Download Evidence for a modulatory effect of sulbutiamine on

Neuroscience Letters 292 (2000) 49±53
www.elsevier.com/locate/neulet
Evidence for a modulatory effect of sulbutiamine on glutamatergic
and dopaminergic cortical transmissions in the rat brain
Fabrice Trovero a,*, Marco Gobbi b, Jeanne Weil-Fuggaza c, Marie-Jo Besson c,
Denis Brochet d, Sylvain Pirot e
a
Key-Obs S.A., Centre d'Innovation, 16, rue Leonard de Vinci, 45074 Orleans, Cedex 2, France
b
Istituto di Ricerche Farmacologiche Mario Negri, Via Eritrea, 62, 20157 Milan, Italy
c
Laboratoire de Neurochimie-Neuroanatomie URA 1488, 9 Quai Saint Bernard, 75005 Paris, France
d
Psypharm S.A., B.P. 0102, 53001 Laval, France
e
A.N.P.P., 25 rue de la Plaine 75020 Paris, France
Received 26 June 2000; received in revised form 2 August 2000; accepted 4 August 2000
Abstract
Chronic treatment of rats by sulbutiamine induced no change in density of N-methyl-d-aspartate (NMDA) and (^)-aamino-3-hydroxy-5-methylisoxazole-4-propionic acid receptors in the cingular cortex, but a signi®cant decrease of the
kainate binding sites, as measured by quantitative autoradiography. In the same treated animals, an increase of D1
dopaminergic (DA) binding sites was measured both in the prefrontal and the cingular cortex, while no modi®cation of
the D2 binding sites was detected. Furthermore, an acute sulbutiamine administration induced a decrease of kainate
binding sites but no change of the density of D1 and D2 DA receptors. Acute sulbutiamine injection led to a decrease of
the DA levels in the prefrontal cortex and 3,4-dihydroxyphenylacetic acid levels in both the cingular and the prefrontal
cortex. These observations are discussed in terms of a modulatory effect of sulbutiamine on both dopaminergic and
glutamatergic cortical transmissions. q 2000 Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Sulbutiamine; Glutamate receptors; Dopaminergic receptors; Prefrontal cortex; Cingular cortex; Autoradiography
Sulbutiamine is a hydrophobic molecule that easily
crosses the blood±brain barrier and gives rise to thiamine
and thiamine phosphate esters in the brain [2,3]. It does not
have psychostimulant properties and is currently used for
treatment of somatic and psychic inhibition. For example, it
improves the functioning of episodic memory in schizophrenic patients [11] and hastens the resorbtion of psychobehavioural inhibition occurring during major depressive
disorder [7]. Moreover, it improves performances in behavioural models of inhibition (induced by an aversive situation) such as the `learned helplessness' and the `forced
swimming test' (Lacroix and Bourin, personal communications). Given the role of dopaminergic and glutamatergic
transmissions in the anterior cortical regions, (e.g. prefrontal
and cingular cortices) where decisions and strategies are
organized, the present study has been conducted in order
to determine the effects of sulbutiamine on these systems.
For this aim, the effects of acute and chronic sulbutiamine
* Corresponding author. Tel.: 133-2-38-64-60-68; fax: 133-238-64-88-73.
E-mail address: [email protected] (F. Trovero).
treatments on glutamatergic and dopaminergic binding sites
in the rat brain were analyzed by autoradiographic analysis
in prefrontal and cingular cortices and compared to a subcortical structure, the nucleus accumbens.
Male Sprague±Dawley rats weighing 200±250 g (Charles
River France, Saint-Aubin les Elbeuf) have been used. For
chronic treatments, rats have been treated intraperitoneally
(i.p.) with sulbutiamine (12.5 mg/kg) or vehicle (Arabic gum,
15% w/v) once a day for 14 days, and killed 2 h or 5 days after
the last injection. For acute treatments, rats have been treated
intraperitoneally with sulbutiamine (12.5 mg/kg) or vehicle
(arabic gum, 15% w/v), and have been sacri®ced after
various times following the injection (15 and 30 min, 1, 4,
12 and 24 h) for autoradiographic analysis of binding sites.
The brain was rapidly dissected out and was frozen in isopentane (2408C) Sections (20 mm thick) were cut with a cryostat
at 2208C at different levels of the brain. They were mounted
onto gelatine-coated glass slides and stored at 2208C until
the day of incubation with ligand. Autoradiographic
experiments of D1 dopaminergic (D1-DA), N-methyl-daspartate (NMDA), (^)-a-amino-3-hydroxy-5-methylisox-
0304-3940/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved.
PII: S03 04 - 394 0( 0 0) 01 42 0- 8
50
F. Trovero et al. / Neuroscience Letters 292 (2000) 49±53
azole-4-propionic acid (AMPA) and kainate were performed
according to the procedures previously described [1,9,14].
For D2 dopaminergic binding sites, sections were incubated
for 60 min at 208C in a 50 mM Tris±HCl buffer (pH 7.4)
containing ( 125I)-iodosulpride (0.2 nM) (from NEN Dupont,
France). Non-speci®c binding was determined by adding
10 26 M unlabelled sulpiride. Sections were washed ®ve
times in ice-cold 50 mM Tris±HCl buffer (pH 7.4) and
dried under a stream of cold air. Finally, the sections were
exposed to 3H-ultro®lm (LKB) for 10 days. After developing,
binding quanti®cation was obtained by densitometric analysis of autoradiographic ®lms. Autoradiograms were transformed into digitised diagrams (512 £ 512 pixels) with an
image analyzer giving 256 grey levels per pixel. Each
value of grey level was transformed into optical density
(OD) by the computer. Autoradiographic ®lm densities
were converted to apparent tissue ligand concentrations on
the basis of ®lm densities overlying the radioactive standards
and the speci®c activity of the radioligand. Differences
between treatments were evaluated using one-factor analysis
of variance for repeated measures (ANOVA), followed by
the Tuckey±Kramer t-test. For estimation of monoamine
levels, rats were sacri®ced 2 h following the injection of
sulbutiamine (12.5 mg/kg) and the brain was rapidly
dissected out and was frozen at 2808C, until the dissection.
On frontal brain sections (400 mm thick) realized with a
microtome, prefrontal cortex, anterior cingular cortex and
nucleus accumbens were dissected and kept at 2808C until
Fig. 1. Representative autoradiograms of rat brain coronal
sections obtained following incubations with speci®c ligands.
(A) D1 dopaminergic binding sites. Sections were incubated in
presence of 1 nM ( 3H)-SCH 23390 and 10 nM spiroperidol to
avoid labelling of 5-hydroxytryptamine tri¯uoroacetate (HT)2
serotonergic binding sites. Sections were exposed to 3H-ultro®lm (LKB) for 15 days. (B) D2 dopaminergic binding sites.
Sections were incubated in presence of 0.2 nM ( 125I)-iodosulpride. Sections were exposed to 3H-ultro®lm (LKB) for 10 days.
(C) AMPA glutamatergic binding sites. Sections were incubated
in presence of 34 nM ( 3H)-glutamate. Sections were exposed to
3
H-ultro®lm (LKB) for 2 weeks. (D) Kainate glutamatergic binding
sites. Sections were incubated in presence of 60 nM ( 3H)-kainic
acid. Sections were exposed to 3H-ultro®lm (LKB) for 6 weeks. (E)
NMDA glutamatergic binding sites. Sections were incubated
with 200 nM ( 3H)-glutamate and of 5 mM AMPA and 1 mM kainic
acid. Sections were exposed to 3H-ultro®lm (LKB) for 6 weeks.
analysis. Tissue samples were put into 100 ml of perchloric
acid 0.1 M containing ethylene diamine tetra-acetic acid
(EDTA) 0.1 mM and 0.01% cysteine and sonicated. Following centrifugation (35 000 £ g, 40 min at 48C), supernatants
were kept at 2808C until dosages. They were puri®ed on
alumine microcolumns and were injected into a high-pressure liquid chromatography C18 column equilibrated with a
Fig. 2. Effect of a chronic treatment by sulbutiamine on the
density of glutamatergic and dopaminergic binding sites in
various cerebral structures. The values are expressed in fmol/
mg of tissue and are the means ^ SEM calculated from data
obtained with six animals per group. (a) Signi®cantly different
from the control-5 days group P , 0:05. (b) Signi®cantly different from the control-2 h group P , 0:05.
F. Trovero et al. / Neuroscience Letters 292 (2000) 49±53
51
Fig. 3. Effect of an acute administration of sulbutiamine (12.5 mg/kg, (i.p.)) on the density of D1 and D2 dopaminergic and kainate binding
sites in the cingular and prefrontal cortex. For each time following sulbutiamine injection, the values are expressed in percent of the
control value (^SEM calculated from data obtained with six animals per group). `Control' column is the mean ^ SEM of control values
obtained at various times. * Signi®cantly different from the control group P , 0:01.
solvent consisting of sodium phosphate buffer (0.1 M),
1-octanesulphonic acid (2.75 mM), triethylamine (0.25
mM), EDTA (0.1 mM), NaH2PO4 (0.1 M) and methanol
(15%), adjusted with phosphoric acid to pH 2.9. The injection
rate was 1 ml/min DA and 3,4-dihydroxyphenylacetic acid
(DOPAC) levels were quanti®ed by electrochemistry using a
carbon electrode (Chromato®eld Eldec 102) set at 0.65 V.
Fig. 1 shows representative autoradiograms obtained
from rat brain sections following incubations with various
ligands at an anteriority level showing the nucleus accumbens and the cingular cortex. The densitometric quantitative analysis of the binding in the different groups is
reported on Fig. 2. For comparison, densitometry has
been systematically performed on the same structures,
except for 3H-glutamate, whose binding is only enriched
in indicated regions. The chronic treatment induced a
signi®cant increase in the density of D1 binding sites in
both the prefrontal and the anterior cingular cortex (126
and 134%, respectively), only 2 h following the interruption of the treatment. No change in the density of D2 DA
receptor was observed. Furthermore, 5 days following the
chronic treatment, a signi®cant decrease in the density of
the 3H-kainate binding was observed in the cingular cortex
and the nucleus accumbens (236 and 228%, respectively).
A signi®cant and similar decrease has also been observed
in the striatum and the hippocampus (data not shown). The
chronic treatment induced no signi®cant change in the
density of other glutamatergic binding sites (NMDA and
AMPA subtypes).
The acute administration of sulbutiamine did not induce
any change in the binding density of 3H-SCH 23390 and of
125
I-iodosulpride (Fig. 3). Fifteen minutes following sulbutiamine administration, a signi®cant decrease of the density
of kainate binding sites was observed in the cingular cortex.
Such a decrease was also observed in the prefrontal cortex
one hour after the sulbutiamine administration. In both these
structures, the decrease was still persistent 24 h after the
injection. Since no change was observed 2 h following the
chronic treatment despite the presence of sulbutiamine
metabolites in the brain [3], the occupation of kainate binding sites by sulbutiamine and/or by its metabolites can be
excluded to explain the binding decrease. No variation in
52
F. Trovero et al. / Neuroscience Letters 292 (2000) 49±53
Table 1
Effects of a single intraperitoneal injection of sulbutiamine (12.5 mg/kg) on levels of dopamine and its metabolites in the prefrontal
cortex, the cingular cortex and in the nucleus accumbens a
Prefrontal cortex
Dopamine
HVA
DOPAC
Cingular cortex
Nucleus accumbens
Control
Sulbutiamine
Control
Sulbutiamine
Control
Sulbutiamine
2.06 ^ 0.21
3.18 ^ 0.27
1.05 ^ 0.09
1.65 ^ 0.08
2.97 ^ 0.15
0.75 ^ 0.03*
3.64 ^ 0.39
2.88 ^ 0.25
1.41 ^ 0.1
2.43 ^ 0.16*
2.66 ^ 0.13
1.05 ^ 0.06*
366 ^ 26
23.3 ^ 1.5
56.6 ^ 3.6
248 ^ 15
22.2 ^ 0.9
53.4 ^ 2.9
a
Data are the mean ^ SEM of ten determinations per group and are expressed in pmoles/mg of protein. * Signi®cantly different from
the control group P , 0:01.
the density of D1, D2 and kainate binding sites has been
observed in the nucleus accumbens (data not shown).
Table 1 shows that following an acute injection of sulbutiamine, DOPAC levels were signi®cantly reduced (230%)
in the prefrontal cortex. In the cingular cortex, DA and
DOPAC levels were decreased (234 and 226%, respectively), as compared to control animals. Finally, no change
in dopaminergic transmission was observed in the nucleus
accumbens. The decreased levels of DOPAC suggest a
reduced release of DA in the prefrontal cortex. The
increased density of D1 DA receptors 2 h following a
chronic treatment (Fig. 2), may result from this chronic
modi®cation and its induced compensatory mechanisms.
These mechanisms disappear with the interruption of the
sulbutiamine treatment, no more modi®cation of D1 binding
sites being observed ®ve days later (Fig. 2). A single injection of sulbutiamine should not be suf®cient to change the
D1 receptor density (Fig. 3). These observations suggest
that changes of receptor sensitivity take place following a
chronic change in the cortical dopaminergic transmission
induced by sulbutiamine.
Thus, the changes in density of kainate receptor in the
cortex lead to suggest that sulbutiamine and/or its metabolites may modulate the cortical glutamatergic transmission.
In fact, the rapid decrease observed immediately following a
single injection suggests a direct effect on the cortical glutamatergic transmission, for instance a modulation of the
intra-synaptic glutamate concentration. Cellular mechanisms responsible for the effects of sulbutiamine are
unknown. Thiamine triphosphate, a derivative signi®cantly
enhanced in rat brain following an injection of sulbutiamine,
could have modulatory effect on neuronal membrane
permeability [2,3]. Such a modulation induces ionic
changes, and further modulate the binding on kainate receptors. Changes on glutamatergic transmission could be at the
origin of the modulation of dopaminergic cortical transmission. Indeed, numerous behavioural, anatomical and electrophysiological studies report the modulation of
dopaminergic cortical transmission by glutamate, notably
through kainate receptors [4,8,10,12]. On the contrary,
chronic blockade of dopaminergic transmission by antipsychotic drugs does not change the density of cortical kainate
binding sites, even though the expression of their mRNA is
affected [6,13]. In the nucleus accumbens kainate receptors
are down regulated without any signi®cant change in dopaminergic transmission.
The interactions between dopaminergic and glutamatergic transmissions within the prefrontal cortex could play a
pivotal role in the therapeutic action of sulbutiamine. Those
results strongly support recent ®ndings that demonstrate
improvement in behavioural, cognitive, attentional and
functional disorders in schizophrenic, alcoholic and
depressed patients [5,7,11]. Other experiments, including
release and/or uptake studies, have to be conducted in
order to further investigate the respective role of the cortical
glutamatergic and dopaminergic cortical transmissions in
the effect of sulbutiamine.
[1] Albin, R.L., Makowiec, R.L., Hollingsworth, Z.R., Dure,
L.S.T., Penney, J.B. and Young, A.B., Excitatory amino
acid binding sites in the basal ganglia of the rat: a quantitative autoradiographic study, Neuroscience, 46 (1992) 35±
48.
[2] Bettendorf, L., Weekers, L., Wins, P. and Schoffeniels, E.,
Injection of sulbutiamine induces an increase in thiamine
triphosphate in rat tissues, Biochem. Pharmacol., 40 (1990)
2557±2560.
[3] Bettendorff, L., Wins, P. and Lesourd, M., Subcellular localization and compartmentation of thiamine derivatives in
rat brain, Biochim. Biophys. Acta, 1222 (1994) 1±6.
[4] Borkowska, H.D., Oja, S.S., Saransaari, P. and Albrecht, J.,
Release of [ 3H]dopamine from striatal and cerebral cortical
slices from rats with thioacetamide-induced hepatic encephalopathy: different responses to stimulation by potassium ions and agonists of ionotropic glutamate receptors,
Neurochem. Res., 22 (1997) 101±106.
[5] Gueguen, B., Aubin, H.J., Guillou, S., Ollat, H., Elatki, S. and
Barrucand, D., Cognitive disorders in alcoholic patients.
ERP evaluation and treatment, Neuropsychobiology,
(2000) in press.
[6] Healy, D.J. and Meador-Woodruff, J.H., Clozapine and haloperidol differentially affect AMPA and kainate receptor
subunit mRNA levels in rat cortex and striatum, Brain
Res. Mol. Brain Res., 47 (1997) 331±338.
[7] Loo, H., Poirier, M.F., Ollat, H. and Elatki, S., Etude des effets
de la sulbutiamine (Arcalion 200) sur l'inhibition psychocomportementale des eÂpisodes deÂpressifs majeurs, L'EnceÂphale, 26 (2000) 70±75.
[8] Meltzer, L.T., Christoffersen, C.L. and Serpa, K.A., Modulation of dopamine neuronal activity by glutamate receptor
subtypes, Neurosci. Biobehav. Rev., 21 (1997) 511±518.
F. Trovero et al. / Neuroscience Letters 292 (2000) 49±53
[9] Mennini, T., Miari, A., Presti, M.L., Rizzi, M., Samanin, R.
and Vezzani, A., Adaptive changes in the NMDA receptor
complex in rat hippocampus after chronic treatment with
CGP 39551, Eur. J. Pharmacol., 271 (1994) 93±101.
[10] Moghaddam, B., Adams, B., Verma, A. and Daly, D., Activation of glutamatergic neurotransmission by ketamine: a
novel step in the pathway from NMDA receptor blockade
to dopaminergic and cognitive disruptions associated with
the prefrontal cortex, J. Neurosci., 17 (1997) 2921±2927.
[11] Patris, M., Kastler, B., Rohmer, J.G., Biringer, F., Elatki, S.
and Ollat, H., Schizophrenic cognitive disorders. Is a pharmacological treatment within reach? (2000) submitted for
publication.
53
[12] Sesack, S.R. and Pickel, V.M., Prefrontal cortical efferents in
the rat synapse on unlabeled neuronal targets of catecholamine terminals in the nucleus accumbens septi and on
dopamine neurons in the ventral tegmental area, J.
Comp. Neurol., 320 (1992) 145±160.
[13] Tarazi, F.I., Florijn, W.J. and Creese, I., Regulation of ionotropic glutamate receptors following subchronic and
chronic treatment with typical and atypical antipsychotics,
Psychopharmacology (Berl.), 128 (1996) 371±379.
[14] Trovero, F., Blanc, G., Herve, D., Vezina, P., Glowinski, J.
and Tassin, J.P., Contribution of an alpha 1-adrenergic
receptor subtype to the expression of the `ventral tegmental area syndrome', Neuroscience, 47 (1992) 69±76.