http://www.kidney-international.
org
& 2006 International Society of Nephrology
Butyrate modulates TGF-b1 generation and function:
Potential renal benefit for Acacia(sen) SUPERGUMTM
(gum arabic)?
N Matsumoto1, S Riley1, D Fraser1, S Al-Assaf2, E Ishimura3, T Wolever4, GO Phillips2 and AO Phillips1
1
Institute of Nephrology, Cardiff University School of Medicine, Heath Park, Cardiff, UK; 2North East Wales Institute, Phillips Hydrocolloids
Research Centre, Wrexham, UK; 3Department of Nephrology, Osaka City University, Osaka, Japan and 4Division of Endocrinology and
Metabolism, St Michael’s Hospital, Toronto, Ontario, Canada
Anecdotal evidence suggests that high fibre
supplementation of dietary intake may have health benefits
in renal disease related to alterations in circulating levels of
short-chain fatty acids. The aim of the study was to examine
the hypothesis that dietary manipulation may increase serum
butyrate and thus have potential beneficial effects in renal
disease. We examined the effect of dietary supplementation
with a gum arabic sample of standardized molecular
characteristics, Acacia(sen) SUPERGUMTM EM2
(SUPERGUMTM), on systemic levels of butyrate in normal
human subjects. In an in vitro study, we also examined the
potential role of butyrate in modifying the generation of
the profibrotic cytokine transforming growth factor-beta
(TGF-b1) by renal epithelial cells. Following 8 weeks of dietary
supplementation with 25 g/day of SUPERGUMTM, there was a
two-fold increase in serum butyrate (n ¼ 7, P ¼ 0.03). In vitro
work demonstrated that exposure of renal epithelial cells to
elevated concentrations of butyrate suppressed both basal
and stimulated TGF-b1 synthesis. The action of butyrate was
mediated by suppression of the extracellular signal-regulated
kinase/mitogen-activated protein kinase signalling pathway.
In addition, butyrate exposures reduced the response of
renal epithelial cells to TGF-b1 as assessed by luciferase
activity of a TGF-b-responsive reporter construct. Attenuation
of TGF-b1 signalling was associated with reduced
phosphorylation of Smad 3 and decreased trafficking of
TGF-b1 receptors into signalling, non-lipid raft-associated
membrane fractions. In conclusion, the data demonstrate
that dietary supplementation with SUPERGUMTM increased
serum butyrate, which at least in vitro has beneficial effects
on renal pro-fibrotic cytokine generation.
Kidney International (2006) 69, 257–265. doi:10.1038/sj.ki.5000028
KEYWORDS: TGF-b; renal; butyrate; SUPERGUM
Correspondence: AO Phillips, Institute of Nephrology, School of Medicine,
Cardiff University, Heath Park, Cardiff CF14 4XN, UK.
E-mail: PhillipsAO@cf.ac.uk
Received 15 June 2005; revised 15 August 2005; accepted 15 September
2005
Kidney International (2006) 69, 257–265
original article
Epidemiological evidence suggests that a high intake of
dietary fibre is associated with numerous health benefits.1 To
date, much of the research interest has focused on their role
in protection against colorectal cancer and inflammatory
bowel disease. However, there is now evidence that altered
systemic levels of short-chain fatty acids (SCFA), resultant
from colonic bacterial fermentation of fibre, may also offer
health benefit beyond the gastrointestinal tract. Studies over
the last decade have linked dietary fibre intake and
cardiovascular risk and also epithelial cell carcinogenesis in
numerous organs including the kidney. These health benefits
of high-fibre diets are thought to be specifically linked to the
generation of the SCFA butyrate, which displays a host of
biologically important effects in vitro, including modulation
of cell proliferation, apoptosis, regulation of angiogenesis and
inflammation.2
Gum arabic is a tree gum exudate, which has been an
important food ‘additive’ since ancient times. Supplementa-
tion of the diet with gum arabic has been shown to increase
faecal nitrogen excretion and lower serum urea nitrogen
concentration in patients with chronic renal failure.3 This
is dependent on an increase in bacterial growth and activity
in the gut. Colonic bacteria produce ureases that hydrolyse
urea to ammonia and CO2. The resultant ammonia can
then be incorporated into bacterial proteins, which are
subsequently excreted in the bacterial mass fraction of the
faeces. The net result is increased N excretion in the faeces.
More recent studies, in rat models of acute renal failure,4
suggest that gum arabic may also improve renal function
independently of its action on faecal bacterial ammonia
metabolism.
It is now clear that in renal disease, regardless of its
aetiology, the prognosis best correlates with the degree of
renal interstitial fibrosis. As a result, interest has focused on
the role of cells, such as the epithelial cells of the proximal
tubule, in the initiation of the fibrotic response. There is now
overwhelming evidence that the pro-fibrotic cytokine trans-
forming growth factor-beta (TGF-b1) is a key mediator of
progressive renal injury. Much of our research work has
257
original article N Matsumoto et al.: Gum arabic, short-chain fatty acids and TGF-b1

focused on the generation and function of this cytokine by a P = 0.03

proximal tubule. 1.2

The aim of this study was to examine the potential use of
1.0
dietary supplementation with gum arabic in progressive renal
disease. More specifically, we have examined the relationship

Butyrate (
mol/l)
0.8
between ingestion of gum arabic and systemic levels of SCFA
in human subjects, and examined the effects of the SCFA 0.6
butyrate and propionate on the generation of, and response
to, the pro-fibrotic cytokine TGF-b1 in vitro in proximal 0.4

tubular epithelial cells.

0.2
A weakness associated with previous dietary fibre studies
involving gum arabic has been the inherent variable nature of 0.0
this natural product.5 This study comprising 75 commercial Prestudy Post-gum arabic
samples showed that their average molecular weight varied
b
from 4.6 to 10.2  105. The traditional gum arabic used in the 3.5

food industry is the exudate from the tree Acacia senegal.

3.0
There are two variety forms available commercially, namely

Propionate conc (
mol/l)

A. senegal var. senegal and A. senegal var. karensis. Both are 2.5
acceptable as food additives and conform to the specification
now approved by the FAO Joint Expert Committee on Food 2.0
Additives and the Codex Alimenarius Commission. Here, we
have used an A. senegal var. senegal sample, which has been 1.5

matured to give a standardized and reproducible test

1.0
material, ensuring that the work can be accurately repeated
and that any future product will be able to reproduce the 0.5
effects reported. Prestudy Post-gum arabic

Figure 1 | Dietary supplementation with SUPERGUMTM increases

RESULTS
Dietary supplementation increases serum butyrate
The SUPERGUMTM was well tolerated, with only minor
symptoms of bloating being reported. All volunteers
completed the 10-week course of fibre supplementation.
There was a significant increase in serum butyrate levels
from a baseline value of 0.4970.08 to 0.7170.11 mmol/l after
10 weeks (Figure 1a). Owing to technical reasons, results
from only seven volunteers were available for butyrate levels.
serum butyrate. At the end of a 10-week study period in which
normal healthy volunteers were fed 25 g/day of SUPERGUMTM, serum
samples were collected and (a) butyrate or (b) propionate were
quantified as detailed in Materials and Methods. Data represent
median and inter-quartile range; for butyrate n ¼ 7 and for
propionate n ¼ 10.
a 24 h b 48 h c 72 h
150
300 400

There was no change in the serum propionate (baseline: *

TGF-1 (ng/ml)

TGF-1 (ng/ml)

TGF-1 (ng/ml)

100 300
1.6370.24 mmol/l; study end: 1.5970.25 mmol/l, n ¼ 10, * 200 *
*
P40.05) during the course of the fibre supplementation 200
50
period (Figure 1b). 100
100
Butyrate decreases basal TGF-b1 generation. Addition of
butyrate led to a dose-dependent reduction in the generation 0
0 0.5 1 5 10
0
0 0.5 1 5 10
0
0 0.5 1 5 10
of TGF-b1 (Figure 2). A significant reduction in this Concentration (mM) Concentration (mM) Concentration (mM)
‘unstimulated’ constitutive generation was seen at concentra- Figure 2 | Butyrate suppresses basalTGF-b1 production by
tions of butyrate of 5 mM and above. In contrast, addition of tubular epithelial cells. Increasing doses of butyrate were added to
propionate over the concentration range 0–10 mM did not confluent monolayers of serum-deprived HK2 cells under serum-free
conditions. Cell culture supernatant samples were collected after
influence TGF-b1 synthesis at any of the time points studied (a) 24, (b) 48, and (c) 72 h and TGF-b1 was quantified by ELISA. Data
(Figure 3). represent mean7s.d.; n ¼ 6. *Po0.05 compared to serum-free
The doses of the SCFA used were non-toxic. Over the medium in the absence of butyrate.
concentration range used, there were no alterations in
fluorescence readings following addition of butyrate or protein (MAP) kinase pathway is involved in the stimulation
propionate, as assessed by alamarBlue assay (data not of promoter activity, whereas the extracellular signal-
shown). regulated kinase (ERK)/MAP kinase pathway may regulate
Our previous studies have demonstrated independent post-transcriptional events.8 The potential involvement of
mechanism regulating transcription and translation of TGF- these two signalling pathways in TGF-b1 generation under
b1.6,7 Furthermore, signalling via the p38 mitogen-activated basal conditions was assessed by determining the impact of

258 Kidney International (2006) 69, 257–265

N Matsumoto et al.: Gum arabic, short-chain fatty acids and TGF-b1 original article

inhibition of these pathways on TGF-b1 concentration.
SB203580 and PD98059 (Calbiochem, Nottingham, UK)
are cell-permeable inhibitors of p38 kinase and MAP kinase
kinase (MEK), respectively. Addition of the MEK inhibitor
PD98059 (Figure 4a), but not the p38 MAP kinase inhibitor
(Figure 4b), mimicked the effect of butyrate and led to a
significant reduction in TGF-b1 concentration in the culture
supernatant (Figure 5). Neither compound over the con-
centration ranges used led to any toxic effect on cells as
a 24 h b 48 h
150
TGF-1 (ng/ml)
assessed by the alamarBlue assay (data not shown). The
hypothesis that butyrate may mediate its effect by inhibition
of ERK/MAP kinase was also supported by immunoblot
analysis demonstrating both dose-(Figure 4c) and time-
dependent (Figure 4d) suppression of ERK phosphorylation
in unstimulated monolayers of HK-2 cells.
Glucose-stimulated TGF-b1 synthesis is inhibited by butyrate.
We have previously demonstrated, in our in vitro model of
the diabetic state, that exposure of HK-2 cells to elevated
concentrations of D-glucose for 7 days stimulates de novo
TGF-b1 synthesis.8 Having determined the effect of SCFA on
c 72 h basal/constitutive TGF-b1 synthesis, we next sought to
400
500
examine their effect on glucose-stimulated TGF-b1 synthesis.
300 Addition of 25 mM D-glucose led to a significant increase

TGF-1 (ng/ml)

400
TGF-1 (ng/ml)

100

200
50
100
0 0
0 0.5 1 5 10
Concentration (mM)
Figure 3 | Propionate has no effect on basal TGF-b1production
by tubular epithelial cells. Increasing doses of propionate were
added to confluent monolayers of serum-deprived HK2 cells under
serum-free conditions. Cell culture supernatant samples were
collected after (a) 24, (b) 48, and (c) 72 h and TGF-b1 was quantified
by ELISA. Data represent mean7s.d.; n ¼ 6. P40.05 for all samples
compared to serum-free medium in the absence of propionate.
300 in the concentration of TGF-b1 in the culture supernatant, as
200
previously demonstrated.8 Glucose-stimulated increase in
TGF-b1 was significantly attenuated when cells were cultured
100
in the presence of 25 mM D-glucose together with butyrate
0
0 0.5 1 5 10 0 0.5 1 5 10
(Figure 5a). Propionate in contrast did not influence glucose-
Concentration (mM) Concentration (mM) stimulated TGF-b1. To confirm that reduced TGF-b1
concentration following addition of butyrate was the result
of decreased synthesis, we assessed de novo synthesis of TGF-
b1 by detection of radioactive amino-acid incorporation into
TGF-b1.8 Consistent with the enzyme-linked immuno-
sorbent assay (ELISA) data, 25 mM D-glucose-stimulated
synthesis of radiolabelled TGF-b1 was also reduced in the

a 125
b
100

100
75
TGF-1 (ng/ml)

TGF-1 (ng/ml)

75 ∗
50
50

25
25

0 0
PD98059 0 0.8 8.0 80
M SB203580 0 30 300 3000 ng/ml

c d

Phospho-ERK

0 0.1 1 mM 0 0.1 1 mM
5 10 15 30 min
Butyrate Propionate Butyrate – + – + – + – +

Total ERK

Figure 4 | Butyrate-mediated suppression of TGF-b1 is associated with suppression of ERK activation. Increasing doses of either (a) the
MEK inhibitor PD98059 or (b) p38 MAP kinase inhibitor SB203580 were added to confluent monolayers of HK-2 cells under serum-free
conditions for 24 h. Subsequently, cell culture supernatant samples were collected for analysis of TGF-b1 concentration by ELISA. Data
represent mean7s.d.; n ¼ 3. *Po0.05 compared to serum-free medium alone in the absence of inhibitor. In parallel experiments, the effect of
butyrate on ERK activation was examined by immunoblot analysis using antibodies to either dually phosphorylated active form of MAPK (p44/
ERK1 and p42/ERK2) or total MAPK. (c) To determine the dose-dependent effect of SCFA, butyrate or propionate at the concentrations shown
was added to confluent HK2 cell monolayers in the absence of serum for 30 min. (d) Time-dependent effects of butyrate were examined by
addition of 1 mM butyrate to confluent monolayers of HK2 cells for up to 30 min. Cell lysis and immunoblot analysis were carried out as detailed
in Materials and Methods.

Kidney International (2006) 69, 257–265 259

original article
a 1500
1000 *
TGF-1 (ng/ml)
500
0 pERK
0 0.1 1
Butyrate/propionate concentration (mM)
b 25 mM D-glucose, 7 days
Radiolablled TGF-1
Butyrate − 0.1 1.0 mM −
Propionate − − − 0.1
Figure 5 | Butyrate suppresses 25 mM D-glucose-stimulated
TGF-b1 synthesis. (a) Confluent monolayers of HK2 cells were
stimulated with 25 mM D-glucose either alone or in the presence of
butyrate (’) or propionate ( ) at the concentrations indicated for a
total of 7 days under serum-free conditions. In control experiments,
serum-free medium containing 5 mM D-glucose ( ) was added to
the cells. TGF-b1 concentration in the cell culture supernatant was
examined by ELISA. (b) Data represent mean7s.d.; n ¼ 6, *Po0.05
compared to 25 mM D-glucose alone. In parallel experiments, de novo
TGF-b1 protein synthesis was examined by stimulation of cells
with 25 mM D-glucose either alone or in the presence of butyrate
or priopionate in tissue culture medium to which 40 mCi of
3
H-radiolabelled amino-acid mixture was added. Following 7 days
of stimulation, TGF-b1 in the cell culture supernatant was
immunoprecipitated and visualized by autoradiography.
presence of butyrate, but was unaffected by the addition of
propionate (Figure 5b).
The involvement of the p38 and MAP kinase pathways in
25 mM D-glucose-stimulated TGF-b1 synthesis was deter-
mined by addition of glucose in the presence of either
SB203580 or PD98059. Stimulation of cells with 25 mM
D-glucose in the presence of PD98059 led to a dose-dependent
decrease in glucose-stimulated TGF-b1 concentration (Figure
6a). In contrast, inhibition of the p38 MAP kinase pathway
did not affect glucose-stimulated TGF-b1 generation (Figure
6b). D-glucose (25 mM)-stimulated activation of the ERK/
MAP kinase pathway was examined by immunoblot analysis
of ERK phosphorylation. A time-dependent increase in
p-ERK was demonstrated 10 min after addition of 25 mM
D-glucose, with maximal stimulation seen at 15 min (Figure 6c).
Addition of 25 mM D-glucose had no effect on the expression of
total ERK. The effect of butyrate on glucose-stimulated ERK
activation was examined by stimulation of HK-2 cells with
25 mM D-glucose in the presence of 1 mM butyrate for 15 min.
These experiments demonstrated abrogation of 25 mM
D-glucose activation of ERK such that there was no difference
in ERK phosphorylation following addition of 25 mM
260
N Matsumoto et al.: Gum arabic, short-chain fatty acids and TGF-b1
a b
1500
1500
*
TGF-1 (ng/ml)
TGF-1 (ng/ml)
1000 **
1000
500 500
0 0
D-glucose 5 25 25 25 25 mM D-glucose 5 25 25 25 25 mM
PD98059 0 0 0.8 8.0 80
M SB203580 0 0 30 300 3000 ng/ml
c d
Total ERK
0 5 10 15 30 180 25 mM D-glucose − − + + + +
Time (min) Butyrate − − − − + +
Figure 6 | Butyrate-mediated suppression of 25 mM
D-glucose-stimulated TGF-b1 is associated with suppression of
− ERK activations. (a) Increasing doses of either the MEK inhibitor
1.0 mM PD98059 or (b) p38 MAP kinase inhibitor SB203580 were added to
confluent monolayers of HK-2 cells stimulated with 25 mM D-glucose
under serum-free conditions for 24 h. Subsequently, cell culture
supernatant samples were collected for analysis of TGF-b1
concentration by ELISA. Data represent mean7s.d.; n ¼ 3. *Po0.05
compared to serum-free medium alone in the absence of inhibitor.
In parallel experiments, the effect of butyrate on 25 mM D-glucose-
stimulated ERK activation was examined by immunoblot analysis
using antibodies to either dually phosphorylated active form of MAPK
(p44/ERK1 and p42/ERK2) or total MAPK. (c) To demonstrate
activation of ERK, HK2 cells were stimulated with 25 mM D-glucose
under serum-free conditions for up to 180 min. At the time points
indicated, cells were lysed and ERK activation was assessed. (d) To
determine the effect of butyrate, HK-2 cells were stimulated with
25 mM D-glucose either alone or in the presence of 1 mM butyrate for
15 min before cell lysis and immunoblot analysis.
D-glucose in the presence of butyrate compared with that
seen in the control cell (Figure 6d).
Butyrate suppresses TGF-b1-dependent signalling. Activa-
tion of the Smad signalling pathway was examined using the
(SBE)4-Lux reporter, which contains four repeats of the
CAGACA sequence identified as a SMAD-binding element.
Addition of TGF-b1 led to a significant increase in luciferase
activity (Figure 7). Stimulation of HK-2 cells transfected with
the reporter construct with TGF-b1 in the presence of
butyrate blunted TGF-b1-dependent activation of Smad
signalling, as luciferase activity was significantly lower in
the presence of either 1 or 10 mM butyrate (Figure 7a).
Stimulation with TGF-b1 in the presence of propionate in
contrast did not affect TGF-b1-dependent activation of the
reporter construct (Figure 7b).
Activation of the Smad signalling pathway was also
examined by immnuoblot analysis of phosphorylated Smad
2/3 (Figure 8). Addition of TGF-b1 increased Smad phospho-
rylation. As with reporter construct activity, TGF-b1-
dependent phosphorylation of Smad was blunted in the
presence of butyrate in a dose-dependent manner, while
propionate did not influence Smad phosphorylation.
Kidney International (2006) 69, 257–265
N Matsumoto et al.: Gum arabic, short-chain fatty acids and TGF-b1 original article

a P = 0.003 a
P = 0.012
pSMAD
2
Fold increase in luciferase (CAGA) activity

Control TGF 1 ng/ml

Butyrate Propionate
1.5 1 10 30 nM 1 10 30 nM

b Raft Non-raft
over control

1 Fraction 2 3 4 5 6 7 8 9 10 11 12

Control

0.5
TGF-

0 TGF-+
0 1 10 butyrate
Butyrate (mM)
c 100
b 2
Fold increase in luciferase (CAGA) activity

Receptor distribution
(percentage of total)
75

1.5 Raft
50
Non-raft
over control

25
1

0
Control TGF- TGF-+
0.5 butyrate

Figure 8 | Suppression of Smad phosphorylation and distribution

0
0 1
Propionate (mM)
Figure 7 | Antagonism of TGF-b1-mediated increase of
SMAD-dependent signalling. HK-2 cells were transfected with a
SMAD-responsive promoter (SBE)4-Lux, before stimulation with
TGF-b1 (1 ng/ml) either alone or in the presence of either (a) butyrate
or (b) propionate. All stimuli were applied for 6 h, subsequently
luciferase content was quantified as described in Materials and
Methods and the results were normalized for transfection efficiency
(using b-galactosidase) expressed as the fold increase above the
non-stimulated control. Data represent mean7s.d.; n ¼ 6.
As previously demonstrated,9,10 addition of TGF-b1 and
stimulation of TGF-b1-dependent signalling increased traf-
ficking of receptors to non-lipid raft membrane-associated
fractions, as assessed by sucrose gradient density separation
of membrane fractions and detection of TGF-b RI by
immunoblot analysis (Figure 8b). Following TGF-b1, 69%
of TGFb RI was in the non-raft membrane fraction as
compared to only 39% in the control cell (Figure 8c).
Addition of TGF-b1 in the presence of butyrate decreased
TGF-b-mediated trafficking of TGF-b1 receptors into non-
lipid raft-associated membrane fractions, which are associated
with endosomal internalization and enhanced signalling.11
Under these conditions, only 38% of total TGF-b RI detected
was in the non-raft-associated fractions.
DISCUSSION
Recent changes in industrialized countries have led to dietary
patterns based on animal foodstuffs and refined cereals, with
of TGF-b receptors. (a) Confluent monolayers of HK2 cells were
stimulated with TGF-b1 (1 ng/ml) either alone or in the presence
10 of increasing doses of either butyrate or propionate as indicated.
Subsequently, 6 h after stimulation, cells were lysed and Smad
phosphorylation was assessed by immuoblot analysis using an
antibody. (b) TGF-b receptor distribution in lipid raft and non-raft
fractions was analysed in HK-2 cells following sucrose gradient
subcellular fractionation to separate lipid rafts from other cellular
components. Confluent monolayers were stimulated with TGF-b1
(1 ng/ml) either alone or in combination with 1 mM butyrate for 24 h
before cell lysis and fractionation as described in Materials and
Methods. In control experiments, cells were exposed to serum-free
medium alone. Membrane fractions were subsequently analysed by
TGF-b RI immunoblot analysis. (c) Following scanning densitometry
of autoradiographs, the distribution of TGF-b receptor into the raft
(fractions 3–6) and non-raft (fractions 7–12) was quantified and the
data from three separate experiments were expressed graphically
and expressed as the percentage of the receptor in the raft-and
non-raft-associated membrane fractions.
a reduced supply of dietary fibre. Until the last 30 years or so,
dietary fibre had long been considered a non-useful food
component. Nutrition scientists, however, have long agreed
that a quality diet consists of eating plenty of high-fibre
foods. Currently, this view is reinforced by numerous
epidemiological, clinical and cell biological observations.
The fibre hypothesis proposed by Burkitt and Trowel12,13
suggested, some three decades ago, that there was a link
between the consumption of a diet rich in fibre and the level
of protection against many of the ‘first world diseases’ such as
constipation, diverticulosis, cancer of the colon diabetes,
obesity, and cardiovascular disease.

Kidney International (2006) 69, 257–265 261

original article
Studies dating back over 20 years have shown that locust
bean gum, a non-digestible polymer of mannose and
galactose derived from the seeds of the Ceratonia siliqua
tree, was an efficient sorbent, which bound to many of the
potential toxic substances found in patients with chronic
renal failure.14 Subsequently, in a study of only two patients,
it was suggested that dietary supplementation with locust
bean gum improved blood pressure control and reduced
serum creatinine.15 These observations, which have not been
investigated further, if confirmed would suggest a mechanism
other than that related to its ‘sorbent’ properties by which
locust bean gum may influence progressive renal disease. In
addition, a single study using dietary supplementation with
the hemicellulosic ispaghula husk also suggested a beneficial
effect on renal function beyond increased faecal bacterial
loss.16
Our interest in the potential health benefit of gum arabic
stems from two studies demonstrating that supplementation
of the diet with gum arabic increases faecal nitrogen excretion
and lower serum urea nitrogen concentration in patients with
chronic renal failure,3 and work in rat models of acute renal
failure, which suggests that gum arabic may improve renal
function independently of its action on faecal bacterial
ammonia metabolism.4
The aim of this study was to examine the hypothesis that
dietary supplementation with gum arabic in humans will
increase systemic levels of the SCFA, which in turn may have
actions at the cellular level and a beneficial effect on the
development and progression of renal fibrosis.
The first significant observation is the demonstration of
elevated serum butyrate levels in human subjects who
received 25 g/day of specific gum arabic (Acacia(sen) SUPER-
TM
GUM ). Completion of the full 10-week protocol by all of
the subjects confirms the tolerability of such an approach.
Butyrate is considered a preferred fuel for colonocytes and is
believed to be selectively extracted by them,17 resulting in
relatively low butyrate concentrations in the portal blood.18
Furthermore, as the vast majority of butyrate is likely to be
extracted by the liver,19 alterations in systemic butyrate are
likely to be at low levels.18 Our data, however, are consistent
with other studies, which suggest that increased colonic
butyrate production in human subjects can be detected by
increased serum butyrate.20 It is interesting to note that in
contrast to the increase in butyrate levels, we were unable to
demonstrate any alteration in serum propionate in the
subjects given the test material. We postulate that this is
related to the fact that the majority of colonic propionate is
taken up by the liver19 where it inhibits the incorporation of
acetate into lipids21 and can be utilized as a gluconeogenic
substrate.22
Many studies have demonstrated that the SCFA butyrate
displays a host of biologically important effects in vitro,
including modulation of cell proliferation, apoptosis, regu-
lation of angiogenesis and inflammation.2
In order to provide a mechanistic basis to the hypothesis
that gum arabic supplementation of the diet may have
262
N Matsumoto et al.: Gum arabic, short-chain fatty acids and TGF-b1
beneficial effects in renal disease, we have examined the in
vitro effects of butyrate on the generation of the pro-fibrotic
cytokine TGF-b1 and also the responsiveness of cells to the
effects of TGF-b1 in vitro. There is clear evidence that this
cytokine is involved in the final common pathway of renal
fibrosis in all renal diseases, and antagonism of its action has
been proposed as a potential therapeutic goal.23–27 There is a
close correlation between rate of progression of renal disease
and the degree of interstitial fibrosis. Furthermore, it is well
established that the epithelial cells of the proximal tubule
contribute to this fibrotic process, both by generation of
cytokines such as TGF-b1 6,7,28 and by a process of
epithelial—mesenchymal transdifferentiation under the in-
fluence of TGF-b1, through which contribute to the
population of renal interstitial fibroblasts, which mediate
deposition of interstitial extracellular matrix in disease.29 We
have therefore focused on the modulation effect of butyrate
on TGF-b1 in this renal epithelial cell. The results clearly
demonstrate that at non-toxic doses, exposure to elevated
concentrations of butyrate prevents TGF-b1 generation.
More specifically, the data demonstrate that both basal/
constitutive TGF-b1 synthesis and glucose-stimulated gen-
eration of TGF-b1 are suppressed. Previously, we have
demonstrated the involvement of both the p38 MAP kinase
and ERK/MAP kinase pathways in the regulation of
transcriptional and translational regulation of TGF-b1,
respectively.8 In this study, the data suggest that butyrate-
mediated inhibition of TGF-b1 synthesis is related to the
specific inhibition of the ERK/MAP kinase pathway. This
coupled to our previous work would suggest that butyrate
may therefore specifically modulate the synthesis of TGF-b1
translation. It is of note that the effects of butyrate in the in
vitro situation were seen at concentrations far in excess of the
circulating serum levels seen in the in vivo feeding study. How
circulating levels relate to tissue levels, and in particular to
levels in the kidney and more specifically in the environment
of the proximal tubular epithelial cell, remains a matter for
speculation. The concentrations used, however, are in line
with previous studies in which the effects of butyrate on renal
cell function have been studied.30,31 These in vitro observa-
tions therefore merit further clinical studies to validate their
potential importance.
TGF-b1 receptors internalize into both caveolin- and
EEA-1-positive vesicles and reside in both lipid raft and non-
raft membrane domains.11 Clathrin-dependent internaliza-
tion into the EEA-1-positive endosome promotes TGF-b1
signalling. In contrast, the lipid raft-caveolar internalization
pathway contains Smad7-bound receptor and is required for
receptor turnover. Altered receptor trafficking between these
two membrane receptor pools may influence TGF-b1
signalling in a stimulus-specific manner resulting in either
enhancement10 or attenuation9 of TGF-b1 signalling. In this
study, we have demonstrated attenuation of TGF-b1 signal-
ling in the presence of butyrate, using a TGF-b1-responsive
promoter construct, and also phosphorylation of Smad
proteins as markers of TGF-b1 response. Furthermore,
Kidney International (2006) 69, 257–265
N Matsumoto et al.: Gum arabic, short-chain fatty acids and TGF-b1 original article

attenuation of TGF-b1 signalling was associated with
inhibition of trafficking of TGF-b RI in the cell membrane.
In conclusion, the data support the hypothesis that dietary
TM
supplementation with Acacia(sen) SUPERGUM (gum ara-
bic), by increasing systemic levels of butyrate, may have a
potential beneficial effect in renal disease by suppression of
TGF-b1 activity. Confirmation that the alterations in
systemic butyrate levels documented in this study may have
comparable in vivo effects to those demonstrated in this cell
culture system therefore warrants further in vivo investiga-
tion. TGF-b1 is a multifunctional cytokine, which controls a
plethora of biological responses, and alterations in its
signalling pathway are associated with a range of human
diseases. It is important in development and normal wound
healing through its effects on cell proliferation, migration,
differentiation and apoptosis.32 In addition, it is a critical
regulator of the normal inflammatory response, as illustrated
by the death of TGF-b1/ mice of a multifocal inflamma-
tory syndrome soon after weaning.33 It has growth inhibitory
effects, which explains its role as a tumor suppressor early in
the course of disease. Its expression by tumor cells, however,
also contributes to cancer progression and metastasis at later
stages of disease.34,35 TGF-b1 also plays a central role in the
pathological process of fibrosis, which may lead to organ
failure.36,37 Although the data presented in this paper have
Table 1 | Percentage loss on drying and specific rotation
of the matured gum arabic
% loss on Specific rotation
Sample name drying (%wt) (deg/dm cm3/g)
Matured gum arabic (SUPERGUMTM EM2) 6.25 31
been generated in a renal epithelial cell line, given this wider
role of TGF-b in health and disease, the general applicability
of our findings to other organ systems may have important
health implications.
MATERIALS AND METHODS
Effect of gum arabic on SCFA in vivo
Test material. The test gum Acacia(sen) SUPERGUMTMEM2
(SUPERGUMTM) has precisely structured molecular dimensions
and chemical composition as shown in Tables 1–6.
Protocol for dietary supplementation. Healthy volunteers
(n ¼ 10) each had their normal diet supplemented with 25 g
SUPERGUMTM daily. The daily amount of SUPERGUMTM was
provided in a dried form in individual sealed foil pouches. The
SUPERGUMTM was reconstituted in 250 ml of water and shaken to
ensure adequate mixing. The volunteers then added flavouring to
taste and consumed the SUPERGUMTM mixture.
Compliance was assessed by counting the foil pouches returned
to the department. Volunteers were also questioned about ill effects
from the fibre supplementation.
Quantitation of butyrate and propionate. Serum for mea-
surement of SCFA was obtained before and 10 weeks after
commencing dietary supplementation with SUPERGUMTM. Volun-
teers attended after an overnight fast, and a 12 ml blood sample was
obtained and placed into two 6 ml tubes containing serum clot
activator (Vacuette, Greiner labortechnik, Kremsmunter, Austria).
Once clotted, the serum was immediately separated by centrifuga-
tion for 7 min at 2000 r.p.m. and 41C. Aliquots of serum were stored
at 701C and batched for measurement of SCFA.
Serum butyrate and propionate were measured in quadruplicate
by gas chromatography as previously described.38 Statistical
significance was assessed by standard non-parametric tests (Wilcox-
on) using SPSS 11 statistical computer package.
Table 5 | Trace metal analysis (ICP) of the matured gum arabic
(lg/g) Supergum EM2

Table 2 | Molecular weight of matured gum arabic by K 7810

GPC-MALLS analysis Na 100
Ca 6580
Molecular Mg 2250
weight % Rg P 4
Sample name Processing (Mw, g/mol) Mass (nm) Fe 6
Zn 1
Matured gum arabic One peak 1.77  106 68 Cu 1
SUPERGUMTM EM2 Two peaks (1, AGP)a 7.84  106 18.2 75 Mn 2
Two peaks (2, AG+GP) 4.16  105 81.8 — Si 39
a
The method allows fractionation of the gum into its components as described by S 97
Al-Assaf et al.5

Table 3 | Sugar composition (mol%) of the matured gum

arabic Table 6 | Intrinsic viscosity of the matured gum arabic
Sample name Gal Ara Rha Uronic acid Sample name (code no.) Intrinsic viscosity (g) (ml/g)
TM
SUPERGUM EM2 32.8 31.4 13.1 24.3 Matured gum arabic (SUPERGUM EM2) 19.4

Table 4 | Amino-acid composition (mol%) of the matured gum arabic

Sample name %N Ala Arg Asp Cys Glu Gly His Hyp Ile Leu Lys Met Phe Pro Ser Thr Tyr Val
SUPERGUM EM2 0.325 2.8 1.0 5.4 – 3.6 5.7 5.4 26.2 1.4 8.3 2.6 0.1 3.5 7.4 13.8 8.1 0.7 4.0

Kidney International (2006) 69, 257–265 263

original article
In vitro effects of SCFA
Cell culture. HK-2 cells (human renal proximal tubular epithelial
cells immortalized by transduction with human papilloma virus (HPV
16 E6/E7 genes39) were cultured as previously described.40–43
Toxicity/cell viability. Cell viability was assessed by the ability
of monolayers to reduce the dye alamarBlue. At the end of each
experimental period/manipulation, alamarBlue reagent was added
to make up a final volume of 10% of the culture medium. The cells
were returned to the incubator for 1 h, after which 100 ml aliquots of
medium were quantified for alamarBlue fluorescence using a
Fluostar Optima Fluorescence Meter of excitation wavelength
544 nm and detection wavelength 590 nm.
Quantitation of TGF-b1. Total TGF-b1 in the cell culture
supernatant was measured by specific ELISA (R&D Systems,
Abingdon, UK) of cell culture supernatant samples collected from
growth-arrested HK-2 cells, stimulated under serum-free condi-
tions. Active TGF-b1 was measured directly and latent TGF-b1 can
be measured indirectly following acid activation of samples.
Assessment of de novo TGF-b1 synthesis. De novo TGF-b1
synthesis was examined by detection of radioactive amino-acid
incorporation into newly synthesized protein by metabolic labelling
and detection by TGF-b1 immunoprecipitation and autoradiogra-
phy as previously described.8
Western blot analysis. Equal numbers of cells were lysed by
addition of sodium dodecyl sulphate (SDS) sample buffer, and
100 ml of the resultant lysates was boiled for 5 min at 951C before
loading onto SDS-polyacrylamide electrophoresis (SDS-PAGE) gel
(5–15%). Electrophoresis was carried out under reducing condi-
tions.44 Separated proteins were transferred to a nitrocellulose
membrane (Amersham, Little Chalfont, UK). Following a blocking
step, membranes were incubated with the appropriate primary
antibody in Tris-buffered saline containing 0.1% Tween 20
(phosphate-buffered saline-Tween) overnight at 41C. Primary
antibodies were as follows:
K rabbit polyclonal anti-TGFb type I receptor antibody (dilution
1:500; Santa Cruz Biotechnology, Wiltshire, UK);
K antibody to dually phosphorylated active form of (mitogen-
activated protein kinase) MAPK (p44/ERK1 and p42/ERK2;
final dilution 1:5000; New England Biolabs (UK) Ltd, Hitchin
Herts, UK;
K antibody to total MAPK (dilution 1:5000; New England Biolabs
(UK) Ltd);
K antibody to phosphorylated SMAD3 final dilution 1:1000;
Cambridge Bioscience, Cambridge, UK).
The blots were subsequently incubated with an appropriate
horseradish peroxidase-conjugated secondary antibody (Sigma,
Poole, UK) in Tris-buffered saline-Tween. Proteins were visualized
using enhanced chemiluminescence (Amersham, UK).
Assessment of TGF-b1 responsiveness. The SMAD-responsive
promoter (SBE)4-Lux45 was a gift from Aristidis Moustakas. For
transfection of the reporter construct, 30  103 cells/well were
seeded onto a 24-well plate (cells at this density produced a 70%
confluence monolayer the following day). The next day, cells were
transfected with 0.2 mg of the SMAD-responsive promoter-luciferase
construct using the mixed lipofection reagent FuGene 6 (Roche, IN,
USA) at a ratio of 0.6 ml Fugene to 0.2 mg DNA in serum-free and
insulin-free medium. Transfection efficiency was monitored by co-
transfection with a b-galactosidase reporter plasmid. At 24 h after
transfection, cells were stimulated with TGF-b1. Following lysis of
264
N Matsumoto et al.: Gum arabic, short-chain fatty acids and TGF-b1
the cells in Reporter Lysis Buffer (Promega, WI, USA), the luciferase
content was quantified by glow-type luminescence assay (Bright-
Glo, Promega), and b-galactosidase activity determined by a
commercial assay (Promega). Luciferase activity was normalized to
b-galactosidase activity.
Detergent-free purification of the lipid raft-rich membrane
fraction. HK-2 cells were grown to near confluence in 100 mm
dishes. After two washes with ice-cold phosphate-buffered saline,
two confluent dishes were scraped into 2 ml of 500 mM sodium
carbonate, pH 11.0. Homogenization was carried out by three 20 s
bursts of ultrasonic disintegrator (Soniprep 150, Fisher Scientific) to
disrupt cellular membranes as previously described.46 The homo-
genates were adjusted to 45% sucrose by addition of 2 ml of 90%
sucrose prepared in MBS (2-(N-morpholino) ethanesulphonic acid-
buffered saline;, 25 mM 2-(N-morpholino) ethanesulfonic acid, pH
6.5, 0.15 M NaCl) and placed at the bottom of an ultracentrifuge
tube. A discontinuous sucrose gradient (4 ml of 35% sucrose, 4 ml of
5% sucrose, both prepared in MBS) was formed above and
centrifuged at 39 000 r.p.m. for 16–20 h, in an SW40 TI rotor
(Beckman Instruments, Palo Alto, CA, USA). A light-scattering
band was observed at the 3–35% sucrose interface. A total of 12 1-ml
fractions were collected from the top of the tubes, and a portion of
each fraction was analysed by SDS-PAGE (10% gel) and immuno-
blot analysis for TGF-b RI, which was quantified by densitometry
(Chemi Doc, Bio-Rad, Hemel Hempstead Herts, UK). Data are
expressed as a percentage of the total TGF-b:receptor complex for
that experiment in each fraction.
Statistical analysis
Statistical analysis was performed using the unpaired Student’s
t-test, with a value of Po0.05 considered to represent a significant
difference. The data are presented as means7s.d. of n experiments
as indicated in the figure legends. For each individual experiment,
the mean of duplicate determinations was calculated.
ACKNOWLEDGMENTS
This work was supported by a research grant from San-Ei Gen Inc.,
Japan. AOP is supported by a GlaxoSmithKline Advanced Fellowship.
The Institute of Nephrology is supported by Kidney Wales
Foundation.
REFERENCES
1. Slavin J. Why whole grains are protective: biological mechanisms.
Proc Nutr Soc 2003; 62: 129–134.
2. Smith JG, Yokoyama WH, German JB. Butyric acid from the diet: actions
at the level of gene expression. Crit Rev Food Sci Nutr 1998; 38:
259–297.
3. Bliss DZ, Stein TP, Schleifer CR et al. Supplementation with gum arabic
fiber increases fecal nitrogen excretion and lowers serum urea nitrogen
concentration in chronic renal failure patients consuming a low-protein
diet. Am J Clin Nutr 1996; 63: 392–398.
4. Ali BH, Al-Qarawi AA, Haroun EM et al. The effect of treatment with gum
arabic on gentamicin nephrotoxicity in rats: a preliminary study. Renal
Failure 2003; 25: 15–20.
5. Al-Assaf S, Phillips GO, Williams PA. Studies on acacia exudate gums, Part
I: the molecular weight of acacia senegal gum exudate. Food
Hydrocolloids 2005; 10: 647–660.
6. Phillips AO, Steadman R, Topley N et al. Elevated D-glucose
concentrations modulate TGF-b1 synthesis by human cultured renal
proximal tubular cells: the permissive role of platelet derived growth
factor. Am J Pathol 1995; 147: 362–374.
7. Phillips AO, Topley N, Steadman R et al. Induction of TGF-b1 synthesis in
D-glucose primed human proximal tubular cells: differential stimulation
by the macrophage derived pro-inflammatory cytokines IL-1b and TNFa.
Kidney Int 1996; 50: 1546–1554.
Kidney International (2006) 69, 257–265
N Matsumoto et al.: Gum arabic, short-chain fatty acids and TGF-b1
8. Fraser DJ, Brunskill NJ, Phillips AO. Long-term exposure of proximal
tubular epithelial cells to glucose induces TGF-b1 synthesis via an
autocrine PDGF loop. Am J Pathol 2003; 163: 2565–2574.
9. Ito T, Williams JD, Fraser DJ et al. Hyaluronan regulates TGF-b1 receptor
compartmentalisation. J Biol Chem 2004; 279: 25326–25332.
10. Zhang XL, Topley N, Ito T et al. IL-6 regulation of TGF-b receptor
compartmentalisation and turnover enhances TGF-b1 signalling. J Biol
Chem 2005; 280: 12239–12245.
11. Di Guglielmo GM, Le Roy C, Goodfellow AF et al. Distinct endocytic
pathways regulate TGF-beta receptor signalling and turnover. Nat Cell
Biol 2003; 5: 410–421.
12. Burkitt DP, Trowell HC. Nutritional intake, adiposity, and diabetes.
BMJ 1979; 1: 1083–1084.
13. Burkitt DP, Trowell HC. Dietary fibre and western diseases. Irish Med J
1977; 70: 272–277.
14. Yatzidis H. Preliminary studies with locust bean gum, a new sorbent with
great potential. Kidney Int 1977; 12: 152A.
15. Yatzidis H, Koutsicos D, Digenis P. Newer oral sorbents in uremia. Clin
Nephrol 1979; 11: 105–106.
16. Rampton DS, Cohen SL, Crammond VB et al. Treatment of chronic renal
failure with dietary fiber. Clin Nephrol 1984; 21: 159–163.
17. Roediger WE. Role of anaerobic bacteria in the metabolic welfare of the
colonic mucosa in man. Gut 1980; 21: 793–798.
18. Cummings JH, Pomare EW, Branch WJ et al. Short chain fatty acids in
human large intestine, portal, hepatic and venous blood. Gut 1987; 28:
1221–1227.
19. Remesy C, Demigne C, Chartier F. Origin and utilization of volatile fatty
acids in the rat. Reprod Nutr Dev 1980; 20: 1339–1349.
20. Wolever TM, Chiasson JL. Acarbose raises serum butyrate in human
subjects with impaired glucose tolerance. Br J Nutr 2000; 84: 57–61.
21. Wolever TM, Spadafora PJ, Cunnane SC et al. Propionate inhibits
incorporation of colonic [1,2-13C]acetate into plasma lipids in humans.
Am J Clin Nutr 1995; 61: 1241–1247.
22. Wolever TM, Spadafora P, Eshuis H. Interaction between colonic acetate
and propionate in humans. Am J Clin Nutr 1991; 53: 681–687.
23. Akagi Y, Isaka Y, Arai M et al. Inhibition of TGF-b1 expression by antisense
oligonucleotides suppressed extracellular matrix accumulation in
experimental glomerulonephritis. Kidney Int 1996; 50: 148–155.
24. Peters H, Border WA, Noble NA. Targeting TGF-b1 overexpression in renal
disease: maximizing the antifibrotic action of angiotenin II blockade.
Kidney Int 1998; 54: 1570–1580.
25. Cheol DC, Hoffman BB, Hong SW et al. Therapy with antisense TGF-b1
oligodeoxynucleotides reduces kidney weight and matrix mRNAs in
diabetic mice. Am J Physiol 2000; 278: F628–F634.
26. Isaka Y, Tsujie M, Ando Y et al. Transforming growth factor b1 antisense
oligodeoxynucleotides block interstitial fibrosis in unilateral ureteral
obstruction. Kidney Int 2000; 58: 1885–1892.
27. Isalm M, Burke JF, McGowan TA et al. Effect of transfroming growth
factor-b antibodies in cyclosporine-induced renal dysfunction. Kidney Int
2001; 59: 498–506.
28. Phillips AO, Topley N, Morrisey K et al. Basic fibroblast growth factor
stimulates the release of pre-formed TGF-b1 from human proximal
tubular cells in the absence of de-novo gene transcription or mRNA
translation. Lab Invest 1997; 76: 591–600.
Kidney International (2006) 69, 257–265
original article
29. Zeisberg M, Hanai J, Sugimoto H et al. BMP-7 counteracts
TGF-beta1-induced epithelial-to-mesenchymal transition and reverses
chronic renal injury. Nat Med 2003; 9: 964–968.
30. Hara I, Miyake H, Hara S et al. Sodium butyrate induces apoptosis in
human renal cell carcinoma cells and synergistically enhances their
sensitivity to anti-Fas-mediated cytotoxicity. Int J Oncol 2000; 17:
1213–1218.
31. Moyer BD, Loffing-Cueni D, Loffing J et al. Butyrate increases apical
membrane CFTR but reduces chloride secretion in MDCK cells.
Am J Physiol 1999; 277: F271–F276.
32. Roberts AB, Flanders KC, Kondaiah P et al. Transforming growth factor b:
biochemistry and roles in embryogenesis, tissue repair and remodeling,
and carcinogenesis. Rec Progr Horm Res 1988; 44: 157–197.
33. Kulkarni AB, Karlsson S. Transforming growth factor beta1 knockout mice:
a mutation in one cytokine gene causes a dramatic inflammatory disease.
Am J Pathol 1993; 143: 3–9.
34. Tang B, Vu M, Booker T et al. TGF-beta switches from tumor suppressor to
prometastatic factor in a model of breast cancer progression. J Clin Invest
2003; 112: 1116–1124.
35. Tang B, Böttinger EP, Jakowlew SB et al. Transforming growth factor b1 is
a new form of tumor suppressor with true haploid insufficiency. Nat Med
1998; 4: 802–807.
36. Border WA, Ruoslahti E. Transforming growth factor b in disease: the dark
side of tissue repair. J Clin Investig 1992; 90: 1–7.
37. Border WA, Noble NA. Transforming growth factor beta in tissue fibrosis.
N Engl J Med 1994; 331: 1286–1292.
38. Wolever TM, Fernandes J, Rao AV. Serum acetate:propionate ratio is
related to serum cholesterol in men but not women. J Nutr 1996; 126:
2790–2797.
39. Ryan MJ, Johnson G, Kirk J et al. HK-2: An immortalized proximal tubule
epithelial cell line from normal adult human kidney. Kidney Int 1994; 45:
48–57.
40. Fraser DJ, Wakefield L, Phillips AO. Independent regulation of
transforming growth factor-\{beta\}1 transcription and translati.
Am J Pathol 2002; 161: 1039–1049.
41. Janssen U, Thomas G, Glant T et al. Regulation of inter-alpha-trypsin
inhibitor (IaI) and tumour necrosis factor-stimulated gene 6 (TSG-6)
expression in human renal proximal tubular epithelial cells. Kidney Int
2001; 60: 126–136.
42. Jones SG, Jones S, Phillips AO. Regulation of renal proximal tubular
epithelial cell hyaluronan generation: implications for diabetic
nephropathy. Kidney Int 2001; 59: 1739–1749.
43. Morrisey K, Evans RA, Wakefield L et al. Translational regulation of renal
proximal tubular epithelial cell transforming growth factorb1 generation
by insulin. Am J Pathol 2001; 159: 1905–1915.
44. Laemmli UK. Cleavage of structural proteins during the assembly the
head of bacteriophage T4. Nature 1970; 227: 680–685.
45. Piek E, Moustakas A, Kurisaki A et al. TGF-(beta) type I receptor/ALK-5 and
Smad proteins mediate epithelial to mesenchymal transdifferentiation in
NMuMG breast epithelial cells. J Cell Sci 1999; 112: 4557–4568.
46. Song KS, Li S, Okamoto T et al. Co-purification and direct interaction
of Ras with caveolin, an integral membrane protein of caveolae
microdomains. Detergent-free purification of caveolae microdomains.
J Biol Chem 1996; 271: 9690–9697.
265