mml

I
r

—
—
__
=
—
—
—
—
—
—
—
—_
—
—

 

THESIS

l IUIIUHI(llIll!ll“IIHIIUIIUIIIIUIHI”UlNHllllHHl

c600 0483875

0’

This is to certify that the

thesis entitled

Fermentation Characteristics of Selected
Dietary Substrates: A Human In Vitro Study

presented by

Julie A. Arnold, MS, RD

has been accepted towards fulfillment
of the requirements for

Ni degree in mm ti on

 

WWW

Major professor

Date June 5, 2000

 

0-7639 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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

moo chIRC/DatoDuo.p65—p.14

 

FERMENTATION CHARACTERISTICS OF SELECTED DIETARY SUBSTRATES:
A HUMAN IN WWO STUDY

By

Julie Ann Arnold, RD

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF SCIENCE
DepartmentofFood SeieneeandHumanNuu'ition

2000

ABSTRACT

FERMENTATION CHARACTERISTICS OF SELECTED DIETARY SUBSTRATES:
A HUMAN IN VIIRO STUDY

By
Julie Ann Arnold, RD

The objective of this study was to use batch fermentation methodology to assess
the fermentability of several dietary substrates as measured by short chain fatty acid
production, organic and dry matter disappearance, and water holding and potential water
holding capacities. Specifically, molecular weight was examined as it relates to the
potential fermentability of B-glucan. The substrates used were arabinogalactan, high
viscosity B-glucan, medium viscosity B-glucan, low viscosity B-glucan, a mixture of
inulin and fi'uctooligosaccharide, a mixed fiber source, oat flour, oat groats, and Solka-
noc". The production of total short chain fatty acids by fermentation with human fecal
inocula ranged from 0.92 mmol per gram original organic matter (Sorrel-Hoe“) to 5.30
mmol per gram original organic matter (high viscosity B-glucan). Substrate organic
matter disappearance ranged ham -1 .79°/o (Sousa-Hoe") to 83.84% (mixed fiber) and I
was positively correlated with short chain fatty acid production (r = 0.77, p = 0.015).
Water holding capacity ranged from 3.79 grams water per gram fermented residue dry
matter (oat greats) to 6.76 grams water per gram fermented residue dry matter
(inulin/F08). Potential water holding capacity was inversely correlated with organic
matter disappearance (r = --0.91, p = 0.001). This data is suggestive of a relationship
between increasing molearlar weight and the fermentability of B-glucan, however further

researchisnecessarytoseeiftheexistenceoftheefi‘ectcanbeconfirmed.

Capyn'sht by
Julie Ann Arnold, RD
2000

I dedicate this thesis to my family: Mom, Dad, Kelly, and Alex.

ACKNOWLEDGMENTS

I would like to acknowledge the following people for the kindness and guidance
extended to me during my research period:
General Mills, Inc.
Department of Food Science and Human Nutrition, Michigan State University
Dr. Norman Hord
Dr. Les Bourquin
Dr. Joseph Schroeder
Dr. Jenny Bond
Dr. Maurice Bennink
- Dr. Sue Hengemuehle
Dr. Mel Yokoyama
Debie Lecato
Amy Malow
Andrea Padgitt
Elizabeth Rondini

MSU Statistical Consulting Service

TABLE OF CONTENTS

Introduction ..................................................................................................................... 1
Chapter 1 -— Literature Review ......................................................................................... 2
Chapter 2 - Materials and Methods
Subjects ...................................................................................................... 14
Fermentation Media Preparation ................................................................. l4
Fermentation Methodology ......................................................................... 16
Fatty Acid Analysis .................................................................................... 18
Statistical Analysis ..................................................................................... 19
Chapter 3 - Results ........................................................................................................ 20
Chapter 4 — Discussion .................................................................................................. 27
Chapter 5 - Conclusion ................................................................................................. 34
Appendix A - Substrate Composition ............................................................................ 37
Appendix B — Chemical Structures of Short Chain Fatty Acids ..................................... 41
Appendix C - Proposed Production, Absorption, and Utilization of Short Chain Fatty
Acids ....................................................................................................... 42
Appendix D — OMD, DMD, WHC, PWHC Raw Data by Donor ................................... 44
Appendix E — SCFA Production Raw Data by Donor .................................................... 46
Appendix F — Human Subjects Approval ....................................................................... 50
Bibliography ................................................................................................................. 52

vi

LIST OF TABLES

Table 1. Composition of anaerobic dilution medium ...................................................... 14
Table 2. Composition of medium used for in vitro fermentation .................................... 15

Table 3. Short chain fatty acid production (mmol / g original organic matter) post-
fermentation of experimental substrates including all donors ............................ 21

Table 4. Molar proportions of short chain fatty acids produced including all donors ...... 22

Table 5. Organic Matter Disappearance (OMB) and Dry Matter Disappearance (DMD)

post-fermentation of experimental substrates including all donors .................... 23
Table 6. Water Holding Capacity (WHC) and Potential Water Holding Capacity (PWHC)

post-fermentation of experimental substrates including all donors .................... 24
Table 7. All parameters of fermentation measured by donor .......................................... 25

vii

LIST OF FIGURES

Figure 1. SCFA production and possible absorption ...................................................... 42
Figure 2. Energy utilization of various respiratory fuels ................................................. 43

viii

LIST OF ABBREVIATIONS

DMD ................................................................ ' ........................ D ry Matter Disappearance
g ............................................................................................................................... Gram
Mmol ............................................................................................................... Millimoles
01W) .................................................................................. Organic Matter Disappearance
PWHC ........................................................................... Potential Water Holding Capacity
SCFA ............................................................................................ Short Chain Fatty Acid
WHC ............................................................................................ Water Holding Capacity

INTRODUCTION

The process of digestion and utilization of nutrients in humans is complex and
multifaceted. Certain fibers of plant origin escape normal digestion, as humans lack the
necessary enzymes for their hydrolysis. These fibers enter the large bowel relatively
unaltered and serve as substrates for fermentation by colonic microflora. The primary
end products of this fermentation are short chain fatty acids (acetate, propionate, and
butyrate), gases (C02, H2, and CHa), and microbial cells.

The objective of this research was to use in vitro batch fermentation methods to
predict the organic matter disappearance (OMD), short chain fatty acid (SCF A)
production, water holding capacity (WHC) and potential water holding capacity (PWHC)
of fiber products potentially of interest for health promotion and disease prevention. The
primary hypothesis to be tested in this study was that difi‘erences in the molecular weights
of B-glucan—containing substrates do not affect the extent of fermentation as measured by
the above indices. Other substrates were included in this study due to the funding
corporation’s proprietary interest in incorporating these substrates into processed cereal

products.

Chapter 1

LITERATURE REVIEW

The wall of the digestive tract consists of four layers. From the irmermost layer
outward, they are the mucosa, submucosa, muscularis extema, and the serosa. The
mucosa itself consists of three layers, the epithelium, the lamina propria, and the
muscularis mucosa. The epithelial layer within the large bowel consists of crypts, which
are approximately fifty cells deep and are the location of cellular growth (Potter, 1999).
Proliferation of cells within the colon is very rapid, as they are replaced every four to
eight days (T appenden et al., 1998). Colonocytes begin their cell divisiOn in the lower
60% of the colonic crypts and, once mature, move towards the surface and become highly
polarized. The health and proliferation of colonic cells requires a fuel source, of which
glucose, short chain fatty acids and glutamine may serve. The brush borders of these
colonocytes secrete enzymes, including hydrolases. The epithelium of the proximal
colon absorbs chloride, sodium, and water and secretes potassium and bicarbonate
(Scheppach, 1994).

The human colon contains about 222 grams of fecal matter distributed along the
length of the bowel and is highly populated with bacterial flora (Cummings, 1997). It is
estimated that 95% of the total cell count in the body consists of colonic bacteria, which
is equivalent to 1012 bacterial cells per gram of dry feces. The species colonization is
location-dependent within the intestine. The upper small bowel is dominated with
facultative anaerobes and aerobes such as streptococci, staphylococci, and lactobacilli

(Gibson et al., 1996). The bacterial population increases as the large bowel is reached, as

the pH is lower and transit time becomes longer (El Oufir et a1, 1996). The predominant
types of bacteria colonizing the lower gut include bacteroides, bifidobacteria, eubacteria,
clostridia, lactobacilli, Gram-positive anaerobic cocci, coliforms, methanogens and
dissimilatory sulphate-reducing bacteria (Gibson, 1998).

In general, intestinal bacteria do not contain mitochondria and do not use oxygen as a
terminal electron acceptor for respiratory processes (Cummings, 1997). Therefore, the
population of colonic microflora is maintained by anaerobic metabolism or the provision
of an exogenous source of bacteria.

A probiotic is defined as a live microbial dietary supplement that transiently increases
large bowel colonization of the selected bacteria. Bifidobacteria, lactic acid bacteria and
saccharomyces are currently the most commonly supplemented. These bacteria are non-
pathogenic, retain viability during storage, and survive passage into the large intestine of
humans (Macfarlane and Cummings, 1999).

Prebiotics are non-digestible oligosaccharides that selectively stimulate growth and/or
activity of existing colonic microflora (Macfarlane and Cummings, 1999). Supplemented
prebiotics are meant to stimulate the growth of beneficial bacteria, which conversely, can
competitively inhibit the growth of pathogenic bacteria. Prebiotic fibers are also utilized
asanutritionsourcebythebacteriainaprocesscalled famentation, whichresultsinthe
production of short chain fatty acids. Several food marmfacturing companies are
currently supplementing their products with prebiotics, such as fi'uctooligosaccharide
(F OS) and inulin.

The chief substrates for fermentation are non—digested dietary carbohydrates, with

amino acids and endogenously produced carbohydrates and glycoproteins contributing to

a lesser degree. The rate of mucin production, an endogenous carbohydrate, appears to
vary with the amount of fiber present in the diet, therefore the amount of host
polysaccharide available for fermentation is dependent on diet (Salyers, 1990). Only
mono— and oligosaccharides can enter the cytoplasm of a bacterial cell for fermentation.
Therefore, the presence of an enzyme to break 1,4-B glucose or xylose linkages is
required for the breakdown and subsequent fermentation of plant cell wall
polysaccharides, such as cellulose or xylan (Soergel, 1982). Therefore, a synergy among
bacteria may exist in the colon, as one bacterial strain may split a carbohydrate to small
units or sugars and other species may then ferment the products to short chain fatty acids
(Edwards, 1997).

The extent of carbohydrate fermentation, as well as the concentration and type of
products produced is dependent on several factors. The synthesis and ratios of short
chain fatty acids vary depending on the arrival rate and characteristics of the substrate,
the transit time of the fecal matter, as well as the composition and total mass of the
microflora present in the gut (Bourquin et al., 1993; El Oufir et al., 1996; McBurney,
1989; Soergel, 1982). Substrates vary in their degree of digestibility in many ways. High
concentrations of lignin present in the cell-wall material can reduce potential digestibility.
In addition, because the breakdown of fiber depends on the accessibility of the polymer
to colonic microflora, an increasing degree of water solubility should increase colonic
fermentation. It has been postulated that the types of sugars, the degree of cross linkages
and the presence of side chains a substrate contains influences the proportion and total
production of short chain fatty acids. For example, those substrates that contain

rhamnose, arabinose, xylose, ribose, galacturonic and glucuronic acids tend to promote

the formation of propionate. Butyrate is formed in higher proportions fiom the
fermentation of substrates containing sorbitol, galacturonic and glucuronic acids, and
ribose to a lesser extent (Mortenson et al., 1988). Other factors affecting substrate
digestibility include the amount of intake and the efi‘ects of food processing on chemical
structure or availability (Cummings, 1982).

The major end products of bacterial fermentation of carbohydrate are organic acids,
hydrogen and C02 (Cummings, 1997). The fermentation of proteins contributes
additional short chain fatty acids, as well as branched chain fatty acids. Intermediates
may be used as substrates for firrther fermentation: formic acid to C02 and H2; succinate
to propionate and H20; lactic acid to acetate, butyrate, propionate, and H2; and
methanogenic bacteria reduce H2 to C02 and methane(Soergel, 1982; Gibson et a1,
1996). Many of these compounds supply energy and have the ability to alter gut
morphology or produce systemic efi‘ects. .

Colonic fermentation can be studied in several ways. In vivo measurements of
fermentation products have significant limitations. It is dificult to obtain human colonic
contents, and those studies that have reported results in this fashion have utilized sudden
death victims (Cummings et al., 1987). In addition, because short chain fatty acids are
absorbed fi‘om the intestine and are metabolized by the liver and colonic epithelial cells, a
peripheral blood measurement of short chain fatty acid concentrations would not be
indicative of production. For the same reason, fecal concentrations of short chain fatty
acids are also not an efi‘ective measurement of production.

Standardized in vitro batch ferrnentations have been widely used to simulate colonic

activity and provides a rapid means of accurately quantifying estimates of fermentation of

specific sources of dietary fiber (Barry et al., 1995). A study examining the eficacy of
using human fecal inocula for batch fermentations found a 24-hour system using at least
three donors provides the most accurate estimates of colonic mechanisms (McBumey and
Thompson, 1989).

Organic matter disappearance (OMD) also estimates substrate fermentability, as it
approximates the loss of organic matter to digestion and fermentation. A study by
Bourquin et al. (1996) reported the production of short chain fatty acids is directly
correlated with organic matter disappearance.

The measurement of water holding capacity (WHC) is a representation of in viva
colonic water retention and is estimated by exerting an osmotic gradient through dialysis
tubing containing the fermented substrates of interest. The value obtained from this
method includes the water held by the substrates as well as the water holding capacity of
the bacteria present in the fecal matter. WHC represents the amount of water in grams
held by one gram of fermented residue dry matter. The measurement of water holding
capacity is ofinterest in human health because it is an attempt to estimate the fecal
bulking efi‘ects of the substrates of interest. Water holding is dependent on the manner in
which water is held by the food matrix and by the way WHC is measured. Fermentation
can alter the anatomy and physical properties of a fiber, therefore it is most accurate to
quantify the WHC of the fermented residue (Robertson and Eastwood, 1981).

Potential water holding capacity (PWHC) is an extrapolation of the WHC estimate
and represents the amount of water that would be excreted in the feces as a result of the
ingestion of one gram of the substrate of interest. PWHC takes the extent of fermentation

into account, and therefore is a measure of the relative WHC that equivalent amounts of

ingested fiber have in the colon (McBumey et al., 1985). PWHC is calculated by
multiplying WHC for the specific substrate by the amount of dry matter remaining of that
substrate following fermentation. PWHC is inversely related to in vitro OMD (Bourquin
et al., 1996).

The in vitro fermentation procedure does not account for variances in subject transit
time, therefore differences may exist in the comparison between these measurements and
actual colonic production of short chain fatty acids, WHC and PWHC. Finally, organic
matter disappearance will be slightly underestimated due to bacterial matter inclusion in
the dry matter recovery estimates.

Rates of short chain fatty acid production in humans range from 300-400 mmol per
day on the assumption of the daily ingestion of 32-42 grams carbohydrate (Cumrrrings,
1994). It is estimated that 95 to 99% ofthe short chain fatty acids produced by bacterial .
fermentation are absorbed fiom the colonic hmlen (Scheppach, 1994). The mechanism
for absorption of short chain fatty acids from the colon is not completely understood,
however certain aspects are understood through experimental evidence (See Appendix C
for detail). First, the transport of short chain fatty acids through the colonic epithelium I
appws to be concentration dependent and independent of chain length In addition,
absorption of short chain fatty acids may be coupled with sodium absorption, probably
via Na+ - H+ exchange. There is a luminal accumulation of HC03' with SCFA
absorption, which may be a byproduct of the CO: produced fi'om the Krebs cycle.
Finally, the rate of absorption of short chain fatty acids is thought to depend on the

location of absorption in the colon (Soergel, 1982; Scheppach, 1994).

The short chain fatty acids produced by fermentation have a number of physiologic
efi’ects. Acetate, propionate and butyrate may influence epithelial cell replication,
maturation, and death (Wijnands et al., 1999). Short chain fatty acids reduce colonic pH,
thus causing precipitation of the bile acids present in the colon. This results in an
inhibition of the formation of secondary bile acids, a product of bacterial metabolism of
primary bile acids. High concentrations of secondary bile acids have been shown to
increase colon cancer risk in high concentrations (Wijnands et al., 1999).

Short chain fatty acids are thought to influence homeostasis within the colon, as
metabolic starvation causes a degeneration and inflammation of the colonic epithelium
This efi‘ect is abated when supplementation of butyrate is administered (Tappenden et al.,
1997). Therefore, short chain fatty acids may prevent the TPN—induced atrophy of
normal and resected intestine by increasing basolateral intestinal nutrient transport (Sagor
et al., 1983; Tappenden et al., 1996; Tappenden et al., 1997). Specifically, short chain
fatty acid exposure increases jejunal GLUT2 mRNA, ileal GLUT2 protein, and ileal
proglucagon mRNA GLUT2 is a basolateral glucose transporter (T appenden et al.,
1997). Proglucagon-derived peptides are associated with increased cellular proliferation
during colonic adaptation (Bloom and Pollak, 1982; Rountree et aL, 1992; Sagor et al.,
1983; Reimer et al., 1996).

Each short chain fatty acid is thought to have several unique physiological fimctions
throughout the body. The concentrations of acetate and propionate in the portal
bloodstream is much higher than that ofbutyrate, which is suggestive ofa selective use of

butyrate as an oxidative fuel source for colonic epithelial cells. The afinity of the

colonocytes for butyrate as an energy source is higher in the distal than the proximal
colon (Scheppach, 1994).

Acetate is thought to act as a minor energy source for body tissues and as a result,
spares fatty acid oxidation in humans. The chemical structure for acetate is shown in
Appendix B. Acetate has also been found to inhibit lypolysis and glycerol release in
adipocytes (Cummings, 1988). Acetate may play a role in carbohydrate metabolism as
well. Studies have shown that acetate fimctions to reduce serum fiee fatty acid
concentrations (W clever, 1988). Because free fatty acids compete with glucose for
uptake within insulin sensitive tissues, fewer fatty acids present in the serum will lead to
improved uptake of glucose into the cells.

A study of dog colon showed that acetate mediated an increase in mucosal blood flow.
The same efl‘ect of acetate was found in surgical patients in a concentration-dependent
manner (Tappenden et al., 1997). Finally, acetate is thought to increase colonic motility
and may be involved in the chemical reflux barrier within the gastrointestinal tract
(Scheppach, 1994).

There is little research on the physiological efl‘ects of propionate in humans.

The chemical structure of propionate is shown in Appendix B. One study showed an
increase in HDL cholesterol levels in those individuals who ingested 7.5 grams of sodium
propionate per day as compared to the control group (V enter et al., 1990). Propionate is
thought to mediate this reaction by inhibiting hepatic cholesterol synthesis. In addition,
propionate is thought to act as a gluconeogenic agent. One study has shown that rectal
infusion of propionate caused an increase in blood glucose, which is consistent with the

action of a gluconeogenic agent (W clever, 1988). However, the rate of propionate

infilsed (180 mmol) is higher than the concentration present in the colon, and therefore
propionate is probably not a significant source of blood glucose in humans.

Butyrate is a linear four-carbon short chain fatty acid that is taken up by colonocytes
where it is metabolized to CO; and ketones. Butyrate is considered the chief energy
source for these cells. The chemical structure of butyrate is shown in Appendix B. It is
estimated that butyrate supplies 70% of the firel for colonic cellular growth and
development (Hague et al., 1997). The importance of butyrate as a respiratory fuel
source appears to increase from proximal to distal colon, while the importance of
glutamine decreases (Roediger WE, 1982).

Butyrate has been found by many studies to induce growth of colonic mucosa] cells.
Butyrate-induced difl‘erentiation leads to the expression of a normal absorptive cell
phenotype (Basson et al., 1998). However, butyrate afi‘ects cancer cells in the opposite
manner. Butyrate has been shown to inhibit growth and induce apoptosis, or
programmed cell death, and difi'erentiation within malignant cells (Hague et al., 1997).
The mechanisms responsible for these seemingly disparate efl‘ects are not fully
understood. Cell number homeostasis in colonic epithelial cells is believed to be
sustained by an equih'brium between cell proliferation and apoptosis. Apoptosis is a non-
pathophysiologic type of cell death inherent in normal cell (Bedi et al., 1995).
Specifically, Bel-2, which inhibits apoptosis, is believed to counteract Bax, a protein that
induces apoptosis of colonic cells within the crypt. One study found that the withdrawal
of butyrate from the colon resulted in an increase in apoptosis and Bax concentration
(Hass et al., 1997). Conversely, in normal cells, treatment with butyrate resulted in

decreased apoptosis and decreased Bax concentration within the colonic epithelium

10

(Hague et al., 1997). Therefore, not only is butyrate the major energy source for the
colonic mucosa, it also acts as a protective agent against cell death.

Butyrate is thought to inhibit DNA synthesis and hinder tumorogenic cell
proliferation in the G1 phase of the cell cycle (Young, 1988). A study reported a
significant decrease in Caco-2 cell proliferation as a result of treatment with 10mM
butyrate (Basson et al., 1998). Acetate and propionate also inhibited cellular
proliferation, however their efi‘ects were not as pronounced as butyrate. In addition to its
efl‘ects of cancer cell proliferation, butyrate inhibited metastatic expansion and adhesion
as well. Another study found that sodium butyrate inhibited the activity of SNB-19 and
SNB-75 cancer cells (Gross et al., 1988).

The process by which butyrate induces cell death or reduced difl’erentiation in
malignant cell lines is not fully understood, however, several hypotheses have been
suggested. First, butyrate is thought to effect cellular proliferation by the
hyperacetylation of histones via the inhibition of histone deacetylase. Histones fimction
to package DNA into nucleosomes. This action causes the loss of the positive charge on
histone and subsequent weakening of the phosphate group within the DNA strand. This
alteration induces a change in the chromatin structure and subsequent arrest of the cell
cycle at the G1 stage, which limits cell proliferation (Young, 1988; Bofl‘a et al., 1992).
Hyperacetylation is also associated with increased transcriptional activity within areas of
the cell which produces the activation of specific genes (Archer et al., 1998).

Histone hyperacetylation may be responsible for the induction of p21 WAFI/Cr’p], a
cell cycle inhibitor (Archer et al., 1998). p21 WAF 1/Cip1 inhibits certain G1 cyclin-

dependent kinases by the suppression of the phosphorylation of retinoblastoma (RB)

11

protein (Nakano et al., 1997). There is some evidence that p21 WAFI/Cipl plays a role
in the assembly and activation of kinase complexes as well. A tumor suppressor protein,
p53, appears to mediate the activity of p21 WAFI/Cipl under certain conditions,
however, butyrate afi‘ects p21 WAFI/Cip] action independently of p53 activation.

Archer et al. found that p21 WAFI/Cipl is required for the growth inhibition that butyrate
mediates (Archer et al., 1998; Nakano et al., 1997). Taken together, these data support
the hypothesis that p21 WAFI/Cipl is a critical mediator of growth control exerted by
butyrate.

There are several oncogenes that are responsible for regulating cellular growth and
difi‘erentiation within the colon. C-myc is believed to control the production of a nuclear
phosphoprotein that is involved in transcription of specific genes involved in cell growth
control. C-fos and c-jun are responsible for activating DNA binding proteins that form
the AP-l transcription factor. AP-l transcription factor may control genes involved in
cellular growth and difi‘erentiation (Tappenden et al., 1998). Therefore these agents may
control for cellular differentiation and growth, as well as apoptosis (Tappenden et al.,
1998). Western blotting of Caco-2 lysates, a cancerous cell line, with antibodies to c-
myc revealed that propionate, acetate, and butyrate downregulate c-myc 1 and 2 protein
levels, with butyrate having the most significant efl‘ect (Basson et al., 1998; Velazquez et
al., 1996).

Prebiotics efi‘ectively stimulate the production of short chain fatty acids, including
butyrate, as measured by in vitro batch fermentation indices. Short chain fatty acids are
thought to impart benefits not only to the health of the colon, but to systemic health as

well. These above data provide a rational basis for the use of in vitro fermentation

12

techniques to investigate the fermentative capacity of the specific substrates used in this

study.

13

Chapter 2

MATERIALS AND METHODS

Subjects

Two healthy adult females (age 25 and 28) and one healthy adult male (age 35)
participated in this study. In addition, one male child of age eleven was included. All
subjects were encouraged to ingest at least twenty-five grams of dietary fiber from
varying sources per day. Subjects had not been on antibiotic therapy during the three
months preceding the study.
Fermentation Media Preparation

The in vitro batch fermentation procedure was adapted fi'om previously used
methodology (Bourquin et al., 1996). Anaerobic dilution (Table 1) and fermentation

media (Table 2) were made according to the following composition guidelines.

Table l: Corrrposition of anaerobic dilution medium (Bryant and Burkey, 1953)

 

 

Component Concentration in medium
Mineral 1‘ 150 mIJL
Mineral 2 " 150 mL/L
Resazurin 0.0003 g/L
Cysteine hydrochloride 0.09 g/L
Sodium carbonate 0.32 g/L
Distilled water Fill to 1L mark

 

' Composition: KI'IzPO4, 0.06 g/L.
" Composition (g/L): KH2P04, 0.06; (thhsor, 0.12; NaCl, 0.05; MgSOa, 0.18; each,
0.01.

One liter of the anaerobic dilution medium was used. After the appropriate amounts

of the above constituents were mixed, the solution was capped with cotton batting and

autoclaved. Following the autoclave process, carbon dioxide was bubbled through the

14

solution for thirty minutes. The solution was then capped with a rubber stopper and
stored at room temperature until use the same day.

Table 2: Composition of medium used for in vitro fermentation
(Bourquin et al., 1996)

 

 

Component Concentration in medium
Solution A ' 330 mIJL
Solution B # 330 mL/L
Trace Mineral Solution ° 10 mL/L
Water-soluble vitaminmix’ 20 mL/L
Folate: biotin solution 5 5.0 mL/L
Riboflavin‘ 5.0 mL/L
Hemin solution § 2.5 mL/L
Short-chain fatty acid mix " 0.4 mL/L
Resazurin solution ## 1.0 mL/L
Distilled water 296 mL/L
Yeast extract 0.5 g/L
Trypticase 0.5 g/L
N32C03 4.0 g/L
Cysteine HCl-H20 0.5 g/L

 

'Composition (g/L): NaCl, 5.4; KH2P04, 2.7; CaClz-HZO, 0.16; MgClg-6H20, 0.12;
MnC12-4H20, 0.06; COC12.6H20, 0.06; (N'HahSOg 5.4.

i’Composition: KH2P04, 2.7 g/L.

°Composition (mg/L): EDTA, 500; FeSOa-7H20), 200; ZnSOa-7H20, 10; MnClgAHgO,
3; H3PO4, 30; COC12'6H20, 20; CDC12°2H20, 1; NrC12-6H20, 2; N32M004°2H20, 3.
*Composition (mg/L): thiamin-HCI, 100; pantothenic acid, 100; niacin, 100; pyridoxine,
100; p-aminobenzoic acid, 5; vitamin B~12, 0.25.

5Composition (mg/L): folic acid, 10; biotin, 2; NHaCO3, 100.

1Composition: riboflavin, 10 mg/L in 5 mmol/L I-IEPES.

§Hemin, 500 mg/L in 10 mmol/L NaOH.

”250 leL each of n-valerate, isovalerate, isobutyrate, and DL-ot-methylbutyrate.
”Resazurin, 1 g/L in distilled H20.

Three and one-half liters of fermentation media were needed for this study. All
components but the water-soluble vitamins, Na2C03 and cysteine hydrochloride were
mixed and transferred to four one-liter bubble flasks. These flasks were individually
heated with a Bunsen burner under a constant stream of C02 until they reached a boil.

All flasks were capped with rubber stoppers and wired shut before sterilizing the medium

15

via autoclave. Just prior to use, the water-soluble vitamins were added via filter syringe.
The other omitted components were added at this time as well.
Fermentation Methodology

Two sets of fermentations were performed simultaneously. The first fermentation was
carried out in one hundred milliliter serum vials and was used to assess organic matter
disappearance and short chain fatty acid production. The second set of fermentations was
done in twenty-five milliliter Balch tubes and was used in the quantification of water
holding capacity and potential water holding capacity. Each substrate was fermented in
duplicate for each subject. .

Substrates used in this study were arabinogalactan, low viscosity B-glucan, medium
viscosity B—glucan, high viscosity B-glucan, mixed fiber, inulin/F08, Solka-FlocO, oat
flour, and oat groats (see Appendix A for substrate composition). Three hundred
milligrams and one hundred and fifteen milligrams of the substrates were weighed and
placed in the serum vials and Balch tubes, respectively. All weights were recorded.
Aliquots (26ml) of fermentation medium were aseptically transferred under a C02 stream
to the serum vials using 50 milliliter pipettes. Ten milliliters of medium were added to
the Balch tubes in the same manner. The serum vials and Balch tubes were capped with
butyl rubber stoppers and stored under refrigeration (4° C) for twelve hours to allow for
hydration of the substrates. Tubes containing no substrates were also included in
duplicate for each subject to allow for corrections of organic matter and short chain fatty
acid content not arising fiom the substrates.

The procedure for the preparation of the fecal slurly is as follows. Each subject

donated a flesh fecal sample from which the slurry was prepared within four hours. The

16

fecal samples were sealed in plastic bags and were held under refi'igeration (4° C) until
use. Twenty-eight grams of feces were blended using a Hamilton Beach blender under
anaerobic conditions with two hundred and eighty milliliters of anaerobic dilution
medium (1:10 dilution) for fifteen seconds. This mixture was filtered through four layers
of cheesecloth into an Erlenmeyer flask and capped under carbon dioxide until use.

The sets of fermentations were staggered by two hours by subject to allow for
completion of all tasks. The hydrated substrates in the serum vials and Balch tubes were
anaerobically inoculated using ten milliliter pipettes with four milliliters and one and one
half milliliters of the fecal slurry, respectively. The tubes were capped with butyl rubber
stoppers and crimped closed using aluminum caps. Tubes were allowed to incubate at
37° C for twenty-four hours with periodic mixing.

After twenty-four hours, a one-milliliter aliquot was removed from the serum vials
using 3 milliliter syringes for short chain fatty acid analysis. one hundred and twelve
milliliters of 95% ethanol were added to the remaining twenty-nine milliliters in each
serum vial and the solutions were allowed to precipitate for one hour. The contents of the
serum vials were then vacuum filtered through tared Whatman 541 filter paper, and
washed with 95% ethanol and acetone. The filter paper and residue were dried in a 105°
C drying oven (Precision Scientific Model 124 drying oven) for twenty-four hours. After
desiccation, the filter paper and residue was weighed and transferred to tared crucibles.
The crucibles were placed in an ashing oven (500° C) for twelve hours. After
desiccation, the crucibles and their contents were weighed. In vitro organic matter

disappearance was calculated as the original organic matter weight minus (residue

17

organic matter weight minus the organic matter weight of the appropriate blank), divided
by the original sample organic matter weight (Bourquin et al., 1996).

Fermentation in the Balch tubes was stopped after twenty-four hours by the addition
of 0.4 milliliters of 20 g/L CuSOa. Dialysis tubing (Spectra-For 6, 45mm, 2000 dalton
cut-ofi‘, Spectrum Medical Industries, Inc., Los Angeles, CA) was cut into 3” strips and
pre-soaked for twelve hours in 0.53/1. NaNa. The contents of the Balch tubes were
transferred to dialysis tubing and clamped using dialysis clips. The clamped dialysis bags
were placed in a solution containing 95g/L polyethylene glycol (MW=3 500) and 0.5g/L
NaNa. The bags were dialyzed at room temperature with periodic mixing for forty-eight
hours. The dialysis solution was changed at this time and the tubes were dialyzed for an
additional twenty-four hours. The bags were then blotted dry with paper towel and cut
open. - The contents were transferred to tared crucibles and weighed. The crucibles were
placed in a 105° C drying oven for 24 hours and reweighed. Water holding capacity was
calculated as the grams of water lost during the drying process, divided by the dry residue
weight. Potential water holding capacity was calculated as the water holding capacity
multiplied by the percentage of dry matter remaining following the twenty-four hour
fermentation period.

Fatty Acid Analysis

The one-milliliter samples extracted for SCFA analysis were treated with 0.25 mL 250
g/L metaphospholic acid and centrifuged at 10,000 x g for 30 minutes (Beckman Allegra
6R Centrifuge). The supernatant was transferred to 2-ml glass GC vials and refiigerated

(4° C) until analysis.

18

Analysis for acetate, propionate, and butyrate content of the samples was done using a
Varian Model 3700 Gas Chromatograph (GC) equipped with a hydrogen flame-ionization
detector. Branched chain amino acids were not quantified because several strains of
anaerobic bacteria require these for growth and therefore they were added to the
fermentation media (Y okoyama and Johnson, 1979). The two-meter glass column used
for analysis was 10% SP1200/ 1% H3P04 on 80-100 Chromosorb. The analysis was done
using a temperature-timed program, in which the initial oven temperature setting was
120° C for one minute, with an increase of one degree per minute until the final
temperature of 127° C was reached and held for one minute. Nitrogen was used as a
carrier gas with a flow rate of 50 milliliters per minute. The injector temperature was
150° C. Two microliter quantities were used to inject samples in duplicate.

Statistical Analysis

Analysis for area under the curve data was done using Peak Simple 11 software,
version 3.7. Areas under the curve for each injection were compared to a known standard
and converted to millimoles per gram original organic matter. Molar proportions of each
short chain fatty acid produced were calculated by dividing the respective amount by the
total short chain fatty acid concentration.

The data analysis was done using a randomized complete block design using the
General Linear Models procedure of SAS (SAS Institute, Inc., 1996). Inoculum donors
were used as blocks. When significant substrate efl‘ects were detected (p<0.05), means
were compared using the Least Significant Difi'erence Method (Carmer and Swanson,

1974). Pearson’s r correlations were computed using SPSS 7.5.

19

Chapter 3

RESULTS

The result for the fermentations with regard to short chain fatty acid production can be
found in Table 3. Please see Appendix D for experimental raw data. The highest
concentration of total short chain fatty acids was produced from the fermentation of high
viscosity B-glucan. However, this yield was not statistically difi‘erent from that of mixed
fiber, oat flour, or inulin/F OS (p<0.05). The lowest short chain fatty concentration
resulted from the fermentation of Solka-Floc‘D and oat groats, which were both
significantly lower than the other fermented substrates (p<0.05). The rank order of
substrates from highest SCF A production to lowest is as follows: high viscosity [3-
glucan> inulin/F05 > oat flour > mixed fiber > arabinogalactan > medium viscosity [3-

glucan > low viscosity B-glucan > oat groats > Solka-Floc".

20

Table 3. Average short chain fatty acid production (mmol / gram original organic matter)
of experimental substrates including all donors

 

 

Acetate Propionate Butyrate Total SCFA
Arabinogalactan 2.05 0.75 0.49 3.29
High Viscosity B-Glucan 3.22 0.75 1.32 5.30
Medium Viscosity B-Glucan 1.87 0.52 0.86 3.25
Low Viscosity B-Glucan 1.74 0.51 0.74 2.99
SoIka-Floc 0.52 0.19 0.21 0.92
Inulin/FOS 2.27 0.44 2.26 4.97
Mixed Fiber 2.22 0.47 1.61 4.30
Oat Flour 2.56 0.55 1.55 4.67
Oat Greats 0.68 0.19 0.39 1.25
SEM 0.33 0.07 0.21 0.40
LSD“ 1.00 0.23 0.65 1.31

 

‘ Least significant difference between any two mean values in the same column (p<0.05)

Fermentation of the high-viscosity B-glucan resulted in higher acetate production with
respect to all substrates except inulin/F OS and oat flour (p<0.05). Fermentation of the
oat groats and Solka-FlocG resulted in the least production of acetate. The yield of
acetate fi'om highest to lowest from the fermentation of the experimental substrates is as
follows: high viscosity B-glucan > oat flour > inulin/FOS > mixed fiber > arabinogalactan
> medium viscosity B-glucan > low viscosity B-glucan > oat groats > Solka-Floc". The
molar proportion (Table 4) of acetate produced was approximately the same for all

substrates (~50-60°/o).

21

Table 4. Average percentage of short chain fatty acids produced post-fermentation of
experimental substrates including all donors

 

Molar Molar Molar

 

Proportion Proportion Proporation
Acetate ' Propionate ‘ Butyrate ‘

Arabinogalactan 62.4% 22.7% 14.9%
High Viscosity B-Glucan 60.5% 14.1% 25.4%
Medium Viscosity B-Glucan 57.4% 17.3% 25.3%
Low Viscosity B-Glucan 58.6% 18.1% 23.3%
Solka-Floc 56.4% 20.9% 22.8%
Inulin/FOS 46.7% 8.7% 44.6%
Mixed Fiber 51.9% 10.9% 37.2%
Oat Flour 52.0% 13.2% 34.8%
Oat Groats 52.3% 15.0% 32.7%
Standard Error of the Mean 4.10 1.90 3.78

LSD“ 13.74 5.85 11.45

 

‘LeastsignificmtdiflmebeMeenmytwomeanvaluesmthemewlumnMDS

Propionate was produced in the highest quantities during the fermentation of the high
viscosity B—glucan and arabinogalactan, though these were not significantly difi‘erent in
comparison to oat flour (p<0.05). Again, oat groats and Solka-Floc° resulted in the least
amount of propionate production (p<0.05). The production of propionate fi'om highest to
lowest concentration is as follows: high viscosity B-glucan > arabinogalactan > oat flour
> medium viscosity B-glucan > low viscosity B-glucan > mixed fiber > inulin/F OS >
Solka-Floc" > oat groats. The molar proportions of propionate ranged from the lowest
for inulin/F08 to the highest for arabinogalactan, at 9 and 23% respectively.

Butyrate was produced in significantly higher concentrations during the fermentation
of inulin/F08 in comparison to all other experimental substrates (p<0.05). The ranking
from highest to lowest butyrate production is as follows: inulin/F OS > mixed fiber > oat

flour > high viscosity B-glucan > medium viscosity B-glucan > low viscosity B-glucan >

22

arabinogalactan > oat groats > Solka-Floc". Molar proportions of butyrate produced
difi‘ered quite significantly with respect to substrate. The fermentation of inulin/F OS
resulted in the highest molar proportion of butyrate (45%), while arabinogalactan
fermentation resulted in the lowest molar proportion of butyrate (15%).

Table 5. Average organic matter disappearance (OMD) and dry matter disappearance
(DMD) post-fermentation of experimental substrates including all donors

 

 

OMD ’ DMD ‘
Arabinogalactan 24.16 ‘ 24.91
High Viscosity B—Glucan 36.49 38.08
Medium Viscosity B-Giucan 17.81 - 19.02
Low Viscosity B—Glucan 15.81 17.27
Solka-Floc -1.79 —0.75
Inulin/FOS 76.93 78.49
Mixed Fiber 83.84 ' 84.98
Oat Flour 45.31 46.45
Oat Groats 7.86 8.08
Standard Error of the Mean 4.00 4.13
LSD“ 12.32 13.17

 

*Leastsignficamdifi'amcebaweenanytwomemvalmmmesamecdeDS)
lOMD=orgnicmatterdisappearance(%),DMD=drymatterdimppearance(%)

Organic matter disappearance was significantly higher for mixed fiber and inulin/FOS
than all other experimental substrates (p<0.0001). Oat groats and Solka-Floc" were the
most poorly fermented substrates as measured by OMD. OMD fiom highest to lowest is
as follows: mixed fiber > inulin/FOS > oat flour > high viscosity B-glucan >
arabinogalactan > medium viscosity B-glucan > low viscosity B-glucan > oat groats >

23

Solka-Floc". Organic matter disappearance was positively correlated with total short

chain fatty acid production (r——- 0.77, p=0.015).

Table 6. Average water holding capacity (WHC) and potential water holding capacity
(PWHC) post-fermentation of experimental substrates including all donors

 

 

WHCl PWHCl
Arabinogalactan 6.20 4.66
High Viscosity B-Glucan 5.70 3.51
Medium Viscosity B-Glucan 5.25 4.34
Low Viscosity B-Glucan 5.65 4.60
Solka-Floc 4.46 4.51
Inulin/FOS 6.76 1.45
Mxed Fiber 6.34 . 0.97
Oat Flour 4.13 2.10
Oat Groats 3.79 3.46
Standard Error of the Mean 0.33 0.34
LSD* 0.69 ' 0.83

 

*LeamsignifimmdiflerencebetweenmytwomeanvaluesmmesamemlumnMDS)
‘ WHc=gH201grermentcdresimtcmymattcn PWHC=-gH20/goriginalsubstraterhymatter

Water holding capacity of fermented residues was highest for inulin/F OS, mixed fiber,
and arabinogalactan. WHC was the lowest for oat flour and oat groats, both around 4
grams water per gram of residue. WHC fiom highest to lowest is as follows: inulin/F OS
> mixed fiber > arabinogalactan > high viscosity B-glucan > low viscosity B-glucan >
medium viscosity B-glucan > Solka—Floc” > oat flour > oat groats.

Potential water holding capacity was highest for arabinogalactan, medium and low
viscosity B-glucan and Solka-Floc”. PWHC was lowest for mixed fiber and inulin/F08.
PWHC fiom highest to lowest is as follows: arabinogalactan? low viscosity B-glucan >

Solka-Floc" > medium viscosity B-glucan > high viscosity B-glucan > oat groats > oat

24

flour > inulin/F OS > mixed fiber. Potential water holding capacity was inversely

correlated with organic matter disappearance (r= -0.91, p=0.001).

Table 7. All parameters of fermentation measured by donor

 

 

 

Donor 1

Item A B C D SE

Total 24-h SCFA production 3.56 " 3.84 " 2.71 ' 3.64 " 0.27
Acetate 1.75 “' 2.36 ' 1.53 " 1.98 "’ 0.22
Propionate 0.41 b 0.46 b 0.45 " 0.62 ' 0.05
Butyrate 1.39 ' 1.03 '1 0.74 " 1.04 "’ 0.14
Percent oftotal - Acetate 2 51.2% ' 60.4% b 55.7% "’ 54.0% '” 2.73
Percent oftotal - Propionate 2 12.8% ' 13.1% ' 18.2% " 18.5% " 1.26
Percent of total - Butyrate 2 36.0% ' 26.5% " 26.1% b 27.5% b 2.52
DMD 3 36.99 ' 44.98 ' 25.97 " 32.74 "’ 2.76
OMD 3 36.14 ' 41.28 ‘ 26.66 " 32.10 "’ 2.67
WHC 3 6.03 ' 5.37” 5.20 " 1 4.87 " 0.22
PWHC’ 364 2.77b 3.65 309‘” 023

 

“Valuesaremeansandsmdarderra’ Memswithinarowfiallomdbymesamesymbolarenotsignificantlydifiarmt(p<0..05)
DMD=drymattudisappearance(%), OMD=agmicmdisappeuance(%),WHC=mholdingcapacity(gH10/g
fermented resichedryrnatta'), PWHC= potaltial water holding capacity(gH20’gorig'nalmbstrutedrymatter)

Donor C produced significantly fewer total short chain fatty acids than all other
donors (p<0.05). Donor D produced significantly more propionate than all other donors
(p<0.05), however the molar proportion of propionate produced was not significantly
higher than donor C. The molar proportion of butyrate produced by donor A was
significantly higher than all other donors (p<0.05).

Organic matter disappearance and dry matter disappearance followed a similar trend,
with donor B contributing to the highest of both indices, though not significantly difi‘erent
from donors A and D. Donor C had the lowest OMD and DMD estimates, however was

not significantly difi‘erent from donor D.

25

Finally, donor A had the highest WHC (p<0.05) than all other donors. Donor D had
the lowest WHC, averaging around 5 grams water per gram of residue dry matter.
Potential water holding capacity was highest for donor C, at an average of 3 .65 grams
water held per gram of original dry matter. PWHC for donor C was not significantly

difi‘erent from donors A and D.

26

Chapter 4

DISCUSSION

The overall results of this study are in agreement with previously published data with
regards to trends of fermentability and in most cases, absolute measurements of several
indices. The nature of this research, however, makes absolute quantification and
comparison between laboratories and substrates dificult. A comparative analysis of
OMD and SCFA production and molar ratios reported significant differences among five
laboratories using identical substrates and assigned to replicate the same batch
fermentation study using human subjects (Barry et al., 1995). However, despite
interlaboratory difi‘erences in the magnitude of various fermentation parameters, there
was agreement with respect to ranldng of fermentability among substrates. Because
difl‘erences exist in comparisons of the same substrates fermented at different
laboratories, it becomes even more dificult to make accurate comparisons with those
studies following different procedural methods. Differences in the amount and brand of
substrate added to the fecal inoculum may result in difi‘erences in end product
fermentation measurements. The preparation of the fecal inoculum, fermentation time or
the degree to which the enviromnerrt for fermentation was kept anaerobic would also
afl‘ect the results. Resazruin, a redox indicator, was added to the media and allowed for
assurance that the fermentations remained anaerobic in this study. In addition, there are
many studies that report SCFA and OMD results in swine or rat models. While
comparisons between humans and these models have been shown to be relatively
consistent, there is still some question as to the legitimacy of this practice (Barry et al.,

27

1995). Lastly, one must acknowledge that experimental error may be a factor that could
explain the lack of agreement of certain measurements with previous studies.

Thefermentationoftheninefibersusedinthisstudyresultedinameanyieldofshort
chain fatty acids in range with other published data acquired using similar techniques. In
this fecal inoculation system, acetate was the primary short chain fatty acid produced,
which agrees with many published data (Bourquin et al., 1996; Flickinger et al., 2000;
McNeil, 1982; Vince et al., 1990). Total short chain fatty acid production was positively
correlated to organic matter disappearance (r = 0.77, p = 0.015). Organic matter
disappearance is an estimate of the extent to which the substrate is utilized by colonic
bacteria. Therefore, as expected, the higher the organic matter disappearance, the greater
the short chain fatty acid production. This finding is in agreement a previous finding
(Bourquin, 1996). Potential water holding capacity was found to be inversely correlated
to organic matter disappearance (r = -0.91, p = 0.001). This result was expected given
that the more fermented the substrate is by bacteria, the less there will be available to
hold water post-fermentation. Given this general agreement, each substrate will be
discussed with respect to the indices measured.

Arabinogalactan is a highly branched water-soluble hemicellulose consisting of a
galactan backbone with side chains of galactose and arabinose sugars (Vince et al., 1990).
In this study, arabinogalactan was found to be moderately fermentable (3.29 mmng
original organic matter), resulting in a high molar proportion of propionate (23%). This
is in agreement with findings by other research groups (Bradbum et al., 1993; Cummings
and Macfarlane, 1997;ng and Gibson, 1993). Vince et al. (1990) reported a higher

molar ratio of butyrate over pr0pionate resulting from the fermentation of

arabinogalactan. However, this study difi'ered in methodology, as the fermentation
duration was forty-eight hours, and thus may account for the difi‘ering findings. Water
holding capacity was relatively high for arabinogalactan (6.20 grams water per gram
fermented residue dry matter). This finding correlates to the relatively low organic
matter disappearance.

Three barley B-glucan substrates of difl‘ering viscosities were used in this study. B-
glucan is a linear, unbranched polysaccharide composed of about 70% 4-O-linked B-D-
glucopyranosyl units and 30% 3-O-linked B-D-glucOpyranosyl units (Wood, 1990). B-
glucan exists in varying amounts in the endosperm region of several plant products. The
largest amount exists in barley (3-11%), followed by oats (3 -7%), rye (1-2%), and wheat
(<1%). Oat and barley Bcglucans appear to have the same structural identity, however
there may be some slight difi‘erences in the oligosaccharide fragments present (Mazza,
1998). There is some evidence that viscosity may play a role in the availability of the
substrate for fermentation, as well as the rate of starch digestion and glucose absorption
(Knudsen et al., 1990; Sundberg et a1, 1996). Varying levels of viscosity are achieved
during the extraction process, which reduces the molecular size relative to the native cell
wall polymer of B-glucan (Wood, 1990).

In this study, the fermentation of the high viscosity B—glucan product produced
significantly more short chain fatty acids than the two lower viscosity B—glucan products.
However, the results fi'om this study do not indicate that high viscosity B-glucan exceeds
other known substrates in its ability to produce short chain fatty acids. The organic
matter disappearance of all three B-glucans products was lower than expected the
substances’ characteristically high solubility. The OMD of the high viscosity B-glucan,

29

36.49%, is within range of previous findings of 46.2% OMD (McBumey, 1991). The
medium and low viscosity B—glucans, at 17.81% and 15.81% OMD respectively, are not
within range of previously published data. This may be a result of the processing of the
barley B-glucan, making it chemically less available for fermentation. Experimental error
may also play a role in these consistent, yet seemingly disparate findings. It should be
noted, however, that the small aniount of B-glucan fermented resulted in a high
production of short chain fatty acids, in comparison to other substrates with higher
organic matter disappearances. The fermentation of the high viscosity B-glucan resulted
in the highest production of short chain fatty acids, at 5.30 mmol per gram original
organic matter. None of the B-glucan substrates, however, were a significant source of
butyrate, which agrees with previous findings in swine (Knudsen, 1993), but disagrees
with other published data (McBumey, 1991). Due to the low disappearance of the B-
glucan substrates, the WHC of all three was moderately high. Potential water holding
capacity of all three substrates was inversely related to OMD.

The fermentation of Solka-Floco, a crystalline cellulose product, led to the expected
lowest production of short chain fatty acids (0.92 mmng original organic matter). This
finding corresponds to those of many other studies of the same nature (Barry et al., 1995;
Ehle et al., 1982; Penner and Liaw, 1990; Sunvold et al., 1995). The highly crystalline
structure of Solka-Floc° makes this substrate resistant to enzymatic hydrolysis within the
colon. The mean OMD for Solka-FlocQ was —1 .79%, illustrating the inert nature of this
compound in this system. The negative value for substrate disappearance has been noted
by other laboratories, and because it is not significantly difi‘erent than zero, can be

explained by experimental error (Barry et al., 1995; Sunvold et al., 1995). There is some

30

evidence that some individuals are able to transiently degrade this cellulose product,
however this was not confirmed fi'om this study (Ehle et al., 1982).

Inulin and fi'uctooligosaccharide (F OS) were used in combination as an experimental
substrate in this study. Both substances are naturally occurring polyfi'uctans found in
several commonly consumed vegetables as storage carbohydrates and have been shown
to be resistant to digestive processes (Brighenti et al., 1999). The chemical structure of
FOS consists of a glucose unit joined by an ((11-32) linkage to two or more fi'uctose
units. Fructose units are joined to one another by 3(2-1) linkages (Tashiro et al, 1997).
The degree of polymerization characterizes inulin and FOS, which is 2-8 and 10-12,
respectively. The OMD for the inulin/F OS product was 76.93%, suggesting virtually
complete fermentation by colonic bacteria as the majority of the remaining organic matter
is probably of bacterial origin. The high fermentability of mulrn/F OS is supported by past
research (Wang and‘Gibson, 1993). Total short chain fatty acids produced were also high
in comparison to other substrates, though not significantly higher than that produced by
the fermentation of high viscosity B—glucan or oat flour (p<0.05). The molar ratio of
butyrate produced fi'om the fermentation of inulin/F OS was significantly higher than all
other experimental substrates (44%). This finding is in agreement with the findings of
Tashiro et al. (1997) who found a significantly higher proportion of cecal butyrate in rats
fed FOS in comparison to guar gum, cellulose, inulin, nystose, and a fiber fi'ee diet. In
contrast to this finding, Wang and Gibson (1993) found inulin to be a relatively poor
source of butyrate, reporting a molar ratio of 8%. However, other publications report
inulin and FOS fermentation to result in significant productions of butyrate and

propionate in both rats and humans (Levrat et al., 1991; Luo et al., 1996; Campbell et al.,

31

1997; Roland et al., 1995; Flickinger et al., 2000). The water holding capacity alter
fermentation of inulin/FOS was high (6.76 grams water per gram fermented residue dry
matter), however due to its high OMD, probably reflects the water holding capacity of the
bacteria present. Again, the potential water holding capacity (1.45 grams water held per
gram of original substrate dry matter) was inversely correlated with OMD.

The mixed fiber substrate, BeFlora, used in this study is a proprietary blend of
fi'uctooligosaccharides (F OS), soy protein extract and glycolate (potato starch). Because
the major fermentative component of this substrate is FOS, the results are similar to the
above inulin/F OS substrate. The fermentation of the mixed fiber substrate resulted in a
high yield of total short chain fatty acids in comparison to other substrates measured
(4.30 mmol / gram original organic matter). The average organic matter disappearance of
mixed fiber (83.84%) suggests virtually complete utilization by the bacteria present in the
fecal inoculum. Similar to the findings for the inulin/F OS substrate, butyrate was a
substantial proportion of the total short chain fatty acids produced (3 7%). Water holding
capacity was high (6.34 grams water per gram fermented dry matter) and again probably
reflects the water holding capacity of the bacteria present.

Oat flour was included as an experimental substrate in this study. Oat flour contains a
high amount of digestible starch (see Appendix A) and therefore may not be of interest
for in viva colonic fermentation, as it is likely that the substrate would be digested prior
to reaching the colon. This substrate was included in this study at the request of the
funding corporation.

The oat groats used in this study were whole cats with the outer hull layer

mechanically removed. The remaining outer aleurone cell wall layer on the groat is thick

32

and makes this product resistant to human digestion without firrther processing (Wood,
1990). The hull is a highly lignified and has a fibrous covering and therefore is not
suitable for human digestion without processing (Mazza, 1998). The disappearance for
this substrate due to fermentation was 7.86% and except Solka-Floc", was significantly
lower than all other substrates used in this study (p<0.05). The resulting short chain fatty
acid production, 1.25 mmon original organic matter, was also the lowest of all
substrates except Solka-Floc° (p<0.05). The water holding capacity of this substrate was
significantly lower than all substrates except oat flour (p<0.05). This occurrence is
probably due to the unavailability of the soluble portion of the cat due to the surrounding
insoluble hull. Bourquin et al. (1993) found similar results with the fermmtion of oat
bran fiber, reporting OMD to be less than 10% and WHC and PWHC to both be around
four grams water per gram residue.

Significant difi‘erences among fermentation indices existed between donors.
Variances in colonic colonization due to diet or genetics may explain this occurrence.
Donor C appeared to have the least ability to ferment the fibers used in this study, as

measured by individual and total short chain fatty acid production as well as organic

matter disappearance.

33

Chapter 5

CONCLUSION

Several of the substrates analyzed in this study demonstrated the ability to
efi‘ectively stimulate the production of short chain fatty acids, specifically butyrate,
acetate, and propionate, as measured by in vitro batch fermentation. The fermentation of
the high viscosity B—glucan, mixed fiber, and inulin/F OS substrates produced the greatest
yields of short chain fatty acids. The fermentation of oat flour also yielded a high
concentration of short chain fatty acids, however the likelihood of this substrate being
fermented in vivo is relatively low due to its high starch content.

The high viscosity B-glucan was fermented to a greater extent than the two lower
viscosity B-glucan products. From this study, it does appear that increasing molecular
weight does increase fermentability of B-containing substances. However, the extent to
which the high viscosity B-glucan was fermented was not as great as other substrates used
in this study. The yield of short chain fatty acids was high, however not statistically
significant from other substrates included in this study. Therefore, it is dificult to assess
whether viscosity plays a large role in the extent of fermentability or production of short
chain fatty acids in a B—glucan product. Further research is required to determine whether
viscosity is a predictor of the fermentability of B-glucan products.

In vitro batch fermentations continue to be an efi'ective method to predict the
fermentability of dietary fibers in vivo. Emerging proprietary substrates can continue to

be analyzed in this fashion to assess their contribution to factors suspected to maintain

34

colonic health. An additional strength of this study is that the results found confirmed
previous research findings.

There are several limitations that must be considered when reviewing results fiom this
type of research. First, the results from this study may not be generalized easily, as each
subject may vary in their ability to fermented carbohydrate. Another limitation to this
study is the small sample size used, as well as the lack of control of the diet ingested by
the donors prior to the donation of fecal samples.

Future directions for this type of research may include the use of probiotics to
condition the subject before an in vitro batch fermentation is done. This would allow the
researcher to assess whether preconditioning with probiotics would significantly change
the donor’s ability to ferment carbohydrate. To strengthen this existing project, the use of
breath hydrogen testing could be use to measure methane production. This information

would help to confirm fermentation findings in this type of study.

35

APPENDICES

APPENDIX A

SUBSTRATE COMPOSITION

All substrates provided by General Mills, Inc., James Ford Bell Technical Center

Information included is limited due to proprietary nature of substrates.

Oat Flour
Composition:
Carbohydrate: 69.0%
Insoluble Fiber: 1.6%
Soluble Fiber: 0.9%
Total fat: 1.64%
Oat Groats
' Composition:
Carbohydrate: 69.2%
Insoluble Fiber. 5.9%
Soluble Fiber: 3.0%
Protein: 12.9%
Total fat: 6.94%
B -Glucan ‘1
Megazyme International
Ireland Limited
Bray Business Park
Bray, Co. Wicklow
Ireland

Barley B-Glucan (High Viscosity) (Lot 40301)

Properties:
Vrscosity: 114 cSt
Starch Content: <0.1%
Protein Content: 0.65%

37

Moisture: 5.0%
Molecular Weight: 327,000

B Glucan2

Megazyme International
Ireland Limited

Bray Business Park
Bray, Co. Wicklow
Ireland

Barley B-Glucan (Low Viscosity) (Lot 41005)

Properties:
Viscosity. 5.9 cSt
Starch Content: <0.0%
Protein Content: 0.85%
Moisture: 6.0%

Molecular Weight: 137,000

B -Glncan 3

Inulin

Megazyme International
Ireland Limited

Bray Business Park
Bray, Co. Wicklow
Ireland

Barley B-Glucan (Medium Viscosity) (Lot 60501)
Properties:

Viscosity: 25.0 cSt
Starch Content: <0. 1%
Protein Content: 0.10%
Moisture: 2.2%

Molecular Weight: 250,000

Imperial Suiker Unie
8016 Hwy. 90A
PO. Box 9

38

Sugar Land, TX 77487-0009
USA

Frutafit® Inulin/FOS-IQ (NG)

Composition:
Soluble dietary fiber (g): 90.00 i 2.2
Total Carbohydrate:
Inulin/FOS (g) 90.00 i 2.2
Disaccharides 4.50 i 027
Glucose 0.75 i 0.25
Fructose 3.10 i 1.3
Moisture (g): 3.00 i 1.5
Mixed Fiber
Triarco Industries, Inc.
400 Hamburg Turnpike
Wayne, New Jersey 07470
_ USA
BeFloraTM

A proprietary blend of FOS, soy extract and glycolate (potato starch)
Composition and properties unavailable
Arabinogalactan

Larex, Inc.

2852 Patton Rd.

St. Paul, MN 55113

USA

FIBERAIDr-

Composition:

Arabinogalactan (anhydrous) 95+%
Water 58%

39

Cellulose

Fiber Sales and Development Corp.
PO Box 88940

St. Louis, MO 63166

USA

Sorta-Hoe” (crystalline cellulose)

APPENDIX B

CHEMICAL STRUCTURES

 

\l/OH
0
Molecular Weight = 60

 

 

 

Figure 1: Chemical Structure of Acetate

 

W014 Molecular Weight = 74
O

 

 

 

Figure 2: Chemical Structure of Propionate

 

W0... Molecular Weight = 88

0

 

 

 

Figure 3: Chemical Structure of Butyrate

41

APPENDIX C

PROPOSED PRODUCTION, ABSORPTION AND UTILIZATION OF SCFA’S

Figure 1. SCF A production and metabolism

 

 

FROM ILEUM

 

 

 

 

 

 

 

 

 

   

 

 

 

 

 

 

 

 

 

 

_1
\ >———<‘,~_ 9°?“
SODIUM 1r ATPase
7 f t
1 ENERGY
com
a 4
J1 ”WM"? GLUCOSE:
ouoosaccrlaaloss l
PEPTIDES r” PYRUVATEe-s LACTATE
MAEROBIC aacrsma e -------- - KETONES
I HC03= C02 )
f—
x
LUIEII , cororloorna 1

 

Adapted from Roediger, 1982.

42

 

Figure 2. Energy utilization of various respiratory firels

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

GLUCOSE * N—Burmra
j. 3, Exocmous
LACIATEH PYRUVATE ACETOACETYL COAF' KET ONE
' * 11 BODIES
ACETYLCOA POOL - ,
. I new.
mass 0x10411011
' . w ’ acaroacarars
SUC cum. “35.5 ICETOGLUTARIC
‘ ACID
PROPIONATE T
GLUTAMINE

 

 

 

 

Adapted fi'om Roediger, 1982.

43

 

APPENDIX D

OMD, DMD, WHC, PHWC RAW DATA BY DONOR

 

 

Donor Substrate OMD DMD WHC PWHC
A Arabinogalactan 16.02 17.04 6.40 5.32
17.35 16.62 6.66 5.54
High Viscosity B-Ghrean 44.27 49.15 7.42 3.92
42.98 45.22 5.04 2.66
Medium Viscosity B-Ghrcan 24.68 25.48 5.32 4.04
21.80 22.65 5.01 3.81
Low V'scosity B-Glucan 23.49 24.04 6.51 5.54
5.97 5.73 6.91 5.88
Solka-Floc -1.46 1.03 4.54 4.63
-3.92 -5.29 4.54 4.64
Imlin/FOS 75.19 78.85 6.63 1.40
77.18 78.86 7.27 1.54
MixedFiber 82.98 83.14 8.00 1.43
82.17 81.16 7.23 129
OatFlour 62.12 62.36 4.87 2.00
55.15 55.46 4.97 2.04
Oat Groats 14.64 14.28 5.77 5.07
9.98 10.10 5.38 4.73
B Arabimgalactan 30.48 33.31 6.16 4.21
25.40 29.89 6.24 4.27
High Viscosity B—Ghrean 40.14 44.29 5.59 3.07
42.13 45.84 5.78 3.17
Medium Viscosity B-Glucan 39.98 45.74 5.81 3.45
32.56 35.41 5.55 3.30
Low Viscosity B-Ghrcan 35.68 39.93 4.60 2.71
38.78 42.36 5.94 3.49
Solka-Floc 1.41 4.46 4.50 4.36
-0.99 1.42 4.77 4.63
Imlin/FOS 72.15 79.50 8.21 1.52
83.21 83.56 6.53 1.21
MixedFiber 83.04 89.02 6.60 0.72
84.88 89.18 4.22 0.46
Oat Flour 61.63 61.97 4.07 1.51
56.94 63.60 5.75 2.14
Oat Groats 9.13 11.88 2.85 2.56
6.50 8.20 3.45 3.11

 

OMD = organic matter disappearance (%)t DMD = dry matter disappearance (%), WHC = g H20/g fermented residue dry
matter, PWHC = g H20/g original substrate dry matter

 

 

Donor Substrate OMD DMD WHC PWHC
C Arabinogalactan 35.37 34.88 6.02 4.53
16.13 14.64 5.64 4.24
High Viscosity B—Gluean 25.70 24.22 6.27 4.83
24.20 21.72 4.86 3.74
Medium Viscosity B—Ghrean -l.66 -3.58 6.16 6.16
2.62 3.47 5.85 5.85
Low Viscosity B-Ghican -3.37 -2.83 5.31 5.45
1.10 -2.61 4.88 5.02
Solka-Floc -4. 15 -3.79 4.95 5.29
-7.83 -9.81 5.11 5.46
Irmlin/FOS 77.12 77.74 6.19 1.52
73.03 73.28 7.26 1.78
Mixed Fiber 84.08 83.89 5.96 0.91
84.39 85.51 6.85 1.05
Oat Flour 30.98 31.37 3.47 2.39
32.40 31.10 3.18 2.19
Oat Groats 2.18 3.13 2.57 2.47
7.64 5.14 3.03 2.91
D Ardrinogalactan 26.89 28.47 6.53 4.81
25.68 24.44 5.95 4.38
High Viscosity B-Ghiean 39.84 41.01 5.33 3.35
32.62 33.22 5.31 3.34
Medium Viscosity B-Ghican 11.76 11.71 5.92 5.24
10.72 11.27 5.62 4.97
Low Viscosity B-Ghrean 13.69 17.36 3.06 2.58
11.13 14.14 4.79 4.03
Solka-Floc -0.55 -0.35 3.24 3.14
3.17 6.32 4.07 3.95
Imlin/FOS 79.57 78.83 5.84 1.28
78.02 77.27 6.17 1.35
Mixed Fiber 85.30 84.90 6.14 0.99
83.91 82.99 5.74 0.92
Oat Flour 32.37 34.29 3.06 2.06
30.90 31.43 3.71 2.49
Oat Groats 2.29 1.39 5.00 4.70
10.50 10.55 2.26 2.12

 

OMD=organicmatterdisappeerance(%), DMD*drymatterdisappearance(%),WHC=gH20/gfermcntedresiduedry
matter, PWHC=gH¢Olg originalsubsuatedrymatter

45

APPENDIX E

SCFA PRODUCTION RAW DATA BY DONOR

 

SCFA Production (mung original organic matter)

 

 

Donor Substrate Acetate Propionate Butyrate Total
A Arabinogalactan 1.72 0.70 0.53 2.95
1.83 0.74 0.47 3.04

1.88 0.77 0.49 3.14

1.74 0.72 0.46 2.93

High Viscosity B-Glucan 2.96 0.35 1.27 4.59
3.10 0.36 1.33 4.80

4.46 0.61 1.97 7.05

4.20 0.57 1.96 6.72

3.17 0.35 1.36 4.89

3.45 0.41 1.47 5.33

Medium Viscosity B-Ghican 1.97 0.28 1.05 3.31
1.82 0.26 0.96 3.05

2.07 0.28 1.14 , 3.49

2.35 0.31 1.28 3.94

Low Viscosity B-Giucan 1.86 0.27 0.92 3.05
1.55 0.24 0.77 2.56

2.30 0.38 1.17 3.85

1.90 0.32 0.95 3.16

1.59 0.25 0.80 2.64

1.59 0.25 0.81 2.65

Solka-Floc 0.48 0.16 0.27 0.91
0.59 0.19 0.24 1.02

0.64 0.20 0.24 1.09

0.62 0.20 0.23 1.04

Inulin/FOS 2.21 0.56 43.37 6.14
2.23 0.54 3.35 6.12

1.85 0.53 3.15 5.53

1.87 0.54 3.08 5.49

Mixed Fiber 1.43 0.44 2.17 4.04
1.69 0.52 2.50 4.71

1.26 0.44 2.15 3.85

1.35 0.49 . 2.36 4.20

Oat Flour 1.94 0.67 2.30 4.91
1.78 0.61 2.21 4.60

2.28 0.73 2.70 5.71

1.89 0.59 2.11 4.59

2.32 0.72 2.59 5.63

2.46 0.76 2.79 6.01

Oat Groats 0.39 0.10 0.30 0.80
0.38 0.09 0.28 0.75

0.36 0.10 0.23 0.69

0.33 0.09 0.22 0.64

 

 

Domr

Substrate

SCFA Production (ml/W nutter)

 

 

 

Acetate Pmpiomte Butyrate Total

B Arabinogalactan 2.24 0.87 0.56 3.67
1.92 0.79 0.48 3.19

2.45 1.11 0.64 4.20

2.20 0.92 0.52 3.64

2.19 0.93 0.53 3.65

High Viscosity B—Glncan 2.03 0.47 1.41 3.91
2.05 0.48 1.47 4.00

2.69 0.67 1.94 5.30

2.56 0.64 1.86 5.07

2.26 0.55 1.61 4.42

2.19 0.53 1.57 4.29

Medium V'scosity B-Glncan 1.98 0.45 1.11 3.54
2.17 0.49 1.19 3.84

2.85 0.69 1.59 5.13

2.81 0.69 1.57 5.07

2.50 0.54 1.43 4.48

2.27 0.48 1.29 4.05

Low Viscosity B-Glucan 2.20 0.51 1.22 3.93
2.69 0.69 1.58 4.97

2.23 0.55 1.28 4.06

2.24 0.46 1.29 3.99

2.10 0.43 1.21 3.75

Solka-Floc 0.58 0.16 0.21 0.96
0.45 0.13 0.15 0.74

0.54 0.19 0.20 0.92

0.49 0.15 0.18 0.82

0.53 0.18 0.18 0.89

[milin/FOS 2.09 0.31 1.46 3.85
2.03 0.32 1.47 3.81

2.50 0.36 1.70 4.56

1.96 0.36 1.70 4.02

Mixed Filer 2.92 0.33 0.69 3.94
3.28 0.37 0.80 4.45

3.25 0.36 0.77 4.38

2.90 0.31 0.70 3.91

Oat Flour 5.02 0.41 1.29 6.72
4.98 0.41 129 6.69

4.96 0.42 1.34 6.72

5.13 0.41 1.34 6.88

Oat Groats 1.42 0.30 0.64 2.36
1.41 0.30 0.62 2.34

1.09 0.25 0.47 1.81

1.11 0.25 0.47 1.82

 

47

 

SCFA Production @an nutter)

 

 

 

Domr SW Acetate Propionate Butyrate Total
C Arabinogabctan 1.90 0.57 0.40 2.86
1.83 0.50 0.39 2.71

1.79 0.53 0.35 2.67

1.80 0.51 0.39 2.70

1.86 0.58 0.38 2.82

High Viscosity B-Glucan 3.12 0.77 0.85 4.75
3.24 0.80 0.88 4.93

3.07 0.95 0.89 4.91

3.10 0.95 0.89 4.94

Medium Viscosity B-Glucan 1.52 0.66 0.49 2.67
1.21 0.52 0.40 2.13

1.30 0.53 0.42 2.25

1.36 0.56 0.44 2.36

Low Viscosity B-Ghrean 1.14 0.47 0.30 1.91
1.07 0.44 0.28 1.79

0.96 0.41 0.25 1.61

0.78 0.33 0.20 1.31

Solka-Floc 0.47 0.23 0.18 0.88
0.45 0.22 0.18 0.85

0.48 0.15 0.14 0.77

0.32 0.14 0.12 0.57

0.40 0.20 0.15 0.75

Inulin/FOS 2.04 0.32 1.56 3.92
2.13 0.32 1.69 4.14

2.33 0.32 1.63 4.28

2.22 0.31 1.64 4.17

Mixedeer 1.98 0.48 1.44 3.91
1.77 0.45 1.25 3.47

1.69 0.43 1.28 3.40

1.86 0.47 1.44 3.76

Oat Flour 1.40 0.46 1.00 2.85
1.51 0.51 1.13 3.15

2.08 0.76 1.69 4.52

1.97 0.73 1.59 4.29

Oat Groats 0.25 0.09 0.17 0.51
0.22 0.07 0.16 0.45

0.31 0.10 0.22 0.63

0.28 0.09 0.21 0.58

 

 

SCFA Production (WW rmtter)

 

 

 

Dam SW6 Acetate Propionate Butyrate Total
D Arabinogalactan 2.46 0.77 0.61 3.84
2.40 0.82 0.60 3.82

2.35 0.82 0.54 3.71

2.52 0.90 0.56 3.98

2.10 0.76 0.47 3.33

High V'scosity B—Glucan 3.77 1.02 1.06 5.86
3.64 1.08 1.11 5.83

3.76 1.16 1.21 6.13

4.12 1.22 1.27 6.61

4.21 1.27 1.29 6.77

Medium Viscosity B-Glucan 1.60 0.68 0.53 2.81
1.53 0.64 0.50 2.67

1.71 0.67 0.55 2.93

1.69 0.73 0.61 3.04

Low vucosity B—Glncan 2.05 0.90 0.57 3.52
2.03 0.87 0.55 3.46

1.78 0.73 0.44 2.95

1.70 0.69 0.42 2.81

Solka-Floc 0.63 0.27 0.37 1.27
0.51 0.19 0.28 0.98

0.53 0.23 0.21 0.97

0.49 0.19 0.17 0.85

Inulin/FOS 2.80 0.60 2.66 6.06
2.53 0.52 2.52 5.57

2.81 0.54 2.64 5.99

Mixedeer 3.07 0.71 2.25 6.04
3.17 0.64 2.04 5.84

2.01 0.45 1.82 4.27

2.19 0.50 1.94 4.62

2.30 0.66 2.15 5.10

Oat Flour 1.35 0.47 0.99 2.81
1.51 0.63 1.20 3.34

1.36 0.55 1.06 2.98

1.25 0.44 1.01 2.70

1.44 0.47 1.13 3.04

Oat Groats 0.70 0.24 0.47 1.41
0.81 0.30 0.45 1.55

0.86 0.30 0.63 1.79

0.93 0.34 0.67 1.94

 

49

APPENDIX F

HUMAN SUBJECTS APPROVAL

50

 

OFFICE OF
RESEARCH
AND
GRADUATE
STUDIES

:rsity Committee on

Research lnvolvi
Human Subie
(UCRIHS)

:higan State University
.drnrnistralion Building
asl Lansing. Michigan

48824-1046

517/355-2180
FAX' 51 7/353-2976

Michigan Stare Universny
:5 Insrn'urzona/ Diversrry
Excellence in Action

J IS an affirmative-action,
al-DDDDrrumrv Institution

MICHIGAN STATE

 

 
 

UNIVERSITY
June1,1999

TO: Dr. Norman HORD

RE: IRB# 99361 CATEGORY: 2-C
APPROVAL DATEzJune 1, 1999
TITLE:WHOLE OAT UTILIZATION IN HUMANS

The University Committee on Research Involving Human Subjects' (UCRIHS) review of
this project is complete and I am pleased to advise that the rights and welfare of the
human subjects appear to be adequately protected and methods to obtain informed
consent are appropriate. Therefore, the UCRIHS approved this project.

RENEWALS: UCRIHS approval is valid for one calendar year, beginning with the
approval date shown above. Projects continuing beyond one year must be renewed
with the green renewal form. A maximum of four such expedited'renewals possible.
Investigators wishing to continue a project beyond that time need to submit it again for
a complete review.

REVISIONS: UCRIHS must review any changes in procedures involving human

_ subjects, prior to initiation of the change. If this is done at the time of renewal, please

use the green renewal form. To revise an approved protocol at any other time during
the year, send your written request to the UCRIHS Chair, requesting revised approval
and referencing the project's IRB# and title. Include in your request a description of the
change and any revised instruments, consent forms or advertisements that are
applicable.

PROBLEMS/CHANGES: Should either of the following arise during the course of the
work, notify UCRIHS promptly: 1) problems (unexpected side effects. complaints, etc.)
involving human subjects or 2) changes in the research environment or new information
indicating greater risk to the human subjects than existed when the protocol was
previously reviewed and approved.

If we can be of further assistance, please contact us at 517 355-2180 or via email:

UCRIHS@piIot.msu.edu. Please note that all UCRIHS forms are located on the web:
http://www.msu.edulunit/vprgs/UCRIHS/

ly

    
  
  
 

   

 

   

D

   

avid E. Wright,.Ph.
CRIHS Chair

DEW: db
CC: Julie Arnold

    

Kristen England

51

BIBLIOGRAPHY

52

Archer SY, Meng s, Shei A, and Hodin RA mm“ is required for butyrate-mediated
growth inhibition of human colon cancer cells. Proc Natl Acad Sci USA. 1998; 95: 6791-
6796.

Asp NG. Resistant starch — an update on its physiological efl‘ects. Adv Exp Med Biol.
1997; 427: 201-10.

Barry JL, Hoebler C, Macfarlane GT, Marcfarlane S, Mathers JC, Rad KA, Mortenson
PB, Nordgaard I, Rowland IR, and Rumney CJ. Estimation of the fermentability of
dietary fibre in vitro: a European interlaboratory study. BrJNutr. 1995; 74: 303-322.

Basson MD, Emenaker NJ and Hong F. Difl‘erential modulation of human (Caco-Z)
colon cancer cell line phenotype by short chain fatty acids. Proc Soc Exp Biol Med
1998; 217(4): 476-483. '

Bedi A, Pasricha PI, Akhtar A], Barber JP, Bedi GC, Giardiello FM, Zehnbauer BA,
Hamilton SR, and Jones R]. Inhibition of apoptosis during development of colorectal
cancer. Cancer Research. 1995; 55: 1811-1816.

Bloom SR and Polak M The hormonal pattern of intestinal adaptation: a major role for
enteroglucagon. Scand J Gastroenterol. 1982; 14 (suppl 74): 93-103.

Bofi‘a LC, Lupton IR, Mariani MR, Ceppi M, Newmark HL, Scalmati A, and Lipkin M.
Modulation of colonic epithelial cell proliferation, histone acetylation, and luminal short
chain fatty acids by variation of dietary fiber (wheat bran) in rats. Cancer Res. 1992; 52:
6906-5912.

Bourquin LD, Titgemeyer EC, Garleb KA, and Fahey GC. Short-chain fatty acid
production and fiber degradation by human colonic bacteria: Efi‘ects of substrate and cell
wall fi‘actionation procedures. J Nutr. 1992; 122: 1508-1520.

Bourquin LD, Titgemeyer EC, F ahey GC, and Garleb KA Fermentation of dietary fibre
by human colonic bacteria: Disappearance of, short-chain fatty acid production from, and
potential water-holding capacity ofi various substrates. Scand J Gastroenterol. 1993; 28:
249-255.

Bourquin LD, Titgemeyer EC, and Fahey GC. Fermentation of various dietary fiber
sources by human fecal bacteria. Nutrition Research. 1996; 16(7): 1119-1131.

53

Bradbum DM, Mathers JC, Gunn A, Burn J, Chapman PD, and Johnston IDA Colonic
fermentation of complex carbohydrates in patients with familial adenamatous polyposis.
Gut. 1993; 34: 630-636.

Brighenti F, Casiraghi MC, Canzi E, and Ferrari A Effect of consumption of a ready-to-
eat breakfast cereal containing inulin on the intestinal milieu and blood lipids in healthy
male volunteers. Eur J Clin Nutr. 1999; 53: 726-733.

Brown I. Complex carbohydrates and resistant starch. Nutrition Reviews. 1996; 54(11):
S 1 l 5-S 1 l9.

Bryant MP and Burkey LA Cultural methods and some characteristics of some of the
more mrmerous groups of bacteria in the bovine rumen. J Dairy Sci. 1953; 36: 205-217.

Campbell JM, Fahey GC, and Wolf BW. Selected Oligosaccharides Afl‘cct Large Bowel
Mass, Cecal and Fecal Short Chain Fatty Acids, pH, and Microflora in Rats. J Nutr.
1997; 127: 130-136.

Carmer SG and Swanson MR An Evaluation of Ten Pairwise Multiple comparison
Procedures by Monte Carlo Methods. J Am Stat Assoc. 1973; 68(341): 66-74.

Cummings JH (1982). In Colon and Nutrition, pp. 91-104. [Kasper H and Goebell H,
editors]. Lancaster. MTP Press Limited.

Cummings JH, Pomare EW, Branch WJ, Naylor CPE, and Macfarlane GT. Short chain
fatty acids in human large intestine, portal, hepatic and venous blood. Gut. 1987; 28:
1221-1227.

Cummings JI-I (1988). In Ross Laboratories Symposium Monograph, pp. 11-16. [E
Silverman, editor] Columbus: Ross Laboratories.

Cummings JH (1994). In Short-chain fatty acids, pp. 11-19. [J Cummings, HJ Binder, K
Soergel, editors]. Boston Kluwer Academic Publisher.

Cummings JH and Macfarlane GT. Role of intestinal bacteria in nutrient metabolism.
Clinical Nutrition 1997; 16(3): 3-11.

Eastwood MA, Robertson JA, Brydon WG, and MacDonald D. Measurement of water-
holding properties of fibre and their faecal bulking ability in man B J Nutr. 1983; 50:
539-547.

54

Edwards CA (1997). In Dietary Fiber in Health ard Disease, pp. 155-168. [Kritchevsky
D and Bonfield C, editors]. New York: Plenum Press.

Ehle FR, Robertson JB, and Van Soest PJ. Influence of Dietary Fibers on Fermentation
in the Human Large Intestine. J Nutr. 1982; 112: 158-162.

E1 Oufir L, Bruley des Varannes S, Barry JL, Cloarec D, Bornet F, and Galmiche JP.
Relations between transit time, fermentation products, and hydrogen consuming flora in
healthy humans. Gut. 1996; 38: 870-877.

Emenaker NJ and Basson MD. Short-chain fatty acids inhibit human (SW1116) colon
cancer cell invasion by reducing urokinase plasminogen activator activity and stimulating
TIMP-1 and TIMP-2 activities, rather than via MMP modulation. J Surg Res. 1998;
76(1): 41—46.

Emenaker NJ. Short-chain fatty acids derived fi'om dietary fiber may protect against
invasive human colon cancer. Oncology Nutrition Dietetic Practice Group Newsletter,
ADA. 1999 Winter; 7(1): 1-9.

F Iickinger EA, WoIfBW, Garleb KA, Chow J, Leyer GJ, Johns PW, and F ahey GC.
Glucose-based oligosaccharides exhibit different in vitro fermentation patterns and afi‘ect
in viva apparent nutrient digestibility and microbial populations in dogs. J Nutr. 2000;
130: 1267-1273. '

Gibson GR, Wrilems A, Reading S, and Collins MI). Fermentation of non-digestible
oligosaccharides by human colonic bacteria. Proceedings of the Nutrition Society. 1996;
55: 899-912.

Gibson GR Dietary modulation of the human gut microflora using prebiotics. B J Nutr.
1998; 80: 8209-8212.

Gross JL, Behrens DL, Mullins DE, Kornblith PL and Dexter DL. Plasminogen activator
and inhibitor activity of in human glioma cells and modulation by sodium butyrate.
Cancer Res. 1988; 48(2): 291-296.

Hague A, Manning AM, Hanlon KA, Huschtscha LI, Hart D and Paraskeva C. Sodium
butyrate induces apoptosis in human colonic tumor cell lines in a p53-independent
pathway: Implications for the possible role of dietary fibre in the prevention of large-
bowel cancer. IntJ Career. 1993; 55: 498-505.

Hague A, Singh B and Paraskeva C. Butyrate acts as a survival factor for colonic
epithelial cells: Further fire] for the in vivo versus in vitro debate. Gastroenterology.
1997; 112: 1036-1040.

55

Hass R, Busche 1L Luciano L, Reale E, and Engelhardt W. Lack of butyrate is associated
with induction of bax and subsequent apoptosis in the proximal colon of guinea pig.
Gastroenterologz. 1997; 112: 875-881.

Knudsen KE, Hansen 1, Jensen BB, and Ostergard K (1990). In New Developments in
Dietay Fiber, pp. 135-150. [Furda I and Brine CJ, editors]. New York: Plenum Press.

Knudsen KE, Jensen BB, and Hansen 1. Oat barn but not a B-giucan oat fiaction
enhances butyrate production in the large intestine of pigs. J Nutr. 1993; 123: 123 5-
1247.

Levrat M, Remesy C and Demigne C. High propionic acid fermentations and mineral
accumulation in the cecum of rats adapted to difi‘erent levels of inulin. J Nutr. 1991; 121:
1 730-173 7.

Luo J, Rizkalla SW, Alamowitch C, Boussairi A, Blayo A, Barry JL, Laflitte A, Guyon
F, Bornet FRJ, and Slama G. Chronic consumption of short-chain fi'uctooligosaccharides
by healthy subjects decreased basal hepatic glucose production but had no efi‘ect on
insulin-stimulated glucose metabolism. Am J Clin Nutr. 1996; 63: 939-945.

Macfarlane GT and Cummings JH. Probiotics and prebiotics: can regulating the activities
of intestinal bacteria benefit health? BM]. 1999; 318: 999-1003.

Mandal M, Wu X and Kumar R Bel-2 deregulation leads to inhibition of sodium
butyrate-induced apoptosis in human colorectal carcinoma cells. Carcinogenesis. 1997;
18 (1): 229-232.

Mazza G (1998). Functional Foods: Biochemical & Processing Aspects. Lancaster:
Technomic Publishing Co., Inc.

McBumey MI, Horvath PJ, Jeraci JL, and Van Soest PJ. Efl'ect of in vitro fermentation
using human faecal inoculum on the water-holding capacity of dietary fibre. B J Nutr.
1985; 53: 17-24.

McBumey MI and Thompson LU. Efl’ect of Faecal Donor on In vitro Fermentation
Variables. Scard J Gastroenterol. 1989; 24: 359-367.

McBumey Nfl. Potential water-holding capacity and short-chain fatty acid production
fiom purified fiber sources in a fecal incubation system. Nutrition. 1991; 7 (6): 421-424.

McIntyre A, Gibson PR and Young GP. Butyrate production from dietary fibre and
protection against large bowel cancer in a rat model. Gut. 1993; 34: 386-391.

56

McNeil NI (1982). In Colon aid Nutrition, pp. 141-144. [Kasper H and Goebell H,
editors]. Lancaster. MTP Press Limited.

Monsma DJ, Thorsen PT, Vollendorf NW, Crenshaw TD, and Marlett JA In vitro
Fermentation of Swine Ileal Digesta Containing Oat Bran Dietary Fiber by Rat Cecal
Inocula Adapted to the Test Fiber Increase Propionate Production But Fermentation of
Wheat Bran Deal Digesta Does Not Produce More Butyrate. J Nutr. 2000; 130: 585-593.

Mortensen PB, Hove H, Clausen MR, and Holtug K. Fermentation to short-chain fatty
acids and lactate in human faecal batch cultures. Seard J Gastroenerology. 1991; 26:
1285-1294.

Nakano K, Mizuno T, Sowa Y, Orita T, Yoshino T, Okuyama Y, Fujita T, Ohtani-Fujita
N, Matsukawa Y, Tokino T, Yamagishi H, Oka T, Nomura H, and Sakai T. Butyrate
activates the WAF 1/Cip1 gene promoter through Spl sites in a p53-negative human
colon cancer cell line. J Biol Chem. 1997; 272 (3 5): 22199-22206.

Penner MH and Liaw E (1990). In New Developments in Dietay Fiber, pp. 169-178.
[Furda I and Brine CJ, editors]. New York: Plenum Press.

Potter, JD. Colorectal Cancer: Molecules and Populations. Journal of the National
Career Institute. 1999; 91: 916-932.

Reimer RA and McBumey MI. Dietary fiber modulates intestinal proglucagon
messenger ribomrcleic acid and postprandial secretion of glucagon-like peptide-1 and
insulin in rats. Endocrinology. 1996; 137: 3948-3956.

Roberfroid MB. Prebiotics and synbiotics: concepts and nutritional properties. British
Journal of Nutrition 1998; 80: $197-$202.

Robertson JA and Eastwood MA An examination of factors which may afl‘ect the water
holding capacity of dietary fibre. B J Nutr. 1981; 45: 83-87.

Roediger WE (1982). In Colon and Nutrition, pp. 11-26. [Kasper H and Goebell H,
editors]. Lancaster". MTP Press Limited.

Roland N, Nugon-Baudon L, Andrieux C, and Szylit 0. Comparative study of the
fermentative characteristics of inulin and difi‘erent types of fibre in rats inoculated with a
human faecal flora. B J Nutr. 1995; 74: 239-249.

Rountree DB, Ulshen MH, Selub S, Fuller CR, Bloom SR, Ghatei MA and Lund PK.
Nutrient-independent increases in proglucagon and omithine decarboxylase messenger
RN As after jejunoileal resection Gastroerrterolog. 1992; 103: 46-468.

57

 

Sagor GIL Ghatei MA, Al-Mukhtar MYT, Wright NA and Bloom SR. Evidence for a
humoral mechanism after small intestinal resection: Exclusion of gastrin but not
enteroglucagon. Gastroenterology. 1983; 84: 902—906.

Salyers AA (1990). In New Developments in Dietay Fiber, pp. 151-158. [Furda I and
Brine CJ, editors]. New York: Plenum Press.

Scheppach W. Efi‘ects of short chain fatty acids on gut morphology and function. Gut.
1994; supplement 1: $35-$38.

Soergel KB (1982). In Colon aid Nutrition, pp. 27-36. [Kasper H and Goebell H,
editors]. Lancaster: MTP Press Limited.

Sundberg B, Wood P, Lia A Andersson H, Sandberg A, Hallmans G, and Aman P.
Mixed-linked B-glucan from breads of difl‘erent cereals is partly degraded in the human
ileostomy model. Am J Clin Nutr. 1996; 64: 878-85.

Sunvold GD, F ahey GC, Merchen NR, Bourquin LD, Titgemeyer EC, Bauer LL, and
Reinhart GA Dietary Fiber for Cats: In vitro Fermentation of Selected Fiber Sources by
Cat Fecal Inoculum and In vivo Utilization of Diets Containing Selected Fiber Sources
and Their Blends. JAnim Sci. 1995; 73: 2329-2339.

Tappenden KA and McBumey MI. Systemic short-chain fatty acids rapidly alter
gastrointestinal structure, function, and expression of early response genes. Digestive
Diseases and Sciences. 1998; 43 (7): 1526-1536.

Tappenden KA, Thomson ABR, Wild GE and McBumey MI. Short-chain fatty acids
increase proglucagon and ornithine decarboxylase messenger RNAs following intestinal
resection in rats. JPEN. 1996; 20: 357-362. '

Tappenden KA, Thomson ABR, Wild GE and McBumey MI. Short-chain fatty acid-
supplemented total parenteral nutrition enhances firnctional adaptation to intestinal
resection in rats. Gastroenterologr. 1997; 122: 792-802.

Tashiro Y et al. (1997). In Dietay Fiber in Health ard Disease, pp. 221-234.
[Kritchevsky D and Bonfield C, editors]. New York: Plenum Press.

Titgemeyer EC, Bourquin LD, Fahey GC, and Garleb KA Fermentability of various
fiber sources by human fecal bacteria in vitro. Am JCIin Nutr. 1991; 53: 1418-24.

Velazquez 0C, Zhou D, Seto RW, Jabbar A, Choi J, Lederer HM and Rombeau JL. In
vivo crypt surface hyperproliferation is decreased by butyrate and increased by

58

 

deoxycholate in normal rat colon' Associated in vivo efl‘ects on c-fos and c-jun
expression. JPEN. 1996; 20: 243-250.

Venter CS, Vorster HH and Cummings JH Efl‘ects of dietary propionate on
carbohydrate and lipid metabolism in healthy volunteers. Am J Gastroenterol. 1990; 85:
549.

Vince AJ, McNeil N1, Wager JD and Wrong OM. The efi‘ect of lactulose, pectin,
arabinogalactan and cellulose on the production of organic acids and metabolism of _
ammonia by intestinal bacteria in a faecal incubation system. Br J Nutr. 1990; 63: 17-
26.

Wang X and Gibson GR Efi‘ects of the in vitro fermentation of oligofi'uctose and inulin
by bacteria in the human large intestine. Journal of Applied Bacteriology. 1993; 75: 373-
380.

Wijnands MV, Appel MJ, Hollanders VMH and Woutersen RA A comparison of the
effects of dietary cellulose and fermentable galacto-oligosaccharide, in a rat model of
colorectal carcinogenesis: fermentable fibre confers greater protection than non-
fermentable fibre in both high and low fat backgrounds. Ca'cinogenesis. 1999; 20 (4):
65 1-65 6.

Wolever TM (1998). In Ross Laboratories Symposium Monogrqoh, pp. 24-27. [E
Silverman, editor] Columbus: Ross Laboratories.

Wood P (1990). In New Developments in Dietary Fiber, pp. 119-128. [Furda I and Brine
CJ, editors]. New York: Plenum Press.

Yokoyama MT and Johnson KA (197 9). In The Ruminart Animal. Digestive Physiology
and Nutrition. [Church DC, editor]. Englewood Cliffs, NJ: Prentice Hall.

Young GP (1998). In Ross Laboratories Symposium Monograph, pp. 39-44. [E
Silverman, editor] Columbus: Ross Laboratories.

59