Efficient synthesis of substituted 2-aminopyrazines: FeCl3-promoted condensation of hydroxyiminoketones with aminoacetonitriles

Article abstract of DOI:10.1016/S0040-4039(02)02375-4

FeCl₃-promoted condensation of hydroxyiminoketones with aminoacetonitriles followed by catalytic hydrogenation afforded the desired pyrazines in moderate-good yields. This protocol provides an efficient and practical synthesis of substituted 2-

Full text of DOI:10.1016/S0040-4039(02)02375-4

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 9287–9290
Efficient synthesis of substituted 2-aminopyrazines:
FeCl3-promoted condensation of hydroxyiminoketones
with aminoacetonitriles
Takahiro Itoh,* Kenji Maeda, Toshihiro Wada, Koji Tomimoto and Toshiaki Mase
Process R & D, Laboratories for Technology Development, Banyu Pharmaceutical Co. Ltd, Kamimutsuna 3-Chome-9-1,
Okazaki, Aichi 444-0858, Japan
Received 25 September 2002; revised 17 October 2002; accepted 24 October 2002
Abstract—FeCl3-promoted condensation of hydroxyiminoketones with aminoacetonitriles followed by catalytic hydrogenation
afforded the desired pyrazines in moderate–good yields. This protocol provides an efficient and practical synthesis of substituted
2-aminopyrazines. © 2002 Elsevier Science Ltd. All rights reserved.
Substituted pyrazines have gained increased attention
in recent years due to their usefulness as important
constituents either of biologically active compounds1 or
functional materials.2 In particular, substituted 2-
aminopyrazines are key synthetic intermediates to the
luciferins and their analogues,3 which have long been
known as chemiluminescent and/or bioluminescent
agents.4,5 Irrespective of their importance, an efficient
synthetic route to these key pyrazine intermediates has
thus far not been fully developed. Of the several
reported aminopyrazine preparation to date,6 direct
condensation of hydroxyiminoketones with aminoace-
tonitriles followed by reduction appears attractive;
however, poor yields in the initial condensation step
reduces the efficiency of this approach.6,7 Stoichiometric
TiCl4 has been reported to accelerate the condensation
with modest yield improvements.8,9 These improve-
ments, however, were still unsatisfactory for our
preparative purposes and were highly dependent on the
nature of the substituents on the product aminopy-
razine.6c Furthermore, stoichiometric or excess use of
TiCl4 would not be practical for large-scale preparation
due to its handling difficulty and environmentally-ill
waste stream. In this article, we wish to report a novel,
practical and economical synthesis of substituted 2-
aminopyrazines by an FeCl3-induced condensation of
hydroxyiminoketones with aminoacetonitriles.
It has been postulated that Lewis acid such as TiCl4
promotes not only formation of the Schiff base but also
helps some of enolization of the resulting b-iminonitrile
3 to generate a reactive species such as aza-allene
intermediate 4. Cyclization of 4 should then occur
relatively easily to afford the desired pyrazine N-oxide.
This hypothesis can also be supported by the fact that
the reaction of 1 with aminomalononitrile 2 (R3=CN)
proceeds smoothly without assistance by Lewis acid
(Scheme 1).6c,d
Therefore, development of a reagent, which can be
more effective for accelerating both Schiff base forma-
tion and, in particular, enolization of iminonitrile 3
than TiCl4, appears to be the key for successful conden-
sation. We thus sought to find more effective conditions
that would further accelerate both functions.
Recently Christoffers reported the FeCl3-catalyzed
Michael reactions of 1,3-dicarbonyl compounds with
enones, presumably involving formation of stable 1,3-
* Corresponding author. Tel.: +81-564-51-5668; fax: +81-564-51-7086;
e-mail: itouta@banyu.co.jp
Scheme 1.
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)02375-4

9288
T. Itoh et al. / Tetrahedron Letters 43 (2002) 9287–9290
Table 1. FeCl3-induced condensation of a-ketoaldoximes
with aminoacetonitriles
dionato iron chelate complexes which further activated
the enones toward addition.10 Interestingly and
uniquely, in the presence of Fe(III) these complexes
were instantly formed even under strongly Brønsted-
acidic media without prior deprotonation. Thus, we
envisioned that Fe(III) salts might promote not only
formation but enolization of 3 to afford 4 effectively
which would then cyclize to the desired pyrazine N-
oxides 5.
In fact, after neutralizing the aminoacetonitrile hydro-
chloride salt with NaOH, treatment of isonitrosoace-
tophenone 1a with 2a in MeOH–H2O (24:1) at 80°C in
the presence of anhydrous FeCl3 (1 equiv.) afforded the
N-oxide 5a6a in 85% yield (Scheme 2). This condensa-
tion without FeCl3 resulted in only 14% yield of 5a
according to the procedure of Sharp and Spring,6a,7 and
in 10% yield even TiCl4 at 0°C. On the other hand,
acidic conditions6e,f (TsOH, H2SO4 and HCl) gave mis-
erable results.
Entry
Oximes 1
Aminonitriles 2 Yielda of 6
(%)
R1
Ph
R2
R3
1
2
3
4
5
6
7
H
1a
1a
1a
1a
1a
1b
1c
H
2a
2b
2c
2d
2e
2a
2a
80
63
67
58
72
60
55
6a
6b
6c
6d
6e
6f
Me
Et
Ph
Bn
H
Other alcohols and DMF were also acceptable as sol-
vents, although aqueous MeOH afforded a homoge-
neous solution. Sodium carbonate and amine bases
such as N-methylmorpholine and diisopropylethyl-
amine were also tolerable for breaking the salt of
aminoacetonitrile.
Ph
Me
Me
H
H
6g
a Yields refer to the average isolated yield of two runs.
which is more difficult to control,7 also worked rela-
tively smoothly to afford the corresponding pyrazine
6d17 in ca. 60% yield (entry 4). For the benzyl substi-
tuted aminoacetonitrile, 72% of 6e18 was obtained by
the present method, while ca. 60% of 2-amino-3-benzyl-
5-(p-methoxy-phenyl)-6-methylpyrazine was produced
with known TiCl4 sequence (entry 5).19 In addition to
entry 1, condensations with unsubstituted aminoace-
tonitrile 2a, which are known to be difficult due to its
instability,6a,16 were also evaluated. Two different
oximes 1b and 1c were converted to the corresponding
pyrazines 6f and 6g20 in 60 and 55% yield, respectively
(entries 6 and 7).
A variety of other metal chlorides were screened for
promotion of this reaction sequence; however, FeCl3
was proven to be the optimal promoter for this particu-
lar reaction.11 FeCl2 was much less effective than FeCl3.
Other Fe(III) salts such as Fe(NO3)3 and Fe2(SO4)3
were also acceptable for this reaction.12 Anhydrous
FeCl3 was compared with its hydrate in MeOH and
they were not distinguishable.13
Table 1 provides results on the FeCl3-induced conden-
sation of a-ketoaldoximes with aminoacetonitriles fol-
lowed by hydrogenation with 10% Pd–C in one flask
under our typical conditions,14 and also using Zn/
aqueous NH4Cl gave same result. Yields of pyrazines
after hydrogenation are listed.15 Reaction of isonitro-
soacetophenone 1a with unsubstituted aminoacetoni-
trile 2a afforded the desired aminopyrazine in 80%
yield. Condensation of 1a with a-substituted aminoace-
tonitriles was also attempted. The reaction with 2-
aminopropionitrile or 2-aminobutyronitrile 2b or 2c
afforded the corresponding pyrazines in 63 and 67%
yield, respectively (entries 2 and 3). It has also been
reported that the 2-amino-3-methyl-5-phenylpyrazine
6b was obtained in only 40% yield even when TiCl4 was
used.8,16 The reaction with 2-phenylglycinonitrile 2d,
It was found that when the reaction was stirred with
excess aminoacetonitrile (>5 equiv.) at higher tempera-
ture (reflux temp.) overnight, the product was the corre-
sponding pyrazine and not the N-oxide (Table 2).
Table 2.
Entry
Oximes 1
Yielda of 6 (%)
R1
R2
1
2
3
Ph
Ph
Me
H
Me
H
1a
1b
1c
72
68
65
6a
6f
6g
Scheme 2.
a Yields refer to the average isolated yield of two runs.

T. Itoh et al. / Tetrahedron Letters 43 (2002) 9287–9290
9289
Some results were slightly better than previous results
of stepwise reaction. In contrast, treatment with 2
equiv. of aminoacetonitrile at 50°C gave only the N-
oxide. These results indicate that direct reduction of the
N-oxide intermediate, to spontaneously produce the
desired pyrazine, can also be promoted under these
reaction conditions.
5. (a) Campbell, A. K. Principle and Applications in Biology
and Medicine; Ellis Harwood: Chichester, UK, 1998; (b)
Shimomura, O.; Inouye, S.; Musicki, B.; Kishi, Y.
Biochem. J. 1990, 270, 309–312; (c) Ohmiya, Y.; Hirano,
T. Chem. Biol. 1996, 3, 337–347.
6. (a) Sugiura, S.; Inoue, S.; Kishi, Y.; Goto, T. Yakugaku
Zasshi 1969, 89, 1646–1651; (b) Sato, N. J. Heterocyclic
Chem. 1978, 15, 665–670; (c) Neilsen, J. B.; Broadbent,
H. S.; Hennen, W. J. J. Heterocyclic Chem. 1987, 24,
1621–1628; (d) Newbold, G. T.; Spring, F. S.; Sweeny,
W. J. Chem. Soc. 1948, 1855–1859; (e) Taylor, E. C.;
Dumas, D. J. J. Org. Chem. 1981, 46, 1394–1402; (f)
Taylor, E. C.; Dumas, D. J. J. Org. Chem. 1981, 46,
116–119.
Furthermore, when 0.1 equiv. of FeCl3 was used, a
prolonged reaction period gave the N-oxide 5a in ca.
60% yield, indicating that FeCl3 can work as a catalyst.
However, the reaction did not lead to completion even
after stirring for 2 days. The reason is still unclear but
might be due to a lower valent iron generated from
reduction of FeCl3 with aminoacetonitrile.21
7. Sharp, W.; Spring, F. S. J. Chem. Soc. 1951, 932–934.
8. Karpetsky, T. P.; White, E. H. J. Am. Chem. Soc. 1971,
93, 2333–2335.
In conclusion, FeCl3 promotes condensation of hydroxy-
iminoketones with aminoacetonitriles to give the
desired pyrazine N-oxides, which are readily reduced
to afford the desired pyrazines in reasonable yields.
This protocol provides efficient and practical access
to a variety of 3-mono or 3,5-disubstituted 2-amino-
pyrazines.
9. Very recently, Kakoi reported that the condensation was
dramatically improved with 5 equiv. of TiCl4 in pyridine
at −5°C after our study was completed, see: Kakoi, H.
Chem. Pharm. Bull. 2002, 50, 301–302. Our study would
be conceptually distinguishable from the work with TiCl4.
Results given with FeCl3 are also equal or superior to
those with TiCl4. From a process chemistry perspective,
our condition is more practical due to ease of handling
and an environmentally friendly waste stream.
Acknowledgements
10. (a) Christoffers, J. Chem. Commun. 1997, 10, 943–944; (b)
Christoffers, J. J. Chem. Soc., Perkin Trans. 1 1997, 21,
3141–3149; (c) Christoffers, J. Eur. J. Org. Chem. 1998, 7,
1259–1266; (d) Christoffers, J. Synlett 2001, 6, 723–732.
11. Reactions with LiCl, MgCl2, AlCl3, TiCl4, NiCl2, CuCl,
CuCl2, ZnCl2, RuCl3, SnCl2 and CeCl3 were tested:
results were not distinguishable with the one from the
reaction without any additive.
The authors are grateful to Dr. R. P. Volante, Merck &
Co., Inc., for critical reading of this manuscript.
References
1. (a) Bigham, E. C.; Smith, G. K.; Reinhard, J. F., Jr.;
Mallory, W. R.; Nichol, C. A.; Morrison, R. W., Jr. J.
Med. Chem. 1987, 30, 40–45; (b) Abignente, E.; De
Caprariis, P.; Rimoli, M. G.; Capasso, F.; Autore, G.
Farmaco 1992, 47, 919–930; (c) Ferrand, G.; Dumas, H.;
Decerprit, J. Eur. J. Med. Chem. 1992, 27, 309–320; (d)
Foks, H.; Wisterowicz, K.; Damasiewicz, B.; Nasal, A.;
Petrusewicz, J.; Radwanska, A. Acta Pol. Pharm. 1997,
54, 55–62; (e) Miranda, F.; Hong, E.; Velazquez-Mar-
tinez, D. N. Pharmacol. Biochem. Behav. 2001, 68, 427–
433; (f) Quirk, K.; Lawrence, A.; Jones, J.; Misra, A.;
Harvey, V.; Lamb, H.; Revell, D.; Porter, R. H. P.;
Knight, A. R. Eur. J. Pharmacol. 2001, 107–112; (g)
Rimoli, M. G.; Avallone, L.; de Caprariis, P.; Luraschi,
E.; Abignente, E.; Filippelli, W.; Berrino, L.; Rossi, F.
Eur. J. Med. Chem. 1997, 32, 195–203.
2. Zhang, C. Y.; Tour, J. M. J. Am. Chem. Soc. 1999, 121,
8783–8790 and references cited therein.
3. Toya, Y.; Kayano, T.; Sato, K.; Goto, T. Bull. Chem.
Soc. Jpn. 1992, 65, 2475–2479.
4. (a) Nakano, M.; Sugioka, K.; Ushijima, Y.; Goto, T.
Anal. Biochem. 1986, 159, 363–369; (b) Koga, S.;
Nakano, M.; Uehara, K. Arch. Biochem. Biophys. 1991,
289, 223–229; (c) Sakurai, T.; Sugioka, K.; Nakano, M.
Biochim. Biophys. Acta 1990, 1043, 27–33; (d) Takahashi,
A.; Nakano, M.; Mashiko, S.; Inaba, H. FEBS Lett.
1990, 261, 369–372; (e) Nishida, A.; Kimura, H.;
Nakano, M.; Goto, T. Clin. Chim. Acta 1989, 179, 177–
181.
12. Fe(III) citrate was less effective.
13. Powdered anhydrous FeCl3 was used with water since the
hydrate was a stony solid and difficult to handle.
14. Typical procedure. The preparation of 2-amino-5-
phenylpyrazine (6a). All experiments were operated under
a nitrogen atmosphere. 12N aqueous NaOH (123 mL,
1.48 mol) was added to a mixture of aminoacetonitrile
hydrochloride (124 g, 1.34 mol) in methanol (4 L), and
then isonitrosoacetophenone (100 g, 0.67 mol) and ferric
chloride (109 g, 0.67 mol) were added to the resulting
solution below 20°C. The resulting mixture was stirred at
50°C for 2 h followed by stirring at reflux for 4 h. The
reaction mixture was then cooled to ambient temperature
and treated with palladium on carbon (10 w/w%) under a
pressure of hydrogen (5.0 atm) at 50°C for 18 h. After
checking consumption of N-oxide by HPLC, the reaction
mixture was basified with 12N aqueous NaOH (pHꢀ10)
and filtered through a plug of Celite. The filtrate was
concentrated to remove MeOH under reduced pressure.
The aqueous layer was extracted with EtOAc twice. The
combined organic layers were washed with 7% aqueous
NaCl and dried (Na2SO4). The filtered EtOAc solution
was concentrated under reduced pressure to 1 L, pre-
treated with an activated carbon (Darco G-60) for 1 h at
ambient temperature and filtered through Celite. The
filter cake (activated carbon) was washed with EtOAc.
The filtrate was concentrated to ca. 100 mL under
reduced pressure which initiated crystallization of the
product, and then the residue was stirred at ambient

9290
T. Itoh et al. / Tetrahedron Letters 43 (2002) 9287–9290
temperature. To the resulting slurry was added dropwise
128.8, 128.9, 129.0, 136.9, 137.5, 137.8, 137.9, 139.7,
152.0.
n-heptane, and the resulting mixture was agitated for 1 h
at ambient temperature. The solid product was collected
by filtration, washed with n-heptane/EtOAc=5:1 (200
mL) and dried in vacuo under nitrogen sweep at 30°C
overnight to afford 6a (77 g, 67%) as a yellow solid.
HPLC analysis showed that 6a in the mother liquor was
obtained in 13% yield (15 g).
Compound 6e: 1H NMR (270 MHz, DMSO, ppm) l 8.44
(s, 1H), 7.92 (d, d=7.3 Hz, 2H), 7.44–7.17 (m, 8H), 6.41
(s, 2H), 4.11 (s, 2H); 13C NMR (67.5 MHz, DMSO, ppm)
l 124.7, 126.1, 127.3, 128.2, 128.6, 128.9, 136.9, 137.1,
138.1, 138.7, 140.0, 152.7.
1
Compound 6f: H NMR (270 MHz, DMSO, ppm) l 7.82
Compound 5a: 1H NMR (500 MHz, DMSO, ppm) l 8.79
(s, 1H), 8.22 (s, 1H), 7.96 (d, J=7.5 Hz, 2H), 7.44 (dd,
J=7.3, 7.5 Hz, 2H), 7.38 (dd, J=7.3 Hz, 1H), 7.11 (s,
2H).
(s, 1H), 7.51 (d, J=6.6 Hz, 1H), 7.43 (d, J=6.6 Hz, 1H),
7.41 (d, J=7.9 Hz, 1H), 7.34 (d, J=6.6 Hz, 1H), 7.32 (d,
J=7.6 Hz, 1H), 6.40 (s, 2H), 2.35 (s, 3H); 13C NMR
(67.5 MHz, DMSO, ppm) l 22.7, 127.2, 128.2, 129.1,
129.2, 139.7, 140.1, 147.8, 154.2.
Compound 6a: 1H NMR (500 MHz, DMSO, ppm) l 8.53
(s, 1H), 8.01 (s, 1H), 7.94 (d, J=7.6 Hz, 1H), 7.93 (d,
J=7.6 Hz, 1H), 7.43 (dd, J=7.4, 7.6 Hz, 2H), 7.32 (dd,
J=7.4 Hz, 1H), 6.59 (s, 1H); 13C NMR (125 MHz,
DMSO, ppm) l 125.0, 127.7, 129.0, 131.8, 137.5, 139.2,
139.4, 155.3.
Compound 6g: 1H NMR (500 MHz, DMSO, ppm) l 7.81
(s, 1H), 7.78 (s, 1H), 6.10 (s, 2H), 2.25 (s, 3H); 13C NMR
(125 MHz, DMSO, ppm) l 20.0, 131.4, 139.6, 140.7,
154.4, 176.5.
15. It is comparatively difficult to isolate pure pyrazine N-
oxides. As hydrogenation of the N-oxide normally pro-
ceeds very cleanly to give the desired pyrazines in high
yield (>90%), yields given in Table 1 should reflect to
yields for N-oxides.
16. Karpetsky, T. P.; White, E. H. Tetrahedron 1973, 29,
3761–3773.
17. Sato, N.; Matsuura, T.; Miwa, N. Synthesis 1994, 9,
931–934.
18. Saito, R.; Hirano, T.; Niwa, H.; Ohashi, M. J. Chem.
Soc., Perkin Trans. 2 1997, 9, 1711–1716.
19. Teranishi, K.; Goto, T. Bull. Chem. Soc. Jpn. 1990, 63,
3132–3140.
Compound 6b: 1H NMR (500 MHz, DMSO, ppm) l 8.40
(s, 1H), 7.94 (d, J=7.4, 8.3 Hz, 2H), 7.43 (d, J=7.6 Hz,
1H), 7.41 (d, J=7.8 Hz, 1H), 7.31 (d, J=7.2 Hz, 1H),
6.35 (s, 2H), 2.40 (s, 3H); 13C NMR (125 MHz, DMSO,
ppm) l 21.1, 125.1, 127.6, 129.0, 136.8, 137.6, 138.8,
138.9, 153.6, 176.5.
1
Compound 6c: H NMR (270 MHz, CDCl3, ppm) l 8.32
(s, 1H), 7.93 (d, J=7.3, 8.6 Hz, 2H), 7.45 (d, J=7.3 Hz,
1H), 7.42 (d, J=7.6 Hz, 1H), 7.34 (dd, J=6.9, 7.6 Hz,
1H), 4.61 (s, 2H), 2.73 (q, J=7.3, 7.6 Hz, 2H), 1.41 (t,
J=7.3, 7.6 Hz, 3H); 13C NMR (125 MHz, CDCl3, ppm)
l 11.0, 26.0, 125.0, 127.5, 129.0, 136.5, 137.8, 138.8,
142.5, 153.0.
Compound 6d: 1H NMR (270 MHz, DMSO, ppm) l 8.57
(s, 1H), 8.00 (d, J=7.3, 8.6 Hz, 2H), 7.82 (d, J=6.9, 8.3
Hz, 2H), 7.57–7.47 (m, 4H), 7.43 (d, J=7.6 Hz, 1H), 7.34
(dd, J=6.9, 7.6 Hz, 1H), 6.31 (s, 2H); 13C NMR (67.5
MHz, DMSO, ppm) l 124.9, 127.8, 128.1, 128.5, 128.6,
20. Sugiura, S.; Kakoi, H.; Inoue, S.; Goto, T. Yakugaku
Zasshi 1970, 90, 441–444.
21. Treatment of the oxide 5a with FeCl2 did not lead to the
corresponding pyrazine 6a while with Fe powder did
work, giving 6a.