The Synthesis and Evaluation of Fluoro-, Trifluoromethyl-, and Iodomuscimols as GABA Agonists

Article abstract of DOI:10.1002/chem.201701443

Halogenated analogues of the neurotoxic alkaloid muscimol were prepared with fluorine, iodine or trifluoromethyl at the 4 position of the isoxazole ring system. These compounds were investigated as agonists for GABA_A receptors. Only the C-4 fluorine-containing analogue proved to be an active compound in these assays. The fluoro analogue was less active than muscimol, however it showed differential activity between synaptic (α₁β₂γ₂) and extrasynaptic (α₄β₂γ) GABA_A receptors, having a similar potency to the neurotransmitter GABA for the extrasynaptic (α₄β₂γ) receptor.

Full text of DOI:10.1002/chem.201701443

A Journal of
Accepted Article
Title: The synthesis and evaluation of fluoro-, trifluoromethyl, and iodo-
muscimols as GABA agonists
Authors: David O'Hagan, Mohd Abdul Fatah Abdul Manan, David
Cordes, Alexandra Slawin, Vivian Liao, Han CHua, Mary
Chebib, and Michael Buehl
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201701443
Link to VoR: http://dx.doi.org/10.1002/chem.201701443
Supported by

Chemistry - A European Journal
10.1002/chem.201701443
FULL PAPER
The synthesis and evaluation of fluoro-, trifluoromethyl, and iodo-
muscimols as GABA agonists
[a]
[a]
[a]
[a]
Mohd Abdul Fatah Abdul Manan, David B. Cordes, Alexandra M. Z. Slawin, Michael Bühl, Vivian
[b]
[b]
[b]
[a]
W. Y. Liao , Han. C. Chua, Mary Chebib and David O’Hagan.*
Abstract: Halogenated analogues of the neurotoxic alkaloid
muscimol were prepared with fluorine, iodine or trifluoromethyl at
the 4 position of the isoxazole ring system. These compounds
were investigated as agonists for GABA receptors. Only the C-4
A
fluorine containing analogue proved to be an active compound in
these assays. The fluoro analogue was less active than muscimol,
reported, despite the obvious aspect that such derivatives may
offer potential radiochemical tools. Fluorine is a particularly
attractive substituent in the field of brain imaging radiotracers due
to the wide use of the short lived 18F-isotope (t1/2 = 109 min) for
positron emission tomography (PET).8–10 Also in medicinal
chemistry when hydrogen is replaced by fluorine, usually
bioactivity is retained, due to the relatively small size and poor
coordinating ability of fluorine.11–13 Fluorination at C-4 of muscimol
is the only reasonable location for the halogen that will give rise
to a stable molecule, thus the fluorinated analogue 3 emerged as
however it showed differential activity between synaptic (
and extrasynaptic ( ) GABA receptors, having a similar
potency to the neurotransmitter GABA for the extrasynaptic
) receptor.
  2
  )



A
 
( 
3
a synthetic target. The CF analogue 4 was also an attractive
Introduction
target, as new methods are emerging for the synthesis of
1
8
groups.14 Finally a synthesis of the iodo
radiolabelled [ F]-CF
3
analogue 5 was explored as the 1 I-isotope is widely
used for single photon emission computed
tomography (SPECT) imaging.15,16 The syntheses of
these compounds are described below, and the
assessment of their biological activities as agonists
23
at three different GABA
discussed.
A
receptor types are also
Results and Discussion
A
number of approaches were considered to
introduce fluorine at C-4 of the isoxazole ring, to progress towards
Muscimol 1 is a neurotoxic alkaloid which is a natural product of
the hallucinogenic fungi Amanita muscaria (Figure 1).1–3
Muscimol 1 has a structural similarity to the neurotransmitter -
4-fluoromuscimol 3. The successful route is illustrated in Scheme
1. The synthesis starts with protection of commercially available
isoxazole 6 as an isopropyl ether 7 and then reduction of the ester
to alcohol 8. Appel bromination gave 9, and this was followed by
aminobutyric acid (GABA) 2 and it activates GABA
A
receptors via
the orthosteric GABA binding site as a potent GABA
A
receptor
3
nucleophilic displacement with NaN to generate azide 10. A
agonist. Muscimol is used as a tool in neurobiology for studying
the physiological and pharmacological characterisations of
Staudinger reduction gave amine 11 which was then Boc
protected resulting in carbamate 12. Fluorination was achieved by
butyllithium double deprotonation, and then quenching with N-
fluorobenzenesulfonimide (NFSI) as an ‘electrophilic’ source of
fluorine. This reaction was not so efficient resulting in a low
(~30%) yield of 13, however it could not be readily improved and
would require to be optimised to make this a useful radiochemical
route. Sequential deprotection of 13, firstly be removing the Boc
protecting group and then ether cleavage generated the desired
fluoromuscimol 3. The structures of intermediates 12, 13, 14 and
fluoromuscimol 3 were supported by X-ray crystallography
receptors in the central nervous system.5–7 There is a
GABA
A
general desire to identify new compounds which cross the blood
brain barrier for the
discovery
generation of brain
imaging radiotracers,
and we have decided to
explore muscimol in this
of
next
[
a]
M. A. F. A. Manan, Dr D. C.
Cordes, Prof. A. M. Z. Slawin, Prof.
D. O’Hagan.
EaStCHEM School of Chemistry,
University of St Andrews,
North Haugh, St Andrews, KY16
context. Although
a
number of analogues of
9ST, UK.
muscimol
previously
prepared, no fluorinated
and iodinated
derivatives were ever
have
been
E-mail: do1@st-andrews.ac.uk
V. W. Y. Liao, H. C. Chua, Prof M.
Chebib.
Faculty of Pharmacy,
The University of Sydney,
Sydney, New South Wales, 2006,
Australia.
(
Figure 2 and Scheme 1).
[
b]
Supporting information for this
article is given via a link at the end
of the document.((Please delete
this text if not appropriate))
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701443
FULL PAPER
Scheme 3. Successful route to trifluoromethylmuscimol 4. Reagents and
conditions: a) NIS, TFA, rt, overnight, 86%; b) PMe O, 0 C, 1 h, 83%;
3
, THF, H
2
Scheme 1. Route to fluoromuscimol 3 and its X-ray structure. Reagents and
c) Boc O, Na CO , dioxane/H O, rt, overnight, 81%; d) Boc O, DMAP, MeCN,
2
2
3
2
2
conditions: a) i-PrBr, K
2
CO
, CBr
O, 0 C, 1 h, 65%; f) Boc
6 h, 90%; g) n-BuLi, NFSI, THF, -78 Crt, 16h, 30%; h) TFA, DCM, rt, 98%;
3
, DMF, 55 C, overnight, 77%; b) NaBH
, rt, 12 h, 58%; d) NaN , DMF, 80 C, 3.5 h,
O, Na CO , dioxane/H O, rt,
4
, MeOH, 0
reflux, 3 h, 87%; e) MFSDA, CuI, HMPA/DMF, 80 C, 24 h, 34%; f) TFA, DCM,
rt, 6 h; g) 33% HBr in AcOH, 60 °C, 48 h, 36% over two steps.
Crt, 16h, 90%; c) PPh
3
4
3
91%; e) PMe
3
, THF, H
2
2
2
3
2
1
suitable crystal was submitted for X-ray diffraction analysis
i) 33% HBr in AcOH, 60 °C, 17 h, 36%.
(
Figure 3), which confirmed its structure. Ether 20 was then
treated in HBr/AcOH in order to cleave the isopropyl group. The
final trifluoromethylmuscimol 4 was obtained as a colourless solid
in 36% yield after purification by preparative HPLC.The
difluorocarbene mechanism for the trifluoromethylation of 18 is
shown in Scheme 4 consistent with that previously proposed by
Chen and Su.17,18
Figure 2. X-ray structures of 12, 13, 14..
Synthesis of 4-trifluoromethylmuscimol 4
The synthetic route to trifluoromethylmuscimol 4 is illustrated in
Scheme 3. It involves an eleven step protocol starting from
isoxazole 6. The first four synthesis steps follow that for the
synthesis of fluoromucimol 3 outlined in Scheme 1. Iodination at
C-4 of azide 10 was accomplished using NIS as the electrophilic
halogen source. Reduction of the azide moiety of 15 to amine 16
via the Staudinger process followed by direct protection with Boc
anhydride afforded the doubly N-protected amine 18 in 87%. The
key step for trifluoromethylation of 18 was achieved using
Scheme 4. Putative mechanism for the trifluoromethylation of 18.17–21
Synthesis of 4-iodomuscimol 5
Iodomuscimol 5 was prepared by direct HBr/AcOH deprotection
of isopropyl iodomuscimol 16. This reaction was not so efficient
and the major product was muscimol 1 which arose from a
significant level of protio-deiodination. However an analytically
pure sample of iodomuscimol 5 was obtained after purification by
preparative HPLC and the product was used for bioassay studies.
methyl fluorosulfonyldifluoroacetate (MFSDA) as a nucleophilic
trifluoromethylating reagent, using the method developed by
Chen and Su.17,18 Deprotection of 19 afforded the TFA salt of
ether 20 as a colourless hygroscopic crystalline solid, and a
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701443
FULL PAPER
Scheme 5. Synthesis of iodomuscimol 5. Reagents and conditions: a) 33% HBr
in AcOH, 60 °C, 17 h, 19%.
GABA receptor agonist assays
The halogenated muscimol analogues were assayed against
human synaptic 122, extrasynaptic 42 and 1 GABA
receptors expressed in Xenopus laevis oocytes. Oocytes were
injected with a mixture of cRNAs encoding the desired GABA
A
A
receptor subunits and were incubated for 3-7 days at 18 C in
modified Barth’s solution (96 mM NaCl, 2.0 mM KCl, 1.0 mM
MgCl
2 2
, 1.8 mM CaCl , 5.0 mM HEPES, 2.5 mM sodium pyruvate,
0.5 mM theophylline, 100 g/ml gentamycin, pH 7.4).
Agonist-elicited current responses were measured using two-
electrode voltage-clamp electrophysiology, with the membrane
potential held at −60 mV using an Axon Geneclamp 500B
amplifier (Molecular Devices). Data acquisition and analysis were
performed using Clampex and Clampfit 10.4 respectively. In the
data presented in Table 1, Emax refers to the maximal current
produced by a saturating concentration of the agonist and EC50
refers to the concentration of agonist that is required to produce
Figure 3. Activities of GABA 2, muscimol 1 and fluoromuscimol 3 at synaptic
A
122 (a), extrasynaptic (b) 42 and 1 GABA (c) receptors.
receptors.
Fluoromuscimol 3 was between 9- and 17-fold less potent than
GABA at both 122 and 42 receptors. Interestingly, while
fluoromuscimol 3 Emax only reached 77 % of GABA Emax at
synaptic 122 receptors, it displayed higher relative efficacy
at extrasynaptic 42 receptors (Emax = 160 %) (Table 1,
Figures 4a and 4d respectively). Fluoromuscimol 3 was an
50% of the Emax response.
Table 1. Potency and relative efficacy of GABA 2, muscimol 1, fluoromuscimol
trifluoromethylmuscimol and iodomuscimol at synaptic 122,
receptors (ND = not determined; NA = not
3
4
5
extrasynaptic 42 and 1 GABA
A
active).
A
inefficient agonist at 1 GABA receptors, evoking current
responses approximately 33% of GABA Emax on average at 1
mM (Figure 3c).

1

2

2
4 2
  
Extrasynaptic
Agonist

1
Synaptic
EC50
E
(%)
Max
EC50
E
Max
EC50
E
Max
n
n
n
(M)
(M)
(%)
100
160
(M)
(%)
100
75
GABA 2
20
100
5
6
2.8
9
9
1.1
1.6
8
6
muscimol 1
3.1
120
77
0.43
F-muscimol
1
75
5
3
2
48
IA
-
160
IA
-
5
3
-
ND
NA
NA
NA
NA
NA
3
6
3
CF -
NA
NA
NA
NA
muscimol 4
I-muscimol
5
3
2
The neurotransmitter GABA 2 was used as a control in this
study. The EC50 values of GABA at 122, 42 and ρ1
receptors were approximately 20, 2.8 and 1.1 μM
respectively (Table 1, Figures 3), consistent with previous
22–25
reports.
receptor subtypes with higher potency and maximal efficacy
relative to GABA. In contrast, muscimol showed
comparable potency (EC50 = 1.6 μM), but lower maximal
efficacy (Emax = 75 %) than GABA at ρ1 GABA
Muscimol 1 activated both 122 and 42
1
A
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701443
FULL PAPER
oxygen side or the H/F side. The approximate barriers for
these two TSs in the fluorinated compound 3 are ~2.5
kcal/mol and ~3.1 kcal/mol and about ~2.0 kcal/mol for both
transitions in 1. Although the transition barrier is a little higher
for rotation in 3, there is no sense that a significantly different
conformation is accessed between these two molecules at
low energy. Given that the introduction of fluorine into 3 will
not significantly change the ionisation of the zwitterionic
state, we conclude that there is no evidence to suggest a
significantly different conformation, and that the poorer
performance of 3 as an agonist is most probably due to an
adverse interaction between the fluorine and the protein,
relative to hydrogen in 1, which means higher concentrations
are required to achieve a maximal effect.
In addition, we find that increasing the size of the ring
substituent to trifluoromethyl or iodine abolished muscimol’s
agonist activity, suggesting a further sensitivity to sterics at
this site. The application of trifluoromethyl and iodo
analogues of muscimol at concentrations up to 300–1000 M
did not result in detectable current responses at α1β2γ2,
A
α4β2δ and 1 GABA receptors (Figures 4b & c, 4e & f and
4h & i respectively).
Figure 4. Current responses of GABA 2, fluoromuscimol 3,
trifluoromethylmuscimol 4 and iodomuscimol 5 at synaptic 122 (a-c),
Conclusions
A
extrasynaptic 42 (d-e) and ρ1 (g-i) GABA receptors.
The
first
syntheses
of
fluoromuscimol
3,
trifluoromethylmuscimol 4 and iodomuscimol 5 have been
achieved in nine, eleven and seven steps respectively from
isoxazole 6. Fluoromuscimol 3 showed greater maximum
response in comparison to GABA at the extrasynaptic
To assess the impact of halogenation on muscimol activity,
we also compared the bioactivity of muscimol and
fluoromuscimol. We found that the introduction of fluorine
increased the EC50 values of muscimol 1 by 56-fold for
GABA receptors 42, but lower overall potency. On the
A
A
122 and 112-fold for 42 GABA receptors, indicating
other hand, trifluoromethylmuscimol 4 and iodomuscimol 5
were inactive and did not evoke any current responses at the
a negative effect of fluorination on muscimol potency.
However, the relative efficacy was largely unaffected by the
introduction of fluorine, as both muscimol and
tested GABA
was no evidence from computational analysis that muscimol
and fluoromuscimol adopt significantly different
A
receptors. All exist as zwitterions and there
fluoromuscimol showed similar
Emax values at 42
1
3
receptors, and only a slight reduction was detected for
fluoromuscimol at 122 receptors. Fluorination had a more
conformations and thus to a first approximation, the poorer
performance of the muscimol analogues most likely arises
due to adverse steric interactions of the ring sibstituent on
receptor binding.
profound impact on muscimol activity at 1 GABA
receptors, as fluoromuscimol showed poor sensitivity
activation detected only at concentrations > 30 μM) and low
A
(
efficacy, 33 % at the highest tested concentration of 1 mM.
Acknowledgements
On possibility to explain these differences in activity is that
muscimol 1 and fluoromuscimol 3 adopt different preferred
conformations in solution, influenced by the presence of
fluorine. To explore this we have carried out a DFT based
conformational analysis (B3LYP/6-311+G**/CPCM(water)
level) with a dielectric continuum to model water. An energy
MAFAM acknowledges the Ministry of Education of Malaysia
and Universiti Teknologi MARA for
a scholarship. MC
acknowledges the funding support of the Australian National
Health and Medical Research Council (APP1081733). We thank
Dr Xiang-Guo Li (Turku University) for suggesting compound 3
as a synthetic target.
profile monitoring rotation around the C-C(NH +
) bond, the
3
only rotatable bond of the molecular skeleton in these
molecules was carried out. The minimum energy conformers
Keywords: muscimol, GABA agonists, GABA receptors
fluorination, trifluoromethyl, iodination.
in each case had the C-NH +
bond approximately
3
perpendicular to the aromatic ring, very similar to the X-ray
structure for 3 (SI Table 1). There were obvious high energy
+
transitions when the C-NH
3
bond became co- planar with
the heterocyclic ring on the rotational profile, either on the
References
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701443
FULL PAPER
[
1]
T. Takemoto, T. Nakajima, T. Yokobe, J. Pharm. Soc. Jpn. 1964, 84,
232–1233.
C. H. Eugster, G. F. R. Müller, R. Good, Tetrahedron Lett. 1965, 6, 1813–
815.
[14] M. Huiban, M. Tredwell, S. Mizuta, Z. Wan, X. Zhang, T. L. Collier, V.
Gouverneur, J. Passchier, Nat. Chem. 2013, 5, 941–944.
1
[
2]
[15] F. Lu, Z. Yuan, Quant. Imaging Med. Surg. 2015, 5, 433–447.
[16] S. Tsartsalis, M. Moulin-Sallanon, N. Dumas, B. B. Tournier, C. Ghezzi,
Y. Charnay, N. Ginovart, P. Millet, Nucl. Med. Biol. 2014, 41, 186–195.
[17] Q.-Y. Chen, S.-W. Wu, J. Chem. Soc. Chem. Commun. 1989, 705–706.
[18] D.-B. Su, J.-X. Duan, A.-J. Yu, Q.-Y. Chen, J. Fluor. Chem. 1993, 65,
11–14.
1
[3]
K. Bowden, A. Drysdale, Tetrahedron Lett. 1965, 6, 727–728.
Amanita muscaria,
[
4].
https://commons.wikimedia.org/wiki/File:Amanita_muscaria_3_vliegenz
wammen_op_rij.jpg
[
[
[
5]
6]
7]
G. A. R. Johnston, Curr. Pharm. Des. 2005, 11, 1867–1885.
M. Chebib, G. A. R. Johnston, J. Med. Chem. 2000, 43, 1427–1447.
J. Krall, T. Balle, N. Krogsgaard-Larsen, T. E. Sørensen, P. Krogsgaard-
Larsen, U. Kristiansen, B. Frølund, Adv. Pharmacol. 2015, 72, 201–227.
P. W. Miller, N. J. Long, R. Vilar, A. D. Gee, Angew. Chem. Int. Ed. 2008,
[19] S. Roy, B. T. Gregg, G. W. Gribble, V. D. Le, S. Roy, Tetrahedron 2011,
67, 2161–2195.
[20] S. L. Clarke, G. P. Mcglacken, Chem. Eur. J. 2017, 23, 1219–1230.
[21] Y. Nakamura, M. Fujiu, T. Murase, Y. Itoh, H. Serizawa, K. Aikawa, K.
Mikami, Beilstein J. Org. Chem. 2013, 9, 2404–2409.
[8]
4
7, 8998–9033.
[22] G. Akk, J. Bracamontes, J. H. Steinbach, J. Physiol. 2004, 556, 387–99.
[23] N. Karim, P. Wellendorph, N. Absalom, G. A. R. Johnston, J. R.
Hanrahan, M. Chebib, Amino Acids 2013, 44, 1139–1149.
[
[
[
[
9]
L. Cai, S. Lu, V. W. Pike, Eur. J. Org. Chem. 2008, 2853–2873.
10] D. Le Bars, J. Fluor. Chem., 2006, 127, 1488–1493.
11]] J. Swinson, Pharmaceuticals 2005, 1, 26–30.
[24] H. J. Lee, N. L. Absalom, J. R. Hanrahan, P. Van Nieuwenhuijzen, P. K.
Ahring, M. Chebib, Brain Res. 2016, 1644, 222–230.
12] B. K. Park, N. R. Kitteringham, P. M. O’Neill, Annu. Rev. Pharmacol.
Toxicol. 2001, 41, 443–470.
[25] L. Y. Hartiadi, P. K. Ahring, M. Chebib, N. L. Absalom, Biochem.
Pharmacol. 2016, 103, 98–108.
[
13] S. D. Banister, J. Stuart, R. C. Kevin, A. Edington, M. Longworth, S. M.
Wilkinson, C. Beinat, A. S. Buchanan, D. E. Hibbs, M. Glass, M. Connor,
I. S. McGregor, M. Kassiou, ACS Chem. Neurosci. 2015, 6, 1445–1458.
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701443
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
FULL PAPER
Mohd Abdul Fatah Abdul Manan, David
B. Cordes, Alexandra M. Z. Slawin,
Vivian W. Y. Liao, Han. C. Chua, Mary
Chebib and David O’Hagan*
Page No. – Page No.
The synthesis and evaluation of
fluoro-, trifluoromethyl, and iodo-
muscimols as GABA agonists
Muscimol is a neurotoxin and potent GABA receptor agonist from the mushroom
Amanita muscaria. The first halogenated analogues (F-, I-, CF -) of muscimol have
3
been prepared, locating the halogen at the C-4 position. Only the mono fluoro
analogue showed activity at GABA receptors, displaying a greater maximum
response in comparison to GABA at the extrasynaptic GABA
overall potency .
A
receptors, but lower
This article is protected by copyright. All rights reserved.