Content uploaded by Jovyn K.T. Frost
Author content
All content in this area was uploaded by Jovyn K.T. Frost on Jun 22, 2016
Content may be subject to copyright.
Food &
Function
PAPER
Cite this: Food Funct., 2016, 7, 2645
Received 10th December 2015,
Accepted 8th May 2016
DOI: 10.1039/c5fo01526j
www.rsc.org/foodfunction
Vegetable dietary fibres made with minimal
processing improve health-related faecal
parameters in a valid rat model†
John Monro,* Suman Mishra, Claire Redman, Sheryl Somerfield and Jovyn Ng
Dietary fibre-induced faecal bulking and hydration are important contributors to large bowel function and
health, and are affected by the dietary fibre structure. To determine faecal bulk-related parameters for
vegetable dietary fibres with retained structure, cold water fragmentation of vegetables was used to make
minimally processed vegetable fibres (MPVF) from swede, broccoli and asparagus. A valid adult rat model
was used to subject the fibres to processes of hind gut fermentation and faecal accumulation similar to
those in humans. All the MPVFs had high faecal bulking indexes (FBIs, mean ± sem: wheat bran (refer-
ence), 100 ± 6.0; asparagus 168 ± 5.7; swede 135 ± 6.1; broccoli 135 ± 5.9; broccoli rind 205 ± 10.4), and
caused large increases in the theoretical colonic water load at 10 g per 100 g diet (increase over baseline
(%): wheat bran, 137 ± 8.3; asparagus, 236 ± 25, swede 193 ± 8.8; broccoli 228 ± 12; broccoli rind 223 ±
8.5). Faecal bulking by MPVFs was much greater than by fermentable extracted polysaccharides such as
pectin and raftilose, or by commercial fibres made from highly processed cell walls. The results show
natural, non-degraded vegetable fibres with retained botanical structure have beneficial effects not
provided by structure-less fermentable dietary fibres. Dietary fibre-deficient diets supplemented with pre-
biotics cannot, therefore, adequately substitute for varied diets containing adequate vegetables, fruits and
wholegrain cereals in which fermentation is associated with enough retained structure to conserve
physicochemical properties of benefit to colonic function.
1 Introduction
The properties of both the pre-faecal and faecal masses in the
large intestine play an important role in the entire digestion
process,
1
as hormonal and neuronal feedback loops from the
colon to the foregut influence processes from appetite and
food intake to gut transit to colonic loading, with diverse
effects on the metabolism of the entire organism.
2
However,
with the current popular and research emphasis on the effects
of fermentable digestion-resistant carbohydrates (dietary
fibres) as prebiotics, the role of non-fermented bulk has been
relatively neglected. Prebiosis involves fermentative destruc-
tion of fibre, so completely fermentable fibres will undergo a
complete loss of bulk, only partially compensated by increased
biomass due to bacterial growth supported by the fermented
fibre. Unless there is some structural component to the fibre
intake that confers resistance to fermentation and provides
enough structural resilience to maintain volume, with its
associated water load, faecal bulk is likely to fall below the
approximately 150–200 g per day indicated for protection
against colorectal cancer.
3
As dietary fibre, by definition
(CODEX),
4
requires the chain length to be greater than only
3–10 sugar units, which is far too short for polysaccharide
alignments that generate resistant structures such as plant cell
walls, many prebiotics that qualify as dietary fibres are func-
tionally limited compared with plant cell wall remnants in the
colon.
Faecal bulking associated with resistance to fermentation is
a physiological property that modulates the effects of pre-
biotics and may be complementary to them in terms of
health benefits.
5
Bulking capacity is achieved largely through
hydration, which reduces chemical activity in the distal colon
by dilution.
6
Colorectal bulk also contributes to the stimuli
that induce the defecation process.
7
Defecation eliminates
wastes that would produce toxic products of protein putrefac-
tion if allowed to stagnate,
8
and is followed by movement of
fresh digesta into the colorectal region, where the protective
benefits of fermentation are most required.
9
Importantly,
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c5fo01526j
The New Zealand Institute for Plant & Food Research Limited, Palmerston North,
New Zealand. E-mail: john.monro@plantandfood.co.nz; Fax: +64 6 351 7050;
Tel: +64 6 355 6137
This journal is © The Royal Society of Chemistry 2016 Food Funct.,2016,7, 2645–2654 | 2645
Published on 11 May 2016. Downloaded by Massey University on 22/06/2016 23:26:31.
View Article Online
View Journal
| View Issue
regular bowel movement also contributes to a sense of
wellbeing.
Much of the perceived need for prebiotics associated with
bowel disorders in the modern diet arises because dietary
fibre intakes are inadequate, as many sources of mixed func-
tion dietary fibre have been removed through ingredient
refinement, unhealthy food formulations and poor food
choices. Using prebiotics to restore fibre intake is not,
however, entirely satisfactory because prebiotics often lack
physical structure, so have a more narrow range of functional-
ity than the “natural”dietary fibres. They cannot provide the
colon with fermentation-resistant plant cell wall and tissue
remnants, whose morphology plays an important role in faecal
bulk.
10
Being plant cell wall material is not, however, sufficient to
ensure functionality. Commercial dietary fibre preparations
derived from plant cell walls and used to increase fibre intake
are often functionally deficient because they have been sub-
jected to destructive processing to make them suitable for use
as ingredients in foods such as baked products. Heating that
causes depolymerisation, extraction and loss of pectin, and
loss of structure as the dietary fibre preparations are milled to
very fine powders, will alter hydration properties and lower
resistance to fermentation.
11
Therefore, unless minimally pro-
cessed, cell wall preparations from vegetables and fruits are
likely to suffer a reduced faecal bulking efficacy during
preparation.
To determine the capacity of minimally processed vegetable
dietary fibres (MPVF) to contribute to distal colonic bulk, a
comparative study was made of the faecal bulking properties
of dietary fibre preparations obtained from vegetables by cold
water maceration. A validated rat model of human faecal
bulking responses was used to assess the potential of the
MPVF to improve faecal bulk and hydration. Several different
types of vegetables were used: a stem (swede), a shoot (aspara-
gus), and a flower head (broccoli). The dose–response relation-
ship was determined and the faecal bulking efficacy compared
with those of a soluble fermentable and a soluble non-fermen-
table polysaccharide and a commercial prebiotic. Finally the
faecal bulking indexes (FBIs) of the materials were determined
to provide a standardised measure of the relative faecal
bulking efficacy of the materials tested.
2 Materials and method
2.1 Dietary fibre preparation
Swedes (Brassica napobrassica), asparagus (Asparagus officina-
lis), broccoli head (Brassica oleracea), onion (Allium cepa),
parsnip (Pastinaca sativa) and carrot (Daucus carota) were
obtained from commercial growers and were all in sound con-
dition. The onion, parsnip and carrot were not included in the
faecal bulking trials but the hydration properties of all the
fibre preparations were measured. For preparation of broccoli
rind MPVF, broccoli frames (the stalk left after harvesting the
edible flower head) were cut about 10 cm above ground level,
trimmed of leaves, and strips of rind were removed and cut
into 1 cm sections ready for MPVF preparation (below).
The faecal bulking reference material was hard red wheat
bran from a local cereal wholesaler. It was milled in a Retsch
mill with a 1 mm sieve plate and then hand-sieved through a
2 mm sieve for use in the rat diets. Fibrex® sugar beet fibre
(grade 595, <125 μm particle size, Danisco Sugar AB, Malmö,
Sweden) was derived from the wastes of sugar beet so, like the
MPVFs, represents non-woody cell wall material. It is widely
used as a bulking agent in the food industry and was used
directly in the rat diets. Other fibres tested in the hydration
experiment or in the faecal bulking assay were pectin
(Mexpectin; Grinsted Products), psyllium husk (Bronson and
Jacobs,Homebush,Australia),applefibre(Vitacel;JRS,
Rosenberg, Germany), Raftilose (Orafti), lactic casein (Fonterra),
starch (New Zealand Starch Ltd, Auckland) and castor sugar
(Chelsea).
2.2 MPVF preparation
The main aim of the minimal processing was to rupture the
plant cell walls, release and wash away the cell contents, and
dry the remaining fibre (ESI Fig. S1†). The vegetables were
washed by hand in cold water containing 100 ppm chlorine.
After washing, they were passed through a Hallde grater with a
4 mm grating disk at a rate of 2 kg per minute. The grated
vegetable was then chopped in a Talsa bowl chopper to
produce a puree. This was added with 2 L water to a disinte-
grator, which chopped at high speed in a sealed container for
5 minutes, producing a fine puree which was then passed
through a colloid mill as the last stage in the size reduction
process. The aperture size was set at 1.75 micron and the vege-
table pulp was manually fed through at a slow speed. By treat-
ing the milled product with Evan’s Blue stain and viewing
under a microscope, the milling process was shown to be
85–90% effective at rupturing the cells. Cells with intact mem-
branes are not stained. The vegetable fibre particles were col-
lected in organza fine mesh bags from the outlet of the colloid
mill, and washed under running cold water until all traces of
colour were removed, leaving a white MPVF preparation. After
washing, the bags were squeezed by hand to remove as much
excess water as possible. The squeezed fibre was spread in
2 cm thick layers in wire mesh drying trays and dried with fan-
forced recirculated air at 50 °C for two days. The dried MPVF
was milled using a Deawner Grinding Mill fitted with a 1 mm
mesh screen and stored in air-tight bags at room temperature
ready for use. The resulting fibre was composed of a range of
particle sizes less than 1 mm.
2.3 Functional intactness of the MPVF
The physical intactness of the MPVF was gauged by micro-
scopic examination and functional intactness by comparing
water retention capacity of the MPVF preparations with
those of commercial dietary fibre preparations using a stan-
dard procedure.
12
Dietary fibre samples (1.0 g) in duplicate
were hydrated in an excess of water for 16 h. The settled
fibre was centrifuged at 3000gfor 20 min in pre-weighed
Paper Food & Function
2646 |Food Funct.,2016,7, 2645–2654 This journal is © The Royal Society of Chemistry 2016
Published on 11 May 2016. Downloaded by Massey University on 22/06/2016 23:26:31.
View Article Online
centrifuge tubes. After aspirating the supernatant, the tubes
were reweighed and the weight of tube and fibre subtracted
to obtain a value for retained water. For microscopic exam-
ination freeze dried MPVF was wetted with a few drops of
water and mounted onto a glass slide. Samples were viewed
and imaged using an Olympus Vanox AHT3 light micro-
scope (Olympus Optical, Tokyo, Japan) under bright field
mode.
2.4 Faecal bulking analysis
2.4.1 Animals. The properties of faeces from rats fed the
vegetable fibre preparations and wheat bran reference
materials were measured under the conditions of the faecal
bulking index (FBI) assay.
13
Adult male Sprague Dawley rats
were raised on a standard pellet diet to a weight of 300 ± 50 g.
They were placed individually in hanging mesh-floored cages
in a rack containing 30 cages per side, in a controlled environ-
ment room (12 h light–dark cycle; temperature, 22 ± 1 °C;
humidity, 60 ± 5%; air exchange, 12 times per h). Water was
provided ad libitum. Ethical approval for the trial was obtained
from the AgResearch Grasslands Animal Ethics Committee,
Palmerston North (Application 13069). The animal trials were
conducted under the AgResearch Limited guidelines stated in
the “Code of Ethical Conduct for the use of Animals in
Research, Testing and Teaching”.
2.4.2 Diets and feeding. The test diets were fed in three
consecutive trials (Table 1). Each trial involved a three day lead
in period followed by a four-day balance period during which
faecal collections were made. After trials 1 and 2 the rats were
randomised and fed a high fibre clean out diet containing
10% sugar beet fibre (Fibrex®) and 5% wheat bran for three
days, before being placed on the lead-in diet for the sub-
sequent trial.
All rats were raised on a standard pelleted diet containing
dietary fibre and were transferred to a powdered high fibre
diet containing 10% sugar beet fibre (Fibrex®) and 5% wheat
bran for 1 week before being fed the trial diets over the course
of three trials. During the assay period, rats (n= 8 per treat-
ment) were fed 25 g per day of a nutritionally complete diet
consisting of a 50% diet base, which provided all the essential
nutrients and was constant for all diets, and 50% sucrose, a
proportion of which was substituted by the component being
tested for its faecal bulking effects (ESI Table S1†). The treat-
ments within the three trials are summarised in Table 1. The
cages were arranged in eight consecutive blocks of eight cages,
with each block containing all diets (baseline, reference and
six test diets). Each trial consisted of a three-day lead in
period, and a four-day balance period, during which food
intakes and spillage were accurately recorded and all faeces
collected. A three-day lead in before the balance period was
sufficient because of the high background fibre content of the
diet and complete consumption of the 25 g daily ration by the
rats, which were accustomed to the trial conditions. Faeces
were collected daily on a double thickness of blotting paper
beneath the cages over the four-day balance period and dried
under vacuum.
2.4.3 Faecal measurements. The weight of faeces produced
was determined dry and after rehydrating the intact pellets by
imbibition of water. Faecal water-holding capacity was deter-
mined by weighing a subsample (12–15 pellets) of the dried
faeces into plastic pots, as a single layer, and adding about
20% more water/sodium azide (0.02%) than was sufficient to
saturate them fully. After 2 days at room temperature excess
water was aspirated from the pots, the pots placed on a slope
to gravity drain in situ, and any further water that drained from
the pellets in 30 min was removed with a Pasteur pipette,
before the hydrated pellets were weighed.
The measurements allowed calculation of a number of
faecal parameters (ESI Table S2†) including faecal dry matter
per 100 g feed intake, faecal water-holding capacity (ml per g
dry faeces), theoretical faecal water load per 100 g diet
(ml per 100 g feed intake), equivalent hydrated faecal output
(a model of human faecal output; g rehydrated faeces
per 100 g feed intake), the increment in dry and hydrated
faecal bulk per gram of added component, the apparent
survival of MPVF after gut passage, and the faecal bulking
index (FBI; %).
Table 1 Faecal bulking trials conducted
Diet Test component
Inclusion
rate (%)
Trial 1:individual fibre effects
1.1 Swede fibre 10
1.2 Whole swede 10
1.3 Asparagus fibre 10
1.4 Broccoli fibre 10
1.5 Psyllium 2.5
1.6 Pectin 5
1.7 (Reference) Wheat bran 10
1.8 (Baseline) None 0
Trial 2: fibre interactions
2.1 Asparagus fibre 10
2.2 Swede fibre 10
2.3 Swede : asparagus fibres 1 : 1 10
2.4 Swede fibre : psyllium 1 : 1 10
2.5 Swede fibre : wheat bran 1 : 1 10
2.6 Psyllium 5
2.7 (Reference) Wheat bran 10
2.8 (Baseline) None 0
Trial 3: dose response and fibre interactions
3.1 Broccoli fibre 5
3.2 Broccoli fibre 7.5
3.3 Broccoli fibre 10
3.4 Broccoli fibre 12.5
3.5 Broccoli fibre 15
3.6 Broc/Asp/Swede fibres 1 : 1 : 1 10
3.7 (Reference) Wheat bran 10
3.8 (Baseline) None 0
Reference materials from previous trials
Psyllium husk 5
Raftilose 12.5
Sugar beet fibre (Fibrex®) 10
Apple fibre 10
Sucrose 50
Starch 50
Casein 20
Food & Function Paper
This journal is © The Royal Society of Chemistry 2016 Foo d Funct.,2016,7, 2645–2654 | 2647
Published on 11 May 2016. Downloaded by Massey University on 22/06/2016 23:26:31.
View Article Online
FBIs were calculated as the increase over baseline in re-
hydrated faecal weight induced by a diet component expressed
as a percentage of the increase over baseline due to consump-
tion of the 2 mm hard red wheat bran reference equal in weight
to the dietary component.
The following formula was used to calculate FBI:
FBI ¼ðFTFB=FRFBÞRp=Tp 100
where: FBI = faecal bulking index; F
T
= mass of rehydrated
faeces per 100 g feed intake for test diet; F
B
= mass of re-
hydrated faeces per 100 g feed intake for baseline diet; F
R
=
mass of rehydrated faeces per 100 g feed intake for reference
diet; Tp = proportion of test material in diet; Rp = proportion
of reference material in diet.
2.4.4 Validity of the rat model. The rat is a useful model
because it is a monogastric hind gut fermenter, like humans,
in which any material that escapes digestion enters a bacterial
ecosystem that contains hundreds of species of bacteria with
the collective capacity to utilise any potentially fermentable
substrate.
14
Hydrated faecal bulk thus depends more on the
characteristics of the dietary fibre than on the monogastric
host. Production of faeces by rats correlates reasonably with
stool production by humans.
15
Although the rat produces
faeces as pellets with a lower dry matter content than human
faeces, when they are allowed to imbibe water passively they
attain a moisture content close to that of humans consuming
a mixed diet.
16
Therefore the model represents the faecal mass
in a non-dehydrated human consuming enough fibre to not be
constipated. All the osmotically active components of the
faecal mass and the original faecal structure remain intact
during the analysis. The procedure is intended to model the
contribution of foods to hydrated bulk in humans consuming
a mixed balanced diet, and not to predict laxation effects in a
system perturbed by very high amounts of osmotic agents and
vigorously fermented carbohydrate.
2.5 Statistical analysis
Standard deviations and standard error of the means were
determined using Microsoft® Excel and post hoc significance
of differences between means was determined as the
maximum least significant difference (α< 0.05) using GenStat
Version 17 (VSN International, UK).
3 Results and discussion
The main proposition of this paper is that retained structural
intactness in minimally processed vegetables preserves
physicochemical properties, and that these may partially
persist through the monogastric gut to influence distal colonic
events. The fact that the MPVF preparations exhibited greater
swelling and water retention capacities than commercial
dietary fibre preparations (Fig. 1) indicated that their hydration
properties before ingestion were more intact than those of the
more extensively processed preparations, so that “minimally
processed”was a valid descriptor for the vegetable prep-
arations used in this study.
The rat has proved to be a useful model for studying the
properties of dietary fibre after monogastric digestion and
exposure to extensive mixed bacterial fermentation in the hind
gut, and there is a high correlation between the faecal bulking
capacity of fibres in man and the rat.
17
In the present experi-
ments the rats were mature and had been exposed to mixed
dietary fibre in pre-trial diets, so they readily accepted, and
would have been pre-adapted to the trial diets. The effects of
all test diets were measured as increases over baseline after
adding them to the baseline diet by substitution of sucrose.
The faecal bulking effects of the MPVFs were well within the
linear dose–response range for faecal bulking effects of
fibre.
18,19
The comparison of the swede fibre (diet 1.1) and whole
swede (diet 1.2) with the baseline in terms of faecal dry matter
(Table 2) and faecal hydration (Table 3) parameters confirmed
that the method of MPVF preparation had had the effect of
concentrating the faecal bulking component of the vegetable.
An idea of the relative faecal bulking efficacy of foods and
food components can be obtained by using wheat bran as a
reference material, as it is well known as an effective, natural,
structured faecal bulking agent in the human diet. Increases
in faecal dry weight induced by the individual MPVFs were sig-
nificantly greater than that induced by wheat bran fed at the
same intake (Table 2: diets 1.1, 10% swede fibre; 1.3, 10%
asparagus fibre; 1.4, 10% broccoli fibre, compared with diet
1.7, 10% wheat bran). The apparent survival of wheat bran was
Fig. 1 Retained functionality indicated by retained water retention
capacity of minimally processed vegetable fibres compared with com-
mercially available processed dietary fibres. Means ± SD.
FBI ¼Increase over baseline in mass of rehydrated faeces=g of diet component consumed
Increase over baseline in mass of rehydrated faeces=g of wheat bran reference 100
Paper Food & Function
2648 |Food Funct.,2016,7,2645–2654 This journal is © The Royal Society of Chemistry 2016
Published on 11 May 2016. Downloaded by Massey University on 22/06/2016 23:26:31.
View Article Online
about 40% averaged across the three trials, which is consistent
with food analyses that have shown wheat bran to be about
43% mainly insoluble dietary fibre, with starch and protein
making up most of the rest of it. It is also consistent with the
fact that wheat bran is a protective seed coating that has
evolved to be highly resistant to bacterial degradation.
Based on the faecal dry weights (Table 2), the apparent
survival of the MPVFs varied from 55 to 71%. In contrast, the
apparent survival of pectin (diet 1.6), a hydrating and fermen-
table polysaccharide, was 14%, while the apparent survival of
psyllium (diet 1.5), a hydrating but relatively non-fermentable
polysaccharide, was 99%. Assuming that pectin was fully
fermented (diet 1.6), we may estimate that every gram of
fermented polysaccharide would increase bacterial biomass
by about 0.14 g. Therefore taking an average apparent survival
(fibre ingested × 100/increase in faecal dry weight) for the
MPVFs of roughly 65% (based on Table 2 values for diets 1.1,
swede fibre; 1.3, asparagus fibre; 1.4, broccoli fibre; 2.1,
asparagus fibre; 2.2, swede fibre; 2.3, swede/asparagus fibres;
3.3, broccoli fibre; 3.6, broccoli/asparagus/swede fibres), we
can estimate conservatively that about 50% of the MPVF had
survived into the faeces. The results of trial 1 show clearly
that in contrast to extracted cell wall polysaccharide in the
form of pectin (apparent survival 14%), the structurally intact
cell walls of MPVF were approximately half fermented, so
were able to make a substantial contribution to residual
faecal mass. Microscopic examination of the faecal material
(Fig. 2) confirmed that the morphology of the parent plant
tissues in the MPVF from broccoli rind survived passage
through the rat gastrointestinal tract, in which it would have
been subjected to gastric and small intestinal digestion, fol-
lowed by fermentation by a full complement of the hind gut
microbiota.
Trial 2 examined combinations of the MPVFs (diet 2.3,
swede/asparagus; diet 2.4, swede/psyllium; diet 2.5, swede/
wheat bran) and it showed that the MPVFs from different
sources had approximately additive effects on faecal dry weight
(Table 2), as well as confirming the results of trial 1. At the
same total fibre intake the diets combining swede MPVF with
asparagus MPVF (diet 2.3) and swede MPVF with wheat bran
(diet 2.5) were intermediate between the values for the individ-
ual fibres (diet 2.1, asparagus; diet 2.2, swede; diet 1.7, wheat
bran). Similarly the effects of psyllium and swede fibre in a
1 : 1 mixture of the two (diet 2.4) were approximately additive,
suggesting that the fibre dose-faecal bulking response was
linear with the mixed fibre sources.
Table 2 Changes in faecal parameters related to dry faecal matter
Diet
b
Faecal DW
a
per
100 g diet (g)
Increase over BL
a
per 100 g diet (g)
Increase over BL
(%)
Increase per g
component (g)
Apparent
survival (%)Mean SD Mean SD Mean SD Mean SD
1.1 11.5 0.34 5.50 0.34 47.9 1.6 0.55 0.03 55
1.2 7.80 0.41 1.83 0.41 23.3 4.0 0.18 0.04 18
1.3 13.1 0.36 7.14 0.36 54.4 1.2 0.71 0.04 71
1.4 12.3 0.47 6.35 0.47 51.5 1.7 0.63 0.05 63
1.5 8.44 0.45 2.47 0.45 29.1 3.8 0.99 0.18 99
1.6 6.65 0.29 0.68 0.29 10.1 3.7 0.14 0.06 14
1.7 10.6 0.78 4.65 0.78 43.5 4.1 0.46 0.08 46
1.8 5.97 0.40
LSD 0.52 0.49 3.33 0.09 9.1
p<0.001 <0.001 <0.001 <0.001 <0.001
2.1 13.1 0.60 7.11 0.60 54.3 2.1 0.71 0.06 71
2.2 10.9 0.88 4.90 0.88 44.7 4.6 0.49 0.09 49
2.3 12.0 1.06 6.00 1.06 49.8 4.6 0.60 0.11 60
2.4 15.8 0.95 9.80 0.95 62.0 2.3 0.98 0.10 98
2.5 10.5 0.45 4.49 0.45 42.8 2.5 0.45 0.05 45
2.6 11.0 0.65 5.07 0.65 45.8 3.2 1.01 0.13 101
2.7 9.60 0.70 3.63 0.70 37.5 5.1 0.36 0.07 36
2.8 5.97 0.34
LSD 0.87 0.85 3.89 0.10 9.7
p<0.001 <0.001 <0.001 <0.001 <0.001
3.1 11.3 0.62 4.79 0.62 42.1 3.1 0.96 0.12 96
3.2 13.2 0.64 6.62 0.64 50.2 2.4 0.88 0.09 88
3.3 14.8 1.12 8.22 1.12 55.5 3.7 0.82 0.11 82
3.4 17.3 1.56 10.8 1.56 62.0 3.3 0.86 0.13 86
3.5 20.0 1.19 13.4 1.19 67.1 2.0 0.89 0.08 89
3.6 14.2 0.61 7.70 0.61 54.0 2.0 0.77 0.06 77
3.7 10.5 0.48 3.95 0.48 37.5 2.9 0.39 0.05 39
3.8 6.54 0.38
LSD 1.09 1.06 3.08 0.11 11
P<0.001 <0.001 <0.001 <0.001 <0.001
a
DW = dry weight; BL = baseline.
b
Summary of diets in Table 1.
Food & Function Paper
This journal is © The Royal Society of Chemistry 2016 Food Funct.,2016,7, 2645–2654 | 2649
Published on 11 May 2016. Downloaded by Massey University on 22/06/2016 23:26:31.
View Article Online
In trial 3 linearity of the MPVF dose-faecal bulking response
relationship was examined up to 15% broccoli rind MPVF in
2.5% steps (diets 3.1–3.5). Faecal dry matter increased almost
linearly with increasing MPVF dose, and the apparent survival
of fibre was between 82 and 96%, being 96% for 2.5% MPVF
and 89% for 15% MPVF, which was not a significant difference
(LSD = 11%; p= 0.05).
Hydrated faecal mass (Table 3) was increased substantially
over the baseline by consuming the MPVFs. Furthermore, the
MPVFs were more effective than an equal weight of wheat bran
(diets 1.1, swede fibre; 1.3, asparagus fibre; 1.4, broccoli fibre
compared with diet 1.7, wheat bran). Diets that contained psy-
llium (diets 1.5, 2.5% psyllium; diet 2.4, 5% psyllium/5%
swede fibre; 2.6, 5% psyllium), however, produced a much
greater increase in hydrated faecal mass per gram of fibre com-
ponent than the MPVFs alone. Pectin fed at twice the intake of
psyllium (diets 1.6, pectin versus diet 1.5, psyllium) had half
the effect on hydrated faecal mass, confirming the persistence
of the physicochemical properties of psyllium and their loss
during fermentation of pectin which, in unfermented form, is
a hydrating polysaccharide.
Increasing MPVF intake in the form of broccoli fibre from
2.5% to 15% of the diet increased the absolute amount of
hydrated faecal mass almost linearly (Table 3: diets 3.1–3.5)
but did not cause an increase in faecal water per gram of
MPVF consumed (Table 4: trial 3). The pre-ingestion hydration
properties of the MPVF preparations did not, therefore, predict
faecal hydration,
20
because fermentation in the hind gut
removes much of the polysaccharide responsible for hydration
of undigested vegetable cell walls. The constancy of faecal
hydration per gram of added broccoli fibre did not change sig-
nificantly between intakes of 2.5% and 15% broccoli MPVF
(Table 3: trial 3), suggesting that at all doses the broccoli fibre
had been reduced to a similar physicochemical state by gut fer-
mentation. That is, the highly hydrating pectic polysaccharide
networks of cell walls do not appear to have survived any more
at high versus low MPVF doses within the range tested. These
results are consistent with earlier findings with wheat bran
and low-methoxyl pectin that neither fibre dose nor particle
size greatly affected dietary fibre fermentation in the rat.
21
The
increase in hydrated faecal bulk due to the MPVF was due
more to the increase in dry matter, while the water holding
capacity remained more or less constant.
All of the fibre preparations in the diets increased faecal
water holding capacity (Table 4) measured by passive water
imbibition to saturation of the dry faeces. However, based on
Table 3 Changes in faecal parameters related to hydrated faecal mass
Diet
b
Faecal WW
a
/100 g Increase over BL
a
(g) Increase over BL (%)
Increase per g
component (g)
Mean SD Mean SD Mean SD Mean SD
1.1 45.7 2.3 28.6 2.3 167 14 2.86 0.23
1.2 29.0 1.9 11.9 1.9 70 11 1.19 0.19
1.3 51.9 2.4 34.8 2.4 204 14 3.48 0.24
1.4 49.4 4.0 32.3 4.0 189 23 3.23 0.40
1.5 37.9 4.8 20.8 4.8 122 28 8.32 1.93
1.6 21.8 2.3 4.7 2.3 28 13 0.94 0.46
1.7 41.0 3.6 23.9 3.6 140 21 2.39 0.36
1.8 17.1 1.0
LSD 3.5 3.5 20 0.87
p<0.001 <0.001 <0.001 <0.001
2.1 57.6 4.5 38.2 4.5 223 26 3.82 0.45
2.2 49.2 5.0 29.8 5.0 174 29 2.98 0.50
2.3 53.3 5.2 33.8 5.2 198 31 3.38 0.52
2.4 103 4.8 83.4 4.8 488 28 8.34 0.48
2.5 45.3 2.7 25.9 2.7 151 16 2.59 0.27
2.6 67.2 8.3 47.8 8.3 279 49 9.55 1.66
2.7 39.4 3.2 19.9 3.2 116 18 1.99 0.32
2.8 19.5 0.9
LSD 5.7 5.6 33 0.82
p<0.001 <0.001 <0.001 <0.001
3.1 44.3 3.5 22.8 3.5 106 16 4.55 0.71
3.2 56.1 4.1 34.5 4.1 160 19 4.60 0.55
3.3 63.3 4.6 41.7 4.6 194 21 4.17 0.46
3.4 81.6 11.6 60.1 11.6 279 54 4.80 0.93
3.5 95.3 10.0 73.7 10.0 342 47 4.91 0.67
3.6 63.6 6.0 42.0 6.0 195 28 4.20 0.60
3.7 44.0 3.1 22.5 3.1 104 14 2.25 0.31
3.8 21.6 1.4
LSD 7.7 7.6 35 0.69
p<0.001 <0.001 <0.001 <0.001
a
WW = wet weight; BL = baseline.
b
Summary of diets in Table 1.
Paper Food & Function
2650 |Food Funct.,2016,7, 2645–2654 This journal is © The Royal Society of Chemistry 2016
Published on 11 May 2016. Downloaded by Massey University on 22/06/2016 23:26:31.
View Article Online
the increase in faecal water per gram of fibre added to 100 g
diet, the fibres fell into four groups: (1) pectin (diet 1.6) had
the least effect (0.81 ml increase per g), presumably because it
was largely fermented; (2) wheat bran caused about 1.9 ml
increase per g, possibly reflecting the limitation to hydration
caused by lignin in its resilient secondary cell wall structure;
(3) MPVFs caused an increase of about 2.5–3 ml of water held
per g included in the 100 g diet; (4) psyllium caused about
8 ml increase per g (Table 4).
Psyllium led to the greatest increase in theoretical faecal
water load of several hundred percent, because, per gram of
psyllium in the 100 g diet it increased both faecal dry matter
output and faecal water holding capacity. MPVF did not
increase water holding capacity (ml per g faecal dry matter)
greatly, but the combination of the increased dry matter and
its bound water caused the percentage increase over baseline
in faecal water load per 100 g diet to move from 120% increase
at 2.5% broccoli fibre to 402% at 15% broccoli fibre (Table 4
and Fig. 3). All MPVFs alone and in combination with one
another led to about 200% increase in FWL when included at
10% of the diet. Such a large increase in dilution could be
expected to have a biologically significant effect on many
colonic processes.
To allow comparison of the faecal bulking potential of the
test components, faecal bulking indexes (FBIs) were calculated
for the test components in all diets, and compared with some
previous measurements. FBI values express the faecal bulking
effect of a test component as a percentage of the faecal
bulking effect of an equal weight of wheat bran, where bulk is
determined as the weight of fully hydrated pellets. Based on
the FBIs, the following can be seen (Fig. 4):
1. MPVFs were at least as effective as wheat bran in their
faecal bulking capacities. As discussed (above), the efficacy of
the MPVFs relative to wheat bran was due partly to slightly
greater output in faecal dry matter than induced by the wheat
bran reference, and partly due to higher water imbibition by
the dry faeces.
2. Psyllium (diets 1.5 and 2.6) was highly effective as a
faecal bulking agent, much more so than the other fibres
tested. The effectiveness of psyllium reflects the findings
(above) that it has a high apparent survival of hind gut transit,
which increased faecal dry matter, but more so because the
psyllium surviving in the faeces retained a high capacity for
water absorption.
3. The faecal bulking effects of the materials tested were
approximately additive in combination, whether the combi-
nation involved vegetable fibres alone, or vegetable fibres com-
bined with wheat bran or psyllium.
4. Pectin and raftilose, which are disperse polysaccharides
lacking in structure, acting as fermentable hindgut controls,
had very similar and relatively small faecal bulking indexes.
5. Under the conditions in which pectin was fermented,
over 50% of MPVFs survived hind gut transit, which indicates
that MPVF has the capacity to resist fermentation enough for
it to contribute equally to hindgut fermentation and faecal
bulking.
6. Commercial apple fibre and Fibrex® (sugar beet), which
are both finely ground highly processed dietary fibres, had
about half the FBI of the MPVF, indicating the importance of
retaining structure by minimal processing to achieve a
balanced hind gut functionality.
7. Materials that were, in theory, fully utilized in the foregut
(controls: sugar, starch and casein) had no detectable effect on
faecal bulking. Sugar replacement was therefore a suitable way
of including test components in the diet to measure faecal
bulking.
8. The faecal bulking indexes for broccoli rind fed at
different concentrations in the diet (diets 3.1 to 3.5) were
almost the same, reflecting the fact that the apparent survival
of MPVF was independent of intake within the 2.5–15% range
tested, and confirming the robustness of the FBI assay.
The present study has demonstrated that minimally pro-
cessed dietary fibres from vegetables are functionally more
balanced than either single fermentable polysaccharides, such
as pectin and raftilose, or highly processed commercial dietary
fibres, such as those derived from sugar beet (Fiberex®) or
apple, because they provide fermentation residues that retain
structure, hydration and bulk. Hydration and bulk influence
rates of digestion and digesta transit, with numerous direct
and indirect connections with health outcomes.
22
The results
suggest that the current practice of making up for deficient
dietary fibre intakes with prebiotic supplements or pro- and
synbiotics may not be as beneficial as increasing consumption
Fig. 2 Structure remaining in commercial apple fibre (A), commercial
sugar beet fibre (B), minimally processed asparagus fibre (C), broccoli
fibre (D), swede fibre (E) and wheat bran (F) after passing through the rat
gastrointestinal tract. The bar represents 100 µM.
Food & Function Paper
This journal is © The Royal Society of Chemistry 2016 Food Funct.,2016,7, 2645–2654 | 2651
Published on 11 May 2016. Downloaded by Massey University on 22/06/2016 23:26:31.
View Article Online
Table 4 Changes in faecal hydration properties
Diet
b
Faecal WHC
a
(ml g
−1
) Faecal moisture (%)
Increase in FWL
a
per
100 g diet (%)
Increase in water per g
added component
Mean SD Mean SD Mean SD Mean SD
1.1 2.99 0.18 74.9 1.1 203 19 2.32 0.22
1.2 2.72 0.11 73.1 0.8 88 14 1.01 0.16
1.3 2.96 0.17 74.7 1.1 243 20 2.77 0.23
1.4 3.01 0.32 75.0 1.8 228 34 2.60 0.39
1.5 3.48 0.40 77.5 2.1 161 39 7.34 1.78
1.6 2.27 0.27 69.3 2.5 34 19 0.81 0.42
1.7 2.86 0.11 74.1 0.8 169 26 1.93 0.29
1.8 1.87 0.27 65.0 3.0
LSD 0.29 2.1 28 0.81
p<0.001 <0.001 <0.001 <0.001
2.1 3.40 0.24 77.2 1.2 230 30 3.11 0.40
2.2 3.52 0.17 77.9 0.9 184 31 2.48 0.41
2.3 3.45 0.15 77.5 0.7 206 31 2.78 0.42
2.4 5.53 0.24 84.7 0.6 545 30 7.36 0.40
2.5 3.33 0.08 76.9 0.4 159 16 2.14 0.22
2.6 5.07 0.46 83.5 1.2 316 57 8.53 1.55
2.7 3.10 0.06 75.6 0.3 121 18 1.63 0.25
2.8 2.26 0.10 69.3 0.9
LSD 0.27 0.97 36 0.75
p<0.001 <0.001 <0.001 <0.001
3.1 2.91 0.17 74.4 1.1 120 20 3.59 0.71
3.2 3.26 0.21 76.5 1.2 186 24 3.72 0.55
3.3 3.29 0.18 76.7 1.0 223 24 3.35 0.46
3.4 3.69 0.26 78.6 1.2 328 67 3.94 0.93
3.5 3.77 0.28 79.0 1.2 402 60 4.02 0.67
3.6 3.46 0.27 77.5 1.3 229 36 3.44 0.60
3.7 3.20 0.18 76.1 1.0 123 18 1.85 0.31
3.8 2.30 0.10 69.6 1.0
LSD 0.25 1.3 44 0.61
p<0.001 <0.001 <0.001 <0.001
a
WHC = water holding capacity; FWL = faecal water load.
b
Summary of diets in Table 1.
Fig. 3 Effect of minimally processed vegetable fibres (MPVF) (broccoli
rind) dose on faecal water load (FWL) per 100 g diet. The baseline diet
contained 3.8% dietary fibre. Adding 15% MPVF to the baseline diet
increased FWL by more than 400%. Values are means ± sem.
Fig. 4 Faecal bulking indexes for the components tested in the three
faecal bulking trials and some comparison materials (mean ± sem). The
dotted line shows faecal bulking index for the reference (wheat bran =
100%).
Paper Food & Function
2652 |Food Funct.,2016,7, 2645–2654 This journal is © The Royal Society of Chemistry 2016
Published on 11 May 2016. Downloaded by Massey University on 22/06/2016 23:26:31.
View Article Online
of natural dietary fibre sources, because most supplements
cannot provide the benefits derived from plant morphology in
the gut. Vegetables, fruits and unrefined cereals will provide
mixed function cell wall residues that the hindgut has natu-
rally evolved to process, as well as contributing numerous
other known and unknown beneficial phytochemicals and
nutrients to the diet. Moreover, the mixed polysaccharide
nature of plant cell walls plays an important role in maintain-
ing bacterial diversity in the colon,
23
conferring resilience on
the microbiome
24
that allows it to retain its diverse contri-
butions to health.
25
There was no evidence provided in this study that consum-
ing non-fermentable hydrating polysaccharides, such as psy-
llium, to increase faecal hydration and laxation, impaired
colonic utilisation of co-consumed MPVF. However, more
detailed research should be conducted to determine whether
the persistent high viscosity milieu that non-fermented hydro-
colloids such as psyllium promote in the colon has the effect
of hindering either the bacterial utilisation of fermentable
polysaccharides or the availability of fermentation products to
the host. Should they do so, the benefits of increased laxation
may be counterbalanced by a reduction in benefits of fermen-
tation mediated by short chain fatty acids, such as gut immu-
nity. The advantage of natural vegetable tissue remnants over a
non-fermentable hydrocolloid such as psyllium is that they are
functionally multidimensional, as they can promote bacterial
activity and diversity by supplying mixed polysaccharide sub-
strates, while maintaining enough structured fermentation-
resistant polysaccharides to promote hydration and gut transit,
as the present study has shown.
The results indicate that highly processed fibres produced
for the food industry as ingredients that are compatible with
consumer tastes and baking quality standards may lack the
functionality of the parent fibres. However, as including
dietary fibres in bakery products is a helpful approach to
improving dietary fibre intakes, ways of including either mini-
mally processed fibres or combinations of fibres with comp-
lementary functions in products need to be explored.
A more thorough test of the contrast between minimally pro-
cessed vegetable fibres and highly processed fibres in their
faecal bulking capacity would have been to base the comparison
on the same vegetable matter subjected to minimal and exten-
sive processing. It was not possible in the research program that
yielded the present results but would be worth considering in a
more focused study that quantifies the relationship between
extent of processing, extent of reductions in particle size, aspect
ratio and changes in faecal parameters, with the research
ultimately extended to a human intervention study.
4 Conclusion
The results of this study have shown that the relatively un-
degraded form in which dietary fibre is consumed in whole
vegetables and wheat bran confers beneficial faecal properties
that are unlikely to be obtained from refined, structureless and
highly fermented “fibres”of which many prebiotics are typical
examples. Therefore, the advice to consume a mixed diet con-
taining vegetables, fruits and wholegrain products should be
heeded before relying on off-the-shelf fibre supplements or
highly processed fibre ingredients to make up any shortfall. A
dietary fibre deficient diet supplemented with structure-less pre-
biotics runs the risk of remaining a functionally-deficient diet.
Conflict of interest
The authors disclose no conflict of interest.
Abbreviations
BL Baseline
DW Dry weight
FBI Faecal bulking index
FWL Faecal water load
WHC Water holding capacity
MPVF Minimally processed vegetable fibres
Acknowledgements
Use of the Food Pilot Plant at Massey University for vegetable
processing with advice from the Pilot Plant Manager, Garry
Radford. Justine Shoemark raised the rats and helped with
their care.
References
1 R. G. Lentle and P. W. M. Janssen, The physical processes of
digestion, Springer, New York, 2011.
2 P. Vernocchi, L. Vannini, D. Gottardi, F. Del Chierico,
D. I. Serrazanetti, M. Ndagijimana and M. E. Guerzoni,
Front. Cell. Infect. Microbiol., 2012, 2, 156–156.
3 J. Cummings, S. Bingham, K. Heaton and M. Eastwood,
Gastroenterology, 1992, 103, 1783–1789.
4 B. V. McLeary, Cereal Foods World, 2010, 55,24–28.
5 G. P. Young, Asia Pac. J. Clin. Nutr., 2000, 9, S76–S82.
6 M. A. Eastwood and E. R. Morris, Am. J. Clin. Nutr., 1992,
55, 436–442.
7 S. Palit, P. J. Lunniss and S. M. Scott, Dig. Dis. Sci., 2012,
57, 1445–1464.
8 K. Windey, V. De Preter and K. Verbeke, Mol. Nutr. Food
Res., 2012, 56, 184–196.
9 A. McIntyre, G. Young, T. Taranto, P. Gibson and P. Ward,
Gastrointesrology, 1991, 101, 1274–1281.
10 J. A. Monro and S. Mishra, Food Dig., 2010, 1(1–2), 47–56.
11 F. Guillon and M. Champ, Food Res. Int., 2000, 33, 233–245.
12 J. Robertson, F. de Monredon, P. Dysseler, F. Guillon,
R. Amadò and J.-F. Thibault, Lebensm. Wiss. Technol., 2000,
33,72–79.
13 J. A. Monro, Asia Pac. J. Clin. Nutr., 2000, 9,74–81.
Food & Function Paper
This journal is © The Royal Society of Chemistry 2016 Foo d Funct.,2016,7, 2645–2654 | 2653
Published on 11 May 2016. Downloaded by Massey University on 22/06/2016 23:26:31.
View Article Online
14 C. D. Davis and J. A. Milner, J. Nutr. Biochem., 2009, 20,
743–752.
15 C. A. Edwards, J. Adiotomre and M. A. Eastwood, J. Sci.
Food Agric., 1992, 59, 257–260.
16 A. M. Birkett, G. P. Jones, A. M. deSilva, G. P. Young and
J. G. Muir, Eur. J. Clin. Nutr., 1997, 51, 625–632.
17 M. Nyman, N. Asp, J. Cummings and H. Wiggins,
Br. J. Nutr., 1986, 55, 487–496.
18 J. A. Monro, Asia Pac. J. Clin. Nutr., 2002, 11, 176–185.
19 J. A. Monro and E. Martinet, J. Sci. Food Agric., 2005, 85,
902–908.
20 A. M. Stephen and J. H. Cummings, Gut, 1979, 20, 722–729.
21 M. Nyman and N. G. Asp, Br. J. Nutr., 1985, 54, 635–643.
22 M. J. Gidley, Curr. Opin. Colloid Interface Sci., 2013, 18, 371–
378.
23 A. Cotillard, S. P. Kennedy, L. C. Kong, E. Prifti, N. Pons,
E. Le Chatelier, M. Almeida, B. Quinquis, F. Levenez,
N. Galleron, S. Gougis, S. Rizkalla, J.-M. Batto, P. Renault,
J. Dore, J.-D. Zucker, K. Clement, S. D. Ehrlich and
A. N. R. M. Consortium, Nature, 2013, 500, 585–588.
24 C. A. Lozupone, J. I. Stombaugh, J. I. Gordon, J. K. Jansson
and R. Knight, Nature, 2012, 489, 220–230.
25 S. Fuller, E. Beck, H. Salman and L. Tapsell, Plant Foods
Hum. Nutr., 2016, 71,1–12.
Paper Food & Function
2654 |Food Funct.,2016,7, 2645–2654 This journal is © The Royal Society of Chemistry 2016
Published on 11 May 2016. Downloaded by Massey University on 22/06/2016 23:26:31.
View Article Online
