Assessment of false transmitters as treatments for nerve agent poisoning

Article abstract of DOI:10.1016/j.toxlet.2019.12.010

Nerve agents inhibit acetylcholinesterase (AChE), leading to a build-up of acetylcholine (ACh) and overstimulation at cholinergic synapses. Current post-exposure nerve agent treatment includes atropine to treat overstimulation at muscarinic synapses, a benzodiazepine anti-convulsant, and an oxime to restore the function of AChE. Aside from the oxime, the components do not act directly to reduce the overstimulation at nicotinic synapses. The false transmitters acetylmonoethylcholine (AMECh) and acetyldiethylcholine (ADECh) are analogs of ACh, synthesised similarly at synapses. AMECh and ADECh are partial agonists, with reduced activity compared to ACh, so it was hypothesised the false transmitters could reduce overstimulation. Synthetic routes to AMECh and ADECh, and their precursors, monoethylcholine (MECh) and diethylcholine (DECh), were devised, allowing them to be produced easily on a laboratory-scale. The mechanism of action of the false transmitters was investigated in vitro. AMECh acted as a partial agonist at human muscarinic (M₁ and M₃) and muscle-type nicotinic receptors, and ADECh was a partial agonist only at certain muscarinic subtypes. Their precursors acted as antagonists at muscle-type nicotinic, but not muscarinic receptors. Administration of MECh and DECh improved neuromuscular function in the soman-exposed guinea-pig hemi-diaphragm preparation. False transmitters may therefore help reduce nerve agent induced overstimulation at cholinergic synapses.

Full text of DOI:10.1016/j.toxlet.2019.12.010

Journal Pre-proof
Assessment of false transmitters as treatments for nerve agent poisoning
Charlotte Whitmore, Christopher D. Lindsay, Mike Bird, Samuel J.
Gore, Helen Rice, Rebecca L. Williams, Christopher M. Timperley,
A. Christopher Green
PII:
S0378-4274(19)30397-2
DOI:
https://doi.org/10.1016/j.toxlet.2019.12.010
TOXLET 10649
Reference:
To appear in:
Toxicology Letters
Received Date:
Revised Date:
Accepted Date:
30 August 2019
4 December 2019
8 December 2019
Please cite this article as: Whitmore C, Lindsay CD, Bird M, Gore SJ, Rice H, Williams RL,
Timperley CM, Green AC, Assessment of false transmitters as treatments for nerve agent
poisoning, Toxicology Letters (2019), doi: https://doi.org/10.1016/j.toxlet.2019.12.010
This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.
©
2019 Published by Elsevier.

Assessment of false transmitters as treatments for nerve agent poisoning
Charlotte Whitmore*, Christopher D. Lindsay, Mike Bird, Samuel J. Gore, Helen Rice,
Rebecca L. Williams, Christopher M. Timperley, A. Christopher Green
Chemical, Biological and Radiological Division, Defence Science and Technology Laboratory
(Dstl), Porton Down, Salisbury, Wiltshire SP4 0JQ, UK
*
Corresponding author: C. Whitmore. E-mail: clwhitmore@dstl.gov.uk
Highlights


AMECh and ADECh act as false transmitters at human acetylcholine receptors.
Administration of MECh and DECh led to recovery in a soman-exposed functional
neuromuscular preparation.

Consistent with activity as false transmitters, the functional recovery was maintained after
washout.
Abstract
Nerve agents inhibit acetylcholinesterase (AChE), leading to a build-up of acetylcholine
(ACh) and overstimulation at cholinergic synapses. Current post-exposure nerve agent
treatment includes atropine to treat overstimulation at muscarinic synapses, a
benzodiazepine anti-convulsant, and an oxime to restore the function of AChE. Aside from
the oxime, the components do not act directly to reduce the overstimulation at nicotinic
synapses. The false transmitters acetylmonoethylcholine (AMECh) and acetyldiethylcholine
(ADECh) are analogs of ACh, synthesised similarly at synapses. AMECh and ADECh are partial
agonists, with reduced activity compared to ACh, so it was hypothesised the false
transmitters could reduce overstimulation.
1

Synthetic routes to AMECh and ADECh, and their precursors, monoethylcholine (MECh) and
diethylcholine (DECh), were devised, allowing them to be produced easily on a laboratory-
scale. The mechanism of action of the false transmitters was investigated in vitro. AMECh
acted as a partial agonist at human muscarinic (M
1
and M ) and muscle-type nicotinic
3
receptors, and ADECh was a partial agonist only at certain muscarinic subtypes. Their
precursors acted as antagonists at muscle-type nicotinic, but not muscarinic receptors.
Administration of MECh and DECh improved neuromuscular function in the soman-exposed
guinea-pig hemi-diaphragm preparation. False transmitters may therefore help reduce
nerve agent induced overstimulation at cholinergic synapses.
1
Introduction
Nerve agents inhibit the enzyme acetylcholinesterase (AChE), which results in an
accumulation of acetylcholine (ACh) at cholinergic synapses, including the neuromuscular
junction. This build-up of ACh causes the toxic effects of nerve agent poisoning, including
but not limited to muscle tremors, fatigue, fasciculations and secretions. There are two
types of ACh receptors in the body, muscarinic ACh receptors (mAChR) which activate G-
proteins and nicotinic ACh receptors (nAChR) which are ligand-gated ion channels. Current
therapeutic approaches to nerve agent poisoning include atropine to reduce the effect of
excess ACh at mAChR, an anti-convulsant to reduce or prevent seizures and an oxime to
restore AChE activity, so it hydrolyses ACh again and clears excess ACh from the synapse.
These countermeasures do not contain any compounds acting directly at nAChR. Thus, aside
from the indirect action of an oxime to reactivate AChE, there is no specific therapy
targeting overstimulation of the nAChR. This can be problematic because the ability of the
oxime to reactivate nerve agent-inhibited AChE varies depending on the oxime, the nerve
agent itself and the time interval between the exposure and the treatment (1). Thus there is
an opportunity to improve nerve agent treatment by identifying new therapeutics which can
reduce the effects of excess ACh at nAChRs. Such therapies, as they are independent of
effects on AChE, would be broad-spectrum, offering the same benefit regardless of the
2

nerve agent used. Previously it has been shown that administration of a nAChR antagonist,
such as the bis-pyridinium compound MB327, is efficacious against nerve poisoning,
although there are drawbacks, as MB327 is toxic itself at similar doses to those which result
in efficacy (2-4). However, these results suggest that alternative therapeutic strategies
which achieve a similar result (reduced effectiveness of nAChR activation) could offer
benefits in treating nerve agent poisoning.
ACh is synthesised in the presynaptic nerve terminal by the enzyme choline
acetyltransferase (ChAT) from choline and acetyl coenzyme A (acetyl-CoA).
Monoethylcholine (MECh) and diethylcholine (DECh) are structural analogs of choline
(Figure 1). Like choline, MECh and DECh can be transported into the presynaptic nerve
terminal and acetylated by the same enzyme to give acetylmonoethylcholine (AMECh) and
acetyldiethylcholine (ADECh) respectively (5-8).
Both AMECh and ADECh are capable of activating postsynaptic AChR if released from the
nerve terminal, although as partial agonists they generate reduced receptor activation
compared to ACh, approximately 40% lower (9, 10). Thus it is said that AMECh and ADECh
act as ‘false transmitters’ and a schematic of this process is shown in Figure 2. In addition to
their use as pharmacological tools for investigating synaptic function and signalling (9), the
false transmitters have been investigated for their potential to decrease cholinergic
transmission (11). Since they are partial agonists of nAChR, it was hypothesised they might
offer therapeutic benefit after nerve agent poisoning by reducing overstimulation of the
nAChR and by normalising the kinetics of endplate potentials at synapses without functional
AChE (10, 12). Indeed, MECh administration has already been shown to be effective in
reducing the mortality caused by the anticholinesterases physostigmine, diisopropyl
fluorophosphate (DFP), and diazinon in mice (13).
First, the synthesis routes for the false transmitters and precursors were reviewed, and then
modified to make them more amenable to laboratory-scale synthesis. Secondly, the agonist
activity of AMECh and ADECh at human AChR was assessed in vitro, using an indirect
calcium fluorescence assay. Thirdly, as choline is known to act as an antagonist at AChR, the
ability of MECh and DECh to antagonise signalling at AChRs was investigated. To date,
published work on the false transmitters has described their activity at AChR from various
3

species including mouse, frog, rat and Torpedo, but not human (6, 9, 14). To cover a range of
different AChR types and subtypes, three different cell lines have been used – CN21 cells
which express both the human muscle-type nAChR (both αβγδ and αβδε) and M
15-17) and CHOm1 and CHOm3 which express the human M and M mAChR subtypes,
respectively. In these fluorescence assays, the addition of an agonist activates the receptor,
3
mAChR
(
1
3
2
+
facilitating an increase of intracellular calcium ion concentration ([Ca ] ) which is quantified
i
as a fluorescent signal. Addition of a partial agonist, such as a false transmitter, would
generate smaller increases than those observed after application of a full agonist like ACh.
Similar assays were also used to determine whether MECh and DECh can act as antagonists
at AChR. Finally, to assess whether administration of the false transmitters could restore
neuromuscular function, MECh and DECh were added to soman-exposed neuromuscular
preparations (guinea-pig phrenic nerve/hemi-diaphragm).
Figure 1. The biological conversion of choline and analogues to their acetyl derivatives.
Choline is converted in the nerve terminal to acetylcholine (ACh) by reaction with acetyl
coenzyme A (acetyl-CoA), catalysed by choline acetyltransferase (ChAT). The coenzyme A
that is generated during this process (not shown) can be re-acetylated in the cell, ready for
another cycle. Monoethylcholine (MECh) and diethylcholine (DECh) are acetylated by this
system to generate the false transmitters: acetylmonoethylcholine (AMECh) and
acetyldiethylcholine (ADECh). Counterions are not shown for ease of structural comparison:
the natural counter-ion of choline and ACh is hydroxide. Choline hydroxide is a non-
4

crystallisable, strongly alkaline syrupy liquid, which deliquesces in air and is soluble in all
proportions in water (18). Choline halides crystallise and are easier to handle. The ACh used
in the study described herein was the chloride salt. MECh and DECh were iodide salts.
5

Figure 2. A schematic of the neuromuscular junction (NMJ) showing the cycle of ACh during neuromuscular transmission (control), how this is
affected by the presence of nerve agents, and the proposed mechanism for using false transmitters as a nerve agent therapeutic. (Left image)
During neuromuscular transmission, choline is transported into the pre-synaptic terminal of the NMJ (1) and combines with acetyl CoA to
synthesise ACh (2), which is then packaged into vesicles for release (3). In response to stimulation, the vesicles are released, ACh diffuses
across the synaptic cleft and activates the nAChR leading to muscle contraction (4). Acetylcholinesterase (AChE) breaks down ACh, terminating
its action at the receptor (5) and generating choline, which is taken up into the pre-synaptic terminal (1). (Middle image) In the presence of
nerve agent, AChE activity is inhibited, resulting in a build-up of ACh at the NMJ and over-stimulation of the muscle (4). (Right image) In the
absence of choline, exogenous monoethylcholine (MECh) can be transported into the synaptic cleft (1) and combined with acetyl CoA to
generate acetylmonoethylcholine (AMECh) (2). Similarly to ACh, AMECh can be packaged into vesicles (3) and released into the synaptic cleft.
As a partial agonist, AMECh is predicted to activate the receptor with reduced activity, decreasing the overstimulation at the muscle (4), and to
have a therapeutic action.
6

2
Materials and methods
2
.1 Materials
MECh, AMECh, DECh and ADECh were synthesised by the methods described below and
solutions at the required concentrations were freshly prepared on the day of the
experiment. Soman was synthesised at Dstl Porton Down and supplied at >95% purity
diluted in isopropanol. The Molecular Devices FLIPR 5 calcium kit (ThermoFischer Scientific
2
+
Ltd., UK) was used for the [Ca ] assays. On the day of the assay, the calcium dye was made
i
up according to the manufacturer’s instructions. All reagents and solvents were from Sigma-
Aldrich Ltd. (Dorset, UK), unless otherwise stated.
2
.2 Chemical synthesis
Anhydrous solvents were used for synthesis: tetrahydrofuran (THF) and diethyl ether
hereinafter ‘ether’). This was important as the quaternary ammonium salts prepared
(
deliquesced and care had to be taken to exclude water during their production. Melting
points were measured using an Electrothermal apparatus and are uncorrected (dec. stands
for ‘with decomposition’). Nuclear magnetic resonance (NMR) spectroscopic data were
collected at 9.4 T using a Bruker Avance III spectrometer equipped with a 5 mm BBFO+
probehead, as solutions in deuterated dimethylsulfoxide (d
6
-DMSO) or deuteriochloroform
(
CDCl ).
3
Caution! The compounds synthesised have marked physiological action and should be
handled cautiously in a fume cupboard. Eye protection and gloves should be worn during all
manipulations, and glassware decontaminated in dilute aqueous bleach.
Monoethylcholine iodide (MECh). Iodoethane (22.5 ml, 280 mmol) was added dropwise over
2
5 min to a stirred solution of 2-(dimethylamino)ethanol (17.8 g, 200 mmol) in THF (200 ml).
During addition the reaction mixture coloured yellow and a thick white precipitate formed.
The mixture was stirred at room temperature overnight. The precipitate was filtered off,
rinsed with THF, and quickly transferred to a flask as it deliquesced readily. It was dried to a
constant weight (44.2 g) under high vacuum. The solid was recrystallized from hot
7

isopropanol (about 200 ml). The white crystals were transferred to a flask to avoid
deliquescence and dried under vacuum to constant weight (35.9 g, 73%). Mp 272 °C. 1H
NMR (600 MHz, d
.3 Hz), 3.38-3.35 (2H, m), 3.05 (6H, s), 1.24 (3H, tt, J = 1.7, 7.3 Hz). ¹³C NMR (151 MHz, d
DMSO) 64.5 (t, J = 2.7 Hz), 60.1 (t, J = 2.7 Hz), 55.4, 50.7 (t, J = 3.8 Hz), 8.4 ppm.
6
-DMSO) 5.25 (1H, dd, J = 4.9, 4.9 Hz), 3.84-3.80 (2H, m), 3.41 (2H, q, J =
7
6
-
Diethylcholine iodide (DECh). Iodomethane (8.7 ml, 140 mmol) was added dropwise over 15
min to a stirred solution of 2-(diethylamino)ethanol (11.7 g, 100 mmol) in THF (100 ml).
During addition the reaction mixture coloured yellow and a white precipitate formed. The
mixture was stirred at room temperature overnight. The white solid was transferred to a
flask to avoid deliquescence, and dried under high vacuum to constant weight (25.4 g). The
solid was recrystallized from hot isopropanol (about 80 ml). The white crystals were filtered
off, rinsed with isopropanol, quickly diverted into a flask, and dried under vacuum to
constant weight (22.6 g, 87%). Mp 276 °C (dec.). ¹H NMR (600 MHz, d
6
-DMSO) 5.24 (1H, t, J =
5
1
5
.0 Hz), 3.82-3.78 (2H, m), 3.36 (4H, q, J = 7.3 Hz), 3.33 (2H, dd, J = 5.2, 5.2 Hz), 2.98 (3H, s),
.22 (6H, tm, J = 7.3 Hz). ¹³C NMR (151 MHz, d
6
-DMSO) 61.7 (t, J = 2.5 Hz), 56.8 (t, J = 2.5 Hz),
5.2, 47.7 (t, J = 3.7 Hz), 8.1 ppm.
2
-(Diethylamino)ethyl acetate (Acetylmonoethylcholine iodide, AMECh). A solution of acetyl
chloride (7.9 g, 100 mmol) in ether (50 ml) was added dropwise over 50 min to a stirred
solution of 2-(diethylamino)ethanol (11.7 g, 100 mmol), triethylamine (15.3 ml, 110 mmol)
and ether (50 ml) at 0-5 °C. A dense white precipitate formed. Additional ether (50 ml) was
added to facilitate stirring and the mixture heated under reflux for 2 h. The mixture was
allowed to cool and was stirred at room temperature overnight. The precipitate was filtered
off and thoroughly rinsed with ether. The solvent was removed from the filtrate to leave a
yellow liquid (16.1 g). This was distilled using a one-piece distillation rig to leave a brown oily
residue and provide a colourless distillate (9.02 g; bp 58-62 °C/10 mmHg). Analysis by 1H
NMR spectroscopy showed this to contain about 85% of the desired compound. The crude
product was therefore dissolved in ether (75 ml) and washed with saturated aqueous
sodium bicarbonate (2 × 75 ml). The organic phase was dried over anhydrous magnesium
sulfate. The drying agent was filtered off and the solvent removed from the filtrate to leave
1
a colourless liquid (4.8 g, 30%). Bp 60-61 °C/10 mmHg. Analysis by H NMR spectroscopy
indicated that this material was around 90% pure; this was used for the next step without
8

further purification. ¹H NMR (600 MHz, CDCl
3
) 4.13 (2H, t, J = 6.2 Hz), 2.70 (2H, t, J = 6.1 Hz),
.58 (4H, q, J = 7.2 Hz), 2.04 (3H, s), 1.03 (6H, t, J = 7.2 Hz). ¹³C NMR (151 MHz, CDCl ) 171.1,
2.5, 50.9, 47.6, 21.1, 11.6 ppm.
2
6
3
Acetyldiethylcholine iodide (ADECh). Iodomethane (2.18 ml, 35 mmol) was added in one
portion to a stirred solution of 2-(diethylamino)ethyl acetate (3.98 g, 25 mmol) in THF (25
ml). The mixture was left to stir at ambient temperature overnight. The solution went milky
and an orange oil separated. The solvent was removed by rotary evaporation to leave a
liquid. This was dried to constant weight (7.9 g) under high vacuum. Recrystallization from
isopropanol was attempted but no crystals formed: signs of solid formation were evident if
the mixture was chilled, but no recrystallization occurred upon warming slowly to room
temperature. The solvent was removed and the liquid dried under vacuum to a constant
weight (7.6 g, 100%). ¹H NMR (400 MHz, CDCl
.3, 2.5 Hz), 3.67 (4H, q, J = 7.3 Hz), 3.34 (3H, s), 2.12 (3H, s), 1.44 (6H, t, J = 7.2 Hz). ¹³C NMR
101 MHz, CDCl ) 170.1, 59.5, 57.9, 57.6, 48.8, 21.2, 8.6 ppm.
3
) 4.57 (2H, dd, J = 4.9, 4.9 Hz), 3.94 (2H, tt, J =
2
(
3
2
.3 Cell lines
Three different cell lines were used: CHO-CHRM1 (ECACC 10110203), T02J-7/10 (CHO-M3
CHRM3)) (ECACC 10031603) and CN21. The first two are Chinese Hamster Ovary (CHO) cells
(
stably transfected with the AChR M1 and M3 receptor cDNA sequences respectively.
Hereinafter, these cell lines are abbreviated to CHOm1 and CHOm3. They were obtained
from the European Collection of Authenticated Cell Cultures (ECACC) maintained by Public
Health England. CN21 cells were obtained from Dr. David Beeson (Institute of Molecular
Medicine, John Radcliffe Hospital, Oxford, UK) (15). These were derived from the TE671
rhabdomyosarcoma cell line by a stable transfection of the ε-subunit to express both the
foetal and adult human muscle-type nicotinic receptor.
CHOm1 and CHOm3 were grown in Ham’s F12 media supplemented with 10% FBS (foetal
bovine serum), 50 U/mL penicillin, 50 μg/mL streptomycin and either 2 mM L-glutamine or
GlutaMAX™ (Life Technologies, California, USA). CN21 cells were grown in DMEM
9

(Dulbecco’s Modified Eagle Medium) supplemented with 10% FBS, 50 U/mL penicillin, 50
μg/mL streptomycin, 2 mM L-glutamine, 0.5 mg/mL geneticin.
2
All cells were grown in 150 cm flasks at 37 °C, 5% CO
2
with the appropriate media until
approximately 70-80% confluent. Cells were then washed with PBS (phosphate-buffered
saline) and harvested with either Trypsin-EDTA or TrypLE Select (Fischer Scientific Ltd.). The
resulting cell suspension was either re-plated into tissue culture flasks at split ratios of 1:6-
1
:10 for continued proliferation, or onto Corning clear-bottomed, black-walled, 96 well half-
well plates for use in the assay. Cells were plated out in 50 μL of media 24-48 h before the
experiment at densities of 20,000 or 10,000 cells per well for CN21 and 30,000 or 15,000
cells per well for CHOm1 and CHOm3.
2
.4 Flexstation assays
On the day of the assay, the growth medium was removed and 50 μL of calcium dye was
added to the cells. The calcium dye, made according to the manufacturer’s instructions, was
further diluted (50:50) in a 50:50 mixture of HEPES-buffered (20 mM HEPES (Fischer
Scientific Ltd.) at pH 7.4) Hanks Buffered Saline Solution (HBSS) and DMEM/Ham’s F12 as
appropriate before use. For experiments with CHOm1 or CHOm3 cells, probenecid (Life
Technologies) at a final concentration of 2.5 mM was added to the calcium dye. For
experiments with CN21 cells, the calcium dye contained either atropine (at a final
concentration of 40 μM) to block the muscarinic response, or tubocurarine (at a final
concentration of 50 μM to block the nicotinic response), ensuring only stimulation at the
nAChR or mAChR respectively was measured.
AMECh, DECh and MECh were made up as stock solutions in HEPES-buffered HBSS on the
day of the experiment, while stock solutions of 250 mM ADECh were prepared in advance,
stored at -20 °C and thawed on the day of the experiment. These were then serially diluted
in compound buffer across a 96 well plate in half-log10 steps to provide the required range
of concentrations.
1
0

Following 30 min incubation with the calcium dye, the cells were analysed using the
FlexStation 3 (Molecular Devices Ltd., Berkshire, UK), a multi-mode plate reader, which was
programmed to measure baseline fluorescence for 15 s before adding the agonist (ACh,
AMECh or ADECh) and measuring the fluorescence for another 45 s. For experiments
assessing the antagonist activity of the false transmitter precursors, MECh and DECh were
added to the cells after 25 min of dye incubation, and ACh diluted in HEPES-buffered HBSS
to give a final concentration of 30 μM (for the CN21 and CHOm1 cells) and 0.3 μM (for the
CHOm3 cells) was used as the agonist.
2
.5 Data analysis
Raw fluorescence values, corrected for background fluorescence, were exported from
SoftMax Pro software (Molecular Devices Ltd.) to GraphPad Prism version 6.00 for
Windows. The concentrations were converted to log10 values and non-linear regression
analysis was performed using the sigmoidal concentration dose response (variable slope
function). Interpolation of this curve generated the EC50 values for ACh, AMECh and ADECh.
A statistical sum-of-squares F test was performed on the output of the linear regression to
compare the IC50 values and the maximal response.
For analysis of MECh and DECh as antagonists, the non-linear regression analysis provided
best-fit values for the top and bottom of the curve, which were used to normalise the data
for presentation. Where a concentration-response curve could not be fitted, values were
normalised to the mean of the maximal response.
Once values were normalised, they were analysed statistically using non-linear regression
analysis, specifically a dose-response curve (log (inhibitor) versus normalised response) on
GraphPad Prism version 6.02 for Windows, which gave the IC50. IC50 values were generated
when the inhibition was greater than 50% at the highest concentration tested.
If IC50 values could not be determined from the curve, the degree of antagonism was
estimated by taking the mean of the normalised fluorescence intensity values at the highest
concentration tested. This will also be provided in tables where appropriate.
1
1

2
.6 Hemi-diaphragm preparations
Male Dunkin-Hartley guinea pigs (250-300 g, Harlan UK) were killed by concussion, a UK
Home Office Schedule 1 method. Guinea pig hemi-diaphragm tissues isolated by dissection
were set up according to a method (19) based on that of Bulbring (20). As both left and right
hemi-diaphragms were prepared, experiments were organised such that each experimental
group had a mixture of left and right preparations. Tissues were suspended in modified
Tyrode’s solution (137 mM NaCl (Sigma-Aldrich/BDH), 5 mM KCl, 12 mM NaHCO
NaH PO ·2H O (Fischer Scientific/Sigma-Aldrich), 25 mM glucose, 2 mM CaCl and 1 mM
MgCl ), continually gassed with 95% O /5% CO and maintained at 37 °C.
3
, 1 mM
2
4
2
2
2
2
2
Muscle contractions were recorded from a baseline tension of 4 g using a 100 g isometric
force transducer (TSD 125D, Buxco Research Systems, Wilmington NC, USA) using Buxco
recording software (Biosystems for Windows version 2.7.5). Diaphragms were stimulated via
the phrenic nerve using a Grass S88 stimulator with 0.2 ms pulses delivered at supramaximal
voltage (1.5 times the maximum response, usually between 2 V and 15 V).
Figure 3 shows the experimental protocol used to test the efficacy of the false transmitters.
Tetanic stimuli were delivered at a frequency of 50 Hz for 3 s every 15 min, with twitch
stimuli delivered at a frequency of 1 Hz. After establishing stable baseline contractions for
3
0 min, contractions were decreased to approximately 5% of control values abolished by
adding an irreversible cholinesterase inhibitor, the nerve agent soman at a concentration of
50 nM. After 30 min (a period of 2 tetanic contractions), the soman was washed out and
1
diaphragm preparations were incubated with MECh and DECh. For control diaphragms, an
equivalent volume of modified Tyrode’s solution was added. Finally, two further tetanic
contractions were assessed following a washout.
2
.7 Data analysis
Force analysis was performed automatically by the Buxco recording system for all tetanic
contractions, except those where soman was present. These were analysed manually, as the
1
2

software did not analyse them accurately. To analyse diaphragm force, the area under the
curve (AUC) of the 50 Hz tetanic contractions was measured, normalised to the control
value (mean of 2 initial control responses) using Microsoft Excel 2010 and analysed using
GraphPad Prism version 6.0 for Windows. A 2-way ANOVA, with a post-hoc Tukey’s multiple
comparisons test, was performed on the maximal functional recovery achieved with MECh
or DECh using GraphPad Prism version 6.0 for Windows.
Figure 3. Protocol for analysis of false transmitters in hemi-diaphragms used to evaluate the
actions of MECh and DECh on guinea pig phrenic nerve/hemi-diaphragm preparations. Each
letter (A to N) represents a tetanic stimulus (delivered at a frequency of 50 Hz), with twitch
stimuli delivered at a frequency of 1 Hz in the 15 min between tetanic stimuli. The labels
under the schematic indicate the experimental conditions, with the washouts (W/O) shown
by vertical arrows.
3
Results
3
.1 False transmitter synthesis
To avoid the use of gaseous or hazardous reagents, we devised alternative routes for the
synthesis of AMECh, MECh, DECh and ADECh. Rather than use the expensive unsymmetrical
amines - dimethylethylamine (Me
2
NEt) or diethylmethylamine (MeNEt ) - as precursors, we
2
started from the readily-available and comparatively non-toxic 2-(dialkylamino)ethanol
compounds.
1
3

Quaternization of 2-(dimethylamino)ethanol and 2-(diethylamino)ethanol in THF with ethyl
iodide and methyl iodide, respectively, gave MECh and DECh iodide salts in acceptable yields
(Figure 4A). These quaternary ammonium salts had the physical properties expected of ionic
salts: they were crystalline solids of defined melting points (272 °C and 276 °C with
decomposition, respectively), dissolved in water or other polar solvents, and were insoluble
in ethers (diethyl ether or THF). The decomposition observed at the melting point is
characteristic of such salts; quaternary ammonium halides generally decompose on heating
to give the tertiary amine and alkyl halide (21). Treatment of 2-(diethylamino)ethanol with
acetyl chloride in diethyl ether in the presence of triethylamine gave 2-(diethylamino)ethyl
acetate (Figure 4B). Quaternization of this colourless liquid with methyl iodide in THF gave
ADECh as a liquid that failed to crystallise; it was used in the biological experiments as a
liquid. Note that all quaternary ammonium salts prepared in this study, whether solid or
liquid, were deliquescent, absorbing quickly moisture from the air. They were isolated in an
anhydrous state upon prolonged drying under high vacuum. The dry material was best
weighed as rapidly as possible for biological experiments to reduce the potential for
stoichiometry errors. This was the approach taken here. Such handling challenges are posed
by many quaternary ammonium salts, not just those reported in this publication, as they are
notoriously hygroscopic as a class: some form stable hydrates and are difficult to prepare in
a completely anhydrous state (21).
1
4

Figure 4. Synthetic routes to make the precursors and false transmitters. (A) Quaternization
chemistry used to produce MECh and DECh. (B) Two step route to ADECh. All routes use
straightforward economical procedures suitable for research laboratories, avoiding toxic
reagents used by industry to effect similar transformations.
1
5

3
.2 Agonist action of the false transmitters
First, a calcium/fluorescence assay was used to test whether AMECh and ADECh can activate
the ACh receptors expressed in CHOm1, CHOm3 and CN21 cells (Table 1 and Figure 5). If
AMECh and ADECh were partial agonists, they would be expected to produce a smaller
maximal response in comparison to ACh. Additionally, the EC50 (concentration which
generated 50% of the maximal response) may be shifted to the right, indicative of a lower
potency.
These experiments showed that both AMECh and ADECh acted as partial agonists at both
the M
1
subtype of mAChR expressed in CHOm1 cells and the M subtype expressed in
3
CHOm3 cells. An extra sum-of-squares F test performed on the curve fits indicated that both
the maximal fluorescent responses and EC50 values for AMECH and ADECh were significantly
different from ACh and each other (Table 1 and Figure 5A-B).
AMECh also acted as a partial agonist at the mAChR expressed in CN21 cells, as although the
EC50 value was slightly higher (Table 1), the maximal fluorescent response generated by
AMECh was lower than that generated by ACh (Figure 5C). In contrast to its effects in the
CHO cells, ADECh at concentrations up to 6 mM did not have any agonist activity at the
mAChR expressed in CN21 cells.
AMECh acted as a partial agonist at the human muscle-type nAChR in CN21 cells, as AMECh
generated much reduced maximal raw fluorescent values when compared to ACh and has a
lower EC50 value than ACh (Figure 5D). Contrastingly, ADECh did not act as an agonist at the
human muscle-type nAChR, displaying only slight activity at 6 mM, the highest
concentration tested.
In summary, AMECh acted as a partial agonist at both human muscle-type nAChR and
mAChR (M
1
and M subtypes), whilst ADECh acted as an agonist at certain mAChR subtypes,
3
but not human muscle-type nAChR. ADECh has reduced binding affinity in comparison to
both ACh and AMECh. As AMECh generated higher maximal responses, had lower EC50
values at mAChR, and also acted at human muscle-type nAChR, it should be considered a
more efficient partial agonist than ADECh (Table 1 and Figure 5).
1
6

A
B
CHOm1 cells
CHOm3 cells
4
00
50
00
50
00
50
00
400
350
300
250
200
150
100
50
AMECh
AMECh
3
3
2
2
1
1
ACh
ACh
ADECh
ADECh
5
0
0
0
-9
-8
-7
-6
-5
-4
-3
-2
-11 -10
-9
-8
-7
-6
-5
-4
Concentration [M]
Concentration [M]
C
D
mAChR in CN21 cells
nAChR in CN21 cells
6
4
2
00
00
00
0
700
AMECh
ACh
ADECh
AMECh
6
5
4
3
2
1
00
00
00
00
00
00
0
ADECh
ACh
-8
-7
-6
-5
-4
-3
-2
-8
-7
-6
-5
-4
-3
-2
Concentration [M]
Concentration [M]
Figure 5. Agonist actions of false transmitters at ACh receptors. The graphs show the agonist
actions of ACh, AMECh and ADECh at mAChR, in (A) CHOm1, (B) CHOm3, and (C) both
muscle-type nAChR, and (D) mAChR in CN21 cells. Data show the raw fluorescence units
recorded on the Flexstation following stimulation with the agonist, with the error bars
showing the standard deviation. In CHOm1 cells, n = 6 for all three agonists; in CHOm3 cells,
n = 3 for AMECh and ACh, n = 5 for ADECh; in CN21 cells, n = 3 for AMECh and ACh and n = 4
for ADECh. Where appropriate, a dose response curve was fitted using GraphPad Prism
version 6.0 for Windows, where the bottom was constrained to 0 as the baseline values had
already been subtracted.
1
7

Cell line EC50 with (95% Confidence Interval) (nM)
Relative efficacy
and
ACh
AMECh
ADECh
AMECh
ADECh
receptor
1
591
3823
16060
86.5%*
67.9%*
66.2%*
CHOm1
AChR
M1
(1074 – 2358)
(2436 – 5999)
(8173 – 31580)
8
.6
25.8
99.4
93.0%*
CHOm3
AChR
M3
(7.16 – 10.3)
(22.4 – 29.8)
(74.9 - 131.9)
3
845
2421 – 6107)
136
1345 – 3392)
14060
NA
NA
56.6%*
81.9%*
NA
NA
CN21
(
(7746 – 25520)
nAChR
2
1096
CN21
(
(744 – 1615)
mAChR
Table 1
Agonist EC50 values and relative efficacy for the false transmitters.
This table shows the calculated EC50 values for ACh and AMECh in CHOm1 and CHOm3 cells
and at muscle-type nAChR in CN21 cells. EC50 values were calculated from a log-dose
response curve generated using a Sigmoidal dose-response (variable slope) function in
GraphPad Prism version 6.02 for Windows, with the bottom constrained to 0 as the basal
response had already been subtracted. For mAChR and nAChR in CN21 cells, there was no
meaningful agonist response with ADECh (indicated by NA and shading in the table). The
9
5% confidence interval is shown. The relative efficacy was calculated from the maximal
value generated in the non-linear regression analysis, with ACh set at 100%. * Denotes
where an extra sum-of-squares F test, performed using GraphPad Prism, indicated that the
top value of the fit was different for each data set.
1
8

3
.3 Antagonist actions of the false transmitter precursors
Over the concentration range tested, neither MECh nor DECh caused appreciable inhibition
of the mAChR-mediated responses in any of the three cell lines (CHOm1, CHOm3 and CN21)
(Table 2 and Figure 6). In contrast, both MECh and DECh acted as antagonists at human
muscle-type nAChR, albeit at low potency, reducing signalling to 42% (± 9.4%) and 41% (±
5
2
.6%) respectively, at the highest concentration tested, with estimated log10 IC50 values of -
.6 and -2.8 M, respectively. As shown in Table 2, these IC50 values were markedly higher
than that achieved with the known muscle-type nAChR antagonist tubocurarine, which
caused a near 100% inhibition of the fluorescent response induced by ACh, with a log IC50 of
-
6.1 M.
In summary, the false transmitter precursors MECh and DECh did not act as antagonists at
the mAChR expressed in CHOm1, CHOm3 and CN21 cells. MECh and DECh acted as weak
antagonists at the human muscle-type nAChR and were approximately 1000 fold less potent
than tubocurarine.
1
9

Antagonism in CHOm1 cells
Antagonism in CHOm3 cells
1
1
25
00
125
100
75
50
25
0
7
5
2
5
0
5
0
DECh
DECh
MECh
MECh
Atropine
Atropine
-10 -9 -8 -7 -6 -5 -4 -3
-10 -9 -8 -7 -6 -5 -4 -3
Concentration [M]
Concentration [M]
Antagonism at nAChR in CN21 cells
Antagonism at mAChR in CN21 cells
1
1
1
50
1
1
25
25
00
00
7
5
2
5
0
5
0
7
5
2
5
0
5
0
DECh
MECh
Atropine
DECh
MECh
Tubocurarine
-8
-7
-6
-5
-4
-3
-11 -10 -9 -8 -7 -6 -5 -4 -3 -2
Figure 6. Antagonism actions of the false transmitter precursors. These graphs show the
antagonist actions of DECh and MECh against ACh stimulation at both human muscle-type
nAChR (CN21 cells) and mAChR (CHOm1, CHOm3 and CN21 cells). Values have been
normalised to baseline fluorescence responses and show mean ± SEM. For comparative
purposes, the dose-response curves for known antagonists at mAChR (atropine) and nAChR
(tubocurarine) are shown. In CHOm1 cells, n = 5 for all three treatments groups; in CHOm3
cells, n = 5 for MECh and atropine, n = 6 for DECh; in CN21 cells at mAChR, n = 4 for MECh
and atropine, n = 5 for DECh; in CN21 cells at nAChR, n = 4 for DECh and tubocurarine, and n
=
5 for MECh.
2
0

IC with (95% Confidence Interval) (μM)
Cell Line
50
MECh
DECh
Atropine
.038
0.029 – 0.051)
.0029
0.0024 – 0.0034)
Tubocurarine
0
CHOm1
(
0
CHOm3
(
2
341
2030
4.29
CN21 - nAChR
(1645 – 3332) (1549 – 2661)
(3.28 – 5.6)
0
.00094
CN21 - mAChR
(0.00082 – 0.0011)
Table 2. Antagonist actions of the false transmitter precursors. Antagonist IC50 values were
determined for MECh and DECh at human muscle-type nAChR, shown as the lC50 value, with
the 95% confidence interval. At mAChR, MECh and DECh were insufficiently active as
antagonists in the concentration range tested to allow the determination of IC50 values.
Instead, to indicate their antagonist actions, the normalised relative fluorescence intensity
(RFI) values at the highest concentration tested ± SD are shown. SD is the standard
deviation. SEM is the standard error of measurement.
2
1

3
.4 Effect of the false transmitters on neuromuscular function
A functional neuromuscular preparation, the guinea-pig phrenic nerve/hemi-diaphragm
preparation, was used to assess whether the false transmitters would improve
neuromuscular function after nerve agent exposure. The build-up of ACh which arises as a
consequence of nerve agent exposure has a pronounced effect on skeletal muscle function,
with muscle fibres no longer able to sustain a tetanic contraction (22). Addition of soman
caused the expected rapid and sustained fade of the tetanic contraction that was
maintained after washout of the nerve agent and is reflected in the marked decrease in the
tetanic AUC measured (Figure 7). Addition of 1 mM MECh in the absence of soman had no
effect on neuromuscular function, with contraction force similar to that seen with addition
of saline (data not shown). Addition of 0.3 mM MECh or DECh to the tissue bath after the
washout of soman led to a continuous improvement in neuromuscular function, with the
maximal improvement in neuromuscular force seen at the end of the 2 h incubation period
(Figure 7A-B). At a concentration of 1 mM, the improvement shown by both MECh and
DECh peaked earlier then plateaued or declined slightly respectively.
When MECh or DECh were applied at the highest concentration tested (3 mM), the tetanic
response initially improved more rapidly than at lower concentrations. This was then
followed by a decline in tetanic contraction to levels lower than those exhibited by the
soman control (Figure 7A-B). Statistical analysis (2-way ANOVA with a post-hoc Tukey’s
multiple comparisons test) of the maximal recovery achieved at each concentration studied
showed that the presence of MECh and DECh significantly improved diaphragm function
over and above saline alone.
Interestingly, for all treatment groups, there was a rapid and marked increase in tetanic
contraction following washout of both MECh and DECh, in comparison to values during the
incubation period (Figure 7A-C). For all groups except the DECh 3.0 mM group, these values
were significantly increased in comparison to saline.
The increase in the AUC after treatment with MECh or DECh appeared to be associated with
the ability of the diaphragm to maintain the tetanic contraction. Whilst the peak contraction
2
2

did not approach that achieved prior to soman addition, the contraction in treated
diaphragms did not show the marked fade normally associated with anticholinesterase
poisoning in this preparation (Figure 7C).
In summary, the experiments with these functional neuromuscular preparations showed
that administration of MECh and DECh improved neuromuscular function in hemi-
diaphragm preparations exposed to soman, and this improvement was maintained, or even
enhanced, upon washout of MECh and DECh.
2
3

A
Neuromuscular recovery with MECh
B
Neuromuscular recovery with DECh
1
0
0
0
0
0
.0
0.3 mM
1.0
0.3 mM
Soman Precursor
C
1
.0 mM
.0 mM
Soman Precursor
1.0 mM
3.0 mM
Vehicle
3
.8
.6
.4
.2
.0
0.8
0.6
0.4
0.2
0.0
Vehicle
0
15 30 45 60 75 90 105 120 135 150 165 180 195
Experiment Time (min)
0
15 30 45 60 75 90 105 120 135 150 165 180 195
Experiment Time (min)
Figure 7. Normalised contraction following false transmitter administration. A and B show normalised diaphragm contraction (gs) following
exposure to 150 nM soman and treatment with either MECh or DECh. After 2 baseline tetanic contractions, diaphragms were incubated with
soman for 30 min, before washing out (W/O) the soman and incubation of 120 min with either vehicle, DECh or MECh at 3 concentrations: 0.3,
1
.0 and 3.0 mM. Contractions were normalised to the average of the 2 baseline tetanic contractions. Data show mean ± SEM, with n = 4 for all
dose groups. C shows example muscle contractions from an individual hemi-diaphragm preparation exposed to soman and treated with 1 mM
MECh, followed by a trace 30 minutes after washout of MECh or DECh.
2
4

4
Discussion
This paper has shown that alternative synthesis routes avoiding the use of toxic chemicals
can be used successfully to generate AMECh, ADECh, MECh and DECh on a laboratory scale.
AMECh was shown to be a partial agonist at human mAChR and muscle-type nAChR, whilst
ADECh was shown to be a partial agonist at mAChR only, with no appreciable activity at
nAChR. The precursors MECh and DECh acted as antagonists at muscle-type nAChR, but not
at mAChR. Finally, it was demonstrated that administration of MECh and DECh to nerve
agent exposed hemi-diaphragm preparations significantly improved neuromuscular function
and this improvement persisted upon washout of the compound.
4
.1 Chemistry
Synthetic procedures that have been used to produce choline include the action of aqueous
trimethylamine on ethylene oxide to yield choline hydroxide, and the reaction between
trimethylamine and 2-chloroethanol to yield choline chloride (mp 302-305 °C dec.) (23, 24).
Similarly, ACh bromide or ACh chloride has been prepared from trimethylamine and 2-
bromoethyl or 2-chloroethyl acetate (24). These procedures, suitable for industrial
manufacture, are possible due to the availability and cheapness of trimethylamine (a gas, bp
3
°C). However, they use as the other reagent, chemicals of an unpleasant nature: ethylene
oxide (a gas, bp 12 °C; carcinogenic), 2-chloroethanol (a liquid, bp 128 °C; toxic) (24), or 2-
haloethyl acetates (volatile liquids, whose vapours are irritating and cause lung oedema)
(25, 26).
To avoid the use of gaseous or hazardous reagents, we devised alternative routes for the
synthesis of AMECh, MECh, DECh and ADECh. Rather than use the expensive unsymmetrical
amines - dimethylethylamine (Me
2
NEt) or diethylmethylamine (MeNEt ) - as precursors, we
2
started from the readily-available and comparatively non-toxic 2-(dialkylamino)ethanol
compounds. It was envisaged that their quaternization reactions – Menshutkin reactions –
with an appropriate alkyl halide would furnish the required ethyl-bearing choline derivatives
as halide salts. The reaction is a replacement of the halogen atom by the nucleophilic
2
5

tertiary amine; its rate depends on the nature of the alkyl halide. Methyl halides are more
reactive than other alkyl halides, and among the halides having the same alkyl group, the
iodide is the most reactive (27, 28). The solvent also influences the rate of reaction; since
the transition states and products have considerable separation of charge whereas the
reactants do not, solution in solvents of high dielectric constant, such as acetonitrile or
tetrahydrofuran (THF), facilitates quaternization by lowering the activation energy (27, 28).
The acetylated false transmitters from an electronic perspective are expected to be similarly
susceptible towards enzymatic or uncatalysed hydrolysis; the acetyl group being activated
towards nucleophilic attack similarly by the distant electron-withdrawing trialkylammonium
entity. The degree of positive charge on the nitrogen atom is expected to decline in the
+
+
+
order –N Me
methyl group) and the size of this grouping increases the more ethyl groups that are present
the ethyl group is larger than a methyl group). These factors might influence the interaction
3
> –N Me
2
Et > –N MeEt
2
(an ethyl group is more electron-donating than a
(
of the acetylated transmitters with AChE and other binding sites involved in
neurotransmission, and this might account for the differences in potency at nAChR observed
for ACh, AMECh and ADECh (6, 9, 14, 29).
4
.2 Agonism at acetylcholine receptors
To date, the activity of AMECh and ADECh has been determined at muscarinic receptors in
the rat (6, 11, 14) and muscle-type nicotinic receptors in the rat (6), mouse (9, 29), frog (9)
and Torpedo electric organ (14). For the false transmitters to be effective in reducing
overstimulation at the NMJ after nerve agent poisoning, they would need to act at human
ACh receptors. Consistent with the results previously reported in the rat (11), here it has
been shown that AMECh activated human mAChR M1 and M3 receptors, and achieved a
comparable maximal response to ACh, albeit with a higher EC50. In contrast, ADECh acted as
a lower potency partial agonist, producing lower maximal responses than both ACh and
AMECh.
Although AMECh acted as an agonist at the mAChR known to be expressed in CN21 cells,
ADECh did not. The CN21 cells are derived from the TE671 cell line (16), which is known to
2
6

express muscarinic receptors (30), specifically the M3 subtype (17). In this study a difference
in agonist activity was identified between the CHOm3 cells and the CN21 cells – the EC50 of
ACh was markedly different between these cell types and additionally ADECh did not
activate the mAChR in CN21 cells. This could be related to differences between the cell
types expressing the receptor, Chinese Hamster Ovary versus human rhabdomyosarcoma,
2
+
and differences in coupling to the Ca response.
Consistent with previous reports on AMECh and ADECh activity at rat and Torpedo nAChR (9,
1
1), AMECh acted as a partial agonist at human muscle-type nAChR, generating lower
maximal responses compared to ACh. Although ACh and AMECh generate unitary current
events of the same amplitude when bound to the receptor, in comparison to ACh, AMECh
has a reduced potency for the nAChR and a reduced channel opening time, approximately
4
0% shorter (6, 9, 10).
In addition, here it was identified that ADECh does not activate human muscle-type nAChR
in CN21 cells, although previous studies showed that ADECh activated nAChR from the
Torpedo electric organ and rat muscle, albeit with much reduced potency compared to ACh
and AMECh (9, 11). Thus, given that ADECh is a weak partial agonist, the difference in
activity is likely due to species differences in the structure of the nAChR.
4
.3 Antagonism at nAChR
One of the hydrolysis products of ACh, choline, can antagonise signalling through the
nAChR, since it blocks the nicotinic ion channel and may act as a partial agonist at human
muscle-type nAChRs (31). Given this dual action of choline, the potential for the false
transmitters to act as antagonists of nAChR was investigated. Here, it was identified that, in
high concentrations, MECh and DECh are antagonists of the human muscle-type nAChR.
Although their potency was much reduced in comparison to the known nAChR antagonist
tubocurarine, their potency is comparable to the parent compound choline, which has an
equilibrium constant for channel block at the muscle-type nAChR of 6.9 mM (31).
Densitisation of nAChR depends on the subunit composition of the receptor and whilst
some nAChR such as α7 desensitise rapidly, others such as the muscle-type nAChR
2
7

investigated here desensitise more slowly (32). As reviewed in Giniatullin et al., 2005
different ligands have different effects on desensitisation of the nAChRs (32), so it is possible
that the false transmitters could alter these kinetics. However, activation of the nAChR by
the false transmitters results in a reduced channel opening time (6, 9, 10), and it is
considered more likely that this, plus the reduced potency of the false transmitters for the
receptor, lead to the partial agonism (and subsequent functional recovery in the
neuromuscular preparation).
4
.4 False transmitter precursors as treatments for nerve agent poisoning.
A previous study showed that administration of MECh and a related false transmitter
precursor, N-aminodeanol (NAD), protected mice from the lethal effects of AChE inhibitors,
when administered at the same time as the inhibitor (13). In this same publication, the
authors also showed in vitro that, at these efficacious doses, MECh and NAD both inhibited
AChE activity and acted as antagonists at both nAChR and mAChR, specifically M
1
and M
2
(13). Thus, the authors’ proposed mechanism of action for the protection seen with MECh
and NAD was through either interaction with AChE or competition with ACh for receptor
occupancy (13). The authors made no mention of the false transmitter effect. Here, an
alternative mechanism of action has been investigated, which could contribute to the
protection seen in the mice (13). This is the manipulation of neuromuscular transmission to
reduce the effects of ACh overstimulation at the NMJ, countering the build-up ACh caused
by AChE inhibitors.
It was predicted that introducing the false transmitter precursor would improve NMJ
function after nerve agent poisoning by (a) reducing the amount of ACh released, (b)
increasing the contribution of the lower efficacy false transmitter to the released pool of
ACh and (c) changing the kinetics of the human muscle-type nAChR channel opening, which
would at least partially correct the kinetics of synaptic potentials in the presence of AChE
inhibitors. Moreover, it is known that incorporation of the false transmitter into releasable
pools occurs more rapidly when AChE is inhibited, as choline is not being generated from
the breakdown of ACh (9).
2
8

The improvement in muscle contraction seen here after administration of MECh and DECh
to the soman exposed neuromuscular preparations is consistent with this hypothesis. The
recovery in muscle contraction develops over time and is concentration-dependent;
administering a larger dose of the precursor results in a quicker recovery time. The most
effective concentration of MECh and DECh restored approximately 40-50% of control
muscle function. This degree of recovery is comparable to that achieved with a muscle-type
nAChR antagonist, like MB327 (33, 34), and greater than that achieved with oximes
including HI-6 (22). Although high concentrations of MECh and DECh have been used and
show this benefit in vitro, potential toxicities in vivo may impact on their utility. Although
this does not impact on their ability to protect against anti-cholinesterases in vivo (13), an
assessment of potential toxicity would be a key aspect of any further exploration of this
approach.
The maintained and enhanced effect seen on washout is also consistent with this
mechanism of action, as the acetylated transmitter will continue to be released from the
pre-synaptic terminal even after the precursor has been washed out. It shows there must be
a persistent change and one not due to a pharmacological action that would reverse on
washout. In contrast, the inhibitory effect of the highest concentration does washout, hence
the rebound in function upon washout of the 3 mM dose. Additionally, the presence of
MECh or DECh improved and restored the tetanic shape of the muscle contraction, even
during the washout period. This is important and suggests the presence of the false
transmitter has corrected some of the fundamental aspects of neuromuscular transmission.
Additionally, it is interesting that MECh and DECh led to comparable improvements in
muscle contraction, given that ADECh did not activate human muscle-type nAChR here. This
positive result suggests that reducing the amount of ACh released is key. Alternatively, the
functional recovery could be a result of the competition between ACh and the false
transmitter, which would reduce ACh stimulation, regardless of whether the false
transmitter activated the receptor.
It should be noted that the in vitro assessment of receptor agonism and antagonism was
performed in cell lines expressing the human ACh receptors, whilst the functional
2
9