Development of a practical synthesis of 7-bromo-8-methoxycarbonyl-3,3-dimethyl-3,4-dihydro-1hquinoxalin- 2-one

Article abstract of DOI:10.3987/COM-15-13240

Synthesis of 7-bromo-8-methoxycarbonyl-3,3-dimethyl-3,4-dihydro- 1H-quinoxalin-2-one from 5-amino-2-bromobenzoic acid was achieved in an overall yield of 27% over four steps. This is a significant improvement from the previous six-step process, which affo

Full text of DOI:10.3987/COM-15-13240

HETEROCYCLES, Vol. 91, No. 8, 2015, pp. - . © 2015 The Japan Institute of Heterocyclic Chemistry
Received, 4th May, 2015, Accepted, 19th June, 2015, Published online, 9th July, 2015
DOI: 10.3987/COM-15-13240
DEVELOPMENT
OF
A
PRACTICAL
SYNTHESIS
OF
7-BROMO-8-METHOXYCARBONYL-3,3-DIMETHYL-3,4-DIHYDRO-1H-
QUINOXALIN-2-ONE
Kazuhiro Kudo,a, b, c* Takahiro Honda,a, b and Noriyoshi Yamamotob
aGraduate School of Materials Sciences, Nara Institute of Science and Technology,
Ikoma, Nara, Japan
bNara Research & Development Center, Santen Pharmaceutical Co., Ltd, 8916-16,
Takayama-cho, Ikoma-shi, Nara 630-0101, Japan
cProduct Supply Division, Santen Pharmaceutical Co., Ltd, 348-3, Aza Suwa
Oaza Shide Taga-cho Inukami-gun, Shiga 522-0314. Tel: +81-749-48-2924;
E-mail: kazuhiro.kudou@santen.co.jp
Abstract – Synthesis of 7-bromo-8-methoxycarbonyl-3,3-dimethyl-3,4-dihydro-
1H-quinoxalin-2-one from 5-amino-2-bromobenzoic acid was achieved in an
overall yield of 27% over four steps. This is a significant improvement from the
previous six-step process, which afforded only a 12% yield. The previous process
was optimized by adding a high-yielding alkylation step under mild conditions as
well as by reducing the number of isolation steps and eliminating purification by
column chromatography.
INTRODUCTION
Glucocorticoids are commonly used to treat inflammatory diseases. Apart from their potent
anti-inflammatory activities, long-term and high dose treatment may lead to serious adverse effects.1
Selective glucocorticoid receptor agonists (SEGRA) that can separate an anti-inflammatory activity, and
their adverse effects are reported in recent years.1,2 SEGRA compounds with 3,3-dimethyl-3,4-dihydro-
1H-quinoxalin-2-one framework (Figure 1) have been developed as candidates for SEGRA and show a
remarkable anti-inflammatory effect after topical dosing to the eye and side-effect dissociation by in vivo
assay.2b-e The structure of SEGRA compounds was not a steroid framework and was reportedly the most
desirable example of drug efficacy to have dimethyl functions in 3, 3-position and carbonyl function at
2-position. To evaluate their activities in preclinical and clinical studies, an efficient, reliable, and
cost-effective synthetic method is required. Thus, a practical synthesis of a 3,3-dimethyl-3,4-dihydro-

1H-quinoxalin-2-one framework must be developed for the construction of SEGRA compounds.
ꢀꢁ
ꢀꢁ
ꢀꢁ
ꢀꢁ
ꢐ
ꢀꢁꢂ ꢂꢃꢄꢅ ꢄꢆꢂꢇꢈꢃꢄꢅ ꢄꢆꢂꢇꢈꢃꢉ ꢄꢊꢆꢂ
ꢇꢈꢊꢋꢄꢌꢍꢎ ꢄꢆꢂꢃꢉ ꢄꢊꢏ
ꢇ
ꢒꢆꢂꢒꢈ!ꢓꢔꢕꢖꢗ ꢄꢈꢒꢆꢂꢘꢈꢙꢓꢗ ꢙꢉꢍꢈꢁꢑ
ꢈꢚꢋꢓꢎꢍꢛꢃꢄꢓꢎꢈꢜꢈꢍꢎꢕꢂꢌꢉꢃꢔꢕꢝꢍꢉꢅ
ꢇ
ꢀꢁ
ꢐ
ꢑ
Figure 1. SEGRA compounds with 3,3-dimethyl-3,4-dihydro-1H-quinoxalin-2-one framework
In the conventional synthetic route, a number of SEGRA compounds were synthesized from key
intermediate 1a (Scheme 1).2b-e A six-step procedure was used to synthesize 1a, the first step was the
reduction of the nitro group on 3 by SnCl2 (94% yield). This was followed by the protection of the amino
group on 4a (98% yield), nitration of 5, and purification by silica gel column chromatography to afford 6
in 62% yield. Deacetylation afforded 7 in quantitative yield, and alkylation with ethyl 2-bromoisobutyrate
(2a) and isolation by silica gel column chromatography afforded 8a in 31% yield. Finally, reduction of
the nitro group and successive intramolecular cyclization and column chromatography provided 1a in
70% yield. Although 1a was successfully synthesized in an overall 12% yield over six steps, this method
has many disadvantages, such as low yield, use of a halogenated solvent, and several purification steps
using silica gel column chromatography. Step 5 was particularly problematic as it resulted in a low yield
(31%), excess use of 2a as a solvent, use of expensive Cs2CO3, high temperature, and long reaction time
(85 °C, 4 d). Purification after this step was also difficult and required two consecutive purifications by
silica gel column chromatography. Since this synthetic method was found to be very difficult, an
inexpensive, stable, and more efficient synthetic route for 1a is desirable.
㼏㼛㼚㼏㻚㻴 㻿㻻㻠
㼏㼛㼚㼏㻚㻴㻺㻻
㻻
㻻
㻻
㻻
㻞
㻭㼏㻯㼘
㻱㼠㻟㻺
㻻
㻻
㻻
㻻
㻿㼚㻯㼘㻞
㻮㻲㻟㻛㻱㼠㻞㻻
㻟
㻮㼞
㻮㼞
㻺㻻㻞
㻺㻴
㻮㼞
㻮㼞
㻹㼑㻻㻴㻘㻌㼞㼑㼒㼘㼡㼤㻘㻌㻞㻌㼔
㻯㻴㻞㻯㼘㻞㻘㻌㻜㻌䉝㻘
㻞㻚㻡㻌㼔
㻥㻤㻌㻑
㻜㻌䉝㻘㻌㻝㻚㻡㻌㼔
㻢㻞㻌㻑
㻹㼑㻻㻴㻘㻌㼞㼑㼒㼘㼡㼤㻘
㻞㻚㻡㻌㼔
㼝㼡㼍㼚㼠㻚
㻺㻴
㻥㻠㻌㻑
㻺㻻㻞
㻺㻴㻞
㻟
㻠㼍
㻡
㻢
㻻
㻻
㻿㼠㼑㼜㻌㻝
㻿㼠㼑㼜㻌㻞
㻿㼠㼑㼜㻌㻟
㻿㼠㼑㼜㻌㻠
㻻㻱㼠
㻮㼞
㻞㼍
㻷㻵
㻯㼟㻞㻯㻻㻟
㻻
㻻
㻻
㻻
㻻
㻴
㻻
㻣
㻻
㻿㼚㻯㼘㻞
㻮㼞
㻺㻻㻞
㻿㻱㻳㻾㻭
㼏㼛㼙㼜㼛㼡㼚㼐㼟
㻮㼞
㻺
㻻
㻮㼞
㻺㻻㻞
㻺㻴㻞
㻱㼠㻻㻴㻘㻌㼞㼑㼒㼘㼡㼤㻘㻌㻡㻌㼔
㻣㻜㻌㻑
㻤㻡㻌䉝㻘㻌㻠㻌㼐
㻟㻝㻌㻑
㻻㻱㼠
㻺
㻴
㻺
㻴
㻻
㻤㼍
㻝㼍
㻿㼠㼑㼜㻌㻡
㻿㼠㼑㼜㻌㻢
Scheme 1. Conventional synthetic route for key intermediate 1a

The reason for the resultant low yield in step 5 was assumed to be that the tertiary bromide 2a is a
hindered electrophile and 7 shows low nucleophilicity due to the presence of a strong electron
withdrawing group in the o-position. To improve the yield of alkylation, we selected 4 as a starting
material, which does not have a strong electron withdrawing group and was easy to procure. Alkylation
with 2 was investigated under mild conditions. We also explored the nitration of 8 to afford 1, completing
this new synthetic route (Scheme 2).
㻻㻾㻞
㻻㻾㻞
㻻㻾㻞
㻻
㻻
㻻
㻻㻾㻞
㻻
㻻㻾㻞
㻮㼞
㻺㻻㻞
㻻㻾㻞
㻴
㻺
㻞
㻻
㻮㼞
㻺㻻㻞
㻮㼞
㻮㼞
㻻
㻮㼞
㻻㻾㻞
㻺
㻴
㻺
㻴
㻺
㻴
㻺㻴㻞
㻻
㻻
㻝
㻤
㻥
㻠
㻾㻞㻌㻩㻌㻴㻌㼛㼞㻌㼍㼘㼗㼥㼘
Scheme 2. Retrosynthesis of key intermediate 1
RESULTS AND DISCUSSION
Alkylation of tertiary, hindered electrophile 2 with 4 was investigated.3 In the case of 4a and 2a,
base/solvent combinations of K2CO3/DMF, Et3N/DMF, and Et3N/2-PrOH at 50 °C did not proceed at all
and only the starting material 4a was recovered (Table 1, entries 1–3). When carboxylic acid 4b was
reacted in K2CO3/DMF, 4b and O-alkylation product A were found in a ratio of 59:41, with no detection
of target material 9b (entry 4). In the case of Et3N/DMF, the conversion was relatively high (76%) and
the remaining 4b, 9b, and A were found in a ratio of 24:41:35 (entry 5).4 In Et3N/2-PrOH, the conversion
was low and mainly recovered 4b, a small amount of 9b, and A (85:12:3, entry 6).
Table 1. Alkylation of 4a and 4b with 2aa
entry
R3
base
K2CO3
Et3N
solvent
DMF
ratio of 4:9:A
100:N.D.:N.D.
100:N.D.:N.D.
conversion (%)
1
2
Me
0
0
DMF

3
4
5
6
Et3N
K2CO3
Et3N
2-PrOH
DMF
100:N.D.:N.D.
59:N.D.:41
24:41:35
0
H
41
76
15
DMF
Et3N
2-PrOH
85:12:3
1
aMolar ratio and conversion were determined from H NMR spectra of the
reaction mixture.5
Conversely, alkylation of 4b and carboxylic acid 2b in K2CO3/DMF showed low conversion to 9d (44%).
However, mainly 4b and 9d were observed in the reaction mixture, with 4b being converted to target
material 9d in high amounts (Table 2, entry 1). In the case of Et3N/DMF, 99% conversion was observed;
however, only 15% was target material 9d, while the rest were impurities (entry 2). The use of 2-PrOH as
the solvent and Et3N as the base gave high conversion with few impurities, affording 9d in 85% yield
(entry 3). In the case of methyl ester 4a, 85% conversion was confirmed and the yield of target material
9c slightly decreased to 68% (entry 4). The use of EtOH as the solvent resulted in a decreased yield of
48% due to residual unreacted 4b (entry 5). When MeOH was used as the solvent, a further decrease in
conversion was observed (24%) and only the remaining 4b and 9d were confirmed. In the reaction
mixtures in Table 2, impurity B was not observed through 1H NMR spectroscopic analysis.
Table 2. Optimization of alkylation
entry R3
base
K2CO3
Et3N
solvent yield (%) of 9 conversion (%)a time (h)
b
1
2
3
4
5
6
H
DMF
DMF
44
99d
89d
85
17
23
20
19
23
21
15c
85c
68e
Et3N 2-PrOH
Me Et3N 2-PrOH
H
Et3N
EtOH
48e
65
f
Et3N
MeOH
24
b
aConversion was determined by 1H NMR spectroscopy. Not isolated, but mainly
1
9d and 4b were confirmed in approximately 1:1 molar ratio by H NMR
c
d
spectroscopy of the reaction mixture. HPLC assay yield. Determined by HPLC.
eIsolated yield. fNot isolated.

The lower yield of the alkylation reaction shown in Table 2 was due to the decreased steric effect of the
alcohol, which was investigated through the addition of H2O to 2-PrOH. The reaction was performed at
50 °C for 18 h and with the same equivalents of reagents. Under these conditions, the ratio of the
remaining 4b increased in proportion to H2O addition and the conversion decreased (Table 3).
Table 3. Investigation of the influence of H2O on alkylation
㻴㻻
㻮㼞
㻻
㻻㻴
㻻
㻻㻴
㻻
㻞㼎㻌㻝㻚㻡㻌㼑㼝
㻮㼞
㻻
㻻㻴
㻱㼠㻟㻺㻌㻠㻚㻜㻌㼑㼝㻌
㻮㼞
㻞㻙㻼㼞㻻㻴㻌㻗㻌㻴㻞㻻㻘
㻡㻜㻌䉝㻘㻌㻝㻤㻌㼔
㻺
㻴
㻺㻴㻞
㻥㼐
㻠㼎
entry
solvent
2-PrOH (dry)
ratio of 4b:9d a
8:92
conversion (%)
1
2
3
93
84
77
2-PrOH/H2O (95:5)
18:82
2-PrOH/H2O (90:10)
27:73
aConfirmed by HPLC.
Therefore, the combination of carboxylic acid 2b and 4b in Et3N/2-PrOH afforded the best results, with
high conversion and high yield of 9d. Furthermore, it was found that the addition of H2O or a low steric
effect solvent decreased the yield of 9d. The mechanism was believed to involve 2b forming highly active
intermediate 10, which could then react with the amino group on 4b (Scheme 3). Active intermediate 10
was able to react with H2O and low steric effect alcohols, resulting in low yields.6 In the case of
O-alkylation to yield B, the retro-O-alkylation proceeds intramolecularly due to the vicinal effect of the
carboxylic acid, making B unstable.
㻻
㻻㻴
㻻
㻻㻴
㻮㼞
㻮㼞
㻻
㻻㻴
㻻㻙
㻻
㻱㼠㻟㻺
㻺㻴㻞
㻗
㻠㼎
㻮㼞
㻻㻴
㻻
㻺
㻴
㻞㻙㻼㼞㻻㻴
㻥㼐
㻞㼎
㻝㻜
㻠㼎
㻴㻞㻻
㻻
㻻
㻻
㻻
㻻㻙
㻴㻻
㻮㼞
㻻㻴
㼕㼚㼍㼏㼠㼕㼢㼑
㻺㻴㻞
㻮
Scheme 3. Plausible mechanism of alkylation

The nitration reaction of 9 was subsequently investigated (Table 4). Reaction of dicarboxylic acid 9d in
both HNO3/H2SO4 and NaNO3/TFA7 yielded a complex mixture with a small amount of remaining 9d
(entries 1 and 2). Nitration of 9c, where R4 is a methyl ester, in NaNO3/TFA also yielded a complex
mixture and no remaining 9c (entry 3). The reaction mixture of dimethyl ester 9e in HNO3/H2SO4
contained not only remaining 9e and target material 8e but also regioisomer 11e and a slight amount of
dinitration product 12e (27:42:23:8, entry 4).8 For the reaction of 9e in NaNO3/TFA, no 9e remained and
the reaction progress was confirmed as the ratio of target material 8e, 11e, and 12e was 62:22:16, with a
57% HPLC assay yield of 8e.9 Impurities were removed through recrystallization in 2-PrOH, affording 8e
in 38% yield and 99% purity. It was found that dimethyl esterification from 9d to 9e was necessary for
the nitration reaction and was accomplished in 80% yield by stirring 9d in MeOH in the presence of
H2SO4.
Table 4. Investigation of nitration of 9
nitro
temp. time
ratio of
entry R4 R5
equiv. solvent
yield of 8e (%)
agent
(°C)
0
(h)
3
9:8:11:12
1
2
3
4
5
H
H
HNO3 15 v/wa H2SO4
complex mixtureb
complex mixtureb
complex mixturec
-
NaNO3
NaNO3
1.05
1.05
TFA
TFA
25
25
0
4
-
Me
H
5
-
Me Me HNO3 15 v/wa H2SO4
3
27:42:23:8e
N.D.:62:22:16g
d
NaNO3 1.0 TFA
50
5
57(38) f
b
c
aWeight of 9. Confirmed a small amount of remaining 9d and many impurities. Confirmed many
impurities and no remaining 9d. dNot isolated. eConfirmed by 1H NMR spectroscopy. HPLC assay yield
f
(isolated yield after recrystallization). gConfirmed by HPLC.
In the final synthesis of 1a, the use of SnCl2 and MeOH under reflux afforded 1a in 70% yield with a
conventional method.2b-e Our investigation revealed that Fe powder (I = 45 μm) with AcOH/MeOH (1:1)
reduced the nitro group, followed by an intramolecular cyclization to provide 1a in 96% yield (Scheme
4).10

Scheme 4. Nitro reduction and successive cyclization of 8e
Optimization of the total synthetic procedure was investigated to further improve the synthesis of 1a.
Products 9d and 9e were found to be an amorphous solid and liquid, respectively. They were thus
difficult to purify by simple methods such as recrystallization (Scheme 5). When the alkylation step was
complete, the reaction solvent was substituted with MeOH and H2SO4 was added, allowing for a
subsequent esterification. Upon completion, water soluble salts and impurities were removed by
extraction. The organic layer was substituted with TFA, and nitration was performed by the addition of
NaNO3 without the isolation of 9e. The reaction mixture was then extracted, the organic layer was
substituted for 2-PrOH, and 8e was recrystallized in 29% yield from 4b11 and 99.0% HPLC purity. This
was then reduced and cyclized with Fe, which was then removed by filtration and extracted with aqueous
HCl using a reduction-cyclization step. Finally, the precipitated solid was washed with n-hexane/AcOEt
(5:1) to afford 1a in 93% yield with over 99.9% HPLC purity.
Scheme 5. Optimization of synthetic procedure toward new route
In the previous synthesis, the total yield of 1a was 12% with six steps. It required six isolation steps and
purification by four silica gel columns. With this improved procedure, 1a was produced in 27% yield with
no need for purification by silica gel column chromatography and with only two isolation steps (Table 5).

Table 5. Comparison of previous and new routes
number of
number of
number of silica gel
total yield
steps
isolation steps
columns
(%)
12
previous route
new route
6
4
6
2
4
0
27
CONCLUSION
A practical, efficient four-step process for the synthesis of 1a was developed. This novel procedure
afforded 1a in high purity and an improved overall yield of 27% compared with the previous yield of
12%. Alkylation with hindered electrophiles 2b and 4b afforded 9d in high yield when a combination of
Et3N and 2-PrOH was used. Nitration with NaNO3 in TFA preferentially afforded 8e at the desired
position with good purity following recrystallization in 2-PrOH. Final product 1a was synthesized using
Fe in the nitro reduction of 8e and cyclization, resulting in high yield and purity. Optimization of this
synthetic procedure resulted in reducing the isolation steps from six to two, and did not require
purification by column chromatography.
EXPERIMENTAL
GENERAL
Unless otherwise noted, all materials were obtained from commercial suppliers and used without further
1
13
purification. H NMR and C NMR spectra were recorded on a JEOL GSX400 or JEOL ECP500 or
JEOL JNM-ECS400 spectrometer at ambient temperature. Chemical shifts are reported as ppm downfield
from tetramethylsilane. Data are reported as follows: chemical shift, multiplicity (s = singlet, d = doublet,
t = triplet, q = quartet, br = broad, m = multiplet), coupling constants, and the number of protons. HPLC
analyses were conducted on a Waters alliance series. In process method for preparations of 9d, 9c and 9e;
C8 column (3.5 μm; 4.6 mm I.D. × 150 mm) (X BridgeTM), Mobile phase A: 0.05% TFA in water,
Mobile phase B: 0.05% TFA in MeCN, Gradient condition: 0 min 95% A, 5 min 95% A, 25 min 10% A,
30 min 10% A, column temperature 40 °C, 1.0 mL/min flow rate, UV detector at 225 nm. Retention
times: 4b (6.3 min), 9d (13.6 min), 9c (16.5 min), 9e (19.0 min). In process method for preparations of 8e
and 1a; C8 column (3.5 μm; 4.6 mm I.D. × 150 mm) (X BridgeTM), isocratic elution with 10 mM
CH3CO2NH4 aq/MeCN (60:40), column temperature 40 °C, 1.0 mL/min flow rate, UV detector at 225 nm.
Retention times: 1a (4.9 min), 9e (9.8 min), 8e (13.7 min), 11e (16.8 min), 12e (18.9 min).
2-Bromo-5-(1-carboxy-1-methylethylamino)benzoic
acid
(9d).
To
a
mixture
of

5-amino-2-bromobenzoic acid 4b (10.0 g, 46.3 mmol), 2-bromoisobutyric acid 2b (11.6 g, 69.5 mmol)
and anhydrous 2-PrOH (100 mL) was added Et3N (26.0 ml, 0.187 mol), and the mixture was stirred for
20 h at 50 °C. When the reaction was completed, the reaction mixture was evaporated under vacuum
pressure. Water (300 mL) and 6M HCl (20 mL) were added and extracted with AcOEt (150 mL×2). The
combined organic layer was dried over anhydrous Na2SO4, filtered, and evaporated under vacuum
pressure to achieve the 9d as a pale yellow solid (12.3 g, 88%). HPLC purity: 95.8%, which was used in
1
the next reaction without any purification. H NMR (DMSO-d6, 400 MHz) δ: 1.42 (s, 6H), 6.51 (dd, J =
8.8, 2.9 Hz, 1H), 6.87 (d, J = 2.9 Hz, 1H), 7.33 (d, J = 8.8 Hz, 1H), 12.86 (br s, 2H); 13C NMR (DMSO-d6,
100 MHz) δ: 25.52, 55.99, 104.57, 115.15, 117.11, 133.40, 133.66, 145.96, 167.67, 176.80; IR (KBr)
cm−1: 1035, 1157, 1218, 1257, 1473, 1601, 1717, 2510, 2769, 2987, 3387; Mp 105 °C; HRMS (EI): calcd
for C11H12BrNO4 [M+]: 300.9950. Found; 300.9952.
Methyl 2-Bromo-5-(1-carboxy-1-methylethylamino)benzoate (9c). To a mixture of methyl
5-amino-2-bromobenzoate 4a (0.436 g, 1.90 mmol), 2-bromoisobutyric acid 2b (0.479 g, 2.87 mmol) and
anhydrous 2-PrOH (4.5 mL) was added Et3N (1.07 mL, 7.68 mmol), and the mixture was stirred for 19 h
at 50 °C. When the reaction was completed, the reaction mixture was evaporated under vacuum pressure.
Water (30 mL) and 1M HCl (5 mL) were added and extracted with AcOEt (15 mL×2). The combined
organic layer was washed with brine and dried over anhydrous Na2SO4. After filtration, the organic
solution was concentrated to dryness under reduced pressure. The residue was was purified by column
chromatography to afford 9c as a pale yellow solution (0.407 g, 68%). 1H NMR (DMSO-d6, 400 MHz) δ:
1.42 (s, 6H), 3.81 (s, 3H), 6.54 (dd, J = 8.8, 3.3 Hz, 1H), 6.88 (d, J = 3.3 Hz, 1H), 7.36 (d, J = 8.8 Hz, 1H),
13
12.47 (s, 1H); C NMR (DMSO-d6, 100 MHz) δ: 25.52, 52.36, 56.04, 104.54, 115.26, 117.54, 132.11,
133.83, 146.09, 166.57, 176.74; IR (KBr) cm−1: 1032, 1154, 1254, 1334, 1437, 1474, 1601, 1725, 2949,
2991, 3388; HRMS (EI): calcd for C12H14BrNO4 [M+]: 315.0106. Found; 315.0090.
Methyl 2-Bromo-5-(1-methoxycarbonyl-1-methylethylamino)benzoate (9e). To a mixture of
2-bromo-5-(1-carboxy-1-methylethylamino)benzoic acid 9d (10.8 g, 35.8 mmol) and anhydrous MeOH
(150 mL) was added 12% HCl (15 mL), and the mixture was stirred for 22 h at 70 °C. When the reaction
was completed, the reaction mixture was evaporated under vacuum pressure. Water (200 mL) was added
and extracted with AcOEt (120 mL×2). The combined organic layer was washed with brine and dried
over anhydrous Na2SO4. After filtration, the organic solution was concentrated to dryness under reduced
pressure. The residue was purified by chromatography to afford 9e as a colorless solution (9.44 g, 80%).
1H NMR (DMSO-d6, 400 MHz) δ: 1.45 (s, 6H), 3.61 (s, 3H), 3.81 (s, 3H), 6.46 (dd, J = 8.7, 2.7 Hz, 1H),
6.51 (s, 1H), 6.85 (d, J = 2.7 Hz, 1H), 7.37 (d, J = 8.7 Hz, 1H); 13C NMR (DMSO-d6, 100 MHz) δ: 25.55,

52.25, 52.37, 56.31, 104.87, 115.51, 117.19, 132.10, 133.99, 145.78, 166.41, 175.64; IR (KBr) cm−1:
1031, 1147, 1247, 1335, 1386, 1436, 1475, 1516, 1601, 1733, 2953, 2994, 3398; HRMS (EI): calcd for
C13H16BrNO4 [M+]: 329.0263. Found; 329.0260.
Methyl 6-Bromo-3-(1-methoxycarbonyl-1-methylethylamino)-2-nitrobenzoate. (8e). To a mixture of
methyl 2-bromo-5-(1-methoxycarbonyl-1-methylethylamino)benzoate 9e (0.861g, 2.61 mmol) and TFA
(4.30 mL) was added NaNO3 (0.224g, 2.63 mmol) at 50 °C, and the mixture was stirred for 5 h at 50 °C.
When the reaction was completed, the reaction mixture was evaporated under vacuum. Water (10 mL)
and 4M NaOH aq (4.0 mL) were added and extracted with AcOEt (15 mL×2). The combined organic
layer was dried over anhydrous Na2SO4, filtered, and evaporated under vacuum pressure. 2-PrOH was
added and the mixture was stirred for 20 h, and the precipitated solid was filtered. The filtered cake was
further washed with 2-PrOH (0.5 mL) and dried in vacuo at 50 °C to afford compound 8e as an orange
solid (0.368 g, 38%). 1H NMR (DMSO-d6, 400 MHz) δ: 1.61 (s, 6H), 3.70 (s, 3H), 3.86 (s, 3H), 6.70 (d, J
13
= 9.3 Hz, 1H), 7.76 (d, J = 9.3 Hz, 1H), 8.01 (s, 1H); C NMR (DMSO-d6, 100 MHz) δ: 25.41, 52.98,
53.23, 57.44, 105.20, 118.35, 130.68, 132.59, 138.86, 141.78, 165.24, 174.15; IR (KBr) cm−1: 1031, 1088,
1151, 1217, 1262, 1358, 1387, 1444, 1498, 1605, 1734, 2958, 3004, 3377, 3455; Mp 119 °C; HRMS (EI):
calcd for C13H15BrN2O6 [M+]: 374.0113. Found; 374.0113.
7-Bromo-8-methoxycarbonyl-3,3-dimethyl-3,4-dihydro-1H-quinoxalin-2-one (1a). To a mixture of
methyl 6-bromo-3-(1-methoxycarbonyl-1-methylethylamino)-2-nitrobenzoate 8e (0.500 g, 1.33 mmol),
MeOH (3.75 mL) and AcOH (3.75 mL) was added Fe (I=45 μm) (0.224 g, 4.01 mmol) at ice cooling,
and the mixture was stirred for 12 h at rt. When the reaction was completed, the MeOH and AcOH were
evaporated under vacuum pressure. AcOEt (20 mL) was added and the mixture was sttired for 20 min at
60 °C. The mixture was filtered by celite® and the filtrate was added H2O (20 mL) and extracted. The
1
organic layer was evaporated under vacuum pressure to achieve a pale gray solid (0.400 g, 96%). H
NMR (DMSO-d6, 400 MHz) δ: 1.22 (s, 6H), 3.84 (s, 3H), 6.46 (s, 1H), 6.72 (d, J = 8.4 Hz, 1H), 7.07 (d, J
13
= 8.4 Hz, 1H), 10.09 (s, 1H); C NMR (DMSO-d6, 100 MHz) δ: 24.68, 52.79, 53.71, 105.98, 116.25,
120.96, 124.53, 126.28, 134.01, 165.51, 170.29; IR (KBr) cm−1: 1027, 1160, 1265, 1286, 1311, 1358,
1387, 1436, 1469, 1591, 1691, 1716, 1855, 2969, 3060, 3113, 3206, 3325; Mp 156 °C; HRMS (EI): calcd
for C12H13BrN2O3 [M+]: 312.0110. Found; 312.0098.
Optimaized process for 1a. To a mixture of 5-amino-2-bromobenzoic acid 4b (500 g, 2.32 mol),
2-bromoisobutyric acid 2b (580 g, 3.47 mol) and 2-PrOH (5.00 L) was added Et3N (1.29 L, 9.26 mol),
and the mixture was stirred for 7 h at 50 °C. When the reaction was completed as indicated by HPLC

(conv. >92%), the reaction mixture was filtered and the filtrate was evaporated under vacuum pressure
and azeotroped with MeOH (2.5 L×2). Dry MeOH (10 L) and 12% HCl (500 mL) were added to the
mixture and the mixture was stirred for 19 h at 70 °C. When the reaction was completed as indicated by
HPLC (conv. >90%), the reaction mixture was evaporated under vacuum pressure and water (5.00 L) was
added and extracted with AcOEt (5.00 L×2). The combined organic layer was evaporated under vacuum
pressure to achieve an oil (641 g). TFA (4.80 L) was added to the oil and stirred at 20 °C. NaNO3 (164.9
g, 1.94 mol) was added to the mixture and stirred for 5 h at 50 °C. When the reaction was completed as
indicated by HPLC (conv. >98%), the reaction mixture was evaporated under vacuum pressure and
neutralized by 4 M NaOH aq (1.44 L). Water (3.6 L) was added to the mixture and extracted with AcOEt
(5.00 L). The organic layer was evaporated under vacuum pressure to achieve an oil. 2-PrOH (1.80 L)
was added to the oil and stirred for 12 h at rt. The precipitated solid was filtered and the filtered cake was
further washed with 2-PrOH (500 mL) and dried in vacuo to a constant weight at 50 °C to afford
compound 8e as an orange solid (247 g, 29% yield from 4b, HPLC purity: 99.0%). To a mixture of
methyl 6-bromo-3-(1-methoxycarbonyl-1-methylethylamino)-2-nitrobenzoate 8e (247 g, 0.658 mol),
MeOH (1.73 L) and AcOH (1.73 L) was added Fe (I=45 μm) (110 g, 1.98 mol) at ice cooling, and the
mixture was stirred for 23 h at rt. When the reaction was completed as indicated by HPLC (conv. >99%),
the reaction mixture was filtered by celite® and the filtrate was evaporated under vacuum pressure. To the
mixture was added 1M HCl (3.75 L) and extracted with AcOEt (3.75 L). The organic layer was washed
by H2O (3.75 L), filtered and evaporated under vacuum pressure to give a solid. The solid was washed
with n-hexane:AcOEt (5:1), (250 mL) and dried in vacuo to a constant weight at 50 °C to afford
compound 1a as a pale gray solid (191 g, 93%). HPLC purity: >99.9%.
ACKNOWLEDGEMENTS
We thank Ms. Eiko Nishioka for NMR analysis.
REFERENCES AND NOTES
1. A. Schottelius, W. D. Döcke, P. Strehlke, S. Jaroch, N. Schmees, H. Rehwinkel, H. Hennekes, and K.
Asadullah, Proc. Natl. Acad. Sci. U.S.A., 2004, 101, 227.
2. (a) S. S. Hadida-Ruah, H. E. Xiaohui, and J. Y. Nagasawa, WO2004/110385 A2; (b) M. Matsuda, T.
Mori, K. Kawashima, M. Yamamoto, M. Kato, M. Takai, M. Nagatsuka, and S. Kobayashi, WO
2007/105766 A1; (c) M. Matsuda, T. Mori, M. Nagatsuka, and S. Kobayashi, WO 2008/146871 A1;
(d) M. Kato, M. Takai, T. Matsuyama, T. Kurose, Y. Hagiwara, M. Matsuda, and T. Mori, WO
2009/035067 A1; (e) M. Kato, M. Takai, T. Matsuyama, T. Kurose, Y. Hagiwara, K. Oki, M.
Matsuda, and T. Mori, WO 2010/029986 A1.

3. Examples of alkylation of 2 and 4 homologs were reported before, but the reported yields were not
stable for every substrate, see: F. Hoffmann, et al. WO2011/128251 A1.; K. Tachibana, et al.
EP1790640 A1.; S. D. Clarke, J. C. Metcalfe, and G. A. Smith, J. Chem. Soc., Perkin Trans. 2, 1993,
1187.
4. Compounds 9b and A were assumed by 1H NMR. 9b: 1H NMR (DMSO-d6, 400 MHz) δ: 1.09 (t, J =
7.0 Hz, 3H), 1.44 (s, 6H), 4.08 (q, J = 7.0 Hz, 2H), 6.42 (s, 1H), 6.46 (dd, J = 8.7, 3.2 Hz, 1H), 6.84
(d, J = 3.2 Hz, 1H), 7.33 (d, J = 8.7 Hz, 1H), 13.10 (br s, 1H). A: 1H NMR (400 MHz, DMSO-d6) δ:
1.20 (t, J = 7.0 Hz, 3H), 1.59 (s, 6H), 4.14 (q, J = 7.0 Hz, 2H), 5.59 (s, 2H), 6.64 (dd, J = 8.4, 2.9 Hz,
1H), 6.92 (d, J = 2.9 Hz, 1H), 7.29 (d, J = 8.4 Hz, 1H).
1
5. Sample preparation of H NMR analysis: sampling of small amount from reaction mixture and
controlling mild acid by HCl aq and H2O. The mixture was extracted with AcOEt and the organic
layer was concentrated to dryness.
6. An addition product of H2O or low steric alcohol was not isolated and not observed by 1H NMR and
HPLC analysis.
7. P. P. Fu, L. S. Von Tungeln, L.-H. Chiu, D.-J. Zhan, J. Deck, T. Bucci, and J.-C. Wang, Chem. Res.
Toxicol., 1998, 11, 937.
8. Compounds 11e and 12e were assumed by 1H NMR. 11e: 1H NMR (DMSO-d6, 400 MHz) δ: 1.63 (s,
1
6H), 3.74 (s, 3H), 3.88 (s, 3H), 6.97 (s, 1H), 8.18 (s, 1H), 8.34 (s, 1H). 12e: H NMR (DMSO-d6,
400 MHz) δ: 1.60 (s, 6H), 3.76 (s, 3H), 3.93 (s, 3H), 8.32 (s, 1H), 8.68 (s, 1H).
9. Although the causes of regioselectivity are unknown, it is assumed to be comprehensive affection of
substituents on the aromatic and vicinal effect in methyl carbonyl function.
10. G. C. Coutts, N. J. Culbert, M. E. Edwards, J. A. Hadfield, D. R. Musto, V. H. Pavlidis, and D. J.
Richards, J. Chem. Soc., Perkin Trans. 1, 1985, 1829.
11. Although 4b can be procured at an appropriate cost, the detailed information is not disclosable. 4b
was synthesized from 3-aminobenzoic acid with NBS in 78% in laboratory scale, see: K. Kudou, N.
Yamamoto, M. Ban, and A. Ohno, WO 2012/108455 A1.