A novel thiolate mediated cyclization to OPC-15161

Article abstract of DOI:10.1016/S0040-4020(00)00670-0

An efficient synthetic route to OPC-15161 (1) was developed via novel pyrazine ring closure promoted by a lithium thiolate anion. The key intermediate (4) was prepared in a one-pot procedure by treating the methyl ester (2) with a lithium arylthiolate. This protocol does not require a free acid intermediate and thus can establish the shortest route to pyrazine dioxide skeleton from tryptophan ester derivatives. In the present transformation, lithium arylthiolates could behave like aluminium arylthiolates, and not like lithium alkylthiolates that often cleave esters to the corresponding acids. One-pot reactions that involve lengthy multiple steps in a single flask are of significant importance in contemporary organic synthesis. Utilization of a catalytic or stoichiometric promoter which can facilitate several transformations is a key to the success of such reactions. In this paper, we would like to disclose an interesting one-pot transformation discovered in our process research, which offered us novel information on the reactivity of metal thiolates. Main feature of our one-pot process is a merged deprotection-cyclization sequence. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4020(00)00670-0

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 7427–7431
A Novel Thiolate Mediated Cyclization to OPC-15161
*
Koichi Shinhama, Katsuhide Matoba, Yasuhiro Torisawa and Jun-ichi Minamikawa
Process Research Laboratory, Second Tokushima Factory, Otsuka Pharmaceutical Co. Ltd, Kawauchi-cho, Tokushima 771-0182, Japan
Received 14 June 2000; accepted 31 July 2000
Abstract—An efficient synthetic route to OPC-15161 (1) was developed via novel pyrazine ring closure promoted by a lithium thiolate
anion. The key intermediate (4) was prepared in a one-pot procedure by treating the methyl ester (2) with a lithium arylthiolate. This protocol
does not require a free acid intermediate and thus can establish the shortest route to pyrazine dioxide skeleton from tryptophan ester
derivatives. In the present transformation, lithium arylthiolates could behave like aluminium arylthiolates, and not like lithium alkylthiolates
that often cleave esters to the corresponding acids. One-pot reactions that involve lengthy multiple steps in a single flask are of significant
importance in contemporary organic synthesis. Utilization of a catalytic or stoichiometric promoter which can facilitate several transforma-
tions is a key to the success of such reactions. In this paper, we would like to disclose an interesting one-pot transformation discovered in our
process research, which offered us novel information on the reactivity of metal thiolates. Main feature of our one-pot process is a merged
deprotection-cyclization sequence. ᭧ 2000 Elsevier Science Ltd. All rights reserved.
OPC-15161 (1) is a novel inhibitor against superoxide anion
generation, isolated in our institute from the product of the
culture broth of fungus Thielavia minor OFR-1561.1
Previous reports have already described successful chemical
syntheses of 1, though they necessitated undesirable multi-
ple steps with low yields.2 These initial difficulties have
been overcome in our plant-scale synthesis through the
beneficial use of a protecting group.3 We now wish to report
here another practical synthetic route to 1, which includes a
novel and simple cyclization protocol for the pyrazine
dioxide skeleton. This new method is notable because the
cyclization can be achieved by the use of a metal thiolate
anion. Furthermore, N,O-diprotected oxime ester (2) was
employed as starting material without extra conversion to
the deprotected hydroxyimino carboxylic acid precursor as
depicted in Scheme 1.
(3b) was also prepared in 44% yield in a similar manner
using benzyl bromide instead of dimethyl sulfate.
Having obtained these precursors, we next investigated the
transformation of the hydroxyimino esters (3) using various
kinds of metal thiolates in an attempt to form an active
thioester intermediate. While aliphatic metal thiolates
gave only complex mixtures, some aromatic or heterocyclic
metal thiolates were found to be an excellent promoter for
the sequential transformation to the cyclized product (4).
The pyrazine (4) was directly obtained from N,O-dipro-
tected ester (2) in a good yield when treated with lithium
2-benzimidazolethiolate generated in situ from 2-mercapto-
benzimidazole and excess lithium hydroxide in DMF. The
pyrazine (4) was also obtained in good yields without the
addition of excess lithium hydroxide when lithium naphtha-
lenethiolate was used. This direct conversion of 2 to 4 did
not take place in CH3CN, while with deprotected substrates
(3a, 3b), reaction proceeded even in CH3CN. The results
obtained are summarized in Table 1. Fig. 1 also delineates
the reaction course monitored by HPLC.
Here again, we utilized 2-cyanoethyl (CE) group as a key
protecting group for the N1-position of pyrazinone, because
it can be introduced and removed by a simple alkaline treat-
ment and does not interfere with the sensitive part of the
molecule (indole ring).3
These experimental observations suggested that the steric
effect of the alkoxycarbonyl group was small, because no
significant difference in yields was observed between 3a and
3b. The resulting hydroxypyrazinone (4) was methylated
with dimethyl sulfate to give the key intermediate (5) in
61% yield. To our delight, 5 was directly obtained in 37%
yield from 2 in one-pot procedure by treating with lithium
2-naphthalenethiolate followed by subsequent methylation.
Same conversion (2 to 5) by lithium 2-benzimidazolethiolate
gave less satisfactory results due to the formation of
by-products. This new sequence was operationally simple
and useful to prepare methoxypyrazinone (5).
The key intermediate N,O-diprotected ester (2) was
prepared without event (for the detailed sequence, see
Experimental). Hydrolysis of this ester (2) with aq. NaOH
followed by the addition of boric acid and dimethyl sulfate
gave the corresponding N-protected methyl esters (3a) in
65% yield. On the other hand, N-protected benzyl ester
Keywords: one-pot reaction; alkylthiolates; aluminium arylthiolate.
* Corresponding author. Tel.: ϩ88-665-2126; fax: ϩ88-637-1144;
e-mail: jm@fact.otsuka.co.jp
0040–4020/00/$ - see front matter ᭧ 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)00670-0

7428
K. Shinhama et al. / Tetrahedron 56 (2000) 7427–7431
Scheme 1. Direct cyclization sequence without deprotection.
Although the partially deprotected hydroxyimino esters (3a
and 3b) could be used for the same purpose, the N,O-
diprotected ester (2) was considered to be a better synthetic
intermediate because of the operational simplicity (less
steps required for preparation) and a higher isolated yield.
It also should be noted that deprotected hydroxyimino
acids were relatively unstable and more prone to isomeriza-
tion at oxime. We observed that 3a and 3b contained
syn-oxime isomer, which could not be cyclized into the
pyrazine (4).
Insights into the Mechanism
The cyclization described above is notable because almost
no reports have disclosed related cyclization between
oxime-nitrogen and ester promoted by a metal thiolate.
We initially speculated that this cyclization from 2 to 4
was brought about in the presence of a catalytic amount of
thiolate (as indicated in the following Scheme 2). However,
the actual reaction required more than 1.1 (and optimally
1.5–2.5) equivalents of the thiolate as shown in Table 1.
Table 1. Thiolate promoted cyclization
SM
ArS (equiv.)a/base (equiv.)
Solvent, time
4 (% HPLC yield)
2
2
2
2
3a
3a
3a
3b
3b
3b
Li–BIT (2.0)/LiOH (1.0)
Li–BIT (2.5)
Li–NPT (1.5)
Li–NPT (2.0)/LiOH (1.0)
Li–BIT (3.0)
Li–BIT (1.6)
Na-NPT (1.4)
Li–BIT (3.0)
Li–BIT (1.6)
Li–NPT (1.3)
DMF, 2.5 h
DMF, 1.5 h
DMF, 0.8 h
DMF, 1.8 h
DMF, 2.0 h
CH3CN, 1.5 h
CH3CN, 1.5 h
DMF, 2.1 h
CH3CN, 1.5 h
CH3CN, 1.5 h
80
8
55
58
60
53
58
79
52
60
a Li–BIT: Lithium 2-benzimidazolethiolate; Li–NPT: Lithium 2-naphthalenethiolate.

K. Shinhama et al. / Tetrahedron 56 (2000) 7427–7431
7429
Figure 1. Cyclization of N,O-diprotected ester (2) with lithium 2-benzimidiazolethiolate (2.5 equiv.) and LiOH (1.0 equiv.) in DMF.
We then assumed that at least one equivalent of the reagent
was consumed in the formation of the stable product,
2-cyanoethyl sulfide. Another possibility was some kind
of complexation with the substrate (as indicated in Scheme
2). In fact, generation of the lithium thiolate for the catalytic
path was not a feasible process. A plausible reaction path-
way was proposed to understand the role of the lithium
thiolate (as shown in Scheme 2). Thus, the reaction started
with C–O bond fission in the CE group to form the oxime
anion under basic conditions. This was followed by the
controlled nucleophilic attack of a thiolate anion on the car-
bonyl carbon to reach to the key intermediate shown.
Another more simple description might be possible, but
we believe, the final step is most likely an intramolecular
nucleophilic attack of the oxime nitrogen on the carbonyl
carbon of the thiol ester. Several unsuccessful attempts have
been made to isolate this intermediate thiol ester so far.
Besides these inspections, it is interesting to note here the
reactivity of the thiolates towards alkyl esters. The reaction
described here is not similar to the reported patterns of
lithium alkylthiolates. Lithium alkylthiolates such as
lithium propanethiolate are reported to cleave esters into
the corresponding acids4,5 In the present cyclization, we
believe, lithium arylthiolates could behave like aluminium
arylthiolates,5 which can produce thioesters in a single step.
Thus, lithium arylthiolates are different from lithium
alkylthiolates in their behavior towards esters.
Additionally, we observed that sodium arylthiolates were
Scheme 2. Thiolate-mediated ring closure from 2 to 4.

7430
K. Shinhama et al. / Tetrahedron 56 (2000) 7427–7431
ineffective for the conversion of 2 into 4, but did promote
the cyclization from 3a as shown in Table 1.
Jˆ6.8 Hz, 2.1H), 0.91 (d, Jˆ6.8 Hz, 2.1H), 1.53–1.61 (m,
0.6H), 1.82–2.14 (m, 1H), 2.19 (dt, Jˆ4.3, 1.8 Hz, 1.4H),
2.42–2.57 (m, 1.4H), 2.64 (dt, Jˆ6.0, 2.2 Hz, 0.6H), 2.72 (t,
Jˆ6.0 Hz, 1.4H), 2.87 (dt, Jˆ4.5, 2.5 Hz, 0.6H), 3.18–3.33
(m, 1H), 3.43–3.67 (m, 2.4H), 3.81–3.87 (m, 3.6H), 4.05
(dt, Jˆ6.0, 2.2 Hz, 0.6H), 4.25 (t, Jˆ6.0 Hz, 1.4H), 4.39
(dd, Jˆ10.4, 5.0 Hz, 0.7H), 5.20 (dd, Jˆ10.0, 4.7 Hz,
0.3H), 6.99–7.17 (m, 3H), and 7.34–7.39 ppm (m, 1H).
Anal. Calcd for C24H29N5O4: C, 63.84; H, 6.47; N, 15.51.
Found: C, 63.82; H, 6.41; N, 15.41.
The subtle assistance of a lithium ion associated with the
steric surroundings of 2 might be key factors involved in this
cyclization reaction, which will be the subject of our further
research.
Summary
In the quest for an efficient synthetic pathway to the dioxy-
genated pyrazines, we developed here a novel thiolate-
mediated ring closure. The key cyclization is triggered by
the deprotection of one CE group by a lithium arylthiolate.
The subsequent cyclization to pyrazine skeleton was
presumably assisted by a lithium cation. Thus, the penulti-
mate intermediate (5) was prepared in a one-pot procedure
by treating the precursor methyl ester (2) with lithium
2-naphthalenethiolate followed by methylation with
dimethyl sulfate in 37% yield. The resulting pyrazinone
(5) was successfully converted to OPC-15161 (1) by the
treatment with sodium hydroxide in DMF–methanol. Our
process research here again revealed the important and
crucial assistance of a protecting group for a new type of
ring closure to inaccessible pyrazinones as well as some
synthetic potential of a thiolate anion for attaining one-pot
process.
N-(2-Cyanoethyl)-N-(2-hydroxyiminoisohexanoyl)-l-tryp-
tophan methyl ester (3a). A mixture of 4N-sodium
hydroxide aq. solution (0.75 ml), 2 (0.45 g, 1.00 mmol),
and THF (20 ml) was stirred at room temperature for 15 h.
After addition of DMF (20 ml) and boric acid (124 mg,
2.01 mmol), the solution was stirred further for 30 min.
Then dimethyl sulfate (0.40 ml, 4.23 mmol) was added to
the solution and stirred further for 2 h. The solution was
diluted with ethyl acetate, washed with saturated aq. sodium
hydrogen sulfate and saturated aq. sodium chloride, dried
over MgSO4, and concentrated. Purification by column
chromatography (silica gel, dichloromethane/methanolˆ
99:1) gave pure 3a (0.26 g, 65%); mp 122–124ЊC. 1H
NMR (CDCl3): 0.73 (m, 1.5H), 0.91 (m, 4.5H), 3.71 (s,
0.75H), 3.76 (s, 2.25H), 4.18 (dd, Jˆ10.8 Hz, 0.25H),
6.9–7.4 (m, 4H), 7.56 (d, Jˆ8 Hz, 1H), 8.00 (s, 0.75H),
8.90 (s, 0.25H), and 9.41 ppm (s, 0.25H). Anal. Calcd for
C21H26N4O4·0.3H2O: C, 62.44; H, 6.44; N, 13.87. Found: C,
62.43; H, 6.50, N, 13.96.
Experimental
N-(2-Cyanoethyl)-N-(2-hydroxyiminoisohexanoyl)-l-tryp-
tophan benzyl ester (3b). Benzyl ester 3b (2.11 g, 44%)
was obtained from 2 by using benzyl bromide (2 equiv
based on 2) instead of dimethyl sulfate in the same manner
All the melting points were uncorrected. Analytical deter-
minations by HPLC were performed on a Shimadzu LC-6A
liquid chromatography with a TSK gel ODS-80TM column.
1H NMR spectra were taken at Varian 200 MHz spectro-
meter. IR spectra were recorded with a Perkin–Elmer 1600
series FTIR apparatus. Mass spectra were recorded with a
Shimadzu GCMS-QP1000 spectrometer at 70 eV.
1
as in the above experiment. H NMR (CDCl3): 0.68 (d,
Jˆ6.4 Hz, 2.4H), 0.85 (d, Jˆ6.6 Hz, 3.6H), 1.6–2.3 (m,
5H), 2.5 (m, 1.2H), 2.7 (t, Jˆ6 Hz, 0.8H), 3.1–3.8 (m,
4.4H), 4.2 (t, Jˆ9 Hz, 0.6H), 5.19 (s, 2H), 7.34 (s, 5H),
6.9–7.4 (m, 4H), 7.58 (d, Jˆ10 Hz, 1H), and 8.25 ppm
(br, s, 1H). MS m/z (rel. int.): 474 (Mϩ, 1), 201 (20), and
130 (100). Anal. Calcd for C27H30N4O4: C, 68.34; H, 6.37;
N, 11.81. Found: C, 68.16; H, 6.48, N, 11.64.
Metal thiolates
Metal thiolates used in the following experiments were
prepared from the corresponding thiols by usual methods
(from ArSH and LiOH).5 They were also prepared in situ
by treating thiols with one or more equivalents of anhydrous
lithium hydroxide in dry DMF.
1-(2-Cyanoethyl)-3-isobutyl-5-hydroxy-6-(indol-3-yl)-
methyl-1,2-dihydropyrazin-2-one 4-oxide (4). Method A:
A mixture of 2-mercaptobenzimidazole (6.67 g, 44.4
mmol), anhydrous lithium hydroxide (1.60 g, 66.6 mmol),
and dry DMF (150 ml) was stirred at room temperature for
30 min. After addition of 2 (10.0 g, 22.2 mmol), the mixture
was stirred further for 3 h at room temperature, then diluted
with ethyl acetate (500 ml), and poured into ice-cold hydro-
chloric acid (500 ml). The organic layer was separated,
washed with water, and extracted with aq. sodium hydrogen
carbonate, which was then acidified with hydrochloric acid,
and extracted with ethyl acetate. The organic extract was
washed with water, dried over MgSO4, and concentrated in
vacuo at room temperature to give crude 4 (5.03 g, 62%) as
an unstable yellow amorphous solid. IR (KBr): 3310, 1600,
N-[2-(2-Cyanoethoxyimino)isohexanoyl]-N-(2-cyanoethyl)-
l-tryptophan methyl ester (2). 2-(2-Cyanoethoxyimino)-
isohexanoyl chloride (2.14 kg, 9.88 mol) was added to a
stirred solution of N-(2-cyanoethyl)-l-tryptophan methyl
ester (1.44 kg, 4.72 mol) in dichloromethane (12 l) during
1 h at 27–28ЊC, and the solution was stirred further for
35 min at 22–25ЊC. Then aq. potassium carbonate (2.4 kg
in 3 l of water) was added to the solution. The organic layer
was separated, washed with aq. potassium carbonate, 20%
hydrochloric acid, and saturated aq. sodium chloride, dried
over MgSO4, concentrated, and recrystallized from 60% aq.
i-PrOH to give 2 (1340 g, 63%); mp 102–103ЊC (recrystal-
lized from acetonitrile). IR (neat): 3398, 2961, 2250, 1728,
1
1174, 1099, and 746 cmϪ1. H NMR (CDCl3): 0.97 (d,
Jˆ7 Hz, 6H), 2.18–2.40 (m, 1H), 2.48 (t, Jˆ6 Hz, 2H),
2.89 (d, Jˆ7 Hz, 2H), 3.0–4.0 (br s, 1H), 4.21 (t, Jˆ6 Hz,
2H), 4.36 (s, 2H), 7.03–7.63 (m, 5H), and 8.30 ppm (br s,
1
1634, 1435, 1266, 961, and 744 cmϪ1. H NMR (CDCl3):
0.61 (d, Jˆ6.8 Hz, 0.9H), 0.65 (d, Jˆ6.8 Hz, 0.9H), 0.90 (d,

K. Shinhama et al. / Tetrahedron 56 (2000) 7427–7431
7431
1H). MS m/z (rel. int.): 366 (Mϩ, 3), 350 (15), and 130
(100).
pyrazin-2-one 4-oxide (OPC-15161) (1). A mixture of 5
(2.16 g, 5.68 mmol), sodium hydroxide (1.50 g, 37.5
mmol), DMF (10 ml), and methanol (10 ml) was stirred at
room temperature for 30 min. The mixture was poured into
cold water, and acidified with hydrochloric acid. The result-
ing crystals were collected by filtration and dried in vacuo to
give 1 (1.79 g, 96%); mp 225ЊC. IR (KBr): 3246, 1627,
Method B: A mixture of 3b (40.9 mg, 0.0862 mmol),
lithium 2-benzimidazolethiolate (40.0 mg, 0.256 mmol),
and dry DMF (0.4 ml) was stirred for 2 h at room tempera-
ture. The yield of 4 (79%) was determined by HPLC
analysis based on the calibration curve.
1
1373, 1247, 1100, 998, 744, 719, and 521 cmϪ1. H NMR
(DMSO): 0.84 (d, Jˆ6.6 Hz, 6H), 1.97–2.12 (m, 1H), 2.60
(d, Jˆ7.1 Hz, 2H), 3.76 (s, 3H), 3.91 (s, 2H), 6.93–7.59 (m,
5H), 10.97 (br s, 1H). These spectroscopic data were
identical with those of authentic sample.
1-(2-Cyanoethyl)-3-isobutyl-5-methoxy-6-(indol-3-yl)-
methyl-1,2-dihydropyrazin- 2-one 4-oxide (5). Method A:
A mixture of 2-naphthalenethiol (8.52 g, 53.2 mmol),
anhydrous lithium hydroxide (1.27 g, 53.2 mmol), and dry
DMF (100 ml) was stirred for 15 min at room temperature.
Then a solution of 2 (12.0 g, 22.6 mmol) in DMF (40 ml)
was added and the mixture was stirred further for 2.3 h at
room temperature. After addition of powdered anhydrous
potassium carbonate (18.4 g, 133 mmol) and dimethyl
sulfate (12.6 ml, 133 mmol), the mixture was stirred further
for 1.8 h, poured into cold water, then extracted with ethyl
acetate. The extract was washed with water, dried over
MgSO4, and concentrated to ca. 1/4 volume. The resulting
precipitate was separated by filtration and dried in vacuo to
give 5 (3.75 g, 37%); mp 223–225ЊC. IR (neat): 3320, 2251,
1659, 1614, 1259, 1614, 1259, 910, and 733 cmϪ1. 1H NMR
(CDCl3): 0.96 (d, Jˆ7 Hz, 6H), 2.18–2.40 (m, 1H), 2.53 (t,
Jˆ7 Hz, 2H), 2.85 (d, Jˆ7 Hz, 2H), 4.00 (s, 3H), 4.11 (t,
Jˆ7 Hz, 2H), 4.31 (s, 2H), 6.90–7.65 (m, 5H), and
8.68 ppm (br, s, 1H). Anal. Calcd for C21H24N4O3: C,
66.30; H, 6.36; N, 14.73. Found: C, 66.36; H, 6.55; N, 14.60.
Acknowledgements
The authors wish to thank Professor T. Mukaiyama of
Science University of Tokyo and Emeritus Professor Sei
Otsuka of Osaka University for their helpful advice.
References
1. Nakano, Y.; Kawaguchi, T.; Sumitomo, J.; Takigawa, T.;
Uetsuki, S.; Sugawara, M.; Kido, M. J. Antibiot. 1991, 44, 52.
2. (a) Ito, Y.; Sato, H.; Murakami, M. J. Org. Chem. 1991, 56,
4864. (b) Jing, H.; Murakami, K.; Aoyagi, Y.; Ohta, A. Hetero-
cycles 1992, 34, 1847. (c) Kita, Y.; Akai, S.; Fujioka, H.;
Taniguchi, Y. Tetrahedron Lett. 1991, 32, 6019. (d) Kita, Y.;
Akai, S.; Fujioka, H.; Tamura, Y.; Tone, H.; Taniguchi, Y.
J. Chem. Soc., Perkin Trans. 1 1994, 875.
3. (a) Tone, H.; Matoba, K.; Goto, F.; Torisawa, Y.; Nishi, T.;
Minamikawa, J. Org. Proc. Res. Dev. 2000, in press. (b) Matoba,
K.; Tone, H.; Shinhama, K.; Goto, F.; Saka, M; Minamikawa, J.
J. Synth. Org. Chem. Jpn 1999, 57, 407.
4. (a) Vaughan, W. R.; Baumann, J. B. J. Org. Chem. 1962, 27,
739. (b) Sheehan, J. C.; Daves, G. D. J. Org. Chem. 1964, 29, 2006.
(c) Bartlett, P. A.; Johnson, W. S. Tetrahedron Lett. 1970, 4459.
5. See for example: Encyclopedia of Reagents for Organic
Synthesis; Paquette, L. A.; Ed., Wiley: New York, 1995; pp 169;
1784–1785; 3161–3163.
Method B: Dimethyl sulfate (64 mg, 0.51 mmol) and
powdered anhydrous potassium carbonate (70 mg,
0.51 mmol) were added to a stirred solution of 4 (93.5 mg,
0.255 mmol) in dry DMF (1.3 ml). The mixture was stirred
at room temperature for 20 min, then diluted with ethyl
acetate (10 ml), washed with cold water, dried over
MgSO4, and concentrated in vacuo. The residue was
purified by TLC (silica gel, chloroform/methanolˆ20:1) to
give 5 (58.8 mg, 61%).
6-(Indol-3-yl)methyl-3-isobutyl-5-methoxy-1,2-dihydro-