Effects of long-term soluble vs. insoluble dietary fiber intake on high-fat diet-induced obesity in C57BL/6J mice.

Abstract

Although most of the proposed beneficial effects of fiber consumption have been attributed to viscous and gel-forming properties of soluble fiber, it is mainly insoluble cereal fiber and whole grains that are strongly associated with reduced diabetes risk in prospective cohort studies, indicating that other unknown mechanisms are likely to be involved. We performed a long-term study investigating potential protective effects of adding soluble guar fiber (10% w/w) vs. insoluble cereal fiber (10% w/w) to an isoenergetic and macronutrient matched high-fat diet in obesity-prone C57BL/6J mice. After 45 weeks, mice fed soluble vs. insoluble fiber showed both significantly increased body weight (41.8+/-3.0 vs. 33.6+/-1.5 g, P=.03) and elevated markers of insulin resistance. In mice fed soluble fiber, energy loss via the feces was significantly lower and colonic fermentation with production of short chain fatty acids (SCFA) was markedly increased. Gene expression analysis in white adipose tissue showed significantly increased levels of the fatty acid target G-protein coupled receptor-40 in soluble fiber-fed mice. Liver gene expression in the insoluble fiber group showed a pattern consistent with increased fatty acid oxidation. The present results show that soluble vs insoluble dietary fiber added to a high-fat, Western-style diet differently affected body weight and estimates of insulin sensitivity in obesity-prone mice. Soluble fiber intake with increased SCFA production significantly contributed to digested energy, thereby potentially outweighing the well known short-term beneficial effects of soluble fiber consumption.

Full text

Page 1

University of Warwick institutional repository: http://go.warwick.ac.uk/wrap

This paper is made available online in accordance with
publisher policies. Please scroll down to view the document
itself. Please refer to the repository record for this item and our
policy information available from the repository home page for
further information.
To see the final version of this paper please visit the publisher’s website.
Access to the published version may require a subscription.
Author(s): Frank Isken, Susanne Klaus, Martin Osterhoff, Andreas F.H.
Pfeiffer and Martin O. Weickert
Article Title: Effects of long-term soluble vs. insoluble dietary fiber
intake on high-fat diet-induced obesity in C57BL/6J mice
Year of publication: 2010
Link to published article:
http;//dx.doi.org/10.1016/j.jnutbio.2008.12.012
Publisher statement: Isken, F. et al. (2010). Effects of long-term
soluble vs. insoluble dietary fiber intake on high-fat diet-induced
obesity in C57BL/6J mice. The Journal of Nutritional Biochemistry, Vol.
21(4), pp. 278-284

Page 2

1
Effects of long term soluble vs. insoluble dietary fiber intake on high fat diet
induced obesity in C57BL/6J mice

Frank Iskena,b, Susanne Klausc, Martin Osterhoffa,b, Andreas F.H. Pfeiffera,b,
Martin O. Weickerta,b,*

aDepartment of Clinical Nutrition, German Institute of Human Nutrition Potsdam-
Rehbruecke, Nuthetal, Germany
bDepartment of Endocrinology, Diabetes and Nutrition, Charité-University-Medicine Berlin,
Berlin, Germany
cDepartment of Pharmacology, German Institute of Human Nutrition Potsdam-Rehbruecke,
Nuthetal, Germany

Running title: Dietary fiber and obesity

Corresponding author: Dr. Martin O. Weickert, Department of Clinical Nutrition, German
Institute of Human Nutrition, Arthur-Scheunert Allee 155, 14558 Potsdam-Rehbruecke,
Germany
Phone: +49(0)33 200 88782
Fax: +49(0)33 200 88777
E-mail: m.weickert@dife.de

Keywords: Obesity, dietary fiber, colonic fermentation, short chain fatty acids, fat oxidation,
G-protein coupled receptors

Disclosure: all authors assured that there were no conflicts of interest.

Page 3

2
Abstract

Although most of the proposed beneficial effects of fiber consumption have been attributed to
viscous and gel-forming properties of soluble fiber, it is mainly insoluble cereal fiber and
whole grains that are strongly associated with reduced diabetes risk in prospective cohort
studies, indicating that other unknown mechanisms are likely to be involved.
We performed a long-term study investigating potential protective effects of adding soluble
guar fiber (10% w/w) vs. insoluble cereal fiber (10% w/w) to an isoenergetic and
macronutrient matched high-fat diet (control) in obesity prone C57BL/6J mice. After 45
weeks, mice fed soluble vs. insoluble fiber showed significantly increased body weight (41.8
± 3.0 vs. 33.6 ± 1.5 g, P = 0.03), and markers of insulin resistance were increased. In mice fed
soluble fiber, energy loss via the feces was significantly lower and colonic fermentation with
production of short chain fatty acids (SCFA) was markedly increased. Gene expression
analysis in white adipose tissue showed significantly increased levels of the fatty acid target
G-protein coupled receptor-40 (Gpr40) in soluble fiber fed mice. Liver gene expression in the
insoluble fiber group showed a pattern consistent with increased fatty acid oxidation. The
present results show that soluble vs insoluble dietary fiber added to a high-fat, high-fat
Western-style diet differently affected body weight and insulin sensitivity in obesity prone
mice. Soluble fiber intake with increased SCFA production significantly contributed to
digested energy, thereby potentially outweighing the well known short-term beneficial effects
of soluble fiber consumption.

Page 4

3
Introduction

Recent meta-analyses of prospective cohort studies report markedly reduced diabetes risk in
subjects consuming diets high in insoluble cereal dietary fiber and whole grains [1, 2]. In
contrast, and surprisingly, there is no strong support that soluble viscous fibers from fruits and
vegetables play a key role in this context [1-3], although most of the proposed protective
mechanisms of fiber consumption are either more likely to be relevant with soluble fiber, or
they are shared by soluble and insoluble fiber [3]. For instance, viscous and gel-forming
properties of soluble fibers are involved in hindering of macronutrient absorption, slowing of
gastric emptying, reducing postprandial glucose responses, and reducing total and LDL
cholesterol levels. Colonic fermentation of naturally available high fiber foods with the
production of short chain fatty acids (SCFA) can also be mainly attributed to soluble fiber
consumption, whereas effects on inflammatory markers [4] and moderate weight loss due to
low energy density and increased satiety have been reported to be comparable with both
soluble and insoluble fiber [3, 5]. Therefore, it is likely that further mechanisms are involved
in conveying reduced diabetes risk after long term insoluble fiber intake. These might include
improved whole-body insulin sensitivity, as recently shown in short-term randomized
controlled intervention studies [6-8] and several cross-sectional studies in humans [9].
However, no causal relationships can be stated from cross-sectional studies, and long term
dietary interventions in humans face the problem of controlling confounding factors and
maintaining dietary adherence [10]. Difficulties investigating signalling pathways in the liver
are another problem in human studies. Several relatively short-term studies in animal models
reported soluble fiber such as guar gum or psyllium being superior in improving insulin
sensitivity, in comparison to insoluble cellulose [11, 12]. In a study investigating male Wistar
rats, short-term feeding with guar gum vs. cellulose or bran had favourable effects on body
weight and carbohydrate tolerance. However, and importantly, in the long-term these effects

Page 5

4
were absent, with a tendency to reduced body weight and significantly lower pancreatic
insulin and glucagon concentrations in the cellulose fed rats after 67 weeks [13]. However,
different effects may be observed in other species, and long-term effects of supplementing a
Western-style high-fat diet with soluble vs. insoluble fiber in obesity prone mouse models are
unknown. Potential fiber induced influences on transcription factors in liver and adipose
tissue are of further interest and might provide important insights for the understanding of
beneficial effects of fiber consumption. Therefore, in the present study we investigated in
C57BL/6J mice whether a long term supplementation of a high-fat, Western-style diet with
soluble viscous guar fiber vs. highly purified insoluble cereal fiber differently affects body
weight, liver fat, insulin sensitivity, and gene expression of metabolic markers in liver and fat
tissue.

Methods and Materials

Animals
The protocol for all animal experiments was approved by the local governmental animal ethic
review board (State of Brandenburg, Germany). The animals were kept in accordance with the
NIH guidelines for care and use of laboratory animals. After 45 weeks, mice were sedated
using ether inhalation and sacrificed by decapitation.
Experiments were performed in adult (16 weeks old) male C57BL/6J mice obtained by
Charles River, Germany. Animals were housed individually at a temperature of 22°C with a
12:12-h light-dark cycle in cages with soft wood bedding. Animals were divided into three
groups (n = 7 per group) and received three different diets (Table 1) with unlimited access to
chow and liquids over 45 weeks. Metabolizable macronutrients of diets were calculated
according to the following energy contents: casein 15.7 kJ/g, carbohydrates 16 kJ/g, and fat
38 kJ/g. After the experimental period animals were sacrificed in fed state. Organs were

Page 6

5
isolated after rapid preparation. Epididymal white adipose tissue and liver tissue were
submerged in nitrogen and immediately stored at -80 °C until further RNA preparation.

Diets
The macronutrient composition of the experimental diets is shown in Table 1. The diet
enriched with soluble fiber contained 10% (w/w) guar gum, which is an established model for
the investigation of soluble fiber diets (64% soluble fiber, 13% insoluble fiber) (Kumar J Nutr
Biochem 2002, Owusu-Asiedu J Anim Sci 2006). The diet enriched with insoluble fiber
contained 10% of the insoluble fraction of oat fiber, as previously used [8, 14, 15] providing a
mixture of insoluble fibers as found in cereal fiber and whole grain products (3 % soluble
fiber, 93 % insoluble fiber (cellulose 70%, hemicelluloses 25%, lignin 3-5%). Food intake
rate was recorded every week.

Digestibility of diets
Feces of all animals were collected at week 5 and food intake was recorded for further
analysis of energy balance. After drying, energy content of diet samples and feces was
determined by bomb calorimetry (IKA C5003, IKA Werke, Germany) and digested energy
(defined as diet energy intake (kJ/g) minus energy loss via the feces (kJ/g)) was calculated for
all dietary groups. Digestibility of the diet (%) was defined as [(digested energy (kJ/g)/diet
energy intake (kJ/g)) ∙ 100]. Cumulative digested energy was calculated over the experimental
period of 45 weeks, by multiplying diet energy intake with digestibility, as measured over one
week.

Body composition

Page 7

6
Body composition (fat mass, lean mass, and free fluids) was measured every four to six weeks
using nuclear magnetic resonance spectroscopy (Mini Spect MQ 10 NMR Analyser Bruker,
Karlsruhe, Germany). Body weight was recorded every week.

Hydrogen (H2) breath test as a marker of colonic fermentation
This test was performed twice in all mice in the fed state between 08:00 and 09:00. In order to
collect hydrogen exhalation samples animals were placed individually into 140 ml syringes.
After 1 min of equilibration 2 samples of 30 ml air breath were drawn of for further analysis.
Air breath samples were collected and H2 concentrations were measured using a breath
hydrogen analyser (Quinton Model-12i-Microlyzer; Quintron Instruments, Milwaukee, USA).

Analysis of hepatic triacylglycerol
Frozen liver tissue was ground in liquid nitrogen to a homogenous powder. 100 mg of tissue
was homogenized in 5 ml of 10 mM sodium phosphate buffer containing 1 mM EDTA and
1% polyoxyethylene 10 tridecylethan, using an Ultra-Turrax (IKA Werke, Germany).
Samples were centrifuged (10 min, 20.000 g) and the supernatant was incubated at 70 °C for 5
min. Triacylglycerols (triglyceride reagent, SIGMA) and protein (DC protein assay, Bio-Rad)
levels were analysed in triplicates.

Insulin tolerance test
Insulin sensitivity was estimated after 45 weeks of dietary intervention in fed mice by
intraperitoneal (i.p.) injection of insulin (0.75 IU/kg body mass, Actrapid®, Novonordisk) as
described previously [16]. There was a six days period between ITT and sacrification of the
animals. Glucose concentrations were measured in tail blood at 0, 15, 30, and 60 min after
insulin injection.

Page 8

7
Glucose tolerance test
Glucose tolerance tests were performed by intraperitoneal glucose injection, as described
(Isken Horm Metab Res 2006). Plasma was collected before and 10, 30, and 60 min after
glucose challenge and immediately frozen at -80°C for measurement of glucose and insulin.

Plasma analyses
Animals were investigated in the over-night fasted state. Mouse plasma insulin levels were
measured by ELISA for rat insulin using a mouse insulin standard (both from Crystal Chem
Inc., Chicago, Illinois, USA), as described [17]. Blood was obtained from the retro-orbital
sinus during anaesthesia using Isoflurane® (Baxter, Unterschleissheim, Germany). Plasma
glucose, plasma triacylglycerols, and plasma total cholesterol were measured using
commercial kits (glucose HK 125; triacylglycerols and total cholesterol: ABX Pentra,
Montpellier France), by using an autoanalyzer (Cobas Mira S, Hoffmann La Roche,
Switzerland).

RNA extraction and real time RT-PCR
Total RNA was extracted from liver and epididymal white fat tissue of animals in non fasted
state by RNeasy lipid tissue kit® (QIAgen GmbH, Germany). DNA digestion was performed
using RNase free DNase® (QIAgen GmbH, Germany). Total RNA (1 μg) was reverse-
transcribed to first-strand cDNA using high capacity cDNA Reverse Transcription Kit
(Applied Biosystems, USA), according to manufacturer’s protocols. Each cDNA sample was
applied at a concentration of 0.626 ng of total RNA to the qRT-PCR assay wells, using
optical 384-well plates, and labelled with Power SYbr® Green master mix (Applied
Biosystems, Germany). All samples were measured in triplicates, and non template controls
were used to confirm specifity. The quantity of target and the housekeeping gene
hypoxanthine-guanine phosphoribosyltransferase (HPRT) were calculated according to a

Page 9

8
standard curve. In liver tissue, expression of forkhead transcription factor Foxa2, Pparγ
coactivator ß (Pgc-1ß), L-carnitine palmitoyl transferase1 (Cpt-1), acyl-CoA oxidase (Aox),
diacylglycerol acetyltransferase-2 (Dgat-2), uncoupling protein2 (Ucp-2), and peroxysome
proliferator-activated receptor-α (Pparα) were measured. Expression of G protein-coupled
receptors 40, 41, and 43 (Gpr40 (free fatty acid receptor (Ffar-1)), Gpr41 (Ffar-3), and Gpr43
(Ffar-2)) was analysed in white adipose tissue.
The oligonucleotide specific primers were:
Hprt:
up5'-CAGTCCCAGCGTCGTGATTA-3', lo5'-AGCAAGTCTTTCAGTCCTGTC-3',
Foxa2:
up5'-GCGGCCAGCGAGTTAAAGTAT-3', lo5'-TCATTCCAGCGCCCACATA-3',
Pgc-1ß:
up5'-ATGAAGGCGACACACCATCCT-3', lo5'-TGCCATCCACCTTGACACAAG-3',
Cpt-1:
up5'-CCTGCATTCCTTCCCATTTG-3', lo5'-CCCATGTCCTTGTAATGTGCG-3',
Aox:
up5'-TACTTGAATGACCTGCCGAGC-3', lo5'-GCAGCAATTTCTACCAATCTGG-3',
Dgat-2:
up5'-CCAAGAAAGGTGGCAGGAGAT-3', lo5'-GCAGGTTGTGTGTCTTCACCA-3',
Ucp-2:
up5'-CCAACAGCCACTGTGAAGTTCC-3', lo5'-TGACTCTCCCCTTGGATCTGCA-3',
Pparα:
up5'-CAGTGCCCTGAACATCGAGTGT-3', lo5'-TTCGCCGAAAGAAGCCCTT-3',
Gpr40:
up5'-TGGCTAGTTTCATAAACCCGG-3', lo5'-TCCCAAGTAGCCATGGACCAGT-3'
Gpr41:

Page 10

9
up5'-CTTGTATCGACCCCCTGGTTTT-3', low5'-GCTGAGTCCAAGGCACACAAGT-3'
Gpr43:
up5'-TGTTCAGTTCCCTCAATGCCA-3', lo5'-CAGGATTGCGGATCAGTAGCA-3'

Statistical analysis
Quantitative date are presented as means ± SEM. Data were analysed using one-way ANOVA
with Bonferroni post hoc test, or two-tailed Student`s t test for unpaired samples (SPSS 14 ,
Chicago, USA). Not normally distributed data were calculated using non parametric Mann
Whitney U-test. Six of seven mice in the fiber groups, but only three of seven mice in the
control group reached the end of the intervention in week 45, indicating that minimum fiber
contents in mouse diet are essential. Therefore, because of low number of animals in the
control group further analysis was restricted to comparisons between fiber groups. P < 0.05
was considered significant.

Results

Different effects of soluble guar fiber vs. insoluble cereal fiber on body weight gain and
insulin resistance in mice fed a Western-style diet
Fig. 1A shows changes in body weight during the experiment. The observed drops in body
weight at weeks 9/10 and 15/16 were induced by overnight fasting procedure for the
performance of GTTs, with AUC glucose showing no difference between soluble and
insoluble fiber fed animals (P = 0.68 and P = 0.4, respectively). After 45 weeks, mice fed the
soluble fiber diet showed significantly increased body weight, in comparison to mice fed the
insoluble fiber diet (41.8 ± 3.0 g 33.6 ± 1.5 g, P = 0.03). Increased body weight was
accompanied by a tendency to increased body fat (Fig. 1B, P = 0.14). The exact body
composition (free fluid, lean mass and fat mass) in week 43 is shown in Fig 1C. In agreement

Page 11

10
with the observed change in body weight, measurements of insulin sensitivity after i.p. insulin
administration at week 45 were significantly lower after soluble vs. insoluble fiber
consumption (Fig. 2; P = 0.044). AUC tended to be lower in insoluble fiber fed animals
(298.8 26.2 vs. 344.5 15.8 mmol x min, P = 0.19). No significant differences were detected
in fasting concentrations of plasma cholesterol, plasma triacylglycerols, plasma glucose, and
plasma insulin (data not shown).

Increased energy digestion with soluble vs. insoluble fiber
For analysis of digestibility, fecal excretion and food uptake were recorded at week 5 for
further analysis of digested energy. Energy content of the respective diets measured in a
calorimeter was 19.5 kJ/g in the fiber free control group, 18.8 kJ/g in the group fed with
soluble fiber and 17.6 kJ/g in the group fed with insoluble fiber. Food intake was multiplied
with these values. Fig. 3A shows no significant difference in diet energy intake during a one
week period. However, fecal excretion was significantly increased with the insoluble fiber
diet, both compared to soluble fiber (P < 0.0001) and fiber free control (P < 0.0001) (Fig.
3B). The difference in feces weight between soluble fiber and fiber free control was also
significant (P = 0.0002). Energy content in feces was significantly higher both in soluble fiber
fed mice (P < 0.0001) and in insoluble fiber fed mice (P < 0.0001), as compared to the fiber
free control group, but not significantly different between the fiber groups (P = 0.16).
Digestibility of the diets was: 94% with soluble fiber, 89% with insoluble fiber, and 97% with
fiber free control. Digested energy between groups was not significantly different in the short
term (Fig. 3D), as measured over one week (soluble fiber 365.0 ± 8.0 vs. insoluble fiber 340.8
± 10.8 kJ/week, P = 0.1). However, long term cumulative energy digestion over 45 weeks was
significantly different between soluble and insoluble fiber consumption (AUC 350577 ± 3919
kJ/mouse vs. 316367 ± 6569 kJ/mouse, P = 0.0025) (Fig. 3E).

Page 12

11
Increased colonic fermentation with soluble fiber
Colonic fermentation of each diet was estimated by using hydrogen breath tests at week 4
(Fig. 4A). As could be expected, hydrogen exhalation was significantly higher in mice fed the
soluble fiber diet, both in comparison with mice fed insoluble fiber (P = 0.002) or fiber free
control (P = 0.003).

Fiber-induced changes in adipose tissue mRNA expression of G-protein coupled receptors
The G-protein coupled receptor (Gpr) family Gpr40, Gpr41, and Gpr43 shares approximately
30% identity among members [18] and has recently been shown to be specifically activated
by SCFA [18, 19]. In agreement with significantly increased colonic fermentation in mice fed
soluble vs. insoluble fiber, expression in white adipose tissue of Gpr40 was significantly
increased after 45 weeks of dietary intervention (P = 0.042), (Fig. 4B). Gpr41 was not
detectable in adipose tissue, as also observed by others [20]. No significant differences were
observed in Gpr43 (Fig. 4B, P = 0.47).

Expression of hepatic transcription factors of fat metabolism
After 43 week hepatic triacylglycerol contents tended to be higher in mice fed with soluble
fiber (Fig. 5A), P = 0.11. We further investigated fiber induced changes of factors involved in
the regulation of liver fat metabolism. Foxa2 is an important transcription factor in liver and
fat tissue, regulating e.g. beta oxidation, triacylglycerol synthesis, and fat cell differentiation
[21]. After intervention with insoluble fiber expression of Foxa2 was significantly increased,
as compared with the soluble fiber fed animals (Fig. 5B, P = 0.031). In agreement with this,
there was also significantly increased expression of Pgc-1ß (Fig. 5B, P = 0.006), which is
another transcription factor known to be recruited by Foxa2. Expression of key enzymes of
beta-oxidation and triacylglycerol synthesis was further analysed. Dgat-2 was significantly
increased in liver tissue of mice fed with insoluble fiber (P = 0.045), whereas Cpt-1 (P =

Page 13

12
0.07), Aox (P = 0.37), and Uxp-2 (P = 0.12) were comparable between groups (Fig. 5B). The
expression of Pparα, which plays a key role in fatty acid catabolism was significantly higher
in insoluble fiber fed animals (P = 0.042).

Page 14

13
Discussion

A high fiber intake is emphasised in the recommendations of most nutritional and diabetes
associations. Although most of the proposed beneficial effects of fiber consumption have
been attributed to viscous properties of soluble, fermentable sorts of fiber, results from
prospective cohort studies consistently show that insoluble cereal fiber and whole grain
consumption, but not soluble fiber intake is strongly linked to reduced risk of diabetes [1-3,
22]. We present novel findings showing that long-term supplementation of a Western-style
diet with soluble, highly fermentable guar fiber vs insoluble, moderately fermentable cereal
fiber resulted in an obese phenotype in C57BL/6J mice. Weight gain and insulin resistance
were significantly different between groups, although dietary energy intake was comparable.
Mice in the present study had free access to chow. Therefore, significantly lower energy loss
via the feces in the soluble fiber group together with comparable dietary energy intake in all
mice suggests that more metabolizable energy was extracted from the soluble, highly
fermentable fiber diet, without affecting satiety and feeding behaviour. This is supported by
earlier studies showing that fiber induced increases of SCFA do contribute to energy intake
both in rodents and humans [23]. Interestingly, colonization of germ-free gnotobiotic mice
with a prominent saccharolytic member of the normal human gut microbiota together with the
dominant human methane producing germ results in markedly improved colonic fermentation
and is associated with an obese phenotype in the host [3, 24]. As could be expected, H2-
measurements indicated that colonic fermentation with the production of SCFA was largely
increased in mice consuming soluble fiber, but not in animals ingesting insoluble cereal fiber.
Fiber induced increases of SCFA production are commonly assumed to be beneficial, e.g. by
reducing hepatic glucose output and beneficially affecting blood lipid profiles [25, 26].
Increased insulin sensitivity after relatively short term consumption of highly fermentable
resistant starch has also recently been shown [7]. However, both highly fermentable insoluble

Page 15

14
resistant starch [7, 27] and only moderately fermentable insoluble cereal fiber [8, 14]
increased insulin sensitivity in humans, indicating that a dose-dependent relation between
fermentability of dietary fibers and insulin sensitivity is unlikely [3]. Further, available studies
indicate that SCFA could contribute to increased de novo lipogenesis [24], probably by
stimulating adipogenesis through Gpr43 [18], although in the present study Gpr43 expression
was not significantly increased in fermentable fiber fed mice. However, expression of Gpr40
representing another recently identified target of fatty acids including SCFA [19] that is
mainly expressed in pancreas and brain, but also in monocytes and in adipose tissue, was
significantly increased in the soluble fiber group. Loss of Gpr40 protects mice from obesity-
induced hyperinsulinemia, hepatic steatosis, increased hepatic glucose output, hyperglycemia,
and glucose intolerance [28]. Vice versa, increased Gpr40 in soluble fiber fed mice could
provide a potential mechanism contributing to the observed obese phenotype in these animals.
A further novel finding of the present study was liver gene expression in the insoluble fiber
group showing a pattern consistent with increased fatty acid oxidation. Mice fed insoluble
fiber showed significantly increased expression levels of Foxa2 and Pgc-1b in comparison to
soluble fiber fed animals. These transcription factors are known to play an important role in
the regulation of hepatic lipid homeostasis. Co-expression of Foxa2 and Pgc-1b has been
reported to concomitantly increase gene expression of key enzymes of mitochondrial beta-
oxidation [29]. Significantly increased expression levels in liver of Pparα and Dgat-2 in
insoluble fiber fed mice in the present study were in good agreement with these findings, thus
potentially contributing to increased beta-oxidation and reduced fat mass in mice fed
insoluble cereal fibers. Differences in liver gene expression were likely caused by fiber
induced differences in body weight and insulin resistance, although a direct influence of fiber
types on liver fat metabolism can not be finally excluded.
It has been suggested that soluble fibers such as guar gum increase postprandial fullness and
reduce food intake, e.g. by increasing the viscosity of the bowel content [30]. However, in a