An Efficient Method for the Synthesis of Laquinimod

Article abstract of DOI:10.1002/jhet.2324

Laquinimod, 5-chloro-1,2-dihydro-N-ethyl-4-hydroxy-1-methyl-2-oxo-N- phenyl-3-quinoline carboxamide, is an oral drug in clinical trials for the treatment of multiple sclerosis. An efficient synthetic method for laquinimod from 2-amino-6-chlorobenzoic acid via four steps was established. The overall yield of laquinimod is up to 82% as compared with 70% reported in literature. It has also been demonstrated that green reagent dimethyl carbonate is not suitable for the N-methylation of 5-chloroisatoic anhydride owing to the ring-cleavage reaction induced by the generated methanol. The ring-cleavage by-products were isolated and characterized by ¹H-NMR and ¹³C-NMR. In addition, in the study of laquinimod derivatives, we found that 5-chloro-1,2-dihydro-N-ethyl-4-hydroxy-1-methyl-2-oxo-N-phenyl-3-quinoline carboxamide (laquinimod) was obtained in much higher yield than 7-chloro-1,2-dihydro-N-ethyl-4-hydroxy-1-methyl-2-oxo-N-phenyl-3-quinoline carboxamide under the same reaction conditions, and it is possibly attributed to a neighboring group effect.

Full text of DOI:10.1002/jhet.2324

Month 2015
An Efficient Method for the Synthesis of Laquinimod
Yanyan Huang, Ying Feng, Wensheng Gao, Chao Zhang, and Ligong Chen *
a
a
a
a
a,b
a
School of Chemical Engineering and Technology, Tianjin University, Tianjin, People’s Republic of China
b
Collaborative Innovation Center of Chemical Science and Engineering (Tianjin), Tianjin, People’s Republic of China
*
E-mail: lgchen@tju.edu.cn
Received June 4, 2014
DOI 10.1002/jhet.2324
Published online 00 Month 2015 in Wiley Online Library (wileyonlinelibrary.com).
Laquinimod, 5-chloro-1,2-dihydro-N-ethyl-4-hydroxy-1-methyl-2-oxo-N- phenyl-3-quinoline carboxamide,
is an oral drug in clinical trials for the treatment of multiple sclerosis. An efficient synthetic method for
laquinimod from 2-amino-6-chlorobenzoic acid via four steps was established. The overall yield of laquinimod
is up to 82% as compared with 70% reported in literature. It has also been demonstrated that green reagent
dimethyl carbonate is not suitable for the N-methylation of 5-chloroisatoic anhydride owing to the ring-cleavage
reaction induced by the generated methanol. The ring-cleavage by-products were isolated and characterized by
1
13
H-NMR and C-NMR. In addition, in the study of laquinimod derivatives, we found that 5-chloro-1,
2
-dihydro-N-ethyl-4-hydroxy-1-methyl-2-oxo-N-phenyl-3-quinoline carboxamide (laquinimod) was obtained in
much higher yield than 7-chloro-1,2-dihydro-N-ethyl-4-hydroxy-1-methyl-2-oxo-N-phenyl-3-quinoline carbox-
amide under the same reaction conditions, and it is possibly attributed to a neighboring group effect.
J. Heterocyclic Chem., 00, 00 (2015).
INTRODUCTION
intermediate 2a formed in the first step may react with
ethyl chloroformate before the addition of acetyl chloride,
which leads to the generation of by-product 2b. Therefore,
this synthetic route was modified in this paper; a more
efficient synthesis of laquinimod was established. Under
optimized reaction conditions, laquinimod was obtained
in 82% overall yield.
Laquinimod, 5-chloro-1,2-dihydro-N-ethyl-4-hydroxy-1-
methyl-2-oxo-N-phenyl-3-quinoline carboxamide (Fig. 1),
also known as ABR-215062, is a novel oral immunomodula-
tor for the treatment of multiple sclerosis. Compared with
its lead compound Linomide (roquinimex), which had
been terminated in the phase III trial because of infrequent
and important toxicities [1,2]; laquinimod demonstrated
high potency and superior toxicological profile [3–6].
Now, laquinimod is just under phase III clinical trial, so
development of an efficient synthetic method for
laquinimod is urgent.
Laquinimod was first synthesized at laboratory scale via
a three-step procedure [7]. However, this process suffered
from several major disadvantages from an industrial per-
spective, including high cost, tedious workup procedures,
low yield, and the use of phosgene. In 2007, another
synthetic route of laquinimod was reported by Johan
Wennerberg (Scheme 1) [7].
RESULTS AND DISCUSSION
As shown in Scheme 3, in order to avoid the use of poison-
ous ethyl chloroformate, triphosgene (bis(trichloromethyl)
carbonate or BTC) was chosen as a carbonyl equivalent in
the present study. As is well known, BTC, which is a stable
solid and also an active acylation agent, can release phosgene
quantitatively in the presence of nucleophiles [8–10]. Thus,
the cyclization reaction of 2-amino-6-chlorobenzoic acid with
BTC proceeds smoothly in tetrahydrofuran (THF) at room
temperature to afford compound 2 in 96% yield. A large
amount of hydrogen chloride was generated, so an effective
absorption apparatus of hydrogen chloride was required.
Moreover, the posttreatment of the cyclization was quite
simple, quenching the reaction and washing the obtained
precipitates just by water. Fortunately, no impurities such as
2a and 2b were detected in the reaction mixture. Compared
with the cyclization described in Scheme 1, shorter reaction
Although this route was more feasible for large-scale
production than the previous method, the usage of
unfriendly ethyl chloroformate and expensive methyl
iodide makes it industrially unattractive. Furthermore, the
undesirable side reaction might happen owing to the cas-
cade reaction nature of the first step, resulting in the hard
purification of compound 2. As described in Scheme 2,
©
2015 HeteroCorporation

Y. Huang, Y. Feng, W. Gao, C. Zhang, and L. Chen
Vol 000
À
were not encouraging. It was found that CH O was gen-
3
erated during the process [11]. As a strong nucleophile, it
possibly attacks the carbonyl carbon of 5-chloroisatoic
anhydride and leads to ring-cleavage reaction (Scheme 4).
As a consequence, by-products 3a and 3b were isolated
Figure 1. Chemical structure of laquinimod.
1
13
and confirmed by H-NMR and C-NMR. Besides, 3a
or 3b was hydrolyzed at reflux temperature (150°C) in
the presence of water to yield the corresponding
aminobenzoic acid. Owing to so many side reactions, the
yield of the target compound was really poor. No choice,
methyl iodide was employed as a methylating agent in this
reaction to afford compound 3 in 87% yield.
The enolate of dimethyl malonate was formed by the
treatment of dimethyl malonate with sodium methoxide
in Scheme 1. However, sodium methoxide and generated
methanol probably induce the ring-cleavage side reaction
of compound 3 according to the mechanism shown in
Scheme 4. Thus, we employed sodium hydride instead of
sodium methoxide to generate the enolate of dimethyl
time, milder reaction conditions, better yield, and easier post-
treatment made it more competitive in industrial production.
As we all know, methyl iodide and dimethyl sulfate
(
DMS) are conventional methylating agents. However,
the genotoxicity of DMS and its derived alkyl sulfates pre-
vent it from being used in pharmaceutical production. With
consideration of the cost and toxicity of methyl iodide, we
tried to use green reagent dimethyl carbonate (DMC) as the
N-methylating agent in the second step. In order to increase
the activity of compound 2 and enhance the N-methylation,
the reaction with DMC as the N-methylating agent was
conducted in the presence of catalytic amounts of base at
high temperature (over 120°C). Unfortunately, the results
Scheme 1. The synthetic route of laquinimod reported by Johan Wennerberg.
Scheme 2. The generation of by-product 2b.
Scheme 3. The efficient synthetic route of laquinimod.
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

Month 2015
An Efficient Method for the Synthesis of Laquinimod
Scheme 4. The mechanism of ring-cleavage reaction.
malonate, which condensed with compound 3 to yield
the desired compound 4. As an inorganic alkaline,
sodium hydride can acquire the proton of dimethyl
malonate at room temperature quickly and release
hydrogen that will not induce side reactions. In fact,
the condensation of compound 3 and dimethyl malonate
consists of four elementary reactions, including nucleo-
philic substitution, nucleophilic substitution, hydrolysis,
and decarboxylation. So, the choice of base is very
important to this condensation. Under the optimized
conditions, compound 4 was finally gained in 97% yield
and 98.9% purity.
CONCLUSION
In conclusion, we have established an efficient synthetic
method for laquinimod from 2-amino-6-chlorobenzoic acid
via four steps. Laquinimod was finally obtained in 82%
total yield and 99.5% purity. It was found that the ring-
cleavage side reaction of compound 2 was induced by the
generated methanol. The ring-cleavage by-products were
1
13
isolated and characterized by H-NMR and C-NMR.
Moreover, in the study of laquinimod derivatives, we found
that the N-acylation of compound 4 with N-ethylaniline
exhibited higher reactivity than that of compound 5, and it
is probably caused by the neighboring group effect.
The last step was the N-acylation of N-ethylaniline with
ester 4 to the amide 1. Esters are normally taken as weak
acylating agents. An equilibrium exists between compound
EXPERIMENTAL
4
and compound 1 [12]. The removal of formed methanol
Reagents and solvents were obtained from commercial sup-
pliers. All reactions were monitored by thin-layer chromatogra-
phy using commercial silica gel plates. The purity of products
was detected by HPLC on Agilent 1100 series. Melting points
were observed on YRT-3 Melting Point Tester and uncorrected.
during the reaction will force the equilibrium shift to the
right and lead to high yield of compound 1. However, in
the study of laquinimod derivatives, we accidentally found
that the N-acylation of compound 5 with N-ethylaniline
1
13
(Scheme 4) proceeded much slower than that of compound
H-NMR and C-NMR spectra were recorded on Bruker
AVANCE III 600 MHz. High-resolution mass spectrometry
HRMS) was measured on MicrOTOF-Q II.
-Chloroisatoic anhydride (2). 2-Amino-6-chlorobenzoic
acid (10.00 g, 58.5 mmol) was suspended in THF (60 mL) at
°C. Triphosgene (6.93g, 23.4mmol) was added, and the mixture
4
at the same reaction conditions. The structures of com-
(
pounds 4 and 5 are similar except the position of the
substituted chlorine, but the conversion of compound 4
was obviously higher than that of compound 5 (Table 1).
As a result, the yield of compound 1 was also higher than
that of compound 6 (Table 1). The different behaviors
may be attributed to the neighboring group effect. The sit-
uation of chlorine at the C5 position of the quinoline ring is
crowded to the hydroxyl group, which disturbs the conju-
gated system, resulting in the increase of the positive
charge of the carbonyl group. Nevertheless, when chlorine
substituted at the C7 position of the quinoline ring, far
away from the hydroxyl group, it does not destroy the pla-
narity of compound 5 (Scheme 5), and the positive charge
of the carbonyl group is dispersed on the whole conjugated
system. Hence, the activity of compound 5 is much lower
than that of compound 4.
5
0
was stirred for 3 h at room temperature. Water (50 mL) was
added. The precipitate was collected by filtration, washed with
water (2 × 20 mL), and dried at 50°C under vacuum over P O5
2
to give product 2 as a white solid (10.97 g, yield 97% ). mp
1
271.9–272.5°C ([5] 272.5–272.9°C); H-NMR [hexadeuterated
6
dimethyl sulfoxide (DMSO-d ), 600 MHz] δ: 7.10 (d, J = 8.4 Hz,
1
H), 7.31 (d, J = 7.8 Hz, 1H), 7.66 (t, J = 8.1 Hz, 1H), 11.86
1
3
(
s, 1H). C-NMR (DMSO-d , 150 MHz) δ: 108.00, 114.50,
6
1
25.64, 134.90, 136.49, 143.85, 146.61, 156.43.
-Chloro-N-methylisatoic anhydride (3).
5
Compound 2
(
(
10.00 g, 52.0 mmol) was dissolved in dimethylformamide
DMF; 60 mL). Potassium carbonate (9.32 g, 67.5 mmol) and
iodomethane (9.54 g, 67.5 mmol) were added, and the mixture
was stirred for 5 h at room temperature. Then, 5% hydrochloric
Table 1
Scheme 5. Synthesis of compound 5.
Reaction of esters with N-ethylaniline.
Ester
Product
Conversion
Yield
4
5
1
6
99%
60%
97%
56%
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

Y. Huang, Y. Feng, W. Gao, C. Zhang, and L. Chen
Vol 000
acid was added dropwise until no gas was emitted. The mixture
was stirred for an additional 10min. The precipitate was filtered,
washed with water (2× 20mL), and dried to constant weight at
Methyl 7-chloro-1,2-dihydro-4-hydroxy-1-methyl-2-oxo-3-
quinolinecarboxylate (5). Compound was similarly
synthesized as compound 4 in total yield of 76%. mp 201.9–202.5°C.
5
1
5
0°C under vacuum over P
O
2 5
to give the title compound 3
3
H-NMR (CDCl , 600 MHz) δ: 3.66 (s, 3H), 4.05 (s, 3H), 7.23
(
9.55 g, yield 87% ) as a white solid. mp 217.8–218.5°C ([5]
(dd, J = 12.6 and 1.8 Hz, 1H), 7.32 (d, 1.8 Hz, 1H), 8.12
1
13
2
7
d
1
17.9–218.2°C). H-NMR (DMSO-d , 600MHz) δ: 3.46 (s, 3H),
(d, J = 13.2 Hz, 1H), 14.11 (s, 1H). C-NMR (CDCl , 150 MHz)
6
3
1
3
.41–7.43 (m, 2H), 7.78 (t, J = 8.4 Hz, 1H). C-NMR (DMSO-
δ: 172.73, 170.97, 159.24, 141.90, 140.84, 126.99, 122.41,
6
, 150 MHz) δ: 32.37, 109.33, 114.07, 125.99, 135.37, 136.54,
114.09, 113.14, 97.66, 53.04, 29.26. HRMS [electrospray
+
44.56, 147.38, 155.46.
A mixture of compound 2 (10.00g, 52.0mmol), sodium hydride
60% in mineral oil, 2.70g, 67.6 mmol), DMC (9.36 g,
04.0 mmol), and DMF (30 mL) was stirred and heated to reflux
ionization (ESI)], calcd for
90.0118; found: 290.0190.
7-Chloro-1,2-dihydro-N-ethyl-4-hydroxy-1-methyl-2-oxo-
N-phenyl-3-quinoline carboxamide (6). A mixture of
C
12
H
10ClNO
4
[M+ Na] m/z:
2
(
1
for 5 h under N atmosphere. After that, the solvent was evaporated
compound 5 (10.00 g, 37.5 mmol) and N-ethylaniline (10 mL,
75.0 mmol) in heptane (200 mL) was heated to reflux. The
methanol formed during the reaction was distilled off together
with some heptane (120 mL) for 6 h. After cooling to room
temperature, the precipitate was filtered. Sodium hydroxide
solution (5%; 50 mL) was added to the filtrate and stirred for
15 min. After filtering, the filtrate was acidified by
hydrochloric acid. The mixture was centrifuged, and the
residue was dried at 50°C under vacuum to give the crude title
2
under reduced pressure, and the title compound was purified by
silica gel column (petroleum ether: ethyl acetate (PE:EA)= 2.5:1).
Besides, 3a and 3b were also isolated during the process.
Methyl 2-amino-6-chlorobenzoate (3a).
Yellow oil [13].
1
H-NMR (CD
3
OD, 600 MHz) δ: 3.86 (s, 3H), 6.63–6.65 (m,
1
3
2
H), 7.04 (t, J = 8.1 Hz, 1H). C-NMR (CD OD, 150 MHz) δ:
3
5
2.81, 116.14, 116.38, 119.58, 133.15, 134.32, 150.50, 169.12.
Methyl 2-chloro-6-(methylamino)benzoate (3b).
Yellow
1
compound (5.34 g, yield 56%) as a white solid (assay 97.7%).
oil [14]. H-NMR (CDCl , 600 MHz) δ: 2.81 (s, 3H), 3.90 (s,
3
1
mp 143.8–146.5°C. H-NMR (CDCl , 600 MHz) δ: 1.22 (t,
3
7
5
H), 6.52 (d, J = 8.4 Hz, 1H), 6.67 (dd, J = 7.8 and 0.6 Hz, 1H),
.16 (t, J = 8.1Hz, 1H). C-NMR (CDCl , 150MHz) δ: 30.00,
3
1.90, 109.08, 114.01, 117.73, 132.39, 133.92, 150.48, 168.00.
3
1
3
J = 7.2 Hz, 3H), 3.20 (s, 3H), 3.99 (q, J = 7.2 Hz, 2H), 7.12–
1
3
7
.27 (m, 7H), 8.00 (d, J = 8.4 Hz, 1H), 12.48 (br, 1H). C-
NMR (CDCl , 150 MHz) δ: 12.90, 28.93, 45.55, 106.07,
3
Methyl 5-chloro-1,2-dihydro-4-hydroxy-1-methyl-2-oxo-3-
1
1
13.49, 114.21, 121.75, 126.03, 126.67, 127.27, 128.62,
quinolinecarboxylate (4).
4.8 mmol) was added slowly to a suspension of sodium hydride
60% in mineral oil, 3.83 g, 94.8 mmol) in DMF (30 mL) under N2
Dimethyl malonate (12.30 g,
38.38, 140.72, 141.66, 159.00, 162.61, 168.64. HRMS (ESI),
9
(
+
calcd for C19
79.0822.
H
17ClN
2
O
3
[M + Na] m/z: 379.0747; found:
3
atmosphere. The mixture was stirred at room temperature until the
evolution of hydrogen gas ceased. A solution of compound 3
(
10.00 g, 47.4 mmol) in DMF (100 mL) was added slowly, and the
REFERENCES AND NOTES
mixture was heated to 80°C for 5 h. The mixture was cooled to
room temperature, poured into ice water (100 mL), and acidified by
cold 5% hydrochloride acid (90 mL). The precipitate was filtered,
washed with water (2 × 20 mL), and dried at 50°C under vacuum
[
1] Tan, I.; Nijeholt, G. L.; Polman, C.; Ader, H.; Barkhof, F. Mult
Scler 2000, 6, 99.
[2] Noseworthy, J.; Wolinsky, J.; Lublin, F.; Whitaker, J.; Linde, A.;
Gjorstrup, P.; Sullivan, H. Neurology 2000, 54, 1726.
[3] Jönsson, S.; Andersson, G.; Fex, T.; Fristedt, T.; Hedlund, G.;
Jansson, K.; Abramo, L.; Fritzson, I.; Pekarski, O.; Runström, A. J Med
Chem 2004, 47, 2075.
to give the title compound (12.65 g, yield 99%) as a white solid in
1
98.9% purity. mp 158.8–159.7°C. ([5] 159.1–160.1°C). H-NMR
(
CDCl , 600 MHz) δ: 3.66 (s, 3H), 4.05 (s, 3H), 7.25–7.30 (m,
H), 7.53 (t, J= 8.1Hz, 3H), 14.96(s, 1H). C-NMR (CDCl ,
3
50 MHz) δ: 30.10, 53.25, 97.92, 112.28, 113.33, 126.00, 134.38,
34.45, 143.44, 158.71, 172.59, 173.31.
3
13
2
1
1
[4] Polman, C.; Barkhof, F.; Sandberg-Wollheim, M.; Linde, A.;
Nordle, O.; Nederman, T. Neurology 2005, 64, 987.
[5] Comi, G.; Pulizzi, A.; Rovaris, M.; Abramsky, O.; Arbizu, T.;
Boiko, A.; Gold, R.; Havrdova, E.; Komoly, S.; Selmaj, K. Lancet 2008,
5
-Chloro-1,2-dihydro-N-ethyl-4-hydroxy-1-methyl-2-oxo-
N-phenyl-3-quinoline carboxamide (1). A mixture of
compound 4 (10.00 g, 37.5 mmol) and N-ethylaniline (10 mL,
5.0mmol) in heptane (200 mL) was heated to reflux. The
371, 2085.
[
[
6] Brück, W.; Wegner, C. J Neurol Sci 2011, 306, 173.
7] Wennerberg, J.; Björk, A.; Fristedt, T.; Granquist, B.; Jansson, K.;
7
Τhuvesson, I. Org Process Res Dev 2007, 11, 674.
8] Goren, Z.; Heeg, M. J.; Mobashery, S. J Org Chem 1991, 56,
methanol formed during the reaction was distilled off together
with some heptane (120 mL) for 6 h. After cooling to room
temperature, the precipitate was filtered, washed with heptane
[
7186.
[9] Wilder, R.; Mobashery, S. J Org Chem 1992, 57, 2755.
(
2× 20mL), and dried at 50°C under vacuum to give the crude
[10] Pasquato, L.; Modena, G.; Cotarca, L.; Delogu, P.; Mantovani, S.
J Org Chem 2000, 65, 8224.
title compound (13.35 g, yield 97%) as a white crystal in 99.5%
purity. mp 200.8–201.5 ([5] 201°C). H-NMR (CDCl
δ: 1.21 (t, J = 6.9Hz, 3H), 3.29 (s, 3H), 3.99 (q, J = 6.9 Hz, 2H),
1
[11] Tundo, P.; Selva, M. Acc Chem Res 2002, 35, 706.
3
, 600MHz)
[
12] Jansson, K.; Fristedt, T.; Olsson, A.; Svensson, B.; Jönsson, S.
J Org Chem 2006, 71, 1658.
13] Maw, G. N.; Rawson, D. J.; Thompson, L. R. WO/2004/
13
7
.12–7.27 (m, 2H), 7.41 (t, J = 8.1 Hz, 1H), 12.64 (br, 1H). C-
NMR (CDCl , 150 MHz) δ: 12.91, 29.82, 45.68, 105.37, 112.72,
13.28, 125.40, 126.71, 126.94, 128.54, 131.64, 132.67, 142.05,
42.57, 157.99, 165.26, 168.69.
[
3
0
14357 A2 2004; Chem Abstr 2004, 140, 181216.
[14] Marom, E.; Mizhiritskii, M.; Rubnov, S. WO/2012/070051 A1
2012; Chem Abstr 2012, 157, 45028.
1
1
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet