming control.

125

Table 13. List of genes up-regulated by sulindac and anthocyanins in small intestine

 

 

Plate ID1 Gene Name2 S/C3 A/C3
H3019H02 TAX INTERACTION PROTEIN 1 1.45 1.46
H3023H01 no hit found 1.36 1.38
H3155E09 hypothetical 1.36 1.39
H3102C12 ADP-ribosylation-like factor 6 interacting protein 1.30 1.22
H3125C12 guanine nucleotide binding protein, alpha inhibiting 3 1.29 1.20
H3145A05 hypothetical 1.27 1.29

mas proto-oncogene,insulin-like growth
H3149AO9* factor type 2 receptor (Igf2r), L41ps, and Au76 peusogenes 1.26 1.13

H3149H02 unknown 1.26 1.17
H3066E10 no hit found 1.23 1.25
H306OE03 no hit found 1.23 1.29
H3077H06 LIM domain binding 1 1.21 1.79
H3026G06 tumor rejection antigen gp96 1.20 1.18
H3148H07 hypothetical 1.19 1.35
H3128B04 liver-speciﬁc bHLH-Zip transcription factor 1.19 1.22
H3108A01 interferon-stimulated protein 1.16 1.29
H3063GO7 unknown 1.16 1.40
H3039G11 unknown 1.13 1.13
H3109B07 unknown 1.13 1.10
H3155C06 heat stable antigen 1.12 1.22

 

TPlate identiﬁcation number: sequence number for NIA 15K mouse cDNA library set
2The list of genes includes those that were signiﬁcantly up-regulated by both
sulindac and anthocyanins compared to control (P<0.05).

Names written capital letters are for the respective homolog of the genes. The total
number of genes that were up-regulated by both sulindac and anthocyanins is 19,
which includes 4 unknown, 3 no hit and 3 hypothetical protein genes.

3 Ratio of intensity score; S/C: sulindac versus control, A/C: anthocyanins versus
control. Ratio represents fold change in gene expression in tissue of mice consuming
sulidac or anthocyanins when compared that of mice consuming control.

*The true identity of the H3149A09 gene cannot be identiﬁed at this time and could
be one of the four.

126

Table 14. List of genes down-regulated by sulindac vs control in small intestine

 

 

Plate ID1 Gene Name2 C/S3
H3071D01 no hit found 1.68
H3086A12 unknown 1.54
H3154Bl 1 adipocyte-speciﬁc protein 5 1 .49
H3 146FO3 hypothetical 1 .46
H3123H07 novel cell death-regulatory protein GRIM19 1.44
H3107D06 unknown 1 .41
H3065F02 unknown 1 .40
H3033F 02 no hit found 1.38
H3134E01 hemoglobin, beta adult major chain 1.36
H3122H02 unknown 1 .36
H3122F10 putative THl protein 1.36
H3 142F09 hypothetical 1 .34
H3087F06 dynein 1.34
H3 108F07 alpha-L-iduronidase 1 .33
H3049B04 no hit found 1.32
H3136F06 unknown 1.32
H3132G02 hypothetical 1 .3 1
H3096H01 no hit found 1.30
H3138B03 unknown 1.30
H3156C08 metaxin 1.30
H3033D08 unknown 1.30
H3108D06 unknown 1 .29
H3 O2OF09 hypothetical 1 .29
H3045C02 ubiquitin-conjugating enzyme 1.28
H3 O60C 1 2 unknown 1 .28
H3025G06 hypothetical 1 .27
H3126H02 no hit found 1.27
H3045HO4 nonmuscle heavy chain myosin 1.27
H3060D01 CCR4-NOT transcription complex, subunit 7 1.27
H3065B01 hypothetical 1 .27
H3078G08 unknown 1.26
H3035D04 GM2 ganglioside activator protein 1.25
H3042BO6 unknown 1 .25
H3042F 12 serine (or cysteine) proteinase inhibitor 1.25
H3033A02 no hit found 1.25
H3108F08 programmed cell death 10 1.24
H3081FO3 hypothetical 1 .24
H3084B06 F -box protein 1.23
H3 O74C07 unknown 1.22
H3 140F02 unknown 1 .22

 

127

 

H3072C 12 unknown 1.21
H3015E11 PDZ and carboxyl terminal LIM domain 1 1.21
H3099H06 unknown 1 .20
H3 O44DO4 unknown 1 .20
H31 12BO6 unknown 1.20
H3020H08 unknown 1 .20
H3065B11 unknown 1.20
H3057D02 RhoGDI-l 1.19
H3016F02 unknown 1. 18
H3047G01 unknown 1.18
H3101F01 hypothetical 1.18
H3104D02 unknown 1 . 17
H3 06OF 1 2 hypothetical 1 .16
H3122A08 unknown 1 . 16
H3 096FO3 unknown 1 . 16
H3109F02 DREV protein 1.15
H3113F11 unknown 1.14
H3100G04 no hit found 1.14
H3 064BO6 no hit found 1 . l 3
H3077A12 hypothetical 1 . 1 1
H3104E04 no hit found 1 .07

 

TPlate identiﬁcation number: sequence number for NIA 15K mouse cDNA library set
2The list of genes includes those that were signiﬁcantly down-regulated by sulindac
compared to control (P<0.05). The total number of genes that were down-regulated
by sulindac is 108, which includes 47 genes co-regulated with anthocyanins (Table
15), 25 unknown, 9 no hit and 10 hypothetical protein genes.

3Ratio of intensity score; C/S: sulindac versus control. Ratio represents fold change
in gene expression in tissue of mice consuming sulidac when compared that of mice
consuming control.

128

Table 15. List of genes down-regulated by sulindac and anthocyanins in small

 

 

intestine

Plate 11)1 Gene Name2 as3 C/A3
H3104D07 fatty acid binding protein 3 2.34 1.78
H31 13812 no hit found 1.97 3.30
H3156A10unknown 1.84 2.16
2038 glutathione S transferase 1.67 1.69
H3007A01 unknown 1.66 1.56
H3091A01 hypothetical 1.65 1.39
H3105B07 hypothetical 1.52 1.38
H3124H12 no hit found 1.49 3.02
H3010H09 unknown 1.43 1.49
H3018G10unknown 1.42 1.39
H3004B03 no hit found 1.37 1.55
H3064F03 hypothetical 1.37 1.28
H3114G02 unknown 1.37 1.33
H3039E10 serine/threonine kinase 1.36 1.26
H3078F01 unknown 1.36 1.33

H3051E02 unknown 1.35 1.37
H3100F10 unknown 1.33 1.27
H3067F05 unknown 1.32 1.44
H3127G08 hypothetical 1.31 1.36
H3098E02 no hit found 1.30 1.26
H310OE10 hypothetical 1.30 1.36
H3143A07 unknown 1.28 1.46
H3097E12 unknown 1.28 1.20
H3084FO7 unknown 1.28 1.39
H3072E11 unknown 1.27 1.25

H3090D11 unknown 1.27 1.26

protein kinase, interferon inducible double

H3024FO8 stranded RNA dependent activator 1.27 1.28

H3117A06 voltage-dependent calcium channel beta-3 subunit 1.26 1.17
H3142G12 hypothetical 1.26 1.13

H3057C12 hypothetical 1.26 1.35

H3088B01 hypothetical 1.26 1.26
H3008CO3 cytotoxic granule-associated RNA binding protein 1.25 1.28
H3047D07 nuclear receptor subfamily 5, group A, member 2 1.25 1.24
H3033D06 no hit found 1.24 1.31

H3142CO6 hypothetical 1.23 1.13

H309OFO7 unknown 1.23 1.14
H3100H05 unknown 1.22 1.30
H3119B07 unknown 1.21 1.23

 

129

 

H3126F07 unknown 1.21 1.19

H3052A08 unknown 1.19 1.26
H3065CO6 unknown 1.15 1.12
H3035G10unknown 1.15 1.19
H3132G08unknown 1.14 1.12
H3067B05 unknown 1.14 1.21
H3127G05 palmitoyl-protein thioesterase 1.13 1.17
H3032D10no hit found 1.12 1.11
H3134H05unknown 1.11 1.19

 

1Plate identiﬁcation number: sequence number for NIA 15K mouse cDNA library set.
Plate ID 2038 is not in the NIA 15K mouse cDNA library and added by Dr. John
Lapres’ lab in Michigan State University.

2This list of genes includes those that were signiﬁcantly down-regulated by both
sulindac and anthocyanins compared to control (P<0.05). The total number of genes
that were down-regulated by both sulindac and anthocyanins is 47, which includes

24 unknown, 6 no hit and 9 hypothetical protein genes.

3 Ratio of intensity score; C/S: sulindac versus control, C/A: anthocyanins versus
control. Ratio represents fold change in gene expression in tissue of mice consuming
sulidac or anthocyanins when compared that of mice consuming control.

130

DISCUSSION

In this study, we conﬁrmed our previous results in which we found that
anthocyanins and cyanidin suppressed tumor development in cecum of ApcMin mice.
In the previous experiments, the compounds (anthocyanins, cyanidin and sulindac)
were administered in drinking water to a relatively small number (n=10) of mice per
treatment group for 10 weeks. In this study, the extent to which dietary anthocyanins
and sulindac inhibited intestinal tumor development was determined in a larger
number of ApcMin mice (n=25) when fed compounds in the diet for 8 weeks. This
approach increased the stability of the administered anthocyanins, increased power
of the experiment by increasing the number of replicates, and reduced the morbidity
of mice by shortening the promotion period.

In this experiment, dietary anthocyanins signiﬁcantly reduced adenoma
number in all three intestinal sections. Cecal adenoma burden (volume) also was
reduced and small intestinal adenoma mean diameter was intermediate in mice
consuming anthocyanins. Sulindac reduced adenoma number and mean diameter in
small intestine, but increased cecal adenoma number when compared to controls.

Koide et a1. (1996) ﬁrst reported that feeding red rice prolonged survival of
Balb/C mice inoculated i.p. with Meth/A lymphoma cells and that hydrolyzed
anthocyanins decreased the grth of HCT-15 cancer cells. Later, they found that
sugar-bonded bioﬂavonoids extracted from red soybeans and red beans prolonged
survival of Balb/C mice bearing tumor Meth/A cells and that anthocyanin fractions

extracted from red soybeans inhibited the growth of HCT-l 5 cells (Koide et al.,

1997). Pool-Zobol et a1. (1999) have shown that anthocyanins reduced H202-

131

induced DNA strand breaks but did not prevent endogenous generation of oxidized
DNA bases in HT 29 clone 19A human colon cancer cells. Recently, it has been
reported that purple corn anthocyanins suppressed the promotion of colon tumors
caused by 2-amino—1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) in rats initiated
with dimethyl hydrazine. However, these researchers did not test the potential of
purple corn color anthocyanins to directly suppress tumor development (Hagiwara et
a1,2001)

The second objective of this study was to determine the inﬂuence of dietary
anthocyanins and sulindac on expression and cellular localization of B-catenin in the
intestine of ApcMin mice. Contrary to our expectations, immunostaning of B-catenin
in intestinal tissues demonstrated that sulindac generally did not inﬂuence the
expression of B-catenin protein, whereas anthocyanins increased B-catenin
expression in some intestinal regions compared to control or sulindac. Intracellular
distribution of B-catenin was primarily in the membrane in all mice. These results
are contradictory to the ﬁndings of other investigators. Mutanen et a1. (2000)

M' -
‘" mice and decreased

reported that rye bran diet prevented tumor formation in Ap‘c
the cytosolic B-catenin level of intestinal epithelial cells. Cellular distribution of B-
catenin also was modiﬁed by sphingolipid consumption (Schmelz et al., 2001).
Sphingolipid feeding reduced adenoma numbers in all regions of the intestine in
ApcMin mice and sphingosine reduced cytosolic and nuclear B—catenin expression
(Schmelz et al, 2001).

Gene expression results from the cDNA microarray analysis showed that

expression of the mRNA transcripts for B-catenin and APC were not altered by

132

treatment, suggesting that transcriptional regulation did not cause the increase in B-
catenin protein expression in the intestine of ApcMin mice consuming anthocyanins.
Other possible explanations for the increase in B-catenin expression by anthocyanins
would be increased stability of B-catenin mRNA, decreased degradation of B-catenin,
orincreased translation. Additional experiments are necessary to address these
possibilities.

APC mutation is permissive for aberrant accumulation of cytoplasmic [3-
catenin, which then translocates to the nucleus and binds ch/Lef transcription
factors and causes increased transcriptional activation of genes in the Wnt/B-catenin
signaling pathway. Known target genes of this signaling include PPARE, C-Myc,
Cyclin D1 and MMP7. These genes have been found to be overexpressed in human
colonic adenocarcinomas and in many human colonic tumor cell lines (He et al.
1999; Zhang et al., 1997; Crawford et al., 1999; Tetsu and McCormick, 2000;
Newell et al., 1994). Our interest in testing the potential of tart cherry anthocyanins
and cyanidin to inhibit tumor development in ApcMin mice stemmed from the
observation that these compounds inhibit the activities of COX enzymes (Wang et al.,
1999). McEntee et a1. (1999) reported that NSAID-induced regression of intestinal
adenomas is associated with reduced levels of B-catenin. NSAIDs also have been
shown to inhibit PPARS activity in HCT116 and SW480 cell lines in a dose-
dependent manner (He et al., 1999). Therefore, we hypothesized that anthocyanins
reduce tumor development in ApcMin mice in a manner similar to sulindac by

suppressing Wnt/B-catenin signaling.

133

In contrast to our initial hypothesis, our results indicate that neither
anthocyanins nor sulindac inﬂuenced the expression of mRNA of B-catenin and most
of the predicted target genes of the Wnt/B-catenin signaling. Anthocyanin treatment
modestly reduced the expression of cyclin D1 compared to sulindac in cecum but not
in the other intestinal tissues. The expression of the other predicted target genes —

MMP7, PPAR5 and C-Myc — were not altered by anthocyanins or sulindac in all
intestinal sections of ApcMin mice. These results suggest that the mechanisms
responsible for reductions in intestinal tumor development by sulindac or
anthocyanins likely are not associated with inhibition of Wnt/B-catenin signaling.
We consider it likely that tumor suppression associated with these compounds is via

mechanisms other than normalized Wnt/B-catenin signaling.

Although most hypotheses for explaining the anticarcinogenic effects of
NSAIDs have focused on their common property to inhibit COX enzymes, the
mechanism of the anticarcinogenic effect of these drugs has not been fully
established. Little information on potential mechanisms whereby anthocyanins may
suppress intestinal tumor development is available. Anthocyanins have been shown
to inhibit the activity of epithermal growth factor receptor (EGFR) kinase. Meiers et
al. (2001) demonstrated that two anthocyanins, cyanidin and delphinidin, are potent
inhibitors of the EGFR kinase. Administration of inhibitors to the EGF R kinase also
has been shown to reduce intestinal adenoma development in ApcMin mice (Torrance
et al., 2000).

A total of 1,492 genes were inﬂuenced by dietary treatment in any of the

intestinal sections — small intestine, ceum or colon. A relatively small subset of

134

these gene effects overlapped across any two intestinal regions, and no single gene
was inﬂuenced by treatment in all regions. To begin narrowing this list of gene
changes to a shorter list of genes that merit future investigation, we have used two
strategies. First, we identiﬁed genes inﬂuenced by treatment in more than one
intestinal region. Second, we focused on gene changes that corresponded to the
observed tumor numbers. For example, we are particularly interested in genes up- or
down-regulated speciﬁcally by anthocyanins relative to sulindac and control
treatments in the cecum or colon.

Although gene changes are ranked in the tables by fold change, it is
important to consider that an arbitrary selection of a fold change value as signiﬁcant
(i.e. 2-fold) is probably not appropriate for two reasons. First, all. gene changes
reported in this experiment are based on statistical analysis of variance. Unlike
many microarray experiments, in this study we have used true replication and valid
statistical approaches to identify genes signiﬁcantly inﬂuenced by treatment.
Secondly, it is well established that the intestinal epithelium is a mixed cell
population comprised by proliferating cells, terminally differentiated cells, and cells
undergoing programmed cell death. Any changes in gene expression that occurred
speciﬁcally in one of these cell types would be diluted by expression in the other
epithelial cell types.

We identiﬁed several clusters of differentially expressed genes in cecum,
colon and small intestine of ApcMin mice using cDNA mcroarray analysis. Tumor

inhibition by anthocyanins was most prominent in cecum. For this reason, we

focused on genes changed speciﬁcally by anthocyanins relative to control and

135

sulindac in the cecum (Table 8). Although hemochromatosis gene and
Ly6/Neurotoxin protein gene were inﬂuenced to the greatest extent by anthocyanins,
there is very little information available on their functions or potential role in tumor
development. Ras-related botulinum C3 substrate 1 also was up-regulated by
anthocyanins. Clostridium botulinum C3 represents a major family of the bacterial
toxins that transfer the ADP-ribose moiety of NAD to speciﬁc amino acids in
acceptor proteins to modify key biological activities in eukaryotic cells, including
protein synthesis, differentiation, transformation, and intracellular signaling (Han et
al., 2001). C3-like transferases have been shown to ADP-ribosylate the small
guanosine triphosphate (GTP)- binding protein family, Rho, resulting in functional
inactivation of Rho (Wilde et al., 2001).

Among the many genes altered by anthocyanins, we are particularly
interested in those genes that potentially are involved in cellular differentiation
(neuroendocrine differentiation factor), DNA synthesis and repair (ribonucleotide
reductase M2), and regulation of apoptosis (translocase of inner mitochondrial
‘ membrane 8). Ribonucleotide reductases catalyze the production of the
' deoxyribonucleotides required for DNA replication and repair. Yamaguchi et al.
(2001) demonstrated that a newly identiﬁed ribonucleotide reductase, p53R2, is
directly regulated by p53 for supplying nucleotides to repair damaged DNA.
Translocases of the inner mitochondrial membrane mediate transport of
mitochondrial proteins from the cytosol into the mitochondria in conjunction with
translocases of the outer mitochondrial membrane (Halestrap et al., 2000).

Translocases of inner mitochondrial membrane also inﬂuence mitochondrial

136

membrane permeability, which is a critical step of apoptotic pathways (Crompton,
2000). Brenner et a1. (2000) demonstrated that adenine nucleotide translocator
(ANT), the most abundant inner mitochondrial membrane protein, interacted with
Bax, a proapoptotic member of the Bcl-2 family to yield an efﬁcient composite
channel. This opening of the mitochondrial permeability transition pore is thought to
be involved in apoptosis.

Altered intracellular signaling by increased expression of initiation factor 5,
serine/threonine kinases (tetratricopeptide repeat domain, nemo-like kinase) and
suppressor of mitogen-activated protein (MAP) kinase (astrocytic phophoprotein)
also merit further investigation. Initiation factor 5 plays a role in translation
initiation following recognition of the start codon (Das and Maitra, 2000). It has
been shown that initiation factor 5 is required for hydrolysis of GTPase and acts as a
GTPase-activator protein (Paulin et al., 2001). Nemo-like kinase is a serine-
threonine protein kinase that connects MAP kinase and Wnt/B-catenin signaling
pathways in Caenorhabditis elegans, Xenopus laevis and Drosophila melanogaster
(Ishitani etal., 1999; Rocheleau et al, 1999; Verheyen et al., 2001). Ishitani et al.
(1999) demonstrated that activation of TAKl-MAP kinase (a kinase activated by V
transforming grth factor-beta) stimulated nemo-like kinase activity and
downregulated Wnt signaling by inhibiting the interaction of the B-catenin-TCF
complex with DNA. Astrocytic phosphoproteins are targets of numerous
extracellular signals received by astrocytes. A major astrocytic phosphoprotein,
PEA-15, has been identiﬁed as a protein kinase C substrate (Estelles et al., 1996).

Formstecher et al. (2001) report that PEA-15 blocked epithelial receptor kinase

137

(ERK)-dependent transcription and proliferation by binding ERK5 and preventing
their localization in the nucleus. Genetic deletion of PEA-15 also resulted in
increased ERK nuclear localization with consequent increased cFos transcription and
cell proliferation. No functional information is available on bladder cancer
associated protein.

Genes that were down-regulated in their expression by anthocyanins relative
to control and sulindac include genes (Table 9) coding for G-protein coupled
receptors (GTP-binding protein NGB, olfactory receptor cluster), cell adhesion
proteins (nectin-3 alpha, beta-1,4-ga1actosyltransferase), enzymes involved in cell
proliferation (omithine decarboxylase), an oncogenic protein (AS oncogene family),
an apoptosis-related protein (Bcl-2-related ovarian killer protein), chromatin
structural proteins (intermediate ﬁlament proteins family, chromatin structural
protein), and a nucloesome assembly protein (spindlin). G-protein coupled receptors
are cell surface receptors that are coupled to heterotrimeric G-proteins (GTP-bindin g
proteins). They have binding sites in the extracellular domain for glycoprotein
hormones. Down-regulated expression of these G-protein coupled receptors may
indicate a reduction in intracellular signaling necessary for tumor growth. Beta-1,4-
galactosyltranferase l is the key enzyme transferring galactose to the terminal N-
acetylglucosamine (GlcNAc) of glycoconjugates (Furukawa and Sato, 1999). It also
serves as a cell adhesion molecule by recognizing and binding to terminal GlcNAc of
glycoconjugates on the adjacent cell surface and matrix. Reduced expression of
omithine decarboxylase is also an important ﬁnding regarding the potential cellular

actions of anthocyanins. Omithine decarboxylase is a rate-limiting enzyme for

138

synthesis of polyamines that ultimately regulate DNA synthesis (Williams-Adams et
al., 1972). It is well known that intracellular polyamine concentrations and omithine
decarboxylase activity are both increased in colorectal cancer tissue and in
premalignant polyps (Wallace and Caslake, 2001).

In colon, anthocyanin feeding reduced adenoma numbers compared to both
control and sulindac treated mice. Anthocyanins also tended to reduce colonic
adenoma volume compared to controls. Anthocyanins speciﬁcally increased the
expression (Table 10) of genes coding for proteins involved in signal transduction
(heat shock protein, transforming protein Rho A, interleukin 1 receptor-associated
kinase, calcineurin B), transcription (GATA binding protein 2), mRN A splicing
(transportin-SR), translation (458 pre rRNA), apoptosis control (BIMPl, inhibitor of
growth family 1) and cell cycle control (cell cycle progression restoration 3). Heat
shock proteins are considered to be molecular chaperones which play a universal role
in maintaining cellular homeostasis (Liu and Steinacker, 2001). Rho genes encoding
small GTP-binding proteins have been implicated in actin organization and the
interaction of the cytoskeleton with intracellular membranes (Lackie and Dow, 1999).
It has been reported that signaling and transforming properties of G protein coupled
receptors cause aberrant grth of cells via activation of Rho family small GTPases
(Whitehead etal., 2001). Interleukin 1 receptor-associated kinase (IRAK-l) is
essential for pro-inﬂammatory cytokine interleukin l-beta signaling and interleukin 1
stimulation of nuclear factor KB (NF ~1cB), inducing an increase in the expression of
many genes with roles in immunity and inﬂammation (Baldwin, 1996; Cooke et al.,

2001). Calcineurin is a calmodulin-stimulated protein phosphatase and has a pivotal

139

role in the calcium/cahnodulin—calcineurin dependent pathway (Aramburu et al.,
2000). Calcineurin, or PP2B, is a highly conserved, Ca2+ calmodulin-dependent
phosphoserine/phosphothreonine-speciﬁc phosphatase and regulates many processes
including NMDA signaling, Na+/K+ ATPase function, cardiac development and
hypertrophy, T-cell activation, and angiogenesis (Cyert, 2001). GATA-binding
proteins are a subfamily of zinc ﬁnger transcription factors with six members
(GATA-L6) that interact with the GATA deoxyribonucleic acid (DNA) sequence.
GATA sequences are found in the regulatory regions of many genes including those
encoding T-helper 2-like cytokines, receptors, adhesion molecules and enzymes
(Caramori et al., 2001). The GATA family of transcription factors also regulates
tissue-speciﬁc patterns of gene expression during development by functioning as
transcriptional effectors of the small GTPase RhoA (Charron et al., 2001). 45 S
rRNA is structurally homologous to mammalian 7 S RNA and plays an essential role
in targeting proteins containing a particular signal peptide to the secretory apparatus
by forming signal recognition-like particle. It also binds independently to protein
elongation factor G (EF-G), and functions in the translation process (N akamura et al.,
2001). Transportin—SR proteins are serine/arginine-rich (SR) protein import
receptors. SR proteins are a group of abundant arginine/serine (RS)—rich proteins
and essential pre-mRNA splicing factors that are localized in the nucleus (Kataoka et
al., 1999). The RS domain of these proteins serves as a nuclear localization signal
and transportin-SR binds speciﬁcally and directly to the RS domains of several SR
proteins (Lai et al., 2001). Inhibitor of growth family member 1(ING1) is a novel

candidate tumor suppressor gene which is involved in the regulation of apoptosis (Ito

140

 

et al., 2002). INGl (p33) is thought to interact with the p53 signaling pathway and
regulate cellular grth (Ito et al., 2002). McAllister-Lucas et al. (2001)
characterized BHVIPI, a new signaling protein that binds BcllO and activates NF-KB.
They found that BIMPl-mediated NF-kB activation required BcllO and IKB kinases.
The involvement of BIMPl in the regulation of apoptosis is not clear. Interestingly,
expression of S-adenosylmethionine decarboxylase (SAMDC) was increased by
anthocyanins. SAMDC is involved in polyamine synthesis with omithine
decarboxylase, implicating potential effects on cell proliferation in regulating colonic
tumor grth (Milovica et al., 2001). The detailed functional information is not
available for cell cycle progression restoration 3.

Anthocyanins caused down-regulation (Table 11) of genes coding for
proteins involved in signal transduction (inositol 1,4,5-triphosphate 3-kinase B,
phospholysine phosphohistidine inorganic phyrophosphate phosphatase, A kinase
anchor protein (yotiao) 9, orphan nuclear receptor (NR2E1), protein tyrosine
phosphatase non-receptor type 21, ADP-ribosylation factor [GTP-binding proteins]),
tumor suppression (deleted in colon cancer (DCC)), translational and post-
translational modiﬁcation (N-myristoyltransferase 1), apoptosis (neuronal apoptosis
inhibitory protein 7), cell proliferation control (ribonuclease/angiogenin inhibitor 2)
and a putative oncogene. Inositol 1,4,5-tn'phosphate 3-kinase mediates cellular
signaling by phosphorylating the Cay-mobilizing second messenger inositol 1,4,5-
triphosphate to yield the putative second messenger inositol 1,3,4,5-tetrakis
phosphate (1P4). A kinase anchor protein (yotiao), a N-methyl-D-aspartate WMDA)

receptor-associated protein, has been found to bind CAMP-dependent protein kinase

141

II (F eliciello et al., 1999). Yotiao has also been found to bind the type 1 protein
phosphatase (PPl) and attaches PPl and CAMP-dependent protein kinase to NMDA
receptors (Westphal et al., 1999). The downstream effect of the NMDA receptors
related to yotiao protein is not clear. We found that anthocyanins decreased
expression of the DCC gene, an unexpected observation since inactivation of DCC
on chromosome 18 is known to be associated with colonic tumorigenesis and
metastasis in humans (V ogelstein and Kinzler, 1993). Down-regulation of DCC by
anthocyanins merits further study. N-myristoyltransferase catalyzes the co-
translational and/or post-translational transfer of myristate to the amino terminal
glycine residue of a number of important proteins. Maurer-Stroh et al. (2002)
reported that N-myristoyltransferase catalyzes myristylation of kinases, phosphatases,
proteasomal regulatory subunit 4, kinase interacting proteins and homologues of
mitochondrial translocase.

We focused on the gene changes caused by sulindac and, to a lesser extent,
anthocyanins in small intestine because these compounds reduced small intestinal
tumor development. Sulindac alone up-regulated genes for ATPase 6 subunit,
nuclear protein (putative nuclear protein), DNA-binding proteins (kruppel-related
zinc ﬁnger protein, zinc ﬁnger protein), microtubule-asscoiated proteins
(microtubule-associated protein 4, cytoplasmic dyenin intermediate chain 2), and
regulation of apoptosis (translocase of inner mitochondrial membrane) (Table 12).
ATPase is an enzyme which yields ADP and inorganic phosphate by releasing the
terminal phosphate from ATP. Sequence variants of the ATPase 6 gene are thought

to play a role in mitochondrial DNA maintenance and integrity in yeast. Maximo et

142

al. (2001) reported that germline polymorphisms of the ATPase 6 gene are associated
with mitochondrial DNA common deletion mutation, which increases thyroid
tumorigenesis. Cytoplasmic dynein is a microtubule associated protein with ATPase
activity involved in microtubule associated movement.

The small intestinal genes up-regulated by both sulindac and anthocyanins
(Table 13) include ones for adhesion (tax interaction protein 1), GTP—binding protein
(ADP-ribosylation-like factor 6 interacting protein), GTP-binding inhibitor (GTP-
binding protein alpha inhibiting 3), immune response (tumor rejection antigen gp96),
and developmental decision (LIM domain binding). Tax protein has been identiﬁed
as an axonal surface glycoprotein involved in cell adhesion. ADP-ribosylation factor
is a ubiquitous GTP—binding protein. ADP-ribosylation is a form of post-translational
modiﬁcation of protein structure involving the transfer to protein of the ADP-ribosyl
moiety of NAD (nicotinamide adenine dinucleotide). ADP-ribosylation is believed
to play a part in normal cellular regulation. Tumor rejection antigen gp96 is a
resident endoplasmic reticulum chaperone protein that directs peptides into the
immune response pathway and assists in protein folding (Linderoth et al., 2000).
Although the mechanism remains unclear, tumor rejection anti gen gp96 can activate
the immune system to slow or stop the progression of tumors by escorting tumor-
derived immunogenic peptides into the endogenous antigen presentation pathway of
antigen presenting cells (Reed and Nicchitta, 2000).

Among a total of 161 genes altered by sulindac in their small intestinal
expression, there were more down-regulated genes (108) (Table 14 and 15) than up-

regulated genes (53). Cell death regulation genes such as novel cell death-regulatory

143

protein GRIM19 and programmed cell death 10 were found to be down-regulated.
Their exact function in cell death regulation has not been identiﬁed. Protein
ubiquination and degradation may have been altered by down-regulation of
ubiquitin-conjugating enzyme and F-box protein. Ubiquination has been identiﬁed
as an important mechanism controling a diversity of cellular regulatory processes,
including ubiquination-mediated protein turnover, protein targeting, and cell fate
decisions. Clark et al. (1989) identiﬁed a developmentally essential gene, chB
encoding ubiquiting conjugating enzyme, involved in substrate recognition in protein
ubiquination. F-box protein is an adapter protein that is involved in associating
proteins with the ubiquitin-driven proteolytic system. The F -box is a motif found in
a wide variety of proteins including many cell cycle regulatory proteins, though
various F-box proteins probably also play a part in regulation of transcription, signal
transduction and development. Spiegelrnan et a1. (2002) reported regulation of F-

box proteins by the Wnt/B-catenin signaling pathway. Activation of the Wnt/B-

catenin signal transduction pathway elevates F-box protein BTrCP levels, whereas it
reduces expression and activity of another closely related F -box protein, HOS. DNA
helicase hFDH has been identiﬁed as an F-box protein (Kim et al., 2002). Down-
regulation of DNA helicase hF DH inﬂuences regulation of DNA metabolism.
Transcription and intracellular signal transduction are also inﬂuenced by altering
chemokine receptor (CCR-NOT transcription complex), serine proteinase inhibitor,
and PDZ and carboxyl terminal LIM domain 1.

Genes that were down-regulated both by anthocyanins and sulindac (Table

15) included fatty acid biding protein (F ABP) 3 and glutathione S transferase. FABP

144

is thought play a role in enterocyte fatty acid trafﬁcking by binding long chain fatty
acids and facilitating exchange of fatty acids between intracellular membranes.
Liver FABP is highly expressed in the both the liver and small intestine and its
expression is likely up-regulated by PPAR5 activity in small intestine (Poirier et al.,
2001). Poirier et al., (2001) demonstrated that the PPARS/PPAROL agonist GW2433
up-regulated the gene encoding liver FABP in the intestine of PPARa-null mice.
Our results would indicate that anthocyanins and sulindac may alter the expression
of genes that are downstream of PPARS signaling but not PPARB expression itself.
Glutathione S transferase (GST) is a major phase II detoxiﬁcation enzyme that
catalyzes the conjugation of electrophilic substrates to glutathione. Although a
protective role of the glutathione biotransforrnation in colorectal carcinogenesis has
been suggested, studies that compared GST polymorphisms in relation to colon
cancer risk are not conclusive with respect to an increased or decreased risk
associated with a particular genotype (Strange and Fryer, 1999; Wormhoudt et al.,
1999; Grubben et al., 2001).

We were also interested in studying genes that were similarly inﬂuenced by
treatment in different intestinal sections to enable us to identify patterns of gene
changes that may be globally related to tumor development in the intestine. It was
somewhat surprising that relatively few genes were similarly inﬂuenced by treatment
in the different intestinal sections. However, the proﬁles of adenoma development
and effects of treatment on tumor development were different throughout the
intestine. No single gene was inﬂuenced by treatment in all three intestinal sections

(small intestine, colon and cecum). The expression patterns of genes inﬂuenced by

145

treatment in more than one region of the intestine oﬁen were not consistent in the
various regions. Our observations suggest that the mechanisms controlling tumor
promotion in ApcMin mice likely differ in the various regions of the intestine. This
hypothesis is supported by the distinct morphology of tumors in the small intestine
versus the cecum and colon.

In summary, we have demonstrated that dietary anthocyanins inhibit tumor
Min

development in small intestine, colon and cecum of Apc mice. Sulindac primarily

inhibited small intestinal tumor development of ApcMm mice. Tumor inhibition by
anthocyanins and sulindac does not appear to be associated with reductions in
cytoplasmic B-catenin expression or reduced activity of the Wnt/B-catenin signaling
pathway. cDNA microarray results showed that most of the predicted target genes of
Wnt/B-catenin signaling were not inﬂuenced by anthocyanins or sulindac, suggesting
the presence of downstream compensatory mechanisms for tumor inhibition by these
compounds. We identiﬁed several genes that were differentially expressed in mice
exposed to the three dietary treatments in intestinal sections. Genes inﬂuenced by
anthocyanins and sulindac are involved in a wide variety of cellular and molecular
processes such as DNA synthesis and repair, transcription, mRNA processing,
protein interactions and turnover, cell proliferation and differentiation, cell cycle
control, and programmed cell death (apoptosis). The patterns of gene expression
changes suggest that alterations of cellular and molecular processes by anthocyanins
and sulindac likely are different in ApcMin mice. Additional conﬁrmatory analysis of

these gene expression results will be necessary to further establish potential

mechanisms whereby anthocyanins and sulindac inﬂuence tumor development.

146

CHAPTER VI

OVERALL SUMMARY AND CONCLUSIONS

147

Anthocyanins, a member of the ﬂavonoid family of phytochemicals, are
widely distributed in plants including beans, fruits, and vegetables (Harbome and
Grayer, 1988). We hypothesized that tart cherry anthocyanins would suppress
intestinal tumorigenesis because they were previously found to inhibit the activities
of COX enzymes (Wang et al., 1999b). Compounds that inhibit COX 2 activity have
been demonstrated to protect against colon cancer development in animal models
and reduce the risk of colon cancer in humans (Giovannucci et al., 1995; Labayle et
al., 1991). However, little research has been conducted to study the potential of
anthocyanins to inhibit colon cancer development.

The ﬁrst objective of this research was to determine the effects of tart cherry
anthocyanins on the development of intestinal adenomas of ApcMin mice. ApcMin
mice are a mutant mouse lineage predisposed to multiple intestinal neoplasia due to a
mutation in the murine homolog of the APC gene (Su et al., 1992). The primary
phenotype of ApcMin mice is the development of multiple adenomas throughout the
small intestine, cecum and colon. Effective use of ApcMin mice in this research
required an assessment of their normal process of tumor development. This was
necessary to establish appropriate tumor endpoints for subsequent studies. There fore,
we ﬁrst examined the time of development and morphology of intestinal adenomas
in ApcMin mice to obtain baseline data on tumor development.

The ﬁrst experiment was conducted to determine the inﬂuence of age and sex
on the numbers and sizes of small intestinal, cecal and colonic tumors in ApcMm mice.

Min

Apc mice were weaned at 28 days of age and fed Teklad 8640 rodent diet

148

throughout the experiment. Mice were sacriﬁced at 30, 45, 60, 75, 90 and 120 days
of age. As expected, the numbers and sizes of adenomas in the small intestine
increased with age. The numbers of adenomas in the cecum and colon did not
increase signiﬁcantly with age, but the volume of colonic tumors increased
significantly after 75 days of age. Sex did not inﬂuence the numbers of small
intestinal, cecal or colonic tumors. Male mice had greater body weights, colon
lengths and colon tumor volumes than female mice. Tumors were already present in
30 day-old‘mice. All small intestinal tumors were sessile (ﬂat), whereas most cecal
and colonic tumors were polypoid. These results provided important information
that intestinal tumors develop early in life and progress rapidly in size as ApcMm
mice exceed 90 days of age.

We next examined the effects of tart cherry diets, anthocyanins, and their

Min

aglycone, cyanidin, on multiplicity and size of intestinal adenomas in Apc mice.

Mi“ mice (n=10/treatment) were exposed

During a 10 week experimental period, Apc
to the following treatments; 1) control diet (modiﬁed AIN 93G diet containing 220
g/kg protein, 150 g/kg soybean oil and 50 g/kg cellulose), 2) control diet + 800 mg/L
anthocyanins in the drinking water, 3) control diet + 200 mg/L cyanidin in the
drinking water, 4) control diet + 200 mg/L sulindac in the drinking water, and 5)
modiﬁed control diet containing 200 g/kg freeze-dried pitted tart cherries. ApcMin
mice consuming anthocyanins, cyanidin, and tart cherries had signiﬁcantly fewer

numbers and smaller size of tumors in the cecum. Mice consuming sulindac had

smaller small intestinal tumors, whereas mice consuming tart cherries had bigger

149

 

 

small intestinal tumors in average diameter. The total numbers and burden of
colonic tumors were not signiﬁcantly inﬂuenced by treatment.

We also determined the potential of tart cherry anthocyanins, their aglycone,
cyanidin, and their degradation products to directly inhibit growth of the human
colon cancer cell lines HT 29 and HCT 116. Both anthocyanins and cyanidin
reduced the growth of human colon cancer cells. The concentrations of anthocyanins
and cyanidin required to cause a 50% inhibition of cell growth was 780 uM and 63
uM for HT 29, respectively and 285 uM and 85 11M for HCT 116, respectively.
Anthocyanin degradation products did not inﬂuence cell growth. These results
indicated that cyanidin directly inhibits the growth of colon cancer cells.

Based on these results, we conducted a second feeding study with ApcM'"
mice to determine the extent to which dietary anthocyanins and sulindac 1) inhibit

intestinal tumorigenesis, 2) inﬂuence expression and cellular localization of B-

catenin, and 3) inﬂuence expression of target genes of the Wnt/B-catenin signaling
pathway as well as global gene expression proﬁles. The dietary treatments were; 1)
control diet (modiﬁed American Institute of Nutrition 93G diet containing 220 g’k g
protein, 150 g/kg soybean oil and 50 g/kg cellulose), 2) control diet + 200 ppm
sulindac, 3) control diet + 750 ppm anthocyanins. The concentrations of sulindac
(200 ppm) and anthocyanins (750 ppm) were based on efﬁcacious concentrations
used in the previous experiment, but were administered in the diet in this study to
increase their stability.

The results of this second study demonstrated that dietary anthocyanins

signiﬁcantly reduced adenoma development in all three intestinal sections — small

150

intestine, cecum and colon. Mice consuming anthocyanins had less cecal adenoma
burden (volume) compared to mice consuming either sulindac or the control diet.
Relative to controls, sulindac-consuming mice had fewer and smaller adenomas in
small intestine, but larger cecal adenomas.

Sulindac did not inﬂuence the expression of B-catenin protein in intestinal
epithelium. Anthocyanins increased B-catenin levels in the middle small intestine
and cecum compared to control or sulindac. Intracellular distribution of B-catenin
was primarily in the cell membrane of mice consuming all treatments. cDNA
microarray analysis showed that the expression of mRNA for APC, B-catenin, and
most of the target genes of Wnt/B-catenin signaling were not inﬂuenced by treatment.
Collectively, these observations indicate that the mechanisms responsible for tumor
inhibition by sulindac in small intestine and by anthocyanins in all three intestinal
sections do not directly involve inhibition of the Wnt/B-catenin signaling pathway.

cDNA microarray analysis was used to screen for global proﬁles of gene
expression that were inﬂuenced by anthocyanins and sulindac. A total of 1,492
genes were inﬂuenced by treatment in any of the intestinal regions. Relatively few
genes were inﬂuenced by treatment in any two of the three intestinal sites — small
intestine, cecum or colon. No single gene was inﬂuenced by treatment in all three
regions. Genes inﬂuenced by anthocyanins and sulindac are involved in a wide
variety of cellular and molecular processes such as DNA synthesis and repair,
transcription, mRNA processing, protein interactions and turnover, cell proliferation
and differentiation, cell cycle control, and programmed cell death. Among the many

genes whose expression was inﬂuenced by treatment, we were particularly interested

151

in the following effects. Anthocyanins and sulindac up-regulated translocase of
inner mitochondrial membrane 8 (regulation of apoptosis) in cecum and small
intestine, respectively. Anthocyanins up-regulated ribonucleotide reductase M2
(DNA synthesis and repair) and nemo-like kinase, astrocytic phosphoprotein (signal
transduction) whereas anthocyanins down-regulated omithine decarboxylase (cell

proliferation).

In summary, a series of experiments was conducted to determine the effects

Min

of anthocyanins on intestinal tumor development in Apc mice. Our most

important ﬁndings were 1) characterization of the early and rapid development and
Min

progression of intestinal tumors in Apc mice, 2) demonstration that anthocyanins

and cyanidin inhibited tumor development in ApcMin mice, 3) tumor inhibition by

anthocyanins throughout the intestine of ApcMin mice, whereas sulindac only
inhibited small intestinal tumors, 4) evidence that tumor inhibition in ApcMin mice by
anthocyanins and sulindac is not likely to be mediated through inhibition of the
Wnt/B-catenin growth signaling pathway, and 5) identiﬁcation of several genes that
may be related to reduced tumor development by anthocyanins and sulindac.

Future experiments to further elucidate the effect of anthocyanins on
intestinal development should include 1) conﬁrmatory analysis using quantitative
PCR or Northern blot analysis is needed to quantitate the changes in expression of
genes identiﬁed by microarray analysis as being inﬂuenced by anthocyanins and/or
sulindac, 2) cellular expression of signaling proteins and other markers associated

with anthocyanin treatment by immunohistochemistry, 3) studies using other colon

cancer models to conﬁrm that anthocyanins inhibit colon cancer development, 4)

152

 

efﬁcacy studies with compounds that are structurally related to anthocyanins but

have greater stability.

153

APPENDICES

154

 

Table A1. Carrier status, body weights, and tissue lengths of mice from experiment I.

 

Animal ID Sex Age Statusa Bwt (g) SI length (mm)b Colon length (mm)

 

l M 30 C 14.0 303 64
2 M 30 C 14.9 317 75
3 F 30 C 12.9 305 79
4 F 30 C 16.8 309 71
5 M 30 C 14.9 316 79
6 F 30 C 10.0 289 59
7 F 30 C 14.3 290 55
8 F 30 C 15.0 288 66
9 M 30 N 16.5 288 66
10 F 30 N 15.3 329 64
11 M 30 N 16.3 320 68
12 F 30 N 16.3 318 70
13 M 30 N 19.7 350 75
14 F 30 N 14.8 314 63
15 F 30 N 14.0 309 49
16 F 30 N 15.0 350 72
17 M 30 N 10.2 265 53
18 M 30 N 10.5 252 64
19 M 30 N 11.6 285 64
20 M 30 N 17.5 334 82
21 M 30 N 14.4 311 76
22 M 30 N 13.9 304 76
23 M 45 C 19.0 342 83
24 F 45 C 18.2 334 80
25 M 45 C 16.7 320 74
26 M 45 C 20.8 320 81
27 M 45 C 18.2 300 80
28 F 45 C 15.8 295 71
29 M 45 C 19.4 333 83
30 F 45 C 16.9 310 73
31 M 45 C 19.0 325 79
32 M 45 C 18.5 335 67
33 M 45 N 20.5 310 72
34 M 45 N 20.5 342 81
35 F 45 N 17.0 302 74
36 M 45 N 21.5 309 85
37 M 45 N 21.0 330 85
38 F 45 N 16.0 325 70
39 M 45 N 18.9 339 74
40 M 60 C 19.5 303 69

 

155

 

41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
6O
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83

6O
60
60
6O
60
60
6O
60
6O
60
6O
6O
6O
60
6O
60
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
90
90
90
90
90

000002ZZZZZZZZZZZZOOOOOOOOOZZ22222200000000

24.3
21.3
17.5
18.9
20.7
18.3
18.8
21.1
21.9
23.9
19.7
22.6
19.5
18.4
18.9
23.5
23.1
19.1
21.2
21.5
19.4
20.7
21.9
18.5
19.0
24.5
20.2
24.5
19.0
24.3
24.8
20.3
20.6
18.3
19.7
21.2
17.8
21.3
25.4
19.2
20.5
21.1
23.2

338
333
324
346
340
322
345
335
333
380
309
325
370
340
349
365
362
357
330
356
330
370
330
339
347
340
355
370
327
374
366
360
350
353
343
356
325
336
360
330
345
362
318

85
94 ‘
69
80
81
70
80
70
80
97
75
76
78
75
82
89
84
78
81
85
80
79
85
80
77
75
78
85
72
80
78
78
82
73
82
83
83
90
81
80
72
75
83

 

156

 

 

84 F 90 C 18.9 337 71
85 F 90 C 19.3 354 75
86 M 90 C 22.2 350 87
87 F 90 C 17.5 300 71
88 M 90 C 21.4 317 89
89 F 90 N 20.4 356 76
90 M 90 N 24.7 379 98
91 M 90 N 25.9 360 55
92 F 90 N 21.7 356 62
93 M 90 N 27.6 391 89
94 F 90 N 20.0 361 83
95 M 90 N 25.2 360 70
96 F 90 N 22.8 373 75
97 M 90 N 27.5 370 75
98 M 90 N 25.1 340 63
99 M 90 N 23.4 356 87
100 M 90 N 25.1 370 80
101 F 90 N 21.1 330 81
102 F 120 C 21.6 366 81
103 F 120 C 21.6 380 70
104 M 120 C 27.3 376 70
105 F 120 C 23.5 365 85
106 M 120 C 22.6 394 76
107 M 120 C 26.7 374 100
108 M 120 C 24.0 405 95
109 F 120 C 19.2 367 86
110 M 120 N 27.0 390 85
111 M 120 N 25.6 371 83
112 F 120 N 23.0 365 70
113 F 120 N 23.1 396 77
114 F 120 N 21.2 382 83
115 F 120 N 22.1 370 84
116 M 120 N 27.3 367 97
117 M 120 N 24.8 389 88
118 F 120 N 21.3 381 80
119 M 120 N 26.4 368 85
120 M 120 N 26.5 374 95
121 F 120 N 21.5 386 85
122 M 120 N 26.8 360 90
123 M 120 N 26.1 375 90
124 F 120 N 20.3 380 86
“Mm

 

aCarrier Status: C; carrier (Apc

bSmall intestine

), N; normal (Apc

 

Table A2. Numbers and sizes of adenomas in ApcMin mice from experiment I.

 

 

 

Adenoma Number Diameter Volumec
(mm) (mm“)
ID3 SIb PSIb MSIb DSIb Cecum Colon SIb Cecum Colon
1 6 3 0 3 2 O 5 3 O
2 8 3 3 2 0 l 8.5 0 l
3 13 6 3 4 0 3 10.5 0 1.5
4 10 3 4 3 l l 11 3 9
5 11 2 5 4 0 3 8 0 2.5
6 l3 4 5 4 2 2 13.5 4 2.8
7 16 5 5 6 0 1 18 0 0.8
8 42 7 l6 l9 0 O 25.5 0 0
23 25 2 10 13 1 3 21 1.5 2
24 27 6 11 10 l 2 20.5 1 5
25 25 6 12 7 1 2 22 2 3
26 35 7 12 16 1 0 29.5 1 0
27 26 3 13 10 l 3 21.5 1.5 2
28 34 9 13 12 1 O 28.5 1 O
29 31 3 16 12 2 2 24.5 2.5 3.5
30 46 11 21 14 3 2 42.5 3.5 l
31 65 10 26 29 0 1 49 0 1
32 46 2 27 17 2 2 31 2.5 2
40 10 4 2 4 O 2 13 O 1
41 9 3 5 1 O 1 11.8 0 0.5
42 24 3 5 16 4 1 24.5 4 0.5
43 25 4 7 14 O 1 21.5 0 l
44 25 4 8 13 O 1 28.8 0 0.5
45 27 3 10 14 0 0 29 0 0
46 34 5 14 15 3 1 33 4 0.5
47 108 9 21 78 2 2 80 1.5 5
48 101 14 34 53 2 4 91 2 4.5
57 29 4 10 15 l 1 29.5 1.5 l
58 25 6 l3 6 1 3 23 l 2.5
59 37 2 13 22 0 2 37 0 2
6O 50 8 15 27 1 l 48 2 1
61 43 5 16 22 2 2 40.5 3.5 l
62 78 5 22 51 2 2 67.5 3 2.5
63 58 7 23 28 2 4 61.5 4.5 3
64 53 9 25 19 0 2 50.5 0 5
65 90 38 38 14 l 3 96.5 1.5 2
79 10 4 2 4 0 2 13 0 1
8O 41 9 13 19 l 4 49 l 5

 

158

 

81
82
83
84
85
86
87
88
102
103
104
105
106
107
108
109

29
37
60
100
113
95
145
163
35
43
44
68
92
71
96
126

6
4
12
ll
l6

17
19
11

"‘0th

0
7

14
15
16
30
32
40
45
62
7

ll
17
23
26
29
33
48

9

18
32
59
65
49
83
82
17
23
20
41
61
36
53
71

wH—‘wNNOWWNNH-bNo

0

MHw—‘OH—dONr—Av—sn—tmom

5

30
39.5
52
104.5
131
121
147.5
231
58.5
83
75.5
109
185.5
111
212.5
236

14
0

5.5
0

4

6

40
93.5
16.5
50.5
6
2.5
0

72
105
90
140
33.5

 

3 Animal ID. Note that Normal (Apcm) mice were excluded.
b SI; small intestine, PSI; proximal small intestine, MSI; middle small intestine, DSI;

distal small intestine

cValues for adenoma volume of cecum and colon represent values calculated by
width x length x height of adenomas. These values were transformed by 7t/ 6 to
convert them to spherical tumor volumes in the ﬁnal analysis.

159

Table A3. Weekly body weights (g) of ApcMin mice from the experiment 11.

 

IDa Sex InitialkalWk2Wk3Wk4Wk5Wk6Wk7Wk8Wk9Wk10

 

Conl M 18.9 22.3 23.5 25.7 28.1 29.2 27.1 29.9 30.4 28.2 26.8
Con2 M 17.9 20.8 22.3 23.5 25.2 26.6 29.4 28.8 28.4 28.2 26.8
Con3 F 12.8 15.9 17.2 18.4 18 20 20.5 20 21.7 22.1 20.9
Con4 F 16.1 17.2 18.3 19.4 20.1 20.4 20.6 22.1 22.5 22.1 22.8
Con5 F 16.9 17.5 18.3 18.6 18.8 19.1 18.7 18.1 17.4 15.8 14.9
Con6 F 16.4 18.4 18.8 20.1 20.2 20.7 21.2 20.8 21.1 21.6 20.8
Con7 M 19.5 21.9 23.1 24.5 26.1 27 27.2 26.4 24.7 24.7 21.2
Con8 M 15.6 20.8 22.6 23.8 23.6 26.5 26 26.8 26.8 25.5 24.9
Con9 M 14.8 20.5 22.4 24.7 26.7 29.8 28.9 30.5 30.7 29.6 29.1
Antl M 19.5 23.5 24.9 25.6 24.2 29.5 31.1 32.2 33.7 35 36

Ant2 M 19.1 21.7 23.3 25.1 26.2 28.3 28.8 30.2 31.6 32.4 33.7
Ant3 F 16.2 18.1 18.9 19.6 20.2 20.6 20.4 17.8 16 NA NA
Ant4 M 18.9 21.6 22.7 23.8 23.3 22.1 20.8 19.2 18.6 17.6 18.6
Ant5 F 15.5 17.2 19.2 20.4 20 21.4 21.4 21.8 19.8 17.1 15.3
Ant6 M 19.9 21.7 23.7 24.3 25.2 26.7 27.9 29.2 29.5 31.7 32.4
Ant7 F 15.6 17.2 19.1 19.4 20 21.2 21.8 22.5 22.6 22.7 22.9
Ant8 F 15.2 17.5 19.3 19.9 20.8 21.7 22 22 21 19.5 18.1
Cyal M 18 21.2 22.1 22.4 23.7 23.9 23.8 22.9 22.8 22.2 20.5
Cya2 F 17.2 19.5 20.3 21.5 23.3 22 23.2 27 26.5 23.8 26.2
Cya3 F 15.4 17.2 18.3 18.9 20.2 19.2 20.1 20.1 20.2 19.3 18

Cya4 M 18 20.7 21.8 22.6 22.5 23.1 22.5 21.6 18.9 18.9 19.5
Cya5 F 16.7 19 20.1 21.2 21.5 22.1 22.9 23.5 22.8 22.8 24

Cya6 F 16.6 16.8 19 20.4 20.5 21.1 22 22.4 22.6 22.5 22.5
Cya7 F 16.3 17.6 19.4 19.8 20.7 21.3 22.3 21.9 22.4 23 22.5
Sull M 16 19.8 21.3 22.3 22.8 23.5 25 25.5 25.9 26.4 27.1
Su12 F 16.2 17.6 19.1 20 21 20 21.1 22 23.1 22.6 22.8
Sul3 F 16.2 17.2 18.6 20.8 20.8 20.6 22.5 22.6 23.3 22.9 22.1
Sul4 M 16.5 19.9 21.4 21.8 22.4 23.4 23.6 24.4 25.4 22.1 25.3
Su15 F 15.7 17.1 18.3 19.5 20.7 21.2 21.9 21.7 22.1 25.3 22.3
Sul6 M 18 20.6 22.2 23.1 23.8 25 25.5 26.5 26 24.8 25.2
Sul7 F 17 17.8 19.4 19.7 20.7 21.4 21.3 21.6 22.1 22.2 22.4
Su18 F 15.3 17.5 19.3 19.7 19.8 20.2 21.5 23.4 23.9 25.8 26.7
Sul9 F 13.2 16.6 17.2 18.2 19.8 19.5 20.6 22.6 22.4 22.3 22.7
Sul 10 M 19 22.6 24.3 25.9 27.5 28.1 29 31.2 31.9 32.9 33.7
Chel M 17.5 20.3 22.2 23.2 23.3 24.2 24 26.1 26.4 26 24.5
Che2 F 13.9 17.9 19.9 19 20.7 21.5 22.2 23 22.3 22.9 23.5
Che3 F 15.9 17.5 18.6 19.9 19.9 17.3 16.7 15.2 15.7 15 NA
Che4 M 19.3 21.3 22.2 22.5 23 22.4 21 17.7 19.7 14.3 NA
CheS M 18.9 21.1 22.1 22.1 22.9 22.7 23.4 23.5 19.5 NA NA
Che6 M 19.4 20.5 20.7 22.4 23 23.5 23.6 24.4 24.4 23.9 18.9

 

 

 

160

 

Che7 M 19.4 20.6 21.2 21.7 22.3 23 23.3 23.9 23.7 23.2 19.2
Che8 F 17.4 17.7 19.5 20.4 21.4 21.6 21 182 16.8 17.2 NA
Che9 F 13.9 17.3 19.2 20.1 21.8 20.5 20.3 21.8 21.3 21 25.3
ChelOF 14 17.5 18.5 19.6 20.3 22 22.3 20.4 22.5 20.8 NA
aAnimal ID: Con; control, Ant; anthocyanins, Cya; cyanidin, Sul; sulindac, Che;
20% Cherry diet.

bInitial; body weight at the beginning of the experiment, week 10; body weight at the
end of the experiment.

NA: Data not available due to death of animal

161

Table A4. Tissue lengths and numbers and sizes of adenomas in ApcMm mice from
experiment II.

 

 

 

 

Lglgth Adenoma Number Diameter Volume

(mm) (mm“) <mm“)
11)“ SI“ Colon 31“ PSI“ M31“ DSI“ CecumColon SI“ CecumColon
Con 1 363 93 44 10 20 14 2 4 60.5 11.5 4
Con 2 365 86 41 12 16 13 1 6 55 9 5.5
Con 3 356 84 33 9 13 11 5 2 45 4.5 1
Con 4 365 75 2 2 O O 1 3 2 0.5 3
Con 5 352 79 157 29 57 71 3 5 217.5 4 3
Con 6 390 85 91 16 13 62 3 3 129.5 11 6.5
Con 7 390 88 47 15 16 16 2 3 69.5 2 3
C011 8 363 78 42 14 14 14 3 3 71 8 3
Con 9 370 87 48 15 22 11 1 4 68 2 12
Ant 1 370 98 10 4 4 2 O 6 13.5 0 5
Ant 2 370 90 12 5 5 2 0 2 12 O 2
Ant 3 325 56 147 27 54 66 2 5 164.5 5.5 30
Ant4 380 87 131 38 32 61 O 4 155.5 0 5.5
Ant 5 375 82 129 34 51 44 O 3 175 0 11
Ant 6 375 95 2 2 0 0 2 2 2.5 4 3
Ant 7 366 63 54 17 20 17 O 1 97 O 1
Ant 8 370 72 121 26 45 50 l 2 198 2 3
Cya 1 377 84 52 25 6 21 1 5 67.5 4 17.5
Cya 2 390 84 96 20 44 32 1 4 122.5 1.5 4.5
Cya 3 350 80 7 5 2 0 O 3 6 O 2.5
Cya4 375 83 70 3 34 33 0 2 104.5 0 2
Cya 5 367 88 35 31 2 2 0 2 42 0 2
Cya 6 360 75 47 7 15 25 2 4 54.5 2 5
Cya 7 370 82 52 18 7 27 O 6 75 0 13.5
Sul 1 372 95 32 6 14 12 3 7 43.5 7 20.5
Sul 2 360 90 O 0 0 O 10 8 O 15 6.5
Su13 366 85 8 5 3 O 9 3 6.5 12.5 6
Sul 4 384 85 50 8 20 22 4 6 40.5 8 6.5
Sul 5 374 83 19 5 9 5 2 7 16.5 2 6
Sul 6 380 78 92 12 3O 50 3 2 88.5 4 2
Sul 7 386 88 17 7 4 6 3 9 13.5 5.5 7
Su18 366 98 15 7 3 5 2 3 13 2.5 2.5
Sul 9 371 80 8 4 4 O 3 5 7 17.5 4
Sul 10 360 92 6 2 4 0 1 3 6 2.5 3
Che 1 405 65 21 2 10 9 1 4 44 2 35
Che 2 350 80 10 3 2 5 0 5 10 0 4
Che 3 356 85 86 10 31 45 1 O 155 24 O

 

 

162

 

Che 4 375 82 63 21 21 21 0 2 111 0 2
Che 5 375 82 43 9 11 23 2 1 83 1.75 2.5
Che 6 336 80 69 9 21 39 1 8 120 4.5 43
Che 7 390 80 27 5 10 12 0 2 62 0 58
Che 8 440 80 80 19 27 34 1 1 142 2 3
Che 9 375 90 5 3 2 O 0 ' 7 6 0 8
Che 10 365 92 12 5 4 3 0 3 11.5 0 4

 

a"Animal ID: Con; control, Ant; anthocyanins, Cya; cyanidin, Sul; sulindac, Che;
Cherry diet.

bSI; small intestine, PSI; proximal small intestine, MSI; middle small intestine, DSI;
distal small intestine

dValues for adenoma volume of cecum and colon represent values calculated by
width x length x height of adenomas. These values were transformed by 7t/ 6 to
convert them to spherical tumor volumes in the ﬁnal analysis.

163

Table A5. Sex, carrier status, and tissue lengths of mice from experiment III.

 

 

ID“ , Sex Carrier Status“ SI length (mm)° Colon length (mm)
Con 1 F N 340 65
Con 2 F C 310 70
Con 3 M C 354 80
Con 4 M N 270 67
Con 5 M C 350 75
C011 6 M N 300 65
Con 7 F C 330 65
Con 8 F N 250 51
Con 9 F C 262 57
Con 10 F N 260 50
Con 12 M C 240 59
Con 13 M N 307 55
Con 14 F N 330 60
Con 15 F C 224 76
Con 16 M C 330 81
Con 17 M N 320 73
Con 18 F C 305 55
Con 19 F N 295 65
Con 20 F N 305 76
Con 21 M C 321 65
Con 22 F N 260 56
Con 23 F C 230 54
Con 24 F N 288 58
Con 25 M C 315 70
Con 26 M N 330 70
Con 27 M - N 285 70
Con 28 F N 300 62
Con 29 M C 270 67
Con 30 M C 292 69
Con 31 M N 316 79
Con 32 M N 323 73
Con 33 F N 321 65
Con 34 M N 336 69
Con 35 F C 299 65
Con 36 F C 305 66

 

164

I:

 

 

Con 37
Con 38
Con 39
Con 40
Con 41
Con 42
Con 43
Con 44
Con 45
Con 46
Con 47
Con 48
Con 49
Con 50
Ant 1

Ant 2

Ant 3

Ant 4

Ant 5

Ant 6

Ant 7

Ant 8

Ant 9

Ant 10
Ant 11
Ant 12
Ant 13
Ant 14
Ant 15
Ant 16
Ant 17
Ant 18
Ant 19
Ant 20
Ant 21
Ant 22
Ant 23
Ant 24

“7133333371”“7““ngZFFZZZFZZZFFZZZZZZZZ””71”“

0202002ZOOOZZOZOZOZZOOOZZOZZZZOZZZOOZO

336
336
286
296
302
345
305
327
311
326
311
302
305
303
338
372
360
384
290
320
315
315
325
261
252
260
313
293
259
330
297
317
305
320
290
290
305
290

71
71
69
70
70
76
72
75
71
76
73
64
68
71
70
95
90
85
60
75
65
65
63
62
55
61
59
68
63
62
71
77
77
78
65
7O
50
55

 

165

 

 

lAnt25
lAnt26
lAnt27
lAnt28
lAnt29
.Ant30
lAnt31
.Ant32
.Ant33
lAnt34
lAnt35
lAnt36
.Ant37
lAnt39
.Ant40
lAnt41
lAnt42
lAnt43
.Ant44
lAnt45
lAnt46
lAnt47
lAnt48
lAnt49
lAntSO
Sull
Su12
Su13
Sul4
Su15
Sul6
Su17
Su18
Sul9
SullO
Sulll
Su112
Sull3

“ZZF’T’FFFZZZEZZFPF““ZZZZZZFT’FF'“ZZZZ'P'PKZ

ZOOOZZZOOOZOOZOOOZOOOOOZOOZ22200202002

290
273
247
265
311
307
309
314
.318
315
314
307
290
310
322
320
294
300

300 .

278
321
298
319
295
311
345
310
310
300
330
375
300
305
330
273
273
247
311

69
6O
57
66
74
70
67
71
61
72
63
61
69
75
76
81
73
68
67
68
67
70
73
59
61
90
78
70
64
67
82
60
65
75
61
61
58
75

 

166

 

 

Sul 14 F N 320
Sul 15 F C 235
Sul 16 M C 346
Sul 17 M N 330
Sul 18 F C 325
Sul 19 F N 328
Su120 M C 346
Su122 M C 305
Su123 M C 338
Su124 F N 303
Su125 F C 305
Su126 M N 293
Su127 F C 297
Su128 F N 298
Sul 29 M C 337
Sul 30 M C 331
Sul 31 F N 328
Sul 32 F C 305
Su133 F N 316
Sul 34 F N 297
Sul 35 F C 319
Sul 36 M C 328
Su137 F C 296
Sul 38 M N 311
Sul 39 F C 319
Sul 40 M N 315
Sul 41 M N 325
Sul 42 F C 300
Sul 43 M C 315
Sul 45 M C 326
Sul 46 M C 331
Sul 47 M C 306
Sul 48 F N 311
Su149 F C 291
Sul 50 F N 294

62
67
67
75
73
72
75
65
77
59
65
69
67
62
71
74
67
70
65
70
64
7O
62
65
67
71
72
64
69
74
67
68
62
63
60

 

 

aAnimal ID: Con; control, Ant; anthocyanins, Sul; sulindac
t’Carrier status: C; carrier (AchMm ), N; normal (Apcm )
cSmall intestine

167

Table A6. Weekly body weights (g) of mice from experiment III.

 

 

ID“ Initial“ Wkl Wk2 Wk3 Wk4 Wk5 Wk6 Wk7 Wk8
Con 1 13.9 17.9 19.8 20.5 21.8 23.5 24 .6 25.1 24
Con 2 14.5 16.5 17.1 18.3 19.2 20.3 20.7 21.3 22
Con 3 14.1 19.4 22 23.8 24.5 25.2 26.1 26.5 27.5
Con 4 13.1 17.9 19.9 21.3 22 23.7 24.2 25 26.4
Con 5 14 19.7 22.2 24 25 25.8 26.4 27.7 27.9
Con 6 14.8 19.1 20.9 22.7 22.9 23.9 24.7 25.4 25.3
Con 7 13.1 17.6 19 20.5 21.4 21.4 21.2 21.3 20.5
Con 8 13.7 17.7 16.9 18.7 18.4 20.7 21.8 22.6 23.3
Con 9 13.2 17.5 17.3 19.2 20.1 21.4 22.9 24.7 25.5
Con 10 13.4 17.6 16.4 18 18.3 20.6 21.9 22.1 22.5
Con 12 14.4 19.8 21.6 22.7 23.9 26.9 27.5 27.7 28.1
Con 13 15 20.6 23.9 26.3 26.5 27.3 28.2 28.8 28.6
Con 14 16.9 18.3 19.9 20.3 22.8 23.5 23.8 24.5 25.5
Con 15 16.9 18 19.1 20.9 21.8 23.4 23.6 24.3 26.8
Con 16 18.4 ' 22.4 24 25.4 26.9 28 28.3 29.8 29.3
Con 17 19.2 22.3 23.9 25.2 26 26.8 27.2 27.8 28.1
Con 18 15.3 17.4 18.9 18.8 20 20.4 21.5 22.5 22.4
Con 19 13.1 17.5 18.1 19.5 21.5 21.8 23.8 25.8 26.6
Con 20 15 18.1 19.4 19.9 21.5 21.5 23.8 25 25.4
Con 21 13.7 20 22.6 23.8 25.1 26.2 27.6 28.7 29.9
Con 22 12.6 17.8 18.2 19.5 20.5 21.1 22.9 22.4 22.7
Con 23 11.1 17.7 17.3 18 18.8 20.2 20.6 20.9 21.2
Con 24 12.9 18.2 18.9 20.2 21.8 21.9 23.1 23.1 23.5
Con 25 14.6 19.4 21.7 23.4 24.7 25.2 26.8 27.7 29.1
Con 26 16.9 21.3 24.9 26.8 28 28.3 30.1 31.6 33
Con 27 13.7 19.3 21.1 22.9 23.8 24.4 25.5 26.2 27.3
Con 28 13.3 19.4 20 23.1 25.4 26.3 28.2 28.6 29
Con 29 12.5 18.6 20.2 21.9 22.9 23.1 23.5 24.3 24
Con 30 12.5 19.3 20.9 22.7 23.5 23.8 24.1 25 24.6
Con 31 15.3 21.7 22.8 24.4 24.9 25.7 27 27.9 28.3
Con 32 14.4 21.5 24.2 25.4 26.8 27.8 28.5 29.7 30.1
Con 33 13.7 18.9 18.7 19.7 20.4 21.1 22.4 22.7 23.3
Con 34 14.6 19.8 22.2 23.3 25 25.5 26.5 28 28.6
Con 35 12.5 16.7 18 18.4 19 19.7 20.7 21 20.7
Con 36 12.7 16.7 17.2 18.3 18.6 20.5 21.2 21.5 20.6

 

168

 

Con 37
Con 38
Con 39
Con 40
Con 41
Con 42
Con 43
Con 44
Con 45
Con 46
Con 47
Con 48
Con 49
Con 50
Ant 1

Ant 2

Ant 3

Ant 4

Ant 5

Ant 6

Ant 7

Ant 8

Ant 9

Ant 10
Ant 11
Ant 12
Ant 13
Ant 14
Ant 15
Ant 16
Ant 17
Ant 18
Ant l9
Ant 20
Ant 21
Ant 22
Ant 23
Ant 24

15.2
14
11.4
13.8
13.1
18
15.7
15.4
13.7
14.1
13.2
13.8
13.8
14.7
19.4
15.4
15.7
9.4
9.3
9.8
9.1
9.2
14.4
16
11.8
13.3
13.8
11.9
10.5
16.7
19.6
17.2
15.7
18.6
17.7
17.4
14.6
14.6

18

19.8
15.6
19.2
19.5
18.5
19.4
19.4
15.3
15.2
13.9
16.8
16.9
19.5
23.1
18.2
18.8
14.1
14.2
14.6
12.8
12.5
20.3
20.6
18.2
17.7
16.9
16.6
14.6
18.2
22

21.4
19.8
22.2
21.7
21.3
18.7
17.9

19
20.1
16.5
20.4
21
21.7
21.7
22.8
20.4
20.4
19.2
18.6
16.9
21.1
25.6
18.7
19
19.2
20
2.5
17
16.6
22.9
23.5
21.5
16.8
16.9
17.3
16
21.1
23.6
23.1
21.2
24.1
24.3
23.4
20.2
19.2

18.9
21.3
17

21.3
21.8
24.4
24.4
25.4
20.5
21.5
21

18.9
18.1
23

26.7
20.1
20.3
21

21.1
21.5
17.2
18.2
24.1
25.5
24.7
19.3
18.8
19.5
16.2
21.7
24.4
23.6
22

25.2
25.9
24.5
21.3
19.4

169

20.7
23

18.2
23

22.5
27.2
26.6
27.9
22.6
23.4
23.4
19.9
19.3
23.6
27.8
20.4
20.9
22.7
22.9
22.9
17.7
18.3
24.7
26.9
26.6
19.4
18.5
18.4
16.2
24.5
24.6
24.8
22.6
27

27.2
25.9
21.4
19.8

22.4
23.8
19.7
23.5
23.6
29.4
27.7
30

24.4
25.4
25.4
20.6
19.9
24.8
29.4
21.8
23.6
24.3
24.1
24.4
19

19.9
26.2
28.2
28

21 . 1
21.4
20.2
17.6
25.9
25.3
25.9
23.1
27.3
27.8
26.7
21.3
21.1

23.9
26

20.1
24.2
24.2
31.2
28

30.4
25.1
26.4
26.5
21.9
20.2
25.7
29

22.3
23.6
25

24.3
24.2
19.2
20.1
26.4
29.8
29.6
22.8
22.5
21.9
19.3
27.3
25.9
26.9
23.2
27.3
28

27.2
22.6
21.4

24.3
28.3
20.5
24.8
24.6
32.5
28.2

29.9

26.6
27.2
27.4
21.9
20.1
26.2
30.7
23.1
26.1
26

25.5
25.1
19.8
20.9
27.1
30.3
29.9
22.1
21.2
23

19.2
27.7
27.4
28.7
24

27.8
29

28.6
23.6
21.8

24.7
27.3
20.9
24.5
24.2
34.8
29.6
30.5
28.7
28

27.7
22.5
20.6
23.6
31 .1
23.6
26.2
26.6
26.2
25.7
23.9
21.2
27.2
30.2
31.1
22.3
22.6
24.1
19.6
28.2
27.5
28.6
24.2
24.9
29.6
17.9
23.5
21.9

 

.Ant25
Amu26
Amu27
-1Ant28
An129
.Ant30
Aun3l
Aun32
Aun33
Aun34
Auu35
Ant36
Aun37
.Ant39
Ant40
Ant4l
.Ant42
Ant43
Auu44
Ant45
Auu46
Auu47
Aun48
.Ant49
AntSO
Sull
Su12
Sul3
Sul4
Su15
Su16
Sul7
Su18
Su19
SullO
Sulll
Su112
Sull3

136
106
1L5
122
169
138
144
143
13J
134
14
139
141
135
134
142
103
92
10
83
l3
14
13
132
138
18
18A
182
9A
99
88
89
94
133
1L9
126
124
135

205
173
162
162
222
18

19A
195
167
176
173
188
185
202
2L2
208
148
144
15

95

142
174
169
162
198
209
205
202
128
149
125
124
132
174
162
179
178
162

222
196
179
174
234
206
222
22

185
188
176
183
192
223
234
232
193
193
194
132
172
186
173
175
206
2L8
22

222
174
186
166
166
172
17

17J
189
188
168

243
2L5
19

187
ZSJ
229
246
243
19A
19A
182
199
19

232
242
252
2L4
2L6
2L5
149
188
183
199
184
2L9
242
24

23A
193
199
171
165
18

192
183
202
202
177

170

256
227
198
19J
263
243
236
246
198
202
185
22

2L3
232
259
269
228
24A
235
134
206
202
195
19

237
252
246
243
195
208
178
165
185
196
189
21

204
181

26

23

202
20

265
25

264
249
209
2L3
20

22

232
264
273
274
238
234
246
165
207
2L9
208
202
231
265
269
261
2L2
2L9
189
176
196
2L1
20

223
2L6
199

26

24

21

20

269
259
273
239
21

23

2L5
226
236
266
285
285
243
265
255
172
2L5
2L4
2L1
2L1
239
273
27

26

2L5
223
198
181
205
2L6
21

232
222
2L3

27

248
2L5
2L7
288
269
284
272
221
228
221
245
233
272
284
279
25

271
262
182
234
233
233
2L6
271
27

274
262
225
235
199
191
208
224
2L4
237
229
224

276
263
236
222
295
277
287
275
226
2L9
2L9
25

263
279
288
284
261
28

269
199
245
222
2L8
206
265
272
272
239
231
234
199
187
2L6
223
231
236
228
222

 

 

Sul 14 11.8 16.4 17.5 19.3 19.4 21.3 22 21.9 22.4
Sul 15 16.4 16.7 18.8 19.7 21 21.8 22.4 23.2 24.5
Sul 16 18.8 21.9 24.4 25.8 26.5 27.3 28.1 28.8 29

Sul 17 15.9 18.6 20.9 23.1 24.5 26.4 26.9 27.4 27.2
Sul 18 15.2 17.5 18.6 20.8 21.7 22.4 23 24.6 25.4
Sul 19 14.5 17.6 18.7 20.8 22.2 22.6 23.4 23.5 24.5
Sul 20 15.1 17.7 21 23 24 24.6 25.3 26.5 26.7
Sul 22 15.3 20.8 22.4 22.9 24.5 23.5 24.5 24.5 24.6
Sul 23 13.8 20.5 23 24.6 26.3 26.3 27.3 27.7 27.6
Su124 12.1 15.2 16.8 17.6 18.2 18.8 19.5 19.7 19.9
Su125 11.6 16 18.2 18.3 19 19.7 20.3 20.9 21.2
Su126 12.1 18.7 20.8 22.8 23.2 24.1 24.4 25 25.3

Su127 11 16 16.3 17.7 19.1 18.5 18.9 19.5 20
Su128 11.5 17.4 18.1 19.5 21 20.3 20.9 20 22.1
Sul 29 14 19.9 21.6 22.4 23.2 23.8 24.3 24.8 24.4

Sul 30 13.8 i 19.2 21.1 22.8 23.9 25.3 25.8 26.8 26.1
Sul 31 15.2 19.3 19.6 21.4 21.8 23.9 25.4 25.7 26.7
Sul 32 12.8 16.6 17.9 18.7 18.7 19.1 20.6 22.2 23.1
Sul 33 12.7 16.7 17.1 17.7 18.2 18.5 18.8 20.3 19.9
Sul 34 11.1 16.6 17.3 17.8 18.2 19.3 21.1 20.8 20.1
Sul 35 13 17.6 17.7 19 19.7 19.8 21.4 21.9 21.9
Sul 36 16.9 20.2 21 22.5 23.4 24 24.9 25.3 25.3
Sul 37 12.1 15.1 17 17.5 18.5 19 19.3 19.4 19.5
Sul 38 14.4 18.4 19.8 20.4 21.7 22.1 22.8 23.2 22.7
Sul 39 13.1 17.1 17.2 18.9 19.7 20.9 21.1 21.7 21.9
Sul 40 12.8 18 19.5 21 21.8 22.5 23 24.3 25.3
Sul 41 13.8 18.8 21.1 22.5 23.7 24.6 25.7 26.1 26.5
Sul 42 11.8 14.9 17.1 18.8 21.2 22 23 23.4 24.1
Sul 43 9.4 14.3 18.3 19.1 20.3 21.6 22.4 22.8 23.4
Sul 45 11.7 13.6 16.9 17.5 18.4 19.1 20.2 21.3 21.3
Su146 16.2 21 22.2 23.9 24.7 25.2 26.8 27.2 28

Su147 12.4 15.7 16.8 17.5 18.9 20.6 21.5 20.6 21.8
Sul 48 14.6 17.1 18.5 19.4 21.6 22.1 24 23.9 23.8
Sul 49 12.6 13.8 14.8 16.2 17.6 18.2 18.8 19.2 19.5

Sul 50 13.7 16.9 17.1 18.1 19.3 20 20.3 20.7 20.8
aAnimal ID: Con; control, Ant; anthocyanins, Sul; sulindac

bInitial; body weight at the beginning of the experiment, Week 8; body weight at the
end of the experiment.

171

Table A7. Numbers and sizes of adenomas in ApcMin mice from experiment III.

 

Adenoma Number Diameter Volumec
(mm) (mm3)
ID“ Total SIb MSIb DSIb Cecum Colon SIb Cecum Colon

 

Con2 18 11 7 1 2 21.5 12 32
Con3 12 12 O 0 0 10.5 0 0
Con 5 9 6 3 1 1 15 0.25 24
C011 7 39 22 17 2 1 50.5 7 6
Con 9 8 5 3 1 2 9 1 1.5
Con 12 8 4 4 2 5 11 2 34
Con 15 20 12 8 1 1 25 0.5 0.5
Ccon 16 41 14 27 1 3 66 4 40
Con 18 23 11 12 O O 31.5 0 0
Con 21 25 17 8 1 3 31 9 3
Con 23 36 19 17 0 0 45 0 0
Con 25 24 14 10 1 3 30.5 1 2.5
Con 29 17 6 11 0 2 21.5 0 2
Con 30 28 15 13 1 1 37.5 1 24
Con 35 52 23 29 1 1 66.5 1.5 1
Con 36 56 3O 26 1 1 65 4 27
Con 37 14 10 4 1 6 17 2 5
Con 39 15 6 9 O 5 18 0 5
Con 40 19 8 11 2 4 22 3 4
Con 44 56 31 24 2 2 84.5 8 7
Con 49 4O 21 19 2 2 48 2 2
Ant 2 8 1 7 O O 9 O 0
Ant 3 6 O 6 1 O 9 0.25 0
Ant 4 44 2O 22 4 6 32.5 8 6.5
Ant 7 9 4 5 O O 11.5 0 0
Ant 9 3 3 O 0 1 4 O 12
Ant 11 9 4 5 2 O 10.5 1 0
Ant 14 21 8 13 0 O 26 O 0
Ant 15 13 8 5 O l 21 0 6
Ant 16 21 5 16 0 1 21.5 0 12
Ant 19 12 7 5 1 O 9 1 0
Ant 20 20 12 8 1 1 25.5 1 9
Ant 22 5 3 2 0 1 4.5 O 9
Ant 24 16 7 9 1 0 17 1 0
Ant 26 . 25 7 18 1 1 27.5 1 4
Ant 27 18 6 12 O O 17 O 0
Ant 29 14 7 7 O 1 20 0 1
Ant 31 25 7 18 O 0 25 O 0

 

 

172

 

 

Ant32 18 15 3 0 1 17.5 0 1
Ant37 6 4 2 0 0 7.5 0 0
Ant39 13 5 8 0 0 25 0 0
Ant4l 23 12 11 1 2 27 4 34
Ant42 6 1 5 1 1 7 4 4
Ant43 7 4 3 0 0 8 0 0
Ant44 11 4 7 0 2 15 0 10
Ant45 34 16 18 1 2 42 9 30
Ant47 18 9 9 1 1 11 2 2
Ant48 13 7 6 2 0 . 12.5 2.5 0
Ant49 41 18 23 0 2 39.5 0 1.5
81111 5 3 2 5 1 4.5 1.25 12
Su12 5 2 3 1 1 5.5 0.5 16
Su14 9 5 4 1 2 13.5 6 4
Su15 0 0 0 1 1 0 2 0.5
Sul6 3 1 2 0 0 6 0 0
Sul 10 7 2 5 3 4 6.5 8.5 2.75
Sulll 1 0 1 2 3 1 10 1.75
Su112 10 6 4 2 2 11.5 8 12
Sul 15 10 7 3 2 1 6.5 1.5 0.5
Sul 16 0 0 0 0 1 0 0 16
Su118 6 3 3 1 0 5.5 2 0
Su120 7 3 4 1 2 5.5 20 9.5
Su122 5 3 2 1 2 4 9 18.5
Su123 7 5 2 3 0 5.5 4 0
Su125 9 4 5 2 1 7 10 6
Su127 16 6 10 1 1 16 12 0.5
Su129 6 3 3 1 1 5.5 12 18
Sul 30 12 4 8 1 4 12.5 12 3.5
'8u132 4 2 2 7 1 5.5 20 1
Su135 28 18 10 1 1 23.5 9 9
Su136 7 3 4 1 2 7 12 2
Su137 8 4 4 2 3 10 10 11
Sul39 7 3 4 0 3 9.5 0 3.5
Sul42 .3 2 1 3 1 5.5 9 2
Sul43 4 2 2 1 2 3 20 2
Sul 45 25 16 7 1 2 19.5 16 10
Sul46 29 13 16 4 3 21 20 9
Sul47 10 6 4 3 1 9.5 12 4
Su149 8 4 4 7 1 11 16.5 1
+

a Animal ID. Note that Normal (Apcw ) mice were excluded.
b SI; small intestine, PSI; proximal small intestine, MSI; middle small intestine, DSI;
distal small intestine

173

cValues for adenoma volume of cecum and colon represent values calculated by

width x length x height of adenomas. These values were transformed by 11/6 to
convert them to spherical tumor volumes in the ﬁnal analysis.

174

 

Table A8. Intensity scores of [El-catenin expression in small intestine of mice from
experiment III.

 

[De Carrier Statusb Proximal Small Intestine

Top° Bottomc
Membrane Cytoplasm Membrane Cytoplasm

 

 

Con 40 C 2 1 1 0
Con 44 C 2 l l 0
Con 49 C 2 O 1 0
Con 42 N 2 1 1 1
Con 50 N 3 1 1 0
Sul 43 C 2 l 2 0
Sul 46 C 2 1 1 1
Sul 47 C 2 1 1 1
Sul40 N 1 0 1 O
Sul 50 N 3 1 l 1
Ant 45 C 3 1 2 1
Ant 48 C 2 1 l 0
Ant 49 C 3 2 1 1
Ant 40 N 1 O 1 0
Ant 50 N 3 1 1 0
Middle Small Intestine
Con 40 C O 0 0 0
"Con 44 C 1 O 1 0
Con 49 C 2 O 1 0
Con 42 N 2 0 1 0
Con 50 N 2 1 1 0
Sul 43 C 2 1 1 0
Sul 46 C 2 1 1 O
Sul 47 C 2 l 1 0
Sul 40 N 1 O 1 O
Sul 50 N 2 1 l 0
Ant 45 C 3 2 2 1
Ant 48 C 3 2 2 1
Ant 49 C 2 1 2 1
Ant 40 N 2 1 2 0
Ant 50 N 3 2 2 0

 

175

 

Distal Small Intestine

 

Con 40
Con 44
Con 49
Con 42
Con 50
Sul 43
Sul 46
Sul 47
Sul 40
Sul 50
Ant 45
Ant 48
Ant 49
Ant 4O
Ant 50 1
aAnimal ID: Con; control, Ant; anthocyanins, Sul; sulindac

bCarrier status: C; carrier (AchMin ), N; normal (Apcm )

cTop; villus region in the top of the small intestine, Bottom; crypt region in the
bottom of the small intestine.

pip—OH
r—IN

>
Z
;>
Z
>
>

ZZOOOZZOOOZZOOO
HNHND—‘b—‘NHb—OI—i

WN-kwwa-DLHNUJNZNNUJ
OOOOOOb—‘OOOOZOOO

Nt—ot—tt—INI—IWNF—‘NH

176

Table A9. Intensity scores of B-catenin expression in colon of mice from experiment
IH.

 

 

 

11)“ Status“ Topc Middle“ Bottomc
M C M C M C
Con 40 C 2 1 1 O 1 0
Con 44 C 3 2 1 0 1 1
Con 49 C 3 2 2 1 2 1
Con 42 N NA NA NA NA NA NA
Con 50 N 3 2 1 0 1 0
Sul 43 C 3 1 2 1 2 1
Sul 46 C 3 1 2 0 2 0
Sul 47 C 3 2 1 1 1 1
Sul 40 N 2 1 0 0 0 0
Sul 50 N 3 1 1 1 1 1
Ant 45 C 2 1 0 1 1 1
Ant 48 C 4 2 2 2 -2 1
Ant 49 C 4 3 1 1 2 1
Ant 40 N 2 1 1 0 1 0
Ant 50 N 3 1 2 1 2 1

 

aAnimal ID: Con; control, Ant; anthocyanins, Sul; sulindac

bCarrier status: C; carrier (Apc+/M‘" ), N; normal (Apcm )

cTop; top one third of the crypts, Middle; middle one third of the crypts, Bottom;
bottom one third of the crypts in the colon, M; membrane, C; cytoplasm

177

Table A10. Intensity scores of B-catenin expression in cecum of mice from
experiment III.

 

 

 

ID“ StatusB Topc Middlec Bottomc
M C M C M C
Con 40 C 3 1 1 O 1 0
Con 44 C 3 2 0 0 l 0
Con 49 C 4 2 1 0 0 0
Con 42 N 2 1 O O O 0
Con 50 N 3 2 2 1 O 0
Sul 43 C 3 1 2 O 1 0
Sul 46 C 2 1 0 O O O
Sul 47 C 3 1 1 O 1 O
Sul 40 N NA NA NA NA NA Na
Sul 50 N 3 1 1 O 0 0
Ant 45 C 3 2 2 O 2 0
Ant 48 C 3 2 1 0 l 1
Ant 49 C 4 2 2 0 1 0
Ant 40 N 4 1 2 0 1 0
Ant 50 N 3 2 2 1 0 O

 

alAnimal ID: Con; control, Ant; anthocyanins, Sul; sulindac

bCarrier status: C; carrier (ApcWlin ), N; normal (Apcm )

°Top; top one third of the crypts, Middle; middle one third of the crypts, Bottom;
bottom one third of the crypts in the cecum, M; membrane, C; cyt0plasm

178

LITERATURE CITED

179

American Cancer Society (2000) Cancer Facts and Figures-1997. ACS.

Aramburu J, Rao A and Klee CB (2000) Calcineurin: from structure to function.
Curr. Top. Cell Regul. 362237—95.

Baker, SM, Harris, AC, Tsao, J -L, Falth, TJ, Bronner, E, Gordon, M, Shibata, D and
Lisky, RM (1998) Enhanced intestinal adenomatous polyp formation in PmsZ"';Min
mice. Cancer Res. 58:1087-1089.

Baldwin, AS (1996) The NF-KB and IKB proteins: new discoveries and insights.
Annu. Rev. Irnmunol. 14: 649-683.

Barnes, CJ and Lee, M (1998) Chemoprevention of spontaneous intestinal adenomas
in the adenomatous polyposis coli Min mouse model with aspirin. Gastroenterology
114: 873-877.

Berenblum, I and Shubik, P (1947) The role of croton oil applications, associated
with a single painting of a carcinogen, in tumor induction of the mouse skin. Br. J.
Cancer 1: 379-382.

Bilger, A, Shoemaker, AR, Gould, KA and Dove, WF (1996) Manipulation of the
mouse germline in the study of Min-induced neoplasia. Sem. Cancer Biol. 7: 249-
260.

Boolbol, SK, Dannenberg, A, Chadbum, C, Martucci, X, Guo, JT, Ramonetti, M,
Abreu-Goris, HL, Newrnark, ML, Lipkin, DeCosse, JJ and Bertagnolli, MM (1996)
Cyclooxygenase-2 overexpression and tumor formation are blocked by sulindac in a
murine model of familial adenomatous polyposis. Cancer Res. 56:2556-2560.

Brenner, C, Cadiou, H, Vieira, HL, Zamzabi, N, Marzo, I, Xie, Z, Leber, B, Andrews,
D, Duclohier, H, Reed, JC and Kroemer, G (2000) Bcl-2 and Bax regulate the
~channel activity of the mitochondrial adenine nucleotide translocator. Oncogene 19:
329-336.

Cademi, G., Bianchini, F, Dolara, P and Kriebel, D (1991) Starchy foods and colon
proliferation in mice. Nutr. Cancer 15: 33-40.

Cademi, G., Bianchini, F, Mancina, A, Spagnesi, MT and Dolara, P (1991) Effect of
dietary carbohydrates on the growth of dysplastic crypt foci in the colon of rats
treated with 1,2-dimethylhydrazine. Cancer Res. 51: 3721-3725.

Caragay, AB. (1992) Cancer-preventive foods and ingredients. Food Tech. 46:65-
68.

180

 

Caramori, G, Lim, S, Ito, K, Tomita, K, Oates, T, J azrawi, E, Chung, KF, Bames, PJ
and Adcock, IM (2001) Expression of GATA family of transcription factors in T-
cells, monocytes and bronchial biopsies. Eur. Respir. J. 182466-73.

Charron, F, Tsirniklis, G, Arcand, M, Robitailla, L, Liang, Q, Molkentin, JD,
Meloche, S, Nemer, MT (2001) Tissue-speciﬁc GATA factors are transcriptional
effectors of the small GTPase RhoA. Genes Dev. 15:2702-2719.

Chiu, C-H, McEntee, MF and Whelan, J (1997) Sulindac caused rapid regression of
preexisting tumors in Min/+ mice independent of prostaglandin biosynthesis. Cancer
Res. 57: 4267-4273,

Chomczynski, P and Sacchi, N (1987) Single-step method for RNA isolation by acid
guanidium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162:156-159.

Clark, A, Nomura, A, Mohanty, S and Firtel, RA (1997) A ubiquitin-conjugating
enzyme is essential for developmental transitions in Dictyostelium. Mol. Biol. Cell
8: 1989—2002.

Cooke, E.-L., Uings, LJ, Xia, CL, Woo, P and Ray, KP (2001) Functional analysis
of the interleukin-1 B-stimulated nuclear factor KB (NF-KB) pathway activation:

IPAK-l associates with the NF-KB essential modulator (NEMO) upon receptor
stimulation. Biochem. J. 359: 403-410.

Cormier, RT, Hong, KH, Halberg, RB, Hawkins, TL, Richardson, P, Mulherkar, R,
Dove, WP and Lander, ES (1997) Secretory phospholipase Pla2g2a confers
resistance to intestinal tumorigenesis. Nat. Genet. 17: 88-91.

Crawford, HC, Fingleton, BM, Rudolph-Owen, LA, Heppner, KJ, Rubinfeld, B,
Polakis, P and Matrisian, LM (1999) The metalloproteinase matrilysin is a target of
B-catenin transactivation in intestinal tumors. Oncogene 18:2883-2891.

Crompton, M (2000) Mitochondrial intermembrane junctional complexes and their
role in cell death. J. Physiol. 529: 11-21.

Cyert, MS (2001) Genetic analysis of calmodulin and its targets in Saccharomyces
cerevisiae. Annu. Rev. Genet. 35:647-672.

Das, S and Maitra, U (2000) Mutational analysis of mammalian translation initiation
factor 5 (eIFS): role of interaction between the [3 subunit of eIF2 and eIFS in eIF
function in vitro and in vivo. Mol. Cell Biol. 20: 3942-3950.

Doll, R (1990) An overview of the epidemiological evidence linking diet and cancer.
Proc. Nutr. Soc. 49: 119-131.

181

Duggan, DJ, Bittner, M, Chen, Y, Meltzer, P and Trent, JM (1999) Expression
proﬁling using cDNA microarrys. Nature Gene Suppl. 21:10-14.

Easwaran, V, Song, V, Polakis, P and Byers, S (1999) The ubiquitin-proteasome
pathway and serine kinase activity modulate adenomatous polyposis coli protein-

mediated regulation of beta-catenin-lymphocyte enhancer-binding factor signaling. 1 .
Bio. Chem. 274: 16641-16645.

Eklof—Spink, K, Fridman, SG and Weis, WI (2001) Molecular mechanisms of beta-
catenin recognition by adenomatous polyposis coli revealed by the structure of an
APC-beta-catenin complex. EMBO J. 20: 6203-6212.

Elliot. (1992) Inhibition of glutathione reductase by ﬂavonoids. A structure-activity
study. Biochem. Pharrnacol. 8:1603-8.

Estelles, A, Yokoyama, M, Nothias, F, Vincent, JD, Glowinski, J, Vernier, P,
Chneiweiss, H (1996) The major astrocytic phosphoprotein PEA-15 is encoded by
two mRNAs conserved on their full length in mouse and human. J. Biol. Chem. 271:
14800-14806.

Fearon, ER and Vogelstein, B (1990) A genetic model for colorectal tumori genesis.
Cell 61: 759-767.

Feliciello, A, Cardone, L, Garbi, C, Ginsberg, MD, Varrone, S, Rubin, CS,
Avvedimento, EV and Gottesman, ME (1999) Yotiao protein, a ligand for the

NMDA receptor, binds and targets CAMP-dependent protein kinase 11(1). FEBS Lett.
464:174-178.

Fodde, R, Edelmann, W, Yang, K, van Leeuwen, C, Carlson, C, Renault, B, Breukel,
C, Alt, E, Lipkin, M, Khan, PM and Kucherlapati, R (1994) A targeted chain-
termination mutation in the mouse Apc gene results in multiple intestinal tumors.
Proc. Natl. Acad. Sci. USA 91: 8969-8973.

Formstecher, E, Ramos, JW, Fauquet, M, Calderwood, DA, Hsieh, J C, Canton, B,
Nguyen, XT, Bamier, JV, Carnonis, J, Ginsberg, MH, and Cheiweiss, H (2001)
PEA-15 mediates cytoplasmic sequestration of ERK MAP kinase. Dev. Cell 1: 239-
250.

Fujimori, M, Ikeda, S, Shimizu, Y, Okajima, M and Asahara, T (2001)
Accumulation of B-catenin protein and mutations in exon 3 of B-catenin gene in

gastrointestinal carcinoid tumor. Cancer Res. 61 :6656-6659.

Furukawa, K and Sato, T (1999) Beta-1,4-ga1actosylation of N-glycans is a complex
process. Biochem. Biophy. Acta. 1473: 54-66.

182

Giardiello, FM, Krush, AJ, Peterson, GM, Booker, SV, Kerr, M, Tong, LL and
Hamilton, SR (1994) Phenotypic variability of familial adenomatous polyposis in 11

unrelated families with identical APC gene mutation. Gastroenterology 106: 1542-
1547..

Giovannucci, E, Egan, KM, Hunter, DJ, Starnpfer, MJ, Colditz, GA, Willett, WC and
Speizer, FE (1995) Asprin and the risk of colorectal cancer in women. N. Engl. J.
Med. 333: 609-614.

Grubben, MJ, Nagengast, FM, Katan, MB and Peters, WH (2001) The glutathione
biotransforrnation system and colorectal cancer risk in humans. Scand. J.
Gastroenterol. Suppl. 234: 68-76.

Hagiwara, A, Miyashita, K, Nakanishi, T, Sano, M, Tamano, S, Kadota, T, Koda, T,
Nakamura, M, Irnaida, K, Ito, N and Shirai, T (2001) Pronounced inhibition by a
natural anthocyanins, purple corn color, of 2-amino-1-methyl-6-phenylimidazo[4,5-
b]pyridine (PhIP)—associated colorectal carcinogenesis in male F344 rats pretreated
with 1,2-dimethylhydrazine. Cancer Lett. 171: 17-25.

Halestrap, AP, Doran, E, Gillespie, JP and O’Toole, A (2000) Mitochondria and cell
death. Biochem. Soc. Trans. 28: 170-177.

Han, S, Arvai, AS, Clancy, SB and Tainer, J A (2001) Crystal structure and novel
recognition motif of rho ADP-ribosylating C3 exoenzyme from Clostridium

botulinum: structural insights for recognition speciﬁcity and catalysis. J. Mol. Biol.
305:95-107.

Harbome, J. B. (1988) Flavonoids: Recent Advances. In: Plant Pigments, Goodwin,
T. ed.; Academic Press; London, 299-343.

Harbome, JB and Grayer, RJ (1988) The anthocyanins, in the ﬂavonoids (Harbome,
J B ed.) pp.1—20, Chapman and Hall, London.

Hartwell, LH and Kastan, MB (1994) Cell cycle control and cancer. Science 266:
1821-1828.

He, T-C, Chan, TA, Vogelstein, B and Kinzler, KW (1999) PPAR8 is an APC-
regulated target of nonsteroidal anti-inﬂammatory drugs. Cell 99: 33 5-345.

He, T-C, Sparks, AB, Rago, C, Herrneking, H, Zawel, L, Costa, LT, Morin, PJ,

Vogelatein, B and Kinzler, KW (1998) Identiﬁcation of c-MYC as a target of the
APC pathway. Science 281: 1509-1512.

183

'I‘.""*n- .

 

Hegde, P, Qi, Rong, Abernathy, K, Gay, C, Dharap, S, Gaspard, R, Earle-Hughes, J,
Snesrud, E, Lee, N and Quackenbush, J (2000) A concise guide to cDNA
microarray analysis 11. Biotechniques 29: 548-562.

Hulsken, J, Birchmeier, W and Bchrens, J. (1994) E-cadherin and APC compete for
the interaction with beta-catenin and the cytoskeleton. J. Cell. Biol. 127: 2061-2069.

IARC (2001) GLOBOCAN 2000: Cancer Incidence, Mortality and Prevalence
Worldwide, Version 1.0, CancerBase No. 5, Lyon, IARCPress.

Ilyas, M and Tomlinson, [PM (1997) The interactions of APC, E-Cadherin and B-
catenin in tumor development and progression. J. Pathol. 182: 128-137.

Ishitani, T, Ninomiya—Tsuj i, J, Nagai, S, Nishita, M, Meneghini, M, Barker, N,
Waterman, M, Bowerman, B, Clevers, H, Shibuya, H and Matsumoto K. (1999) The
TAKl-NLK-MAPK-related pathway antagonizes signalling between beta-catenin
and transcription factor TCF. Nature 399: 798-802.

Ito, K, Kinjo, K, Nakazato, T, Ikeda, Y, Kizaki, M (2002) Expression and sequence
analyses of p33(ING1) gene in myeloid leukemia. Am. J. Hematol 69: 141-143.

J acoby, RF, Marshall, DJ, Newton, MA, Navarouic, K, Tutsch, K, Cole, CE, Lubet,
RA, Kelloff, GJ, Verma, A, Moser, AR and Dove, WF (1996) Chemoprevention of
spontaneous intestinal adenomas in the APC Min mouse model by the nonsteroidal
anti-inﬂammatory drug proxicam. Cancer Res. 56: 710-714.

Kataoka, N, Bachorik, J L and Dreyfuss, G (1999) Transportin-SR, a nuclear import
receptor for SR proteins. J. Cell Biol. 145:1145-1152.

Kennedy, BP, Payette, P, Mudgett, J, Vadas, P, Pruzanski, W, Kwan, M, Tang, C,
Rancourt, DE and Cromlish, WA (1995) A natural disruption of the secretory group
II phospholipase A2 gene in inbred mouse strains. J. Biol. Chem. 270:22378-22385.

Kim, J, Kim, JH, Lee, SH, Kim, DH, Kang, HY, Bae, SH, Pan, ZQ and Seo, YS
(2002) The novel human helicase hFDH is an F-box protein. J. Biol. Chem. In press.

King, A and Young, G (1999) Characteristics and occurrence of phenolic
phytochemicals. J. Am. Diet. Assoc. 99: 213-218.

Kinzler, KW, Nilbert, MC, Vogelstein, B, Bryan, TM, Levy, DB, Smith, KJ,
Preisinger, AC, Hamilton, SR, Hedge, P and Markham, A (1991) Identiﬁcation of a

gene located at chromosome 5q21 that is mutated in colorectal cancers. Science
251:1366-1370.

184

Kinzler, AR and Vogelstein, B (1996) Lessons from hereditary colorectal cancer.
Cell 87:159-170.

Kitagawa, M, Hatakeyama, S, Shirane, M, Matsumoto, M, Ishida, N, Hattori, K,
Nakamichi, I, Kikuchi, A, Nakayarna, K and Nakayama, K (1999) An F-box protein,
FWDl, mediates ubiquitin-dependent proteolysis of beta-catenin. EMBO J. 18:
2401-2410.

Kennedy, AR, Beazer-Barclay, Y, Kinzler, KW and Newbeme, PM (1996)
Suppression of carcinogenesis in the intestines of Min mice by soybean-derived
Bowman-Birk inhibitor. Cancer Res. 56:679-682.

Koide, T, Hashimoto, Y, Kamei, H, Kojima, T, Hasegawa, M, Terabe, K (1997)
Antitumor effect of anthocyanins extracted from red soybeans and red beans in vitro
and in vivo. Cancer Biother. Radiopharm. 12: 277-280.

Koide, T, Kamei, Y, Hashimoto, T, Kojima, M, Hasegawa, M (1996) Antitumor
effect of hydrolyzed anthocyanins from grape rinds and red rice. Cancer Biother.
Radiopharm. 11: 173-177.

Kristiansen, E, Meyer, 0 and Thorup, I (1996) Reﬁned carbohydrate enhancement
of aberrant crypt foci induced by the foodbome carcinogen 2-amino-3-methyl-
imidazo[4,5-f]quinoline (IQ). Cancer Lett. 105: 147-151.

Lai, MC, Lin, RI and Tarn, WY (2001) Transportin-SR2 mediates nuclear import of
phosphorylated SR proteins. PNAS 98:10154-10159.

Labayle, D, Fischer, D, Vielh, P, Drouhin, F, Pariente, A, Bories, C, Duhamel, O,
Trousset, M and Attali, P (1991) Sulindac causes regression of rectal polyps in
familial adenomatous polyposis. Gastroenterology 101: 635-639.

Lackie, JM and Dow, JAT (1999) The dictionary of cell and molecular biology
(third edition), Academic Press, London

Lal, G, Ash, C, Hay, K, Redston, M, Kwong, E, Hancock, B, Mak, T, Kargman, S,
Evans, JF and Gallinger, S (2001) Suppression of intestinal neoplasia mice by a

speciﬁc cyclooxygenase 2 inhibitor and by a dual cyclooxygenase 1/2 inhibitor.
Cancer Res. 61 :6131-6.

Labarca, C and Paigen, K (1980) A simple, rapid, and sensitive DNA assay
procedure. Anal. Biochem. 102: 344-352.

Leister, I, Weith, A, Bruderlein, S, Cziepluch, C, Kangwangpong, D, Schlag, P and

Schwab, M (1990) Human colorectal cancer: high frequency of deletions at
chromosome 1p35. Cancer Res. 50: 7232-7235.

185

Leppert, M, Burt, R, Hughes, JP, Samowitz, W, Nakamura, Y, Woodward, S,
Gardner, E, Lalouel, JM and White, R (1990) Genetic analysis of an inherited
predisposition to colon cancer in a family with a variable number of adenomatous
polyps. N. Engl. J. Med. 322: 904-908.

Levi, S, Goodland, RA, Lee, CY, Stamp, G, Walport, MJ, Wright, NA, Hodgson, HJ
(1990) Inhibitory effect of nonsteroidal anti-inﬂammatory drugs on mucosal cell
proliferation associated with gastric ulcer healing. Lancet 336: 840-843.

Levy, DB, Smith, KJ, Beazer-Barclay, Y, Harniltom, SR, Vogelstein, B and Kinzler,
KW (1994) Inactivation of both APC alleles in human and mouse tumors. Cancer
Res. 54: 5953-5958.

Linderoth, NA, Popowicz, A and Sastry, S (2000) Identiﬁcation of the peptide-
binding site in the heat shock chaperone/tumor rejection antigen gp96(Grp94). J.
Biochem. Chem. 25: 5472-5477.

Lipken, M, Blattner, E, Fraumeni, JP, Lynch, HT, Deschner, E and Winawer, S
(1983) Tritiated thymidine labeling distribution as a marker for the identiﬁcation of
hereditary predisposition to colon cancer. Cancer Res. 43: 1899-1904.

Liu, CH, Chang, SH, Narko, K, Trifan, OC, Wu, MT, Smith, E, Haudenschild, C,
Lane, TF and Hela, T (2001) Overexpression of cyclooxygenase 2 is sufﬁcient to
induce tumorigenesis in transgenic mice. J. Biol. Chem. 276: 18563-18569.

Liu, Y and Steinacker, JM (2001) Changes in skeletal muscle heat Shock proteins:
pathological signiﬁcance. Front. Biosci. 6: D12-D25.

Luongo, C, Moser, AR, Gledhill, S and Dove, WF ( 1994) Loss of Apc+ in intestinal
adenomas from Min mice. Cancer Res. 54: 5947-5952.

Martinsson, T, Weith, A, Cziepluch, C and Schwab, M (1989) Chromosome 1
deletions in human neuroblastoma: generation and ﬁne mapping of microclones from
the distal 1p region. Genes Chromosom. Cancer 1: 67-78.

Mahmoud, NN, Carothers, AM, Grunberger, D, Bilinski, RT, Churchill, MR,
Martucci, C, Newmark, HL and Bertagnolli, MM (2000) Plant phenolics decrease
intestinal tumors in an animal model of familial adenomatous polyposis.
Carcinogenesis 21: 921-927.

Mas, T, Susperregui, J, Berke, B, Cheze, C, Moreau, S, Nuhrich, A and Vercauteren,
J (2000) DNA triplex stabilization property of natural anthocyanins.
Phytochemistry 53: 679-687.

186

Maurer-Stroh, S, Eisenhaber, B and Eisenhaber, F (2002) N-terrninal N-

myristoylation of proteins: prediction of substrate proteins from amino acid sequence.
J. Mol. Biol. 317:541-557.

Maximo, V, Soares, P, Lima, J, Carneselle-Teijeiro, J and Sobrinho—Simoes, M
(2002) Mitochondrial DNA somatic mutations (point mutations and large deletions)
and mitochondrial DNA variants in human thyroid pathology: a study with emphasis
on hurthle cell tumors. Am. J. Pathol. 160: 1857-1865.

McAllister-Lucas, LM, Inohara, N, Lucas, PC, Ruland, J, Benito, A, Li, Q, Chen, S,
Chen, FF, Yamaoka, S, Verma, 1M, Mak, TW and Nunez, G (2001) Bimpl , a
MAGUK family member linking protein kinase C activation to BcllO-mediated NF-
kappaB induction. J. Biol. Chem. 276:30589-30597.

McCracken, JD, Wechter, WJ, Liu, Y, chase, RL, Kantoci, D, Murray, D, Quiggles,
DD and Mineyama, Y (1996) Antiproliferative effects of the enantiomers of
ﬂurbiprofen. J. Clin. Pharrnacol. 36: 540-545.

McEntee, MF, Chiu, C-H and Whelan, J (1999) Relationship of B-catenin and Bc1-2
expression to sulindac-induced regression of intestinal tumors in Min mice.
Carcinogenesis 20: 635-640.

McPhee, M, Chepenik, KP, Liddell, RA, Nelson, KK, Siracusa, LD and Buchberg,
AM (1995) The secretory phospholipase A2 gene is a candidate for the Mom 1 locus,
a major modiﬁer of ApcM‘"- induced intestinal neoplasia. Cell 81: 95 7-966.

Meiers, S (2001) The anthocyanidins cyanidin and delphinidin are potent inhibitors
of the epidermal growth-factor receptor. J. Ag. Food Chem. 49: 95 8-62.

Miller, EC and Miller, JA (1981) Mechanisms of chemical carcinogenesis. Cancer
47:1055-1064.

Milovica, V, Turchanowa, L, Khomutov, AR, Khomutov, RM, Caspary, WF and
Stein, J (2001) Hydroxylamine-containing inhibitors of polyamine biosynthesis and
impairment of colon cancer cell growth. Biochem. Pharmacol. 61: 199-206.

Miyazawa, T, Nakagawa, K, Kudo, M, Muraishi, K and Someya, K (1999) Direct
intestinal absorption of red fruit anthocyanins, cyanin-3-glucoside and cyanin-3,5-
diglucoside, into rats and humans. J. Agric. Food Chem. 47 :1083-1091.

Mistry, TV, Cai, Y, Lilley, TH and Haslarn, E (1991) Polyphenol interactions, Part 5.
Anthocyanin co-pigrnentation. J. Chem. Soc. Perkin. Trans. 2: 1287-1296.

187

Miyoushi, Y, Nagase, H, Ando, H, Horil, A, Ichii, S, Nakatsuru, S, Aoki, T, Miki, Y,
Tori, T and Nakamura, Y (1992) Somatic mutations of the APC gene in colorectal
tumorszmutation cluster region in the APC gene. Hum. Mol. Genet. 12229-233.

Moolgavkar, SH and Knudson, AG (1981) Mutation and cancer: A model for human
carcinogenesis. J. Natl. Cancer Inst. 66: 1037-1052.

Morin PJ, Sparks AB, Korinek V, Barker N, Clevers H, Vogelstein B, Kinzler KW

(1999) Activation of beta-catenin-ch signaling in colon cancer by mutations in beta-
catenin or APC. 275: 1787-1790.

Moser, AR, Pitot, HC and Dove, WF (1990) A dominant mutation that predispose to
multiple intestinal neoplasia in the mouse. Science 247: 322-324.

Munemitsu, S, Souza, B, Muller, O, Albert, 1, Rubinfeld, B and Polakis, P (1994)
The APC gene product associates with microtubules in vivo and promotes their
assembly in vitro. Cancer Res. 54: 3676-3681.

Munemitsu, S, Albert, 1, Souza, B, Rubinfeld, B and polakis, P (1995) Regulation of
intracellular B-catenin levels by the adenomatous polyposis coli (APC) tumor
suppressor protein. Proc. Natl. Acad. Sci. USA 92: 3046-3050.

Muscat, J E, Stellrnan, SD, Wynder, EL (1994) Nonsteroidal anti-inﬂammatory
drugs and colorectal cancer. Cancer 74: 1847-1854.

Mutanen, M, Pajari, A-M and Oikarinen, SI (2000) Beef induces and rye bran
prevents the formation of intestinal polyps in ApcM‘" mice: relation to B-catenin and
PKC isozyrnes. Carcinogenesis 21: 1167-1173.

Nakamura, Y (1993) The role of the adenomatous polyposis coli (APC) gene in
human cancers. Adv. Cancer. Res. 62:65-87.

Nakamura, K, Miyamoto, H, Suzuma, S, Sakarnoto, T, Kawai, G, Yamane, K (2001)
Minimal functional structure of Escherichia coli 4.5 S RNA required for binding to
elongation factor G J. Biol. Chem. 276:22844-22849.

Newell, KJ, Witty, JP, Rodgers, WH and Matrisian, LM (1994) Expression and
localization of matrix-degrading metalloproteinases during colorectal tumori genesis.
Mol. Carcinogenesis 10: 199-206.

Nishida, T, Kaneko, F, Katagawa, M and Yasuda, H (2001) Characterization of a
novel mammalian SUMO-1/Smt3-speciﬁc isopeptidase, a homologue of rat axam,

which is an axin-binding protein promoting beta-catenin degradation. J. Biol. Chem.
276: 39060-39066.

188

h-m

 

Ocke, MC, Bueno-de-Mesquita, I-IB, Feskens, EJ, van-Staveren, WA and Kromhout,
D (1997) Repeated measurements of vegetables, fruits, beta-carotene, and vitamin C
and E in relation to lung cancer. Am. J. Epidem. 145: 358-65.

Olsson, NO, Caignard, A, Martin, MS, Martin, F (1984) Effect of indomethacin on
the grth of colon cancer cells in syngeneic rats. Int. J. Irnmunopharrnacol. 6:329-
334.

Oshima, M, Oshima, H, Kitagawa, K, Kobayashi, M, Itakura, C and Taketo, M
(1995) Loss of APC heterozygosity and abnormal tissue building in nascent
intestinal polyps in mice carrying a truncated Apc gene. PNAS 92: 4482-4486.

Park, BH, Vogelstein, B and Kinzler, KW (2001) Genetic disruption of PPAR5
decreases the tumorigenicity of human colon cancer cells. PNAS 98:2598-2603.

Paul, P, Letteboer, T, Gelbert, L, Groden, J, White, R and C0ppes, MJ (1993)
Identical APC exon 15 mutations result in a variable phenotype in familial
adenomatous polyposis. Hum. Mol. Genet. 2: 925-931.

Paulin, FEM, Campbell, LE, O’Brien, K, Loughlin, J and Proud, CG (2001)
Eukaryotic translation initiation factor 5(eIF5) acts as a classical GTPase-activator
protein. Curr. Biol. 11:55-59.

Paulsen, J E, Elvaas, I-KO, Steffenson, I-L an Alexander, JA (1997) Fish oil derived
concentrate enriched in eicosapentanoic and docosahexanoic acid as ethyl ester

suppresses the formation and growth of intestinal polyps in the Min mouse.
Carcinogenesis 18: 1905-1910.

Pierre, F, Perrin, P, Champ, M, Bomet, F, Meﬂah, K and Menanteau, J (1997)
Short-chain fructo-oligosaccharides reduce the occurrence of colon tumors and
develop gut associated lymphoid tissue in Min mice. Cancer Res. 57: 225-228.

Pisani, P, Parkin, DM and Ferlay, J (1993) Estimates of the worldwide mortality
from eighteen major cancers in 1985. Implications for prevention and projections of
future burden. Int. J. Cancer 55:891-903.

Pitot, HC (1989) Progression: The terminal stage in carcinogenesis. Jpn. J. Cancer
Res. 80: 599-607.

Pool-Zobel, BL, Bub, A, Schroder, N and Rechkemmer, G (1993) Anthocyanins are
potent antioxidants in model systems but do not reduce endogenous oxidative DNA
damage in human colon cells. Eur. J. Nutr. 38:227-34.

189

Potten, CS (1996) What is an apoptotic index measuring? A commentary. Br. J.
Cancer 74: 1743-1748.

Potten, CS (1997) Epithelial cell growth and differentiation. 11. Intestinal apoptosis.
Am. J. Physiol. 273: G253-G257.

Potter, JD, Slattery, ML, Bostick, RM and Gapstur, SM (1993) Colon cancer: A
review of the epidemiology. Epidemiol. Rev. 15: 499-545.

Prescott, SM and White, RL (1996) Self—promotion? Intimate connections between
APC and prostaglandin H synthase-2. Cell 87:783-786.

Ramirez-Tortosa, C, Anderson, OM, Gardner, PT, Morrice, PC, Wood, SG, Duthie, 51
SJ, Collins, AR, Duthie, GG (2001) Anthocyanin-rich extract decreases indices of

lipid peroxidation and DNA damage in vitamin E-depleted rats. Free Radic. Biol.

Med. 31:1033-1037. “"

 

Reed, RC and Nicchitta, CV (2000) Chaperone-mediated cross-priming: a
hitchhiker’s guide to vesicle transport (review). Int. J. Mol. Med. 6: 259-264.

Reeves, PG., Nielsen, F H and Fahey, Jr., G.C (1993) AIN-93 puriﬁed diets for
laboratory rodents: ﬁnal report of the American Institute of Nutrition ad hoc writing
committee on the reformulation of the AIN-76A rodent diet. J. Nutr. 123: 1939-
195 1 .

Renner, HW (1990) In vivo eﬂects of single or combined dietary antimutagens on
mutagen-induced chromosomal aberrations. Mutat. Res. 244: 185-188.

Rice-Evans, CA, Miller, NJ, Bolwell, PG, Brarnley, PM and Pridham, JB (1995)
The relative antioxidant activities of plant-derived polyphenolic ﬂavonoids. Free
Radical Res. 22:3785-93.

Rigau, J, Pique, J, Rubio, E, Planas, R, Tarrech, JM, Bordas, JM (1991) Effects of
long-terrn sulindac therapy on colonic polyposis. Ann. Intern. Med. 115: 952-955.

Ritland, SR and Gendler, SJ (1999) Chemoprevention of intestinal adenomas in the
ApcMin mouse by piroxicam: kinetics, strain effects and resistance to
Chemoprevention. Carcinogenesis 20:51-58.

Rocheleau, CE, Yasuda, J, Shin, TH, Lin, R, Sawa, H, Okano, H, Priess, JR, Davis,
RJ and Mello, CC (1999) WRM-l activates the LIT-1 protein kinase to transduce
anterior/posterior polarity signals in C. elegans. Cell 97: 717-726.

 

Rosin-Arbesfeld, R, Townsley, F and Bienz, M (2000) The APC tumor suppressor
has a nuclear export function. Nature 406:1009-1012.

190

Roy, HK, Karoski, WJ, Ratashak, A and Smyrl, TC (2001) Chemoprevention of
intestinal tumorigenesis by nabumetone: induction of apoptosis and Bel-2
downregulation. Br. J. Cancer 18:1412-1416.

Rubinfeld B, Souza, B, Albert, 1, Muller, O, Chamberlain, SH, Masiarz, FR,
Munemitsu, S, Polakis P (1993) Association of the APC gene product with beta-
catenin. Science 10:1731-1734.

Rubinfeld, B, Albert, 1, Porﬁri, E, Fiol, C, Munemitsu, S and Polakis, P (1996)

Binding of GSK 313 to the APC-B-catenin complex and regulation of complex
assembly. Science 272: 1023-1026.

Salic, A, Lee, E, Mayer, L, Kirschner, MW (2000) Control of beta-catenin stability:

reconstitution of the cytoplasmic steps of the wnt pathway in Xenopus egg extracts.
Mol. Cell 5: 523-532.

Sarrna, AD and Sharma, R (1999) Anthocyanin-DNA copigrnentation complex:
mutual protection against oxidative damage. Phytochemistry 52: 1313-1318.

Sasai, H, Masaki, M and Wakitani, K (2000) Suppression of polypo genesis in a new

mouse strain with a truncated ApcA474 by a novel COX-2 inhibitor, J TE-522.
Carcinogenesis 21:953-958.

Seeram, N. P., Momin, R. A., Bourquin, L. D. and Nair, M. G. (2001)
Cyclooxygenase inhibitory and antioxidant cyanidin glycosides from cherries and
berries. Phytomedicine, 8: 362-369.

Seeram, NP, Bourquin, LD and Nair, MG. (2001) Degradation products of cyanidin
glycosides from tart cherries and their bioactivities. J. Agric. Food Chem. 4924924-
4929.

Schmelz, EM, Roberts, PC, Kustin, EM, Lemonnier, LA, Sullards, C, Dillehay, DL
and Merill Jr, AH (2001) Modulation of intracellular B-catenin localization and

intestinal tumorigenesis in vivo and in vitro by sphingolipids. Cancer Res. 61 :6723-
6729.

Sheng, H, Shao, J, Williams, CS, Pereira, MA, Taketo, MM, Oshima, M, Reynolds,
AB, Washington, MK, DuBois, RN and Beauchamp, RD (1998) Nuclear

translocation of B-catenin in hereditary and carcinogen-induced intestinal adenomas.
Carcinogenesis 19: 543-549.

Shoemaker, AR, Moser, AR and Dove, WF (1995) N-Ethyl-N-nitosourea treatment

of multiple intestinal neoplasia (Min) mice: Age-related effects on the formation of
intestinal adenomas, cystic crypts, and epiderrnoid cycts. Cancer Res. 55:4479-4485.

191

Shiff, SJ, Qiao, L, Tsai, LL and Rigas, B (1995) Sulindac sulﬁde, an aspirin-like
compound, inhibits proliferation, causes cell cycle quiescence, and induces apoptosis
in HT-29 colon adenocarcinoma cells. J. Clin. Invest. 96:491-503.

Skinner, SA, Penney, AG, O’Brien, PE (1991) Sulindac inhibits the rate of growth
and appearance of colon tumors in the rat. Arch. Surg. 126: 1094-1096.

Smith, KJ, Levy, DB, Maupin, P, Pollard TD, Vogelstein, B and Kinzler, KW
(1994) Wild-type but not mutant APC associates with the microtubule cytoskeleton.
Cancer Res. 54:3672-3675.

Sorenson, IL, Kristiansen, E, Mortensen, A, Nicolaisen, GM, Wijnands, J AH, van
Kranen, H], van Kreijl, CF (1998) The effect of soy isoﬂavones on the development
of intestinal neoplasia in ApcMin mouse. Cancer Lett. 130: 217-225.

Speigelman, VS, Tang, W, Katoh, M, Slaga, TJ and Fuchs, SY (2002) Inhibition of
HOS expression and activities by Wnt pathway. Oncogene 21: 856-860.

Steffensen, I-L, Paulsen, JE, Eide, TJ and Alexander, J (1997) 2-amino-1-methyl-6-
phenylimidazo[4,5-b]pyridine increase the numbers of tumors, cystic crypts and
aberrant crypt foci in multiple intestinal neoplsia mice. Carcinogenesis 18:1049-
1054.

Strange, RC and Fryer, AA (1999) Chapter 19. The glutathione S-transferase:
inﬂuence of polymorphism on cancer susceptibility. IARC Sci. Pub]. 148: 231-249.

Su, LK, Kinzler, KW, Vogelstein, B, Preisinger, AC, Moser, AR, Luongo, C, Gould,
KA and Dove, WF (1992) Multiple intestinal neoplasia caused by a mutation in the
murine homolog of the APC gene. Science 256: 668-670.

Su, LK, Johnson KA, Smith, KJ, Hill, DE, Vogelstein, B and Kinzler, KW (1993a)
Association between wild type and mutant APC gene products. Cancer Res. 53:
2728-2731 .

Su, LK, Vogelstein, B and Kinzler, KW (1993b) Association of the APC tumor
suppressor protein with catenins. Science 262: 1734-1737.

Sugauma, M, Ohkura, Y, Okabe, S and Fujiki, H (2001) Combination cancer

chemoprevention with green tea and sulindac shown in intestinal tumor formation in
Min mice. J. Cancer Res. Clin. Oncol. 127:69-72.

Takeichi, M (1988) The cadherins: cell-cell adhesion molecules controlling animal

morphogenesis. Development 102:639-655.

192

Tanaka, TS, J aradat, SA, Lim, MK, Kargul, GJ, Wang, X, Grahovac, MJ, Pantano, S,
Sano, Y, Piao, Y, Nagaraja, R, Doi, H, Wood, WH, Becker, KG and Ko, MS (2000)
Genome—wide expression proﬁling of mid-gestation placenta and embryo using a
15,000 mouse developmental cDNA microarray. PNAS 97: 9127-9132.

Taya, S, Yamamoto, T, Kanai-Azuma, M, Wood, SA and Kaibuchi, K (1999) The
deubiquitinating enzyme Farn interacts with and stabilizes beta—catenin. Genes.
Cells. 12: 757-767.

Tetsu, O and McCormick, F (1999) B-catenin regulates expression of cyclin D 1 in
colon carcinoma cells. Nature 398:422-426.

Terpstra TO, van Blankenstein, M, Dees J, Eilers, GAM (1987) Abnormal pattern of
cell proliferation in the entire colonic mucosa of patients with colon adenoma or
cancer. Gastroenterology 92: 704—708.

Torrance, CJ, Jackson, PE, Montgomery, E, Kinzler, KW, Vogelstein, B, Wissner, A,
Nunes, M, Frost, P and Discafani, CM (2000) Combinatorial chemoprevention of
intestinal neoplasia. Nat. Med. 6:1024-1028.

Tsuda, T, Horio, F and Osawa, T (1999) Absorption and metabolism of cyanidin 3-
O-B-D-glucoside in rats. FEBS Lett. 449: 179-182.

Tsuda, T, Ohshima, K, Kawakishi, S and Osawa, T (1996) Oxidation products of
cyanidin 3-O-B-D-glucoside with a free radical initiator. Lipids 31: 1259-1263.

Tsuda T. (2000) The role of anthocyanins as an antioxidant under oxidative stress in
rats. Biofactors 13:133-9.

Tsuda T. (1998) Dietary cyanidin 3-O-beta-D-glucoside increases ex vivo oxidative
resistance of serum in rats. Lipids 33:583-8.

Van Kranen, HJ, van Iersel, PWC, Rijnkels, JM, Beems, DB, Alink, GM, van Kreijl,
CF (1998) Effects of dietary fat and a vegetable-fruit mixture on the development of
intestinal ne0plasia in the APC“ mouse. Carcinogenesis 19: 1597-1601.

Verheyen, EM, Mirkovic, I, MacLean, SJ, Langrnann, C, Andrews, BC and
MacKinnon, C (2001) The tissue polarity gene nemo carries out multiple roles in

patterning during Drosophila development. Mech. Dev. 101: 119-132.

Vogelstein, B and Kinzler, KW (1993) The multistep nature of cancer. Trends
Genet. 9: 138-139.

Wallace, HM and Claslake, R (2001) Polyarnines and colon cancer. Eur. J.
Gastroenterol. Hepatol. 13: 1033-1039.

193

Wang, H, Nair, MG, Iezzoni, AF, Strasburg, GM, Booren, AM and Gray J I (1997)
Quantiﬁcation and characterization of anthocyanins in Balaton tart cherries. J. Agric.
Food Chem. 45: 2556-2560.

Wang, H, Nair, MG, Strasburg, GM, Booren, AM, and Gray, J I (1999a) Novel
antioxidant compounds ﬁ'om tart cherries (Prunus cerasus). J Nat. Prod. 62: 86-88.

Wang, H, Nair, MG, Strasburg, GM, Chang, Y.-C. Booren, AM. Gray, J I and DeWitt,
DL (1999b) Antioxidant and anti-inﬂammatory activities of anthocyanins and their
aglycone, cyanidin, from tart cherries. J. Nat. Prod. 62: 294-296.

Weiser, MM (1973) Intestinal epithelial cell surface membrane glycoprotein
synthesis. 1. An indicator of cellular differentiation. J. Biol. Chem. 248:2536-2541.

Westphal, RS, Tavalin, SJ, Lin, JW, Alto, NM, Fraser, ID, Langeberg, LK, Sheng, M
and Scott, JD (1999) Regulation of NMDA receptors by an associated phosphatase-
kinase signaling complex. Science 285: 93-96.

WCRF/AICR (1998) Food, Nutrition and the prevention of cancer: A global
perspective.

Wilde, C, Chhatwal, GS, Schmalzing, G, Arktories, K and Just, I (2001) A novel
C3-like ADP-ribosyltransferase from Staphylococcus aureus modifying RhoE and
Rnd3. J. Biol. Chem. 276:9537-9542.

Willet, WC (2001) Diet and cancer: one view at the start of the millennium. Cancer
Epidemiol. Biomarkers Prev. 10: 3-8.

Williarns-Adams, HG, Jana, J, Coppocc, GL, Geroch, ME and Sherone, AC (1972)
New aspects of polyamine biosynthesis in eukaryotic organisms. Advances in
Enzyme Regulation 10: 225-245.

Whitehead, IP, Zohn, IE and Der, CJ (2001) Rho GTPase-dependent transformation
by G protein-coupled receptors. Oncogene 20: 1547-1555.

Worrnhoudt, LW, Commandeur, JN and Vermeulen, NP (1999) Genetic
polymorphisms of human N-acetyltransferase, cytochrome P450, glutathione-S-

transferase, and epoxide hydrolase enzymes: relevance to xenobiotic metabolism and
toxicity. Crit. Rev. Toxicol. 29: 59-124.

Wright, NA and Irwin, M (1982) The kinetics of villus cell populations in the mouse

small intestine: normal villi-the steady state requirement. Cell Tissue Kinet. 15:595-
609.

194

Yamaguchi, T, Matsuda, K, Sagiya, Y, Iwadate, M, Fujino, MA, Nakamura, Y and
Arakawa, H (2001) p53R2-dependent pathway for DNA synthesis in a p53-
regulated dell cycle checkpoint. Cancer Res. 61 :8256-8262.

Zhang, T, Nanney, LB, Luongo, C, Lamps, L, Heppner, KJ, DuBois, RN and
Beauchamp, RD (1997) Concurrent overexpression of cyclin D1 and cyclin-
dependent kinase 4(cdk 4) in intestinal adenomas from multiple intestinal neoplasia

(Min) mice and human familial adenomatous polyposis patients. Cancer Res. 57:
169-175.

195

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

1.11111lllljlllllllljllllljljlll