5...... .. 355:1
gum... . . rhmmuamu. .
0- .u V . .

“A
v n
2.. V
{K
V . I 1
£3.25:

. . . $5“.an #3

M9. .1
V c
. 1.4g

..
RN

. z
x .
, 5&an
. h.
. . . .1 m
_‘y S“ V V . V . V
V .ﬁ 1.... . ,
Sr. .45.. ,
y :n

|.-->ER
v-‘I.

F in V
c‘-..v h V . ”ITV 5‘ a;
urn i. 31,. , . . .t
. “Vain... V
V p. I
V... . .

innit;
. a
r ’V
3..

13K“

, «a: 5.

v?‘ «3»... 1......
. a ‘1

Va... 1.

VSV. $1.2... . . V3-..
.. Y . u! . t _ . 973‘
. .VVVV VVVVVV . .VV... .VVVV
, .. . :- , 1&3. , 3......
. .. . . . 3...»...
V? .V. . . . VFW... V... .
V ﬁnnvaVJ§althV .. , . . V V , V". . .5 ﬁring». 1
V. V. . . . 1
. . . . v... V . a ,
5,353.?” .

.3"
2.: JV! V .2 2 .

...

. . a“..- . V
_ . . . . gm...

z is Enuhr.
ind...» . .V. u........5...w...%
. ._ V V;- . 7
: 53 3V... in} . maﬁa...
.9... .. .. , LE6.» .. . . 3V). 3.
Vininimvcm GA» an”). .5“ ”vamV. anvmaﬁlmeﬁ‘. .
3:53.11..- F 9? .
«:1. at»... . Qﬁvm
hr.- .. «w .. ﬂinch}.
V? . .ﬁVrVw. ..
nVVWt 831:5!
. I v. at
. .\ .Iﬂtlﬁlnuofumnhw.) \«ﬂnk:u\.
.E

rpéﬂur‘.
.. Y .. .
5 V.

 

 

 

«a... draw;

 

 

THEsss
l

V: CC)

This is to certify that the
thesis entitled

The Influence of Fermentable Fibers on Intestinal
Growth and Tumorigenesis in Min+ Mice

presented by

Elizabeth Ann Rondini

has been accepted towards fulﬁllment
of the requirements for

Masters degree in Science

Major professor

Date 12V” 7’00

0-7 639 MS U is an Afﬁrmative Action/Equal Opportunity Institution

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

11m WM

 

THE INFLUENCE OF FERMENTABLE FIBERS ON INTESTINAL GROWTH AND
TUMORIGENESIS IN Min+ MICE

By

Elizabeth Ann Rondini

A THESIS

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

MASTER OF SCIENCE
Department of Food Science and Human Nutrition

2000

ABSTRACT

THE INFLUENCE OF FERMENTABLE FIBERS ON INTESTINAL GROWTH
AND TUMORIGENESIS IN Min“ MICE

BY

ELIZABETH ANN RONDINI

Fermentation of dietary ﬁber and starch in the colon produces short-chain fatty
acids (SCFA), namely acetate, propionate, and butyrate. Butyrate has anti-neoplastic
eﬂ°ects in vitro. This study examined the effects of the fermentable dietary ﬁbers
ﬁ'uctooligosaccharides (F OS), modiﬁed starch, and gum arabic on intestinal tumors and
tissue parameters in Min+ mice, a murine model of familial adenomous polyposis (F AP)
in humans. Min+ mice (4 weeks old) were separated into one of ﬁve treatment groups
and fed a powdered AIN-93G diet with 15% fat and 6% of a dietary ﬁber or an inert bulk
source. The dietary ﬁber/bulk sources used were 1) cellulose, 2) F08, 3) modiﬁed
starch, 4) gum arabic, and 5) kaolin After 10 weeks of treatment, the animals were
sacriﬁced and the colon and small intestine (SI) were removed, measured, and then ﬁxed.
The cecum was emptied and then ﬁxed and weighed. The number and size of adenomas
in each tissue were recorded, and short chain fatty acids (SCFA) were quantiﬁed in cecal
contents. The highly fermentable ﬁbers (groups 2, 3, and 4) produced signiﬁcantly
greater cecum weights and SI and colon lengths, and had signiﬁcantly greater
concentrations of SCFA compared to the less fermentable cellulose and kaolin. There
were no signiﬁcant diﬁ‘erences (P < 0.05) between treatment groups for small intestinal
(71/mouse), colon (l/mouse), or cecal (O/mouse) tumors. It was concluded tlmt
fermentable ﬁbers are trophic to the small and large intestines and cecum through SCFA

preduction, but do not inﬂuence tumorigenesis in Min+ mice.

ACKNOWLEDGEMENTS

I would like to thank my major advisor, Dr. Bennink, for his help, support, and
guidance. I would also like to extend my gratitude to my committee members. To Dr.
Chenoweth for her continued support of my academic achievements, and Dr. Bourquin,
for giving me the opportunity to do this experiment.

I would also like to thank my family, especially my parents, for their love, faith,
and encouragement. Also for their support, both emotionally and ﬁnancially, which has

made my education both possible and meaningful.

TABLE OF CONTENTS

LIST OF TABLES ...................................................................
LIST OF FIGURES ..................................................................
INTRODUCTION ...................................................................
REVIEW OF LITERATURE ......................................................
A. EPIDEMIOLOGY OF COLON CANCER ..................................
A.1 Incidence of colon cancer .................................................
A.2 Diet and colon cancer ......................................................
B. MOLECULAR BASIS OF COLON CANCER ............................
B.1 Physiology of the colon ...................................................
8.2 Genetic alterations in colon cancer ......................................
B.3 Inherited predispositions to colon cancer ...............................
B.4 Cellular ﬁmctions of APC .................................................
B.4.1 Interactions between E-cadherins, B-catenin, and APC .......
3.5 Animal model of colon cancer ............................................
C. DIETARY FIBER ...............................................................
C.l Components of dietary ﬁber ...............................................
C.2 Non-fermentable ﬁber ......................................................
C.3 Fermentable ﬁbers ........................................................
C.3.1 Resistant Starch .....................................................

C.3.2 Oligosaccharides ....................................................

iv

vii

1

3

3

3

3

4

4

5

7

8

10

13

14

15

l6

16

16

17

C.3.3 Gums ................................................................. 18

D. FERMENTATION OF DIETARY FIBER .................................. 18
DJ Products of anaerobic fermentation in the colon ....................... 19

D. 1 .1 Gas production ..................................................... 21

D.1.2 Short chain fatty acids ............................................. 21

D2 SCFA absorption and utilization ........................................ 22
D.3 Clinical signiﬁcance of SCFA ........................................... 25
D4 Trophic effects of SCFA on gut morphology .......................... 26
D.4.1 Trophic effects of SCFA in the cecum and colon ............. 26

D42 Trophic eﬁ‘ects of SCFA in the small intestine ................ 27

E. FERMENTATION AND COLON CANCER ................................ 27
El Alterations in bile acid metabolism ....................................... 27
E2 Alterations in ammonia metabolism ..................................... 29
E.3 Direct actions of butyrate .................................................. 30

F. FERMENTATION AND EXPERIMENTAL COLON

CARCINOGENESIS ............................................................ 32
G. RATIONALE .................................................................... 34
H. OBJECTIVES .................................................................... 36
1. NULL HYPOTI-[ESIS ............................................................ 36
MATERIALS AND METHODS .................................................. 37

Animal Care ....................................................................... 37

Diets ................................................................................ 37

Experimental Design ........................................................... 37

Scoring of adenomas ............................................................. 39

SCFA analysis .................................................................... 40
Statistical analysis ............................................................... 40
RESULTS .............................................................................. 41
Mouse body weights .............................................................. 42
Tumor numbers .................................................................... 44
Tissue parameters ................................................................. 45
SCFA analysis .................................................................... 46
DISCUSSION ......................................................................... 49
SUMMARY AND CONCLUSIONS ............................................. 53
RECOMMENDATIONS FOR FUTURE RESEARCH ........................ 54
LITERATURE CITED .............................................................. 56

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

Table 6.

LIST OF TABLES

Dietary ﬁber polysaccharides. Adapted from Phillips 6.0.,
1998 .........................................................................

Dietary composition of the modiﬁed AIN—93G diets. The
compositions are presented as g/ 100 g diet ............................

Small intestinal, colonic, and cecal tumor numbers in Min+ mice
fed either a ﬁber source or inert bulk source (6g/kg diet) .............

Mean tissue lengths (mm) in the small intestine and colon and
cecal wet weights (g) in Min+ and Min' mice fed diﬁ’erent sources
of ﬁber or an inert bulk source (6 g/ 100 g diet) ........................

Millimolar (umol/g) concentrations of individual SCFA
SCF A per gram cecal contents in Min+ and Min' mice fed diﬁ‘erent
sources of ﬁber or an inert bulk (6g/ 100g diet) .........................

pH of cecal contents and molar proportions of SCF A in Min+ and
Min" mice fed different sources of ﬁber or an inert bulk (6g/ 100g
diet) ...........................................................................

15

38

45

46

47

48

LIST OF FIGURES

Figure 1. Genetic alterations in colon cancer. Adapted from
Vogelstein et a1, 1988 ....................................................... 6

Figure 2. Potential functional domains of APC. Modiﬁed ﬁom Polakis,
1997. Abbreviations: Min+, multiple intestinal neoplasia; DLG,

discs large protein ........................... 9
Figure 3. End products of anaerobic fermentation in the colon. Adapted
ﬁom Hill, 1995 ............................................................... 20

Figure 4. SCFA absorption from the proximal and distal colon. Modified
from von Engelhardt, 1995 ................................................ 23

Figure 5. Body weight gain of male Min+ and Min' mice during dietary
intervention. There were no signiﬁcant treatment differences
(P < 0.05) at any time point. There was a signiﬁcant effect due to
carrier status (Min+ > Min'), which was signiﬁcant at 6 weeks of
treatment (P < 0.05) ......................................................... 43

Figure 6. Body weight gain of female Min+ and Min' mice during dietary
intervention. There were no signiﬁcant treatment differences
(P < 0.05) at any time point. There was a signiﬁcant effect due to
carrier status (Min+ > Min'), which was signiﬁcant at 8 weeks of
treatment (P < 0.05) ........................................................ 44

I. INTRODUCTION

Colon cancer is currently the third leading cause of cancer-related deaths in the
US. Burkitt proposed in the late 1960s that ﬁber is protective against colon cancer and a
range of gastrointestinal disorders common to Westemized populations (Burkitt, 1971).
This hypothesis stimulated numerous studies over the past 30 years. The results of this
research, however are' conﬂicting and inconclusive. In recent years there have been many
developments that have led to a better understanding of the genetic mechanisms of colon
cancer.

Colon cancer develops through a sequence of mutations (V ogelstein et al, 1988).
The APC gene is mutated in the inherited condition familial adenomatous polyposis
(F AP) in hurmns and is also one of the earliest, and most ﬁequent, mutations thought to
occur in the adenoma-carcinoma sequence. The protein encoded by the APC gene is
important for regulation of cell proliferation. Min+ (multiple intestinal neoplasia) mice,
the murine homologue of PAP in humans, have a fully penetrant germline mutation in the
APC gene and develop numerous benign tumors throughout the intestine (Su et al, 1992).
Therefore, the Min+ mouse may be a reasonable in vivo model to study the effects of
metabolites that can possibly alter gene transcription and inﬂuence the promotion stage of
colon cancer development.

Bacterial fermentation of dietary ﬁber in the colon results in the production of
short-chain fatty acids (SCFA), primarily acetate, propionate, and butyrate. Butyrate has
been shown to induce differentiation in various colon cancer cell lines rendering the cells
non-cancerous. The exact mechanism whereby butyrate induces differentiation is not

ﬁilly known, but may involve histone hyperacetylation and p21 up—regulation (McBain et

a1, 1997; Hague et a1, 1993). In contrast to their possible anti-carcinogenic effects in
vitro, SCF A are also trophic to the small intestine and colon. Butyrate is the primary
energy source for colonocytes and locally stimulates cellular proliferation. SCFA, after
absorption in the colon, have also been shown to be trophic to the small intestine, likely
through a systemically-mediated mechanism. Because SCFA stimulate proliferation in
normal tissue, it has been suggested that SCF A may also enhance growth of neoplasms
(Jacobs, 1987). Therefore, this study was designed to determine the inﬂuence of

fermentation on intestinal tumorigenesis in Min+mice.

II. REVIEW OF LITERATURE

A. EPIDEMIOLOGY OF COLON CANCER
A.l Incidence of colon cancer

Cancer is currently the second leading cause of death in the United States
(Greenlee et al, 2000). Of the different types of cancer, colorectal cancer accounts for
11% of all cancer deaths, with an expected incidence of 93,800 new cases in the US in
2000 (Greenlee et al, 2000). Approximately 5% of colon cancer cases have a well-
deﬁned genetic pattern whereas 80-90% of cases occur sporadically (Burt, 1995).
Environmental factors identiﬁed in the etiology of colon cancer include diet, inadequate

physical activity, obesity, and ulcerative colitis (Shike, 1999).

A.2 Diet and colon cancer

Dietary and environmental factors are thought to contribute to 80-90% of all
cancers (Drasar and Hill, 1972). High intakes of dietary fat, especially from animals, and
low intakes of vegetable and cereal ﬁbers, vitamin D, and calcium are associated with an
increased risk of colon cancer development and may account for much of the worldwide
variation in colon cancer incidence (Macrae, 1999). Burkitt proposed in the late 19605
that the high ﬁber diets common in Aﬁ'ican populations are associated with lower colon
cancer incidences (Burkitt, 1971). Epidemiological studies examining dietary ﬁber
intake and colon cancer incidence, however, have not been conclusive (reviewed in
Bingham, 1997). Case-control studies consistently show an inverse association between

vegetable and ﬁ'uit intake and colon cancer risk. The inverse relationship between

vegetable ﬁber intake and colon cancer risk is strongest for raw, green, and cruciferous
vegetables (Potter, 1999). Bingham sumrmrized thirty case-control studies that
examined the relationship between dietary ﬁber intake and colon cancer risk from 1969-
89 (Bingham, 1990). Eleven of 22 studies showed a reduced risk associated with
increased intakes of ﬁber, 12/ 19 for vegetable ﬁber intake, and 3/13 for cereal ﬁber
consumption (Bingham, 1990). Recently published case-control studies examining the
effect of wheat bran supplementation on either colon cancer risk or adenoma recurrence
have also been inconclusive. Two showed no effect on adenoma recurrence, one showed
no effect on adenoma recurrence but a decrease in adenoma size with wheat bran, and
one showed wheat bran was protective only for rectal polyp development (Alberts et al,
2000; Schatzkin et al, 2000; Macrae, 1999). In short, no strong conclusions between

dietary ﬁber intake and colon cancer can be made ﬁ'om epidemiological studies.

B. MOLECULAR BASIS OF COLON CANCER:
B.l. Physiology of the Colon:

The human colon is approximately 150 cm (5 ft.) in length with an overall surface
area of 1300 cm2 (Eastwood, 1983; Cummings and Macfarlane, 1991). Histologically,
the colon contains multiple invaginations that extend into the mucosa layer to form crypts
(Kinzler and Vogelstein, 1998). The mucosa is divided into three portions, namely the
epithelium, lamina propria, and muscularis mucosae (Eastwood, 1983). The epithelium
lines the mucosa and is maintained by immature stem cells located in the middle crypt of
the proximal colon and at the base of the crypts in the distal region (Karam, 1999).

Proliferation occurs in the lower 2/3 of the crypts and as cells migrate towards the lumen,

they differentiate into absorptive, goblet, or enteroendocrine cell types (Eastwood, 1993).

After approximately 3-6 days differentiated cells at the top of the crypt are shed into the

lumen (Lipkin et al, 1963).

3.2 Genetic alterations in colon cancer

A balance between cell division and death is important for maintaining the
epithelial lining. The earliest biomarker for colon cancer development is a
hyperproliferative state of the mucosa associated with an expansion of the proliferation
zone to the upper part of the crypt and a concomitant decrease in diﬂ‘erentiation
(Wmawer et al, 1991). Neoplastic progression in colon cancer as well as other cancer
types involves genetic mutations resulting in the cumulative loss of functional tumor
suppressor genes and/or dominant activation of proto-oncogenes (F earon and Vogelstein,
1990). Vogelstein et a1 (1988) have examined the ﬁ'equency of genetic mutations in
various tumors and proposed a sequence of events important for the development of
colon cancer. These include, in order, mutations in adenomatous polyposis coli (APC), k-
ras, deleted in colon cancer (DCC), and p53 genes (Figure 1).

APC, located on chromosome 5q21 in humans, encodes a protein with tumor
suppressive properties (Kinzler et al, 1991). As discussed later, inactivation of both
alleles of the gene is permissive for adenoma forrmtion, suggesting APC acts as a
“gatekeeper” of colonic epithelial proliferation (Powell et al, 1992; Kinzler and
Vogelstein, 1996). The ras prom-oncogenes encode proteins with intrinsic GTPase

activity and are located on chromosome 12p (Kinzler and Vogelstein, 1998). K-ras

 

.32 ._n 8 EoEowo> ..EE~€8&§V .8058 :28 E macaw—23 0:050 A 9:.me

 

 

«80:03 «8983 «80:03 Esmoﬁﬁo Sam—egg“.
$36802 «803980 Bug 2338.283 beam o>tﬂom=8auo§m H.502

 

 

 

3:26 and DDQ 8.3‘ UK“... 050
£85m m8— mmB gauges $2 53832
$50 a: 63 mg cm 08808020

 

mutations generally occur later in adenoma growth and have been detected in 50% of
colorectal adenomas larger than 1 cm (Vogelstein et al, 1988). DCC is a tumor
suppressor gene located on chromosome 18q (Kinzler and Vogelstein, 1998). DCC
mutations generally occur late in tumor progression, and are found in 47% of advanced
adenomas and 73% of carcinomas (V ogelstein et al, 1988). Mutations in the p53 gene
occur late in tumor progression, most likely in the adenoma-carcinoma transition

(V ogelstein et al, 1988). In normal cells, the p53 protein is upregulated in response to
genomic damage, and is associated with cell cycle arrest, either in G1 and/or G2. This

ensures repair of damage and/or leads to apoptosis of rapidly proliferating cells.

B.3 Inherited Predispositions to Colon Cancer

Inherited predispositions to colon cancer are responsible for 5-10% of colon
cancer cases and have led to a better understanding of genetic events which occur in
tumor formation (Rudy and Zdon, 2000). Familial adenomatous polyposis (F AP) is an
autosomal dominantly inherited condition affecting approximately 1 in 7,000 people
(Kinzler and Vogelstein, 1996). Affected individuals inherit a germline mutation in one
of the alleles of the APC gene, increasing their susceptibility to colon cancer development
(N ishisho et al, 1991). The initiating event in tumor formation involves a somatic
mutation in the remaining wild type (WT) allele of the APC gene (Groden et al, 1991).
Individuals with F AP will commonly develop hundreds to thousands of benign adenomas
throughout the colon, usually by 20-30 years of age (Su et al, 1992). If left untreated,

colon cancer will develop by the age of 40 (Rudy and Zdon, 2000).

Mutations in both alleles of the APC gene occur in approximately 60-80% of
sporatic colon tumors, and has been proposed to be the initiating event in the pathway
that leads to colon cancer (V ogelstein et al, 1988; Fearon and Vogelstein, 1990). The
following section will discuss the ﬁmctional domains of APC and potential tumor

suppressive properties.

B.4 Cellular functions of APC

A schematic diagram of the functional domains of the WT APC protein is shown
in Figure 2 (Polakis, 1997). APC is a 2843 base pair gene encoding a multi-ﬁmctional
protein found in a variety of cell types (Shoemaker, 1997). Along the colonic crypt there
exists differences in APC expression, increasing as cells migrate towards the lumen
(Smith et al, 1993). The central region of the protein contains two binding sites for [3-
catenin, consisting of a 15 amino acid (a) repeat region and a 20 a repeat region, the
latter of which contains phosphorylation sites necessary for B-catenin binding and its
down regulation (Shoemaker et al, 1997; Polakis, 1997; Ilyas and Tomlinson, 1997).
Analysis of regions commonly mutated in sporatic colon cancer as well as germ-line
mutations indicate that mutations occur most frequently in codons 1286-1513, named the
mutation cluster region (MCR) (Miyaki et al, 1994, Shoemaker et al, 1997). More than
two-thirds of mutations result in ﬁameshiﬁ, and one-third involve nonsense point
mutations, both which result in premature truncation of the APC protein (Polakis, 1997).

In the human condition'FAP, the severity of colon tumors as well as the presence

of extra-colonic manifestations varies with the location of the mutation (Shoemaker et al,

830.8 0mg mommc .049 ”Ema—Q00: 85388 032:8 HS: ...E0.:0.§0$0V
.33 ammo—30m 80¢ 005002 6%.? 2:083. 1805056 1503.5 .N 0.53m

@0008
200 08:8
52.8 as 8:382 0382
£8800 wig—.0 .8300 0:0 200 0880
wSUSD 01.530.5me wanna §0u$m m C wave—5 aomuﬁtoaoumo
Em 08800.m

 

if}. a ‘1? Li
.m{l_iﬂlnlﬁr«_ .m

3:85 common .8520 803808 0082 8002
GAO 58:1 803082 +33 05008.2.

 

 

 

1997). Congenital hypertrophy of the retinal pigment epithelium occurs in PAP patients
with mutations aﬁer codon 500 (Dobbie et al, 1996). Mutations in APC occurring after
codon 1400 cause an increased frequency of osteomas, epidermoid cysts, and desmoid
tumors, but decreased frequency of colonic tumors (Dobbie et al, 1996; F odde et al,
1999; Olschwang et al, 1993; Shoemaker et al, 1997). Because both germline and
somatic mutations occur in the 5 ' region of APC, the functional regions for B-catenin
degradation and binding of microtubules, EBl, and DLG may all be involved in the
tumor suppressive properties of APC. However, as discussed above, mutations occurring
after the MCR are associated with decreased incidence in colon tumors, and a higher
ﬁ'equency of extra-colonic manifestations (Wallis and Macdonald, 1996). Mutations
occurring prior to the MCR often cause higher tumor multiplicity as seen in germline
mutations. Therefore, the B-catenin binding and degradation sites appear to be most

important for inhibition of colon tumor development.

B.4.l Interaction between E-cadherins, B-catenin, and APC

Cell adhesion and cell-cell communication is important in regulating contact
inhibition between cells, tissue homeostasis, and morphogenesis (Miller and Moon,
1996). This process is mediated by cellular adhesion molecules (CAMS), notably
irnmunoglobulin-type, adherins, and cadherins (Miller and Moon, 1996). E-cadherins,
which are involved in adheren junctions in epithelial cells, are calcium-dependent
transmembrane proteins which homotypically bind adjacent cells (Ilyas et al, 1997). The
interaction of these proteins between cells forms tight cell-cell junctions important in

establishing cell polarity, shape changes, and migration (Peifer, 1997). The intracellular,

10

carboxyterminus domain of E-cadherin binds the 92 kDa protein B-catenin (Drosophilia
homologue Armadillo), which connects E-cadherin to a-catenin, attaching the complex
to the actin cytoskeleton (Yamada and Geiger, 1997; Ilyas et al, 1997). This attachment
is necessary for cadherin ﬁmction, and serves an architectural role for the cell.

Dissociation of the cadherin complex occurs through phosphorylation of the
arnino-terminal binding domain of B-catenin, which displaces it from the cadherin
complex (Ilyas et al, 1997). Several tyrosine kinase receptors have been identiﬁed and
shown to interact directly with membrane-bound B-catenin (Morin et al, 1999). Two
growth factors, epidermal (EGF) and hepatocyte growth factors (HGF) act via the
receptors to stimulate dissociation of the B-catenin/E-cadherin complex (Ilyas et al,
1997). Loss of cadherin ﬁinction either through tyrosine phosphorylation of B-catenin,
mutations in the binding site of the cadherins, or mutations in the N-terminal domain of
B-catenin loosens cell-cell junctions and allows cellular migration. This is an important
process in tissue healing, angiogenesis, and in oncogenic transformation (Ilyas et al,
1997)

Aside from the role in adherens junctions, B-catenin exists in a monomeric (ﬁ'ee)
form in the cytoplasm which, upon stimulation, ﬁmctions in a signal transduction
pathway. Under normal conditions, the cytosolic level of B-catenin is tightly controlled
by APC. Aberrant increases in the cytosolic level of B-catenin induces the activity of
glycogen synthetase kinase 30 (GSK 30), a serine/threonine kinase which phosphorylates
both B-catenin and APC (Orford et al, 1997; Polakis, 1997). Phosphorylation results in

an APC/B-catenin/GSK-3B complex which acts to degrade B-catenin by the ubiquitin-

11

proteosome pathway (Orford et al, 1997). Another protein, axin, serves as a scaffold for
the complex and is necessary for the activity of GSK-3B (Hedgepeth et al, 1999).

In steady-state conditions, B-catenin is thought to be in equilibrium with APC and
cadherins, so low cytoplasmic levels are maintained. Initiation of the Wnt/wingless
signal transduction pathway, which ultimately inhibits GSK 3B, or mutations in the B-
catenin binding site on APC results in stabilization of cytosolic B-catenin (Morin, 1999).
When this occurs, B-catenin can freely translocate into the nucleus, where it interacts
with a family of high mobility group-box (HMG) containing transcription factors (TCF-
1, TCF-2, TCF-4, LEF-l) (Yokoya et al, 1999; Kuhl and Wedlich, 1997). Association
with these proteins forms a ternary complex important for transcription to occur (Morin
et al, 1999). A few of the target genes have been found to be those involved in cell cycle
control and apoptosis, including c-Myc, cyclin D1, c-Jun, and Fra-I (Mann et al, 1999;
Mayer et al, 1999). Cadherins also possess LEF promoters on their transcription domain.

Although B-catenin signaling is important in embroygenesis, there is compelling
evidence that the nuclear localization of B-catenin is also a common event in colon
cancer. Because APC negatively regulates the free cytosolic form of B-catenin, and this
protein is mutated in 60-80% of colon tumors, it has been suggested that its tumor
suppressive activity is through regulation of B—catenin levels in the cell. Supporting this
hypothesis, restoration of wild-type APC to cells deﬁcient in this protein results in
recruitment of B-catenin ﬁ'om the nucleus (Korinek et a], 1997). Also, overexpression of
E-cadherin in cancer cells results in increased cytosolic and membrane-bound B-catenin
and restoration of cell adhesion (Orsulic et al, 1999). In addition, there appears to be a

strong inverse relationship between cancer cell progression and nuclear localization of B-

12

catenin (Hao et al, 1997). Cancer cells located at the invasion ﬁont of tumors posses
more nuclear staining of B—catenin compared to cells in the tumor center which retain a
more membranous expression, similar to norrml epithelium (Brabletz et al, 1998).
Although the regulation of cytosolic B-catenin is likely controlled by several proteins, a
ﬁmctional APC protein appears to be among the most important components in the

cytosol.

B.5 Animal Model of Colon Cancer:

In any animal study, the biological plausibility, or ability to extrapolate ﬁndings
to human conditions, is an important consideration. Relevant to colon cancer, the Min+
(multiple intestinal neoplasia) mouse shares many similarities to the human condition
FAP. Min)“ mice exhibit a ﬁilly penetrant germline mutation on codon 850 in the murine
homolog of the adenomous polyposis coli (APC) gene (Figure 2) (Su et al, 1992). As a
result, mice carrying the mutation develop, on average, 50+ adenomas throughout the
intestinal tract. (Su et al, 1992, Moser et al, 1990). Studies examining intestinal tumor
formation in Mirir mice suggest that tumors are initiated in the ﬁrst few weeks of life, and
tumor multiplicity is established by 67 days of age (Shoemaker et al, 1995). The
initiation of intestinal tumor development is thought to preceed somatic loss of the
remaining wild type APC allele (Oshima et al, 1995). The lifespan of Min+ mice is
relatively short, 119213] days, and the cause of death is commonly severe anemia due to
intestinal bleeding (Moser et al, 1990).

There are many similarities between Min+ mice and PAP patients both at a

genetic level and phenotypicaﬂy. The murine and human APC genes share 86% and 90%

13

similarities at the nucleotide and amino acid levels, respectively (Su et al, 1992). The
Min+ mutation is the result of a TTG-TAG transversion, causing premature truncation of
the APC protein, which is common to mutations observed in humans (Su et al, 1992).
Extracolonic features, including desmoid tumors and epidermoid cysts, which are seen in
some forms of FAP have also been reported in Min+ mice (Shoemaker et al, 1995). The
main difference between the Min+ mutation compared to human FAP is the distribution
of tumors. In humans, most of the tumors develop in the large intestine, whereas in Min+
mice, the majority of tumors are found in the small intestine. Notwithstanding this,
because of the mutational similarities in the APC gene and the impact on tumor
development, the Min+ mouse model may provide insight into chemopreventative agents

for intestinal carcinogenesis.

C. DIETARY FIBER

Dietary ﬁber, as initially deﬁned by Trowell in 1972, is the “skeletal remains of
plant cell walls incapable of digestion by human alimentary enzymes” (Trowell, 1976).
This deﬁnition has been expanded to include a broad range of polysaccharides (storage
polysaccharides and plant gums and mucilages) and lignin (Chesson, 1995). For
purposes of this paper, resistant starches and indigestible oligosaccharides will also be
included in the traditional deﬁnition as dietary ﬁber components. The recommended
intake of dietary ﬁber is currently 25-30 g/day, however in populations consuming
Westemized diets, intake is 11-14 g/day (Greenwald and Lanza, 1986). The main

sources of dietary ﬁber, components, and food types are shown in Table 1.

14

Table 1.

Dietary ﬁber polysaccharides. Adapted from Phillips G.O., 1998.

 

 

Main Soluble in Major Groups Components Foods
Source Water
Storage Partly Starch Amylose Fruits, seeds,
material Amylopectin tubers
Structural No Cellulose All cell walls
material
Yes Non-cellulose Pectic substances Mainly ﬁ'uits
and vegetables
Slightly Hemicelluloses Cereals
Fruits and
vegetables
Non- Yes Mucilages Algal seaweed,
structural Yes Gums exudates, seeds
material and ﬁ'uits

 

C.1 Components of dietary ﬁber:

The types of polysaccharides found in plant cell walls vary with the plant origin

and its maturity (Selvendran and Verne, 1988). A major portion of dietary ﬁber present

in the human diet comes ﬁom parenchymous tissues in ﬁ'uits and vegetables, whereas

wheat bran and other cereal products contribute more of the ligniﬁed tissues (Selvedran

and Verne, 1988). Analytical methods used for determining dietary ﬁber commonly

involve removal of soluble sugars, lipophillic materials, protein, and starch and then

separation with either ethanol or centrifugation to yield soluble and insoluble ﬁactions

(reviewed by Asp, 1995). The sugar moieties vary with the polymer present. Hydrolysis

of cellulose yields glucose, hemicellulose yields xylose and glucose (in dichots), and

15

pectic substances yield uronic acid, galactose, and arabinose (Selvedran and Verne,

1988)

C.2 Non-fermentable Fibers

Insoluble, non- or poorly fermentable ﬁbers present in the diet include lignins,
cellulose, and some hemicelluloses (Johnson and Southgate, 1994). Lignins are
polyphenolic compounds found mainly in mature plants and cereal ﬁbers (Johnson and
Southgate, 1994). Cellulose is a high molecular weight compound composed of linear
polymers of glucose linked by [3 1-4 glycosidic bonds (Johnson and Southgate, 1994).
Hydrogen bonding tightly packs cellulose chains into ﬁbrils, making them relatively
inaccessible to digestion (Johnson and Southgate, 1994). In vitro indigestibility with
human fecal ﬂora for cellulose (Solka Floc) and wheat bran have been previously

reported to be 97-99% and 58-63%, respectively (Ehle et al, 1982).

C.3 Fermentable Fibers:
C.3.l Resistant Starches

Starches consumed in the diet are hydrolyzed in the small intestine by a-amylase.
However, some starches, termed resistant starches, may escape digestion and become
substrates for fermentation in the colon (Cummings and Macfarlane, 1991). Resistant
starch (RS) is classiﬁed into three categories, namely RSI, RSII, and RSIII (Cummings et
al, 1987) Starches in the RSI category are resistant to digestion due to limited amylase
accessibility to the starch molecules because of physical constraints. Examples include

starches ﬁ'om whole or coarsely ground grains and seeds. In native (raw) starch

16

granules (RSII), amylose and amylopectin chains are packed into a crystalline structure,
making them less susceptible to enzymatic degradation (Gallant, 1995). Because cooking
disrupts (gelatinizes) the crystalline structures, only foods consumed in the raw form and
containing RSII are nutritionally signiﬁcant (Englyst and Cummings, 1986). The third
class (RSIH) form as a result of food processing (Cummings and Macfarlane, 1991).
Slow, prolonged cooling of cooked starches allow some of the amylose chains to
crystallize, forming “retrograded starch” If these crystalline, retrograded starch
structures are not gelatinized during the reheating process that typically occurs before
consumption, the retrograded starch will not be digested (McWilliams, 1993).
Approximately 8-40 g/day of resistant starches become available for fermentation in the

colon (Curmnings and Macfarlane, 1991).

C.3.2 Oligosaeeharides

Approximately 2-8 g of oligosaccharides are consumed per day and pass to the
colon for fermentation in populations consuming Western diets (Cummings and
Macfarlane, 1991). One class of oligosaccharides, oligoﬁ'uctoses, are low molecular
weight polymers composed of B-1-2-D ﬁ'uctose units (Roberfroid, 1999). Inulin, (degree
of polymerization (DP) between 10-60) and ﬁ'ucto-oligosaccharides (DP < 10) are found
in large quantities in chicory roots, garlic, leeks, and Jerusalem artichokes (Roberﬁ'oid et
al, 1993; Roberfroid, 1999). Oligoﬁ'uctoses, unlike gums, do not appear to reduce
cholesterol or bile acid absorption in the small intestine, but are rapidly fermented in the
colon (Andersson et al, 1999). Fructo-oligosaccharides (F OS) also have been shown in

vitro to selectively stimulate the growth of Biﬁdobacterium at the expense of bacteriodes,

l7

clostridia, and coliforrns (Gibson and Roberﬁoid, 1995). Due to the perceived beneﬁts of
Biﬁdobacteria, FOS is currently the only oligosaccharide classiﬁed as a “prebiotic” food

(Gibson and Wang, 1994).

C.3.3 Gums

Plant gums are a diverse group of heteroglycans produced by plants under
conditions of stress (Stephen, 1995). They are composed of a variety of large molecular
weight compounds used by the food industry as thickeners and stabilizers (Stephen,
1995). The sugar composition varies between the different types of gums, although most
contain a high proportion of galactose residues (McWilliams, 1993b). Variations in sugar
residues dictate the degree of solubility in water, viscosity, and ability to form gels in
food systems (Stephen, 1995). Among the different types of gums, gum arabic is a large
molecular weight (MW 850,000) polysaccharide composed primarily of rhamnose,
arabinose, glucuronic acid, and galactose residues (Phillips, 1998). Gum arabic is not
digested in the small intestine, but is fermented in the colon. Ross et a1 (1984) reported
that gum arabic was completely fermented, primarily in the cecum of rats (Ross et al,

1984).

D. FERMENTATION OF DIETARY FIBER

Short chain fatty acids (SCFA) are the primary end products of fermentation by
anaerobic bacteria residing in the colon. Relative proportions of the SCFA produced are
estimated to be 60:25:10 for acetate, propionate, and butyrate respectively (Scheppach et

al, 1995). Dietary ﬁber is the main substrate for fermentation, with smaller contributions

18

ﬁom protein, mucin, and epithelial cells (Cummings and Macfarlane, 1991). In recent
years it Ins become evident that the quantity and relative proportions of SCFA can be
manipulated by varying the diet. The signiﬁcance of this and the recognition that SCFA
are the primary metabolic fuels for the colonocyte has led to research on SCFA utilization
in a variety of pathological conditions (Roediger, 1990). Of particular interest is that
butyrate is a differentiating agent in various colon cancer cell lines, leading some to
believe that the fermentation of ﬁber can potentially alter colon cancer risk. The
following will review what is known regarding SCFA production and absorption, and the

possible role of fermentation in colon cancer prevention.

DJ Products of anaerobic fermentation in the colon

There are over 400 species of bacteria residing in the colon, with total colonic
bacteria numbering 10”'12 /g dry feces (Cummings, 1981). Most bacteria are saccrolytic,
and use the Embden-Meyerhof glycolytic pathway for hexose degradation (Cummings,
1981). As shown in Figure 3, the major product of this pathway is pyruvate, which is
further metabolized to produce the three main SCFA, acetate, propionate, and butyrate
(Hill, 1986). Lactate, hydrogen, carbon dioxide, formate, and urethane are also produced
(Miller and Wolin, 1979). The overall equation for the reaction derived from rumen

metabolism is:

58 (CHzO)6 -) 62 CH3COOH + 22 CH3CH2COOH + 16 CH3(CH2)2COOH +

60.5 C02 + 33.5 CH4 + 27 H20 (Miller and Wolin, 1979).

19

 

 

 

 

 

 

.32 £5

 

 

 

 

 

 

 

 

 

 

 

 

..E0a\h&&03\ .0200 05 8 02880888 0308008“ .«0 9.00008 00m .m 0.53.:

3800<

8.25.5
a
<0U-_b00< 88030.5
60...: algae... a! A .8 i
353...!” Y 808025
8300‘— A\J\\1\ a N
E .50

EN

80030

20

D.1.1 Gas Production

Gases (hydrogen, methane, carbon dioxide) produced during fermentation are
rapidly absorbed and expelled through the lungs or excreted as ﬂatus (Cunnnings and
Macfarlane, 1991). Fermentation is the only signiﬁcant source of hydrogen gas in the
body, so breath hydrogen measurements are sometimes used as indices of fermentation.
Hydrogen that is not excreted is ﬁirther metabolized by methanogenic, acetogenic, and
sulphate-reducing bacteria (Cummings and Macfarlane, 1991). Approximately 30-60%
of healthy hmnans favor methane production over H2 excretion and methane levels can
also rise in pathological conditions such as when polyps and colon cancer occur
(Cummings and Macfarlane, 1991; Haines et al, 1977). Lactate production is usually low
in humans, but concentrations increase in malabsorptive conditions. Rapid fermentation
of readily available sugars causes a signiﬁcant decrease in luminal pH, which inhibits

lactate-utilizing enzymes and increases lactate concentrations (Cummings, 1981).

D.1.2 Short Chain Fatty Acids:

As mentioned previously, the primary substrates for fermentation in the colon are
starches that escape digestion in the small intestine and dietary ﬁbers (Cummings and
Macfarlane, 1991). The SCFA produced in the highest quantity is acetate, followed by
propionate and butyrate (Cummings and Macfarlane, 1991). Although fermentation
favors acetate production, propionate and butyrate yields are sensitive to changes in the
diet. Overall, fermentation patterns are inﬂuenced by the composition of the substrate,

transit time, and fermentability (Edwards and Eastwood, 1994; Bergreen et al, 1993).

21

Substrates which favor acetate production include pectins and ﬁbers with low
ferrnentabilities (cellulose and oat husk) (Bergreen et al, 1993; Ehle et al, 1982). Shifts to
propionate production occur with fermentation of gums and resistant starches from wheat
and maize (Bergreen et al, 1993; Cummings et al, 1996). Butyrate concentrations tend to
increase with the amount of starch available. Sheppach et al (1988) showed that
administrating arcarbose, a drug that inhibits starch digestion in the small intestine, to
humans resulted in an increase of fecal butyrate concentrations compared to controls.
Fructo-oligosaccharides, oligosaccharides, and potato starch have also been shown to

favor butyrate production (Campbell et al, 1997; Fergusen et al, 2000).

D.2 SCFA Absorption and utilization

More than 95% of SCFA produced are efﬁciently absorbed by humans, and
absorption is thought to be both concentration-dependent and pH-independent
(Scheppach, 1994; Rechkernmer and von Engelhardt, 1988). The pKa values for acetate,
propionate, and butyrate are 4.75, 4.87, and 4.81 respectively, and therefore at
physiological pH (6.0-7.0) are predominately in the anionic form (Rechkemmer and von
Engelhardt, 1988). SCFA absorption is associated with an increase in bicarbonate
secretion into the lumen and Na+ and Cl+ absorption (Binder and Mehta, 1989).
Although there may be marked species differences, the following model for SCFA
absorption in the guinea-pig colon has been proposed by von Engelhardt and
Rechkemmer (1992), von Engelhardt et al (1994), and extended by von Engelhardt

(1995) (Figure 4).

22

Proximal Colon

Lumen Apical Membrane

 

 

 

 

'1 """" 1'?
i Pr
3
l
SCFA' + H+ (—) SCFAH
i
l
l
k """"" " SCFA‘
HCO'3
Distal Colon
Na+
E K+ \\ {--- H+
l ‘ H* + HCO'3
V . \ '
SCFA' + 11* <—) SCFAH SCFA + W 5
: g
: \\‘_~ I
a IIIII a
‘ --------- -> SCFA' ----- E. SC F A‘
”CO-3 ‘- ............... ”

 

Figure 4. SCFA absorption from the proximal and distal colon. Modified from
von Engelhardt, 1995.

23

In the proximal colon, approximately 30-50% of SCF A are absorbed in the lipid
soluble form and H+ ions are derived ﬁom an active Na+-H' exchange in the apical
membrane. The remainder of the ionized fatty acids are absorbed by a SCFA'-HCO'3
exchange mechanism. Intracellular carbonic anhydrase provides both H+ and HCO'3 ions
necessary for exchanges to occur through the membrane. In the distal colon, lipid
solubility and chain length are thought to be more important than in the proximal colon
for SCF A absorption (Rechkemmer and von Engelhardt, 1988). In the apical membrane,
K+-H* ATPase provides hydrogen ions necessary for luminal protonation of the acids and
for non-ionic diffusion to occur. Ionized forms of SCFA are exchanged for HCO'3 by a
SCFA-HCO'3 transport mchanism. The intracellular pH in both regions of the colon is
thought to be regulated by a Na+-H+ exchange located on the basolateral membrane,
however SCFA transport through the basolateral membrane has not been fully elucidated
(von Engelhardt, 1995).

After absorption, butyrate is primarily used by the colonocyte for B-oxidation
(Roediger, 1982). Utilization of SCFA by the colonocytes is in the order
butyrate>propionate>acetate. Studies in sudden death victims have shown propionate
and acetate travel though the blood stream to the liver, although some may also be used
locally in the colon (Cummings et al, 1987). Propionate acts as a substrate for
gluconeogenesis via its metabolism to succinate. Studies done on the utilization of
propionate in viva have also suggested that it has cholesterol-lowering effects after its
metabolism in the liver, possrbly by inhibiting HMG-reductase. Acetate is likewise

absorbed and used for hepatic ketogenesis and synthesis of phospho lipids, cholesterol,

24

and long-chain fatty acids. Acetate is the only SCFA found in signiﬁcant amounts in
peripheral blood and can be used by skeletal muscles and the heart as an energy source

(Cummings et al, 1987).

0.3 Clinical signiﬁcance of SCFA

SCFA are the main energy source for colonocytes (Roediger, 1982). Deﬁciencies
in the production and/or utilization of SCFA have been implicated in a range of clinical
conditions, emphasizing the importance of these acids for colonic health. Lack of
luminal nutrition, as occurs in famines and starvation, decreases energy provision to the
small intestine and colon (Roediger, 1990). Although systemic circulation supplies
alternate ﬁlC] supplies to the colon, long-term starvation results in decreased oxidative
capacity, decreases in sodium and water absorption, and mucosa] atrophy (Roediger,
1990). This is commonly seen in patients on long-term total parenteral nutrition (TPN)
and has been demonstrated with colonic bypasses in the rat. Diverting the fecal stream
away ﬁom the colon causes mucosa] inﬂammation that is histologically similar to
ulcerative colitis (U C), and which can be partially alleviated with butyrate enemas (Harig
et a], 1989). Long-term TPN also causes mucosa] atrophy to both the small intestine and
colon. Incorporation of SCFA into TPN solutions has been shown to prevent some of the
mucosa] atrophy in the small intestine, but not the colon, and may lead to a better

prognosis in patients with short bowel syndrome (Koruda et al, 1990).

25

D.4 TROPHIC EFFECTS OF FIBER AND SCFA ON GUT MORPHOLOGY
D.4.1 Trophic effects of SCFA in the cecum and colon

SCF A are known trophic agents for epithelial cell growth in the small intestine,
cecum, and colon, however the mechanism differs between tissues (Sakata, 1987; Lupton
and Kurtz, 1993). Fermentation of ﬁber in the cecum and colon both decrease luminal
pH and increase SCFA concentrations. Lupton et a] (1993) showed that a decrease in pH
was highly correlated to epithelial proliferation in the cecum. This resulted in a greater
cecal surface area, and an increase in crypt depth, circumference, and cellularity (Lupton
and Kurtz, 1993; Lupton et a], 1988). In the colon, SCFA act locally to stimulate
epithelial cell growth. Infusion studies strongly suggest that SCFA are the main trophic
agents for proliferation. Sakata et a] (1982) reported that SCFA inﬁised into the colon of
fasted rats rapidly increased epithelial mitotic and labeling indices, possibly by
shortening the G2 phase of the cell cycle. Butyrate produces the greatest
hyperproliferative response, followed by propionate and acetate (Sakata, 1987). The
effect on colonic cell growth is also dose-dependent (Kripke et al, 1989; Sakata, 1987).
Proliferation is conﬁned to the lower 23 of the crypt and is associated with an increase in
crypt depth and cellularity (V elaquez et al, 1996; Lupton and Kurtz, 1993). SCFA also
stimulate colonic and hepatic blood ﬂow. In perfused dog colons, SCFA infusions
increased both blood ﬂow and oxygen uptake to the colon with acetate being the most

effective (Kvietys and Granger, 1981).

26

D.4.2 Trophic effects of SCFA in the small intestine

Luminal nutrition, pancreato-biliary secretions, and hormonal stimuli all maintain
the epithelial lining of the small intestine (Dowling, 1978). SCFA also are known trophic
agents for epithelial growth in the small intestine, most likely through a systemic
mechanism (Thompson et al, 1996). Infusional studies suggest that, unlike in the colon,
acetate and propionate are more trophic to the small intestine (Koruda et a], 1988).
Although the mechanism by which this occurs has not been fully elucidated, possible
SCFA-mediated agents include stimulation of the autonomous nervous system and

release of the hormones gastrin and enteroglucagon (Frankel et a], 1988).

E. FERMENTATION AND COLON CANCER

Fermentable ﬁbers, if protective, most likely modulate colon carcinogenesis
through end-products of fermentation. Aside from the possible beneﬁts of stimulating
the growth of certain strains of bacteria, the production of SCFA lowers the pH of
intestinal contents. This can alter luminal metabolism, including the formation of
secondary bile acids and ammonia (Gibson and Roberfroid, 1995; Fergusen and Harris,
1996). Butyrate, and to a lesser extent propionate, may also act independently to protect
the mucosa against malignant transformation, although these effects have been primarily

studied in vitro.

E.1 Alterations in bile acid metabolism

Populations at high risk for colon cancer development excrete higher levels of bile

acids in the stool (Hill et a], 1971). Approximately 2-5% of bile acids escape

27

reabsorption in the small intestine and are metabolized by bacteria in the colon
(Nagengast et a], 1995). The bacterial enzyme B-glucuronidase hydrolyzes the
glucuronide derivative in the cecum and colon, and bacterial 7-a dehydroxylase is
responsible for converting primary bile acids into secondary acids, namely lithocholic

and deoxycholic acid (DCA) (N agengast et a], 1995). The enzyme 7-a dehydroxylase
becomes increasingly unstable at pH values higher or lower than 7.0, and is inhibited at
pH values below 6.0, resulting in decreased production of secondary bile acids (F ergusen
and Harris, 1996).

Raﬂer et a] (1986) showed in perﬁlsed rat colons that 5 mM deoxycholic acid is
damaging to the colonic epithelium, resulting in a more cuboidal morphology and erosion
of cells ﬁ'om the lamina propria. Decreasing the pH of the perfusate from 7.9 to 5.5 or
increasing the concentration of calcium from 0 to 4 mM lessened the damage to the
colonic surface. Rectal infusion of deoxycholic acid has also been shown to enhance
experimentally induced colon carcinogenesis and is associated with an increase in
proliferating cells in the upper part of the colonic crypt (V elaquez et a], 1996).
Fermentation lowers luminal pH, and could therefore decrease the activity of 7-a
dehydroxylase, resulting in an increased excretion of primary with a subsequent decrease
in secondary bile acids in the stool. Precipitation of bile acids out of the aqueous phase
of the stool or a decreased concentration of secondary bile acids would decrease exposure
and cytotoxicity to the colonocytes.

There is some evidence in humans that addition of dietary ﬁber can alter bile acid
metabolism. Addition of 13-15 g of wheat bran/day for 8 weeks in humans decreased

concentrations of secondary bile acids in the stool and caused a signiﬁcant reduction in

28

fecal B-glucuronidase and 7-a dehydroxylase activities (Reddy, 1992). In another study,
adding 45 g native amylomaize (resistant) starch/day to the diet of healthy vohmteers
resulted in a signiﬁcant increase in primary bile acid excretion and a trend towards lower
excretions of secondary bile acids (Van Munster et al, 1994). Although the stool pH did
not change signiﬁcantly, total soluble bile acids decreased in the aqueous phase of the
stool. Proliferation of colonic crypts also signiﬁcantly decreased during the amylomaize

feeding compared to the control period.

E.2 Alterations in ammonia metabolism

Another mechanism by which SCFA may protect against colon carcinogenesis is
by altering luminal metabolism of ammonia. Based on studies in ileostomy patients,
approximately 2 g/day of nitrogen-containing compounds enter the large intestine,
primarily from endogenous substrates. Levels can also increase with dietary protein
consumption (Cummings and Macfarlane, 1991). Urea from systemic circulation also
supplies ammonia nitrogen after metabolism in the bowel by bacterial urease (Royall et
al). Protein is the major source of nitrogen, and Bacteroides spp, Propionibacterium spp,
Streptococcus, and Clostridium present in the colon all exhibit proteolytic activity
(Curmnings et a], 1987). Proteolysis results in accumulation of SCFA, ammonia, and
phenolic acids in vitro (Cummings et a], 1987).

Ammonia is cytotoxic to colonocytes and stimulates colonocyte proliferation (Lin
and Visek, 1991). Lowering the luminal pH through fermentation has been proposed to
increase the proportion of the less soluble ammonium ion, which can then be used locally

for bacterial synthesis (Visek, 1978). This would result in a decreased exposure of

29

colonocytes to ammonia. Birkett et a] (1996) reported decreased fecal ammonia and
phenol metabolites in the feces of humans fed resistant starch. Scheppach et a] (1988)
also showed that administration of acarbose increased fecal nitrogen excretion in humans,
which was associated with an increase in bacterial mass. Thus, fermentation may protect
against colon carcinogenesis by altering luminal metabolism, decreasing ammonia and
phenol metabolites and attenuating the cytotoxic effects of these nitrogen by-products in

the lumen.

E.3 Direct actions of butyrate

The possible direct link between fermentation of dietary ﬁbers and colon cancer
prevention has been demonstrated most extensively in vitro, and more consistently for
butyrate than acetate or propionate. Butyrate, at physiological concentrations of 1-5
mM, has been shown to induce differentiation and apoptosis, modify gene expression,
and alter chromatin ultrastructure (Smith, 1986; Barnard and Warwick, 1993; Whitehead
et al, 1986). This has been shown to occur in various colon cancer cell lines and other
cell types, including white blood cells and breast cancer cells. There is, however, limited
in viva evidence to support a direct effect of butyrate on colon carcinogenesis.

In the colon, cells near the upper third of the’crypt lose the ability to proliferate
and gain cell-speciﬁc differentiation markers which dictates the phenotype of progeny
cells. A characteristic of many colon cancer cell types is the loss of cell cycle control,
resulting in uncontrolled proliferation and a decrease in differentiation. Butyrate induces
cell cycle arrest early in the G1 phase of the cell cycle (Barnard and Warwick, 1993).

Barnard et a1 (1993) showed that growth arrest was rapidly reversible and associated with

30

a decrease in c-myc protonco gene expression and an increased expression of alkaline
phosphatase. Whitehead et a] (1986) showed in LIM1215 colon cancer cells that butyrate
treatment resulted in an increase in doubling time ﬁ'om 26 to 72 hours and suppression of
cloning eﬂiciency. The effects on cell grth were associated with an increase in
diﬂ'erentiation markers, as butyrate-treated cells had increased activities of alkaline
phosphatase, CEA, and dipeptidyl amino peptidase over controls.

Inhibition of apoptosis is thought to occur early in colon carcinogenesis, and is
associated with loss of APC protein functioning Morin et al, 1996). Butyrate has been
shown in several colon cancer cell lines to induce apoptosis. Heerdt et a1 (1997) showed
that butyrate, but not isobutyrate or heptoﬂuorobutyric acid derivatives, induced
apoptosis in the carcinoma cell lines HT 29 and SW260. Hague et a] (1995) conﬁrmed
these results in both adenoma and carcinoma colon cancer cell lines. However, 2 of the 4
carcinoma cell lines were shown to be more resistant to butyrate-induced apoptosis.

The mechanisms by which butyrate can induce apoptosis and/or differentiation
are not fully known. Histone hyperacetylation, which alters DNA accessibility to
endonucleases and repair proteins, can be altered with butyrate treatment in vitro (Smith
et al, 1986). The mechanism appears to involve direct inhibition of deacetylase enzymes,
primarily through a histone H4 acetylation (McBain et al, 1997). McBain et a] (1997)
showed that the time required for butyrate-induced apoptotic death correlated to the time
required for tertracetylation of histone H4. They also reported that carboxylic acids 3-5
carbons in length were more effective in inducing hyperacetylation. Other possible

mechanisms involving butyrate-induced apoptosis include upregulation of the p21 cell

31

cycle inhibitor and downregulation of the epidermal grth factor receptor (EGF)

(Hague et a], 1993; Archer et al, 1998).

F. Fermentation and experimental colon carcinogenesis

Colon cancer is induced in rodent models by administration of carcinogens (DMH
or AOM) either subcutaneously, intraperitoneally, or by gavage. Depending on the time
of carcinogen exposure, one can examine the effect of diet during initiation (before
carcinogen) or promotion (after carcinogen exposure). Differences in dosage, type, and
route of administration, however, cause many inconsistencies between study designs
(Whiteley and Klurfeld, 2000). End-point measurements used to assess the effect of diet
on colon cancer development in this model include abberant crypt foci (ACF) and/or
number of colon tumors. Another model for colon cancer studies is the Min+ mouse
model, as discussed previously.

There is limited evidence to suggest that fermentable ﬁbers are protective against
experimentally induced colon carcinogenesis. However, proposed mechanisms involved
in chemoprevention vary widely. In carcinogen-induced colon cancer, oat bran, corn
bran, and guar gum do not appear to inﬂuence tumor development, whereas wheat bran is
more consistently associated with a protective effect (Zoran et a], 1997; Compher et al,
1999; McIntyre et al, 1993). In one of these studies, there was a negative association
found between intraluminal butyrate concentrations and tumor mass regardless of the
dietary treatment (McIntyre et al, 1993). Reddy et a1(1997) reported that the chicory
ﬁ'uctans, oligoﬁ'uctose and inulin, when fed at 10% of the diet, inhibited ACF

precancerous lesions and crypt multiplicity in AOM-induced carcinogenesis. Both potato

32

and resistant starches also have been shown to inhibit the development of ACF in rats
when fed during the promotion phase of tumorigenesis. The protective effect of resistant
starch was associated with an increase in intraluminal butyrate concentrations over
controls (Thorup et a], 1995; Cassand et al, 1997). Using Min+ mice as a model, one
study reported that FOS (6% of diet) inhibited colon tumor incidence and was associated
with an increase in gut associated lymphoid tissue development in this group (Pierre et al,
1997). However, in two studies using resistant starch and one study using wheat bran, no
protective effect on colon tumor incidence was reported (Pierre et al, 1997). Thus, inulin,
F OS, and wheat bran appear to consistently protect against colon carcinogenesis, whereas
there are inconsistencies between models for the effect of resistant starches on colon

carcinogenesis.

33

III. RATIONALE:

The addition of ﬁber to the diet can inﬂuence fermentation patterns in the colon,
and the concentration of butyrate is particularly sensitive to the type and amount of ﬁber
in the diet (Rombeau and Kripke, 1990). Dietary ﬁber includes a heterogenous group of
compounds with varying physiochemical properties. Therefore, it becomes increasingly
difﬁcult to correlate ﬁber per se with colon cancer risk. Examination of fermentation
patterns, at the site of production, allows more quantitative associations to be made and
can give insight into possible mechanistic roles of fermentation metabolites in colon
cancer prevention. The types of fermentable ﬁbers used in the present study were fructo-
oligosaccharides (F OS), modiﬁed starch, and gum arabic. FOS yields high amounts of
butyrate, and resistant starch yields either high amounts of butyrate or propionate
(Campbell et al, 1997; Cummings et al, 1996; Berggren et a], 1993). Gum arabic, a plant
extrudate, yields higher propionate at the expense of acetate (Annison et al, 1995;
Titgemeyer et a], 1991). The controls used in this study include cellulose, which is
minimally fermentable in the cecum and colon, and kaolin, an inert bulking agent that is
non-fermentable. The two bulk sources were included in the diet to minimize muscular
atrophy in the small and large intestines.

Min+ mice were chosen as the model in this study because, similar to humans
with F AP, they have a genetic predisposition to colon cancer. However, colon cancer
rarely occurs in Min+ mice due to their short life span (120 i 31 days), making it
necessary to study dietary effects at earlier stages of adenomatous growth. In humans,

about 70-90% of colon cancers arise from adenomatous polyps, and the larger the polyp,

34

the more likely it is for cancer to develop (Rudy and Zdon, 2000). Therefore,
interventions that result in reduced tumor burden (decreased number and/or reduced

polyp size) are more likely to delay the progression to cancer.

35

IV. OBJECTIVES

The primary objective in this study was to determine if the concentrations and
patterns of SCFA in the cecum of mice are alterable by varying the ﬁber content in the
diet. The second objective was to determine whether fermentation metabolites decrease
intestinal tumor incidence and size in Min+ mice. Certain ﬁber sources and SCFA
produced by fermentation in the large bowel have previously shown trophic effects
throughout the intestinal tract. Therefore, the third objective was to differentiate trophic
effects of SCF A ﬁom unfermented bulk by measuring intestine lengths and cecal

weights.

V. NULL HYPOTHESES

The null hypotheses to be tested are: 1) fermentation does not protect against tumor
development in the intestinal tract of Min+ mice; 2) SCFA amounts or types in the cecum
do not differ among treatment groups; and 3) fermentable and non-fermentable ﬁber

sources are not trophic to the small intestine, cecum, or colon.

36

VI. MATERIALS AND METHODS

Animal Care: C57BL/6J Min+ and Min' mice, aged 3-4 weeks, were obtained
from a colony maintained at Michigan State University. Mice were not genotyped for the
APC mutation prior to the experiment, but were divided into groups under the assumption
that halfthe offspring would be Min +. The animals (n=128) were grouped by litter, sex,
and weight and randomly assigned to treatment groups. Mice were housed in plastic
cages (2-3 mice/cage) in temperature (23°C :1: 2°) and humidity (40-60%) controlled
rooms with a 12 hour off/on light cycle. All mice had continual access to food and water
and were observed daily for health status. This study was approved by the Michigan State
University All-University Committee on Animal Use and Care.

Diets: Mice were fed a powdered version of the AIN-93G diet with slight
modiﬁcations, however, the nutrientzenergy ratio was similar for all diets and comparable
to the AIN-93G diet (Reeves, 1993). All diets contained 15% fat (wt/wt) as soybean oil
and 6% (wt/wt) of either a ﬁber source or inert bulk (Table 2). LoDex 15 was used as the
cornstarch source to minimize resistant starch reaching the colon for fermentation.
Cellulose (Solka Floc) and kaolin, an inert bulk, were used as the controls.

Experimental Design: Mice began nutritional intervention at 4 weeks of age.
Due to the relatively short life span of these mice (120 i 31 days), they were fed

experimental diets for a total of 10 weeks. At 10 weeks post-treatment, the animals were

37

8.82 8.8. 0.82 O82 Q82 ".33
8.8 882
8.8 282 .80

88 88m 8282

88 $08 8888888585

88 282.232 82800

888 888 8.8.8 88.8 88.8 8%: 0888888925

 

8.8 88 8.8 88 8.8 $880 22.3 9885 0880
8.8 88 8.8 8.0 8.8 820.2
8.2 8.2 8.2 8.2 8.2 82 5.825 x2032?
88 22 88 88 88 82 3.556 22-08-22

8.2 8.2 8.2 8.2 8.2 3808
8.; 8.; 8.2. 8.3. 8.: £05.88 888638 2 593
8.2 8.2 8.2 8.2 8.2 @0888 22m: =0 82:8
2.8 2.8 2.8 2.8 2.8 8228 22: 588
2588
93.5, 225 @822
2.82 .56 8282 mom 8282 .5828—

 

.68 w ootw mm @8533 0.8 22282228 23. .2.on 03-22 vogue”: 25.20 2033383 529 .N 033—.

38

sacriﬁced by C02 inhalation and exsangunatation around the same time of day (6-10
am.) to control for diurnal variation in epithelial cell kinetics. The small intestine and
colon were excised and rinsed with lukewarm tap water. Small intestinal and colon
lengths were measured to the nearest mm under ﬁxed tension (8 g). The small intestines
were then cut into thirds, opened, and rinsed with tap water. After excision of the cecum,
the contents were immediately removed, weighed, and kept at -80° C until analysis of
SCF A could be performed. All tissues were placed in a phosphate-buffered saline
solution (PBS, pH 7.4) immediately following excision, pinned ﬂat on cardboard, and
then ﬁxed in a 10% neutral-buffered formalin solution (pH 7.4) overnight (20-24 h).
Aﬂer ﬁxation, all tissues were stored in 1% neutral buffered formalin and reﬁ'igerated

(4 °C) until counting was completed. Cecums were subsequently blot-dried with paper
towels and weighed to the nearest hundredth of a gram.

Scoring of Adenomas: To avoid variation between researchers, all counting was
performed by one person who was blinded to treatments. Tissues were stained in a 0.2%
methylene blue phosphate-buffered solution (pH 7.4), and viewed with a stereo
microscope to detect adenomas. Carriers (Min +) were determined by presence of
adenomas. When carrier status was ambiguous, DNA was extracted from a piece of
liver, obtained at sacriﬁce, and polymerase chain reaction (PCR) was performed as
previously described (Luongo et al, 1994). Intestinal adenomas and average diameter for
each adenoma were recorded. Colonic and cecal tumors were recorded as the total
number per tissue, and a three dimensional measurement (mm?) was recorded. The three
dimensional measurements for colonic tumors were then converted to spheric volumes

(41tlwh/3).

39

SCFA Analysis: Ceca] contents were thawed and hydrated with 3 or 4 volumes
of double-distilled water (ddeO). Individual pHs were obtained after dilution using a
Fischer pH probe. Meta-phosphoric acid (25% wt/wt) was added to the samples in a
1:1 :4 ratio (m-phosphoric acidzsample wtzddeO), and samples were centrifuged at
15,000 x g for 5 minutes. The supernatant (4uL) was injected into a gas chromatograph
(column: 10% SP1200/ 1% H3PO4 on 80-100 Chromosorb, glass column 3.2 mm ID. by
2 m long). Nitrogen was the carrier gas. The injection and detector temperatures were
150 °C, and the column oven was temperature-programmed (initial temperature: 110 °C
for 5 minutes, increased by 3 °C/min until ﬁnal temperature of 127 °C). SCFA were
detected by ﬂame ionization and quantiﬁed by comparison to known standards of acetate,
propionate, isobutyrate, butyrate, valerate, and isovalerate. The mean of two sample
injections differing by < 5% was used to quantify the SCFA.

Statistical Analysis: Statistical analyses were performed using SAS statistical
software (SAS Institute, Inc. Cary, NC, Version 7.0). All statistical differences were
detected using a critical value of P < 0.05 with the F statistic. When signiﬁcant
treatment, sex, and carrier status effects were detected, appropriate least squares means
(LSM) were compared by the least signiﬁcant difference method. The total number of
tumors for the small intestine, colon, and cecum were totaled, ranked, and analyzed by
the Kruskal-Wallis test. Small intestine and colon lengths and cecum wet weights were
adjusted for metabolic body size (kg3/4). After adjustment, the tissue parameters and
SCFA concentrations were analyzed as a 2X5X2 factorial design (sex, treatment, and
carrier status). Body weights were analyzed by repeated measures armlysis of variance

of body weight/ sex/carrier status/time. All data are reported as mean 2 SEM.

40

VII. RESULTS

Mouse Body Weights: Dietary treatment did not signiﬁcantly inﬂuence body
weight at any time point in the study. As expected, males weighed signiﬁcantly more
than females (P < 0.0001 ). Carriers (Mir?) weighed signiﬁcantly less than non-carriers
(P < 0.001). This difference was signiﬁcant between male carriers and non-carriers at 7
weeks of treatment (P < 0.002) and for fermles, this difference was signiﬁcant by week 8
(P < 0.04) (Figures 5 and 6, respectively).

Tumor numbers: There was no signiﬁcant effect of treatment on adenoma
numbers in the small intestine or polyp numbers in the colon or cecum (Table 3).

Tissue Parameters: Small intestinal and colonic tissue lengths (mm) and the
cecum wet weights (g) are shown in Table 4. After adjustments were made for metabolic
body size, there was a signiﬁcant treatment effect for small intestinal lengths (P <
0.0001), colon lengths (P < 0.0015), and cecum wet weights (P < 0.0001) with the
fermentable ﬁbers (F OS, resistant starch, and gum arabic) having heavier cecums, and
the FOS and gum arabic groups producing longer small intestines and colons than the
controls (cellulose and kaolin). Females had signiﬁcantly greater cecum weights and
small intestinal and colon lengths than males. Non-carriers (Min') had signiﬁcantly

shorter small intestines and colons than carriers.

41

Amod v E 2258: Me 8.83 o «a 282%:me mm? 523 AS: A
+533 2288 22.23 8 26 Soho 28%? a 83 822. .228. 08: >5 8 God v A: 80:20ng 25:58.:
223me6 o: 803 82:. deﬁning: .9206 mats“. 028 -3: 28 +2.2: 22: .3 Saw 2203 boom .m 9.53%

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A285 82- 2283 25
3 w c v m o S w o v N o
r . . _ _ od 2 _ _ . . .
L §8MIII _ ﬂog II] o c
‘ 282. 86¢? ,- -l-|-z.-i- ow 282. 861.? ca
.
noHSm BEBE lal . scam 82.2262 IT .
z monTou 2 82 m monTou 1 82 m
_ 8235+ T. -,]....--III 0.2 A 8o3=oololT » of A
r_l i . _ m. I]- . : m.
. odm mm. - odm am.
98 mm» - 8.8M,
i-lll i]- odm ( cdm (
1‘ ll 1 l 1i 1- 9mm - 9mm
odv odv

 

 

 

«+35% £02.50 0...: 72.5: ﬂoEaUéez 9?:

42

.390 v a: “conga“ mo 8.33 w «a “50$:me mm? 523 .55: A
15$ 38% Ego 3 26 «coho “SB—.336 a mm? 205. .aﬁa 0:5 has 6 God v E mooceummw 2253:
38%? o: 0.83 222- doucotouﬁ E06 waist 028 SE 93 me: 038$ mo Eww 2303 30m .e 95»:—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

AﬂooBV oEC- Ana—83v 08:-
e w o ¢ N o S w e v N
p r _ b _ o.o r h _ _ p O.O
anMIol amouMI-I
053. §0IXI 9m 3952 830* -ll 1 11 ed
535 8582+ W 53% BEBEIT
8 H
lol
men-.0.- 3: m. m8 _ 3: ma
owe—230+ M 82230 0 V ,m.
w. 1- . . A
gram. 1.1- l I {£111 o2 m.
h m
2am 2.2m
9mm 1 ‘ odm
can 93

«3.5» 20.5.6 225-,—

ES» EEG-52 ciao-.—

43

8808808 8080 000808-38 885:me o: 0803 085. 8028 09 280 80808808

882808—0008“ a 08:3 8083 8000808 8083 8800 2:... :28 28m .8083 828 28 8: .8 82m 05 08

8085 :28 802. .9828 28 .2228 408283 82800 00.8 mo 13805 083 8083 8200008 =88m 80880828885
.8 8388 n c 28 .Emm H 808 8 880008 08 3800.— 820 802:8 8.2 88 2:83.10me 22>» 808280 083 Sam .

 

 

ASHE
3 H 3 S H a... 2 H S 3 H he :8. S H 8 882
GT5
3H 3 : HS «:on SHE :8 aHS 055.89
ans
2 H We on H H We H 3 no H 0.0 :9. 2 H a 650 8802
$7.5
No H 0.0 2 H E 2 H 2 2 H S :8 S H NH 2:
Eng
No H 2 S H Hm 3 H 3 2 H 3 $8 2 H 2 322.00
FEEV 08m 82.852 82.852 080.20:—
80:8: Z 88am. 885,—. 88.5. 885,—. 82.882

.885. .300 8200 8200 2.0m 8.00 .309 8200 885,—. 5 80880.;

..§8° a 8:0 8 a
0088 2:5 802 .8 8pm 0 80:20 8.2 028 +8: 8 8388 888 2800 28.0 .0828 823.008 =8m .m 030,—.

44

020 80: 0:80.08 .8 3&8 8 88 00 00000882 .000me 020 .802 0:30.08 8.. 80.02.30 0
.0880 02.800 8 .80 808.000. 00: God v .6 0008008: 08:00 0 .0 .n .0 0.800085 0
80800082088. .8 0028:: u : 0:0 8mm H :008 00 0080008 000 02:00.. .<>OZ< 80.. 80.02080
0003 0.0m .88 :20:0. 0006 002:. 880008 2.0 0.220000 w830=2 20.08088 :82. 0003 028:2 0:00; _

 

 

 

 

 

0. H000. .000H0. .000H00 308000800 .530 .5an

0. H2... 000.0 H 0.. .. 00.0 H 0.0 000.15 03.50.82
.. .00 H 0.. 0 00.0 H 0.. .. 00.0 H 0.0 00000 0.0800

00m
. 00.0 H 0... . .000 H E . 00.0 H 0.0 A00ns 0.02
. 00.0 H 0.. . 00.0 H 00.. 0.00 H 0.0 Guns 0080.
0 cod H 0.2 3 mod H 32 02.0 H 0.0 Gané 030.2. 8:0
.. 00.0 H 0.. .. 3.0 H 0.... .. 0.0 H 0.0 sans 02000 000.002 05.5020.
. 00.0 H 0.0 . 00.0 H 00.. . 0.0 H 0.0 Gang 00..
0. 00.0 H 8 . 00.0 H 00.. . .0 H 0.0 .008 00.03000
0380 0088. 0598..— 0 A88. 023:0.—

08203 00>? 8:000 0:005. 8200 08.28 08.008— =08m

 

.2308 w 820m 8 00.800 2...: .008 :0 8 00am .8 000.800 820%
00.0 028 -8: 80 +53. 8 A3 08203 .03 2000 0:0 :28 2.0 08000.2 :80 0:. 8 0880 059.2 0:02. :00: .v 030,—.

45

SCFA Analysis: The millimolal (mm) concentrations of SCFA (umole/g cecal
contents) are shown in Table 5. Dietary treatments affected both total SCFA and
individual SCF A concentrations. For all three fermentable ﬁbers, there were
signiﬁcantly greater concentrations of SCFA in cecal contents than for either the
cellulose or kaolin treatments (P < 0.0001). The cecal contents from mice fed kaolin had
the smallest concentration of acetate followed by cellulose. Propionate concentration was
highest in mice fed the F OS and modiﬁed starch, whereas mice fed F08 and gum arabic
had greater amounts of butyrate in the cecal contents. The concentration of valerate was
highest in mice fed modiﬁed starch, and within this treatment, females produced
signiﬁcantly more valerate than nmles (P < 0.0001). Carrier status also affected both
individual and total SCFA concentrations. For both acetate (P < 0.01) and total SCFA (P
< 0.03), Min’ produced higher concentrations than Min+ mice. Carrier status also aﬂected
the branch-chain fatty acids isobutyrate (P < 0.0001) and isovalerate (P < 0.0001) with
carriers producing higher amounts than non-carriers.

Table 6 shows the pH of cecal contents and molal proportions of individual SCFA
for each dietary treatment. The FOS treatment yielded both higher propionate and
butyrate levels. The lower total mm SCFA among all carriers (Table 5) was most likely
due to a decrease in acetate production. The FOS and modiﬁed starch groups had
signiﬁcantly lower molal proportions of acetate than the other treatment groups. In the
FOS group, there was a shift towards propionate and butyrate production, and towards
propionate and valerate production in the modiﬁed starch group. The three fermentable

ﬁber groups also had lower cecal pH values compared to controls (Table 6).

46

£38m 3E8 5 £8 €258: com God v AC moococo§@ 80:3 @ .o .n d marombasw N
.28: v
AC 953 £035 @oE@oE 2: E 882? 0.58 @8305 why—«Sum 320:3 .GooodVa .Soodv AG Sago—goﬂ @5 25.53%
we 3285588 H232 @a: 832 .298 v m dcod v m .. 5E A +EV<V cake—«>3 @§ SEbBOmm 8% @§ .F 5:
A . £33 «mam :38 @5 8808 How Soho 5E8 986 v m @5 Ed v A: “5053mm m 83 205. $3502 @qm $582
05 E @ontomo@ mu 3:058 :88 Eob @8388 983 <mUm .36 v m ”a agoﬂawmm 35233 203 Sam 4:253:

com mg mo 398:: n a @§ .Emm H :38 me @8585 0.8 838.— .< >0 Z< maﬁa @ogoﬁoq 203 mosh—«Sc. Romucﬁm _

 

 

 

 

 

 

 

«Wm H ow Ed H 9N and H M: awed H n; 9o H ow 9o H 3w F; H mm ASN.5 E50 253m
DON H we cod H _.N smod H 0; .36 H 2 ed H 06 We H ad gm; H me ch5 5:30-82 3:30
Wm H no a Ed H N.N a vod H m; amod H «4 ed H M; We H 3w n; H om ASHE £25m
mm H 8 ._ 86 H 3 ._ 8d H S a 8d H 2 9o H 3 3 H ed 2 H 3 ASHE 2H: 5m
me H 9. Hood H M: 2.36 H N; 58.0 H 0; .ud H Ow .50 H 3. “MN H mm GNHE 533m
HEM H on 16 H ON amod H o; Rod H mg 36 H S and H 5.5 Rama H 3 Gmncv 058< 8:0
a: H on Ld H QM acod H \L good H o; a: H 0.x no; H 2 god H mv Avmucv sebum @02
pea H E. sood H w; h.mod H M: 58.0 H 5; 36 H 3 smd H 2 3m.m H 9 Gang mo;
.2 H S 1.0 H 5 .86 H 3 s50 H 3 no.0 H Ow .wd H we aim H 2 Gmncv 32260 2.258.;

:35. 8325» 8a..2a>3— 8.2.8.53— 825—5 3.2839:— 8303..

:2:
.~ ._Com@ wcotwcv £3 tea .8 .8 “one me $958 505%
@om 3:: -5: @5 +33. 5 328:8 @800 89% con «ﬁx—Um @3239: we 3285588 AwB—oﬁsv 30% .m 03:.

47

.835 82.80 .8 £8 3:258: 8Q Amod v Ac 8288b? 80:3 @ .o d d mam—889m N

.225... v e .22.

52: 3829» Mo 8220qu 59: 852 @E macaw spasm @0592: 05 E 852$ @5 A36 v m .mood v .3 880332
9.. 2.53820 .\. ES. 5&2 HE .232 .225 v .2 ms: v -53: 28.3.8. 2.. 3.5382 .2 8. 5s: A

-5}: 280% .x. com 80.20 82.80 Sodv as 2.85%? a 83 22:. .36 v m E 280$:me 88288 083 Sea .8258:

8a mg Mo 8955 u a @S. .Emm H :88 8 @2883 new 3382 .<>OZ< maﬁa @28023 083 8&3.“ Botmzﬂm _

 

 

 

 

 

 

 

2 H 2 .3 H 2 .3 H 3 2 H 2 to H 2 . 3 H 8 2. Gene .556 .23...
3 H m... .3 H tn .3 H 2 no H 2 2 H .1 . 3 H we as some 8.2.5.52 3E5
3 H 3 . S H E . S H E 9o H 2 to H 3 no H 8 .2 sons 32...;

3 H 3 . 3 H 3 . 3 H 2 2 H 2 2 H 2 2 H S .3 Gone 2.2 5m
. S H 2 .3 H 2.. . S H ..m . .3 H: . 2 H 2 . .3 H we .3 GT5 £82
. 3 H 3 . S H 3 . S H 2 . 2 H 2 . 3 H : . 2. H 8 .3 Gang 05...... use
. 3 H S. .S H 3 . No H .2 . 3 H 2 . 3 H 2 . 2 H G .3. sans 535 .82
.3 H 2 . S H 2 . S H .H . .3 H 2 . 3 H 2 . .3 H a .3 smug 8.2
. 2 H 3 . S H 2 . g H mm . 2 H 2 . 2 H : . 3 H 8 .3 GT5 0.22.8. Hogans

382.3
898:5 82395,.— 8255222 8.23.5 8282.9..— 833< .33 me la
.x. 3.:
. .32. .52ro as a... s. 8

Sam mo 88:8 Eo~o§@ @o.« 82: La: @5 +32 3 «ﬂow mo 83:3on

508 @5. 28:80 @800 .«o In

.9 03:.

48

VIII. DISCUSSION

In this study, the effect of three highly fermentable ﬁber sources on tumor
development, SCFA production, and tissue parameters were examined in mice with a
genetic predisposition to colon cancer. It has been previously reported that resistant
starch and F08 are protective against colon carcinogenesis in some animal models
(Pierre, 1997; Thorup, 1995). Pierre et al (1997) showed a lower number of total and
small colon tumor numbers in Min+ mice fed a 6% FOS diet. This effect, however, was
signiﬁcant only when compared to a 2% ﬁber control (Pierre, 1997). In the present
study, there was no effect of dietary treatment on tumor development in either the small
intestine, cecum, or colon. Differences between this study and that of Pierre et a1 (1997)
may have been due both to the level of fat in the diets and the length of dietary treatment.
Wasan et al (1997) showed that increasing the level of fat ﬁom 3 to 15% of the diet
signiﬁcantly increased tumor numbers in both the small and large intestine in Min+ mice.
In this study, a 15% (wt/wt) fat diet was used compared to an 8% fat diet in the study by
Pierre et a1 (1997). Also, in the present study, mice were fed for a total of 10 weeks and
were sacriﬁced at 98 days of age, compared to 42 and 84 days, respectively, in the study
by Pierre et a1 (1997). Although tumor multiplicity is thought to be established by about
70 days of age in Min+ mice, the size of detectable tumors, especially in the colon, could
have been inﬂuenced by age (Shoemaker et al, 1995). The length of the study may have
also confounded possible treatment eﬂ‘ects due to weight loss, which was signiﬁcant for

males in this study at 7 weeks and 8 weeks of treatment for females.

49

There has been some concern that soluble ﬁbers, through fermentation to SCF A,
may enhance colon carcinogenesis (Jacobs, 1987). Both a decrease in luminal pH and an
increase in SCF A production stimulate cellular proliferation in both the cecum and colon,
with butyrate being highly correlated to site-speciﬁc proliferation indices in the distal
colon (Lupton and Kurtz, 1993). The effect of SCFA on epithelial growth in the normal
colon, however is conﬁned to the lower crypt compartments (V elaquez et al, 1996). In
contrast, factors which stimulate hyperproliferation in the upper crypt, such as secondary
bile acids, promote colon carcinogenesis. Although cellular proliferation was not
measured in the present study, the fermentable ﬁbers all signiﬁcantly increased the molal
concentrations of SCFA in the cecal contents of Min+ mice. However, increased
concentrations of SCFA were not associated with an increase in polyp incidence or size
in either the cecum or colon, the primary sites of fermentation.

The effect of fermentable ﬁbers on tissue parameters and SCFA concentrations is
noteworthy. Mice fed the fermentable ﬁbers had signiﬁcantly longer SI and colon
lengths and heavier cecum wet weights than controls. Small intestinal growth is
stimulated both by the presence of viscous ﬁbers in the small intestine and by SCFA-
mediated systemic effects. Gee et 31(1996) examined the effect of viscosity and
fermentation on small intestinal mucosal cell proliferation in rats. Compared to both
cellulose and lactulose controls, the more viscous polysaccharides
(hydroxypropylmethylcellulose and guar gum) signiﬁcantly increased srmll intestinal
lengths in rats in a manner which was independent of their fermentation (Gee et al, 1996).
Goodland et al (1995) examined trophic effects of ﬁber in the small intestine, stomach,

and colon in both germ-ﬂee and conventional rats. They reported that, compared to

50

ﬁber-ﬁee controls, ﬁber was trophic to both the srmll intestinal muscle and gastric
epithelium. In contrast, ﬁber stimulated crypt cell production in both the ileum and colon
only in conventional rats. Because fermentation was necessary, it was suggested by the
authors that this effect was due to SCFA production (Goodland et al, 1995).

In this study, the three fermentable ﬁbers used were not highly viscous, and
cellulose and kaolin were included as poorly or non-fermentable bulk sources. After
adjusting for metabolic body size, tissue lengths in all three fermentable ﬁber groups was
different than controls. Because this effect was seen only in the fermentable ﬁber
treatment groups, this suggests that SCF A likely are more trophic to the small intestine
than a bulk source alone. There were also signiﬁcant differences in colon lengths
between the fermentable ﬁber groups and controls. Although the SCFA concentrations
were not measured throughout the colon in this study, infusion studies show that SCF A
are locally-acting trophic agents to the colonic epithelium (Sakata, 1982). Although the
three fermentable ﬁber sources used in this study are thought to be rapidly fermented, the
trophic eﬁ'ect on colon length is suggestive that the luminal SCF A were present distally
ﬁom the cecum. The weight of empty cecums were also signiﬁcantly heavier in the
fermentable ﬁber groups than in controls. This effect has been reported in both rats and
mice fed fermentable ﬁber sources and is likely due to an increase in proliferation or
increase in cecum size, both from a decrease in pH and an increase in SCFA
concentrations (Lupton and Kurtz, 1993).

The increased millimolal concentrations of total and individual SCFA in cecal
contents of mice fed fermentable ﬁbers supports the view that fermentation patterns are

alterable by dietary modiﬁcations. Kaolin, the non-fermentable bulking agent used as a

51

control in this study resulted in the lowest overall SCFA concentrations in the cecum.
The most signiﬁcant differences were in the major SCF A. F OS yielded greater amounts
of both propionate and butyrate at the expense of acetate. This pattern is in agreement
with ﬁndings ﬁ'om Bergeen et al (1993).

Fermentation of resistant starch and gum arabic resulted in higher propionate and
butyrate concentrations, respectively, than controls. Starches have been shown to yield
relatively large butyrate concentrations in vitro (Cummings, 1991 ). However, several
studies suggest that the type of resistant starch may inﬂuence the proportions of
propionate and butyrate produced (Bergeen, 1993, Cummings et a], 1986). The starch
source used in this study was chemically modiﬁed to be resistant to amylases in the small
intestine and the precise composition was not known because of propriety reasons.
However, based on the molal proportions of SCFA in this study, the modiﬁed starch
source used is consistent with fermentation patterns previously observed from wheat and
maize starch (Bergreen, 1993; Cummings et al, 1996). Valerate production was also
signiﬁcantly higher in the modiﬁed starch group, but the signiﬁcance of this is not
known.

Carrier status also inﬂuenced fermentation patterns. There were smaller
concentrations of acetate and total SCFA and a signiﬁcant increase in the concentrations
of isovalerate and isobutyrate. Production of the branch-chain fatty acids isobutyrate and
isovalerate have been previously shown to increase with protein fermentation
(Macfarlane, 1992). Severe anemia due to gastrointestinal bleeding is common in Min+
mice, and is likely to have increased endogenous protein sources to the colon accounting

for these differences.

52

IX. SUMMARY AND CONCLUSIONS

This study demonstrated that fermentation patterns in mice are inﬂuenced by
differences in dietary ﬁbers and that fermentation has trophic effects to the small and
large intestines and to the cecum. The increase in cecum weight was most likely due to
increased concentrations of SCF A in the cecal contents. The increase in SI lengths may
be due to systemic events stimulated by SCFA. Since the differences in butyrate
concentrations occurred in the cecum whereas the majority of neoplasms occurred in the
SI and distal colon, this study could not adequately evaluate the presumptive anti-tumor
effect of butyrate on intestinal tumorigenesis. However, fermentation did not alter tumor

multiplicity or size in Min+ mice.

53

X. RECOMMENDATIONS FOR FUTURE RESEARCH

The Min+ mouse model was used in this study because genetically, it is similar to
the human condition of FAP. Tumors occur spontaneously and there is no need to
administer carcinogens. Thus, this genetic model of human colon cancer eliminates the
variability associated with the type, amount, and method of administration of chemical
carcinogens required to induce colon cancer in normal rodents. Although polyp
formation in the colon (approximately l/mouse) was observed in this study, the majority
of adenomas were distributed throughout the small intestine. The proposed method of
chemoprevention was through short chain fatty acid production, which occurs primarily
in the cecum and colon. Although there were clearly systemic trophic effects in the small
intestine with fermentable ﬁbers, short chain fatty acids are not found in high
concentrations in portal or systemic blood, and would not likely act by this mechanism to
inhibit tumor formation in the small intestine. SCFA, if protective most likely act from
direct luminal exposure in the colon. Therefore, with a lower colon tumor multiplicity
per animal and the relatively small number of carriers per treatment in this study, it may
have been difﬁcult to detect differences, if any, in colon tumor multiplicity and size.
Future studies speciﬁcally examining fermentation and colon cancer with this model
should consider more carriers per treatment and attempt to suppress tumor growth in the
small intestine, possibly by reducing the level of fat in the diet.

Although we showed that the type of ﬁber fed to the mice in this study resulted in
both different concentrations and proportions of SCFA in the cecum, it was unclear as to

whether this effect continued throughout the large intestine. F OS, resistant starch, and

54

gum arabic are all highly fermentable and could have been completely fermented before
reaching the distal colon, where most polyps developed. A combination of ﬁber sources
that delay fermentation to the distal bowel should be considered in ﬁrture studies.

It is still unresolved as to whether SCF A, in particular butyrate, are protective against
colon carcinogenesis in vivo. Although some evidence supports a protective role for
fermentable ﬁbers that yield high butyrate concentrations, other studies including ours
did not ﬁnd an effect (Pierre et al, 1997 ). Thus, it also remains possible that higher
levels of dietary ﬁber, continual luminal exposure to butyrate, or butyrate concentrations
above which can be achieved through dietary treatment may be needed for butyrate to be
an effective chemopreventive agent in the colon. Future studies should be designed with

these limitations in mind.

55

XI. LITERATURE CITED

Alberts, D.S., Martinez, M.E., Roe, D.J., et a1. (2000) Lack of effect of a high-ﬁber
cereal supplement on the recurrence of colorectal adenomas. New Engl. J. Med.
342: 1156-62.

Andersson, H.B., Ellegard, L.H., and Bosaeus, LG. (1999) Nondigestibility
characteristics of inulin and oligoﬁ'uctose in humans. J. Nutr. 129(7 Suppl):
1428S-3OS.

Annison, G., Trimble, RP, and Topping, D.L. (1995) Feeding Australian acacia gums
and gum arabic leads to non-starch polysaccharide accumulation in the cecum of
rats. J. Nutr. 125:283-92.

Archer, S., Meng, S., and Wu, J. (1998) Butyrate inhibits colon carcinoma grth
through two distinct pathways. Surgery 124: 248-53.

Asp, N.G. (1995) Dietary ﬁbre analysis-an overview. Eur. J. Clin. Nutr. 49S: S42-S47.

Barnard, J .A. and Warwick, G. (1993) Butyrate rapidly induces growth inhibition and
diﬂ‘erentiation in HT -29 cells. Cell Growth & Differentiation. 4:493-501.

Berggren, A.M., Bjorck, I.M., Margareta, E., et a]. (1993) Short chain fatty acid content
and pH in ceacum of rats given various sources of starch. J. Sci. Food Agric. 68:
241-48.

Berggren, A.M., Bjorck, I.M., Margareta, E., et al. (1995) Short chain fatty acid content
and pH in ceacum of rats given various sources of carbohydrates. J. Sci. Food
Agric. 63: 397-406.

Ben-Ze’ez, A. and Geiger, B. (1998) Diﬂ‘erential molecular interactions of B-catenin and
plakoglobin in adhesion, signaling and cancer. Curr. Opin. Cell Biology 10: 629-
39.

Binder, H.J. and Mehta, P. (1989) Short-chain fatty acids stimulate active sodium and
chloride absorption in vitro in the rat distal colon. Gastroenterology 96: 989-95.

Bingham, S. (1990) Mechanisms and experimental and epidemiological evidence relating
dietary ﬁbre (non-starch polysaccharides) and starch to protection against large
bowel cancer. Proc. Nutr. Soc. 49(2): 152-157.

Bingharn, S. (1997) Meat, starch, and non-starch polysaccharides, are epidemiological

and experimental ﬁndings consistent with acquired genetic alterations in sporatic
colorectal cancer? Cancer Letters 114: 25-34.

56

Birkett, A., Muir, J., Phillips, J., Jones, G., and O Dea, K. (1996) Resistant starch lowers
concentrations of ammonia and phenols in humans. Amer. J. Clin. Nutr. 63: 776-
87.

Boland, C.R., Sato, 1., Appelman, H.D., Bresalier, R.S., and Feinberg, A.P. (1995)
Microallelotyping deﬁnes the sequence and tempo of allelic losses at tumour

suppressor gene loci during colorectal cancer progression. Nature Medicin
1:902-09. '

Brabletz T, Jung A, Hermann K, et al. (1998) Nuclear overexpression of the oncoprotein

B-catenin in colorectal cancer is localized predominately at the invasion front.
Pathol Res Pract. 194: 701-704.

Burkitt, WP. (1971) Epidemiology of cancer of the colon and rectum. Cancer 28: 3-13.

Burt, R.W. (1983) Inheritance—general issues (Cohen, A.N., Winawer, S.J., Friedman,
M.A., et al., eds). McGraw Hill, New York.

Campbell, J .M., Fahey, G.C., and Wolf, B.W. (1997) Selected indigestible
oligosaccharides aﬂ‘ect large bowel mass, cecal and fecal short-chain fatty acids,
pH and microﬂora in rats. J. Nutr. 127: 130-36.

Cassand, Pierrette, Maziere, S., Champ, M., et al. (1997) Effects of resistant starch and
Vitamin A-supplemented diets on the promotion of precursor lesions of colon
cancer in rats. Nutrition and Cancer 27(1): 53-59.

Chesson, A. (1995) Food Polysaccharides and their applications (Stephen, A.M., ed.), pp.
547-70. Marcel Dekker, New York.

Compher, C.W., Frankel, W.L., Tazelaar, J, et a1. (1999) Wheat bran decreases abberant
crypt foci, preserves normal proliferation, and increases luminal butyrate levels in
experimental colon cancer. J.P.E.N. 23: 269-78.

Cummings, J.H. (1981) Short chain fatty acids in the human colon. Gut 22(9): 763-79.

Cummings, J.H. and Macfarlane, GT. (1991) The control and consequences of bacterial
fermentation in the human colon. J. Appl. Bacter. 70: 443-59.

Cummings, J.H. Pomare, B.W., Branch, W.J., et a1. (1987) Short chain fatty acids in
human large intestine, portal, hepatic, and venous blood. Gut 28: 1221-27.

Cummings, J.H., Beatty, E.R., Kingman, S.M., et a1. (1996) Digestion and physiological
properties of resistant starch in the human large bowel. Br. J. Nutr. 75: 733-47.

57

Dobbie, Z., Spycher, M., Mary, J., et al. (1996) Correlation between the development of
extracolonic manifestations in FAP patients and mutations beyond codon 1403 in
the APC gene. J. Med. Gen. 33: 274-80.

Dowling, RH. (1978) Small bowel adaptation and its regulation. Scand. J. Gastroenterol.
Suppl. 74: 53-74.

Drasar, 1.8. and Hill, M.J. (1972) Intestinal bacteria and cancer. Am. J. Clin. Nutr. 25:
1399-1404. '

Eastwood, G.L. (1983) (Bustos-Fernandez, L., ed.), pp. 1-8. Plenum Medical Book
Company, New York.

Edwards, C.A. and Eastwood, M.A. (1995) Caecal and faecal short-chain fatty acids and
stool output in rats fed on diets containing non-starch polysaccharides. Br. J. Nutr.
73: 773-81.

Ehle, F.R., Robertson, J.B., and Van Soest, P.J. (1982) Inﬂuence of dietary ﬁbers on
fermentation in the human large intestine. J. Nutr. 112: 158-66.

Englyst, H.N. and Kingman, S.M., in Dietary Fiber in Health and Disease. (Kritchevsky,
D., Bonﬁeld, C., Anderson, J .W., eds.), Plenum Press, New York, 1988, pp. 49-
59.

Englyst, H.N. and Cummings, J.H. (1986) Digestion of the carbohydrates of banana
(Musa paradisiacal sapientum) in the human small intestine. Am. J. Clin. Nutr.
44: 42-50.

Fearon E. R. and Vogelstein B. (1990) A genetic model for colorectal tunorigenesis.
Cell.6l: 757-67.

Ferguson, L.R., & Harris, P.J. (1996) Studies on the role of speciﬁc dietary ﬁbres in
protection against colorectal cancer. Mutation Research 350: 173-84.

Ferguson, L.R., Tasman-Jones, C., Englyst, et al. (2000) Comparative effects of three
resistant starch preparations on transit time and short chain fatty acid production
in rats. Nutr. Cancer 36(2): 230-37.

Frankel, W.L., Zhang, W., Singh, A., et al. (1994) Mediation of the trophic effects of
short-chain fatty acids on the rat jejunum and colon. Gastroenterology 106: 375-
80.

Fodde, K, Smits, R., Hoﬂand, N., et al. (1999) Mechanisms of APC-driven
tumorigenesis: lessons ﬁ'om mouse models. Cytogenet. Cell Genet. 86: 105-111.

58

Gallant, D.J., Bouchet, B., Buleon, A., and Perez, S. (1992) Physical characteristics of
starch granules and susceptibility to enzymatic degredation. Eur. J. Clin. Nutr.
46(Suppl. 2): S3-Sl6.

Garnet, L., Daviaud, D., Denis-Pouxviel, C., et al. (1992) Effects of short-chain fatty
acids on grth and differentiation of the human colon-cancer cell line HT29. Int.
J. Cancer 52: 286-89.

Gee, J .M., Lee-Finglas, W., Wortley, G.W., et al. (1996) Fermentable carbohydrates
elevate plasma enteroglucagon but high viscosity is also necessary to stimulate
small bowel mucosal cell proliferation in rats. J. Nutr. 126: 373-79.

Gibson, GR. and Wang, X. (1994) Enrichment of biﬁdobacteria from human gut
contents by oligoﬁ'uctose using continuous culture. FEMS microbiology letters
118: 121-28.

Gibson, G.R. & Roberﬁ'oid, MB. (1995) Dietary modulation of the human colonic
microbiota: introducing the concept of prebiotics. J. Nutr.125: 1401-12.

Goodland, RA., Ratcliﬂ‘e, B., and Wright, NA. (1995) Dietary ﬁbre and the
gastrointestinal tract: diﬂ‘ering trophic effects on muscle and mucosa of the
stomach, small intestine, and colon. Eur. J. Clin. Nutr. 49: 8178-81.

Govers, M.J., Gannon, N.J., Dunshea, FR. (1999) Wheat bran affects the site of
fermentation of resistant starch and luminal indexes related to colon cancer risk:
a study in pigs. Gut 45: 840-47.

Greenlee, RT., Murray, T., Bolden, S., et aL (2000) Cancer Statistics, 2000. CA Cancer
J. Clin. 50:7-33.

Greenwald, P. and Lanza, E. (1986) Role of Dietary Fiber in the prevention of cancer.
In: Important advances in oncology. Lippincott, J.B., Philadelphia, pp. 37-54.

Groden, J, Thliveris, A., Samowitz, W., et a1. (1991) Identiﬁcation and characterization
of the F arnilial Adenomatous Polyposis Gene. Cell 66: 589-600.

Haines, A., Metz, G., Dilawari, J ., et a]. (1977) Breath-methane in patients with cancer
of the large boweL Lancet 2(8036): 481-3.

Harig, J .M., Soergel, K.H., Komorowski, R.A., et al. (1989) Treatment of diversion
colitis with short-chain fatty acid irrigation. N. Engl. J. Med. 320(1): 23-8.

Hague, A., Manning, A.M., Huschtscha, L.I., et al. (1993) Sodium butyrate induces
apoptosis in human colonic tumor cell lines in a p-53 independent pathway:
implications for the possible role of dietary ﬁber in the prevention of large bowel
cancer. Int. J. Cancer 55: 498-505.

59

Harris, P.J., & Ferguson, LR. (1993) Dietary ﬁber: its composition and role in
protection against colorectal cancer. Mutation Research. 290:97-110.

Hao, X., Tomlinson, I., Ilyas, M., et al. (1997) Reciprocity between membranous and
nuclear expression of B-catenin in colorectal tumors. Virchows Arch. 431: 167-
72.

Heerdt, B.G., Houston, M.A., and Augenlicht, L.H. (1997) Short-chain fatty acid-
initiated cell cycle arrest and apoptosis of colonic epithelial cells is linked to
mitochondrial ﬁmction. Cell Growth & Differentiation 8: 523-32.

Hedgepeth, C.M., Deardorff, M.A., and Klein, RS. (1999) Xenopus axin interacts with
glycogen synthase kinase-3 beta and is expressed in the anterior midbrain. Mech.
Dev. 80(2): 147-51.

Hill, M.J. Microbial metabolism in the digestive tract, (Hill, M.J. ed.). CRC Press, Boca
Raton, 1986, p. 33.

Hill, M.J., Crowther, J.S., Drasar, B.S., et al. (1971) Bacteria and the aetiology of cancer
in the large bowel. Lancet I: 95-100.

Ilyas, I. and Tomlinson, LP. (1997) The interactions of APC, E-cadherin, and B-catenin
in tumor development and progression. J. Path. 182: 128-37.

Jacobs, LR (1987) The effect of dietary ﬁber on colonic cellular proliferation and its
relationship to colon carcinogenesis. Prev. Med. 16(4): 566-7.

Johnson, LT. and Southgate, D.A.T., Dietary ﬁbre and related substances, (Johnson, LT.
and Southgate D.A.T., eds.), Chapman and Hill, 1994, pp. 14-38.

Kararn, SM. (1999) Lineage commitment and maturation of epithelial cells in the gut.
Frontiers in Bioscience 4: d286-98.

Kinzler, K.W., Nilbert, M.C., Su, L.K., et al. (1991) Identiﬁcation of FAP locus genes
from chromosome 5q21. Science 253: 661-663.

Kinzler, K.W. and Vogelstein, B. (1996) Lessons from heredity colorectal cancer. Cell
87: 1159-70.

Kinzler, K.W. and Vogelstein, B. In the genetic basis of human cancer, (V ogelstein, B.
and Kinzler, K.W., eds.), McGraw Hill, New York, 1998, pp. 565-82.

60

 

Klein, 8., Alpers, D., Grand, R.J., et al. (1997) Advances in nutrition and
gastroenterology: summary of the 1997 A.S.P.E.N. research workshop. J.P.E.N.
22: 3-13.

Korinek, V., Barker, N., Morin, P., et al. (1997) Constitutive transcriptional activation

by a B-catenin-ch complex in APC -/- colon carcinoma. Science 275(21) 1784-
87.

Koruda, M.J., Rolandelli, RH., Settle, R.G., et al. (1988) Effect of parenteral nutrition
supplemented with short-chain fatty acids on adaptation to massive small bowel
resection. Gasteroenterology 95(3): 715-20.

Kripke, S.A., Fox, A.D., Berman, J.M., et al. (1989) Stimulation of mucosal growth
with intracolonic infusion of short-chain fatty acids. J.P.E.N. 13: 109-16.

Kuhl, M. and Wedlich, D. (1997) Wnt signaling goes nuclear. Bioassays 19(2): 101-4.

Kvietys, PR. and Granger D.N. (1981) Effects of solute-coupled ﬂuid absorption on
blood ﬂow and oxygen uptake in the dog colon. Gastroenterology 81(3): 450-7.

Lee, S. (1995) Determination of total, soluble, and insoluble dietary ﬁber: collaborative
study. Eur. J. Clin. Nutr. 49: $153-$157.

Lin, H-C. and Visek, W.J. (1991) Colon mucosal cell damage by ammonia in rats. J.
Nutr. 121: 887-93.

Lipkin, M., Bell, B., and Shelrock P. (1963) Cell proliferation kinetics in the
gastrointestinal tract of man. J Clin Invest. 42: 767.

Luongo, C., Moser, A.R, Gledhill, S., and Dove, W.F. (1994) Loss of Apc+ in intestinal
adenomas ﬁ'om Min mice. Cancer Res. 54(22): 5947-52.

Lupton, JR, Coder, D.M., and Jacobs, LR. (1988) Long-term effects of fermentable
ﬁbers on rat colonic pH and epithelial cell cycle. J. Nutr. 118: 840-845.

Lupton, J .R. and Kurtz, RP. (1993) Relationship of colonic luMinal short-chain fatty
acids and pH to in vivo cell proliferation in rats. J. Nutr. 123: 1522-1530.

Macﬁrlane, G.T., Gibson, G.R., Beatty, E., et al. (1992) Estimation of short-chain fatty
acid production from protein by human intestinal bacteria based on branched-
chain fatty acid measurements. FEMS Microbiology Ecology 101: 82-88.

Macrae, L. (1999) Wheat bran ﬁber and development of adenomatous polyps: evidence
from randomized, controlled clinical trials. Am. J. Med. 106(1A): 385-425.

61

Mahmouns, N.N., Boolbol, S.K., Dannenberg, A.J., et al. (1998) The sulﬁde metabolite
of sulindac prevents tumors and restores enterocyte apoptosis in a murine model
of familial adenomatous polyposis. Carcinogenesis 19(1): 87-91.

Mann, B., Gelos, M., Siedow, A., et al. (1999) Target genes of B-catenin-T cell-
factor/lymphoid-enhancer-factor signaling in human co lorectal carcinomas. Proc
Natl Acad Sci USA 96: 1603-08.

Mayer, K., Hieronymus, T., Castrop, J., et al. (1997) Ectopic Activation of
Lymphoid High Mobility Group-Box Transcription factor TCF-l and
overexpression in colorectal cancer cells. Int. J. Cancer 72: 625-630.

McBain, J.A., Eastman, A., Nobel, C.S., et a1 (1993) Apoptotic death in adenocarcinoma
cell lines induced by butyrate and other histone deacetylase inhibitors.
Biochemical Pharmacology 53: 1357-68.

McWilliarns, M. Foods, 2"d edition, (Davis, K. ed.), Macmillan, New York, 1993a, p.
188.

McWilliarns, M. Foods, 2'“I edition, (Davis, K. ed.), Macmillan, New York, 1993b, p.
142.

Miller, T.L. and Wolin, M.J. (1979) Fermentations by saccharolytic intestinal bacteria.
Am. J. Clin. Nutr. 32: 164-72.

Miller, LR. and Moon, R.T. (1996) Signal transduction through B-catenin and
speciﬁcation of cell fate during embryogenesis. Genes and Development 10:
2527-2537

Miller, T.L. and Wolin, M.J. (1979) Fermentation by saccharolytic intestinal bacteria.
Am. J. Clin. Nutr. 32: 164-72.

Miyaki, M., Konishi, M., Kikuchi-Yanoshita, R, et al. (1994) Cancer Res.
54: 3011-20.

Morin, P.J., Vogelstein, B., and Kinzler, K.W. (1996) Apoptosis and APC in colorectal
tumorigenesis. Proc. Natl. Acad. Sciences USA 93: 7950-7954.

Morin, P. (1999) B-catenin signaling and cancer. BioEssays 21 :1021-30.

Moser, A.R., Pitot, HQ, and Dove, W.F. (1990) A dominant mutation that predisposes to
multiple intestinal neoplasia in the mouse. Science 247: 322-24.

Nagengast, F .M., Grubben, M.J.A.L., and Van Munster, LP. (1995) Role of bile acids in
colorectal carcinogenesis. Eur J Cancer.3l: 1067-1070.

62

Nishisho, 1., Nakarnura, Y, Myoshi, Y., et al. (1991) Mutations of chromosome 5q21
genes in FAP and colorectal cancer patients. Science 253: 665-69.

Olschwang, S., Tiret, A., Laurent-Puig, P., et a]. (1993) Restriction of ocular ﬁlndus
lesions to a speciﬁc subgroup of APC mutations in adenomatous polyposis coli
patients. Cell 75: 959-68.

Orford K.,Crockett, C., Jensen, J .P., et al. (1997) Serine phosphorylation-regulated
ubiquitination and degregation of beta-catenin. J. Biol. Chem. 272(40): 24735-8.

Orsulic, S., Huber, 0., Aberle, H., et a]. (1999) E—cadherin binding prevents B-catenin
nuclear localization and B-catenin/LEF-l-mediated transactivation. J .Cell Science
112: 1237-1245.

Oshima, M., Oshima, H., Kitagawa, K. (1995) Loss of APC heterozygosity and abnormal
tissue building innascent intestinal polyps in mice carrying a truncated APC gene.
Proc. Natl. Acad. Sci. USA 92(10): 4482-6.

Peifer M. (1997) B-catenin as oncogene: the smoking gun. Science 275(21): 1752-53.

Phillips, GD. (1998) Acacia gum (gum arabic): a nutritional ﬁbre; metabolism and
caloriﬁc value. Food Additives and Contaminants 15(3): 251-64.

Pierre, F., Perrin, P., Champ, M., Bomet, F ., Meﬂah, K, and Menanteau, J. (1997)
Short-chain ﬁ'ucto-oligosaccharides reduce the occurrence of colon tumors and
gut-associated lymphoid tissue in Min mice. Cancer Res.. 57: 225-28.

Polakis, P. (1997) The adenomatous polyposis coli (APC) tumor suppressor. Biochirnica
et Biophysica Acta 1332: F127-47.

Potter, J .D. (1999) Colorectal cancer: molecules and populations. J. Natl. Cancer
Institute 91: 916-32.

Powell, S.M., Zilz, Y.B., Bryan, T.M., eta]. (1992) APC mutations occur early during
colorectal tumorigenesis. Nature 359: 235-7.

Prosky, L. and Hoebregs, H. (1999) Methods to determine food inulin and oligoﬁ'uctose.
J. Nutr. 129: 14188-14238.

Rechkemmer, G., Ronnau, K, and Engelhardt, W. (1988) Fermentation of
polysaccharides and absorption of short chain fatty acids in the hind gut. Comp.
Biochem. Physiol. 90A: 563-68.

Rechkemmer, G. and Engelhardt, W. (1988) Concentration and pH-dependence of

short-cl‘min fatty acid absorption in the proximal and distal colon of guinea pig.
Comp. Biochem. Physio]. 91A: 659-63.

63

Reddy, BS. (1992) Effect of dietary ﬁber on colonic bacterial enzymes and bile acids in
relation to colon cancer. Gastroenterology 102: 1475-82.

Reddy, BS. (1993) Dietary fat, calories, and ﬁber in colon cancer. Prev. Med. 22:73 8-
49.

Reilly, K.J., Frankel, W.L., & Rombeau, LL. (1995) Colonic short chain fatty acids
mediate jejunal grth by increasing gastrin. Gut 37: 81-86.

Reeves, P.G., Nielsen, RH, and Fahey, G.C. (1993) AIN-93G 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: 1939-51.

Roberfroid, M., Gibson, GR, and Delzenne, N. (1993) The biochemistry of
oligoﬁ'uctose, a nondigestible ﬁber: an approach to calculate its caloric value.
Nutr. Rev. 51(5): 137-46.

Roberfroid, MB. (1999) Caloric value of inulin and oligoﬁ'uctose. .1. Nut. 129(7 Suppl):
1436S-7S.

Roediger, W.E. (1980) Role of anaerobic bacteria in the metabolic welfare of the
colonic mucosa in man. Gut 21: 793-98.

Roediger, W.E. (1982) Utilization of nutrients by isolated epithelial cells of the rat
colon. Gastroenterology 83: 424-29.

Roediger, W.E. (1990) The starved colon-diminished mucosal nutrition, diminished
absorption, and colitis. Dis. Colon Rectum 33: 858-62.

Rombeau, IL, and Kripke, S.A. (1990) Metabolic and intestinal effects of short-chain
fatty acids. J.P.E.N. 14(8): 1818-1858.

Ross, A.H., Eastwood, M.A., Brydon, W.G., et a]. (1984) A study of the effects of
dietary gum arabic in the rat. Br. J. Nutr. 51: 47-56.

Rozen P., Reich CB, and Winawer SJ. (1993) Large Bowel Cancer: Policy,
Prevention, Research and Treatment. S. Karger AG, Basel, Switzerland, p. 13.

Rudy, DR. and Zdon M.J. (2000) Update on colorectal cancer. American Family
Physician 61(6): 1759-70.

Sakata, T. and von Engelhardt, W. (1982) Stirnulatory effect of short chain fatty acids
on the epithelial cell proliferation in the rat large intestine. Comp. Biochem.
Physiol. 74A(2): 459-62.

Sakata, T. (1987) Stirnulatory effect of short-chain fatty on epithelial cell proliferation
in the rat intestine: a possible explanation for trophic effects of fermentable ﬁbre,
gut microbes and luminal trophic factors. Br. J. Nutr. 58:95-103.

Schatzkin, A., Lanza, E., Corle, D., eta]. (2000) Lack of effect of a low-fat, high-ﬁber
diet on the recurrence of colorectal adenomas. New Engl. J. Med. 342: 1149-55.

Scheppach, W., Bartram, H.P., & Richter, F. (1995) Role of short-chain fatty acids in the
prevention of colorectal cancer. Eur J Cancer. 31A:1077-1080.

Scheppach, W. (1994) Effects of short-chain fatty acids on gut morphology and
function. Gut S]: 835-38.

Scheppach, W., Fabian, C., Sachs, M., eta]. (1988) The eﬁ'ect of starch malabsorption
on fecal short-chain fatty acid excretion in man. Scand. J. Gastroenterology 23:
755-59.

Selvendran, RR. and Verne, A.V., (Kritchevsky, D., Bonﬁeld, C., Anderson, J.W., eds.),
Plenum Press, New York, 1988, pp. 1-4.

Shike, M. (1999) Diet and Lifestyle in the prevention of colorectal cancer: an overview.
Am. J. Med. 106(1A): 118—19S.

Shoemaker, A.R., Moser, A.R., Dove, W.F. (1995) N-ethyl-N-nitrosourea treatment of
multiple intestinal neoplasia (Min) mice: age-related effects on the formation of
intestinal adenomas, cystic crypts, and epidermoid cysts. Cancer Res. 55(19):
4479-85.

Shoemaker, A.R., Gould, K.A., Luongo, C., Moser, A.R., & Dove, W.F. (1997) Studies
of neoplasia in the Min mouse. Biochirnica et Biophysica Acta. 1132:F25-F48.

Smith, K.J., Johnson, K.A., Bryan, T.M. (1993) The APC gene product in normal and
tumor cells. Proc. Nat]. Acad. Sciences USA 90: 2846-50.

Smith, P.J. (1986) n—Butyrate alters chromatin accessability to DNA repair enzymes.
Carcinogenesis 7(3): 423-29.

Stephen, A.M. and Churms, S.C. Food polysaccharides and their applications, (Stephen,
A.M. ed.). Marcel Dekker, New York, 1995, p. 377-385.

Sn, L., Kinzler, K.W., Vogelstein, B., Preisinger, A.C., Moser, AR, Luongo, C., Gould,

K.A., & Dove, W.F. (1992) Multiple intestinal neoplasia caused by a mutation in
the murine homolog of the APC gene. Science 256: 668-72.

65

Thompson, J.S., Quigley, E.M., Palmer, J .M., et a]. (1996) Luminal short-chain fatty
acids and postresectional intestinal adaptation. J.P.E.N. 20: 338-43.

Thorup, 1., Meyer 0., Kristiansen, E. (1995) Effects of potato starch, cornstarch, and

sucrose on aberrant crypt foci in rats exposed to azoxymethane. Anticancer
research 15: 2101-2106.

Titgemeyer, E.C., Bourquin, L.D., Fahey, G.C., et al. (1991) Fermentability of various
ﬁber sources by human fecal bacteria in vitro. Am. J. Clin. Nutr. 53: 1418-24.

Topping, D.L. (1991) Soluble Fiber polysaccharides: Effects on Plasma Cholesterol and
Colonic Fermentation. Nutr Rev. 47(7): 195-203.

Trock, B., Lanza, E., Greenwald, P. (1990) Dietary ﬁber, vegetables, and colon cancer:
critical review and meta-analyses of the epidemiologic evidence. J Natl Canc Inst
82: 65 0-6].

Trowell, H. (1976) Deﬁnition of dietary ﬁber and hypotheses that it is a protective factor
in certain diseases. Am J Clin Nutr. 29:417-27.

Van Munster, I.P., Tangerman, A., and Nagengast, F.M. (1994) Effect of resistant starch
on colonic fermentation, bile acid metabolism, and mucosal proliferation. Dig.
Dis. Sci. 39(4): 834-42.

Velaquez, O.C., Zhou, D., Seto, R.W., et al. (1996) In vivo crypt surface
hyperproliferation is decreased by butyrate and increased by deoxycholate in

normal rat colon: associated in vivo effects on c-Fos and c-Jun expression.
J.P.E.N. 20: 243-50.

Visek, W.J. (1978) Diet and cell growth modulation by ammonia. Am. J. Clin. Nutr. 31:
8216-20.

Vogelstein, B., Fearon, E.R., Hamilton, S.R., et a]. (1988) Genetic alterations during
colorectal-tumor development. New Engl. J. Med. 319: 525-31.

Von Engelhardt, W., Gros, G., Burmester, M., et a]. (1994) Functional role of bicarbonate
in propionate transport across guinea-pig isolated caecum and proximal colon. J.
Physio]. 477(Pt 2): 365-71.

Von Engelhardt, W. and Rechkemmer, G. (1992) Segmental differences of short-chain
fatty acid transport across guinea-pig large intestine. Exp. Physio]. 77(3): 491-9.

Von Engelhardt, W.V. In Physiological and clinical aspects of short-chain fatty acids,

(Cummings, J.H., Rombeau, J.L., and Sakata, T., eds.), Cambridge University
Press, New York, 1995, p. 149-165.

66

Wallis, Y. and Macdonald, F. (1996) The genetics of inherited colon cancer. J. Clin.
Path. 49: M65-73.

Wasan H.S., Novelli, M., Bee J ., et a]. (1997) Dietary fat inﬂuences on polyp phenotype
in multiple intestinal neoplasia mice. Proc. Natl. Acad. Sci. USA 94(7): 3308-13.

Whitehead, R.H., Young, GR, and Bhatal, RS. (1986) Effects of Short-Chain Fatty
Acids on a new human colon carcinoma cell line (LIM1215). Gut 27: 1457-63.

Whiteley, LC. and Klurfeld, D.M. (2000) Are dietary ﬁber-induced alterations in
colonic epithelial cell proliferation predictive of ﬁber’s effect on colon cancer?
Nutrition and Cancer 36(2): 131-49.

Whiteley, L.O., Higgins, J.M., Purdon, M.P., et a]. (1996) Evaluation in rats of the dose-
response relationship among colonic mucosa] growth, colonic fermentation, and
dietary ﬁber. Dig. Dis. and Sciences 41(7): 1458-67.

Winawer, S.J., Zauber, A.G., Steward, E., et a]. (1991) The natural history of colorectal
cancer: opportunities for intervention. Cancer 67: 1143-49.

Yarnada, KM. and Geiger, B. (1997) Molecular interactions in cell adhesion complexes.
Curr. Opin. Cell Biol. 9(1): 76-85.

Yokoya, F., Imamoto, N., Tachibana, Y., et al. (1999) Beta-catenin can be transported
into the nucleus in a Ran-unassisted manner. Mol. Cell. Biol. 10(4): 1119-31.

Zoran, D.L., Turner, N.D., Taddeo, SS. (1997) Wheat bran diet reduces tumor

incidence in a rat model of colon cancer independent of effects on distal luminal
butyrate concentrations. J. Nutr. 127: 2217-25.

67

     

GANS BIRARE

HllllHll \Hl \llllllHlH H] H

3 1293 03574 6514