Homologation of mexiletine alkyl chain and stereoselective blockade of skeletal muscle sodium channels

Article abstract of DOI:10.1016/S0223-5234(00)00115-X

The optical isomers (-)-(S)- and (+)-(R)-3-(2,6-dimethylphenoxy)-2- methyl-1-propanamine (Me2), homologues of the antiarrhythmic and antimyotonic drug mexiletine (Mex), were synthesized and assayed as new potential antimyotonic agents. As observed with Mex, Me2 exhibits an enantioselective behaviour. Tests carried out on sodium currents of single muscle fibres of Rana esculenta demonstrated that (-)-(S)- and (+)-(R)-Me2 were less potent than Mex in producing tonic block, but showed a higher use-dependent block. (-)-(S)-Me2 and (-)-(R)-Mex were also used to study the excitability of muscle fibres of myotonic ADR mice, a phenotype of a recessive form of low G(Cl) myotonia. (-)-(S)-Me2 reduced spontaneous discharges and after discharges better than (-)-(R)-Mex in agreement with the use-dependent block of sodium currents. (C) 2000 Editions scientifiques et medicales Elsevier SAS.

Full text of DOI:10.1016/S0223-5234(00)00115-X

Eur. J. Med. Chem. 35 (2000) 147−156
© 2000 Éditions scientifiques et médicales Elsevier SAS. All rights reserved
S022352340000115X/FLA
147
Original article
Homologation of mexiletine alkyl chain and stereoselective blockade of skeletal
muscle sodium channels
Andrea Durantia, Carlo Franchinib, Giovanni Lentinib, Fulvio Loiodiceb, Vincenzo Tortorellab*,
Annamaria De Lucac, Sabata Piernoc, Diana Conte Camerinoc
aIstituto di Chimica Farmaceutica e Tossicologica, Facoltà di Farmacia, Piazza del Rinascimento 6, Università degli Studi di Urbino,
61029 Urbino, Italy
bDipartimento Farmacochimico, Facoltà di Farmacia, via Edoardo Orabona 4, Università degli Studi di Bari, 70125 Bari, Italy
cDipartimento Farmacobiologico, Facoltà di Farmacia, via Edoardo Orabona 4, Università degli Studi di Bari, 70125 Bari, Italy
Received 26 April 1999; revised 21 December 1999; accepted 22 December 1999
Abstract – The optical isomers (-)-(S)- and (+)-(R)-3-(2,6-dimethylphenoxy)-2-methyl-1-propanamine (Me2), homologues of the antiarrhyth-
mic and antimyotonic drug mexiletine (Mex), were synthesized and assayed as new potential antimyotonic agents. As observed with Mex, Me2
exhibits an enantioselective behaviour. Tests carried out on sodium currents of single muscle fibres of Rana esculenta demonstrated that
(-)-(S)- and (+)-(R)-Me2 were less potent than Mex in producing tonic block, but showed a higher use-dependent block. (-)-(S)-Me2 and
(-)-(R)-Mex were also used to study the excitability of muscle fibres of myotonic ADR mice, a phenotype of a recessive form of low GCl
myotonia. (-)-(S)-Me2 reduced spontaneous discharges and after discharges better than (-)-(R)-Mex in agreement with the use-dependent block
of sodium currents. © 2000 Éditions scientifiques et médicales Elsevier SAS
antiarrhythmic drug / mexiletine / chiral analogues / myotonia / sodium channel tonic block / sodium channel use-dependent block
1. Introduction
same, mexiletine has a narrow therapeutic ratio, many
adverse effects (mainly cardiovascular and nervous) be-
ing dose-related and sometimes severe enough to neces-
sitate discontinuation of therapy [9].
A few years ago, by virtue of its pharmacokinetic
properties [10] and, above all, of its mechanism of action,
Mex was clinically proposed in the symptomatic treat-
ment of myotonic syndromes, which comprise a series of
dominant and recessive genetic channelopathies of skel-
etal muscle characterized by membrane hyperexcitability
and delayed muscle relaxation after voluntary contrac-
Mexiletine [1-(2,6-dimethylphenoxy)-2-propanamine]
(Mex) (figure 1) is an established antiarrhythmic drug
belonging to the IB class [1].
The pharmacodynamic theory, common for local
anaesthetic-like drugs, provides for the block of voltage-
gated sodium channels [2] through interaction with seg-
ment 6 of domain IV of the sodium channel α-subunit [3].
According to the ‘modulated receptor’ hypothesis [4, 5],
the affinity for sodium channel receptors increases in
those channel conformations (open and/or inactivated)
favoured by high frequency of stimulation [2, 6]. This
behaviour leads to the so-called ‘use-dependent’ or pha-
sic block and is due to an accumulation of blocking on
depending the slow recovery from inactivation of the
drug-bound channels during membrane repolarization [2,
3, 7, 8]. The use-dependency of mexiletine action sup-
ports its selectivity towards the arrhythmic heart; all the
*Correspondence and reprints: vtor@farmchim.uniba.it
Figure 1. Structures of mexiletine (Mex) and tocainide (Toc).

148
tion [11]. Mexiletine has also been employed in some
hereditary non-dystrophic myotonias [12]: myotonia con-
genita (chloride channel disease) [11] and paramyotonia
congenita (sodium channel disease) [13]. When adminis-
tering Mexiletine as an antimyotonic drug, doses are as
large as those used in the antiarrhythmic treatment. Thus
the antimyotonic therapy, which has to be prolonged
through the whole patient’s life, might elicit adverse
reactions similar to those observed during the antiarrhyth-
mic therapy [11].
Figure 2. Structure of 3-(2,6-dimethylphenoxy)-2-methyl-1-
propanamine (Me2).
vivo [24] studies on cardiac muscle cells evidenced a
stereoselective behaviour of optical isomers of Mex,
being the (-)-(R)- enantiomer more potent than the
(+)-(S)- one and that a stereoselective site for sodium
channel blockers on adult skeletal muscle fibres has been
evidenced too [3, 15], we prepared the optical isomers
(-)-(S)- and (+)-(R)-3-(2,6-dimethylphenoxy)-2-methyl-
1-propanamine (Me2) (figure 2). The effects of Me2 on
sodium currents of single frog muscle fibres, and the
ability to suppress the pathological hyperexcitability of
skeletal muscle isolated from myotonic arrested develop-
ment of righting response (ADR) mice, a phenotype of a
recessive form of low GCl myotonia [25–27], were evalu-
ated. The stereoselectivity of action of Me2 enantiomers
in comparison to that of Mex enantiomers was evaluated.
Conceivably, mexiletine analogues provided with
higher use-dependency of action should present a wider
therapeutic ratio, being more selectively active on hyper-
excited tissues, i.e. arrhythmic heart and myotonic
muscle fibres. As part of a program aimed at preparing
use-dependent sodium channel blockers useful as anti-
myotonic drugs [14], we started from the observations in
favour of a common stereospecific domain [15] for Mex
and tocainide (Toc), an α-amino xylidide belonging to the
IB antiarrhythmic class (see structure in figure 1), in the
pore of Na+ channels [3]: some hints might be extrapo-
lated from the α-amino xylidide series structure–activity
relationships (SAR) [16] and applied these to the rela-
tively less explored xylyloxy alkyl amines, such as Mex.
Thus, â-homologation of Toc improved pharmacokinetic
and pharmacodynamic properties [17] and this was ration-
alized invoking an increased basicity which, in turn,
would both favour the presence of the drug in charged
form in the biophase and discriminate the passage
through the blood–brain barrier, so diminishing the side
effects on the central nervous system. On the other hand,
an increased lipophilicity, that may be supposed to be
caused by the homologation, does favour the access to the
cytoplasmic side of the membrane where the binding site
is accepted to be located [18–20]: the drugs are normally
applied outside the cell and the increased lipophilicity
would facilitate the diffusion through the membrane,
increasing the amount of drug that has access to the
binding site [21]. Thus we speculated that, given the
bioisosteric relationship holding between Toc and Mex,
and the similarities between heart and muscle sodium
channel isoforms which make both Toc and Mex valuable
antiarrhythmic as well as antimyotonic drugs, the ho-
mologation would improve, once again, the pharmacoki-
netic and pharmacodynamic profile of the latter.
2. Chemistry
Mex homologues (5a–b) were obtained through the
stereospecific synthesis described in figure 3, starting
from (-)-(R)- and (+)-(S)-3-bromo-2-methyl-1-propanol
(1a–b); 1a and 1b were converted to the corresponding
phthalimido alkanols (2a–b) by reaction with potassium
phthalimide. The halogenation was led, slightly modify-
ing reaction conditions used in the past to convert benzyl
alcohols into the corresponding bromides, in good to low
yields [28]: the modified procedure gave 3a–d in high
yields. Compounds 3a–d were coupled with 2,6-
dimethylphenol to give the xylyloxy derivatives
4a–b [29]. The target compounds 5a–b were obtained by
deprotection of the amino group with N2H4 [30] and,
finally, conversion into their hydrochloride salts [(-)-(S)-
and (+)-(R)-3-(2,6-dimethylphenoxy)-2-methyl-1-propa-
namine (Me2)]. The enantiomeric purity (e.p.) of 5a and
5b was evaluated by HPLC analysis of the corresponding
N-acetyl derivatives (6a and 6b) on chiral stationary
phase (CSP) and it was 99% and 97%, respectively;
[CSP: CHIRALCEL OD, (DAICEL Co.), mobile phase:
n-hexane/i-PrOH = 95/5, flow rate: 1.0 mL/min]. (-)-(R)-
and (+)-(S)-Mex were obtained by optical resolution of a
racemic mixture and the corresponding e.p. was evalu-
ated as previously described [31].
As expected, the calculated pKa and LogP values [22]
for Me2 (9.7 ± 0.1 and 2.8 ± 0.2, respectively) were
higher than the corresponding values for Mex (8.6 ± 0.1,
and 2.2 ± 0.2, exp. 2.15 [23], respectively). Therefore, in
this work we homologated Mex, moving the amino group
away from the stereogenic centre by insertion of a
methylenic group. Given that in vitro [15] and in

149
Figure 3. Conditions: (i) DMF, 70–80 °C, 24 h; (ii) HX, toluene –5 °C → room temperature, 5 h; (iii) 2,6-dimethylphenol, NaH, DMF,
room temperature, 3 h; (iv) N2H4 AcOH, EtOH, reflux, 1.5 h; (v) HCl, Et2O, room temperature.
3. Pharmacology
enantiomers of Me2 in comparison with that of (-)-(R)-
Mex, the eutomer of Mex. (-)-(S)-Me2 was more potent
than its enantiomer, both being less potent than (-)-(R)-
Mex in producing the tonic block of INa max. The
calculation of the half-maximal concentrations (figure 6)
confirmed the stereoselective behaviour; in fact the eud-
ismic ratios [IC50distomer/IC50 eutomer] were 2 and 1.7
for Mex and Me2, respectively. In figure 6 it is also
shown that Mex was about twice as potent as Me2 in
producing the tonic block. The use-dependent behaviour
of each compound was evaluated with trains of test pulses
from the h.p. to –20 mV at the frequency of 2 and 10 Hz.
With this protocol in the presence, but not in the absence,
of the test compounds, a further reduction in peak sodium
current over the tonic block was observed that progres-
3.1. Effects of Mex and its analogue on
sodium currents of single muscle fibres
The tonic block exerted by in vitro application of the
test compounds was calculated as percent reduction of the
inward sodium transient current (INa max) elicited by
infrequent depolarizing stimulation from the holding
potential of –100 to –20 mV to evaluate drug interaction
with the channels at the holding potential, i.e. in the
resting state. Both Mex and Me2 produced a block of
INa max in this condition (figure 4), in a concentration-
dependent manner. As already observed with Mex [14],
the effects of Me2 were stereoselective. Figure 5 shows
the concentration–response curves of (-)-(S)- and (+)-(R)-

150
Figure 4. Inward sodium transients recorded by the three
vaseline gap voltage clamp method from single frog muscle
fibres and effects of (-)-(R)-Mex (50 µM) and (-)-(S)-Me2 (50
µM). After recording the control trace elicited by a single test
pulse from –100 to –20 mV, the test compounds were applied
(a). After 10 min of incubation, a single test pulse was again
applied. The reduction of the current was the result of the tonic
block by the drug (b). The application of a 30 s train of pulses
at 10 Hz showed a cumulative reduction of sodium currents
over the tonic block due to the use-dependent behaviour of the
drug (c).
Figure 6. Each bar shows the mean ± SE of the concentrations
of Mex and Me2 enantiomers able to exert a 50% reduction of
sodium currents (IC50 values) during tonic and use-dependent
block. The IC50 values have been obtained from the curves
fitting the concentration–response experimental points [25]. At
each drug concentration the use-dependent block of sodium
currents has been calculated at the end of the 10 Hz stimulation,
as percent reduction with respect to the current in the absence
of drug. The statistical differences between fitted values have
been calculated using a Student’s t distribution using a number
of degrees of freedom equal to the total number of values
determining the curve, minus the number of means minus two
for the free parameters. *P < 0.01; **P < 0.05.
sively cumulated until a new equilibrium was reached
(figure 4). The value of the current at this equilibrium
normalized with respect to the current in the absence of
the drug was used to calculate the potency of the drug in
blocking the channels under excessive stimulation con-
ditions (e.g., high-frequency firing).
produced by 50 µM (-)-(S)-Me2 was comparable to that
produced by the same concentration of (-)-(R)-Mex
(figure 4). Consequently, the IC50 value calculated for
both enantiomers of Me2 was remarkably lower than that
measured in tonic block (figure 6). The IC50 for (-)-(S)-
Me2 was 6.2 times lower with respect to that of tonic
block vs. the 2.4-fold lowering of IC50 observed for
(-)-(R)-Mex. During the high-frequency stimulation at
10 Hz the stereoselectivity of the two compounds attenu-
ated. In fact the eudismic ratio of Mex decreased from 2.1
to 1.2, whereas that of Me2 decreased from 1.7 to about
1 (figure 6) [32].
In these conditions a remarkable use-dependent behav-
iour of Me2 was observed. In fact at 10 Hz the final block
3.2. Effects of (-)-(R)-Mex and (-)-(S)-Me2
on the excitability characteristics of musle fibres
of myotonic ADR mice
The myotonic ADR mouse shows a severe recessive
form of myotonia due to an abnormally low resting
chloride conductance caused by a genetic defect in the
muscle chloride channels [25]. The myotonic hyperexcit-
ability of sarcolemma can be clearly detected by in vitro
recordings on intercostal muscle fibres. Spontaneous
myotonic activity, such as high frequency discharges of
action potentials (> 10 Hz) on insertion of the recording
electrode, was detected in about 60–80% of fibres in each
Figure 5. Concentration–response curves for the tonic block
exerted by (-)-(R)-Mex and by both enantiomers of Me2. Each
point is the mean ± SEM of the percentage block of sodium
currents evaluated at each concentration from 3–5 muscle
fibres.

151
ADR muscle preparation (figure 7). As typical of myo-
tonic state, the muscle fibres of ADR mice were able to
generate a significantly higher number of action poten-
tials (11.7 ± 0.7; no. of fibres = 25) than healthy ones (5.1
± 0.5; no. = 20) when the depolarizing stimulus was just
slightly increased over the threshold values. In myotonic
muscle fibres the stimulus-evoked train of action poten-
tials was followed by self-sustained after discharges at
the end of the depolarizing current pulse in about
75–100% of the fibres sampled (figure 7).
Both Mex and Me2, tested in the eutomeric form,
reduced the hyperexcitability of myotonic muscle fibres
and in particular relieved, at low concentrations, the
typical myotonic manifestations (figure 7). (-)-(R)-Mex
started to have a membrane-stabilizing effect at 10 µM,
the concentration at which a significant reduction of the
spontaneous discharges was observed. At 50 µM these
discharges were completely suppressed, along with a
significant reduction in the occurrence of after discharges,
although the stimulus-evoked firing of the membrane was
not modified (figure 7). The latter was appreciably re-
duced by 20% at 100 µM (-)-(R)-Mex, and about 200 µM
were needed to reduce the firing capability by 50%.
(-)-(S)-Me2, in agreement with the strong use-
dependent block of sodium currents described above, was
more potent than (-)-(R)-Mex on the hyperexcitability of
myotonic muscle fibres. In fact (-)-(S)-Me2 was able to
decrease dose-dependently both spontaneous discharges
and after discharges in the range of 5–10 µM; so
(-)-(S)-Me2 was more potent than (-)-(R)-Mex on the
after discharges. At the above concentrations the firing
capability of the membrane was also slightly reduced
(figure 7). (-)-(S)-Me2 at 50 µM completely suppressed
spontaneous myotonic signs and significantly reduced the
maximum number of spikes that the membrane could
generate from 13 ± 0.8 (no. = 6) to 8.3 ± 0.8 (no. = 9) (P
< 0.005), the half maximal concentration of this param-
eter being about 60 µM [33].
Figure 7. Effects of (-)-(R)-Mex and (-)-(S)-Me2 on the hyper-
excitability of intercostal muscle fibres of myotonic ADR
(arrested development of righting response phenotype) mice. In
each panel the first two groups of bars from the left side show
the percent of fibres sampled exhibiting spontaneous dischar-
ges, after discharges in control conditions, and further applica-
tion of 10 and 50 µM of (-)-(R)-Mex (A) or 5 and 10 µM of
(-)-(S)-Me2 (B). The last group of bars shows the percent
reduction of the maximum number of fibres being elicited by a
depolarizing stimulus in the presence of the test compounds at
the above concentrations, with respect to the control taken as
100%. Statistical evaluation was based on the Chi-square test.
As far as the frequency of occurrence of spontaneous dischar-
ges and of after discharges in the presence of (-)-(R)-Mex and
(-)-(S)-Me2 is concerned there is a significant difference
between respective groups with ø2 > 6.
5. Discussion
The use-dependent behaviour of local anaesthetic-like
drugs, correlated with the affinity to and the dissociation
rate from the receptor of sodium channels, is affected by
variations of physical-chemical properties. A relatively
high pKa value (≥ 7.4), increasing the ratio between
charged and uncharged forms at physiological pH, can
improve access to the sodium channel active site by
permeating the hydrophilic pathway of the open channel
from the cytoplasm [19, 34], and by virtue of the voltage
gradient across the membrane [35, 36]. An increase in
pKa and lipophilicity values plays a role in tocainide
analogues [17]. In a series of lidocaine analogues, ob-
tained by insertion of methylene groups in the arylamide-
amine chain, the affinity for the cardiac sodium channel
was optimal for the next superior homologue of
lidocaine. The mean IC50 values of lidocaine and its
homologues suggested that an optimal length of the
chain, per se, is important in drug binding [37]. Finally,

152
the increase of both molecular weight and pKa, in another
series of lidocaine and procainamide derivatives [38], as
well as in a particularly high affinity series of benzocaine
analogues [39], slowed the time constant governing the
dissociation rate. This process, according to the ‘modu-
lated receptor’ hypothesis, should take place during the
membrane repolarization for the transition of the channel
from the high affinity inactivated-state to the low-affinity
resting state [2].
In agreement with the above-mentioned observations,
the homologation of Mex with a methylene group leads to
a use-dependent behaviour stronger than Mex. We hy-
pothesize that the reduction of the electron-withdrawing
effect of the xylyloxy group on the amino group, lower in
Me2 than in Mex, should produce a higher charged/
uncharged ratio and this could improve the access to
sodium channel receptors. On the other hand, an in-
creased lipophilicity, caused by the homologation, does
favour the access to the cytoplasmic side of the mem-
brane where the binding site is accepted to be loca-
ted [18–20]; consequently, Me2 acts at lower concentra-
tions than Mex on high-frequency discharges of action
potentials of myotonic muscle fibres.
general law about primary amine LA-like drugs: the
blocking property of an LA-like drug may result at least
from the combination of the contribution of pKa and
LogP values. Nevertheless, other aspects, such as an
optimal length of the chain, may not be discarded a
priori [37]. The results here presented do parallel previ-
ous findings and confirm the bioisosteric relationship
between Toc and Mex series; these observations suggest
a common mechanism of action for both series. Further-
more, the Me2 eutomer, in spite of its stereochemical
descriptor, does share the same configuration of the
eutomers of Mex and analogues in muscle fibre sodium
channel studies [41]. Thus a common binding site for
Me2 and Mex may be hypothesized. This hypothesis is
currently under investigation by means of eudismic
analysis on newly synthesized Me2 analogues.
In conclusion, the results reported in this study dem-
onstrate that Me2 is more potent than Mex for a use-
dependent block and the candidate Me2 must undergo
more complex in vivo studies which will evaluate its
potential usefulness as a specific antimyotonic agent
possibly overcoming Mex side effects.
The optical isomers of Me2 bind to the receptor of
sodium channels in a stereoselective manner as well as
Mex [(+)-(R)-/(-)-(S)-Me2 eudismic ratio about 1.6 vs. 2
for optical isomers of Mex]; (-)-(S)-Me2 was less potent
than (-)-(R)-Mex in producing tonic block but it showed
a more relevant use-dependent behaviour. We recorded an
attenuation of stereoselectivity when elevating the fre-
quency of stimulation; this result might imply that the
receptor on the muscle fibre sodium channel has different
stereospecific requirements depending on the channel
states, or that such a dissipation of stereoselectivity is due
to a poor recovery from inactivation state of the drug-
blocked channel during the brief interpulse duration at the
10 Hz frequency [33].
The question if there is a specific receptor for Mex and
its analogues, being shared or not with α-amino xylid-
ides, or only binding sites of a multiple nature, remains
unanswered and is far from obvious. The aromatic ring of
local anaesthetic-like drugs has been proposed to bind to
a tyrosine residue (at position 1 586 in the rat skeletal
muscle sodium channel) while the terminal amine group
of the drugs may associate with a neighbouring phenyl-
alanine residue through π electron interaction (position
1 579) [3]. Very recently, site-specific mutagenesis stud-
ies have suggested that an asparagine residue may play a
role in bupivacaine binding to cardiac sodium chan-
nels [40]. Further work is needed to define the local
anaesthetic-like drug receptor, if any. From a general
point of view, our and others’ past findings argue for a
5. Experimental protocols
5.1. Chemistry
Melting points were determined on a Gallenkamp
melting point apparatus in open glass capillary tubes and
1
are uncorrected. H-NMR spectra were recorded on an
FT Spectrometer Bruker Aspect 3000 (300 MHz) spec-
trometer using CDCl3 as solvent. MS spectra were
recorded with a Hewlett-Packard 5995c gas chro-
matograph/mass spectrometer at low resolution. Elemen-
tal analyses were performed on a Carlo Erba mod 1106
instrument, and the data for C, H, N were within ± 0.4%
of theoretical values. Optical rotations were measured on
a Perkin Elmer 241 MC spectropolarimeter, concentra-
tions are expressed in g/100 mL and the cell length is 1
dm. HPLC was performed on a Waters chromatograph
(Waters Assoc., Milford, MA) model 600 equipped with
a U6K model injector and a 481 model variable wave-
length detector. Chromatographic separations were per-
formed on silica gel columns by flash chromatography
(Kieselgel 60, 0.040–0.063 mm, Merck). TLC analyses
were performed on precoated silica gel on glass or
aluminium sheets (Kieselgel 60 F254, Merck). Solvents
were RP grade, unless otherwise indicated. All chemicals
were purchased from Aldrich in the highest quality
commercially available. (±)-Mex was kindly donated by
Boehringer Ingelheim KG.

153
5.1.1. (-)-(S)-2-(3-hydroxy-2-methylpropyl)-
1H-isoindole-1,3(2H)-dione (2a)
5.1.3. (+)-(S)-2-(3-iodo-2-methylpropyl)-
1H-isoindole-1,3(2H)-dione (3c)
Into a stirred, ice-cooled solution of 2a (6 mmol) in dry
toluene (30 mL) under N2 atmosphere, HI was bubbled
until saturation occurred. The mixture was stirred at
–5 °C for 1 h and at room temperature for 4 h. The
solvent was evaporated in vacuo, the residue was taken
up with EtOAc and washed with 6 N NaOH, 2 N HCl,
and H2O. The organic phase was dried (Na2SO4) and
concentrated in vacuo. Purification of the crude solid
residue by column chromatography (petroleum ether/
EtOAc 8:2) and recrystallization from MeOH gave 3c as
To a stirred solution of (-)-(R)-3-bromo-2-methyl-1-
propanol (18.9 mmol) (1a) in dry DMF (36 mL) under N2
atmosphere, potassium phthalimide (6 mmol) was added
and the mixture was heated to 70–80 °C. In 1 h, 6, 4 and
2 mmol of potassium phthalimide (18 mmol on the
whole) were added, maintaining the temperature for 24 h.
The mixture was filtered and the organic layer was
concentrated in vacuo. Purification of the crude oil
residue by column chromatography (petroleum ether/
EtOAc 7:3) and recrystallization from CH2Cl2/petroleum
ether gave 2a as a white solid. Yield: 82% (72 xx). M.p.
79–80 °C. [α]20D = –16.0 (c 2, CHCl3). 1H-NMR: δ 0.97
(d, 3H, J = 7.0 Hz, CH3–CH); 1.96–2.11 (m, 1H,
CH2–CH–CH2); 2.74 (bs, 1H, CH2OH); 3.34 (dd, 1H, J =
11.9 Hz and J = 6.0 Hz, CH–CHH–NR2); 3.50 (dd, 1H,
J = 11.9 Hz and J = 4.2 Hz, CH–CHH–NR2); 3.70 (d, 2H,
J = 6.4 Hz, OH–CH2–CH); 7.68–7.75 (m, 2H, Ar HC-5,
6); 7.81–7.87 (m, 2H, Ar HC-4, 7) ppm. MS: m/z 219 M+
(4),189 (46), 160 (100).
a white solid. Yield: 92% (75 xx). M.p. 83 °C. [α]20
=
D
1
7.2 (c 2, CHCl3). H-NMR: δ 1.05 (d, 3H, J = 6.7 Hz,
CH3–CH); 2.06–2.18 (m, 1H, CH2–CH–CH2); 3.10 (dd,
1H, J = 10.5 Hz and J = 7.5 Hz, I–CHH–CH); 3.23 (dd,
J = 10.5 Hz and J = 6.0 Hz, 1H, I–CHH–CH); 3.55–3.69
(m, 2H, CH–CH2–NR2); 7.68–7.75 (m, 2H, Ar HC-5, 6);
7.81–7.88 (m, 2H, Ar HC-4, 7) ppm. MS: m/z 329, M+
(< 1), 202 (77), 160 (100).
The same reaction run on 2b gave the (-)-(R) enanti-
omer (3d) as a white solid. Yield and non-chiroptical
physical characteristics are identical. [α]20 = –7.8 (c 2,
The same reaction run on 1b gave the (+)-(R) enan-
tiomer (2b) as a white solid. Yield and non-chiroptical
physical characteristics are identical. [α]20D = +16.5 (c 2,
CHCl3).
D
CHCl3).
5.1.4 (+)-(S)-2-[3-(2,6-dimethylphenoxy)-2-
methylpropyl]-1H-isoindole-1,3(2H)-dione (4a)
To a stirred solution of 2,6-dimethylphenol (5.5 mmol)
in dry DMF (30 mL) under N2 atmosphere NaH, 60% oil
dispersion (5.25 mmol), was added. After H2 evolution
ceased, 3a (5 mmol) was added. The solution was then
stirred at room temperature for 3 h then poured into a
funnel separator with EtOAc and H2O and stirred. The
organic layer was washed with 2 N NaOH, and H2O,
dried (Na2SO4) and concentrated in vacuo. Purification of
the crude oil residue by column chromatography (petro-
leum ether/EtOAc 8:2) and recrystallization from
5.1.2. (+)-(S)-2-(3-bromo-2-methylpropyl)-
1H-isoindole-1,3(2H)-dione (3a)
Into a stirred, ice-cooled solution of 2a, (6 mmol) in
dry toluene (30 mL) under N2 atmosphere, HBr was
bubbled until saturation occurred. The mixture was
stirred at –5 °C for 1 h and at room temperature for 4 h.
The solvent was evaporated in vacuo, the residue was
taken up with EtOAc and washed with 6 N NaOH, 2 N
HCl, and H2O. The combined organic layers were dried
(Na2SO4) and concentrated in vacuo. Purification of the
crude solid residue by recrystallization from MeOH gave
benzene/hexane gave 4a as a white solid. Yield: 45% (35
1
xx). M.p. 73–74 °C. [α]20 = +5.5 (c 2, CHCl3). H-
D
3a as a white solid. Yield: 90% xx. M.p. 89–90 °C. [α]20
NMR: δ 1.09 (d, 3H, J = 6.9 Hz, CH3–CH); 2.23 (s, 6H,
2CH3–Ar); 2.51–2.62 (m, 1H, CH2–CH–CH2); 3.63–3.92
(m, 4H, O–CH2–CH and CH–CH2–NR2); 6.84–6.89 (m,
1H, ArO HC-4≠); 6.94–6.97 (m, 2H, ArO HC-3≠, 5≠);
7.67–7.72 (m, 2H, Ar HC-5, 6); 7.81–7.87 (m, 2H, Ar
HC-4, 7) ppm. MS: m/z 323 M+ (3), 160 (99), 202 (100).
D
= +5.3 (c 2, CHCl3). 1H-NMR: δ 1.06 (d, 3H, J = 6.8 Hz,
CH3–CH); 2.33–2.45 (m, 1H, CH2–CH–CH2); 3.29–3.43
(m, 2H, Br–CH2–CH); 3.59–3.75 (m, 2H, CH–CH2–
NR2); 7.68–7.74 (m, 2H, Ar HC-5, 6); 7.81–7.87 (m, 2H,
Ar HC-4, 7) ppm. MS: m/z 281 M+ (7), 202 (21), 160
(100).
The same reaction run on 3b gave (-)-(R) enantiomer
(4b) as a white solid. Yield and non-chiroptical physical
The same reaction run on 2b gave the (-)-(R) enanti-
omer (3b) as a white solid. Yield and non-chiroptical
characteristics are identical. [α]20 = –5.7 (c 2, CHCl3).
D
physical characteristics are identical. [α]20 = –5.3 (c 2,
When repeating the reaction using 3c or 3d as sub-
D
CHCl3).
strates, similar results were obtained.

154
5.1.5. (S)-3-(2,6-dimethylphenoxy)-2-methyl-
1-propanamine (Me2)
1H, CH–CHH–NHR); 3.46–3.54 (m, 1H, CH–CHH-
–NHR); 3.63–3.73 (m, 2H, O–CH2–CH); 6.24 (bs, 1H,
CH2–NH–CO); 6.89–6.92 (m, 1H, Ar HC-4); 7.00 (ap-
parent d, 2H, Ar HC-3, 5) ppm. MS: m/z 235 M+ (< 1),
121 (15), 114 (100).
To a stirred solution of 4a (1 mmol) in absolute EtOH
(10 mL), glacial AcOH (3 mmol) and N2H4 H20
(3 mmol) were added. The mixture was heated under
reflux for 1.5 h and was filtered on a Celite pad; the
solvent was evaporated in vacuo. The residue was taken
up with EtOAc and extracted with 2 N HCl, then the
aqueous phase was made alkaline with 2 N NaOH and
extracted twice with EtOAc. The combined organic
layers were dried (Na2SO4) and concentrated in vacuo.
The final product is a colourless oil whose purity, checked
by GC-MS, was considered sufficient for the subsequent
reaction. MS: M/z 193 M+ (11), 122 (54), 72 (100).
The same reaction run on 4b gave the (R) enantiomer
as a colourless pure oil.
The same reaction run on (R)-Me2 gave the (R)
enantiomer (6b). Yield and non-chiroptical physical char-
acteristics are identical. [α]20 = –6.4 (c 2, CHCl3).
D
5.2. Pharmacology
5.2.1. Recordings of sodium current using the
three vaseline gap voltage clamp technique
Voltage clamp recordings of sodium currents were
performed on 1–2 cm long segments of single muscle
fibres obtained by microsurgery from the ventral branch
of the semitendinosus muscle of Rana esculenta. The cut
end fibre was superfused with an internal solution of
105 mM CsF, 5 mM MOPS, 2 mM MgSO4, 5 mM EGTA
and 0.55 mM Na2ATP (pH = 7.2 with NaOH concen-
trated solution) and transferred into the recording cham-
ber, filled with the same solution, as described in detail
elsewhere [14, 42]. Four KCl/agar bridge electrodes con-
nected each of the four pools of the recording chamber to
the voltage clamp amplifier [14]. Then the solution in one
of the central pools (70–100 µm wide) was replaced by an
external solution of 77 mM NaCl, 38 mM Coline-Cl,
1.8 mM CaCl2, 2.15 mM Na2HPO4 and 0.85 mM
NaH2PO4. Recordings were performed at 10 °C. The
circuit of the voltage clamp amplifier was based on that
described by Hille and Campbell [42]. The holding po-
tential (h.p.) was –100 mV. The voltage clamp amplifier
was connected via a 12-bit AD/DA interface (Digidata
1200, Axon instruments, Forster city, CA, USA) to an
80486 DX2/66 PC. The stimulation protocols and data
acquisitions were driven by the Clampex program
(pClamp 6 software package, Axon Instruments). The
sodium currents flowing in response to depolarizing
command voltages were lowpass filtered at 10 kHz
(Frequency Devices, USA) visualized on an oscilloscope,
sampled at 20 kHz and stored on the hard disk. When
necessary leak and capacity currents were subtracted
using the P/4 method. The acquired traces were later
analysed with the Clampfit program (pClamp 6 software
package Axon Instruments). Maximal sodium currents
(INa max) were elicited with test pulses from the h.p. to
–20 mV for 10 ms. The tonic block exerted by the test
compounds was evaluated as percent reduction of the
peak sodium current elicited by single test pulses. The
evaluation of the use-dependent block by the drugs was
made by using a 10 Hz train of test pulses for a period of
30 s and by normalizing the residual current at the end of
5.1.6. (-)-(S)-3-(2,6-dimethylphenoxy)-2-methyl-
1-propanamine hydrochloride (5a)
Into a solution of (S)-3-(2,6-dimethylphenoxy)-2-
methyl-1-propanamine in dry Et2O was bubbled HCl
until saturation was reached. The organic phase was
concentrated. Purification of the crude residue solid by
recrystallization from CH2Cl2/petroleum ether gave 5a as
a white solid. Yield from 4a to 5a: 50% xx. M.p.
1
133–135 °C. [α]20 = –2.95 (c 2, MeOH). H-NMR: δ
D
1.16 (d, 3H, J = 6.8 Hz, CH3–CH); 2.24 (s, 6H,
2CH3–Ar); 2.45–2.56 (m, 1H, CH2–CH–CH2); 3.11 (dd,
1H, J = 12.8 Hz and J = 6.8 Hz, CH–CHH–NH2); 3.31
(dd, 1H, J = 12.8 Hz and J = 6.6 Hz, CH–CHH–NH2);
3.64–3.75 (m, 2H, O–CH2–CH); 6.94–6.97 (m, 3H, Ar);
8.45 (bs, 2H, CH2–NH2) ppm. Anal. (C12H20ClNO) C, H,
N.
The same reaction run on 4b gave the (+)-(R) enanti-
omer (5b) as a white solid. Yield and non-chiroptical
physical characteristics are identical. [α]20D = +2.75 (c 2,
CHCl3). Anal. (C12H20ClNO) C, H, N.
5.1.7. (+)-(S)-N-acetyl-3-(2,6-dimethylphenoxy)-
2-methyl-1-propanamine (6a)
To a stirred solution of (S)-Me2 (1 mmol, obtained by
extraction from 5a) in dry THF (3 mL) AC2O (1 mmol)
and Et3N (1 mmol) were added and the solution was
stirred at room temperature for 8 h. The solvent was
evaporated in vacuo and the residue was taken up with
EtOAc and washed with a saturated aqueous solution of
NaHCO3 and brine. The organic layer was dried
(Na2SO4) and concentrated in vacuo. Yield from 5a to 6a:
52%. M.p. 84–86 °C. [α]20 = +6.6 (c 2, CHCl3).
1H-NMR: δ 1.05 (d, 3H, J = D7.0 Hz, CH3–CH); 1.98 (s,
2H, 2CH3–CO); 2.14–2.22 (m overlap s at 2.26, 1H,
CH2–CH–CH2); 2.26 (s, 6H, 2CH3–Ar); 3.27–3.36 (m,

155
this stimulation protocol with respect to that in the
absence of the drug [14].
final concentrations to be tested in vitro were obtained by
further diluting the stock solution as needed. On both
mouse intercostal muscle and frog muscle fibres, no more
than three concentrations of the same compound were
tested, and the preparations were exposed to each con-
centration for at least 10 min before recording to allow
the maximum effect of the drug to be reached.
5.2.2. Current clamp recordings of macroscopic
membrane excitability of ADR myotonic mice
Myotonic ADR (genotype adr/adr) mice of 3 months
were used for all the experiments [25, 26]. The intercostal
muscle was dissected under urethane anaesthesia (1.2 g/
kg i.p.), and immediately placed in a muscle bath
containing a normal physiological solution of 148 mM
NaCl, 4.5 mM KCl, 2.0 mM CaCl2, 10 mM MgCl2,
12 mM NaHCO3, 0.44 mM NaH2PO4 and 5.55 mM glu-
cose. The perfusing solution was continuously gassed
with 95% O2 and 5% CO2 (pH = 7.2–7.4) and the
electrophysiological recordings were performed at 30 °C.
The excitability parameters of sarcolemma were recorded
with the two intracellular microelectrode technique [14,
43]. A voltage sensing electrode and a current passing
electrode were placed 50 µm apart in the central region of
randomly selected superficial fibres. After the recording
of the membrane resting potential (RMP), the fibres were
current-clamped at –80 mV. The excitability characteris-
tics of the sampled fibres were determined by recording
the intracellular membrane potential response to square-
wave depolarizing constant current pulses. The current
intensity was gradually increased until the depolarization
was just sufficient to elicit a single action potential. The
current intensity was then further increased to generate
two or more action potentials. In this way it was possible
to record and measure the various excitability para-
meters [14]. The antimyotonic activity of the test com-
pounds has been evaluated at the maximum number of
action potentials that the muscle fibres could generate (N
spikes) which is abnormally increased in myotonic
muscle [26, 44]. Other peculiar features of the myotonic
fibres were the occurrence of spontaneous discharges
(SpD) of action potentials on the insertion of the voltage
sensing microelectrode and of self sustained after dis-
charges (AD), i.e. discharges of action potential after the
end of the application of the depolarizing stimulus [26,
44]. Thus, in each preparation, the number of fibres
showing either spontaneous discharges or after discharges
was calculated as a percentage of the total number of
fibres sampled both in the absence and in the presence of
the drugs in order to evaluate their ability to suppress
these pathological manifestations in each individual
muscle preparation.
5.2.4. Statistical analysis
The data were expressed as mean ± SEM. The statis-
tical significance of the differences between groups of
means was calculated by unpaired Students t-test. The
molar concentrations of the drugs producing a 50% block
of INa max (IC50) were determined by using a non-linear
least square fit of the concentration–response curves to
the logistic equation described in detail elsewhere [14].
The estimates of SE and n for normalized percent values
were obtained as described by Green and Margeri-
son [45].
Acknowledgements
This work was supported by grants from the Ministero
dell’Università e della Ricerca Scientifica e Tecnologica
(MURST) and Telethon-Italy (no. 901). The authors
thank Dr Giuseppe Carbonara for his helpful hints about
chemical-physical constant calculations.
References
[1] Vaughan Williams E.M., in: Simmonds A.B. (Ed.), Advances in
Drug Research, Academic Press, London, 1974, pp. 69–101.
[2] Catterall W.A., Trends Pharmacol. Sci. 8 (1987) 57–65.
[3] Ragsdale D.S., McPhee J.C., Scheuer T., Catterall W.A., Science
265 (1994) 1724–1728.
[4] Hondeghem L.M., Katzung B.G., Biochim. Biophys. Acta 472
(1977) 373–398.
[5] Hondeghem L.M., Katzung B.G., Annu. Rev. Pharmacol. 24 (1984)
387–423.
[6] Grant A.O., Wendt D.J., Trends Pharmacol. Sci. 13 (1992) 352–358.
[7] De Luca A., Pröbstle T., Brinkmeier H., Rüdel R.,
Naunyn-Schmiedeberg’s Arch. Pharmacol. 344 (1991) 596–601.
[8] Sunami A., Fan Z., Sawanobori T., Hiraoka M., Br. J. Pharmacol.
110 (1993) 183–192.
[9] Martindale– The Extra Pharmacopeia 29th ed., The Pharmaceutical
Press, London, 1986, p. 80.
[10] Abounassif M.A., Mian M.S., Mian N.A.A., in: Florey K. (Ed.),
Analytical Profile of Drug Substances, Academic Press, London,
1991, pp. 433–474.
5.2.3. Drugs
[11] Rüdel R., Lehmann-Horn F., Physiol. Rev. 65 (1985) 310–356.
[12] Lehmann-Horn F., Rüdel R., Curr. Opin. Neurol. 8 (1995) 402–410.
Stock solutions of each enantiomer of Mex and Me2
were prepared daily in physiological or external solutions
for current clamp and voltage clamp experiments. The
[13] Jackson C.E., Barohon R.J., Ptacek L.J., Muscle Nerve 17 (1994)
763–768.

156
[14] De Luca A., Natuzzi F., Lentini G., Franchini C., Tortorella V., Conte
Camerino D., Naunyn-Schmiedeberg’s Arch. Pharmacol. 352 (1995)
653–661.
[30] Roubini E., Laufer R., Gilon C., Selinger Z., Roques B.P., Chorev
M., J. Med. Chem. 34 (1991) 2430–2438.
[31] Franchini C., Cellucci C., Corbo F., Lentini G., Scilimati A.,
Tortorella V., Stasi F., Chirality 6 (1994) 590–595.
[15] Hill R.J., Duff H.J., Sheldon R.S., Mol. Pharmacol. 34 (1988)
659–663.
[32] De Luca A., Pierno S., Natuzzi F., Franchini C., Duranti A., Lentini
G. et al., J. Pharmacol. Exp. Ther. 282 (1997) 93–100.
[16] Byrnes E.W., McMaster P.D., Smith E.R., Murray R.B., Boyes R.N.,
Duce B.R. et al., J. Med. Chem. 22 (1979) 1171–1176.
[33] De Luca A., Natuzzi F., Falcone G., Duranti A., Lentini G.,
[17] Tenthorey P.A., Dirubio R.L., Feldman H.S., Takman B.H., Byrnes
E.W., McMaster P.D., J. Med. Chem. 22 (1979) 1182–1186.
Franchini
C.,
Tortorella
V.,
Conte
Camerino
D.,
Naunyn-Schmiedeberg’s Arch. Pharmacol. 356 (1997) 777–787.
[18] Frazier D.T., Narahashi T., Yamada M., J. Pharmacol. Exp. Ther. 171
(1970) 45–51.
[34] Schwarz W., Palade P.T., Hille B., Biophys. J. 20 (1977) 343–368.
[35] Wang G.K., Simon R., Bell D., Mok W.M., Wang S.Y., Mol.
Pharmacol. 44 (1993) 667–676.
[19] Hille B., J. Gen. Physiol. 69 (1977) 499–515.
[20] Postma S.W., Catterall W.A., Mol. Pharmacol. 25 (1984) 219–225.
[36] Wang G.K., Mok W.M., Wang S.Y., Biophys. J. 51 (1994) 123–135.
[21] Snyders D.J., Bennet P.B., Hondeghem L.M., in: Fozzard H.A.
(Ed.), The Heart and Cardiovascular System, Raven Press, New
York, 1992, pp. 2165–2193.
[37] Sheldon R.S., Hill R.J., Taouis M., Wilson L.M., Mol. Pharmacol.
39 (1991) 609–614.
[38] Ehring G.R., Moyer J.W., Hondeghem L.M., J. Pharmacol. Exp.
Ther. 244 (1988) 479–492.
[22] ACD/I - Lab Service, ACD.
[23] Mannhold R., Dross K.P., Rekker R.F., Quant. Struct. -Act. Relat. 9
(1990) 21–28.
[39] Quan C., Mok W.M., Wang GK., Biophys. J. 70 (1996)
194–201.
[24] Turgeon J., Uprichard A.C.G., Bélanger P.M., Harron D.W.G.,
Grech-Bélanger O., J. Pharm. Pharmacol. 43 (1991) 630–635.
[40] Nau C., Wang S.Y., Strichartz G.R., Wang G.K., Mol. Pharmacol. 56
(1999) 404–413.
[25] Gronemeier M., Condie A., Prosser J., Steinmeyer K., Jentsch T.J.,
Jockusch H., J. Biol. Chem. 269 (1994) 5963–5967.
[41] De Luca A., Natuzzi F., Desaphy J.F., Loni G., Lentini G., Franchini
C., Tortorella V., Conte Camerino D., Mol. Pharmacol. 57 (2000) in
press.
[26] Mehrke G., Brinkmeier H., Jockusch H., Muscle Nerve 11 (1988)
440–446.
[42] Hille B., Campbell D.T., J. Gen. Physiol. 67 (1976) 265–293.
[27] Steinmeyer K., Klocke R., Ortland C., Gronemeier M., Jockush H.,
Gründer S., Jentsch T.J., Nature 354 (1991) 304–308.
[43] Bryant S.H., Conte Camerino D., Pflügers Arch. 417 (1991)
605–610.
[28] Ford W.T., Pietsek D.J.J., J. Am. Chem. Soc. 97 (1975) 2194–2198.
[44] Adrian R.H., Bryant S.H., J. Physiol. 258 (1974) 125–143.
[29] Butera J.A., Spinelli W., Anantharaman V., Marcopulos N., Parsons
R.W., Moubarak I.F., Cullinan C., Bagli J.F., J. Med. Chem. 34
(1991) 3212–3228.
[45] Green J.R., Margerison D., in: Statistical Treatment of Experimental
Data, Elsevier, NY, 1978, pp. 86–90.