Prog. ~euro.Psychopharmacol.

& Bml. Psychrat.

Copyright

2000, Vol. 24, pp. 1007-1015

0 2000 Elsevier

Printed in the USA

0278.5846/00/$-see

ELSEVIER

Science

All nghts

Inc.

reserved

front matter

PII: SO278-5846(00)00120-2

MODULATION OF KAINATE - ACTIVATED CURRENTS BY

DIAZOXIDE AND CYCLOTHIAZIDE ANALOGUES (IDRA) IN
CEREBELLAR GRANULE NEURONS
GIULIA PUIA, GABRIELE LOSI, GIORGIA RAZZINI, DANIELA BRAGHIROLI,
MARIA DJ BELLA and MARIO BARALDI
Department

of Pharmaceutical

Sciences, Modena, ITALY

(Final form, July 2000)

Abstract

Puia Giulia, Gabriele Losi, Giorgia Razzini, Daniela Braghiroli, Maria Di Bella and Mario Baraldi:
Modulation of kainate-activated currents by d&oxide and cyclothiazide analogues (IDRA) in cerebellar
granule neurons. Prog. Neuro-Psycbopharmacol.&
BioLPsychiat. 2000,~, pp. 1007-1015.
02000

Elsevier Science Inc.

1. Patch-clamp

2.

3.

4.

5.

technique was used in primary cultures of cerebellar granule neurons to study the
modulation of the cyclothiazide analogue (IDRA21) and of the d&oxide derivative (IDRA 5) on
KA-evoked currents.
The dose-response of kainic acid (KA) reveals an E&o=90 PM and an Hill coefficient of 1.3.
IDRA 21 and cyclothiazide
potentiate KA-evoked current
in a dose dependent way, being
cyclothiazide more potent but less efficacious than IDRA 21. Conversely IDRA 5 acts as a negative
modulator of KA evoked -current
Application of IDRA 21 and cyclothiazide
results in a current potentiation of 125+18 % and 80 +
12 % respectively, while IDRA 5 decreases KA-current (-21+5 %). Coapplication of cyclothiazide
and IDRA 21 produces a potentiation
of 1 IO+ 17 %, suggesting a competition of the two drugs for
the same site.
In the same experimental model we studied the ability of IDRA compounds of promoting toxicity
through AMPA-receptor
activation. Under basal conditions AMPA treatment (50 PM for 1 hour)
results in a negligible excitotoxicity.
In contrast similar treatment with AMPA + IDRA 21 (1 mM) or + IDRA 5 (1 mM) or + cyclothiazide
(100 l.tM) induces citotoxicity. The neurotoxic damage induced by IDRA 21 and cyclothiazide is
blocked by GYKI 53655 (50 PM) and by NBQX (10 pM). Interestingly GYKI and NBQX are
ineffective in reducing IDRA 5 toxicity.

Keywords:
benzothiadiazine,
desensitization.

Glutamate

receptor,

neurotoxicity,

patch-clamp

technique,

receptor

Abbre latlaas : a-amino-3-hydroxy-5-methylisoxazolepropionic

acid (AMPA) receptor (AMPAR),
centray nervous system (CNS),
dimethhyl sulfoxide (DMSO), glutamate receptor (GLUR), l-(4aminophenyl)-4-methyl-7,8-methylenedioxi-5H-2,3
benzodiazepine(N-methyl)carbamate
(GYIU
53655), 7-chloro-3-methyl-3,4-dihydro-2H-l,2,4-benzo-thiadi~ine
S,S-dioxide (IDRA 21), kainic acid
receptor (KAR), 2,3-dihydroxy-6-nitro-7-sulphamoil-benzo(F)
quinoxalinedione
(NBQX), N-methyl-Daspartate (NMDA), holding potential (V,).

1007

1008

G. Puia et al.
Introduction

L-Glutamate mediates fast synaptic transmission

acting at

three different

of ionotropic

typeS

at the majority of excitatory

glutamate

receptors stimulation produces fast synaptic response,

slow synaptic response while the physiological
1993). Recently it has been demonstrated
activate KA receptors
Collingridge

in hippocampal

receptors:

AMPA,

synapses
NMDA

NMDA receptors activation

in the CNS by

and KA. AMPA

is responsible

for the

functions of KA receptors are still under debate ( Seeburg

that high frequency

intemeurons

stimulation

resulting

of mossy fibers patway could

in a slow synaptic

response

(Vignes

and

1997, Castillo, Malenka and Nicoll 1997).

A distinctive

feature of AMPA and KA receptors

is that they undergo

a rapid desensitization

after

exposure to glutamate (Kiskin et al. 1986, Lerma et al. 1993). Recent sudies on memory and cognitionenhancing drugs implicate

AMPA receptors

cyclothiazide,

are thought

aniracetam)

removal of the receptor desensitization

be the most potent compound

to work by potentiating

in vitro able to remove desensitization

efficiently

cognition-enhancer
experiments

than cyclothiazide.

(Zivkovic

showed

personal observation).

that

characteristics

et al. 1994, Thompson

IDRA 5 worsened

In previous studies the modulation

investigated in hippocampal

(Vyklicky et
does not

derivatives

synthesized

that allow it to cross the blood-brain

studies

evidentiate

its efficacy
some

induced by scopolamine

as a

preliminary

in rats

(Baraldi,

by IDRA 21 of glutamate receptor function was

slices (Bertolino et al. 1993).

In the present study the authors analyze IDRA 21 and IDRA 5 modulation
granule neurons

primary culture of cerebellar

augmentation

to

that manage to reach

et al. 1995). Conversely

the amnesia

through a

Since cyclothiazide

analogues of this compound

Also behavioural

currents

has been demonstrated

effects. Among the diverse benzothiadiazine

in our laboratory, IDRA 21 shows physicochemical

more

synaptic

(i.e. diazoxide,

from AMPA receptors

in this way synaptic transmission.

barrier, it is important to develop

the CNS and do not have peripheral

barrier

glutamate

(Yamada and Tang 1993). Cyclothiazide

al 1991, Patneau et al. 1993) enhancing

cross the blood-brain

as their target of action. These compounds

of KA-evoked

using cyclothiazide

current can ultimately lead to neuronal

of KA-evoked

as a reference

currents in

compound.

Since

death the authors have also studied

the potential toxicity of IDRA derivatives.

Methods
Primal?, Cultures qf Cerebellar Granule Cells
Primary culture of cerebellar
previously

described

mg/ml)(SIGMA,
with poly-L-lisine

granule neurons were prepared

from 7-8 days old Sprague-Dawley

(Gallo et al. 1992). Briefly, cells from cerebella were dispersed

rats as

with trypsin (0.24

St. Louis, MO) and plated at a density of lo6 cells/ml on 35 mm Falcon dishes coated
(lougiml,

Santaha, CA), supplemented

100 ug/ml gentamycin

SIGMA).

Cells were grown in basal Eagles Medium

(Irvine Scientific

with 10% fetal bovine serum (Hyclone Lab, Utah), 2mM glutamine,

(SIGMA)

and maintained

at 37C in 5% CO2

Cytosine

arabinofuranoside

PM; SIGMA) was added to the cultures 24 hours after plating to prevent astroglial proliferation.

and
(10

Modulation

Recordings

were performed

of KA currents

1009

by IDRA compounds

using voltage -clamp in the whole-cell

configuration

(Hamill et al. 1981)

on single neuron after 7 days in culture.

Electrodes
Narishige)

were

pulled

from borosilicate

and had a resistance

amplified with an Axopatch

glass

(Hidelberg,

FRG)

on a vertical

puller

(PB-7,

of 5-7 MOhm when filled with KC1 internal solution. Currents were

1D amplifier (Axon Instruments,

digitized at 10 kHz. Pclamp software (Axon Instruments)

Foster City . CA), filtered at 5 kHz and

was used for the analysis.

Solutions and Drugs

Intracellular solution contains (mM): KC1 140, MgC12 3, EGTA 5, Hepes 5, ATP-Na 2, ; pH 7.3 with
KOH. Cells were continuosly

perfused

with the external solution (mM): NaCl 145, KC1 5, CaClr 1,

Hepes 5, Glucose 5, Sucrose 20, pH 7.4 with NaOH.

Kainic acid,

was

purchased

from SIGMA. GIKI 53655 and cyclothiazide

were

a gift from Lilly

Resarch Laboratories.
IDRA 5 and IDRA 21 were synthesized

according to the method of Bertolino et al. (1993)

(2)

(1)
0

C
(3)

Fig. 1 Chemical structures of cyclothiazide

The chemical structures of cyclothiazide
IDRA-derivatives

were dissolved

medium (DMSO final concentration

Kainic acid was dissolved

(l), IDRA 21 (2) and IDRA 5 (3)

, IDRA21 and IDRA 5 are shown in Fig.1. Cyclothiazide

in DMSO and diluted at the final concentration

and

in extracellular

less than 0.1%).

in the extracellular

through a Y-tube perfusion system (Murase

solution. All drugs were applied directly by gravity

et al 1989).

Excitotoxicitv Exoeriments
For the experiments

of toxicity the culture medium was removed,

solution was added and cultures were incubated

the final concentration

whereafter

for 1 hr at room temperature.

the appropriate

drug

Drugs were diluted to

in Mg-free Lokes buffer (NaCl 154 mM, KC1 5.6 mM, CaClz 2.3 m, NaI-ICOj

1010

G. Puia et al.

3.6 mM D-glucose

5.5 mM, Hepes 5 mM, pH 7.4). Drug exposure was stopped by washing the culture

once with drug-free Lockes Buffer and replaced later with the old culture medium. Neuronal cell injury
was assessed 24 hrs later.
Neuronal cell death was extimated by the MTT tetrazolium
MTT tetrazolium

salt was dissolved

isopropanol was added

in serum-free

The adsorbance

salt assay (Ankarcrona

medium

at 37C. Medium

et al. 1995) In brief,

was then aspirated,

were recorded at 570 run and 630 nm. Results were expressed

as percentage of the control.

Data AnaJvsis
Data were analysed using the software pClamp6.3
relationship was performed

(Axon Instrument).

using the logistic equation

%I,,=

100/I,,,

The fitting of the dose-response

(l+(ECs0&4]~))

the maximal current elicited by the agonist, EC50 is the agonist concentration
response,

and nh is the Hill coefficient.

(Microcal Software, Northampton,

Results

are expressed

eliciting the half maximal

as mean + Standard

MA) was used for figure preparation

where I,,, is

Error. Origin

and statistical analysis.

Dose Response of KA
Application of increasing concentrations
at -60 mV, evoked

of KA (1 to 1000 pM) on granule neurons voltage clamped

inward currents of increased amplitude.

The analysis of the dose-response

of the KA currents

derived an E&s= 90 pM and a Hill coefficient

1.3 (Fig. 2). The ECss for Kainate was similar to that determined

in hippocampal

neurons by Patneau

and Mayer (1990).

EC50= 90 pM
Hill axff = 1.3

,1

Fig 2. Dose response curve of KA-activated

current. Each point is the mean + SE of 8-10 neurons.

of

Modulation of KA currents

1011

by IDR4 compounds

Fig 3. Representative recording showing the potentiation of KA-evoked

currents by cyclothiazide and
IDRA 21. Bars on the top of the current traces show the duration of the drug application. Vh=-60 mV.
Modulation of KA-Resnonse
Application

bv IDRA 2 1. and Cvclothiaz&.

of KA on granule

cyclothiazide

( 100 PM)

neuron

elicits

a non desensitizing

and by IDRA21 (In&l)

current

that is potentiated

by

(Fig.3). The effect in both cases is reversible

and

time- dependent being maximal after few seconds

The onset of cyclothiazides

and IDRA2ls

action is slow and gradual as shown in Yamada and Tang

(1993) suggesting that multiple binding is required for their action.

Dose-Denendent

Potentiation

of KA bv IDRA21 and Cvclothi&&

The current evoked by the application

increasing concentrations

of KA 100 uM was potentiated

of IDRA21 and cyclothiazide

in a dose-dependent

as shown in the histogramm

fashion by

of Fig 4.

+30

KA100pM

+10

+30

+100

+300

+100

+1000

+3co

Fig 4.Dose dependent effect of IDRA 21 and Cyclothiazide on IL4 (100 FM)-evoked current.
Each bar is the mean current + SE measured on at least 8 cells. The asterisks mean signiticative
differences compared to control (Dunnet Test, p< 0.05)

1012

G. Puia et al.

Cyclothiazide

is more potent but less efficacious than IDRA 21. The maximal potentiation

evoked currents was of 125518%

(n=l7)

for 1mM IDRA 21 and of 80&12% (n=ll)

of KA-

for 100 pM

cyclothiazide.
In order to understand
recognition

whether

any

interaction

site we applied simultaneously

In 6 cells we coapphed

the two drugs

IDRA 21 and cyclothiazide

single drug. The potentiation

of the KA-evoked

IDRA 21 (143214%) and cyclothiazide

IDRA 5 reduces KA-evoked
high as 1 mM. Coapplication

occurs

between

cyclothiazide

at concentrations

and compared

21

that elicit maximal effect.

this effect to that induced by the

current was intermediate

(110217 %) between

that of

alone (78?15 %) ( Fig 5).

current of 21~5 % and its effect is detectable

of IDEA 2 1 and IDRA 5 potentiates

CYCLO

and IDRA

only at concentrations

as

KA current of 132_+10% ( Fig 5).

IDRA 21 +

Fig 5. IDRA 21, IDRA 5 and Cyclothiazide share a common site of action?
Each bar is the mean + SE (n=6) of the % potentiation of KA-evoked current by IDRA 21 (1mM) ,
cyclothiazide (CYCLO, 100pM) and IDRA 5 (1mM) alone or applied together.
Neurotoxcitv
1

of IDRA Derivatives

In the cerebellar
compounds

cultures

of promoting

the authors
toxicity

studied

through

AMPA-receptor

treatment (50 PM for 1 hour in Mg-free -Lakes)

treatment with AMPA + cyclothiazide
marked citotoxicity.
neurotoxicity

At 100 pM

ability of cyclothiazide
activation.

In control

results in a negligible

citotoxicity.

and of IDRA

conditions

AMPA

In contrast similar

(100 PM), IDRA21 (1 mM) or + IDRAS (1 mM) produces

IDEA 21 produces

induced by cyclothiazide

(10 PM). Interestingly

the potential

low neuronal

damage

(data not shown).

The

and IDRA21 is blocked by GYKI 53655 (50 pM) and by NBQX

GYKI 53655 and NBQX are ineffective

in reducing IDRAS toxicity (Fig 6).

Discussion
In the present
cyclothiazide,

work

the authors

characterized

the modulation

IDRA21 and IDRA 5 in primary culture of cerebellar

of KA-activated

currents

by

granule neurons. Kainate elicits a

Modulation

non desensitizing

of KA currents

current (Fig 3) mediated

by IDRA compounds

mainly by AMPA receptors (Lerma et al. 1998).

At the E&O value for Kainate we have tested increasing

Even at low concentrations

concentrations

(10 PM, 30 PM) cyclothiazide

current, while ID%%21 starts to produce a signiticative

detected for cyclothiazide

120
110
100
90
80

&

1013

of IDRA 21 and cyclothiazide.

produces a significative

potentiation

of KA

effect at 100 PM. The maximal potentiation

is achieved at 100 uM while that of IDRA 21 is at 1mM.

0
m

+AMPA
+AMPA+NBQX

+AMPA+GYKI

60

50

40

30
20
10

700 i

IDRA 5

CYCLOTHIAZIDE

Fig.6 Bar plot showing the cell viability (expressed as a percetage of control) after treatment with
AMPA alone (empty bar) or together with NBQX (grey bar) or GYKl 53655 (stripped bar) in the
presence of IDRA 2 l( ImM) or IDRAS (1mM) or cyclothiazide (100uM). Asterisks mean signiticative
differences compared to control (Dunnet Test p< 0.01).
Coapplication
intermediate

of

between

IDRA

21 and cyclothiazide

those of

the single drugs,

produces

a potentiation

suggesting

of KA current

an action at a common

that

site within

is
the

AMPA receptor stucture.

IDRA 21 and IDRA 5 have shown
avoidance

test): the first compound

opposite

effects when tested in behavioural

ameliorates

(Zivkovic et al 1994,) and by alprazolam

the cognitive

in momkeys(

experiments

deficit induced by scopolamin

Thompson

(passive
in rats

et al 1995) while IDPA 5 worsens

these deficit (Baraldi, personal observations).

The present results revealed
positive

also in vitro an opposite action of the two compunds,

and IDRA 5 a negative

IDRA 5 slightly decreased

antagonism

the potentiation

could be due to a competition

effect at a different
conversely

modulator

of AMPA-receptor

function.

Futhermore,

for a common site of action although we cannot

site. The latter hypothesis

that other mechanisms

application

of

of KA- current elicited by IDRA 21 (Fig. 5). This partial

is supported

by the marked toxicity of

to that elicited by IDRA 21, was not blocked by AMPAR antagonist

or NBQX suggesting

being IDRA 21 a

rule out an

IDRA 5 that,

such as GYKl 53655

could be involved in producing cell death.

1014

At doses

G. Puia et al.

that are able to modulate AMPA receptor desensitization

toxicity than cyclothiazide

These results confirm

as shown also by Impagniatiello

KA-evoked

The authors also

et al. (1997).

previous data of the effect of IDRA 21 in hyppocampal

1993) but expand the earlier work indicating

potentiating

(100 FM) IDRA 21 shows lower

slices (Bertolino

et al.

a lower potency and higher efficacy than cyclothiazide

in

current.

provided

some evidence

of the common

site of action between

IDRA 21 and

cyclothiazide
Substitution

of an hydrogen

with a chloride in the methyl group in position

double bond in IDRA 21 structure lead to a compound

current and with a diverse mechanism

(IDRA 5) with different

3 and the presence

of a

activity on ISA-evoked

of action in producing neuronal damage.

ANKARCRONA, M. DYPBUKT, J.M. BONFOCO, E., ZHIVOTOVSKY, B. ORRENIUS, S.

LIPTON, S. A. and NICOTERA, P. (1995) Glutamate -induced neuronal death: a succession
necrosis or apoptosis depending on mitocondrial function. Neuron Is: 961-973.

of

BERTOLINO, M. BARALDI, M. PARENTI, C. BRAGHIROLI, D., DIBELLA, M. VICINI, S. and

COSTA, E. (1993): Modulation of AMPA/Kainate
receptors by analogues of Diazoxide and
Cyclothiazide in thin slices of rat hippocampus. Receptor channels 1: 267-278.
CASTILLO, P.E. MALENKA,
R.C. and MCOLL, R. (1997) Kainate
postsynaptic current in hippocampal CA3 neurons. Nature m:182-185.

receptors

mediate

a slow

GALLO, V. CIOTTI, M.T. COLETTI, A. ALOISI, F. and LEVI, G. (1982) Selective release of
glutamate
from cerebellar granule cells differentiating in culture. Proc.Nat1.Acad.Sci.U.S.A.
B :
7919-7923.
HAMILL, O.P. MARTY, A. NEHER, E. SACKMANN, B. SIGWORTH, F. J. (1981): Improved patchclamp techniques for high-resolution
current recording from cell and cell-free membrane patches.
Pflugers Arch =:85-100.
IMPAGNATIELLO,
F. OBERTO, A. LONGONE, P. COSTA, E. and GUIDOTTI, A. (1997) 7chloro-3-methyl-3,4-dihydro-2H-l,2,4-benzo-thiadiazine
S,S-dioxide : a partial modulator of AMPA
receptor desensitization devoid of neurotoxicity. Proc. Natl. Acad. Sci . USA &l :7053-7058.
KISKIN, NI. KRISHTAL, 0. A. TSYNDRENKO, A. Y. (1986): Exitatory aminoacid
hippocampal neurons: kainate fails to desensitize them. Neurosci lett 63: 225-230.

receptors

in

LERMA, J. PATERNAIN, A. V. NARANJO, J. and MELLSTRO, B. (1993): Functional kainate-selective

glutamate receptors in cultured hippocampal neurons, Proc. Nat. AC. Sci. USA al 1688-11692.
LERMA, J. PATERNAIN, A.V. SALVADOR, N. SOMOHANO, F. MORALES, M. and CASADO, M.
(1998) Excitatory aminoacid-activated channels. Ion Channel Pharmacology ,Oxford Press 399-42 1.
MURASE, K. RYU, P.D. RANDIC, M. (1989) Excitatory and inhibiting aminoacids and peptideinduced responses in acutely isolated rat spinal horn neurons. Neurosci. Lett. m: 56-63.

Modulation of KA currents by IDRA compounds

1015

PATNBAU, D.K. and MAYER, M.L. (1990) Structure activity relationships for aminoacid transmitter
candidates acting at N-methyl-D-aspartate and quisqualate receptors. J. Neurosci. ltj: 2385-2399.
PATNEAU, D.K. VYKLICKY, L. and MAYER, M.L. (1993) Hippocampal neurons exhibit
cyclothiazide-sensitive rapidly desensitizing responses to kainate. J.Neurosci. 1s: 3496-3509.
SEEBURG, P.H. (1993) The molecular biology of mammalian glutamate receptor channel. TINS 16: 359.
365.
THOMPSON, D.M. GUIDOTTI, A. DIBELLA M. and. COSTA E. (1995) Idra 21, a congener of
aniracetam, potently abates pharmacologically induced cognitive impairments in patas monkeys .
Proc.Natl.Acad.Sci. USA x:7667-7671.
VIGNES, M. and COLLINGRIDGE, G.L. (1997) The synaptic activation of kainate receptors. Nature
m:179-182.
VYKLICKY, L. PATNEAU, D.K. and MAYER, M.L. (1991) Modulation of exitatory synaptic
transmission by drugs that reduce desensitization. Neuron 1: 971-984.
YAMADA, K.A. and TANG, C.M. (1993) Benzothiadiazines inhibit rapid glutamate receptor
desensitization and enhance glutamatergic synaptic currents. J.Neurosci 1l : 3904-3915.
ZIVKOVIC, I. THOMSON, D.M. BERTOLMO, M. UZUNOV, D. DIBELLA, M. COSTA, E. and
GUIDOTTI, A. (1994) 7-Chloro-3-Methyl-3-4-2H-1,2,4-Benzothiadiazine S,S-Dioxide (IDRA 21): a
benzothiadiazine derivative that enhances cognition by attenuating DL-a-amino-2,3-dihydro-5-methyl-3oxo-4-isoxazolepropanoic acid (AMPA) Receptor desensitization. J Pharmacol. Exp. Ther. 272: 300-309.

Inquiries and reprint requests should be addressed to :

Dr. Giulia Puia

Department of Pharmacology
Universita di Modena
Via Campi, 183
41100 MODENA
ITALY