See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/26663524

Dehydroepiandrosterone sulphate improves cholestasis-associated fatigue in

bile duct ligated rats

Article  in  Neurogastroenterology and Motility · August 2009

DOI: 10.1111/j.1365-2982.2009.01356.x · Source: PubMed

CITATIONS READS

17 162

6 authors, including:

Roger Butterworth Robert Lalonde

Université de Montréal Université de Rouen
480 PUBLICATIONS   21,698 CITATIONS    331 PUBLICATIONS   10,445 CITATIONS   

SEE PROFILE SEE PROFILE

Halima Gamrani Samir Ahboucha

Cadi Ayyad University Université Hassan 1er
105 PUBLICATIONS   1,572 CITATIONS    45 PUBLICATIONS   1,215 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Nasal inoculation of an adenovirus vector encoding 11 tandem repeats of Aβ1-6 upregulates IL-10 expression and reduces amyloid load in a Mo/Hu APPswe PS1dE9
mouse model of Alzheimer disease View project

personal research View project

All content following this page was uploaded by Robert Lalonde on 23 June 2021.

The user has requested enhancement of the downloaded file.

Neurogastroenterol Motil (2009) 21, 1319–1325 doi: 10.1111/j.1365-2982.2009.01356.x

Dehydroepiandrosterone sulphate improves

cholestasis-associated fatigue in bile duct ligated rats
R. F. BUTTERWORTH ,* R. LALONDE ,  C. POWER ,à G. B. BAKER ,§ H. GAMRANI – & S. AHBOUCHA *,§,–

*Neuroscience Research Unit, Hôpital Saint-Luc, CHUM, Montreal, QC, Canada

 Behavioural Neurobiology Unit, Pavillon Hôtel-Dieu, CHUM, Montreal, QC, Canada
àDepartment of Medicine, Medical Microbiology & Immunology Neuro-AIDS Clinics, University of Alberta, Edmonton,
AB, Canada
§Neurochemical Research Unit, Department of Psychiatry, University of Alberta, Edmonton, AB, Canada
–Equipe Neurosciences et Pharmacologie, Faculté des Sciences Semlalia, Département de Biologie, Université Cadi Ayyad,
Marrakech, Maroc

Abstract Fatigue is a common debilitating symptom cantly and highly correlated with those of plasma
in patients with primary biliary cirrhosis (PBC). The DHEAS and brain dehydroepiandrosterone (DHEA).
mechanism of fatigue is still poorly understood. Substitutive therapies with DHEAS or DHEA could
However, it has been reported that levels of the steroid represent novel approaches in the management
dehydroepiandrosterone sulphate (DHEAS) are of fatigue due to cholestasis-induced liver failure.
reduced in plasma of patients with PBC, and substi-
Keywords bile duct ligation, cholestasis, cirrhosis,
tutive therapy has been suggested to improve fatigue
dehydroepiandrosterone sulphate, fatigue, neuroactive
symptoms experienced during the course of this dis-
steroids.
ease. In this study, we tested the effect of DHEAS on
whole body fatigue in rats following bile duct ligation
(BDL). Fatigue was estimated by the time spent on an INTRODUCTION
electrified grid as a result of falling off a treadmill and
Fatigue is a common complication of primary biliary
by performance of rats on an infrared beam monitor
cirrhosis (PBC) which impacts considerably on quality
which allows the assessment of travelled distance and
of life of affected patients.1,2 The lack of correlates
stereotypic movement activities. On day 5 after BDL
between fatigue and traditional markers of liver disease
surgery, cholestatic rats exhibited increased whole
favour the concept that fatigue in PBC is central in
body fatigue as reflected by significantly increased
origin.3 Thus, specific changes within the brain are
time spent on the electrified grid, reduced travelled
likely to reflect fatigue severity in this disorder. The
distance and reduced stereotypic movements.
precise pathophysiologic mechanisms responsible for
Administration of 5 mg kg)1 of DHEAS to BDL rats for
fatigue in PBC have not been completely established,
three consecutive days significantly normalized their
and abnormal neuroendocrine function of the hypo-
behaviour. Fatigue scores were also found to be
thalamic–pituitary–adrenal axis, alterations of seroto-
reduced in cirrhotic rats 4 weeks after BDL surgery,
ninergic neurotransmission, and increased sensitivity
and DHEAS treatment for 3 days reduced fatigue
to pro-inflammatory cytokines have been proposed,3
scores at this stage. Dehydroepiandrosterone sulphate
although the relevance of these mechanisms in
treatment was sufficient to increase brain levels of
humans is not yet established.
DHEAS in the BDL rats in a manner that is signifi-
We recently reported that circulating levels of the
neuroactive steroid dehydroepiandrosterone sulphate
Address for correspondence (DHEAS) were significantly correlated with the degree
Samir Ahboucha, PhD, Equipe Neurosciences et Pharmaco- of fatigue in PBC patients,4 suggesting that substitutive
logie, Faculté des Sciences Semlalia, Université Cadi Ayyad, therapy with DHEAS or its precursor may be an alterna-
Avenue Prince My Abdellah, B.P. 2390 Marrakech, Maroc. tive option to treat fatigue in these patients. In humans,
Tel: +212 524434649 post 528; fax: +212 524437412;
e-mail: sahboucha@hotmail.com
DHEAS is an abundant steroid in the circulation and is
Received: 7 March 2009 generated mainly in the adrenals and liver from dehy-
Accepted for publication: 30 May 2009 droepiandrosterone (DHEA) by the enzyme DHEA

 2009 Blackwell Publishing Ltd 1319

R. F. Butterworth et al.
sulphotransferase (DHEA ST).5 Dehydroepiandrosterone
sulphate is also synthesized in the brain independently
from peripheral sources, and is consequently known as a
neurosteroid,6,7 although recent reports suggest that
brain DHEAS is undetectable in rodent, suggesting that
DHEAS levels were overestimated in rodent brains
using indirect methods.8 Many functions have been
attributed to DHEA/S, being the precursors of androgen
and oestrogen steroids, and having antiageing, anti-
inflammatory, immunomodulatory, antiatherogenenic,
anticancer and neurotrophic effects.9–12 Dehydroepiand-
rosterone sulphate is also associated with modulation of
brain function. For example, increased levels of DHEA/S
in animals and humans elicit memory enhancement,13
increased neuronal excitability,14 antidepressant ef-
fects,15 together with decreased depression rating scores
and cognition enhancement.16,17 Therefore, alterations
of brain levels of DHEA/S would be expected to influence
neuronal function, including effects on cognition, mem-
ory and mood.
Fatigue is a subjective symptom that makes its evalu-
ation a considerable challenge. As with any behavioural
score, measurements of whole body fatigue are an amal-
gam of several underlying factors. In rats, time spent
running on a treadmill18 and time spent floating in a water
tank19 have been employed. Although muscle or mental
fatigue is considered to be the main component of these
measures, other factors may be involved, such as senso-
rimotor coordination and depression. In this study, we
evaluate whole body fatigue as time spent on an electrified
grid after falling off a treadmill, and as travelled distance
and stereotypic activity measured on an infrared beam
monitor 5 days after bile duct ligation (BDL) where rats are
cholestatic, and 4 weeks after BDL where rats develop
cirrhosis. We also evaluate the effect of intraperitoneal
administration of DHEAS for 3 days on whole body fatigue
in BDL rats. The effect of DHEAS treatment on brain levels
of DHEAS or DHEA was also assessed using, respectively,
direct methods of liquid chromatography/mass spectrom-
etry and gas chromatography/mass spectrometry.
ANIMALS AND METHODS
Animals
Male Sprague–Dawley rats weighing 175–200 g were obtained from
Charles River (Pointe Claire, QC, Canada). Bile duct ligation was
performed as described by Cameron and Oakley20 with minor
modifications. Briefly, laparotomy was performed under halothane
anaesthesia and the bile duct was isolated, doubly ligated and
resected between the ligatures. The sham-operated group (sham)
consisted of laparotomy and bile duct identification and manipu-
lation without ligation or resection. Experiments were performed
5 days after surgery where animals are cholestatic. Another set of
1320
Neurogastroenterology and Motility
sham and BDL rats was kept for 4 weeks after surgery at which time
rats were cirrhotic. Liver function was confirmed by pathological
and biochemical analysis. Animals were killed and plasma, brain
and liver were sampled for biochemical and histological assays. All
animals were treated in compliance with the guidelines of the
University of Montreal (CHUM).
Serum liver biochemistry
Serum alanine aminotransferase (ALT), aspartate aminotransfer-
ase (AST) and total bilirubin levels were determined using
commercially available kits (Sigma Chemical, St Louis, MO,
USA).
Drug treatment
Dehydroepiandrosterone sulphate was purchased from Sigma
(Oakville, ON, Canada). A single injection at 10:00 a.m. of
5 mg kg)1 of DHEAS or saline was performed intraperitoneally for
three consecutive days. Animals were killed on day 3 immediately
after assessment of fatigue.
Assessment of fatigue
Infrared beam activity monitor Whole body fatigue was assessed
in BDL rats 1 h after the third injection using an infrared beam
activity monitor equipped with a computerized auto-track system
(Columbus Instruments, Columbus, OH, USA) as previously
described.21 Animals were individually placed for 20 min in
Plexiglas cages (29 · 22 · 22 cm) and cumulative distances trav-
elled (cm) were recorded during the last 15 min. As the rat moves
within the cage, the invisible infrared beams are broken and the
machine records ambulatory count when two or more beams are
broken as a measure of locomotor activity (travelled distance).
The machine simultaneously records the number of non-simul-
taneous broken beams and generates a stereotypic count when
only one beam is broken as a measure of stereotypic activity such
as grooming and exploring (stereotypic time). The animals were
exposed to the open field on the day before the experiment for 1 h
and tested on the following day.
Treadmill After the open field test, animal performance was
immediately assessed in a treadmill on the same animals as
described previously.22 Fatigue was estimated with a treadmill
(LETICA, S.A.-Scientific Instruments, LI 8706, Barcelona, Spain)
which consists of a motor-driven conveyor belt divided into two
parallel sectors, each with an electrified grid, enabling the testing of
two rats at the same time. At room temperature, the animals were
placed on the stationary belt inclined at 5 with gradually increasing
speed of up to 20 cm s)1 during a 5-min segment and continued at
that speed for a further 15 min (20-min trial overall). To avoid
footshocks (0.5 mV), the animals had to locomote forward without
falling off the belt. Otherwise, a sensor accumulated time spent on
the grid; the value recorded during the final 15 min of each trial was
noted. The time spent on the grid after the rats fall off the running
belt is referred to as fatigue score. The animals were first pretrained
on a single day and then tested on the following day.
Measurement of dehydroepiandrosterone by gas
chromatography/mass spectrometry
Brain samples (200 mg) were homogenized in methanol (75% in
H2O) using a small electric pellet pestle motor (Kontes, Vernon
 2009 Blackwell Publishing Ltd
Volume 21, Number 12, December 2009
Hills, IL, USA) on ice and then centrifuged (6000 g for 10 min).
Supernatants were then diluted with deionized water to a final
concentration of 5% methanol. Samples were extracted with C18
columns previously equilibrated with pure methanol and water
successively. Samples were passed through the columns under
vacuum and the eluate was discarded. The columns were then
washed with 50% methanol/H2O. The steroid fraction was eluted
with pure methanol and evaporated to dryness at 40 C in a speed-
vacuum system. The samples were then derivatized with hepta-
fluorobutyrylimidazole (HFBI) (Sigma-Aldrich, Oakville, ON,
Canada) and the resultant derivatives analysed by gas chromatog-
raphy/mass spectrometry (an Agilent 6890 GC coupled to a 5973N
mass selective detector; Agilent Technologies, Missisauga, ON,
Canada). Deuterated (d4) pregnenolone was carried throughout the
entire procedure as internal standard. Ions selected for monitoring
were m/z 464 for DHEA and m/z 495 for d4-pregnenolone which
had a retention time of 22.27 and 23.9 min respectively.
Measurement of dehydroepiandrosterone
sulphate by liquid chromatography/mass
spectrometry
Brain samples (200 mg) were homogenized in potassium phos-
phate buffer (5 mmol L)1, pH 7) using a small electric pellet pestle
motor (Kontes). Dehydroepiandrosterone sulphate was extracted
in 40% methanol and centrifuged for 15 min at 8000 g at 4 C.
Supernatants were then passed through micro-elution cartridges
previously conditioned with 100% methanol and water respec-
tively. After washing with 5% methanol, DHEAS was eluted with
60% acetonitrile in methanol. The samples were then analysed by
liquid chromatography/mass spectrometry (Waters Micromass
ZQ 2000 LC Mass Spectrometer, Milford, MA, USA). Deuterated
(d4) pregnenolone sulphate was carried throughout the entire
procedure as internal standard. Ions selected for monitoring were
m/z 366.69 for DHEAS and m/z 398.97 for d4-pregnenolone
sulphate which both had retention time around 1.52 min.
Data analysis
Data expressed as means ± SEM were compared using the ANOVA
test followed by a post hoc Tukey test. P-values < 0.05 were
considered significant.
RESULTS
Bile duct ligation rats showed a significant reduction of
their body weight 5 days after surgery, and they mani-
fested a significant increase of the liver weight and the
liver/body weight ratio (Table 1). At this stage, BDL rats
were cholestatic with dark urine, elevated bilirubin
(BDL 160.21 ± 22.31 lmol L)1 vs sham 10.22 ± 2.14
lmol L)1; P < 0.001, n = 5 per group), and elevated
alanine transaminase (BDL 120.55 ± 21.14 IU L)1,
n = 6; sham 72.10 ± 6.16 IU L)1, n = 5; P < 0.001). On
the treadmill, cholestatic rats exhibited increased elec-
trified grid contact (BDL 10.30 ± 2.33 s, n = 6; sham
2.10 ± 0.78 s, n = 5; P < 0.003) (Fig. 1).
In addition to the use of the treadmill, we also
evaluated fatigue using an infrared beam activity monitor
 2009 Blackwell Publishing Ltd
Brain DHEAS and fatigue in BDL rats
Table 1 Animal characteristics
Sham-vehicle BDL-vehicle BDL-DHEAS
5 days after BDL
Body weight (BW) (g) 314.0 ± 8.81 260.20 ± 6.09* 276.70 ± 3.90*
Liver weight (LW) (g) 11.86 ± 0.45 14.68 ± 0.96* 14.40 ± 0.47*
LW/BW ratio (%) 3.78 ± 0.06 5.65 ± 0.37* 5.21 ± 0.15*
4 weeks after BDL
Body weight (BW) (g) 365.30 ± 7.13 319.32 ± 14.96 311.85 ± 14.10
Liver weight (LW) (g) 13.74 ± 0.26 26.43 ± 3.39* 20.85 ± 1.13
LW/BW ratio (%) 3.76 ± 0.08 8.33 ± 1.08* 6.72 ± 0.45
Values are expressed as means ± SEM. Significant differences from
controls indicated by *P < 0.05 vs Sham group; ANOVA test followed by
post hoc Tukey test.
using a computerized auto-track system previously used
to measure fatigue in cholestatic rats.22 In an open field,
cholestatic rats showed significantly reduced travelled
distance (BDL 4.29 ± 1.38 m, n = 6; sham 10.04 ±
0.96 m, n = 5; P < 0.005; Fig. 2A) and time spent in
performing stereotypic movements used during beha-
viours such as grooming and exploring (BDL 94.22 ±
14.39 s, n = 6; sham 154.70 ± 19.11 s, n = 5; P < 0.02;
Fig. 2B). Dehydroepiandrosterone sulphate administra-
tion for three consecutive days to cholestatic animals
was able to restore electrified grid contact time
(1.93 ± 0.89, n = 6; Fig. 1), travelled distance (9.53 ±
1.51, n = 6; Fig. 2A) and time spent in stereotypic
movements (153.80 ± 15.59, n = 6; Fig. 2B) to levels
comparable with those of saline-treated sham controls.
Dehydroepiandrosterone sulphate did not have any sig-
nificant effect on electrified grid contact time
(1.89 ± 0.66 s), travelled distance (11.21 ± 1.34 m), or
time spent in stereotypic movements (161.72 ± 14.54 s)
in sham-operated animals.
Four weeks after BDL surgery, rats showed a reduc-
tion of their body weight which did not reach signifi-
Figure 1 Effect of DHEAS on treadmill performance. Fatigue scores are
shown in sham vehicle-treated rats (n = 5, white panel) and in chole-
static rats 5 days after BDL followed by vehicle (n = 5, black panel) or
DHEAS (n = 6 grey panel) treatments. Fatigue scores are expressed as
the grid contact time. *P < 0.001 vs sham, #P < 0.01 vs vehicle-treated
BDL (ANOVA test followed by post hoc Tukey test).
1321
R. F. Butterworth et al.
Figure 2 Effect of DHEAS on locomotor performance in an open field
test. Travelled distance (A), and time spent in stereotypic movements
(B) are shown in sham vehicle-treated rats (n = 5, white panel) and in
cholestatic rats 5 days after BDL followed by vehicle (n = 5, black
panel) or DHEAS (n = 6 grey panel) treatments. *P < 0.01 vs sham,
#
P < 0.05 vs vehicle-treated BDL (ANOVA test followed by post hoc
Tukey test).
cance, but had a significant increase of the liver weight
and the liver/body weight ratio (Table 1). At this stage,
BDL rats had elevated bilirubin (BDL 141.10 ±
27.46 lmol L)1; sham 11.33 ± 3.16 lmol L)1; P < 0.001,
n = 5 per group) and elevated plasma AST (BDL
500.00 ± 58.46 IU L)1; sham 223.30 ± 9.26 IU L)1;
P < 0.001, n = 5 per group) and ALT (BDL 93.75 ±
8.60 IU L)1; sham 76.0 ± 7.02 IU L)1; P < 0.01, n = 5
per group).
Similarly to cholestatic rats, BDL rats 4 weeks after
BDL surgery exhibited increased electrified grid contact
time (BDL-vehicle 10.12 ± 0.85 s; sham-vehicle
0.82 ± 0.32 s; P < 0.001; Fig. 3). Administration of
5 mg kg)1 of DHEAS to BDL rats for 3 days diminished
grid contact time (1.91 ± 0.57 s; P < 0.001 vs BDL-
vehicle group) (Fig. 3). Dehydroepiandrosterone sul-
phate treatment was also found to induce a significant
increase of the distance travelled (BDL-vehicle
2.25 ± 0.25 m; sham-vehicle 4.36 ± 0.46 m, BDL-
DHEAS 4.04 ± 0.26 m; P £ 0.01 BDL-DHEAS vs BDL-
vehicle, ANOVA test) and stereotypic movements (BDL-
vehicle 56.60 ± 4.13 s; sham-vehicle 76.56 ± 3.46 s,
BDL-DHEAS 71.80 ± 3.15 s; P £ 0.05 BDL-DHEAS vs
BDL-vehicle, ANOVA test) in BDL rats 4 weeks after
surgery.
1322
Neurogastroenterology and Motility
Figure 3 Effect of DHEAS on treadmill performance: Fatigue scores are
shown in sham vehicle-treated rats (n = 5, white panel) and in rats
4 weeks after BDL followed by vehicle (n = 5, black panel) or DHEAS
(n = 6 grey panel) treatments. Fatigue scores are expressed as the grid
contact time in seconds. *P < 0.001 vs sham, #P < 0.001 vs vehicle-
treated BDL (ANOVA test followed by post hoc Tukey test).
Administration of similar amounts of DHEAS to
sham-operated animals (n = 4) did not significantly
affect grid contact time (0.89 ± 0.36 s), distance trav-
elled (4.50 ± 0.65 m), or time spent in stereotypic
movements (81.50 ± 4.48 s) relative to saline-treated
sham animals.
Basal levels of DHEAS were only detected in two out
of five plasma and brain samples of saline-treated BDL
rats, while DHEAS was under limit of detection in all
brain and plasma sample from sham-operated controls.
As expected, DHEAS treatment significantly increased
plasma levels of DHEAS in all BDL rats (Fig. 4A).
Intraperitoneal treatment with DHEAS significantly
increased brain levels of this neuroactive steroid
(Fig. 4B) in a manner that is highly and significantly
correlated with those found in plasma (r2 = 0.978,
P < 0.002; Fig. 4C), a finding which supports the notion
that this neuroactive steroid readily crosses the blood/
brain barrier. Dehydroepiandrosterone sulphate treat-
ment of BDL rats significantly increased the brain
levels of its immediate metabolite DHEA (Fig. 5A) in a
manner that is highly correlated with those of brain
DHEAS (r2 = 0.923, P < 0.01; Fig. 5B), suggesting that
part of DHEAS could be converted to DHEA.
DISCUSSION
This study demonstrates poorer performance on a tread-
mill, and reduced travelled distance and stereotypic
activity in BDL compared with sham-operated rats.
These findings are concordant with previous data indi-
cating low locomotor performances of BDL rats in
swimming19 and open field21 tests. Our findings show
that locomotor abnormalities between BDL and sham-
operated rats are detected using both the treadmill test
 2009 Blackwell Publishing Ltd
Volume 21, Number 12, December 2009
A A
B B
Figure 4 Plasma (A) and brain (B) levels of DHEAS in sham vehicle-
treated rats (n = 5, white panel) and in cholestatic rats 5 days after BDL
following vehicle (n = 5, grey panel) or DHEAS (n = 5, black panel)
treatments. *P < 0.01 vs sham, #P < 0.01 vs vehicle-treated BDL (ANOVA
test followed by post hoc Tukey test). (C) Plasma and brain levels of
DHEAS in the BDL-DHEAS-treated rats are highly correlated.
r = Pearson coefficient.
and the open field test, suggesting that the treadmill may
be used as a tool to evaluate fatigue in experimental
models of chronic liver disease. Based on our data, the
treadmill allows a better discrimination between BDL
and sham-operated groups. The high fatigue scores
obtained in the treadmill test could be related to the
ability of the test to simultaneously evaluate different
components of fatigue. For example, the open field test
certainly allows the evaluation of a physical (travelled
distance) component of fatigue. However, cognitive
components of fatigue due, for example, to sensorimotor
 2009 Blackwell Publishing Ltd
Brain DHEAS and fatigue in BDL rats
Figure 5 (A) Brain levels of DHEA in sham vehicle-treated rats (n = 5,
white panel) and in cholestatic rats 5 days after BDL following vehicle
(n = 5, grey panel) or DHEAS (n = 5 black panel) treatments. *P < 0.001
vs sham and vehicle-treated BDL (ANOVA test followed by post hoc
Tukey test). (B) Brain levels of DHEA in the BDL-DHEAS-treated rats
are highly correlated with those of DHEAS. r = Pearson coefficient.
abnormalities are certainly not, or at least marginally,
assessed with the open field test, especially because BDL
rats did not appear on gross examination to locomote in
an unusual fashion and displayed no obvious neurologic
impairment. In contrast with the open field test, the
treadmill test allows the evaluation of both physical
components of fatigue18 and cognitive disabilities due to
impairment of sensorimotor activities.21 An alternative
explanation for the high fatigue scores found using the
treadmill test could involve highly iterative ÔforcedÕ
activity during the treadmill test in contrast to the less
iterative ÔspontaneousÕ activity when animals are
exposed to an open field. Another argument in favour
of the specificity for the treadmill test to evaluate
fatigue in BDL rats is given by our finding that the
extent of performances of cholestatic and cirrhotic rats
are similar, consolidating the generally recognized view
that fatigue is independent of the stage of liver disease.
This study describes a significant amelioration of
fatigue symptoms following intraperitoneal treatment
with DHEAS in BDL rats. At this time we cannot rule
out the possibility that the beneficial effects seen with
1323
R. F. Butterworth et al.
the DHEAS are the consequence of non-specific stim-
ulatory effects of the drug. Improvement of fatigue in
BDL rats after DHEAS treatment was concomitant with
significantly increased brain DHEAS levels. Plasma
levels of DHEAS were highly correlated with those
within the brain supporting the notion that DHEAS is a
neuroactive steroid that readily crosses the blood/brain
barrier.23 Dehydroepiandrosterone sulphate is a neuro-
active steroid known to affect neurotransmission in the
brain through its allosteric effects on the function of
several membrane receptors. A well-characterized effect
of DHEAS in the brain is its negative allosteric modu-
latory action on the c-aminobutyric acid (GABA)-A
receptor24; GABA is the most prevalent inhibitory
neurotransmitter system in the mammalian CNS.
Increased levels of DHEAS therefore have the potential
to induce excitatory effects. In mice, intraperitoneal
administration of DHEAS induces hyperexcitability and
aggressive behaviour via the GABA-A receptor,25 and
circulating levels of DHEAS are predictive of responses
to electroconvulsive therapy in depressed psychotic
patients.26 Increased brain levels of DHEAS could
increase the tonic negative allosteric modulatory activ-
ity on the GABA-A receptor, resulting in a net inhibition
of GABA-A receptor function, and consequently coun-
teract potential neural inhibition in brains of BDL rats. It
is possible that increased positive allosteric modulatory
activity on the GABA-A receptor may cause fatigue and
other depressive symptoms in cholestatic rats and
patients with liver diseases. Support for this view is
provided by our recent findings that DHEAS is reduced
in plasma of PBC patients with severe fatigue.4 Further-
more, the most potent positive allosteric modulator of
GABA-A receptors, namely allopregnanonone (3a,5a-
tetrahydroprogesterone), and its less potent isomer
pregnanolone (3a,5b-tetrahydroprogesterone) are accu-
mulated in plasma of PBC patients with severe fatigue.27
In animals, whether depressive symptoms seen in
cholestatic rats, such as anhedonia which is reflected
by significantly reduced saccharin preference,28 are due
to changes of brain neuroactive steroid homeostasis
needs to be determined. Support for such a view is
provided by recent findings demonstrating that intra-
cranial administration of allopregnanolone significantly
reduces glucose intake of rats.29 Further studies are
needed to address these issues.
This study demonstrates increased brain levels of
DHEA following DHEAS treatment in a manner that is
REFERENCES biliary cirrhosis in Ontario, Canada.
Hepatology 1990; 12: 98–105.
1 Witt-Sullivan H, Heathcote J, Cauch 2 Huet PM, Deslauriers J, Tran A,
K et al. The demography of primary Faucher C, Charbonneau J. Impact of
1324
Neurogastroenterology and Motility
highly and significantly correlated with those of
DHEAS, suggesting that part of DHEAS is converted
to DHEA in treated rats. Such conversion could occur
in the periphery or following DHEAS transport into the
brain where it could be converted by hydrolysis of the
sulphated group in presence of the enzyme sulpha-
tase.30 These findings suggest that the effects of
DHEAS could be partly mediated through DHEA, a
neurosteroid which has different neuroactive proper-
ties in the brain. For example, DHEA has weaker
allosteric properties on GABA-A receptors compared
with DHEAS.31 Furthermore, DHEA has less activity
at both N-methyl-D-aspartate (NMDA) and sigma 1
receptors compared to DHEAS.32 However, while
DHEA seems to be less potent on membrane recep-
tor-mediated neurotransmission compared with its
sulphated form, DHEA has more efficient neuropro-
tective properties. For example, DHEA, but not
DHEAS, mediates neuroprotection following NMDA
toxicity by significantly inhibiting nitric oxide (NO)
production and Ca2+-sensitive NO synthase activity
caused by NMDA stimulation in rat hippocampal
neurons.33 Furthermore, DHEA, but not DHEAS,
decreases apoptosis in neural culture.34 These findings
suggest that other mechanisms for the beneficial
effects of DHEAS could be involved either directly
through NMDA and/or sigma 1 receptors35,36 or indi-
rectly via neuroprotective properties of DHEA. Further
studies are needed to address these issues.
In conclusion, results of this study demonstrate that
cholestatic liver injury resulting from BDL leads to
significant fatigue and that the neuroactive steroid
DHEAS improves fatigue scores in these animals.
These findings further support the notion that substi-
tutive therapies with DHEAS or its precursor DHEA
could represent novel approaches in the management
of fatigue in patients with PBC.
ACKNOWLEDGMENTS
The authors thank the Canadian Institutes of Health Research
(CIHR) financial support (to RFB, CP and GBB) and are grateful to
Gail Rauw for her excellent technical assistance with neuroster-
oid measurements. SA is a recipient of the IBRO Return Home
Programme Fellowship.
COMPETING INTERESTS
The authors have no competing interests.
fatigue on the quality of life of patients
with primary biliary cirrhosis. Am
J Gastroenterol 2000; 95: 760–7.
 2009 Blackwell Publishing Ltd
Volume 21, Number 12, December 2009
3 Swain MG. Fatigue in liver disease:
pathophysiology and clinical man-
agement. Can J Gastroenterol 2006;
20: 181–8.
4 Ahboucha S, Pomier-Layrargues G,
Hassoun Z et al. Reduced plasma de-
hydroepiandrosterone sulfate levels
are significantly correlated with fati-
gue severity in patients with primary
biliary cirrhosis. Neurochem Int
2008; 52: 569–74.
5 Luu-The V, Bernier F, Dufort I. Ste-
roid sulfotransferases. J Endocrinol
1996; 150(Suppl): S87–97.
6 Baulieu EE, Robel P. Dehydroepiand-
rosterone (DHEA) and dehydroepi-
androsterone sulfate (DHEAS) as
neuroactive neurosteroids. Proc
Natl Acad Sci USA 1998; 95: 4089–
91.
7 Aldred S, Waring RH. Localisation of
dehydroepiandrosterone sulphotrans-
ferase in adult rat brain. Brain Res
Bull 1999; 48: 291–6.
8 Liere P, Pianos A, Eychenne B et al.
Novel lipoidal derivatives of preg-
nenolone and dehydroepiandroster-
one and absence of their sulfated
counterparts in rodent brain. J Lipid
Res 2004; 45: 2287–302.
9 Svec F, Porter JR. The actions of
exogenous dehydroepiandrosterone in
experimental animals and humans.
Proc Soc Exp Biol Med 1998; 218: 174–
91.
10 Yen SS. Dehydroepiandrosterone sul-
fate and longevity: new clues for an
old friend. Proc Natl Acad Sci USA
2001; 98: 8167–9.
11 Nawata H, Yanase T, Goto K, Okabe
T, Ashida K. Mechanism of action of
anti-aging DHEA-S and the replace-
ment of DHEA-S. Mech Ageing Dev
2002; 123: 1101–6.
12 Dillon JS. Dehydroepiandrosterone,
dehydroepiandrosterone sulfate and
related steroids: their role in inflam-
matory, allergic and immunological
disorders. Curr Drug Targets Inflamm
Allergy 2005; 4: 377–85.
13 Johnson DA, Wu T, Li P, Maher TJ.
The effect of steroid sulfatase inhibi-
tion on learning and spatial memory.
Brain Res 2000; 865: 286–90.
14 Steffensen SC. Dehydroepiandroster-
one sulfate suppresses hippocampal
recurrent inhibition and synchro-
nizes neuronal activity to theta
rhythm. Hippocampus 1995; 5: 320–
8.
15 Reddy DS, Kaur G, Kulkarni SK. Sig-
ma (sigma 1) receptor mediated anti-
depressant-like effects of neuroster-
 2009 Blackwell Publishing Ltd
View publication stats
oids in the Porsolt forced swim test.
NeuroReport 1998; 9: 3069–73.
16 Wolkowitz OM, Reus VI, Roberts E
et al. Dehydroepiandrosterone (DHEA)
treatment of depression. Biol Psychi-
atry 1997; 41: 311–8.
17 Wolkowitz OM, Reus VI, Keebler A
et al. Double-blind treatment of ma-
jor depression with dehydroepiand-
rosterone. Am J Psychiatry 1999; 156:
646–9.
18 Soares DD, Coimbra CC, Marubay-
ashi U. Tryptophan-induced central
fatigue in exercising rats is related
to serotonin content in preoptic
area. Neurosci Lett 2007; 415:
274–8.
19 Swain MG, Maric M. Improvement in
cholestasis-associated fatigue with a
serotonin receptor agonist using a
novel rat model of fatigue assess-
ment. Hepatology 1997; 25: 291–4.
20 Cameron GR, Oakley CL. Ligation of
the common bile duct. J Pathol Bac-
terial 1932; 35: 769–98.
21 Burak KW, Le T, Swain MG. In-
creased sensitivity to the locomotor-
activating effects of corticotropin-
releasing hormone in cholestatic rats.
Gastroenterology 2002; 122: 681–8.
22 Le Marec N, Lalonde R. Treadmill
performance of mice with cerebellar
lesions: 2. Lurcher mutant mice.
Neurobiol Learn Mem 2000; 73: 195–
206.
23 Asaba H, Hosoya K, Takanaga H et al.
Blood–brain barrier is involved in the
efflux transport of a neuroactive ste-
roid, dehydroepiandrosterone sulfate,
via organic anion transporting poly-
peptide 2. J Neurochem 2000; 75:
1907–16.
24 Majewska MD, Demirgoren S, Spi-
vak CE, London ED. The neuroster-
oid dehydroepiandrosterone sulfate
is an allosteric antagonist of the
GABA(A) receptor. Brain Res 1990;
526: 143–6.
25 Nicolas LB, Pinoteau W, Papot S,
Routier S, Guillaumet G, Mortaud S.
Aggressive behavior induced by the
steroid sulfatase inhibitor COU-
MATE and by DHEAS in CBA/H
mice. Brain Res 2001; 922: 216–22.
26 Maayan R, Yagorowski Y, Grupper D
et al. Basal plasma dehydroepiand-
rosterone sulfate level: a possible
predictor for response to electrocon-
vulsive therapy in depressed psy-
chotic inpatients. Biol Psychiatry
2000; 48: 693–701.
27 Ahboucha S, Butterworth RF, Pomier-
Layrargues G, Vincent C, Hassoun Z,
1325
Brain DHEAS and fatigue in BDL rats
Baker G. Neuroactive steroids and fa-
tigue severity in patients with primary
biliary cirrhosis and hepatitis C.
Neurogastroenterol Motil 2008; 20:
671–9.
28 Swain MG, Le T. Chronic cholestasis
in rats induces anhedonia and a loss
of social interest. Hepatology 1998;
28: 6–10.
29 Martin-Garcı́a E, Darbra S, Pallarès
M. Intrahippocampal allopregnano-
lone decreases voluntary chronic
alcohol consumption in non-selected
rats. Prog Neuropsychopharmacol
Biol Psychiatry 2007; 31: 823–31.
30 Li XM, Salido EC, Gong Y et al. Clon-
ing of the rat steroidsulfatase gene(Sts),
a non-pseudoautosomal X-linked gene
that undergoes X inactivation. Mamm
Genome 1996; 7: 420–4.
31 Park-Chung M, Malayev A, Purdy
RH, Gibbs TT, Farb DH. Sulfated and
unsulfated steroids modulate gamma-
aminobutyric acidA receptor function
through distinct sites. Brain Res
1999; 830: 72–87.
32 Maurice T, Grégoire C, Espallergues J.
Neuro(active)steroids actions at the
neuromodulatory sigma1 (sigma 1)
receptor: biochemical and physiolog-
ical evidences, consequences in neu-
roprotection. Pharmacol Biochem
Behav 2006; 84: 581–97.
33 Kurata K, Takebayashi M, Morinobu
S, Yamawaki S. b-Estradiol, dehy-
droepiandrosterone, and dehydroepi-
androsterone sulfate protect against
N-methyl-D-aspartate-induced neu-
rotoxicity in rat hippocampal
neurons by different mechanisms.
J Pharmacol Exp Ther 2004; 311:
237–45.
34 Zhang L, Li B, Ma W et al. Dehydro-
epiandrosterone (DHEA) and its sul-
fated derivative (DHEAS) regulate
apoptosis during neurogenesis by
triggering the Akt signaling pathway
in opposing ways. Brain Res Mol
Brain Res 2002; 98: 58–66.
35 Johansson T, Le Greves P. The effect
of dehydroepiandrosterone sulfate
and allopregnanolone sulfate on the
binding of [3H]ifenprodil to the N-
methyl-D-aspartate receptor in rat
frontal cortex membrane. J Steroid
Biochem Mol Biol 2005; 94: 263–6.
36 Monnet FP, Mahe V, Robel P, Baulieu
EE. Neurosteroids, via sigma recep-
tors, modulate the [3H]norepineph-
rine release evoked by N-methyl-D-
aspartate in the rat hippocampus.
Proc Natl Acad Sci USA 1995; 92:
3774–8.