Novel lipoic acid analogues that inhibit nitric oxide synthase.

Article abstract of DOI:10.1016/S0960-894X(02)00216-0

The synthesis and biological activity of novel lipoic acid analogues are reported. Lipoic acid and structural homologues coupled to arylthiophene amidine via carboxamide linkers are metabolic antioxidants capable of protecting neuronal cells against glutamate cytotoxicity, preventing loss of intracellular glutathione, and inhibit nitric oxide synthase.

Full text of DOI:10.1016/S0960-894X(02)00216-0

Bioorganic & Medicinal Chemistry Letters 12 (2002) 1439–1442
Novel Lipoic Acid Analogues that Inhibit Nitric Oxide Synthase
Jeremiah J. Harnett,a,* Michel Auguet,b Isabelle Viossat,b Christine Dolo,a
Dennis Bigga and Pierre-E. Chabrierb
aDepartment of Medicinal Chemistry, Beaufour-Ipsen Research Laboratories, Institute Henri Beaufour,
5, Avenue du Canada, 91966 Les Ulis Cedex, France
bDepartment of Biology, Beaufour-Ipsen Research Laboratories, Institute Henri Beaufour,
5, Avenue du Canada, 91966 Les Ulis Cedex, France
Received 27 December 2001; revised 8 March 2002; accepted 3 April 2002
Abstract—The synthesis and biological activity of novel lipoic acid analogues are reported. Lipoic acid and structural homologues
coupled to arylthiophene amidine via carboxamide linkers are metabolic antioxidants capable of protecting neuronal cells against
glutamate cytotoxicity, preventing loss of intracellular glutathione, and inhibit nitric oxide synthase. # 2002 Published by Elsevier
Science Ltd.
Alpha-lipoic acid (LA) has been termed the ‘ideal’ anti-
oxidant,1 a readily absorbed and bioavailable com-
pound capable of scavenging a number of free radicals.
LA is readily absorbed from the diet, transported, taken
Figure 1. Lipoic acid is reduced to dihydrolipoic acid in vivo.
up by cells and reduced to dihydrolipoic acid (DHLA)
in various tissues including brain and is considered a
metabolic antioxidant (Fig. 1).1,2
l-arginine and di-tert-butyl-hydroxybenzoic acid pro-
duced a synergistic protective effect in the reduction of
neuronal damage during ischemia-reperfusion.8 More
recently, we demonstrated that the co-administration of
LA 2 and NOS inhibitors, notably compound 3 pro-
vided a greater protective effect against 1-methyl-
4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) induced
depletion of brain dopamine in mice9 than the expected
sum of the compounds administered individually.
MPTP is a neurotoxin which induces selective destruc-
tion of dopaminergic neurons in the substantia nigra
pars compacta and leads to a dramatic decrease of
dopamine.10
The antioxidant activity of LA is attributed to its capa-
city to regenerate endogenous antioxidants for example
vitamin E, C and notably glutathione (GSH), and thus
maintains an intracellular antioxidant balance.3,4 GSH
is the most important thiol antioxidant but it cannot be
directly administrated whereas LA can.1,3 Since LA
raises intracellular GSH levels, it has been used for
treating diseases in which oxidative stress plays a major
role.
The enzyme nitric oxide synthase (NOS) plays various
roles both in normal and pathological processes.5 Over-
production of nitric oxide (NO) is implicated in different
inflammatory and neurologic disorders and represents a
potential therapeutic target.6,7
Based on these in vivo results, the neurotoxicity of
NO,11ꢀ13 and the observed decrease in intracellular
GSH in certain neurodegenerative models14 we have
synthesised new lipoic acid derivatives that inhibit NOS
and are metabolic antioxidants.
We have previously reported that co-administration of a
NOS inhibitor and an antioxidant, notably NG-nitro-
In this report, the design, synthesis and a preliminary
examination of the in vitro activity of these novel lipoic
acid analogues will be discussed.
*Corresponding author. Fax: +33-169-07-3802; e-mail: jeremiah.
harnett@beaufour-ipsen.com
0960-894X/02/$ - see front matter # 2002 Published by Elsevier Science Ltd.
PII: S0960-894X(02)00216-0

1440
J. J. Harnett et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1439–1442
Figure 2. Design of lipoic acid analogues possessing nNOS activity.
Chemistry
The concept is to covalently link lipoic acid and struc-
tural homologues of lipoic acid 2 to a selective neuronal
NOS (nNOS) inhibitor15 3 via a chemical linker as
shown in Figure 2.
Scheme 1. (i) HOBT, Et3N, 1-ethyl-3-[3-(dimethylamino)propyl]-
carbodiimide hydrochloride (EDCl), DCM; (ii) HCl, ethanol/diethyl-
ether/acetone/dichloromethane; (iii) iPrOH, S-methyl-2-thiophene-
imidate 7, 60 ꢁC.
Lipoic acid (2, m=3) and trisnorlipoic acid16 (2, m=0)
were coupled to the corresponding N-Boc-protected
phenylamines 4a–e (n=0–2) using peptide coupling
conditions as shown in Scheme 1, followed by removal
of the Boc protecting group of 5a–e under acidic condi-
tions to afford intermediates 6a–e. The anilines 6a–e
were then coupled with S-methyl-2-thiophene-imidate17
7 to give the desired lipoic acid analogues18 1a–e.
Removal of the tert-butyl carbamate protecting groups
of 5 under acidic conditions led to moderate yields due
to low solubility of 5 in organic solvents as well as, in
some cases, extensive polymerisation due to the oxida-
tion of the 1,2-thiolane moiety producing disulphide
polymers.19 This was overcome by coupling 2 in the
final step, for example in the synthesis of compound 1b
as shown in Scheme 2. The amine function of p-nitro-
benzylamine 8 was protected as a tert-butyl carbamate,
followed by catalytic hydrogenation of the nitro moiety
of 9 using 10% Pd/C to the required aniline. The aniline
10 was then reacted with S-methyl-2-thiophene-imidate
7, followed by acidic deprotection of 11 to afford the
benzylamine 12, which was coupled to lipoic acid to
provide the desired lipoic acid analogue 1b. The analo-
gues described in Schemes 1 and 2 were prepared in
racemic form.
Scheme 2. (i) BOC2O, Et3N, DCM, 70%; (ii) 10% Pd/C, H2, DCM/
THF, 68%; (iii) iPrOH, 7, 60 ꢁC, 99%; (iv) HCl, EtOAc, 92%;
(v) HOBT, Et3N, EDCl, DCM, 40%.
death called oxidative glutamate toxicity.22 Toxicity is
due to the inhibition of cystine uptake by the glutamate
which leads to GSH depletion and cell death. The pro-
tective effects of the compounds against glutamate toxi-
city is expressed as EC50, calculated relative to the
viability of cells which have not been subjected to the
action of glutamate, considered as 100% viability.23
Compounds 1a–e (Table 1) are comparable and in some
cases as active as LA 2 in protecting HT22 cells from
oxidative stress, the parent compound 3 being inactive.
Glutamate challenge causes a marked decrease in the
intracellular concentration of GSH. Thus using the
enzymatic glutathione assay, based on the formation of
5-thio-2-nitrobenzoic acid as a result of the oxidation of
Pharmacology
Results and discussion
The compounds were evaluated in vitro for their capa-
city to inhibit nNOS. Inhibition was determined by
measuring their effect on the conversion by NOS of
[3H]l-arginine into [3H]l-citrulline.20 As can be seen
from Table 1 the lipoic acid derivatives 1a–e have
retained nNOS activity and have comparable IC50
values to that of the parent compound 3 while LA 2
itself is inactive.
Table 1. nNOS inhibition and glutamate induced cytotoxicity results
for compounds 1a–e, 2, 3
Compd
nNOS
inhibition
IC50 (mM)a
Glutamate toxicity
HT22 cells
Prevention of
GSH depletion
IC50 (mM)a
EC50 (mM)a
3
2
4.00
na
na
19.6
11.1
6.3
11.7
9.6
na
121.6
46.9
64.8
37.1
11.9
12.3
The neuroprotective effect was measured by their ability
to protect mouse hippocampal neuronal HT22 cell
lines21 from oxidative stress caused by glutamate. The
cells do not express NOS and lack ionotropic glutamate
receptors; thus, when exposed to high concentrations of
glutamate the cells die via a form of programmed cell
(1a) m=3, n=0
(1b) m=3, n=1
(1c) m=3, n=2
(1d) m=0, n=0
(1e) m=0, n=2
3.50
2.30
1.04
4.40
1.15
1.8
ana, not active, up to 100 mM.

J. J. Harnett et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1439–1442
1441
GSH by 5,50-dithiobis(2-nitrobenzoic acid), total GSH
was assayed.24 Examination of Table 1 reveals that
there is a trend towards the protection of HT22 cells
with restoration of intracellular GSH levels (R2=0.6).
Especially interesting is the trisnorlipoic acid derivative,
1e, which is 10 times more active in protecting neuronal
cells against oxidative stress and 10 times more active in
maintaining GSH levels than lipoic acid.
15. Harnett, J. J.; Auguet, M.; Chabrier, P-E. WO 00/59899;
Chem. Abstr. 2000, 133, 281650.
16. Chen, Y.-S.; Lawton, R. G. Tetrahedron. Lett. 1997, 38, 5785.
17. Bercot, V. Ann. Chim. 1962, 7, 303.
18. Typical experimental procedure, Synthesis of 1a: N-{4-
[(1,1-dimethylethoxy)carbonylamino]phenyl}-1,2-dithiolane-3-
pentanamide (5a). N-Boc-1,4-Phenylenediamine 4a (1.84 g,
8.81 mmol), triethylamine (1.6 mL), hydroxybenzotriazole
(1.70 g, 12.6 mmol), 1-ethyl-3-[3-(dimethylamino)propyl]-
carbodiimide hydrochloride (EDCl), (4.82 g, 25.2 mmol) and
additional triethylamine (1.6 mL) were added successively to a
solution of lipoic acid (2.00 g, 9.694 mmol) in dichloromethane
(40 mL). The reaction mixture was stirred overnight at 25 ꢁC
and diluted with water (100 mL). The product was extracted
with dichloromethane (3Â100 mL), dried over magnesium
sulphate, filtered and concentrated under reduced pressure.
The reddish brown powder obtained was suspended in diethyl
ether (100 mL), filtered and rinsed with the same volume of
ether to afford a salmon pink powder in 80.5% yield. Melting
point: 190.0–195.0 ꢁC. 1H NMR(DMSO- d6, 400 MHz, d):
1.39 (m, 2H, CH2); 1.57 (s, 9H, BOC); 1.61 (m, 4H, CH2); 1.88
(m, 1H, CH2); 2.27 (m, 2H, CH2); 2.50 (m, 1H, CH2); 3.15 (m,
2H, CH2); 3.63 (m, 1H, –S–CH–); 7.34 (d, 2H, arom,
J=8.70 Hz); 7.45 (d, 2H, arom, J=9.00 Hz); 9.24 (s, 1H,
CONH); 9.77 (s, 1H, CONH-BOC). 13C NMR(DMSO- d6,
100 MHz, d): 35.00; 55.70; 101.52; 108.09; 123.78; 129.09;
130.42; 146.49; 148.15; 162.83; 163.91; 169.56.
While the neuroprotective effect of the coadministration
of 2 and 3 was shown in an in vivo model9 it now
remains to demonstrate the in vivo biological activity of
these new dual activity compounds 1.
In conclusion, in this preliminary study we have
designed and synthesised molecules with dual activity,
capable of inhibiting nNOS in vitro and acting as
metabolic antioxidants. Some of these compounds are
more effective than lipoic acid in protecting cells against
glutamate cytotoxicity and in preventing glutamate
induced loss of cellular glutathione.
Further studies are continuing to improve on nNOS
inhibition and using other structural homologues of
lipoic acid, for example, bisnor and tetranorlipoic
acids.25
N-(4-Aminophenyl)-1,2-dithiolane-3-pentanamide (6a).
A
stream of hydrogen chloride gas was slowly bubbled through a
solution of intermediate 5a (6.5 g; 16.4 mmol) in a mixture of
ether/ethanol/acetone/dichloromethane (1/1/1/1, 200 mL) at
0 ꢁC for 15 min. The temperature was allowed to rise to ambi-
ent temperature and left to stir overnight. A stream of argon
was passed through the reaction mixture and the solvents
evaporated under reduced pressure. The residue was dissolved
in dichloromethane (200 mL), washed with a cold saturated
solution of sodium bicarbonate (3Â100 mL) followed by brine
(100 mL). The organic phase was dried over magnesium sul-
phate, filtered and concentrated under reduced pressure. Pur-
ification was carried out on a silica-gel column (mobile phase:
50% ethyl acetate in heptane followed by 5% ethanol in di-
chloromethane ) to afford a beige coloured solid in 29% yield.
Acknowledgements
The authors would like to thank Veronique Roubert,
Jose Camara and teams for their helpful discussions and
advice.
References and Notes
1. Packer, L.; Tritschler, H. J.; Wessel, K. Free. Rad. Biol.
1997, 22, 359.
Melting point: 55.0–60.0 ꢁC. NMR H (DMSO-d6, 400 MHz,
1
d): 1.38 (m, 2H, CH2); 1.57 (m, 4H, CH2); 1.87 (m, 1H, CH2);
2.22 (m, 2H, CH2); 2.40 (m, 1H, CH2); 3.18 (m, 2H, CH2); 3.62
(m, 1H, –S–CH–); 4.78 (s, 2H, NH2); 6.48 (d, 2H, arom, J=8.64
Hz); 7.20 (d, 2H, arom, J=8.64 Hz); 9.39 (s, 1H, CONH). 13C
NMR(DMSO- d6, 100 MHz, d): 27.34; 30.64; 36.43; 38.29;
40.37; 58.40; 116.03; 123.14; 130.83; 146.82; 172.27.
2. Packer, L.; Witt, E. H.; Tritschler, H. J. Free. Rad. Biol.
Med. 1995, 19, 227.
3. Suzuki, Y. J.; Tsuchiya, M.; Packer, L. Free. Rad. Res.
Commun. 1991, 15, 255.
4. Tirosh, O.; Sen, C. K.; Roy, S.; Kobayashi, M. S.; Packer,
L. Free. Rad. Biol. 1999, 26, 1418. Murphy, T. H.; Schnaar,
R. L.; Coyle, J. T. FASEB 1990, 4, 1624. Davis, J. B.; Maher,
P. Brain Res. 1994, 652, 169.
5. Kerwin, J. F.; Lancaster, J. R.; Feldmann, P. L. J. Med.
Chem. 1995, 38, 4342.
6. Billiar, T. R. Ann. Surg. 1995, 221, 339.
7. Kerwin, J. F.; Heller, M.; Feldmann, P. L. Med. Res. Rev.
1994, 14, 23.
8. Spinnewyn, B.; Cornet, S.; Auguet, M.; Chabrier, P.-E.
J. Cereb. Blood Flow Metab. 1999, 19, 139.
9. Auguet, M.; Harnett, J. J.; Chabrier, P.-E. WO 00/59448;
Chem. Abstr. 2000, 133, 276363.
10. Burns, R. S.; Chiueh, C. C.; Markey, S. P.; Ebert, M. H.;
Jacobowitz, D. M.; Kopin, I. J. Proc. Natl. Acad. Sci. U.S.A.
1983, 80, 4546.
11. Boje, K. M.; Arora, P. K. Brain. Res. 1992, 587, 250.
12. Youdim, M. B. H. Adv. Res. Neurodeg. 1993, 1, 67.
13. Youdim, M. B. H.; Lavie, L. Life. Sci. 1994, 55, 2077.
14. Kojima, S.; Matsuki, O.; Nomura, T.; Yamaoka, K.;
Takahashi, M.; Niki, E. Free. Rad. Biol. Med. 1999, 26, 388
and references cited within.
N-{4-[[(2-thienyl)(imino)methyl]amino]phenyl}-1,2-dithiolane-
3-pentanamide hydrochloride (1a). Intermediate 6a (0.703 g;
2.37 mmol) was dissolved in 2-propanol (15 mL) and
S-methyl-2-thiophene thiocarboximide hydroiodide 7 (1.014 g,
3.56 mmol) was added in one portion. After heating at 60 ꢁC
for 15 h, the reaction mixture was concentrated to dryness
under reduced pressure. The residue was taken up in di-
chloromethane (100 mL), washed with a saturated solution of
sodium bicarbonate (3Â100 mL) followed by brine
(3Â100 mL). The organic phase was dried over magnesium
sulphate, filtered and evaporated under reduced pressure. The
free base was dissolved in dichloromethane (30 mL) and the
solution cooled to 0 ꢁC before the dropwise addition of a 1 N
hydrochloric acid solution in anhydrous diethyl ether
(6.3 mL). After stirring for 15 h at 25 ꢁC, the solid formed was
filtered and rinsed with diethyl ether and dried to afford a light
beige solid in 77% yield. Melting point: 258.7–258.9 ꢁC. 1H
NMR(DMSO- d6, 400 MHz, d): 1.43 (m, 2H, CH2); 1.62 (m,
4H, CH2); 1.88 (m, 1H, CH2); 2.38 (m, 3H, CH2); 3.18 (m, 2H,
CH2); 3.63 (m, 1H,–S–CH–); 7.20–8.20 (m, 7H, arom.); 8.79

1442
J. J. Harnett et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1439–1442
(broad s, 1H, NH+); 9.78 (broad s, 1H, NH+); 10.36 (s, 1H,
CONH); 11.49 (broad s, 1H, NH+). NMR 13C (DMSO-d6,
100 MHz, d): 25.10; 28.54; 34.37; 36.44; 38.33; 56.34; 120.18;
126.58; 128.81; 129.08; 129.29; 134.63; 134.95; 139.68; 156.77;
22. Kumiko, I.; Schubert, D.; Sagara, Y. Free. Rad. Biol.
Med. 2001, 30, 433.
23. The HT22 cells were cultured in Dulbecco’s modified
Eagle’s medium (DMEM) with 10% of foetal calf serum at
37 ꢁC and plated in 96-well plates 24 h prior to treatment. 24 h
after glutamate and compound treatment cell viability was
determined by the WST-1 method (Roche, catalogue number
1644807).
24. Tietze, F. Anal. Biochem. 1969, 27, 502. Owens, C. W. I.;
Belcher, R. V. J. Biochem. 1965, 94, 705.
25. Suzuki, Y. J.; Tsuchiya, M.; Packer, L. Free. Rad. Res.
Commun. 1993, 18, 115.
171.71. IR:nC¼N (amidine):1580 cm1; nC¼O (amide):
1659 cmꢀ1
.
.
Elemental analysis for C19H23N3OS3 0.25H2-
.
O HCl: Theoretical: C50.59%, N9.32%, S21.32%, H5.59%.
Found: C50.91%, N9.15%, S21.12%, H5.56%.
19. Thomas, R. C.; Reed, L. J. Chem. Soc. 1956, 78, 6148.
20. Bredt, D. S.; Snyder, S. H. Proc. Natl. Acad. Sci. U.S.A.
1990, 87, 682.
21. Davis, J. B.; Maher, P. Brain. Res. 1994, 652, 169;
Murphy, T. H.; Neuron 1989, 2, 1547.