Synthesis and evaluation of the permeability transition inhibitory characteristics of paramagnetic and diamagnetic amiodarone derivatives

Article abstract of DOI:10.1016/j.bmc.2005.01.028

Several amiodarone analogues were synthesized varying the 2-substituent on the benzofuran ring and diethylaminoethyl side chain of phenolether by introducing 2,2,5,5-tetramethyl-2,5-dihydro-1H-pyrrole and 1,2,5,6- tetrahydropyridine nitroxides or their amino or hydroxylamino precursors. The new compounds were screened on isolated mitochondria and perfused heart and their toxicity was evaluated on WRL-68 liver cells and H9C2 cardiomyocytes. Most of the newly synthesized derivatives exerted uncoupling effect on the mitochondrial oxidative phosphorilation at higher concentrations, compared to amiodarone and one of the modified amiodarone analogues showed an effect similar to that of amiodarone on the mitochondrial permeability transition and on restoring of mitochondrial high-energy phosphate metabolites in perfused hearts. This amiodarone analogue can be new leading compound among the experimental amiodarone analogues with the same or enhanced efficiency of amiodarone, but with less side effects.

Full text of DOI:10.1016/j.bmc.2005.01.028

Bioorganic & Medicinal Chemistry 13 (2005) 2629–2636
Synthesis and evaluation of the permeability transition
inhibitory characteristics of paramagnetic and
diamagnetic amiodarone derivatives
a
Tamas Kalai, Gabor Varbiro, Zita Bognar, Anita Palfi, Katalin Hanto,
b
b
c
c
´
´
´
´
´
´
´
Balazs Bognar, Erzsebet Osz, Balazs Sumegi andKa lman Hideg
´
a
b
b,d
a,
*
}
´
´
´
´
´
´
¨
a
´
Department of Organic and Medicinal Chemistry, University of Pecs, 12 Szigeti street, H-7624 Pecs, Hungary
Department of Biochemistry and Medical Chemistry, University of Pecs, 12 Szigeti street, H-7624 Pecs, Hungary
´
b
´
´
c
´
First Department of Medicine, Division of Cardiology, Faculty of Medicine, University of Pecs, 12 Szigeti street,
´
H-7624 Pecs, Hungary
Hungarian Academy of Sciences, Research group for mitochondrial function and mitochondrial diseases, University of Pecs,
´
12 Szigeti street, H-7624 Pecs, Hungary
d
´
Received28 September 2004; accepted14 January 2005
Abstract—Several amiodarone analogues were synthesized varying the 2-substituent on the benzofuran ring and diethylaminoethyl
side chain of phenolether by introducing 2,2,5,5-tetramethyl-2,5-dihydro-1H-pyrrole and 1,2,5,6-tetrahydropyridine nitroxides or
their amino or hydroxylamino precursors. The new compounds were screened on isolated mitochondria and perfused heart and their
toxicity was evaluated on WRL-68 liver cells and H9C2 cardiomyocytes. Most of the newly synthesized derivatives exerted uncou-
pling effect on the mitochondrial oxidative phosphorilation at higher concentrations, compared to amiodarone and one of the modi-
fied amiodarone analogues showed an effect similar to that of amiodarone on the mitochondrial permeability transition and on
restoring of mitochondrial high-energy phosphate metabolites in perfused hearts. This amiodarone analogue can be new leading com-
pound among the experimental amiodarone analogues with the same or enhanced efficiency of amiodarone, but with less side effects.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
I
O
N
O
Amiodarone (2-butyl-3-benzofuranyl-4-[2-(diethylamino)-
ethoxy]-3,5-diiodophenyl-ketone hydrochloride) is a
benzfuran derivate (Fig. 1) with dominantly class III
antiarrhythmic activity.1 It is frequently usedin control-
ling intractable cardiac arrhythmias.2 There is also data
suggesting, that an important component of amioda-
roneꢀs antiarrhythmic action might be mediated via inhi-
I
O
1
Figure 1. Chemical structure of amiodarone.
3
bition of thyroidhormone action in the heart.
fatal cases were also reported.4 The most serious long-
term toxicity of amiodarone derives from its kinetics
of distribution and elimination.5
Amiodarone, however, has several toxic adverse effects,
the most important of which are thyroid, pulmonary
andhepatic toxicity. Pulmonary toxicity is reversible if
it is recognizedin time, but still fatal in 10% of the cases.
Liver injury is also common, but mild, although several
Apart from its antiarrhythmic properties, it also
possesses beneficial effect on ischemia-reperfusion
injuries.6,7 This effect of amiodarone is due—at least in
part—to the inhibition of mitochondrial permeability
transition (MPT) in lower concentrations.8 However,
when administrated in higher concentrations, it induced
mitochondrial swelling as well as the collapse of the
Keywords: Amiodarone; Nitroxides; Mitochondria; Permeability tran-
sition; Langendorff-perfused heart.
*
Corresponding author. Tel.: + 36 72 536220; fax: + 36 72 536219;
e-mail: kalman.hideg@aok.pte.hu
0968-0896/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.01.028

´
T. Kalai et al. / Bioorg. Med. Chem. 13 (2005) 2629–2636
2630
mitochondrial membrane potential (Dw).8 N-Dese-
thylamiodarone, the major metabolite of amioda-
rone—reportedby some authors to be the major cause
of the amiodarone administration induced toxicity9—
does not present these biphasic characteristics, with no
inhibitory effect on the MPT in lower concentrations
andthe induction of swelling andthe collapse of the
membrane potential in higher concentrations.10 The dif-
ference between the effect of amiodarone and N-dese-
thylamiodarone is due to the absence of an ethyl side
chain from the amino group of N-desethylamioda-
rone.10 This ledus to the conclusion that the structural
modification of amiodarone can improve its inhibitory
effect on MPT as well as its beneficial effect in ische-
mia-reperfusion injuries. In concert with this, recently,
numerous agents were shown to have an effect on the
MPT, like heterocyclic antidepressants and antipsycho-
tics with structural similarity presenting inhibitory
effect,11 or curcumin andits edrivates showing
the induction of mitochondrial swelling.12
able 2-methylbenzofuran. Compound 12 was synthe-
sizedonly for coupling with the most promising side
chain (6a, see biological data). We got mesylate 6a by
treatment of 11 paramagnetic alcohol with metha-
nesulfonyl chloride in the presence of Et3N. Compound
11 was synthesizedby alkylation of N-ethylethanol-
amine with 3a allylic bromide. To achieve better
water-soluble derivatives for biological study from 7a,
8a, 10a and 13a N-oxyl derivatives we reduced the dia-
magnetic N-hydroxylamine hydrochloride salts 7b, 8b,
10b and 13b by refluxing them in ethanol saturatedwith
HCl gas.15 These derivatives in buffer solution at basic
or at neutral pH are readily oxidized back to nitroxide
by O2 andby reactive oxygen species (ROS). On the
other hand, nitroxides are reduced to N-hydroxylamines
by any putative reducing agents of a living organism
such as ascorbate or thiols; ultimately we can assume
an equilibrium between nitroxide ꢁaꢀ form and
N-hydroxylamine ꢁbꢀ form (Scheme 1, Eq. 1), albeit N-
hydroxylamines may act as H donor antioxidants imme-
diately upon administration, eliminating harmful radical
species giving a non-reactive, resonance stabilizednitr-
oxide free radical (Eq. 1a).
Considering the advantages of amiodarone, a lot of ef-
fort has been made in the past decade to improve phar-
macokinetic properties of amiodarone, mainly by
chemical alterations of the original molecule such as
synthesis of monoiodo derivatives,3 introduction of
carboxymethoxy side chain instead of tertiary amine13
or substitute the original n-butyl group for an isobutyl
ester.14 During the past decade, it has turned out that
drugs modified with nitroxides or their precursors have
an additional, more advantageous role beyond their ori-
ginal purpose. Namely, they have some antioxidant
properties capable of scavenging most of the toxic
ROS andRNS highly damaging to biomolecules (e.g.,
proteins, lipids and nucleic acids).15–17 This prompted
us to connect the trend of amiodarone modifications
with the idea of incorporating 2,2,5,5-tetramethyl-2,5-
dihydro-1H-pyrrole, 1,2,5,6-tetrahydropyridine nitrox-
ides or their amine precursors to evaluate their influence
on pharmacological properties of amiodarone in the
hope that these new derivatives may reduce the undesir-
able side effects of the original drug.
The other methodfor improving water solubility and
antioxidant properties of the new derivatives is reducing
nitroxide to the secondary amine 7c, 8c, 10c and 13c. It
21
couldbe executedwith Fe powder in acetic acid
with-
out any alteration on the diiodobenzene ring. These sec-
ondary amine derivatives may also act as ROS
scavengers by oxidizing to nitroxide as proved experi-
mentally in the case of a cardioprotective drug22 and
in singlet oxygen scavenging in plant (Eq. 1).23 From
compound 7c we made the N-methyl derivative 7d by
refluxing this molecule with excess MeI in the presence
of K2CO3 (Scheme 2).
2.2. Pharmacology
2.2.1. Effect of amiodarone analogues on the mitochon-
drial permeability transition and the membrane potential
(Dw). Amiodarone 1 inhibits the mitochondrial perme-
ability transition (MPT) with an IC50 value of
3.9 0.8 lM, however, in concentrations above
10 lM, it induces a swelling of its own.8 In the case of
the modified amiodarone analogues only compounds
8c and 10c exerteda biphasic effect as they induced
mitochondrial swelling. The other compounds did not
induce mitochondrial swelling below the concentration
of 100 lM. Furthermore, while amiodarone exerts a
temporary inhibitory effect on the Ca2+-induced mito-
chondrial swelling,8 compounds 7a, 8a, 8b and 10b pre-
sentedan inhibitory effect similar to the one causedby
Our current studies were focused on synthesis of several
new amiodarone analogues with varying 2-substituent
on benzofuran ring and diethylaminoethyl side chain
of phenolether, andinvestigation of their effect on
MPT in isolated mitochondria and Langendorff-per-
fusedheart andmonitoring their toxicity on WRL-68
liver cells andH9C2 cardiomyocytes.
2. Results and discussion
2.1. Chemistry
thiols, ascorbate
ROS
(1)
N
OH
R., O2, ROS
N
H
N
O
The paramagnetic amiodarone derivatives were
achievedby alkylating 213 and 123 diiodophenoles in
the presence of K2CO3 and18-crown-6 in dry acetone
with the corresponding paramagnetic bromides 3a18
and 4a19 or with N-chloroethylpiperidine 520 or mesylate
such as 6a. The primary starting compoundwas 2,
which is readily available from the commercially avail-
c
a
b
(1a)
+
RH
N OH
+
N O
R
Scheme 1. Possible transformations of nitroxides or their precursors.

´
T. Kalai et al. / Bioorg. Med. Chem. 13 (2005) 2629–2636
2631
I
O
I
Q
7, 8, 10, 13
O
a
OR
+
R
X
OH
a
O
b
I
b
OH
c
X = Br (3a, 4a),
Cl (5), OMs (6a)
I
O
c
O
H
7-10
d
d
CH3
2
4, 8
5, 9
6, 10
3,7
N
N
R
N
Q
N
Q
N
Q
N
f
e
OH
6a
3a
N
O
11
I
I
I
O
N
O
O
a
OH
N
+
6a
Q
I
O
O
13
12
Scheme 2. Synthesis of paramagnetic and diamagnetic amiodarone derivatives. Reagents and conditions: (a) acetone, K2CO3 (1.1 or 2.1 equiv in case
of 5), 18-crown-6 (0.05 equiv), stirring at rt for 30 min then addition of 3a–6a (1.1 equiv) reflux for 24 h, 35–62%; (b) EtOH/HCl, reflux for 30 min,
55–71%; (c) Fe (10 equiv)/AcOH, 70 ꢁC, 1 h, then H2O, K2CO3, extraction with CHCl3 36–51%; (d) 8c, MeI (2 equiv), K2CO3 (2 equiv), CHCl3,
reflux, 24 h, 44%; (e) 3a (1.1 equiv), N-ethylethanolamine, K2CO3 (1.1 equiv), CHCl3, reflux, 4 h, 75%; (f) MsCl (1.1 equiv), Et3N (1.1 equiv),
CH2Cl2, (0 ꢁC ! rt), 1 h, 66%.
the classical MPT inhibitor, cyclosporine A (CsA) when
the permeability transition was monitoredfor more than
0
30 min (data not shown). Amiodarone 1, due to its
uncoupling effect, also induces the collapse of the mem-
1
brane potential with an ED50 value of 4.2 0.7 lM.10
While a moderate uncoupling proved to be beneficial
50
in decreasing the ROS production,24 total uncoupling
leads to energetic failure of the mitochondria due to
3
100
2
the cessation of ATP production. The effect of the differ-
ent amiodarone analogues on the permeability transi-
5 min.
tion andmembrane potential in isolatedmitochondria
with IC50 (Fig. 2) andED 50 (Fig. 3) values, respectively,
are presentedin Table 1. Basedon the results obtained
Figure 2. Mitochondrial permeability transition was induced by
the administration of 60 lM of Ca2+ (indicated with the arrow) in
the absence (line 2) andpresence of inhibitors. Line 3 represents the
swelling in the presence of agents showing IC50 inhibitory effect, line 1
represents the control.
from these experiments, three agents hadIC
values
50
below 10 lM, 8c, 10b and 10c, showing similarity with
amiodarone, however, each of these drugs exerted
uncoupling effect on the mitochondrial oxidative phos-
phorylation in higher concentrations, comparedto ami-
odarone, as it is characterized by elevated ED50 values.
nificantly decreased above the concentration of 10 lM
(Fig. 4A). In H9C2 cells 8c and 10b exhibiteda similar
effect as amiodarone, showing no or only minimal toxic-
ity up 100 lM. Compound 10c however provedto be
toxic above 10 lM (data not shown), showing similar ef-
fect as did 7a. When toxicity was assessedin WRL-68
cells, 8c decreased the viability more than amiodarone
above the concentration of 10 lM. In contrast to this,
10b was less toxic in WRL-68 cells when comparedto
8c above the concentration of 10 lM (Fig. 4B). Based
on the results obtainedfrom the viability test, 8c and
10b provedto be non-toxic on the cardiomyocyte cells.
This is also in accordwith the results obtainedfrom
2.2.2. Effect of amiodarone analogues on the viability of
cultured cells. Viability of WRL-68 liver cells andH9C2
cardiomyocytes was detected by MTT+ method( Fig. 4).
The sensitivity of the cell lines towards amiodarone
shows a difference with extracardiac cells in general,
being more susceptible to amiodarone toxicity.8
Amiodarone 1 up to 100 lM did not decrease the viabil-
ity of H9C2 cells, however in WRL-68 the viability sig-

´
T. Kalai et al. / Bioorg. Med. Chem. 13 (2005) 2629–2636
2632
higher (54 5%), comparedto the effect of 10 lM of
amiodarone (Table 2). When compound 8c was admin-
isteredin the concentration of 5 lM, it facilitatedthe
recovery of creatine phosphate to 51 4% of the initial
value. Although this was higher comparedto the con-
trol, it does not reach the level achieved by the adminis-
tration of 10 lM of amiodarone (74 6%). When 8c
was administered in the concentration of 10 lM, the
level of creatine phosphate returnedto 54 4% of the
initial value (Table 2). We can conclude that 8c did
not have any beneficial effect on the recovery of ATP,
as it was foundto be 27 3% and34 3% when admin-
isteredin 5 or 10 lM, respectively, comparedto
28 5% in the case of the control (Table 2). Basedon
the results obtainedfrom the heart perfusion experi-
ments, we foundthat 5 lM of 10b hada similar effect
as 10 lM of amiodarone, however 8c did not reach the
level of efficiency in facilitating the recovery of high-
energy phosphate metabolites of the mitochondria
comparedto equimolar concentration of amio-d
arone 1.
100
2
50
0
3
1
1 min.
Figure 3. Mitochondrial membrane potential was monitored by the
release of the fluorescent dye Rhodamine 123. Mitochondria, when
added (indicated by the black arrow) sequestered the fluorescent dye
(control, line 1). Following the administration of 60 lM of Ca2+
(indicated with the arrow) Rhodamine 123 was released from the
mitochondria (line 2). Line 3 represents the fluorescent dye released
following the administration of Ca2+ in the presence of agents
exhibiting ED50 effect.
Table 1. IC50 andED
values of amiodarone analogues presenting
50
inhibitory effect on the mitochondrial permeability transition and
inducing the collapse of the mitochondrial membrane potential,
respectively
CompoundIC
(lM)
ED50 (lM)
50
3. Conclusions
Amiodarone 1
7a
3.9 0.8
8.1 1.4
6.1 0.5
16.8 2.1
15.8 2.1
35.4 5.1
2.1 0.2
>100
4.2 0.7
20.1 2.2
13.4 3.0
27.8 2.4
21.8 2.2
51.3 6.1
7.9 0.8
We report the synthesis of new amiadorone analogues,
which were modified on benzofuran ring, for example,
2-methyl versus 2-butyl, andin aliphatic moiety of 2,6-
diiodophenolether by incorporating 2,2,5,5-tetramet-
2,5-dihydro-1H-pyrrole, 1,2,5,6-tetrahydropyridine nitr-
oxides or their hydroxylamine or amine and precursors.
It was foundthat 2-butylbenzofurane derivatives were
less potent than 2-methylbenzofurans, although most
derivatives synthesized exhibited weaker effect on mito-
chondrial permeability transition than amiodarone 1 it-
self. Compounds 8c, 10b and 10c hadsimilar inhibitory
effect on mitochondrial permeability transition as amio-
darone, but compound 10b lackedthe biphasic charac-
teristic, as this agent did not induce swelling up to the
concentration of 100 lM. In addition to this, com-
pounds 7a, 8a, 8b and 10b presenteda total inhibition
of MPT as did 1 lM of CsA. Elevateduncoupling con-
centrations, indicated by the higher ED50 values were
also a feature of most of the compounds. This would
allow the administration of these agents in higher con-
centration without deteriorating the mitochondrial
energy production by uncoupling the oxidation and
the phosphorylation. Among the other parameters tox-
icity andrestoring of the level of high-energy phosphate
metabolites in perfusedhearts, compound 10b have sim-
ilar effects as amiodarone 1, however this effect couldbe
achievedin administering this compoundin a lower con-
centration. Compound 10b, therefore seems a good
starting point for further toxicology studies and rat
pharmacokinetic studies.
7c
7d
8a
8b
8c
9
123.5 9.7
36.2 3.2
11.9 1.8
29.7 4.8
39.2 4.3
10b
10c
13b
13c
4.9 0.5
3.1 0.4
55.9 6.4
52.7 4.9
the effect of these agent on the permeability transition
andmembrane potential in isolatedmitochondria.
2.2.3. Effect of amiodarone analogues on the ischemia-
reperfusion injury in Langendorff-perfused rat hearts.
Basedon the results obtainedfrom the previous experi-
ments, amiodarone analogues 8c and 10b were selected
for further study in the perfused heart model. The re-
sults are summarizedin Table 2. Amiodarone 1 proved
to have beneficial effects on the level of mitochondrial
high-energy phosphate metabolites in perfusedhearts
31
as detected by P NMR.10 During reperfusion the level
of creatine phosphate returnedto about 35 4% of the
normoxic level (Table 2). However, following the
administration of 10 lM of amiodarone, the concentra-
tion of creatine phosphate returnedto 74 6% of the
initial level (Table 2). In addition, amiodarone adminis-
tration significantly (p < 0.01) increasedthe recovery of
ATP levels as well (45 6% vs 28 5%; Table 2). Fol-
lowing the administration of compound 10b in the con-
centration of 5 lM, it facilitatedthe recovery of the
creatine phosphate level to 77 6% of the initial value
(Table 2). The rate andextent of the recovery of creatine
phosphate was slightly higher comparedto the one in-
duced by amiodarone. The recovery in the level of
ATP following 10c administration was also faster and
4. Experimental
4.1. Chemical synthesis
Melting points were determined with a Boetius micro
melting point apparatus andare uncorrected. Elemental

´
T. Kalai et al. / Bioorg. Med. Chem. 13 (2005) 2629–2636
2633
Figure 4. The concentration dependent effect of amiodarone and its analogues on the viability of WRL-68 liver (A) and H9C2 cardiomyocyte cell (B)
lines following a 48 h treatment as detected by MTT+ method.
Table 2. Effect of 10 lM of amiodarone 1, 10 lM of 8c and5 lM of
10b on the levels of creatine phosphate andATP following ischemia-
reperfusion in Langendorff-perfused rat hearts
mercially preparedplates (20 · 20 · 0.02 cm) coated
with Merck Kieselgel GF254. Compounds 2,13 3a,18
4a,19 520 and 123 were preparedaccording to published
procedures.
Control (%) Amiodarone (%) 8c (%) 10b (%)
Creatine
phosphate
ATP
35
28
4
5
74
45
6
6
54
34
4
3
77
54
6
5
4.1.1. Alkylation of diiodophenols, general procedure (7a,
8a, 9, 10a, 13a). A mixture of compound 2 (1.0 g,
2.0 mmol) or 5 (1.09 g, 2.0 mmol) finely powdered
K2CO3 (203 mg, 2.2 mmol or 580 mg, 4.2 mmol in case
of 5), 18-crown-6 in acetone (30 mL) was stirredfor
10 min, then alkyl-halogenide or mesylate 3a, 4a, 6a
(2.2 mmol) dissolved in acetone (10 mL) or HCl salt of
compound 5 (2.2 mmol) was added in one portion and
the mixture was stirredandrefluxedfor 24 h. After cool-
ing the inorganic salts were filteredoff, the acetone was
evaporated, the residue was dissolved in EtOAc
(20 mL), the organic phase was washedwith brine
(10 mL), separated, dried (MgSO4), filteredandevapo-
rated. The remaining oil was purified by flash column
chromatography (hexane/Et2O or hexane/EtOAc) to
give amiodarone derivatives in 35–62% yield.
analyses (C, H, N, S) were performedon Fisons EA
1110 CHNS elemental analyzer. Mass spectra were re-
corded on Finnigan Automass-Multi instrument in the
1
EI mode. H NMR spectra of diamagnetic compounds
were recorded with Varian Unity Inova 400 Wb spec-
trometer; chemical shifts were referencedto TMS.
ESR spectra were taken on Miniscope MS 200 in
10À4 M CHCl3 solution andall monoradicals gave trip-
let line aN = 14.0–14.4 G. Flash column chromatogra-
phy was performedon Merck Kieselgel 60 (0.040–
0.063 mm). Qualitative TLC was carriedout on com-

´
T. Kalai et al. / Bioorg. Med. Chem. 13 (2005) 2629–2636
2634
4.1.1.1. 2-Methyl-3-[3,5-diiodo-4-(1-oxyl-2,2,5,5-tetra-
4.1.2.4. 2-Butyl-3-(3,5-diiodo-4-{2-[N-ethyl-N-(1-oxyl-
2,2,5,5-tetramethyl-2,5-dihydro-1H-pyrrol-3-ylmethyl)-
ethyl]}oxybenzoyl)benzofuran, 2HCl salt (13b). 463 mg
(55%), mp 112–114 ꢁC. Anal. Calcdfor C 32H42Cl2I2N2O4:
C, 45.57; H, 5.02; N, 3.32. Found: C, 45.40; H, 4.93; N,
3.22.
methyl-2,5-dihydro-1H-pyrrol-3-ylmethyl)oxybenzoyl]ben-
zofuran, radical (7a). 551 mg (42%), mp 179–182 ꢁC.
Anal. Calcdfor C 25H24I2NO4: C, 45.75; H, 3.69; N,
2.13. Found: C, 45.71; H, 3.64; N, 2.05.
4.1.1.2. 2-Methyl-3-[3,5-diiodo-4-(1-oxyl-2,2,6,6-tetra-
methyl-1,2,5,6-tetrahydropyridine-4-ylmethyl)oxybenzo-
yl]benzofuran, radical (8a). 469 mg (35%), mp 167–
168 ꢁC. Anal. Calcdfor C 26H26I2NO4: C, 46.59; H,
3.91; N, 2.09. Found: C, 46.55; H, 3.88; N, 1.94.
4.1.3. General procedure for reduction of nitroxides to
amines (7c, 8c, 10c, 13c). To a solution of nitroxide 7a or
8a or 10a or 13a (1.0 mmol) in AcOH (8 mL) Fe powder
(560 mg, 10 mmol) was added and the mixture was
warmedup to 70 ꢁC until the reaction started. The mix-
ture was stirredat room temperature for 1 h, diluted
with water (15 mL), decanted and the decanted aq solu-
tion made alkaline with solid K2CO3. The mixture was
4.1.1.3. 2-Methyl-3-{3,5-diiodo-4-[2-(2,2,6,6-tetrame-
thylpiperidin-1-yl)ethyl]oxybenzoyl}-benzofuran, radical
(9). 698 mg (52%), mp 151–153 ꢁC (HCl). Anal. Calcd
for HCl salt C27H32ClI2NO3: C, 45.82; H, 4.56; N,
1.98. Found: C, 45.76; H, 4.52; N, 1.82.
extractedwith CHCl (3 · 15 mL), dried (MgSO4), fil-
3
tered, evaporated and after chromatographic purifica-
tion (CHCl3/MeOH) we got the title amines in 36–51%
yield.
1H NMR (CDCl3): d 1.42 (6H, s, CH3), 1.60 (2H, m,
CH2), 1.77 (2H, m, CH2), 1.79 (6H, s, CH3), 2.57 (3H,
s, CH3), 2.73 (2H, m, CH2), 3.45 (2H, t, J = 8.6 Hz,
CH2), 4.72 (2H, t, J = 8.6 Hz, CH2), 7.29 (2H, m, aro-
matic CH), 7.46 (2H, m, aromatic CH), 8.18 (2H, s, aro-
matic CH).
4.1.3.1. 2-Methyl-3-[3,5-diiodo-4-(2,2,5,5-tetramethyl-
2,5-dihydro-1H-pyrrol-3-ylmethyl)oxybenzoyl]benzofuran
(7c). 256 mg (40%), mp 222–224 ꢁC (HCl). Anal. Calcd
for C25H26ClI2NO3: C, 44.30; H, 3.87; N, 2.07. Found:
C, 44.21; H, 3.75; N, 1.91. 1H NMR (DMSO-d6): d
1.55 (6H, s, CH3), 1.66 (6H, s, CH3), 2.48 (3H, s,
CH3), 4.60 (2H, s, CH2), 6.05 (1H, s, @CH), 7.30 (1H,
m, aromatic CH), 7.37 (1H, m, aromatic CH), 7.45
(1H, m, aromatic CH), 7.66 (1H, m, aromatic CH),
8.19 (2H, s, aromatic CH), 9.28 (2H, br s, NH HCl).
4.1.1.4.
2-Methyl-3-(3,5-diiodo-4-{2-[N-ethyl,N-(1-
oxyl-2,2,5,5-tetramethyl-2,5-dihydro-1H-pyrrol-3-ylmeth-
yl)ethyl]}oxybenzoyl)benzofuran, radical (10a). 843 mg
(58%), oil. Anal. Calcdfor C 29H33I2N2O4: C, 47.89; H,
4.57; N, 3.85. Found: C, 47.81; H, 4.55; N, 3.71.
4.1.1.5. 2-Butyl-3-(3,5-diiodo-4-{2-[N-ethyl-N-(1-oxyl-
2,2,5,5-tetramethyl-2,5-dihydro-1H-pyrrol-3-ylmethyl)-
ethyl]}oxybenzoyl)benzofuran, radical (13a). 600 mg
(39%), mp 31–33 ꢁC. Anal. Calcdfor C 32H39I2N2O4:
C, 49.95; H, 5.11; N, 3.64. Found: C, 49.85; H, 5.03;
N, 3.56.
4.1.3.2. 2-Methyl-3-[3,5-diiodo-4-(2,2,6,6-tetramethyl-
1,2,5,6-tetrahydropyridine-4-ylmethyl)oxybenzoyl]benzo-
furan (8c). 235 mg (36%), mp 231–233 ꢁC (HCl). Anal.
Calcdfor C 26H26ClI2NO3: C, 45.14; H, 4.08; N, 2.02.
1
Found: C, 45.03; H, 4.00; N, 1.94. H NMR (DMSO-
d6): d 1.45 (6H, s, CH3), 1.50 (6H, s, CH3), 2.41 (2H,
br s, CH2), 2.48 (3H, s, CH3), 4.48 (2H, s, CH2), 5.94
(1H, s, @CH), 7.29 (1H, m, aromatic CH), 7.37 (1H,
m, aromatic CH), 7.44 (1H, m, aromatic CH), 7.65
(1H, m, aromatic CH), 8.18 (2H, s, aromatic CH),
8.88 (2H, br s, NH HCl).
4.1.2. General procedure for N-hydroxylamine salt for-
mation (7b, 8b, 10b, 13b). A solution of radicals 7a or 8a
or 10a or 13a (1.0 mmol) was refluxedin EtOH (satu-
ratedwith HCl) for 30 min. The solvent was evaporated
off, the residue was crystallized with acetone or Et2O to
give white solids in 55–71% yield.
4.1.3.3. 2-Methyl-3-(3,5-diiodo-4-{2-[N-ethyl-N-(2,2,5,5-
tetramethyl-2,5-dihydro-1H-pyrrol-3-ylmethyl)ethyl]}-
oxybenzoyl)benzofuran (10c). 363 mg (51%), mp 132–
134 ꢁC (2HCl). Anal. Calcdfor C 29H36Cl2I2N2O3: C,
44.35; H, 4.62; N, 3.57. Found: C, 44.31; H, 4.55; N,
3.44. 1H NMR (CDCl3): d 1.10 (3H, t, J = 7.0 Hz,
CH3), 1.29 (6H, s, CH3), 1.34 (6H, s, CH3), 2.56 (3H,
s, CH3), 2.68 (2H, q, J = 7.0 Hz, CH2), 3.06 (2H, t,
J = 6.5 Hz, CH2), 3.16 (2H, br s, CH2), 4.12 (2H, t,
J = 6.5 Hz, CH2), 5.56 (1H, s, @CH), 7.24 (1H, m, aro-
matic CH), 7.30 (1H, m, aromatic CH), 7.42 (1H, m,
aromatic CH), 7.47 (1H, m, aromatic CH), 8.20 (2H,
s, aromatic CH).
4.1.2.1. 2-Methyl-3-[3,5-diiodo-4-(1-hydroxy-2,2,5,5-
tetramethyl-2,5-dihydro-1H-pyrrol-3-ylmethyl)oxybenzo-
yl]benzofuran, HCl salt (7b). 471 mg (68%), mp 164–
166 ꢁC. Anal. Calcdfor C 25H26ClI2NO4: C, 43.28; H,
3.78; N, 2.02. Found: C, 43.31; H, 3.70; N, 1.95.
4.1.2.2. 2-Methyl-3-[3,5-diiodo-4-(1-hydroxy-2,2,6,6-
tetramethyl-1,2,5,6-tetrahydropyridine-4-ylmethyl)oxy-
benzoyl]benzofuran, HCl salt (8b). 424 mg (60%), mp
201–203 ꢁC. Anal. Calcdfor C 26H28ClI2NO4: C, 44.12;
H, 3.99; N, 1.98. Found: C, 44.06; H, 3.91; N, 1.86.
4.1.2.3.
2-Methyl-3-(3,5-diiodo-4-{2-[N-ethyl-N-(1-
4.1.3.4. 2-Butyl-3-(3,5-diiodo-4-{2-[N-ethyl-N-(2,2,5,5-
tetramethyl-2,5-dihydro-1H-pyrrol-3-ylmethyl)ethyl]}-
oxybenzoyl)benzofuran (13c). 294 mg, (39%), mp 127–
130 ꢁC (2HCl). Anal. Calcdfor C 32H42Cl2I2N2O3: C,
46.45; H, 5.12; N, 3.39. Found: C, 46.36; H, 4.99; N,
3.24. 1H NMR (CDCl3): d 0.90 (3H, t, J = 7.5 Hz,
hydroxy-2,2,5,5-tetramethyl-2,5-dihydro-1H-pyrrol-3-yl-
methyl)ethyl]}oxybenzoyl)benzofuran, 2HCl salt (10b).
504 mg (63%), mp 127–130 ꢁC. Anal. Calcdfor
C29H36Cl2I2N2O4: C, 43.4; H, 4.53; N, 3.50. Found: C,
43.40; H, 4.45; N, 3.39.

´
T. Kalai et al. / Bioorg. Med. Chem. 13 (2005) 2629–2636
2635
CH3), 1.09 (3H, t, J = 7.0 Hz, CH3), 1.25 (6H, s, CH3),
water andrat chow were providedadlibitum. Animals
were treatedin compliance with approvedinstitutional
animal care guidelines.
1.29 (6H, s, CH3), 1.35 (2H, m, CH2), 1.75 (2H, q,
J = 7.5 Hz, CH2), 2.67 (2H, q, J = 7.0 Hz, CH2), 2.83
(2H, t, J = 7.5 Hz, CH2), 3.05 (2H, t, J = 6.6 Hz,
CH2), 3.15 (2H, s, CH2), 4.11 (2H, t, J = 6.6 Hz, CH2),
5.54 (1H, s, @CH), 7.22 (1H, m, aromatic CH), 7.28
(1H, m, aromatic CH), 7.40 (1H, m, aromatic CH),
7.46 (1H, m, aromatic CH), 8.19 (2H, s, aromatic CH).
4.2.2. Isolation of mitochondria. Liver andheart mito-
chondria were prepared according to standard proto-
col.25 Only difference among the organs were in the
primary homogenization protocol; liver (n = 5) was
squeezedthrough a liver press, while pooledheart tissue
from five rats (n = 15) was mincedwith a blender. All
isolated mitochondria were purified by Percoll gradient
centrifuging.26 All experiments were repeatedat least
three times from the same pool of mitochondria.
4.1.4. 2-Methyl-3-[3,5-diiodo-4-(1,2,2,5,5-pentamethyl-
2,5-dihydro-1H-pyrrol-3-ylmethyl)oxybenzoyl]benzofuran
(7d). To a stirredsolution of amine
7c (712 mg,
1.0 mmol), K2CO3 (276 mg, 2,0 mmol) in CHCl3
(20 mL), MeI (710 mg, 5.0 mmol) was added and the
mixture was stirredandrefluxedfor 24 h. After cooling
the inorganic salt was filteredoff, the solvent was evap-
orated, the residue was dissolved in CHCl3 (20 mL),
washed with brine, separated, dried (MgSO4), filtered,
evaporatedandthe residue was purifiedby flash column
chromatography (CHCl3/Et2O) to give compound
288 mg (44%) 7d, mp 157–159 ꢁC (HCl salt). Anal.
Calcdfor C 26H28ClI2NO3: C, 45.14; H, 4.08; N, 2.02.
4.2.3. Mitochondrial permeability transition. Mitochon-
drial permeability transition was monitored by follow-
ing the accompanying large amplitude swelling via the
decrease in absorbance at 540 nm27 measuredat room
temperature by a Perkin–Elmer fluorimeter (London,
UK) in reflectance mode. Briefly, mitochondria at the
concentration of 1 mg protein/mL were pre-incubated
in the assay buffer (70 mM sucrose, 214 mM mannitol,
20 mM N-2-hydroxyethyl piperasine-N0-2-ethanesulf-
onic acid, 5 mM glutamate, 0.5 mM malate, 0.5 mM
phosphate) containing the studied substances for 60 s.
Mitochondrial permeability transition was induced by
the addition of either 60 lM of Ca2+, or amiodarone
or desethylamiodarone at the indicated concentration.
Decrease of E540 was detected for 20 min. The results
are demonstrated by representative original registration
curves from at least five independent experiments, each
repeatedthree times using mitochondria preparedfrom
the same liver or pool of rat hearts, respectively. The re-
sults are demonstrated by representative original regis-
tration curves from five independent experiments, each
repeatedthree times using mitochondria preparedfrom
the same liver or pool of rat hearts, respectively.
1
Found: C, 45.25; H, 4.01; N, 1.96. H NMR (CDCl3):
d 1.22 (6H, br s, CH3), 1.31 (6H, br s, CH3), 2.35 (3H,
br s, CH3), 2.56 (3H, s, CH3), 4.59 (2H, s, CH2), 5.95
(1H, s, @CH), 7.23 (1H, m, aromatic CH), 7.30 (1H,
m, aromatic CH), 7.41 (1H, m, aromatic CH), 7.47
(1H, m, aromatic CH), 8.21 (2H, s, aromatic CH).
4.1.5. 3-{[Ethyl-(2-hydroxyethyl)-amino]-methyl}-2,2,5,5-
tetramethyl-2,5-dihydro-1H-pyrrol-1-yloxyl, radical (11).
A solution of 2-(ethylamino)ethanol (891 mg, 10.0
mmol), compound 3a (2.54 g, 11.0 mmol) andK CO3
2
(1.51 g, 11.0 mmol) was stirredandrefluxedin CHCl
3
(25 mL) for 4 h. After cooling the inorganic salt was fil-
teredoff, the filtrate washedwith brine (10 mL), the or-
ganic phase was separated, dried (MgSO4), filtered,
evaporatedthe residue was purifiedby flash column
chromatography (CHCl3/MeOH) to give the title com-
pound 11 as a yellow oil 1.8 g (75%). MS (m/z, %):
241 (M+, 4), 211 (5), 210 (7), 138 (28), 102 (100). Anal.
Calcdfor C 13H25N2O2: C, 64.69; H, 10.44; N, 11.61.
Found: C, 64.60; H, 10.32; N, 11.51.
4.2.4. Mitochondrial membrane potential. Mitochondrial
membrane potential was monitoredby fluorescence of
Rh123, releasedfrom the mitochondria following the
induction of permeability transition at room tempera-
ture by using a Perkin–Elmer fluorimeter (London,
UK) at an excitation wavelength of 495 andan emission
wavelength of 535 nm. Briefly, mitochondria at the con-
centration of 1 mg protein/mL were pre-incubatedin the
assay buffer (70 mM sucrose, 214 mM mannitol, 20 mM
N-2-hydroxyethyl piperasine-N0-2-ethanesulfonic acid,
5 mM glutamate, 0.5 mM malate, 0.5 mM phosphate)
containing 1 lM Rh123 andthe studiedsubstances for
60 s. Alteration of the mitochondrial membrane poten-
tial (Dw) was induced by the addition of either 60 lM
of Ca2+, or amiodarone or desethylamiodarone at the
indicated concentration. Changes of fluorescence inten-
sity were detected for 4 min. The results are demon-
stratedby representative original registration curves
from five independent experiments, each repeated three
times using mitochondria prepared from the same liver
or pool of rat hearts, respectively.
4.1.6. Procedure for synthesis of mesylate (6a). To a stir-
redsolution of alcohol 11 (1.20 g, 5.0 mmol) andEt N
3
(550 mg, 5.5 mmol) in CH2Cl2 (10 mL) methanesulfonyl
chloride (630 mg, 5.5 mmol) dissolved in CH2Cl2 (2 mL)
was added dropwise at 0 ꢁC andthe mixture was stirred
at ambient temperature for 1 h. The reaction mixture
was washedwith water (5 mL), the organic phase was
separated, dried (MgSO4), filteredandevaporatedto
give the crude mesylate 6a 1.06 g (66%), which were
not purifiedfurther, but dissolvedin acetone andimme-
diately used for the synthesis of compounds 10a and
13a.
4.2. Biology
4.2.5. Cell culture. WRL-68 liver cells andH9C2 mouse
cardiomyocytes cells were from American Type Culture
Collection. WRL-68 andH9C2 cells were culturedin
DMEM containing 1% antibiotic–antimycotic solution
4.2.1. Animals. Wistar rats were purchasedfrom Charles
River Hungary Breeding Ltd (Budapest, Hungary). The
animals were kept under standardized conditions; tap

´
T. Kalai et al. / Bioorg. Med. Chem. 13 (2005) 2629–2636
2636
(Sigma) and10% FCS. Cells were passagedat intervals
of three days.
References and notes
1. Singh, S. N.; Fletcher, R. D.; Fisher, S. G.; Singh, B. N.;
Lewis, H. D.; Deedwania, P. C.; Massie, B. M.; Colling,
C.; Layyeri, D. N. Eng. J. Med. 1995, 333, 77.
2. Herre, J. M.; Sauve, M. J.; Malone, P.; Griffin, J. C.;
Helmy, I.; Langberg, J. J.; Goldberg, H.; Scheinmann,
M. M. J. Am. Coll. Cardiol. 1989, 13, 442.
3. Ha, H. R.; Stieger, B.; Grassi, G.; Altorfer, H. R.; Follath,
F. Eur. J. Clin. Pharmacol. 2000, 55, 807.
4. Vrobel, T. R.; Miller, P. E.; Mostow, N. D.; Rakita, L.
Prog. Cardiovasc. Dis. 1989, 31, 393.
5. Myers, J. L.; Kennedy, J. L.; Plump, V. J. Hum. Pathol.
1987, 18, 349.
6. Nokin, P.; Jungbluth, L.; Mouton, J. J. Mol. Cell. Cardiol.
1987, 19, 603.
7. Pfisters, M.; Kiowski, W.; Burckhardt, D.; Follath, F.;
Burkhart, F. Am. J. Cardiol. 1992, 69, 1399.
4.2.6. Cell viability assay.28 WRL-3A andH9C2 cells
were seeded into 96-well plates at a starting density of
2 · 104 cells/well andculturedovernight. The following
day, amiodarone and amiodarone analogues at the indi-
cated concentrations were added to the medium.
Twenty-four hours later, 0.5% of the water-soluble
mitochondrial dye, 3-[4,5-dimethylthiazol-2-yl]-2,5-di-
phenyl-tetrazolium bromide (MTT+) was added. Incu-
bation was continuedfor three more hours, the
medium was removed, and the water insoluble blue for-
masan dye formed stochiometrically from MTT+ was
solubilized by acidic isopropanol. Optical densities were
determined by an ELISA reader (Anthos Labtech 2010)
at 550 nm wavelength. All experiments were run in at
least four parallels andrepeatedthree times.
´
´
´
8. Varbiro, G.; Toth, A.; Tapodi, A.; Veres, B.; Sumegi, B.;
¨
Gallyas, F. Biochem. Pharmacol. 2003, 65, 1115.
9. Wilson, B. D.; Lippmann, M. L. Am. Rev. Respir. Dis.
1990, 141, 1553.
4.2.7. Heart perfusion. Studies were performed with iso-
lated, Langendorff-perfused rat hearts (n = 12), which
´
´
´
10. Varbiro, G.; Toth, A.; Tapodi, A.; Veres, B.; Sumegi, B.;
¨
were either treatedwith amioadrone analogues
8c
Gallyas, F. J. Pharm. Exp. Ther. 2003, 307, 615.
11. Starovkaya, I. G.; Naryanan, M. V.; Zhang, W.; Krasni-
kov, B. F.; Heemskerk, J.; Young, S. S.; Blass, J. P.;
Brown, A. M.; Beal, M. F.; Friedlander, R. M.; Kristal,
B. S. J. Exp. Med. 2004, 200, 211.
12. Ligeret, H.; Barthelemy, S.; Zini, R.; Tillement, J. P.;
Labidalle, S.; Morin, D. Free Radical Biol. Med. 2004, 36,
919.
(n = 4) or 10b (n = 5) or untreated( n = 3) as control
group. Hearts were perfusedvia the aorta as described
before.29 After a washout (non-recirculating periodof
15 min), hearts were perfused under normoxic condi-
tions for 10 min; the flow was subsequently discontinued
for 30 min by inflating a balloon (ischemia), which was
followedby 15 min of reperfusion. Levels of high-energy
phosphate intermediates were monitored in the magnet
of a 31P NMR spectroscope during the entire perfusion.
13. Carlsson, B.; Singh, B. N.; Temciuc, M.; Nilsson, S.; Li,
Y.-L.; Mellin, C.; Malm, J. J. Med. Chem. 2002, 45,
623.
14. Druzgala, P.; Milner, P. G. Patent US 2002/0193428A1.
´
15. Krishna, M. C.; Degraff, W.; Hankovszky, H. O.; Sar, P.
C.; Kalai, T.; Jeko, J.; Russo, A.; Mitchell, J. B.; Hideg, K.
4.2.8. NMR spectroscopy. NMR spectra were recorded
with a Varian Unity Inova 400 Wb instrument. 31P mea-
surements (161.90 MHz) of perfusedhearts were run at
37 ꢁC in a ZÆSPEC 20-mm broadband probe (Nalorac
Co. Martinez, CA), applying WALTZ proton decou-
pling (cB2 = 1.6 kHz) during the acquisition only. Field
}
´
J. Med. Chem. 1998, 41, 3477.
16. Li, H.; Xu, K. Y.; Zhou, L.; Kalai, T.; Zweier, J. L.;
´
Hideg, K.; Kuppusamy, P. J. Pharmacol. Exp. Ther. 2000,
295, 563.
´
17. Halmosi, R.; Deres, P.; Berente, Z.; Kalai, T.; Sumegi, B.;
Hideg, K.; Toth, K. J. Cardiovasc. Pharm. 2002, 40,
¨
1
homogeneity was adjusted by following the H signal
´
(w1/2 = 10–15 Hz). Spectra were collectedwith a time
resolution of 3 min by accumulating 120 transients in
each free induction decay. Flip angle pulses (45ꢁ) were
usedafter a 1.25-s recycle delay, andtransients were ac-
quiredover a 10-kHz spectral width in 0.25 s, andthe
acquireddata points (5000) were zero-filledto 16384.
Under the above circumstances, relative concentrations
of the species are proportional to the corresponding
peak areas, since interpulse delays exceeded 4–5 times
the T1 values of the metabolites that were analyzedin
the 31P experiments. Data were acquiredfrom five inde-
pendent experiments for each drug.
854.
18. Hankovszky, H. O.; Hideg, K.; Lex, L.; Kulcsar, G.
´
Synthesis 1980, 914.
}
´
19. Hideg, K.; Cseko, J.; Hankovszky, H. O.; Sohar, P. Can.
J. Chem. 1986, 64, 1482.
20. Robertson, J.; Biel, J. H.; DiPerro, F. J. Med. Chem. 1963,
6, 381.
´
´
´ ´
21. Sar, P. C.; Kalai, T.; Baracz, M. N.; Jerkovich, Gy.;
Hideg, K. Synth. Commun. 1995, 25, 2929.
22. Twomey, P.; Taira, J.; DeGraff, W.; Mitchell, J. B.; Russo,
A.; Krishna, M. C.; Hankovszky, H. O.; Frank, L.; Hideg,
K. Free Radical Biol. Med. 1997, 22, 909.
´
´
23. Hideg, E.; Ogawa, K.; Kalai, T.; Hideg, K. Physiol. Plant.
2001, 112, 10.
24. Kahlert, S.; Reiser, G. FEBS Lett. 2002, 529, 351.
25. Schneider, W. C.; Hageboom, G. H. J. Biol. Chem. 1950,
183, 123.
Acknowledgements
26. Sims, N. R. J. Neurochem. 1990, 55, 698.
´
27. Varbıro, G.; Veres, B.; Gallyas, F.; Sumegi, B. Free
This work was supportedby a grant from the Hungar-
ian National Research Foundations (OTKA T34307,
T023076 andM 045190), Hungarian Ministry of Health,
Social andFamily Affairs (ETT 512/2003 and35/2000),
PTE-DD-KKK andNKFP 1/026. The authors thank
´
´
Radical Biol. Med. 2001, 31, 548.
¨
28. Cassarino, D. S.; Parks, J. K.; Parker, V. D., Jr.; Benett, J.
P., Jr. Biochim. Biophys. Acta 1999, 1453, 49.
29. Szabados, E.; Fisher, M. G.; Gallyas, F.; Kispal, G.;
Sumegi, B. Free Radical Biol. Med. 1999, 27, 1103.
¨
´ ´
Noemi Lazsanyi for elemental analysis.