5-[1′-(2′-N-arylsulfonyl-1′,2′,3′4′-tetra hydroisoquinolyl)]-4,5-dihydro-2(3H)-furanones: Positive allosteric modulators of the GABAA receptor with a new mode of action [3]

Full text of DOI:10.1021/jm001049i

J . Med. Chem. 2000, 43, 4363-4366
4363
5
-[1′-(2′-N-Ar ylsu lfon yl-1′,2′,3′,4′-
tetr a h yd r oisoqu in olyl)]-4,5-d ih yd r o-
(3H)-fu r a n on es: P ositive Alloster ic
also associated with their use. These include amnesia,
tolerance, dependence, alcohol potentiation, and over-
sedation (e.g. when only an anxiolytic effect is desired).
To obtain drugs showing a more restrained pharmaco-
logical profile and/or decreased levels of side effects, a
generally admitted strategy is to develop receptor
subtype-selective ligands for one of the types of allosteric
binding sites. A convincing example of this is zolpidem
which binds to the benzodiazepine recognition site but
2
Mod u la tor s of th e GABAA Recep tor w ith
a New Mod e of Action
†
‡
#
Rodolphe Razet, Urs Thomet, Roman Furtm u¨ ller,
Ang e` le Chiaroni, Erwin Sigel,
Werner Sieghart, and Robert H. Dodd*
†
‡
#
,†
which shows high selectivity for the R1â2γ2 GABAA
receptor subtype and, consequently, predominantly
Institut de Chimie des Substances Naturelles,
Centre National de la Recherche Scientifique,
1198 Gif-sur-Yvette Cedex, France, Department of
12
sedative effects in vivo. The search for subtype-
selective ligands can also be approached from the point
of view of new allosteric modulatory sites. Indeed, the
large number of such sites on the GABAA receptor
suggests that there may in fact be many others. Ligands
for these sites could perhaps modulate the activity of
GABA in more subtle ways than those presently known
9
Pharmacology, University of Bern, Friedb u¨ hlstrasse 49,
CH-3010 Bern, Switzerland, and Section of Biochemical
Psychiatry, Department of Psychiatry, University of Vienna,
A-1090 Vienna, Austria
Received August 4, 2000
(
e.g. as benzodiazepines do compared to barbiturates)9
In tr od u ction . γ-Aminobutyric acid (GABA) is the
major inhibitory neurotransmitter of the central nervous
system. It acts at the GABAA receptor, a pentameric
supramolecular complex which forms a ligand-gated ion
channel controlling chloride ion flux in the neuron.
Binding of GABA to its recognition site on the receptor
has as principal effect the opening of the channel,
allowing chloride ions to flow, leading to inhibition of
neuronal activity. The subunits composing the GABAA
or they may be restricted to only certain receptor
subunit combinations, thereby giving rise to the sought-
after receptor subtype selectivity. The discovery of such
novel allosteric modulatory sites can be accomplished
using classical radioactive ligand displacement studies
in combination with a functional assay, for example, the
electrophysiological measurement of the potentiation of
GABA-elicited currents by a compound.
We have recently reported the synthesis and prelimi-
nary pharmacological evaluation of (()-ROD185, the
N-benzyloxycarbonyl (N-Cbz) derivative of 5-(1-iso-
receptor have been classified as R, â, γ, δ, ꢀ, π, and, more
recently, θ based on sequence homology.1
-6
These
subunits are also generally present in the form of
several isoforms (e.g. R1, R2, etc.), leading to a poten-
tially high level of receptor heterogeneity, though the
most common receptor composition has been found to
consist of R1, â2, and γ2 subunits.7
13
quinolyl)furan-2-one (1). This novel compound was
shown by in vitro radioactive ligand displacement
studies using rat brain tissue to be a good ligand of the
benzodiazepine recognition site of the GABAA receptor
,8
(
IC50 ) 60 nM in cerebellum). In electrophysiological
In addition to the heterogeneity of the GABAA recep-
tor system, each receptor also has a variety of allosteric
sites, which, upon binding of their specific ligands, leads
to modulation of GABA activity.9 Thus, compounds
which stimulate GABA-induced chloride ion currents
are referred to as positive allosteric modulators. These
include the therapeutically useful benzodiazepines which
bind to the benzodiazepine recognition site. The latter
in fact also binds a wide variety of compounds structur-
ally unrelated to benzodiazepines and which have the
same potentiating effect on GABA.10 Other positive
allosteric binding sites of the GABAA receptor include
the barbiturate, neurosteroid, and loreclezole sites. On
the other hand, negative allosteric modulation of the
GABAA receptor (that is, inhibition of the neuroinhibi-
tory action of GABA) is also possible with certain ligands
as, for example, certain â-carbolines which bind to the
benzodiazepine recognition site.1
studies, ROD185 strongly potentiated GABA-elicited
currents in cells expressing the R1â2γ2 receptor subtype
(351% at 100 µM). We now show that replacement of
the N-Cbz group of ROD185 by an arylsulfonyl group
e.g. 2) leads to compounds which are still strong
(
positive allosteric modulators of the GABAA receptor but
which do so mainly by acting in a novel fashion with
this receptor.
0,11
Resu lts a n d Discu ssion . The N-sulfonyl derivatives
Positive allosteric modulators of the GABA receptor
generally display, to varying degrees, anticonvulsant,
anxiolytic, sedative-hypnotic, anesthetic, and muscle-
relaxant activities. Certain undesirable side effects are
2
were prepared by exploiting the vinylogous Mannich
1
4
reaction previously used for the synthesis of ROD185.
Thus, treatment of 3,4-dihydroisoquinoline 3 with ben-
zyl chloroformate in acetonitrile for 15 min gave the
iminium salt 4 which was reacted in situ with 2-(tert-
*
To whom correspondence should be addressed. Tel: 33-1-69824594.
Fax: 33-1-69077247. E-mail: Robert.Dodd@icsn.cnrs-gif.fr.
15
butyldimethylsiloxy)furan (5) to afford a mixture of
†
Institut de Chimie des Substances Naturelles.
University of Bern.
University of Vienna.
‡
the threo and erythro diastereomers 6 and 7, respec-
tively (Scheme 1). Overall yield was 66%, with a threo:
#
1
0.1021/jm001049i CCC: $19.00 © 2000 American Chemical Society
Published on Web 10/28/2000

4
364 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23
Communications to the Editor
Sch em e 1
this ligand at 100 µM (50% and 60%, respectively). In
the case of [ H]muscimol, either no displacement (2d,f,h )
3
or a 5-15% stimulation of binding was observed.
Moreover, binding to the benzodiazepine recognition site
was considerably diminished by replacement of the
benzylcarbamate group of ROD185 by an arylsulfonyl
group. Thus, 100 µM benzenesulfonyl derivative 2a was
3
able to displace only 74% and 66% of [ H]flunitrazepam
in the cerebellum and forebrain, respectively, whereas
ROD185 had IC50 values of 60 and 160 nM in the same
tissues. All the other N-sulfonyl compounds synthesized
3
were even weaker displacers of [ H]flunitrazepam than
2
a .
In view of this loss of binding affinity for the benzo-
diazepine recognition site, it was very surprising to
observe in the electrophysiological studies that the
arylsulfonyl derivatives were in most cases more potent
stimulators of GABA-elicited currents in Xenopus
oocytes than ROD185. In particular, the N-p-toluene-
sulfonyl analogue 2c stimulated currents by 579 ( 61%
at 100 µM (mean of three experiments ( SEM unless
otherwise stated), compared to 351 ( 10% for ROD185
at the same concentration. While the N-benzenesulfonyl
and N-p-chlorobenzenesulfonyl derivatives 2a ,b were
slightly less active in this assay (423 ( 32% and 479 (
1
2% stimulation, respectively), both were still signifi-
cantly more active than ROD185. Replacement of the
p-methyl group of 2c by a p-trifluoromethyl group (2d )
led to considerable loss of stimulatory activity (90 ( 8%
stimulation), as did suppression of the aromatic ring as
in the N,N-dimethylaminosulfonamide 2e (62 ( 7%
stimulation) or the methylsulfonamide 2f (32 ( 4%
stimulation). Aryl groups other than phenyl were also
active. Thus, though the naphthylsulfonyl derivative 2g
could not be assayed at 100 µM because of its insolubil-
ity, it demonstrated significant current stimulation even
at 10 µM concentration (118 ( 24%). The thiophene
derivative 2h was in turn as active as ROD185 in
stimulating GABA currents, though again it displayed
a much weaker interaction with the benzodiazepine
recognition site. None of the N-sulfonyl compounds
produced significant currents in the absence of GABA,
indicating that they are not acting as GABA agonists.
1
erythro ratio of 3:2 based on the H NMR spectrum of
the mixture. Purification of the mixture on silica gel
heptane-ethyl acetate, 1:1) allowed isolation of pure
(
isomers 6 and 7. The relative configurations of the two
asymmetric centers of 6 and 7 were established by
comparison of their 13C NMR spectra and by X-ray
crystallography of a derivative of 6 (see below). Because
our previous studies showed that the threo derivatives
of isoquinolylfuranones were generally more potent than
the erythro isomers in stimulating GABA-elicited cur-
rents, compound 6 was chosen for further transforma-
tions. Hydrogenolysis of the N-Cbz group of 6 using
palladium-on-carbon in ethyl acetate led to simulta-
neous reduction of the lactone double bond to afford
compound 8 in quantitative yield. The latter was
efficiently N-sulfonylated by reaction with a variety of
sulfonyl chlorides in the presence of triethylamine
providing the target compounds 2a -h (Table 1).
Compounds 2a -h were first evaluated in vitro for
their capacity to bind to the GABA, benzodiazepine, and
16
1
3
The question thus remained at this point as to which
of the various allosteric modulatory sites of the GABAA
receptor is involved in the stimulation of GABA currents
by these arylsulfonyl derivatives. Further pharmaco-
logical profiling was thus undertaken on the most active
of these compounds, the N-tosyl derivative 2c (code
named ROD188). However, because ROD188 (as well
as all the compounds prepared via Scheme 1) is a
racemic mixture, it was decided at this point to isolate
each enantiomer of ROD188 and to study only the most
active one. Separation was accomplished by subjecting
ROD188 to HPLC on a chiral OD column. The pure
enantiomers were obtained as solids and (-)-ROD188
was crystallized in a form suitable for X-ray analysis.
As shown in Figure 1, (-)-ROD188 corresponds to the
9
picrotoxin (a chloride channel blocker) recognition sites
of the GABAA receptor by measuring their potency in
displacing the appropriate specific radioligand (respec-
3
3
35
tively [ H]muscimol, [ H]flunitrazepam, and [ S]-tert-
3
5
17,18
butylbicyclophosphorothionate ([ S]TBPS)).
Table 1) are given as percent displacement of the
radioactive ligand by 100 µM compound in whole rat
Results
(
3
35
brain tissue ([ H])muscimol, [ S]TBPS) or in cerebellum
and forebrain tissue ([ H]flunitrazepam). Compounds
a -h were also evaluated electrophysiologically for
their capacity to stimulate GABA-elicited currents using
3
2
2
1
SS configuration of the two chiral centers. That of (+)-
ROD188 must therefore correspond to the RR configu-
ration.
Xenopus laevis oocytes expressing rat brain recombinant
R1â2γ2 GABAA receptors.1
9,20
As in the case of ROD185, the N-sulfonyl analogues
3
5
showed at best only weak displacement of [ S]TBPS.
Only the benzenesulfonyl and p-chlorobenzenesulfonyl
derivatives (2a ,b) showed significant displacement of
Comparison of the current stimulations produced by
each enantiomer showed that (+)-ROD188 was consid-
erably more potent than (-)-ROD188 (859 ( 43% and

Communications to the Editor
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23 4365
Ta ble 1. Displacement of Radioactive Ligands from Rat Brain Membranes and Effect on GABA-Elicited Currents in Xenopus Oocytes
for Compounds 2a -h
a
Determined in rat brain tissue as previously described.17,18 Values are given as percent displacement of the radioactive ligand by
b
each substance. Experiments were performed at least three times in triplicate. ce refers to rat cerebellum; fo refers to rat forebrain
c
tissue (whole brain minus cerebellum). Determined electrophysiologically in X. laevis oocytes expressing R1â2γ2 GABAA receptors as
previously described.
than 100 µM, as indicated.
1
9,20
Average of three assays ( SEM. d Values in parentheses refer to assays performed at a concentration other
1
84 ( 25%, respectively). Threshold of stimulation was
receptors, (+)-ROD188 (30 µM) was found to strongly
stimulate currents elicited by pentobarbital (100 µM)
and 5R-pregnan-3R-ol-20-one (3 µM) (588 ( 218% and
644 ( 47%, respectively) suggesting that the binding
sites of these two ligands are different from that of (+)-
ROD188. Furthermore, while an R1â2 receptor combi-
nation responds to current stimulation by loreclezole,
the point-mutated form R1â2Ν265S is only weakly
about 1 µM for (+)-ROD188. Both compounds showed
approximately the same weak affinities for the benzo-
diazepine recognition sites (<70% displacement of [ H]-
3
flunitrazepam at 100 µM), consistent with the hypoth-
esis that current stimulation is not principally due to
interaction with these sites. This was furthermore
corroborated by the observation that at a concentration
that would normally completely inhibit current stimula-
tions by diazepam, co-administration of the benzodi-
azepine antagonist Ro15-1788 (1 µM) resulted in inhi-
bition of only about 34% of the current stimulation
produced by (+)-ROD188 (20 µM). Thus, stimulation
was reduced from 207 ( 42% to 136 ( 20% (n ) 5).
The possibility that (+)-ROD188 interacts with other
known allosteric binding sites on the GABAA receptor
was then investigated. Both barbiturates (e.g. pento-
barbital)2 and steroids (e.g. 5R-pregnan-3R-ol-20-one)23
modulate GABAA receptor currents and, moreover,
activate the chloride ion channel in the absence of
GABA. In the oocyte preparation expressing R1â2γ2
2
4,25
responsive to loreclezole.
However, both the wild-
type and mutated receptor respond almost equally well
to (+)-ROD188, current stimulation being 859 ( 75%
in the first case and 600 ( 46% (n ) 4) in the second.
The mode of action of (+)-ROD188 thus also appears to
be distinct from that of loreclezole. Finally, the current
stimulation produced by R-EMTBL, which acts via a
2
6
postulated γ-butyrolactone binding site, was also
decreased by the same mutation (unpublished results)
indicating a mode of action different from that of (+)-
ROD188.
2
Su m m a r y. In conclusion, N-arylsulfonylisoquinolyl-
furanones of general structure 2 are a new class of

4
366 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23
Communications to the Editor
(
5) Bonnert, T. P.; McKernan, R. M.; Farrar, S.; Le Bourdell e` s, B.;
Heavens, R. P.; Smith, D. W.; Hewson, L.; Rigby, M. R.;
Sirinathsinghji, D. J . S.; Brown, N.; Wafford, K. A.; Whiting, P.
J . θ, a novel γ-aminobutyric acid type A receptor subunit. Proc.
Natl. Acad. Sci. U.S.A. 1999, 96, 9891-9896.
(6) Barnard, E. A.; Skolnick, P.; Olsen, R. W.; M o¨ hler, H.; Sieghart,
W.; Biggio, G.; Braestrup, C.; Bateson, A. N.; Langer, S. Z.
Subtypes of γ-aminobutyric acid
basis of subunit structure and receptor function. Pharmacol. Rev.
998, 50, 291-313.
7) McKernan, R. M.; Whiting, P. J . Which GABA
really occur in the brain. Trends Neurosci. 1996, 19, 139-143.
8) Sieghart, W.; Fuchs, K.; Tretter, V.; Ebert, V.; J echlinger, M.;
H o¨ ger, H.; Adamiker, D. Structure and subunit composition of
A
receptors: classification on the
1
(
(
A
receptor subtypes
GABA
9) For a review, see: Sieghart, W. Structure and pharmacology of
γ-aminobutyric acid receptor subtypes. Pharmacol. Rev. 1995,
7, 181-233.
A
receptors. Neurochem. Int. 1999, 34, 379-385.
(
A
4
(
10) Sigel, E.; Buhr, A. The benzodiazepine binding site of GABA
A
receptors. Trends Pharmacol. Sci. 1997, 18, 425-429.
11) Macdonald, R. L.; Twyman, R. E. Kinetic properties and regula-
(
A
tion of GABA receptor channels. In Ion Channels; Narahashi,
T., Ed.; Plenum Press: New York, 1992; Vol. 3, pp 315-343.
12) Dennis, T.; Dubois, A.; Benavides, J .; Scatton, B. Distribution
(
of central ω
1 1 2 2
(benzodiazepine ) and ω (benzodiazepine ) receptor
subtypes in the monkey and human brain. An autoradiographic
3
3
1
study with [ H]flunitrazepam and the ω selective ligand [ H]-
zolpidem. J . Pharmacol. Exp. Ther. 1988, 247, 309-322.
13) Razet, R.; Thomet, U.; Furtm u¨ ller, R.; J ursky, F.; Sigel, E.;
Sieghart, W.; Dodd, R. H. Use of bicuculline, a GABA antagonist,
as a template for the development of a new class of ligands
(
A
showing positive allosteric modulation of the GABA receptor.
Bioorg. Med. Chem. Lett., in press.
(
14) For a review of vinylogous Mannich reactions using 2-trialkyl-
silyloxyfurans, see: Casiraghi, G.; Rassu, G. Furan-, pyrrole-,
and thiophene-based siloxydienes for syntheses of densely func-
tionalized homochiral compounds. Synthesis 1995, 607-626.
15) Rassu, G.; Zanardi, F.; Battistini, L.; Gaetani, E.; Casiraghi, G.
Expeditious syntheses of sugar-modified nucleosides and collec-
tions thereof exploiting furan-, pyrrole-, and thiophene-based
siloxy dienes. J . Med. Chem. 1997, 40, 168-180.
F igu r e 1. ORTEP drawing of (-)-ROD188 ((-)-2c) showing
the 5S,1′S configuration.
(
positive allosteric modulators of the GABAA receptor.
Preliminary experiments conducted on the most potent
of the compounds synthesized, (+)-ROD188 ((+)-2c),
showed that while this compound presents weak re-
sidual affinity for the benzodiazepine recognition site,
binding to this site cannot entirely account for the
strong current stimulation observed. Furthermore, it
appears unlikely that (+)-ROD188 interacts with the
barbiturate, steroid, picrotoxin, loreclezole, γ-butyro-
lactone, or GABA binding sites. It is thus concluded that
(16) Martin, S. F.; Corbett, J . W. The stereochemical course of
nucleophilic additions of 2-trialkylsiloxyfurans to cyclic N-
acyliminium ions. Synthesis 1992, 55-57.
(
17) Sieghart, W.; Schuster, A. Affinity of various ligands for ben-
zodiazepine receptors in rat cerebellum and hippocampus.
Biochem. Pharmacol. 1984, 33, 4033-4038.
(18) Zezula, J .; Slany, A.; Sieghart, W. Interaction of allosteric ligands
A
with GABA receptors containing one, two or three different
subunits. Eur. J . Pharmacol. 1996, 301, 207-214.
(19) Sigel, E. Properties of single sodium channels translated by
Xenopus oocytes after injection with messenger ribonucleic acid.
J . Physiol. 1987, 386, 73-90.
(
+)-ROD188 and its analogues interact with the GABAA
2
7
receptor by binding to a new allosteric modulatory site
on this receptor in addition to binding to the benzodi-
azepine recognition site. Confirmation of this possibility
will be the object of further study.
(20) Sigel, E.; Baur, R.; Trube, G.; M o¨ hler, H.; Malherbe, P. The effect
A
of subunit composition of rat brain GABA receptors on channel
function. Neuron 1990, 5, 703-711.
(
21) X-ray crystallographic data for compound (-)-2c (i.e. (-)-
ROD188) have been deposited at the Cambridge Crystal-
lographic Data Centre, University Chemical Laboratory, Lens-
field Road, Cambridge CB2 1EW, U.K. (deposition number
CCDC 149408).
Ack n ow led gm en t. We thank the European Union
(
Grants B104-CT96-0585 and BBW 96.0010) and
(
(
(
22) Inomata, N.; Tokutomi, N.; Oyama, Y.; Akaike, N. Intracellular
the Swiss National Science Foundation (Grant 3100-
picrotoxin blocks pentobarbital-gated Cl- conductance. Neurosci.
0
53599.98/1) for financial support.
Res. 1988, 6, 72-75.
23) Majewska, M. D. Neurosteroids: endogenous bimodal modulators
A
of the GABA receptor. Mechanism of action and physiological
significance. Prog. Neurobiol. 1992, 38, 379-395.
Su p p or tin g In for m a tion Ava ila ble: Full experimental
details and spectroscopic data for compounds 2a -h and 6-8.
This material is available free of charge on the Internet at
http://pubs.acs.org.
24) Wingrove, P. B.; Wafford, K. A.; Bain, C.; Whiting, P. J . The
modulatory action of loreclezole at the γ-aminobutyric acid type
A receptor is determined by a single amino acid in the â
-subunit. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 4569-4573.
25) Stevenson, A.; Wingrove, P. B.; Whiting, P. J .; Wafford, K. A.
â-carboline γ-aminobutyric acid receptor inverse agonists modu-
2
- and
â
3
(
Refer en ces
A
(
1) Burt, D. R.; Kamatchi, G. L. GABA
pharmacology to molecular biology. FASEB J . 1991, 5, 2916-
923.
2) Dunn, S. M. J .; Bateson, A. N.; Martin, I. L. Molecular biology
of the GABA receptor. Int. Rev. Neurobiol. 1994, 36, 51-96.
3) Hedblom, E.; Kirkness, E. F. A novel class of GABA receptor
subunit in tissues of the reproductive system. J . Biol. Chem.
A
receptor subtypes: from
late γ-aminobutyric acid via the loreclezole binding site as well
as the benzodiazepine site. Mol. Pharmacol. 1995, 48, 965-969.
2
(26) Williams, K. L.; Tucker, J . B.; White, G.; Weiss, D. S.; Ferren-
delli, J . A.; Covey, D. F.; Krause, J . E.; Rothman, S. M. Lactone
modulation of the γ-aminobutyric acid A receptor: evidence for
a positive modulatory site. Mol. Pharmacol. 1997, 52, 114-119.
(27) A complete description of the electrophysiological properties of
(+)-ROD188 will appear elsewhere.
(
(
A
A
1
997, 272, 15346-15350.
A
(4) Macdonald, R. L.; Olsen, R. W. GABA receptor channels. Annu.
Rev. Neurosci. 1994, 17, 569-602.
J M001049I