Highly efficient synthesis of fused bicyclic 2,3-diaryl-pyrimidin-4(3H)-ones via Lewis acid assisted cyclization reaction

Article abstract of DOI:10.1016/j.tetlet.2008.01.071

An expedient one-pot synthesis of fused bicyclic 2,3-diaryl-pyrimidin-4(3H)-ones from three readily available components is described. The key step is a Lewis acid assisted cyclization reaction.

Full text of DOI:10.1016/j.tetlet.2008.01.071

Available online at www.sciencedirect.com
Tetrahedron Letters 49 (2008) 1725–1728
Highly efficient synthesis of fused bicyclic 2,3-diaryl-pyrimidin-
4(3H)-ones via Lewis acid assisted cyclization reaction
Kunyong Yang *, Xiaohui He, Ha-soon Choi, Zhicheng Wang,
David H. Woodmansee, Hong Liu
Medicinal Chemistry, Genomics Institute of the Novartis Research Foundation, 10675 John Jay Hopkins Drive, San Diego, CA 92121, USA
Received 29 November 2007; revised 16 January 2008; accepted 16 January 2008
Available online 20 January 2008
Abstract
An expedient one-pot synthesis of fused bicyclic 2,3-diaryl-pyrimidin-4(3H)-ones from three readily available components is
described. The key step is a Lewis acid assisted cyclization reaction.
Ó 2008 Elsevier Ltd. All rights reserved.
Fused bicyclic pyrimidin-4(3H)-ones (Fig. 1), such as
1H-purin-6(9H)-one 11 and 1H-pyrazolo[3,4-d]pyrimidin-
4(5H)-one 2,2 are important core structures to many bio-
logically active small molecules. As part of an ongoing
project, we were interested in developing an efficient method
to construct fused bicyclic 2,3-diaryl-pyrimidin-4(3H)-ones
3. As a requirement to our lead optimization program, we
desired a short route from the readily accessible starting
materials that was flexible enough to allow the exploration
of the C2- and N3-positions as well as to the fused hetero-
aromatic ring. Herein, we report a new synthetic strategy
for the synthesis of fused bicyclic 2,3-diaryl-pyrimidin-
4(3H)-ones.
reported (Scheme 1).3 Both started with a cyclic amino acid
derivative 4, but differed in the order of introducing the C2-
and N3-substituents. Approach A introduced the N3-sub-
stituent first by converting 4 into amino amide 5, which
was then converted into diamide intermediate 6 by mono-
acylation with R2COCl and then cyclized to give the
There are two synthetic approaches for the fused bi-
cyclic 2,3-disubstituted-pyrimidin-4(3H)-ones that have been
Fig. 1.
*
Corresponding author. Tel.: +1 858 332 4711; fax: +1 858 812 1648.
Scheme 1.
E-mail address: kyang@gnf.org (K. Yang).
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.01.071

1726
K. Yang et al. / Tetrahedron Letters 49 (2008) 1725–1728
Table 1
desired fused bicyclic 2,3-disubstituted pyrimidin-4(3H)-
ones 7.3a,b In some cases, the mono-acylation and cycliz-
ation steps were conducted in one pot and the diamide
Lewis acid effects on the cyclization reactiona
intermediate
6
was not isolated or characterized.3c
Approach B introduced the C2-substituent first by mono-
acylation of 4 to provide amide 8, which was then con-
verted into the same diamide intermediate 6 and then
cyclized to give 7.3b,d In an alternate but more common
version of approach B, amide 8 was first converted into
oxazinone 9, which was then converted into 7 when treated
with the corresponding amine R1NH2.3e,f
Both of the traditional routes are linear and require
sequential installation of the fused ring, then the C2-,
N3-substituents. One major drawback to these approaches
is that they are not general in scope. Most of the reported
examples are the syntheses of quinazolin-4-ones. The
detailed reaction sequences, conditions, and yields varied
considerably when the fused ring was heteroaromatic.
Acylation of the amino group in 4 often led to a mixture
of mono- and bis-acylation products. The ring opening of
oxazinone 9 by an aniline was sluggish, which could be
one of the reasons that the reported examples of 2,3-diaryl
substitution are rare. The ring closing reaction conditions
from 6 to 7 required optimization based on individual sub-
strates and led to an undesired cyclization product 10 in
some cases.4
Entry
Lewis acid
3ab (%)
15ab (%)
1
2
3
4
TMSCl
SnCl4
TiCl4
0
25
66
44
16
59
78
65
7
48
0
0
0
0
0
0
BF3ÁOEt2
5
AlCl3
6c
7d
8e
a
TiCl4
TiCl4
TiCl4
Conditions: (1) SOCl2, 80 °C, 2 h; (2) 4a (1.2 equiv), ClCH2CH2Cl,
Lewis acid (4.0 equiv), microwave, 170 °C, 20 min.
b
Reported yields determined via HPLC analysis of the crude reaction
mixtures.
Our goal was to obtain an efficient one-pot synthesis of
c
1.0 equiv of TiCl4 was used.
Microwave irradiation time was extended to 1 h.
Reaction was conducted in an oil bath at 150 °C overnight.
fused bicyclic 2,3-diaryl-pyrimidin-4(3H)-ones
3
by
d
e
employing three readily available components: aniline 11,
benzoyl chloride 12, and a heteroaromatic amino ester 4
(Scheme 2). Amide 13 was synthesized from aniline 11
and benzoyl chloride 12 under standard conditions. After
heating in SOCl2 at 80 °C for 2 h, 13 was converted into
imidoyl chloride 14, which underwent the cycloconden-
sation reaction when treated with various heteroaromatic
amino esters 4 to provide compounds of the general struc-
ture 3. The cyclocondensation occurred in two steps, the
nucleophilic attack of the imidoyl chloride by the amino
group to form intermediate 15 followed by intramolecular
cyclization. In some cases, intermediate 15 could be iso-
lated. To the best of our knowledge, even though some
examples of cyclocondensation between an imidoyl chlo-
ride and an amino ester have been reported, these reactions
were limited to electron rich amino esters like anthranilic
acid ester,5a–g thiophene amino ester,5h and amino crotonic
ester.5i These reactions were normally conducted under
basic or neat conditions. However, for electron deficient
heteroaromatic amino esters, such as 4a (Table 1), the
cyclocondensation was difficult due to the poor nucleo-
philicity of the amino group. Fortunately, we discovered
that adding an appropriate Lewis acid into the reaction
system could activate the imidoyl chloride and therefore
make the aforementioned cyclocondensation feasible. We
were also pleased to find that microwave irradiation could
be employed to accelerate the reaction.
Our initial study was focused on screening a few com-
mon Lewis acids for the cyclocondensation reaction and
the results are listed in Table 1. It was observed that the
strength of the Lewis acid strongly affected the outcome
of this reaction.6 As illustrated in Table 1, the weak Lewis
acid TMSCl (entry 1) gave only 7% of the uncyclized inter-
mediate 15a with no desired cyclized product 3a. The mod-
erately stronger Lewis acid SnCl4 successfully catalyzed the
nucleophilic addition; however, it was not effective enough
to promote the cyclization, leading to the identification of
15a as the major product (entry 2), which was not obtained
when strong Lewis acids were used. For example, TiCl4
gave 66% of the desired compound 3a as the only product,
as did BF3ÁOEt2 and AlCl3 (entries 3–5). However, the
stronger Lewis acids BF3ÁOEt2 (entry 4) and AlCl3 (entry
5) gave lower yields of 3a. Presumably, these strong Lewis
Scheme 2.

K. Yang et al. / Tetrahedron Letters 49 (2008) 1725–1728
1727
Table 2
acids deactivated the nucleophile and hindered the nucleo-
philic addition step. Overall, TiCl4 gave the highest yield of
3a among all the Lewis acids tested and was the catalyst of
choice for further optimization of the reaction conditions.
Next, we investigated the amount of TiCl4, the reaction
times, and the temperature to find the most optimal reac-
tion conditions (Table 1). We found that 1 equiv of TiCl4
was effective in catalyzing the reaction (59%, entry 6), but
was less satisfactory than using a slight excess. Employing
8 equiv of TiCl4 led to decomposition of the starting mate-
rials (not shown). Extended reaction time (1 h) slightly
improved the conversion (entry 7). Extended reaction times
under conventional heating gave good results (ꢀ14 h, entry
8). Finally, the optimal condition for parallel synthesis was
determined to be 2 equiv of TiCl4 at 170 °C under micro-
wave irradiation for 20 min (Table 2, entry 1).
With the optimal reaction condition in hand, a variety
of amides were surveyed and the results are summarized
in Table 2. In general, the aromatic amides, after having
first been converted into imidoyl chlorides, reacted very
effectively with 4a to give the desired products. We noted
that the substitution pattern of both R1 (3b,c) and R2
(3i–k) had minimal impacts on the yields. There was a gen-
eral trend that the aromatic amides with strong electron
donating groups gave a higher overall yield (in two steps)
than those with strong electron withdrawing groups. This
was true for both R1 (3d vs 3a,e,f) and R2 (3g vs 3h). It
Synthesis of 1H-pyrazolo[3,4-d]pyrimidin-4(5H)-ones from various
amidesa
Entry
Product
R1
R2
Yieldb (%)
1
2
3
4
5
6
7
8
9
3a
3b
3c
3d
3e
3f
4-Cl–Ph
2-Me–Ph
4-Me–Ph
4-MeO–Ph
4-CN–Ph
4-NO2–Ph
4-Cl–Ph
4-Br–Ph
4-Cl–Ph
4-Cl–Ph
4-Cl–Ph
Me
4-Br–Ph
4-Br–Ph
4-Br–Ph
4-Br–Ph
4-Br–Ph
4-Br–Ph
4-MeO–Ph
4-CF3–Ph
4-Me–Ph
3-Me–Ph
2-Me–Ph
4-Br–Ph
4-Br–Ph
60c
74
79
86
43d
53e
60
36
61
63
61
56
92f
3g
3h
3i
10
11
12
13
a
3j
3k
16
16
Me
General procedures refer to Ref. 7.
Isolated yield with >95% purity.
Reaction was conducted at 150 °C overnight.
b
c
d
Formation of imidoyl chloride: amide was treated with SOCl2 at 80 °C
for 2 h with the addition of one drop of DMF.
e
Formation of imidoyl chloride: amide was treated with SOCl2 at 80 °C
for 24 h.
f
2 equiv of amide was used.
Table 3
Preparation of fused bicyclic 2,3-disubstituted-pyrimidin-4(3H)-onesa
Entry
Product
R1
R2
4
Yieldb (%)
1
2
3
4
5
6
7
8
9
3l
4-Cl–Ph
4-Cl–Ph
4-Cl–Ph
4-Cl–Ph
4-Cl–Ph
4-MeO–Ph
4-Cl–Ph
Me
4-MeO–Ph
4-NO2–Ph
4-Br–Ph
4-iPr–Ph
4-Br–Ph
4-Br–Ph
4-Br–Ph
4-Br–Ph
4-Br–Ph
4-Br–Ph
4b
4b
4c
4c
4d
4d
4e
4f
73
3m
3n
3o
3p
3q
3r
17
3s
3t
50c
67
74
40d
60d
73e
85
4-Cl–Ph
4-Cl–Ph
4g
4h
63
Trace
10
a
General procedures refer to Ref. 7.
Isolated yield with >95% purity.
Formation of imidoyl chloride: amide was treated with SOCl2 at 80 °C for 2 h with the addition of one drop of DMF.
Microwave irradiation time was extended to 25 min.
b
c
d
e
Reaction was heated at 170 °C in a microwave reactor for 20 min, then 130 °C in an oil bath for 14 h.

1728
K. Yang et al. / Tetrahedron Letters 49 (2008) 1725–1728
was observed that the conversion of aromatic amides into
imidoyl chlorides was lower with strong electron withdraw-
ing groups at either R1 or R2. In these cases, prolonged
heating or addition of DMF was required to assist the con-
version (3e,f). We suspect that the inefficient formation and
poor stability of the imidoyl chloride contributed to the
lower overall yield of the one-pot reaction. Fortunately,
in the presence of excess amide, the yields were improved
(entry 12 vs 13).
References and notes
1. For selected recent examples, see: (a) Manikowski, A.; Verri, A.;
Lossani, A.; Gebhardt, B. M.; Gambino, J.; Focher, F.; Spadari, S.;
Wright, G. E. J. Med. Chem. 2005, 48, 3919; (b) Quintini, G.; Tenor,
H.; Baer, T.; Hatzelmann, A. WO2004089953, 2004; Chem. Abstr.
2004, 141, 379745; (c) Piazza, G. A.; Pamukcu, R. US6200980, 2001;
Chem. Abstr. 2001, 134, 222726; (d) Niewo¨hner, U.; Bischoff, E.;
Hutter, J.; Perzborn, E.; Schutz, H. EP0771799, 1997; Chem. Abstr.
¨
¨
1997, 127, 34234.
2. (a) For selected recent examples, see: Inoue, H.; Murafuji, H.;
Hayashi, Y. WO2004111054, 2004; Chem. Abstr. 2004, 142, 74598; (b)
Markwalder, J. A.; Seitz, S. P.; Sherk, S. R. WO2003063764, 2003;
Chem. Abstr. 2003, 139, 164800; (c) Bacon, E. R.; Daum, S. J.; Singh,
B. WO9628429, 1996; Chem. Abstr. 1996, 125, 301016.
3. For selected examples, see: (a) Hisano, T.; Ichikawa, M.; Nakagawa,
A.; Tsuji, M. Chem. Pharm. Bull. 1975, 23, 1910; (b) Skibo, E. B.;
Huang, X.; Martinez, R.; Lemus, R. H.; Craigo, W. A. J. Med. Chem.
2002, 45, 5543; (c) Tani, J.; Yamada, Y.; Oine, T.; Ochiai, T.; Ishida,
R.; Inoue, I. J. Med. Chem. 1979, 22, 95; (d) Bavetsias, V.; Skelton, L.
A.; Yafai, F.; Mitchell, F.; Wilson, S. C.; Allan, B.; Jackman, A. L. J.
Med. Chem. 2002, 45, 3692; (e) Vaidya, N. A.; Blanton, C. D., Jr. J.
Org. Chem. 1982, 47, 1777; (f) Nielsen, F. E.; Pedersen, E. B.
Tetrahedron 1982, 38, 1435.
To further demonstrate the effectiveness of these reac-
tion conditions, several more heterocyclic amino esters
(4b–h) were investigated and the results are listed in Table
3. These amino esters were either commercially available or
prepared by the known literature procedures.8–12 Most of
these heterocyclic amino esters reacted effectively with the
imidoyl chloride to provide the desired bicyclic 2,3-diaryl-
pyrimidin-4(3H)-ones in good yields. The reactivity heavily
depended on the nature of the heterocyclic amino ester.
For example, in the case of 4b,8 the additional electron
donating methylthio group enhanced the nucleophilicity
of the amino group relative to 4a, leading to higher yields
(3l,m). Ethyl 4-amino-3-phenyl-1H-pyrazole-5-carboxylate
4c9 also proved to be a better substrate for this reaction
than 4a. The reactions between 4c and the imidoyl chloride
normally went to completion in both steps and gave a
higher isolated yield (3n,o). On the other hand, it was
observed that 4d10 and 4e11 required prolonged heating
to force the cyclization (3p–r). In the case of 4f, the amino
group was so nucleophilic that heating was not required for
the nucleophilic attack step to occur, but the intramole-
cular cyclization step still needed heating to facilitate the
cyclization (17). An example was also given to show that
an ester group could survive this reaction condition (3s).
Among all of the heterocyclic amino esters investigated,
only the acid labile 4h12 failed to give any significant
amount of the desired product (3t).
4. He, F.; Snider, B. B. J. Org. Chem. 1999, 64, 1397.
5. (a) Helmholz, F.; Schroeder, R.; Langer, P. Synthesis 2006, 2507; (b)
Langer, P.; Helmholz, F.; Schroeder, R. Synlett 2003, 2389; (c)
Langer, P.; Wuckelt, J.; Do¨ring, M.; Go¨rls, H. Eur. J. Org. Chem.
2001, 1503; (d) Wuckelt, J.; Do¨ring, M.; Beckert, R.; Langer, P.
Synlett 1999, 1100; (e) Nesterova, I. N.; Radkevich, T. P.; Granik, V.
G. Khim.-Farm. Zh. 1991, 28; (f) Chervonyi, V. A.; Zyabrev, V. S.;
Kharchenko, A. V.; Drach, B. S. Zh. Org. Khim. 1989, 2597; (g)
Smolii, O. B.; Brovarets, V. S.; Pirozhenko, V. V.; Drach, B. S. Zh.
Obshch. Khim. 1988, 2465; (h) Manhas, M. S.; Amin, S. G. J.
Heterocycl. Chem. 1976, 13, 903; (i) Staskun, B.; Stephen, H. J. Am.
Chem. Soc. 1956, 4708.
6. For relative strength of various Lewis acids, see: Childs, R. F.;
Mulholland, D. L.; Nixon, A. Can. J. Chem. 1982, 60, 801.
7. General procedure for the synthesis of fused bicyclic 2,3-diaryl-
pyrimidin-4(3H)-ones 3: A mixture of amide 13 (0.25 mmol) and
SOCl2 (0.5 mL) was heated at 80 °C till the formation of imidoyl
chloride was complete (normally within 2 h). Excess SOCl2 was then
removed under vacuo. The residue was dissolved in anhydrous 1,2-
dichloroethane (1.5 mL) and transferred into a solution of amino
ester 4 (0.30 mmol) in anhydrous 1,2-dichloroethane (1.0 mL). After
TiCl4 (0.50 mmol) was added, the reaction mixture was heated at
170 °C by a microwave reactor for 20 min. The mixture was quenched
with water (30 mL) and extracted with chloroform (3 Â 15 mL). The
combined organic layer was washed with brine, dried over MgSO4,
and purified by chromatography on silica gel (hexanes/EtOAc).
8. Yokoyama, M.; Hongo, S. Bul. Chem. Soc. Jpn. 1973, 46, 699.
9. Yuan, J.; Gulianello, M.; Lombaert, S. D.; Brodbeck, R.; Kieltyka,
A.; Hodgetts, K. J. Bioorg. Med. Chem. Lett. 2002, 12, 2133.
10. Collins, M.; Ewing, D.; Mackenzie, G.; Sinn, E.; Sandbhor, U.;
Padhye, S.; Padhye, S. Inorg. Chem. Commun. 2000, 3, 453.
11. L’abbe, G.; Beenaerts, L. Tetrahedron 1989, 45, 749.
In summary, we have developed an efficient one-pot syn-
thesis of fused bicyclic 2,3-diaryl-pyrimidin-4(3H)-ones.
These molecules can be obtained in two steps from commer-
cially or readily available starting materials in moderate to
good yields. The key step is a Lewis acid assisted cyclization
reaction between imidoyl chloride intermediate 14 and
heteroaromatic amino ester 4. This reaction appeared to
be general, robust, and amenable for parallel synthesis.
Acknowledgments
12. Frietze, W. E. WO2000011003, 2000; Chem. Abstr. 2000, 132,
180596.
We thank Dr. Dean Phillips and Dr. Jon Loren for
helpful discussion upon preparing this Letter.