Nitrone derivatives of trolox as neuroprotective agents

Article abstract of DOI:10.1016/j.bmcl.2005.04.043

Synthesis of nitrone derivatives of trolox is described. Their biological evaluation was performed in vitro for scavenging different free radicals, inhibiting Fe²⁺-induced lipid peroxidation, and in vivo in a permanent middle cerebral artery occlusion model in mice. New compounds exert pharmacological activities comparable to or better than those of trolox or nitrone-type reference compounds.

Full text of DOI:10.1016/j.bmcl.2005.04.043

Bioorganic & Medicinal Chemistry Letters 15 (2005) 3012–3015
Nitrone derivatives of trolox as neuroprotective agents
a,
Gy o¨ rgy T. Balogh, Krisztina Vukics, Arp a´ d K o¨ ncz o¨ l, Agnes Kis-Varga,
a,
b
a
*
*
´
´
a
Anik o´ Gere and J a´ nos Fischer
a
a
Gedeon Richter Ltd, Budapest H-1475, 10. PO Box 27, Hungary
b
Budapest University of Technology and Economics, Budapest, M }u egyetem rkp. 3. H-1111, Hungary
Received 22 February 2005; revised 31 March 2005; accepted 21 April 2005
Available online 17 May 2005
Abstract—Synthesis of nitrone derivatives of trolox is described. Their biological evaluation was performed in vitro for scavenging
different free radicals, inhibiting Fe -induced lipid peroxidation, and in vivo in a permanent middle cerebral artery occlusion model
2
+
in mice. New compounds exert pharmacological activities comparable to or better than those of trolox or nitrone-type reference
compounds.
Ó 2005 Elsevier Ltd. All rights reserved.
1. Introduction
Much evidence suggests that biological oxidation in the
human body generates highly pathogenic free radicals
and other reactive oxygen species such as hydroxyl rad-
Å
Å
ꢀ
ical ( OH), superoxide anion ( O ), peroxynitrite
2
(
ꢀ
Å
ONOO ) and lipid peroxide radicals (ROO ), causing
cellular injury. These pathological events have impor-
tant roles in many degenerative disorders, for example,
atherosclerosis, rheumatoid arthritis, and several neuro-
degenerative diseases such as AlzheimerÕs disease, Par-
1
kinson disease, ischemic conditions and stroke.
Therefore, protection against these toxic molecules
would have a benefit in the treatment of the diseases
mentioned herein.
Figure 1.
Nitrone compounds are among the widely used free rad-
2
ical trapping agents (Fig. 1). They trap especially C-
centred radicals on their a-C atom, producing a more
stable nitroxyl radical. The beneficial effect of a-phen-
yl-N-tert-butylnitrone (PBN) was discovered in the
Research for more effective and therapeutically accept-
able free radical trapping agents for medicinal purposes
has resulted in new compounds in recent years such as S-
PBN and NXY-059, of which the latter currently is in
Phase III clinical trials for the indication of stroke.
4
1
990s. Its administration decreased the mortality rate
3
in several ischemia models in rodents. However, be-
cause of its effective dose and therapeutic window
PBN is not acceptable as a medicine.
5
During our work on nitrone compounds, we aimed at
the synthesis of new molecules keeping the nitrone com-
ponent, but combining it with other types of antioxidant
fragment the possessing ability to trap O- or N-centred
radicals as well. Trolox (1), an a-tocoferol (Vitamin E)
derivative, is one of the most powerful antioxidants,
with relatively high selectivity for scavenging peroxyni-
trite and hydroxyl radical. We hoped that the combina-
Keywords: Neuroprotection; Free radical scavenging; Lipid peroxida-
tion; Permanent middle cerebral artery occlusion; Nitrone; Trolox.
*
Corresponding authors. Fax: +36 1 2606650; e-mail addresses:
gy.balogh@richter.hu, k.vukics@richter.hu
0
960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.04.043

G. T. Balogh et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3012–3015
3013
ꢀ
tion of these two different types of free radical trapping
abilities would be favourable. For this purpose, we pro-
posed to synthesize and perform biological evaluation of
compounds 2, derivatives of trolox holding a nitrone
moiety in the position 2. We intended to separate the
free radical trapping ability of the nitrone component
from that of trolox itself; therefore we synthesized ÔsoftÕ
The ONOO scavenging capacity of the compounds was
measured by pyrogallol red bleaching assay (542 nm).
ONOO was prepared essentially by the method of
ꢀ
8
,9
Hughes and Nicklin.
The NO scavenging capacity of the compounds was
tested by direct capturing of NO from cupferron. Free
NO levels were measured indirectly by its stable, sponta-
neously arising oxidative metabolites, nitrite and nitrate
(
O-acetyl) and ÔhardÕ (O-methyl) O-protected derivatives
of 2.
with Griess–Ilosvay reagent at
1
a wavelength of
0
540 nm.
2. Chemistry
Total peroxyl radical-trapping antioxidant parameter
(TRAP) of the compounds was determined by the spec-
trophotometric measurement (417 nm) of a relatively
Trolox nitrones 2a–f were synthesized as depicted in
Scheme 1. Trolox 1 was esterified to ethyl ester 3 in
refluxing ethanol with catalytic p-toluenesulfonic acid.
Trolox aldehyde 4 was prepared from 3 in two steps,
first reduced to the corresponding alcohol with diisobu-
tyl aluminium hydride at ꢀ70 °C, followed by Swern
oxidation. Compound 4 was converted into its O-acetyl
derivative 5 by acetylation with acetyl chloride and tri-
ethyl amine in methylene chloride at 0 °C. Methoxy tro-
lox aldehyde 7 was obtained in a manner similar to 4
from methyl ester 6 prepared via methylation of trolox
with methyl iodide in the presence of potassium carbon-
ate in dimethyl formamide. Trolox nitrones 2a–f were
prepared by the condensation of the appropriate alde-
0
stable cation radical of 2,2 -azino-bis(3-ethylbenzthiazo-
line-6-sulfonic acid) (ABTS), which is generated in a
one-electron oxidation. Peroxyl radical was formed by
thermal decomposition of the water-soluble azo com-
0
pound, 2,2 -azobis(2-methylpropion-amide) dihydro-
,11
chloride (AAPH).
8
2
+
The ability of 2a–f to inhibit Fe -induced lipid peroxi-
dation (LPO) in mitochondria prepared from rat brain
was determined by spectrophotometric measurement
1
2,13
of thiobarbituric acid reactive substances (TBARs).
5
,6
hyde and N-alkylhydroxyl amine. According to the
A permanent middle cerebral artery (MCA) occlusion
1
1
14,15
homonuclear H– H NOE experiments of nitrones 2a–
f, each nitrone proved to be Z-isomer.
test,
gate the efficacy of some selected compounds (2a, 2b and
d) in vivo.
a model of focal ischemia, was used to investi-
2
3. Biology
4. Results and discussion
Free radical scavenging properties of compounds 2a–f
and reference molecules (PBN, NXY-059 and trolox)
were tested against pathophysiologically relevant reac-
Our in vitro and in vivo biological results suggest that
the combination of trolox with a nitrone moiety was
beneficial (Table 1). Our new compounds 2a–f exerted
better efficacy in in vitro tests than reference compounds
having nitrone component alone, although NXY-059 is
considered a modest free radical trapping agent accord-
ing to the literature, and other mechanisms are sup-
posed to be involved explaining its good efficacy in
vivo. Nevertheless, even O-protected compounds, espe-
Å
ꢀ
tive oxygen and nitrogen species ( OH, ONOO , NO
and ROO ).
Å
Å
The OH scavenging capacity of the compounds was
determined by the inhibition of OH induced damage
Å
4
of 2-deoxyribose. Arising thiobarbituric acid reactive
7
products were measured at a wavelength of 535 nm.
,8
Scheme 1. Reagents and conditions: (a) EtOH, pTsOH, reflux, 5 h; (b) DIBAL-H, toluene, ꢀ70 °C; (c) (1) oxalyl chloride, DMSO, CH
2
Cl
2
, ꢀ70 °C,
(
2) Et N; (d) acetyl chloride, Et N, CH Cl CO , DMF, 40 °C; (f) R NHOH, EtOH, reflux, 3 h.
3
3
2
2
, 0 °C, 1.5 h; (e) MeI, K
2
3
2

3014
G. T. Balogh et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3012–3015
Table 1. Free radical scavenging properties and inhibition of lipidperoxidation for compounds 2a–f
Å
a
ꢀ
a
a
a
a
Compds
OH IC50 (mM)
ONOO IC50 (mM)
NO IC50 (mM)
TRAP IC50 (lM)
LPO IC50 (lM)
2
2
2
2
2
2
a
b
c
d
e
f
0.17 (±0.03)
0.15 (±0.01)
3.2 (±1.4)
1.1 (±0.3)
na
6.8 (±2.2)
12 (±1.8)
19 (±1.2)
7.4 (±0.8)
22 (±4)
5.5 (±2.0)
6.0 (±1.2)
895 (±110)
117 (±9)
2.0 (±0.6)
0.68 (±0.04)
10 (±1.3)
3.7 (±0.6)
23 (±2.4)
20 (±1.8)
25 (±0.6)
16 (±1.3)
5491 (±58)
na
na
0.59 (±0.22)
3.7 (±0.7)
1.2 (±0.3)
0.55 (±0.09)
na
b
na
na
2940 (±350)
296 (±48)
10 (±0.6)
b
11 (±1.4)
19 (±1.2)
77 (±4.2)
na
Trolox
PBN
NXY-059
0.06 (±0.01)
43 (±11)
1.5 (±0.12)
3065 (±480)
437 (±123)
na
a
Values are means of three experiments, standard deviation is given in parentheses (na = not active).
Solubility problems.
b
cially cyclopropyl-nitrones (2d and 2f) where only the
nitrone function is responsible for free radical trapping,
exceeded the efficacy of PBN and NXY-059.
model. Therefore, 2a–f are promising candidates for fur-
ther development.
In comparison with trolox, 2a and 2b were comparable
to trolox in vitro, slightly weaker for OH, but slightly
Acknowledgments
Å
stronger for peroxyl radical trapping (TRAP). The inef-
ficacy of O-protected derivatives 2c–f in OH scavenging
We are grateful to Tam a´ s G a´ ti and B e´ la Heged u} s for
spectroscopic measurements. We also thank Mrs. Edit
B e´ res for skilful technical assistance.
Å
capacity measurements indicates the favoured role of the
free hydroxyl group of trolox in the OH capturing pro-
Å
cess. The comparable activities of 2d–f to trolox in
ONOO test are yet to be explained. Moreover, in the
LPO test, there was a marked difference in efficacy, pro-
ducing 13 and 37 times lower IC50 values for 2a and 2b
than trolox, respectively.
ꢀ
References and notes
1
. (a) Halliwell, B. Drugs 1991, 42, 569; (b) Rice-Evans, C.
A.; Diplock, A. T. Free Rad. Biol. Med. 1993, 15, 77; (c)
Ginsberg, Myron D. Drug News Perspect. 2001, 14, 81; (d)
Ringel, F.; Schmid-Elsaesser, R. Expert Opin. Ther.
Patents 2001, 11, 987; (e) Behl, Ch.; Moosmann, B. Free
Rad. Biol. Med. 2002, 33, 182.
. (a) Janzen, E. G.; Blackburn, B. J. J. Am. Chem. Soc.
1968, 90, 5909; (b) Janzen, E. G.; Blackburn, B. J. J. Am.
Chem. Soc. 1969, 91, 4481; (c) Janzen, E. G. Acc. Chem.
Res. 1971, 4, 3; (d) Maples, K. R.; Ma, F.; Zhang, Y.-K.
Free Rad. Res. 2001, 34, 417; (e) Becker, D. A.; Ley, J. J.;
Echegoyen, L.; Alvarado, R. J. Am. Chem. Soc. 2002, 124,
According to our pMCAo measurements trolox had no
effect. NXY-059 decreased infarct area by 18%, compa-
rable to O-methylated trolox-nitrone 2d causing a 17%
of reduction in infarct. The combined trolox–nitrone
molecules 2a and 2b had slightly superior efficacy when
compared to NXY-059, reducing the infarct area by
2
2
7% and 26%, respectively (Fig. 2).
In conclusion, we synthesised new nitrone derivatives of
trolox (2a–f) that exert good efficacy both in vitro and in
vivo. They are comparable to trolox in scavenging free
radicals and exceed nitrone-type references, for example,
PBN or NXY-059. Compounds 2a–f proved to be signif-
icantly more effective than reference compounds against
lipid peroxidation and in permanent focal ischemia
4
678; (f) Barclay, L. R. C.; Vinqvist, M. R. Free Rad. Biol.
Med. 2000, 28, 1079; (g) Lin, S.; Rhodes, Ph. G.; Lei, M.;
Zhang, F.; Cai, Z. Brain Res. 2004, 1007, 132.
. (a) Floyd, R. A. FASEB J. 1990, 4, 2587; (b) Zhao, Q.;
Pahlmark, K.; Smith, M. L.; Siesj o¨ , B. K. Acta. Physiol.
Scand. 1994, 152, 349.
3
4. (a) Maples, K. R.; Ma, F.; Zhang, Y.-K. Free Rad. Res.
001, 34, 417; (b) Zhao, Z.; Cheng, M.; Maples, K. M.;
2
Ma, J. Y.; Buchan, A. M. Brain Res. 2001, 909, 46; (c)
Green, A. R.; Ashwood, T.; Odergren, T.; Jackson, D. M.
Pharm. Ther. 2003, 100, 195; (d) Marshall, J. W. B.;
Green, A. R.; Ridley, R. M. Brain Res. 2003, 972, 119.
. (a) Vukics, K.; T a´ rk a´ nyi, G.; Dravecz, F.; Fischer, J.
Synth. Commun. 2003, 33, 3419(b) Vukics, K.; Balogh, Gy.
T.; Gere, A.; Stadler, K.; T a´ rk a´ nyi, G.; Fischer, J.
Abstracts of Papers, XVIIth International Symposium on
Medicinal Chemistry, Barcelona, Spain, 1–5 September
5
6
2
002; p 531.
. General procedure for the preparation of trolox nitrones
a–f: The appropriate trolox aldehyde (5 mmol) and N-
2
alkyl hydroxylamine (12.5 mmol) in ethanol (20 mL) were
stirred for 3 h under reflux. The solvent was evaporated in
vacuum, and then methylene chloride and water were
added to the residue. The layers were separated and
the organic phase was washed with water, dried over
Figure 2. Neuroprotective activity of compounds 2a, 2b, 2d and
reference molecules in the pMCAo model in mice, single dose given
30 min after occlusion at 10 mg/kg intraperitoneally, *p < 0.05,
n = 4–10.
2 4
anhydrous Na SO and concentrated under vacuum.

G. T. Balogh et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3012–3015
3015
The products were crystallized from an appropriate solvent,
generally from diethyl ether/n-hexane, after purification by
flash chromatography eluting with CH Cl /MeOH, if
7. Taldolini, B.; Cabrini, L. Mol. Cell. Biochem. 1990, 94, 97–
104.
Å
ꢀ
Å
8. OH, ONOO and ROO scavenging assays were carried
out with our modified procedure of Refs. 7,9,11. The
modified methods were described in Balogh, Gy. T.; Ill e´ s,
J.; Sz e´ kely, Zs.; Forrai, E.; Gere, A. Arch. Biochem.
Biophys. 2003, 410, 76.
9. (a) Hughes, M. N.; Nicklin, H. G. J. Chem. Soc. A 1968,
450–452; (b) Bhat, V. B.; Madyastha, K. M. Biochem.
Biophys. Res. Commun. 2001, 285, 262.
10. NO was generated by thermal decomposition of cupferron
at 40 °C. Increasing concentrations (0.3–100 mM) of
samples in 0.5 mL ethanol were mixed with Griess–Ilosvay
reagent consisting of 0.5 mL of 0.2% N-(1-naphthyl)eth-
ylenediamine in 0.4 M HCl and 0.5 mL of 2% sulfanil-
amide in 4 M HCl and 0.5 mL of 0.2% cupferron in
ethanol. After 30 min of incubation at 40 °C, while the NO
formation reached its maximum, the absorption was
measured at 540 nm.
2
2
necessary. Physical properties of compounds 2a–f are as
1
follows: (2a): Mp: 170–172 °C (diethyl ether). H NMR
(
DMSO-d
6
, 300 MHz, 30 °C): 7.47 (s, 1H), 6.62 (s, 1H),
2
2
1
1
.90–2.76 (m, 1H), 2.56–2.40 (m, 1H), 2.36–2.17 (m, 1H),
.06 (s, 3H), 2.04 (s, 3H), 1.99 (s, 3H), 1.85–1.68 (m, 1H),
.55 (s, 3H), 1.32 (s, 9H). IR (KBr): 1592, 1452, 1234,
ꢀ1
+
165, 1108, 925, 785, 678 cm . MS(EI): m/z 305 (M ).
305.1990. Found:
05.1996. 2b: Mp: 152–154 °C (n-hexane/diethyl ether).
HRMS: Calcd for
C
18
H27NO
3
:
3
1
H NMR (DMSO-d
6
, 500 MHz, 30 °C): 7.46 (s, 1H), 7.02
(
(
(
0
1
1
1
2
s, 1H), 3.90–3.81 (m, 1H), 2.89–2.77 (m, 1H), 2.55–2.45
m, 1H), 2.32–2.20 (m, 1H), 2.07 (s, 3H), 2.04 (s, 3H), 1.99
s, 3H), 1.77–1.65 (m, 1H), 1.55 (s, 3H), 1.14–1.00 (m, 2H),
1
3
6
.67–0.54 (m, 2H). C NMR (DMSO-d , 125 MHz): 4.0,
1.7, 12.6, 20.9, 21.4, 27.9, 43.4, 75.3, 96.0, 117.6, 120.5,
21.3, 122.7, 139.1, 144.0, 145.9. IR (KBr): 2927, 1603,
455, 1326, 1258, 1165, 1089, 921, 816 cm . MS(EI): m/z
ꢀ
1
11. Bartosz, G.; Janaszewska, A.; Ertel, D.; Bartosz, M.
Biochem. Mol. Biol. Int. 1998, 46, 519.
+
89 (M ). HRMS: Calcd for C H NO : 289.1679.
1
7
23
3
Found: 289.1687. 2c: Mp: 98.0–98.5 °C (petroleum ben-
12. (a) Takei, M.; Hiramatsu, M.; Mori, A. Neurochem. Res.
1991, 19, 1199; (b) Hara, H.; Kogure, K.; Kato, H.; Ozaki,
A.; Sukamoto, T. Eur. J. Pharmacol. 1991, 197, 75; (c)
Barta-Szalai, G.; Gere, A.; Tak a´ cs-Nov a´ k, K.; Dom a´ ny,
Gy. Arzneim.-Forsch. 2003, 53, 722.
1
zine). H NMR (CDCl
3
, 500 MHz, 30 °C): 6.69 (s, 1H),
.07–2.95 (m, 1H), 2.63–2.52 (m, 1H), 2.43–2.32 (m, 1H),
3
2
1
2
.32 (s, 3H), 2.13 (s, 3H), 2.03 (s, 3H), 1.95 (s, 3H), 1.96–
.85 (m, 1H), 1.69 (s, 3H), 1.44 (s, 9H). IR (KBr): 2984,
ꢀ
1
934, 1759, 1570, 1465, 1369, 1212, 1077, 927 cm
.
13. Mitochondria were prepared from whole brains of male
LATI Wistar rats. Mitochondrial suspension (0.3 mg
protein), 25 mM Tris–HCl buffer (pH 7.4), and 0.2 mM
ascorbic acid were incubated in the presence or absence of
reference substances and test compounds. Lipid peroxida-
tion was induced by the addition of 0.2 mM FeSO4
solution and the mixture was incubated for 60 min at
37 °C. The reaction was terminated with 20% trichloro-
acetic acid solution, and the samples were centrifuged. The
supernatant was boiled with 0.5% thiobarbituric acid
solution for 20 min. Optical density was determined at
532 nm. The inhibitory effect of the test compounds was
expressed as the inhibition of TBARs formation.
+
MS(FAB): m/z 348 (MH ). HRMS: Calcd for
C H NO : 347.2096. Found: 347.2112. 2d: Mp: 188–
2
0
29
4
1
1
5
2
2
1
2
91 °C (n-hexane/diethyl ether).
3
H NMR (CDCl ,
00 MHz, 50 °C): 6.75 (s, 1H), 3.06–2.96 (m, 1H), 2.63–
.53 (m, 1H), 2.48–2.36 (m, 1H), 2.31 (s, 3H), 2.13 (s, 3H),
.04 (s, 3H), 1.96 (s, 3H), 1.92–1.81 (m, 1H), 1.69 (s, 3H),
.40–1.28 (m, 2H), 0.76–0.62 (m, 2H). IR (KBr): 3076,
ꢀ
1
.
936, 1755, 1592, 1369, 1216, 1193, 1083, 927 cm
+
MS(FAB): m/z 332 (MH ). HRMS: Calcd for
: 331.1786. Found: 331.1789. 2e: Mp: 110–
19 4
C H25NO
1
1
3
2
2
12 °C (petroleum benzine). H NMR (CDCl
3
, 300 MHz,
0 °C): 6.69 (s, 1H), 3.63 (s, 3H), 3.06–2.93 (m, 1H), 2.64–
.49 (m, 1H), 2.45–2.28 (m, 1H), 2.19(s, 3H), 2.12 (s, 3H),
.11 (s, 3H), 1.99–1.83 (m, 1H), 1.68 (s, 3H), 1.44 (s, 9H).
14. (a) Backhauss, C.; Karkoutly, Ch.; Welsch, M.; Kriegl-
stein, J. J. Pharm. Methods 1992, 27, 27; (b) Cramer, W. C.
M.; Toorop, G. P. Gen. Pharm. 1998, 30, 195.
IR (KBr): 2980, 1576, 1453, 1357, 1256, 1093, 1002, 884,
7
ꢀ
1
+
34 cm . MS(FAB): m/z 320 (MH ). HRMS: Calcd for
15. MCA occlusion was performed by electrical coagulation
on the right MCA of male NMRI mice. Test compounds
were administered i.p. at a dose of 10 mg/kg 30 min after
occlusion. Animals were maintained for 48 h under
normal conditions, and then their brains were perfused
transcardially with an injection of 2,3,5-triphenyltetrazo-
lium chloride (TTC). Animals were decapitated and the
infarct area of the brain surface was determined plani-
metrically by means of an image analyzer system (Digicell
5.0 ASK).
C H NO : 319.2145. Found: 319.2162. 2f: Mp: 90–91 °C
1
9
29
3
1
n-hexane). H NMR (DMSO-d
(
(
6
, 500 MHz, 30 °C): 7.08
s, 1H), 3.92–3.84 (m, 1H), 3.53 (s, 3H), 2.88–2.79 (m, 1H),
2
3
1
1
.57–2.47 (m, 1H), 2.33–2.22 (m, 1H), 2.10 (s, 3H), 2.05 (s,
H), 2.04 (s, 3H), 1.79–1.69 (m, 1H), 1.57 (s, 3H), 1.14–
.02 (m, 2H), 0.67–0.56 (m, 2H). IR (KBr): 2935, 1582,
ꢀ
1
459, 1403, 1256, 1200, 1102, 950, 740 cm . MS(EI): m/z
+
03 (M ). HRMS: Calcd for C H NO : 303.1834.
3
Found: 303.1841.
1
8
25
3