Anionic cyclophanes as potential reversal agents of muscle relaxants by chemical chelation

Article abstract of DOI:10.1016/S0960-894X(02)00017-3

A series of carboxyl-containing cyclophanes have been designed and synthesised as chemical chelators (or host molecules) of cationic muscle relaxant drugs (or guest molecules). Three of these cyclophane derivatives, 1-3, have been shown by NMR to form 1:1 complexes with the muscle relaxants pancuronium, 4 and gallamine, 5 in D₂O, with association constants up to 10⁴ M^-1. When tested in an in vitro chick biventer muscle preparation, the cyclophanes reversed the neuromuscular block induced by pancuronium and gallamine, with 3 having the most effective reversal against pancuronium (EC₅₀ 40 μM).

Full text of DOI:10.1016/S0960-894X(02)00017-3

Bioorganic & Medicinal Chemistry Letters 12 (2002) 753–755
Anionic Cyclophanes as Potential Reversal Agents of
Muscle Relaxants by Chemical Chelation
Kenneth S. Cameron,c Lee Fielding,c Rona Mason,b Alan W. Muir,b
David C. Rees,a,* Simon Thorna and Ming-Qiang Zhanga
aDepartment of Medicinal Chemistry, Organon Laboratories Ltd., Newhouse, Lanarkshire ML1 5SH, Scotland, UK
bDepartment of Pharmacology, Organon Laboratories Ltd., Newhouse, Lanarkshire ML1 5SH, Scotland, UK
cDepartment of Analytical Chemistry, Organon Laboratories Ltd., Newhouse, Lanarkshire ML1 5SH, Scotland, UK
Received 23 October 2001; accepted 14 December 2001
Abstract—A series of carboxyl-containing cyclophanes have been designed and synthesised as chemical chelators (or host mole-
cules) of cationic muscle relaxant drugs (or guest molecules). Three of these cyclophane derivatives, 1–3, have been shown by NMR
to form 1:1 complexes with the muscle relaxants pancuronium, 4 and gallamine, 5 in D2O, with association constants up to 104
M
ꢀ1. When tested in an in vitro chick biventer muscle preparation, the cyclophanes reversed the neuromuscular block induced by
pancuronium and gallamine, with 3 having the most effective reversal against pancuronium (EC50 40 mM). # 2002 Elsevier Science
Ltd. All rights reserved.
Inspired by the fascinating developments in the area of
supramolecular chemistry,1 the aim of this study was to
investigate whether chemical chelation (or complexation,
encapsulation, sequestration) of muscle relaxant drugs
could be used as a mechanism for reversing their phar-
macological actions.
example, bradycardia, hypotension, increased salivation
and so on. Therefore, in practice, these reversal agents
are usually used in combination with a mAChR
antagonist such as atropine which itself has also a
number of side effects, for example dry mouth, blurred
vision, tachycardia, and so on.
Skeletal muscle relaxants (neuromuscular blocking
agents, NMBA’s) are widely used surgically during
general anesthesia particularly to paralyse laryngeal
muscles prior to airway intubation and also to paralyse
other skeletal muscles before general surgery.2 Most
currently used muscle relaxants in surgery today achieve
their effect by blocking the physiological effects of acetyl-
choline (ACh) at nicotinic acetylcholine receptors
(nAChR) on skeletal muscle. To reverse the blockade in
patients, for example to assist recovery of muscle func-
tion after surgery, it is most common to administer
acetylcholinesterase (AChE) inhibitors (e.g., neostig-
mine, pyridostigmine or edrophonium). However, this
mechanism of action of the reversal agent increases
ACh levels induced by AChE inhibition and leads to
non-selective activation of muscarinic acetylcholine
receptors (mAChR) causing many side effects, for
We hypothesised that chemical chelation of NMBAs by
an exogenous host molecule would reverse neuromus-
cular block. Since this mechanism of action does not
involve direct activation of cholinergic systems it would
be expected to circumvent the undesired clinical side
effects of the currently used AChE inhibitors.
Complexation of quaternary ammonium guests by
negatively charged cyclophanes is known in the litera-
ture.3 Also, complexation of steroidal guests (neutral or
negatively charged) with cyclophanes is known.4 We
sought to synthesise negatively charged cyclophanes 1–3
and to investigate their chelation of quaternary ammo-
nium NMBAs such as the aminosteroid pancuronium 4
and gallamine 5 (Fig. 1).
The carboxyls on the periphery of the cyclophanes 1–3
were designed to provide potential for electrostatic
interaction with the quaternary ammoniums as well as
water-solubility (required for intravenous administra-
tion). The rigid cyclophane backbones provide a
*Corresponding author at present address: Medical Chemistry Dept.
AstraZeneca, R&D Molndal, S-431 83, Sweden. Fax: +46-31-776-
3868; e-mail: david.c.rees@astrazeneca.com
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00017-3

754
K. S. Cameron et al. / Bioorg. Med. Chem. Lett. 12 (2002) 753–755
hydrophobic cavity for NMB (guest) chelation in an
aqueous environment. The cyclophanes 1–3 were syn-
thesised as outlined in Scheme 1. The commercially
available 5,50-methylenedisalicylic acid 6 was protected
as the methyl ester 7. Bis alkylation of the phenol to
give 8 was followed by macrocyclisation with the phenol
7 using syringe pump techniques to give the cyclophane
9. With the increasing hydrophobicity of the macro-
cycles, the final hydrolysis step became more difficult,
with longer reaction times and use of DMSO as co-sol-
vent in order to obtain 1–3.5
reported in literature.7 The values in Table 1 are the aver-
age from independent observations of several protons of
the muscle relaxants. For pancuronium, 4 the chemical
shift changes of both A-ring and D-ring N–CH3 were
observed. In addition, the changes of D-ring acetate
CH3 were also clearly visible in the titration with 2 and
3 and changes of 19-CH3 were observed during titration
with 1 and 3. In the case of gallamine binding, all three
cyclophanes formed symmetrical complexes as seen
1
from the retention of the 2:1 ratio of signals in the H
NMR spectra. This implies that complexation takes
place with the central ammonium ion to give an axis of
symmetry.
The binding affinity of the cyclophanes 1–3 with the
muscle relaxants pancuronium, 4 and gallamine, 5 was
1
determined by H NMR (400 MHz). Titration of phos-
The NMR binding affinity of cyclophanes 1–3 with the
NMBs (Table 1) correlates with the size and the aro-
matic area of the cyclophane ring, that is 3>2>1. This
is in agreement with the previous observation that host
molecules with a high aromatic content tend to have
high affinity for quaternary ammonium compounds,3aꢀc
probably due to an ion-dipole attraction between the
positive charge of the ammonium and the polarisable p
bonds of the aromatic parts of the host. In addition, the
increased aromatic area in the cyclophane ring increases
the lipophilicity of the cavity, which favours interaction
with the lipophilic part of the guest molecules, for
phate buffered (pH 7.6) D2O solutions of 4 or 5 with the
cyclophanes 1–3 (0–4 mM) resulted in concentration
dependent changes of chemical shifts of protons of the
muscle relaxants. Job plot analysis by the method of
continuous variation6 indicates that all three cyclo-
phanes form 1:1 complexes with 4 and 5. The associa-
tion constants (Ka’s, Table 1) were calculated by an in-
house curve fitting method that closely resembles those
Table 1. Association constants of cyclophanes 1–3 to the muscle
relaxants 4 and 5, and the reversal of 4-and 5-induced muscle relaxa-
tion by 1–3
Compd
Pancuronium (4)
Gallamine (5)
Assoc.
constanta
Ka Mꢀ1
Reversal
EC50, mMb
Assoc.
constanta
Ka, Mꢀ1
Reversal
EC50, mMb
1
2
3
167
2500
10,450
142
292
40
608
3275
13,200
1030
563
104
aMeans of at least three independent calculations based on chemical
shift changes of different protons of 4 or 5.
bConcentrations that produced 50% reversal of pancuronium-or gal-
lamine-induced neuromuscular block in the chick biventer assay.10
Data are means of two independent determinations.
Figure 1. Structures of target cyclophanes (1–3) and muscle relaxants
pancuronium (4) and gallamine (5).
Scheme 1. Synthesis of anionic cyclophanes 1–3. Reagents and conditions: (i) MeOH, HCI, reflux, 11 h, (22%); (ii) K2CO3, Br(CH2)5Br, DMF, 1 h
(60%) or NaH, BrCH2ArCH2Br, 50 ꢁC; (iii) 7 (over 3 h), K2CO3 10 equiv (for 8a), DMF, 80 ꢁC, 4.5 h (10%) or NaH (for 8b and 8c), DMF, 120 ꢁC
(10–25%); (iv) for 1: NaOH 20 equiv, MeOH–H2O (3:1), reflux, 3 h (45%); for 2: NaOH, MeOH–H2O–DMSO (1:1:1), 60 ꢁC, 12 h; for 3: NaOH,
THF–DMSO–H2O (1:1:1), 60 ꢁC, 48 h; (v) HCI (40%).

K. S. Cameron et al. / Bioorg. Med. Chem. Lett. 12 (2002) 753–755
755
example, the steroid backbone of pancuronium 4. Pre-
References and Notes
vious work in the area of crown ethers8 and work by
Koga9 has shown that preorganisation favours good
binding. The increased rigidity and therefore better
defined cavity from 1 to 3 may also contribute to the
increased affinity from 1 to 3 for the guest molecules.
1. Balzani, V.; Credi, A.; Raymo, F. M.; Fraser Stoddart, J.
Angew. Chem., Int. Ed. Engl. 2000, 39, 3348.
2. Mirakhur, R. K. Acta. Anaesthesiol. Scand. Suppl. 1995,
106, 62.
3. (a) Petti, M. A.; Shepodd, T. J.; Barrans, R. E., Jr.;
Dougherty, D. A. J. Am. Chem. Soc. 1988, 110, 6825. (b)
Kearney, P. C.; Mizoue, L. S.; Kumpf, R. A.; Forman, J. E.;
McCurdy, A.; Dougherty, D. A. J. Am. Chem. Soc. 1993, 115,
9907. (c) Dougherty, D. A. Science 1996, 271, 163. (d) Meric,
R.; Lehn, J.-M.; Vigneron, J.-P. Bull. Soc. Chim. Fr. 1994, 131,
579. (e) Scneider, H.-J.; Guttes, D.; Schneider, U. Angew.
Chem., Int. Ed. Engl. 1986, 25, 647. (f) Garel, L.; Lozach, B.;
Dutasta, J.-P.; Collet, A. J. Am. Chem. Soc. 1993, 115, 11652.
4. (a) Peterson, B. R.; Diederich, F. Angew. Chem., Int. Ed.
Engl. 1994, 33, 1625. (b) Peterson, B. R.; Wallimann, P.; Car-
canague, D. R.; Diederich, F. Tetrahedron 1995, 51, 401. (c)
Peterson, B. R.; Mordasini-Denti, T.; Diederich, F. Chemistry
and Biology 1995, 2, 139. (d) Wallimann, P.; Seiler, P.; Die-
derich, F. Helv. Chim. Acta 1996, 79, 779. (e) Breslow, R.;
Duggan, P. J.; Wiedenfeld, D.; Waddell, S. T. Tetrahedron
Lett. 1995, 36, 2707.
Encouraged by these results we then investigated the
ability of these guest molecules to reverse neuromus-
cular block in vitro as determined in the isolated chick
biventer cervicis muscle preparation.10 In these experi-
ments, reversal activity was measured against an ꢂ90%
block of muscle twitch induced by cumulative additions
of pancuronium, 4 or gallamine, 5. The concentrations
of 1–3, which produced 50% reversal (EC50), are listed
in Table 1. For comparison, neostigmine, which is
commonly used surgically as a NMB reversal agent via
AChE inhibition, reverses vecuronium-induced block in
chick biventer with an EC50 of 0.028 mM.
Although much less potent than the standard AChE
inhibitor neostigmine, these cyclophane-based chelators
do produce significant reversal of pancuronium, 4 or
gallamine, 5 induced neuromuscular block in vitro.
When cyclophane 3 was tested at the higher concentra-
tion of 2.3 mM it produced almost full reversal (>90%)
of both 4-and 5-induced block in this chick biventer
preparation.11
5. Compound 1: MS (EI) m/z 711 (MꢀH); 1H NMR (DMSO)
d 1.51 (m, 4H), 1.68 (m, 8H), 3.81 (s, 4H), 3.99 (t, J=5.6 Hz,
8H), 6.96 (m, 4H), 7.20 (m, 4H), 7.43 (m, 4H), 12.25 (br s,
4H); 13C NMR (DMSO) d 21.57, 27.81, 38.72, 68.47, 113.99,
121.77, 130.28, 132.60, 133.21, 155.61, 167.27. Compound 2:
1
MS (EI) m/z 779 (MꢀH); H NMR (DMSO) d 3.75 (s, 4H),
5.15 (s, 8H), 6.91 (d, J=8.58 Hz, 4H), 7.20 (m, 4H), 7.32 (m,
8H), 7.49 (m, 4H), 12.40 (br s, 4H); 13C NMR (DMSO) d
68.69, 84.57, 114.52, 126.54, 129.98, 132.32, 133.81, 136.28,
154.60, 167.50. Compound 3: MS (EI) m/z 879 (MꢀH); 1H
NMR (DMSO) d 3.76 (s, 4H), 5.30 (s, 8H), 6.90 (m, 4H), 7.19
(m, 4H), 7.49 (m, 8H), 7.74 (m, 4H), 7.83 (m, 4H), 12.60 (br s,
4H); 13C NMR (DMSO) d 69.21, 83.84, 114.73, 121.88,
125.25, 125.37, 127.96, 130.34, 132.04, 132.67, 133.97, 135.03,
154.91, 167.52.
The cyclophanes 1–3 have also been examined for their
intrinsic activities on isolated muscle by cumulative
addition (1 mM–1 mM) to the tissue bath of unblocked
chick biventer muscles. None of the compounds mod-
ified the height of the electrically-induced twitch
response or affected the resting baseline tension of the
biventer preparation.12
6. Connors, K. A. Binding Constants, The Measurement of
Molecular Complex Stability; Wiley-Interscience: New York,
1987, p 24.
7. (a) Loukas, Y. L. J. Pharm. Pharmacol. 1997, 49, 944. (b)
Bisson, A. P.; Hunter, C. A.; Morales, J. C.; Young, K. Chem.
Eur. J. 1998, 4, 845.
8. Cram, D. J. Angew. Chem., Int. Ed. Engl. 1986, 25, 1039.
9. Odashima, K.; Soga, T.; Koga, K. Tetrahedron Lett. 1981,
22, 5311.
Taken together, these results suggest that the reversal
activity of these cyclophanes seems to stem from their
ability to form complexes with the muscle relaxants.
For example the cyclophane 3 with the highest affinity to
the blockers has the highest reversal potency, although
there are some discrepancies of 1 and 2 in this respect.
10. Ginsborg, B. L.; Warriner, J. Br. J. Pharmacol. 1960, 5, 410.
11. An attempt to determine the in vivo reversal activity of
compound 3 in an anesthetised guinea-pig failed due to insuf-
ficient potency.
12. An anti-AChE component to the reversal activity of com-
pounds 1–3 is ruled out because they show no AChE inhibi-
tion when tested in concentrations up to 50 mM (cf.
neostigmine IC50 0.6 mM).
13. Note added in proof: see Bom, A.; Bradley, M.; Cameron,
K; Clark, J. K.; van Egmond, J.; Fielden, H.; MacLean, E. J.;
Muir, A. W.; Palin, R.; Rees, D. C.; Zhang, M.-Q. Angew.
Chem. Int. Ed. Engl. 2002, in press.
In summary, this study aims at reversing the pharmaco-
logical effects of NMBAs by a mechanism involving che-
mical chelation (or sequestration). The cyclophane 3 binds
to pancuronium 4 in an aqueous environment with an
NMR binding constant of 104 and it reverses the biologi-
cal effect of 4 and 5 in an in vitro muscle preparation.
These results came from the first attempts in our labora-
tories to design and synthesise host molecules acting as
reversal agents for NMBAs. Subsequent studies, which
further develop the concept, are reported elsewhere.13