Synthesis and biological evaluation of (-)-kainic acid analogues as phospholipase D-coupled metabotropic glutamate receptor ligands

Article abstract of DOI:10.1039/c4ob02002b

(-)-Kainic acid potently increases stretch-induced afferent firing in muscle spindles, probably acting through a hitherto uncloned phospholipase D (PLD)-coupled mGlu receptor. Structural modification of (-)-kainic acid was undertaken to explore the C-4 substituent effect on the pharmacology related to muscle spindle firing. Three analogues 1a-c were synthesised by highly stereoselective additions of a CF₃, a hydride and an alkynyl group to the Re face of the key pyrrolidin-4-one intermediate 5a followed by further structural modifications. Only the 4-(1,2,3-triazolyl)-kainate derivative 1c retained the kainate-like agonism, increasing firing in a dose-dependent manner. Further modification of 1c by introduction of a PEG-biotin chain on the 1,2,3-triazole fragment afforded compound 14 which retained robust agonism at 1 μM and appears to be suitable for future use in pull-down assays and far western blotting for PLD-mGluR isolation. This journal is

Full text of DOI:10.1039/c4ob02002b

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Synthesis and biological evaluation of (−)-kainic
acid analogues as phospholipase D-coupled
metabotropic glutamate receptor ligands†
Cite this: Org. Biomol. Chem., 2014,
12, 9638
Chiara Zanato,*a Sonia Watson,b Guy S. Bewick,b William T. A. Harrisonc and
Matteo Zanda*a,d
(−)-Kainic acid potently increases stretch-induced afferent firing in muscle spindles, probably acting
through a hitherto uncloned phospholipase D (PLD)-coupled mGlu receptor. Structural modification of
(−)-kainic acid was undertaken to explore the C-4 substituent effect on the pharmacology related to
muscle spindle firing. Three analogues 1a–c were synthesised by highly stereoselective additions of a CF3,
a hydride and an alkynyl group to the Re face of the key pyrrolidin-4-one intermediate 5a followed by
further structural modifications. Only the 4-(1,2,3-triazolyl)-kainate derivative 1c retained the kainate-like
agonism, increasing firing in a dose-dependent manner. Further modification of 1c by introduction of a
PEG-biotin chain on the 1,2,3-triazole fragment afforded compound 14 which retained robust agonism
at 1 μM and appears to be suitable for future use in pull-down assays and far western blotting for
PLD-mGluR isolation.
Received 19th September 2014,
Accepted 14th October 2014
DOI: 10.1039/c4ob02002b
www.rsc.org/obc
(−)-Kainic acid (Fig. 1), which selectively activates the
kainate family of ionotropic channels but has no known
Introduction
Muscle spindles are skeletal muscle sensory organs (or activity at any metabotropic receptors (mGluRs), potently
mechanoreceptors) that play a key role in motor control and in increases muscle spindle stretch-induced afferent firing at
the regulation of muscle length during movement. Recently, 1 μM concentration (n = 7, p < 0.01). (−)-Kainic acid is not
we presented evidence1 that the glutamate receptor in acting through the classical kainate group of glutamate recep-
mechanically sensitive nerve terminals of the muscle spindle tors, as the kainate receptor-specific antagonist NBQX (2,3-
is pharmacologically distinct from any of the cloned receptors dioxo-6-nitro-1,2,3,4-tetrahydrobenzo[f]quinoxaline-7-sulfona-
but its pharmacology strongly resembles that of the uncloned mide) had no effect either alone (n = 4) or on the (−)-kainic
phospholipase D (PLD)-coupled mGluR described in the acid-mediated increase in afferent firing (n = 7). However, the
hippocampus.2 A similar receptor seems to be present in the enhancement by (−)-kainic acid was inhibited by PCCG13
blood pressure sensors (baroreceptors) of the cardiovascular [(2R,1′S,2′R,3′S)-2-(2′-carboxy-3′-phenylcyclopropyl)glycine],
a
system, where it may prove a useful drug target for treating high selective competitive antagonist of the PLD-coupled mGluR,
blood pressure (hypertension). We therefore aimed to produce (n = 4, p > 0.05).3 Thus, the enhancement by (−)-kainic acid
ligands that could be used to isolate and localise the receptor, seemed to be through activation of the PLD-mGluR.
with a view to sequencing, cloning and characterising it.
The aim of the current study was the synthesis of (−)-kainic
acid analogues suitable to be used as tools for characterising
and isolating the enigmatic PLD-mGluR in future studies.
aKosterlitz Centre for Therapeutics, Institute of Medical Sciences, University of
Aberdeen, Foresterhill, AB25 2ZD Scotland, UK. E-mail: m.zanda@abdn.ac.uk
bSchool of Medical Sciences, Institute of Medical Sciences, University of Aberdeen,
Foresterhill, AB25 2ZD Scotland, UK
cDepartment of Chemistry, University of Aberdeen, AB24 3UE Scotland, UK
dC.N.R.-I.C.R.M., via Mancinelli 7, 20131 Milano, Italy
†Electronic supplementary information (ESI) available: Experimental methods
and characterisation of compounds 1–14, biological tests on compounds 1a–c
and 14, and copies of the NOESY and HOESY NMR experiments. CCDC 1019225.
For ESI and crystallographic data in CIF or other electronic format see DOI:
10.1039/c4ob02002b
Fig. 1 (−)-Kainic acid.
9638 | Org. Biomol. Chem., 2014, 12, 9638–9643
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
literature for the latter step were rather modest,5 it was
Results and discussion
decided to alkylate directly the ketone 2 using the conditions
Most mGluR ligands, both cyclic and acyclic, feature a gluta- previously reported by Greene6 and Schafmeister.7 Disappoint-
mate-like structural framework, incorporating one amino ingly, the enolisation of compound
2 with n-BuLi in
group and two free acidic functions.4 We therefore hypo- THF-HMPA at −78 °C, followed by treatment with tert-butyl-
thesised that the positions 1 to 3 of the (−)-kainic acid scaffold bromoacetate at −40 °C, provided a mixture of mono- and di-
could be very important for binding to the target PLD-mGluR, alkylated products. Although the desired compound 3 was
in analogy with other mGluR ligands. Based on these consider- obtained as single diastereoisomer having the desired R con-
ations, it was decided to focus on the modification of the posi- figuration at C3, the yield was still too low (30%).
tion 4 of (−)-kainic acid (Fig. 1), to explore the influence of the
After this explorative attempt, we changed strategy and
C4 substituent – which is a prop-2-ene group in native decided to replace the benzoyl protecting group with a N-phe-
(−)-kainic acid – on the pharmacology related to muscle nylfluorenyl group (Pf), which had been previously employed
spindle firing.
on the same substrate giving excellent results.6–8
Since (−)-kainic acid is an expensive chemical, it was
Therefore, the N-Pf protected ketone 4 was synthesised
deemed not to be an ideal starting substrate for the synthesis starting from trans-4-hydroxy-L-proline (Scheme 3) through
of analogues for the present structure–activity relationship Fischer esterification, N-Pf protection and Swern oxidation
(SAR) study, therefore we synthesised from inexpensive trans-4- (that provided a simplified work-up procedure in comparison
hydroxy-L-proline a common precursor or key intermediate with the Ruthenium tetroxide oxidation).
(Scheme 1) which could be easily transformed into our target
kainic acid analogues 1a, 1b and 1c.
As expected,7 the alkylation of ketone 4 gave 5a in a good
yield (80%) and excellent diastereoselectivity (dr 16 : 1).
After silica gel chromatography separation of the two dia-
stereoisomers 5a and 5b, the stereochemistry of each ketone
was confidently assigned by comparing the respective 1H NMR
spectra, and specifically the chemical shifts and the coupling
constants between H2 and H3 in 5a,b with those reported pre-
viously in the literature for similar trans/cis derivarives6,7,8b
Synthesis
This work took advantage of the efficient chemistry developed
by Baldwin et al.5 to convert, in good yield and on a multi-
gram scale (Scheme 2), trans-4-hydroxy-L-proline into ketone 2
through a three step synthesis: Fisher esterification, N-benzoy-
lation under standard Schotten–Baumann conditions and
ruthenium tetroxide Sharpless oxidation.
(5a: H2 appears as a doublet at 3.50 ppm with JH2β–H3α
6.1 Hz; 5b: H2 appears as doublet at 4.02 ppm with JH2β–H3β
=
=
8.2 Hz). This assignment was subsequently confirmed by X-ray
analysis of the derivative 9 (see Fig. 5).
In order to obtain compound 3, Baldwin’s pathway envi-
saged the formation of an enamine from ketone 2 and sub-
With the ketone precursor 5a in hand, we proceeded with
the synthesis of the target analogues. The synthesis of com-
pound 1a (Scheme 4) started with a trifluoromethylation reac-
tion using the Ruppert–Prakash reagent9 which afforded the
alcohol 6 as a single diastereoisomer in moderate yield (48%).
Next, the Pf group was removed via hydrogenolysis (Pd/C, H2,
in EtOAc at rt, quantitative yield) and, using the procedure
reported by Baldwin et al. (HCl 6 N in water at reflux for
sequently
a diastereoselective alkylation to install the
substituent in position 3. Since the yields reported in the
Scheme 1 Retrosynthetic approach to the target analogues 1a–c.
Scheme 2 First synthetic approach to a candidate key intermediate 3.
Reagents and conditions: (i) SOCl2, MeOH, 45 °C, overnight, (90%); (ii)
BzCl, NaHCO3, H2O/THF, rt, 1.5 h, (98%); (iii) RuCl3·H2O, NaIO4, CH3CN/
CCl4/H2O, rt, 4 h, (80%); (iv) nBuLi, THF/HMPA at −78 °C, 30 min then
Scheme 3 Second synthetic approach: synthesis of the key intermedi-
ate 5a. Reagents and conditions: (i) SOCl2, MeOH, 45 °C, overnight,
(90%); (ii) Me3SiCl, TEA, DCM, reflux 1 h then MeOH, DCM, rt, 1 h then
PfBr, Pb(NO)3, rt, 96 h, (90%); (iii) DMSO, (COCl)2, DCM, −78 °C, 1 h,
(90%); (iv) nBuLi, THF/HMPA at −78 °C, 30 min then NaI, BrCH2COOtBu
from −60 °C to −42 °C, 1 h, (80%, dr 16 : 1).
NaI, BrCH2COOtBu from −60 °C to −42 °C,
1 h, (30% as single
diastereoisomer).
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 9638–9643 | 9639

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 4 Synthesis of the analogue 1a. Reagents and conditions: (i)
Me3SiCF3, Cs2CO3, DMF, rt, 2 h then TBAF, rt, 30 min (48% as a single
diastereoisomer); (ii) H3PO4 85% wt in H2O, toluene, rt, 2 h; (iii) LiOH
0.5N in H2O, rt, 6 h (48% in two steps).
16 h),10 both the carboxylic acidic functions were deprotected.
Unfortunately the C2 stereocenter underwent epimerization
during the acidic hydrolysis. To circumvent this undesired
event, we opted to use a milder hydrolytic strategy consisting
in the removal of the acid-labile protecting group (Pf and Fig. 3 19F–1H HOESY spectrum of compound 6.
t-butyl ester) with aqueous H3PO4,11 while the methyl ester was
cleaved using aqueous LiOH which delivered the unprotected
CF3-derivative 1a in stereopure form.
The stereochemical configuration of compound 1a at C4
was assigned by performing a 1H NOESY NMR experiment12
on the alcohol precursor 6 (Scheme 4). The NOE peaks for H2
and H3, and their corresponding integration values (Fig. 2A),
are in accordance with the already established configuration of
these two stereocenters, since the observed coupling signals
are stronger when the two protons are in cis (α–α and β–β) and
weaker if they are in trans (α–β) relationship. Analysing the
relative intensity of the OH coupling signals (Fig. 2B) it was
noticed that they are weaker in the case of the couplings with
Hβ protons (H2 and H5β) while they are stronger for the coup-
Scheme 5 Synthesis of analogue 1b. Reagents and conditions: (i)
lings with Hα (H3 and H5α). This evidence allowed the assign-
ment of an S configuration to the newly formed stereocenter at
C4 (OHα and CF3β).
NaBH4, MeOH, rt, 2 h, (90% as a single diastereoisomer); (ii) NaH, DMF,
0 °C, 30 min then MeI, rt, 3 h, (70%); (iii) H3PO4 85% wt in H2O, toluene,
rt, 2 h; (iv) LiOH 0.5N in H2O, rt, 4 h (60% in two steps).
To support this stereochemical assignment, a heteronuclear
19F–1H HOESY NMR experiment12 was performed on the same
compound 6 for confirming the spatial arrangement between
CF3 and non-bonded nuclei around the fluorine atoms. As
shown in Fig. 3, the coupling between fluorides and H5α has a
weaker intensity (β–α) than that with H5β (β–β): this is consist-
ent with the NOESY experiment and supports the previous
stereochemical assignment.
The synthesis of the 4-methoxy analogue 1b (Scheme 5)
also started from the key ketone 5a. Hydride reduction (NaBH4
in MeOH) afforded the alcohol 7 as a single diastereoisomer in
excellent yield (90%). Subsequently, O-methylation with NaH/
MeI gave the ether 8 that was subjected to the previously used
(see Scheme 4) acidic-basic deprotection sequence of the ester
groups to give compound 1b.
Although the stereochemical outcome of the reduction
from ketone 5a to alcohol 7 could be confidently predicted
based on literature reports on structurally related com-
pounds,8b,13 a NOESY spectrum of the alcohol 7 was per-
formed to confirm its configuration. The intensity of the
coupling peaks integrals for H3 and H2 is consistent with the
stereochemistry of C3 and C2 (Fig. 4A). The OH in C4 (Fig. 4B)
displays stronger signals for α–α couplings (with H5α and H3)
and weaker signals for the α–β coupling with H5β, while H4
(Fig. 4C) shows similarly low signals for the interactions with
H3 and H5α (which are both in a trans correlation) and a stron-
ger coupling with H5β (in a cis correlation). From this NMR
experiment, supported by literature data, an S configuration at
the C4 of compound 1b could be therefore safely assigned.
Fig. 2 2D NOESY spectroscopic data for compound 6 showing the
relative intensity of through-space coupling signals: (A) couplings
among H2, H3 and H5 and (B) the same protons with the OH in 4.
9640 | Org. Biomol. Chem., 2014, 12, 9638–9643
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 5 ORTEP of compound 9.16
Fig. 4 2D NOESY spectroscopic data for compound 7 showing the
relative intensity of through-space coupling signals: (A) coupling among
H3, H2 and H5, (B) the same protons with the OH in 4 and (C) the same
proton with the H in 4.
absolute structure parameter. The conformation of the five-
membered ring is an envelope with the C4-carbon atom
bearing the OH group as the flap. This facilitates the formation
of an intramolecular O–H⋯O hydrogen bond, featuring an
H⋯O distance = 2.083 (17) A.
The synthesis of a number of additional kainic acid deriva-
tives modified in position 4 was also unsuccessfully attempted.
For example, dehydroxy-fluorination of 7, as well as of other
protected 4-hydroxy analogues, failed to produce the corres-
ponding 4-fluoro-kainic derivatives under a range of con-
ditions (DAST, Deoxofluor, mesylation followed by treatment
with TBAF) affording either no reaction or complex reaction
mixtures. Attempts to perform reductive amination on 2a for
installing a 4-amino group also failed under a variety of con-
ditions. Analogously, attempts to achieve alkynylation of 5a
with lithiated phenyl-acetylene afforded only complex reaction
mixtures. Finally, attempts to achieve copper-catalysed azide-
alkyne cycloaddition from 9 using alkyl azides, such as methyl
azide, failed too. Due to the above described synthetic difficul-
ties, the pool of kainic acid analogues available for biological
studies was limited to compounds 1a–c.
Scheme 6 Synthesis of analogue 1c. Reagents and conditions: (i)
SiMe3CH, nBuLi, THF, −10 °C, 1 h, (70% as a single diastereoisomer); (ii)
Me3SiN3, CuSO4, Na ascorbate, DMF/H2O, microwave at 120 °C for
30 min, (70%); (iii) H3PO4 85% wt in H2O, toluene, rt, 2 h; (iv) LiOH 0.5N
in H2O, rt, 6 h (65% in two steps).
The 4-(1,2,3-triazolyl) kainic analogue 1c was obtained
(Scheme 6) through the alkynylation14 of ketone 5a using (tri-
methylsilyl)acetylene to give the alkyne 9. The triazole moiety
was then installed by means of a copper-catalysed azide-alkyne
cycloaddition15 employing trimethylsilylazide under micro-
wave irradiation. The newly formed triazole diester 10 was
transformed into the final unprotected compound 1c via the
usual deprotection steps.
In order to establish the stereochemical configuration at C4
of the analogue 1c, a crystallographic analysis on the alkyne 9
(Fig. 5) was performed. The configurations of atoms C2 (S),
C3 (R) and C4 (R) were established by refinement of the Flack
Fig. 6 Kainate increases stretch-induced afferent firing in muscle spin-
dles by acting at the PLD-mGluR and not the kainate receptor. (A)
Kainate at 1 µM 10 µM and 100 µM increases stretch-induced afferent
firing in muscle spindles. (B) This effect is blocked by PCCG-13 but not
NBQX, showing kainate’s actions are mediated by the PLD-mGluR and
not the iGluR kainate receptor.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 9638–9643 | 9641

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 7 Compound 1c increases stretch-induced afferent firing in
muscle spindles by acting at the PLD-mGluR. (A) and (B) Neither com-
pound 1a or 1b are pharmacologically active on muscle spindle stretch-
evoked activity. (C) In contrast, 1c increases firing at 1 µM and above. (D)
Even at 10 µM, the enhancement by 1c is blocked by PCCG-13.
Scheme 7 Synthesis of biotinylated compound 14. Reagents and con-
ditions: (i) N-hydroxysuccinimide, EDC·HCl, DMF, rt, 24 h; (ii) O-(2-amino-
ethyl)-O’-(2-azidoethyl)pentaethylene glycol, TEA, DMF, rt, 24 h (98% in
two steps); (iii) alkyne 9, tBuOH/H2O/THF, CuSO4, Na ascorbate, rt, 36 h,
(83%); (iv) H3PO4 85% wt in H2O, toluene, rt, 2 h; (v) LiOH 0.5N in H2O, rt,
12 h (50% in two steps).
Biological evaluation
As shown in Fig. 6, and discussed above, (−)-kainic acid
potently increases stretch induced afferent firing in muscle
spindles. This action is not inhibited by the ionotropic kainic
acid receptor agonist NBQX, but is inhibited by the specific
PLD mGluR antagonist PCCG-13. These data suggest
(−)-kainic acid is acting via the PLD mGluR in muscle spindles
to increase afferent firing.
Fig. 7 shows that compounds 1a (n = 8) (Fig. 7A) and 1b (n =
8) (Fig. 7B) are completely inactive in muscle spindles while
compound 1c has a potency similar to that of (−)-kainic acid
(n = 8) (Fig. 7C). Like (−)-kainic acid the ability of 1c to
increase stretch-induced afferent firing is inhibited by
PCCG-13 (n = 6) (Fig. 7D), suggesting it also acts via the PLD
mGluR.
Compound 14’s biological activity differed slightly from
that of 1c, as it had a bell-shaped response curve, perhaps
reflecting receptor desensitisation or internalisation at higher
concentrations. Thus, firing rates increased significantly at
1 µM (n = 8) but not at 10 µM (n = 8) (Fig. 8). This retention of
pharmacological action at modest ligand concentrations
means that compound 14 will be a useful tool in further inves-
tigations into the receptor.
Further development
To achieve further insight into the pharmacological properties
of the most active analogue 1c, with the idea of using it for a
pull-down experiment on the target mGlu receptor, we syn-
thesised a biotinylated version 14 (Scheme 7) of compound 1c
carrying a PEG chain as a spacer between the biotin and the
triazole fragments. The synthesis started with the formation of
the activated biotin ester 11 (N-hydroxysuccinimide and
EDC·HCl)17 that was directly coupled with the amino linker,
affording the azide 12. The following click reaction18 provided
the 1,2,3-triazole 13 that was fully deprotected to give the
target biotinylated compound 14.
Fig. 8 Compound 14 reversibly enhances stretch-evoked spindle firing.
There is a bell-shaped response, with 1 μM producing a robust enhance-
ment. Increasing concentration further to 10 μM returned firing to
baseline.
9642 | Org. Biomol. Chem., 2014, 12, 9638–9643
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
7 S. Gupta and C. E. Schafmeister, J. Org. Chem., 2009, 74,
3652–3658.
Conclusions
To investigate the pharmacology of the hitherto uncloned
phospholipase D-coupled mGlu receptor, we undertook a
modification of the C4 position of (−)-kainic acid, which is
known to selectively and potently increase muscle spindle
stretch-induced afferent firing. Highly stereoselective trifluoro-
methylation, hydride reduction/O-methylation and alkynyla-
tion/“click” reaction, all occurring from the Re-face of the key
common C4-oxo intermediate 5a, produced the three kainic
analogues 1a–c. Only the triazole compound 1c retained the
kainate-like agonism, increasing firing in a dose-dependent
manner. The further modification of 1c by introduction of a
PEG-biotin chain on the C4-1,2,3-triazole fragment afforded
compound 14 which retains robust agonism at 1 μM, although
care must be taken at higher concentrations, due to the bell-
8 (a) R. Sharma and W. D. Lubell, J. Org. Chem., 1996, 61,
202–209; (b) M.-J. Blanco, M. R. Paleo, C. Penide and
F. J. Sardina, J. Org. Chem., 1999, 64, 8786–8793; (c) For a
review on the utility of the Phenylfluorenyl group see:
E. J. Karppanen and A. M. P. Koskinen, Molecules, 2010, 15,
6512–6547.
9 (a) M. Barker, M. Clackers, R. Copley, D. A. Demaine,
D. Humphreys, G. G. A. Inglis, M. J. Johnston, H. T. Jones,
M. V. Haase, D. House, R. Loiseau, L. Nisbet, F. Pacquet,
P. A. Skone, S. E. Shanahan, D. Tape, V. M. Vinader,
M. Washington, I. Uings, R. Upton, I. M. McLa and
S. Macdonald, J. Med. Chem., 2006, 49, 4216–4231;
(b) G. K. Surya, A. Prakash and K. Yudin, Chem. Rev., 1997,
97, 757–786.
shaped response. However, notably biotinylation did not 10 J. E. Baldwin, S. J. Bamford, A. M. Fryer, M. P. W. Rudolph
destroy the receptor binding capabilities, meaning compound and M. E. Wood, Tetrahedron, 1997, 53, 5255–5272.
14 is suitable for use in pull-down and other assays, such as 11 B. Li, M. Berliner, R. Buzion, C. K.-F. Chu, S. T. Colgan,
far western blotting, for PLD-mGluR isolation.
T. Kaneko, N. Keene, W. Kissel, T. Ler, K. R. Leeman,
B. Marquez, R. Morris, L. Newell, S. Wunderwald, M. Witt,
J. Weaver and Z. Zhang, J. Org. Chem., 2006, 71, 9045–9050.
12 D. Neuhaus and M. P. Williamson, The Nuclear Overhauser
Effect in Structural and Conformational Analysis, Wiley-VCH,
New York, 2nd edn, 2000.
Acknowledgements
SW was supported by a SULSA Bioskape/Eli Lilly studentship.
13 (a) J. E. Baldwin, G. J. Pritchard and D. S. Williamson,
Tetrahedron, 2001, 57, 7991–7997; (b) J. E. Baldwin,
G. J. Pritchard and D. S. Williamson, Bioorg. Med. Chem.
Lett., 2000, 10, 1927–1929; (c) H. L. Hao, P. Sun,
G. X. Wang, K. Fruta and M. Suzuki, Chin. Chem. Lett.,
2008, 19, 269–272; (d) K. Furuta, G. X. Wang, T. Minami,
M. Nishizawa, S. Ito and M. Suzuki, Tetrahedron Lett., 2004,
45, 3933–3936.
Notes and references
1 S. Watson, C. Zanato, S. Dall’Angelo, R. W. Banks, I. Greig,
M. Zanda and G. S. Bewick, Proc. Physiol. Soc., 2012, 27,
PC46.
2 D. E. Pellegrini-Giampietro, S. A. Torregrossa and F.
Moroni, Br. J. Pharmacol., 1996, 118, 1035–1043.
3 S. Albani-Torregrossa, S. Attucci, M. Marinozzi,
14 R. Delatouche, M. Mondon, A. Gil, G. Frapper, C. Bachmann
and P. Bertrand, Tetrahedron, 2011, 67, 401–407.
R. Pellicciari, F. Moroni and D. E. Pellegrini-Giampietro, 15 N. G. Aher, V. S. Pore, A. Kumar, P. K. Shukla, A. Sharma
Mol. Pharmacol., 1999, 55, 699–707.
and M. K. Bhat, Bioorg. Med. Chem. Lett., 2009, 19, 759–763.
4 P. J. Conn and J.-P. Pin, Annu. Rev. Pharmacol. Toxicol., 16 CCDC 1019225 contains the supplementary crystallo-
1997, 37, 205–237.
graphic data for this paper.
5 J. E. Baldwin, S. J. Bamford, A. M. Fryer, M. P. W. Rudolph 17 P.-C. Lin, S.-H. Ueng, S.-C. Yu, M.-D. Jan, A. K. Adak,
and M. E. Wood, Tetrahedron, 1997, 53, 5233–5254.
C.-C. Yu and C.-C. Lin, Org. Lett., 2007, 11, 2131–2134.
6 J.-F. Poisson, A. Orellana and A. E. Green, J. Org. Chem., 18 H. S. Tae, J. Hines, A. R. Schneekloth and C. M. Crews, Org.
2005, 70, 10860–10863.
Lett., 2010, 12, 4308–4311.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 9638–9643 | 9643