Aza-THIP and related analogues of THIP as GABAC antagonists

Article abstract of DOI:10.1016/j.bmc.2003.09.016

The potency of a series of eight compounds structurally related with 4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol (THIP), a potent GABA _A partial agonist exhibiting GABA_C ρ₁ antagonist effect (K_i=25 μM), was d

Full text of DOI:10.1016/j.bmc.2003.09.016

Bioorganic & Medicinal Chemistry 11 (2003) 4891–4896
Aza-THIP and Related Analogues of THIP as
GABAC Antagonists
Dorte Krehan,a Bente Frølund,a Bjarke Ebert,b Birgitte Nielsen,a
Povl Krogsgaard-Larsen,a,* Graham A. R. Johnstonc
and Mary Chebibd
aCentre for Drug Design and Transport, Department of Medicinal Chemistry, The Danish University of Pharmaceutical Sciences,
DK 2100 Copenhagen, Denmark
bDepartment of Neurobiology, H. LundbeckA/S, DK 2500 Valby, Denmark
cAdrien Albert Laboratory of Medicinal Chemistry, Department of Pharmacology, Sydney, NSW 2006, Australia
dDepartment of Pharmacy, The University of Sydney, Sydney, NSW 2006, Australia
Received 10 January 2003; accepted 15 September 2003
Abstract—The potency of a series of eight compounds structurally related with 4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol
(THIP), a potent GABAA partial agonist exhibiting GABAC r1 antagonist effect (Ki=25 mM), was determined electro-
physiologically using homomeric human GABAC r1 receptors expressed in Xenopus oocytes. Protolytic properties (pKa values for
the acidic bioisosteric groups) and the presence of steric bulk in the molecules appear to be structural parameters of importance for
blockade of the GABAC r1 receptor. Within this series of moderately potent GABAC antagonists, only 4,5,6,7-tetra-
hydropyrazolo[5,4-c]pyridin-3-ol (Aza-THIP) does not interact detectably with GABAA receptors, and Aza-THIP has the potential
of being a useful tool for molecular and behavioural pharmacological studies.
# 2003 Elsevier Ltd. All rights reserved.
Introduction
The GABA system is believed to be involved in several
neurological and psychiatric disorders, and in particular
the GABAC receptors appear to be involved in a num-
ber of inherited diseases of the eye.2,4,5 These aspects
have focused interest on the GABAC receptor as a novel
therapeutic target making the search for ligands as
experimental tools and potential drugs important. In
addition to the reference antagonist ligand for the
GABAC receptors, (1,2,3,6-tetrahydropyridine-4-yl)
methylphosphinic acid (TPMPA),6 the GABAA partial
4-Aminobutyric acid (GABA), the major inhibitory
neurotransmitter in the central nervous system (CNS),
inhibits neuronal activity through two classes of GABA
receptors, ionotropic and metabotropic GABA recep-
tors. Based on molecular biology and pharmacology,
the ionotropic GABA receptors can be devided into two
groups, GABAA and GABAC receptors.1,2 The iono-
tropic GABA receptors are ligand gated ion channels
conducting chloride. GABAA receptors are hetero-
oligomeric receptors being assembled of different sub-
units and are believed to have a pentameric structure
made up of five subunits.3 GABAC receptors, however,
form homooligomeric receptors from the r subunits,
although some evidence for heterooligomeric GABAC
receptors exists.1,2 To date, several subunits have been
identified that make up the GABAA receptors including
agonist
4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol
(THIP)7,8 (Fig. 1) has been shown to be a moderately
potent competitive GABAC antagonist.9
In this paper we describe the pharmacology of eight
compounds structurally related to THIP (Fig. 1) at func-
tional GABAC r1 receptors expressed in Xenopus oocytes
to elucidate the structural determinants for blockade of
the GABAC receptors. Anew and improved synthetic
route was developed for the synthesis of 4,5,6,7-tetra-
hydroisothiazolo[5,4-c]pyridin-3-ol (Thio-THIP) (Scheme
1) making it possible to extend the pharmacological
characterization of the series of THIP analogues.
a
1À6, b1À4, g1À4, d, e, p, while the GABAC receptors are
1,2
built only by r1À3
.
*Corresponding author. Tel.: +45-353-06511; fax: +45-353-06040;
e-mail: ano@dfh.dk (P. Krogsgaard-Larsen)
0968-0896/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2003.09.016

4892
D. Krehan et al. / Bioorg. Med. Chem. 11 (2003) 4891–4896
Results
pound 4 was hydrolized to the corresponding thiol,
oxidized to the disulphide which was then cyclized using
sulphuryl chloride, resulting in protected Thio-THIP
(5). Compound 5 was deprotected with 33% hydrogen
bromide in glacial acetic acid giving Thio-THIP in 85%
yield.
Briefly, isonicotinamide was alkylated with benzyl-
bromide giving a pyridinium salt (1) followed by reduc-
tion with sodium borohydride to provide 2 in good yield
(Scheme 1). After reduction the benzyl group was
replaced by the methoxycarbonyl protecting group (3).
Sulphur was introduced by a Michael addition of
thioacetic acid giving thioester 4 in good yield. Com-
The overall yield based on commercially available iso-
nicotinamide was markedly improved and involved
fewer steps as compared with the previously described
synthesis of Thio-THIP.10
The activities of the analogues of THIP, including the
synthesized Thio-THIP, on functional GABAC r1
receptors expressed in Xenopus oocytes and their affi-
nities for GABAA receptors in rat brain membrane pre-
parations using either [3H]muscimol or [3H]GABA as
the radioligand, are summarized in Table 1. The com-
pounds under study were shown to be either inactive or
competitive antagonists at functional r1 receptors
expressed in Xenopus oocytes.
In this study the reference compound TPMPAwas
shown to be a potent r1 receptor antagonist with a Ki
value of 3.2 mM, whereas THIP was markedly weaker
as an antagonist at r1 receptors (Ki=25 mM). Thio-
THIP, Aza-THIP and Iso-THIP showed activity at
GABAC receptors ranging from weak to potent antag-
onism. Iso-THIP was a potent antagonist with a Ki
value of 7.9 mM only some three times weaker than
TPMPA, whereas Thio-THIP was a weak antagonist
(Ki=91 mM). Aza-THIP was approximately equipotent
with THIP having a Ki of 31 mM. The shift of the
GABA concentration–response curve by Aza-THIP is
shown in Figure 2. Both Iso-THIP and Thio-THIP
show affinity for the GABAA receptors, whereas Aza-
THIP shows no detectable GABAA receptor affinity
(Table 1), hence Aza-THIP was shown to be a selective
GABAC antagonist. The inhibition of a 3 mM GABA
response by 300 mM Aza-THIP is shown in Figure 3.
The two methylated Aza-THIP analogues showed no
activity at GABAC r1 receptors and no affinity for
GABAA receptors (Table 1).
Figure 1. Structures of GABA and a number of heterocyclic analo-
gues of GABA characterized as GABAC antagonists in the present
study.
THAZ, a very weak glycine receptor antagonist,11 was a
moderately potent antagonist (Ki=23 mM) at r1 recep-
tors, whereas the monocyclic analogue, AEMI, of
THAZ was shown to be a weak antagonist (Ki=142
mM). In contrary to AEMI, THAZ showed some affi-
nity for GABAA receptors. THIA, an isomer of THAZ
and also a glycine receptor antagonist,11 shows no affi-
nity for the GABAA receptors and showed no activity at
GABAC r1 receptors.
Discussion
THIP is a conformationally restricted analogue of
GABA, which adopts a well-defined conformation. Of
the eight compounds tested in this study, Iso-THIP,
Aza-THIP and Thio-THIP are structurally and con-
formationally closely related to THIP, adopting essen-
tially rigid conformations. All of these compounds are
Scheme 1.

D. Krehan et al. / Bioorg. Med. Chem. 11 (2003) 4891–4896
4893
Table 1. Effects of TPMPA, THIP and a number of THIP analogues at GABAC r1 receptors and on GABAA receptor bindinga
Compd
PKa values
Human r1 receptors Ki
(mM) (pKiÆSEM)
GABAA receptor binding Ki
(mM) (pKiÆSEM)
TPMPA—
THIP
THAZ
AEMI
THIA4.6; 8.6
Iso-THIP
Thio-THIP
Aza-THIP
1-Me-Aza-THIP
2-Me-Aza-THIP
3.2 (5.49
4.4; 8.5b
4.8; 9.2c
Æ0.11)
25 (4.60Æ0.06)
23 (4.63Æ0.04)
142 (3.87Æ0.07)
NAe
39 (4.40Æ0.01)
0.16 (6.79Æ0.03)
16 (4.78Æ0.04)
>100
5.1; 10.4d
b
>100
3.0, 9.1f
6.1; 8.5c
6.3; 9.9f
7.1; 9.2f
5.8; 9.8f
7.9 (5.10Æ0.06)
91 (4.04Æ0.03)
31 (4.51Æ0.02)
NAe
90 (4.05Æ0.03)
52 (4.29Æ0.04)
>100
>100
>100
NAe
Ki values are calculated from pKi values found in brackets. The pKi values represent mean ÆSEM of at least three experiments.
aStandard [3H]muscimol binding on rat brain synaptic membranes determined using a method described previously.18
bData from Krogsgaard-Larsen (1977).17
cData from Krogsgaard-Larsen et al. (1983).10
dData from Hjeds and Krogsgaard-Larsen (1976).16
eNAfor no activity at 300 mM with or without 1 mM of GABA.
fData from Krogsgaard-Larsen and Roldskov-Christiansen (1979).14
Figure 3. At homooligomeric r1 receptors expressed in Xenopus
oocytes, GABA (3 mM) (duration indicated by filled bar) activates the
receptor while Aza-THIP (300 mM) does not activate the receptor
(duration indicated by unfilled bar). However, when Aza-THIP (300
mM) is co-applied with GABA (3 mM), the GABA response is reduced.
Figure 2. Concentration–response curves for GABA and GABA in
the presence of 300 mM Aza-THIP in Xenopus oocytes injected with r1.
Data points are mean valuesÆstandard errors from three individual
oocytes. Curves were fitted using the equation described in ref 16.
antagonists at the GABAC r1 receptor and most likely
bind to the receptor in a manner similar to that of
THIP.
proposed effect of steric bulk is in agreement with the
pharmacophore model proposed by Chebib et al.,13
where the binding site at GABAC receptors is envisaged
to be narrow. This implies that larger substituents may
interact unfavorably with the binding site. Furthermore,
the different pKa values of THIP and Thio-THIP sug-
gest, that a larger fraction of the molecules of the latter
compound at physiological pH contains a nonionized 3-
isothiazolol group, assumed to be pharmacologically
inactive.
Within this series of GABAC antagonists, Iso-THIP is
the most potent being four times more potent than
THIP. Aza-THIP was approximately equipotent with
THIP at the GABAC r1 receptor and shows no affinity
for the GABAA receptor (Ki>100 mM). This indicates
that Aza-THIP is selective for GABAC r1 receptors.
Thio - THIP was less potent at GABAC r1
receptors, being three times weaker than THIP. The
lower activity of Thio-THIP may be due to the intro-
duction of the sulphur atom in the 3-isothiazolol ring
representing additional steric bulk as compared with
oxygen in the 3-isoxazolol ring of THIP, as elucidated
crystallographically,12 and to the lower acidic character
of the 3-isothiazolol ring (pKa 6.1) as compared to the
3-isoxazolol ring of THIP (pKa 4.4) (Table 1). This
Neither of the two methylated analogues of Aza-THIP
show activity at GABAC r1 receptors, and the lack of
activity of 1-Me-Aza-THIP and 2-Me-Aza-THIP may
be explained by steric interaction and by low acidic
character of the 3-pyrazolol nuclei of these two com-
pounds (pKa 7.1 and 5.8, respectively) (Table 1). In
these two compounds, the methyl group may protrude
into an unfavorable area in the receptor binding pocket

4894
D. Krehan et al. / Bioorg. Med. Chem. 11 (2003) 4891–4896
in agreement with the receptor model proposed by
Chebib et al.13
3-ol (THAZ),15 4-(2-aminoethyl)-5-methylisoxazol-3-ol
(AEMI),16 THIP17 and 5,6,7,8 - tetrahydro - 4H - iso-
xazolo[5,4-c]azepin-3-ol (THIA).17
The homologue of THIP, THAZ, contains a larger
seven membered ring and is therefore a more flexible
molecule. THAZ is a moderately potent antagonist
(Ki=23 mM) at GABAC r1 receptors. Being equipotent
with THIP at this receptor, THAZ may be able to attain
a conformation similar to that of THIP in order to fit
into the GABAC binding pocket. In contrast, AEMI,
which is an open ring analogue of THAZ, is markedly
more flexible, and a conformation of AEMI comparable
with those of THIP and THAZ is unlikely to be ener-
getically favourable. Additionally, the methyl group in
the 5-position of the 3-isoxazolol ring of AEMI may
protrude into an area of steric hindrance causing AEMI
to be some seven times weaker than THAZ (Table 1).
Melting points were determined in open capillary
tubes and are corrected. Elemental analyses were per-
formed at Analytical Research Department, H. Lund-
beck A/S, Denmark, or by Mr. J. Theiner,
Department of Physical Chemistry, University of
Vienna, Austria, and are within Æ0.4% of the calcu-
lated values unless otherwise stated. Nuclear Magnetic
Resonance Spectra (NMR) were recorded on
300 MHz Varian Gemini spectrometer.
a
4-Aminocarbonyl-1-benzylpyridiniumbromide (1). To a
solution of isonicotinamide (10.0 g, 81.9 mmoles) in
ethanol (250 mL) was added benzyl bromide (29.2 mL,
245.6 mmoles) and the reaction mixture was refluxed
(90ꢀ C) for 4 h. After cooling, diethylether (250 mL) was
added, and the reaction mixture was then placed at 4ꢀ C
overnight for crystallization to provide crude 1 as white
crystals (19.3 g, 80%). An analytical sample was
recrystallized (methanol): mp 243.5–246.0ꢀ C (decom-
THIA, an isomer of THAZ (Fig. 1) shows no activity at
GABAC r1 receptors and no affinity for the GABAA
receptors. THIAmay not be able to adopt a conform-
ation where the amino group is in the right position to
interact with the receptor.
1
posed); H NMR (D2O) d 8.93 (2H, d, J=6.6 Hz), 8.19
(2H, d, J=6.6 Hz), 7.35 (5H, s), 5.73 (2H, s), 4.65 (3H,
s); 13C NMR (D2O) d 166.62, 148.65, 145.47, 132.15,
130.11, 129.62, 129.30, 126.45, 66.33. Analysis
(C13H13BrN2O) C, H, Br, N.
Conclusion
TPMPAhas so far been the most useful antagonist for
studying GABAC receptors. However, Aza-THIP,
which has now been shown to be a specific competitive
GABAC antagonist, may be a supplement to TPMPAas
a reference GABAC compound. TPMPAshows strong
antagonist activity at GABAC receptors and, in contrast
to Aza-THIP, weak activity at both GABAA (Table 1)
and metabotropic GABAB receptors.6
1-Benzyl-1,2,3,6-tetrahydropyridine-4-carboxylic
acid
amide (2). Asolution of crude 1 (19.3 g, ca. 65.8
mmoles) in methanol (350 mL) was cooled to 0ꢀ C.
Sodium borohydride (2.62 g, 69.1 mmoles) was added
over 3 h and the mixture was then allowed to stir at
0ꢀ C for 30 min and at 25ꢀ C for 16 h. The mixture
was evaporated in vacuo and the residue was extrac-
ted with CH2Cl2. The combined organic phases were
dried (MgSO4) and evaporated in vacuo to give crude
2 (8.87 g, 62%). An analytical sample was dissolved
in 2 M hydrochloric acid, evaporated in vacuo and
crystallized (methanol) to provide analytically pure
hydrochloride of 2: mp 236.0–238.0ꢀ C (decomposed);
1H NMR (D2O) d 7.34 (5H, m), 6.38–6.35 (1H, m),
4.23 (2H, s), 3.74–3.67 (2H, m), 3.50 and 3.07 (1H,
two broad signals), 2.50–2.43 (2H, m); 13C NMR
Using a series of compounds related to THIP, we have
shown that only minor structural changes affect their
potency as GABAC antagonists. The acidic character of
the 3-isoxazolol ring of Iso-THIP (pKa 3.0) is more
pronounced than that of the 3-isoxazolol ring of THIP
(pKa 4.4) (Table 1), suggesting that this difference in
potency may be explained by a stronger electrostatic
interaction of Iso-THIP than of THIP with the receptor.
The structural and protolytic similarity between THIP
and Aza-THIP suggests that the latter compound, like
THIP, is capable of penetrating the blood-brain-barrier.
Thus, although Aza-THIP is only moderately potent as
a GABAC antagonist, it may be a useful compound for
behavioural pharmacological studies.
(D2O) d 170.50, 131.25, 130.25, 130.13, 129.38,
128,75, 126.00, 59.00, 49.38, 48.00, 21.25. Analysis
(C13H17ClN2O) C, H, Cl, N.
4-Aminocarbonyl-1,2,3,6-tetrahydropyridine-1-carboxylic
acid methyl ester (3). Methyl chloroformate (6.13 mL,
79.7 mmoles) was added through a condenser to crude 2
(8.62 g, ca. 39.9 mmoles). The mixture was stirred for 7
h, evaporated in vacuo and extracted continuously with
CH2Cl2 for 23 h. The CH2Cl2 phase was dried (MgSO4),
evaporated in vacuo and the residue recrystallized (eth-
anol–diethylether) to give 3 (3.22 g, 44%). An analytical
sample was recrystallized (ethyl acetate): mp
Experimental
Chemistry
The following compounds were synthesized by published
procedures: 4,5,6,7-tetrahydropyrazolo[5,4-c]pyridin-3-ol
(Aza-THIP),14 1-methyl-4,5,6,7-tetrahydropyrazolo[5,4-
c]pyridin-3-ol (1-Me-Aza-THIP),14 2-methyl-4,5,6,7-tet-
rahydropyrazolo[5,4-c]pyridin-3-ol (2-Me-Aza-THIP),14
4,5,6,7 - tetrahydroisoxazolo[3,4 - c]pyridin - 3 - ol (Iso-
THIP),14 5,6,7,8-tetrahydro-4H-isoxazolo[4,5-d]azepin-
ꢀ
1
128.0–130.5 C; H NMR (CDCl3) d 6.60 (1H, s), 6.07
(2H, broad s), 4.19–4.04 (2H, m), 3.73 (3H, s), 3.65–3.49
(2H, m), 2.47–2.31 (2H, m); 13C NMR (CDCl3) d
169.00, 156.60, 131.50, 130.50, 52.85, 43.90, 40.00,
24.00. Analysis (C8H12N2O3) C, H, N.

D. Krehan et al. / Bioorg. Med. Chem. 11 (2003) 4891–4896
4895
.
3-Acetylthio-4-aminocarbonylpiperidine-1-carboxylic acid
methyl ester (4). Thioacetic acid (1.11 mL, 15.5 mmoles)
was added to crude 3 (2.9 g, ca. 15.5 mmoles). Ethyl
acetate (6 mL) was added and the mixture was stirred at
25ꢀ C for 20 h. Additional thioacetic acid (0.4 mL, 5.61
mmoles) was added and the mixture was stirred for
another 72 h. The reaction mixture was evaporated in
vacuo and recrystallized (ethyl acetate) affording 4 (2.92
g, 72%). An analytical sample was further recrystallized
(ethyl acetate): mp 130.0–131.0 C; H NMR (CDCl3)
d 6.20–5.80 (2H, m), 4.22–3.98 (3H, m), 3.74–3.68 (3H,
m), 3.43–3.30 (1H, m), 3.12–3.00 (1H, m), 2.84–2.76
(1H, m), 2.35 (3H, s), 2.00–1.88 (1H, m), 1.80–1.63
(1H, m); 13C NMR (CDCl3) d 195.5, 174.0, 156.2, 53.0,
48.9, 44.9, 42.0, 31.0, 25.8. Analysis (C10H16N2O4S) C,
H, N, S.
MgCl2 6H2O, 1.8 mM CaCl2, 5 mM HEPES, pH 7.5)
supplemented with 2.5 mM pyruvate, 0.5 mM theo-
phylline and 50 mg/mL gentamycin. Stage V–VI oocytes
were collected.
Human r1 cDNAin pcDNA(Invitrogen, San Diego,
CA, USA) was provided by Dr. George Uhl (National
Institute for Drug Abuse, Baltimore, MD, USA). After
linearization of the plasmid containing r1 cDNAwith
restriction enzyme Xba 1, cRNAwas synthesized using
the ‘Mmessage Mmachine’ kit from Ambion (Austin,
TX, USA). r1 cRNA(10 ng/50 nL) was injected into
defolliculated stage V–VI Xenopus oocytes and the
oocytes were then incubated at 16ꢀ C on an orbital
shaker for 3 to 8 days prior to recording.
ꢀ
1
Receptor activity was measured by two electrode
voltage clamp recording using a Geneclamp 500
amplifier (Axon Instruments, Foster City, CA, USA),
a MacLab 2e recorder (AD Instruments, Sydney,
NSW, Australia) and Chart version 3.5 program.
Oocytes were voltage clamped at À60 mV and con-
tinuously superfused with frog Ringer solution (96
3-Hydroxy-4,5,6,7-tetrahydroisothiazolo[5,4-c]pyridine-
6-caboxylic acid methylester (5). Asolution of sodium
hydroxide (1.08 g, 26.9 mmoles) in water (9 mL) was
added to crude 4 (3.5 g, ca. 13.5 mmoles). After 1 h of
stirring the reaction mixture was acidified to pH 5 with
sulphuric acid (4 M) and extracted with CH2Cl2. The
combined CH2Cl2 phases were dried (MgSO4) and eva-
porated in vacuo. The residue (2.55 g) was dissolved in
water (10 mL), and H2O2 (33%) (0.57 mL) was added
dropwise at 45ꢀ C. After stirring for 4 h at 45ꢀ C the
reaction mixture was cooled and evaporated in vacuo to
give 2.55 g of crude product. Part of this residue (1.89 g)
was suspended in ClCH2CH2Cl (22 mL), and SO2Cl2
(1.06 mL, 13.1 mmoles) was added dropwise during 25
min. After stirring at room temperature for 24 h the
reaction mixture was filtered. The crystals were washed
with ClCH2CH2Cl and recrystallized (ethanol) affording
.
mM NaCl, 2 mM KCl, 1 mM MgCl2 2H2O, 1.8
.
mM CaCl2
2H2O, 5 mM HEPES). For receptor
activation measurements, the indicated concentrations
of drug were added to the buffer solution.
Antagonist activity was measured as a pKi or an IC50
value. The pKi value was determined on the basis of a
GABA concentration-response curve as a control fol-
lowed by a GABA concentration-response curve in the
presence of a fixed antagonist concentration on the same
oocyte (nꢁ3). The IC50 value was determined by mea-
suring the activation of the receptor by GABA in the
presence of different antagonist concentrations. The
GABA concentration was chosen as the concentration
which produces 50% of max response, e.g., EC50, based
on a standard GABA concentration–response curve for
each oocyte prior to the inhibition curve, thereby using
a fixed relative response level of GABA.
5 (0.67 g, 23%): mp 179.5–181.5ꢀ C; H NMR (CDCl3)
1
d 9.08 (1H, broad s), 4.72–4.65 (2H, m), 3.77 (3H, s),
3.74–3.68 (2H, m), 2.52–2.62 (2H, m); 13C NMR
(CDCl3) d 159.2, 156.0, 153.6, 119.6, 53.2, 42.8, 41.0,
22.6. Analysis (C8H10N2O3S) C, H, N, S.
4,5,6,7-Tetrahydroisothiazolo[5,4-c]pyridin-3-ol hydro-
bromide (Thio-THIP, HBr). To 5 (0.47 g, 2.19 mmoles)
was added 33% hydrobromide in acetic acid (47 mL)
and the mixture was stirred for 48 h. Evaporation and
re-evaporation twice with toluene in vacuo gave crude
Thio-THIP, HBr. Heating the residue in methanol (15
mL) afforded the product as white crystals after fil-
tration (0.44 g, 85%): mp 221.0–222.0ꢀ C (decomposed);
1H NMR (D2O) d 4.46–4.43 (2H, m), 3.52 (2H, t, J 6.3
and 6.0 Hz), 2.72–2.66 (2H, m); 13C NMR (D2O) d
169.50, 146.50, 118.20, 41.30, 40.60, 18.95. Analysis
(C6H9BrN2OS) C, H, N.
Analysis of kinetic data
Current (I) as a function of GABA concentration was
fitted by least squares to I=Imax[GABA]nH/(EC5n0H+
[GABA]nH), where Imax is the maximum current, EC50
is the effective concentration that activates 50% of the
maximum current and nH is the Hill coefficient. The
EC50 and the nH were determined by fitting data from
individual oocytes using Kaleidagraph 3.0 (1993). The
pKi value was calculated from pKi=log(DR-
1)Àlog[Antagonist], where DR is the dose-ratio
(EC50(GABA+antagonist)/EC50(GABA)).
Electrophysiological recording
Xenopus laevis was anaesthesized with 0.17% ethyl 3-
aminobenzoate and a lobe of the ovaries was carefully
removed. The lobe of the ovary was placed in oocyte
releasing buffer 2 (OR2) (82.5 mM NaCl, 2 mM KCl, 1
The IC50 was determined by fitting data from individual
oocytes using Kaleidagraph 3.0 (1993). Current as a
function of antagonist concentration was fitted by least
squares to I=ImaxÀ(Imax[Antagonist]n/(IC5n0H+[Anta-
gonist]n), where IC50 is the effective concentration of
antagonist that inhibits 50% of the maximum current,
Imax is maximum response, n is the slope of the inhibi-
tion curve and nH is the Hill coefficient. The pKi value
.
mM MgCl2 6H2O, 5 mM HEPES, pH 7.5) with 2 mg/
mL Collagenase A(Boehringer Mannheim) for 2 h.
Defolliculated oocytes were then rinsed with frog
Ringer solution (96 mM NaCl, 2 mM KCl, 1 mM

4896
D. Krehan et al. / Bioorg. Med. Chem. 11 (2003) 4891–4896
can be estimated from the IC50 by using Ki=IC50/(1+
([GABA]/EC50)nH)), where [GABA] is the concen-
tration of GABA used to determine the IC50, EC50 is the
effective concentration that activates 50% of the max-
imum current and nH is the Hill coefficient of the GABA
standard concentration-response curve. pKi values are
expressed as meanÆSEM (nꢁ3).
5. Mehta, A. K.; Ticku, M. K. Brain Res. Rev. 1999, 29, 196.
6. Murata, Y.; Woodward, R. M.; Miledi, R.; Overman, L. E.
Bioorg. Med. Chem. Lett. 1996, 6, 2073.
7. Krogsgaard-Larsen, P.; Johnston, G. A. R.; Lodge, D.;
Curtis, D. R. Nature 1977, 268, 53.
8. Krogsgaard-Larsen, P.; Hjeds, H.; Curtis, D. R.; Lodge,
D.; Johnston, G. A. R. J. Neurochem. 1979, 32, 1717.
9. Woodward, R. M.; Polenzani, L.; Miledi, R. Mol. Phar-
macol. 1993, 43, 609.
10. Krogsgaard-Larsen, P.; Mikkelsen, H.; Jacobsen, P.;
Falch, E.; Curtis, D. R.; Peet, M. J.; Leah, J. D. J. Med.
Chem. 1983, 26, 895.
Acknowledgements
11. Krogsgaard-Larsen, P.; Hjeds, H.; Curtis, D. R.; Leah,
J. D.; Peet, M. J. J. Neurochem. 1982, 39, 1319.
12. Brehm, L.; Ebert, B.; Kristiansen, U.; Wafford, K. A.;
Kemp, J. A.; Krogsgaard-Larsen, P. Eur. J. Med. Chem. 1997,
32, 357.
Dr. George Uhl (Baltimore, MD, USA) is gratefully
acknowledged for kindly providing us with human r1
cDNA. Dr. Hue Tran, Mr Kong Li, and Miss Cristina
Høy Fischer are acknowledged for excellent technical
assistance.
13. Chebib, M.; Vandenberg, R. J.; Johnston, G. A. R. Br. J.
Pharm. 1997, 122, 1551.
14. Krogsgaard-Larsen, P.; Roldskov-Christiansen, T. Eur. J.
Med. Chem. 1979, 14, 157.
References and Notes
15. Krogsgaard-Larsen, P.; Hjeds, H. Acta Chem. Scand.
1974, B28, 533.
16. Hjeds, H.; Krogsgaard-Larsen, P. Acta Chem. Scand.
1976, B30, 567.
17. Krogsgaard-Larsen, P. Acta Chem. Scand. 1977, B31, 584.
18. Frølund, B.; Jørgensen, A. T.; Tagmose, L.; Stensbøl,
T. B.; Vestergaard, H. T.; Engblom, C.; Kristiansen, U.; San-
chez, C.; Krogsgaard-Larsen, P.; Liljefors, T. J. Med. Chem.
2002, 45, 2454.
1. Chebib, M.; Johnston, G. A. R. J. Med. Chem. 2000, 43,
1427.
2. Zhang, D.; Pan, Z.-H.; Awobuluyi, M.; Lipton, S. A.
Trends Pharmacol. Sci. 2001, 22, 121.
3. Nayeem, N.; Green, T. P.; Martin, I. L.; Barnard, E. A. J.
Neurochem. 1994, 62, 815.
4. Bailey, M. E. S.; Albrecht, B. E.; Johnson, K. J.; Darlison,
M. G. Biochim. Biophys. Acta 1999, 1447, 307.