A facile synthesis of an oxazolo[5,4]pyrimidin-2-one and a pyrimido[5,4][1,3]oxazin-2-one

Article abstract of DOI:10.1055/s-2006-941604

A facile synthesis is reported for the construction of the five- and six-membered fused carbamate rings of an oxazolo[5,4]pyrimidin-2-one and a pyrimido[5,4][1,3]oxazin-2-one, respectively. The method utilises a controlled two-step procedure in which a re

Full text of DOI:10.1055/s-2006-941604

LETTER
1519
A Facile Synthesis of an Oxazolo[5,4]pyrimidin-2-one and a
Pyrimido[5,4][1,3]oxazin-2-one
F
acile Route toPro
a
duce
F
use
r
m
c
Carbamate–Pyrim
e
idine Scaff
n
olds P. Dijkstra, Catherine Gaulon, Dan Niculescu-Duvaz, Caroline J. Springer*
Centre of Cancer Therapeutics, The Institute of Cancer Research, 15 Cotswold Road, Sutton, Surrey, SM2 5NG, UK
Fax +44(208)7224205; E-mail: caroline.springer@icr.ac.uk
Received 20 September 2005
Our interest in kinase inhibitors led us to develop a novel
route towards oxazolo[5,4]pyrimidin-2-ones 1 and pyrim-
ido[5,4][1,3]oxazin-2-ones 2 (Figure 1) with R2 = benzyl
or functionalised benzyl-groups. To our knowledge, no
Abstract: A facile synthesis is reported for the construction of the
five- and six-membered fused carbamate rings of an oxazo-
lo[5,4]pyrimidin-2-one and a pyrimido[5,4][1,3]oxazin-2-one, re-
spectively. The method utilises a controlled two-step procedure in
which a reactive p-nitrophenylcarbamate intermediate ring closes compounds of types 1 and 2 are known in which N-7 and
upon treatment with base, affording the bicyclic pyrimidine-car-
bamate scaffolds in good yields.
N-9, respectively, are substituted. This was rather surpris-
ing since scaffolds 1 and 2 possess many characteristics of
Key words: pyrimidines, cyclic carbamates, ring-closing, hetero-
cyclic chemistry
a good pharmacophore for binding in the ATP-pocket of
kinases. We believe that the absence of compounds of
type 1 and 2 in the literature is due to the lack of good and
reproducible methods to produce their bicyclic pyrimi-
Pyrimidine derivatives play an important role in many dine-carbamate ring systems. Herein we report a new
drug discovery programmes in the pharmaceutical indus- proof of concept methodology to construct the bicyclic
try and amongst them bicyclic pyrimidine ring systems pyrimidine-carbamate scaffolds of both an oxazo-
are a very important class of compounds.1 Bicyclic pyri- lo[5,4]pyrimidin-2-one (1) and a pyrimido[5,4][1,3]ox-
midine ring systems also play a crucial role in mammalian azin-2-one (2) bearing a benzyl substituent (R2, Figure 1)
systems: the DNA base pairs adenine and guanine are pu- on the positions N-7 and N-9, respectively.
rine derivatives whilst ATP (adenosine triphosphate), a
R2
N7
key energy provider substrate in biological signalling
pathways, contains a purine moiety. More specifically, the
R2
N
N9
N
O
purine group plays an important role in the binding of
ATP to protein kinases (an important class of enzymes
mediating most signal transduction pathways),1b,2 result-
ing in activation of downstream enzymes in the signalling
pathway through phosphorylation.3 It is known that many
forms of cancer develop due to mutations of one or several
enzymes in these signalling pathways. For this reason, a
considerable body of research is directed toward the
development of ATP-competitive inhibitors for mutated
protein kinases thereby preventing activation of these
enzymes and thus blocking the signalling pathway that
can lead to cancers.4 Examples of protein kinase inhibitors
in the treatment of cancer are Gleevec© (for leukaemia),
Iressa© (for lung cancer) and Sorafeni© (for renal cancer).5
Thus developing protein kinase inhibitors that strongly
compete with ATP have a huge potential in cancer treat-
ments. Pyrimidines have already been shown to be very
successful in this area because they possess many key
functionalities for crucial binding (mainly hydrogen
bonds) to the backbone (hinge region) of the constrained
and rather well-defined ATP-pocket of many protein
kinases. In addition, due to their versatile chemistry,
pyrimidines can be readily decorated with various func-
tionalities to fine-tune their biological activity.6
R1
N
O
O
2
2
R1
N
1
O
2
Figure 1 Oxazolo[5,4]pyrimidin-2-ones 1 and
pyrimido[5,4][1,3]oxazin-2-ones 2
Initial screening of the literature for potential suitable
methods to create the bicyclic pyrimidine-carbamate ring
systems resulted in only two publications by Wetzel et al.
describing compounds of type 1 in which R2 = H.7 They
reported the synthesis of a variety of oxazolo-pyrimidine
derivatives of type 1 (R2 = H) as intermediates in the syn-
thesis of b-lactam antibiotics. In their synthesis, 5-amino-
4-hydroxypyrimidines were reacted with phosgene to pro-
vide the desired oxazolopyrimidines in moderate to good
yields. As mentioned earlier, from a medicinal chemistry
point of view, we were keen to introduce substituents,
preferably (functionalised) benzyl substituents, on the
positions N-7 of 1 and the N-9 of 2. To explore suitable
routes we decided to use a model reaction with
R2 = benzyl. Applying Wetzel’s reaction conditions to
construct the bicyclic pyrimidine-carbamate scaffold of 1
(R2 = benzyl) starting from 5-(benzylamino)uracil8 and
phosgene resulted in irreproducible results with poor
yields. In the best case, 1-benzyl-5-hydroxyoxazolo[5,4-
d]pyrimidin-2(1H)-one (5) was isolated in 8% yield. Re-
placing the phosgene with triphosgene, another regularly
used reagent for this type of reaction resulted in equally
SYNLETT 2006, No. 10, pp 1519–1522
2
1
.0
6
.2
0
0
6
Advanced online publication: 12.06.2006
DOI: 10.1055/s-2006-941604; Art ID: D28805ST
© Georg Thieme Verlag Stuttgart · New York

1520
H. P. Dijkstra et al.
LETTER
poor yields. Since we needed a reproducible method to found to be crucial in order to get satisfying yields and re-
access products of type 1, we started to explore a different producible results. This is probably due to the fact that
method to prepare the central pyrimidine-carbamate skel- water is necessary to solubilise all reagents under the ap-
eton. We decided to perform the ring-closing step in a plied reaction conditions.
controllable two-step procedure. The synthesis of 1-ben-
zyl-5-hydroxyoxazolo[5,4-d]pyrimidin-2(1H)-one (5) is
O
HN
HN
N
H
Ph
i
outlined in Scheme 1. First the reactive 4-nitrophenyl
carbamate 4 was synthesised from 5-benzylaminouracil 3,
which is available from 5-bromouracil and benzylamine.8
Uracil 3 was found to be insufficiently soluble to react
with 4-nitrophenyl chloroformate to afford intermediate
4: only up to 30% conversion was observed even after
prolonged reaction times (>72 h) and elevated tempera-
tures (reflux in THF). To overcome this problem, 3 was
first treated with an excess of N,N-diethyltrimethylsilyl-
amine, giving the completely THF-soluble bis(silyloxy)
ether, and subsequent treatment of this soluble intermedi-
ate with 4-nitrophenyl chloroformate in the presence of
triethylamine followed by an aqueous work-up, gave the
4-nitrophenyl carbamate 4 in 77% yield.9 Subsequent
treatment of 4 with potassium tert-butoxide in THF at
reflux conditions resulted in the formation of the desired
bicyclic carbamate structure 5 in 77% yield9 and in 60%
overall yield starting from 3.9 Using the ethyl carbamate
intermediate, which is more commonly used in the litera-
ture to synthesise cyclic carbamates, instead of the 4-
nitrophenyl derivative, failed to the give the final cyclic
product 5. This is probably a result of the rather poor
nucleophilicity of the 4-OH functionality of the pyrimi-
dine combined with the poor leaving group ability of the
ethoxide group relative to the 4-nitrophenol group.
+
+
H2N
Ph
65%
H
O
H
O
N
H
O
N
H
6
O
Ph
ii
iii
N
N
O
N
N
Ph
82%
84%
O
HO
N
7
OH
HO
N
8
O
O
NO2
Scheme 2 Reagents and conditions: i. EtOH–H2O (4:1), reflux,
20 h; ii. 4-nitrophenyl chloroformate, THF, Et3N, 0 °C, r.t., 20 h; iii.
t-BuOK, THF, reflux, 2 h.
Treatment of uracil derivative 6 with 4-nitrophenyl chlo-
roformate in THF in the presence of triethylamine afford-
ed the reactive carbamate intermediate 7 in 82% yield.
Reaction of 7 with potassium tert-butoxide in THF at
reflux conditions resulted in the formation of the desired
pyrimido-oxazinone 8 in 84% yield.11 Also here, we first
attempted to synthesise 8 directly from 6 using phosgene
or triphosgene and the conditions described by Wetzel et
al., however, in this case we were unable to isolate any de-
sired product. Via the new controlled two-step procedure,
however, we were able to obtain the desired product 8 in
67% overall yield starting from uracil derivative 6.
H
Br
O
N
Ph
ii
In conclusion, we report a facile two-step procedure to
synthesise the bicyclic pyrimidine-carbamate skeleton of
oxazolo[5,4]pyrimidin-2-ones and pyrimido[5,4][1,3]ox-
azin-2-ones. For the first time an oxazolo[5,4]pyrimidin-
2-one and a pyrimido[5,4][1,3]oxazin-2-one are described
with an alkyl substituent at N-7 and N-9, respectively.
Whereas the more conventional methods (using phosgene
and triphosgene) failed, this method afforded the desired
fused pyrimidine-carbamate products in good yields,
creating the possibility to access a wide variety of these
bicyclic carbamate substrates. This in combination with
the rich chemistry already developed for functionalising
pyrimidines6,12 and the known key hydrogen bond donat-
ing/accepting abilities of bicyclic pyrimidine systems for
binding to biological targets, make these fused pyrimi-
dine-carbamate systems interesting building blocks in
medicinal chemistry.
HN
HN
i
+
H2N
Ph
90%
77%
O
N
O
N
H
3
O
H
Ph
Ph
N
O
N
O
iii
77%
N
N
O
O
HO
N
HO
N
OH
NO2
4
5
Scheme 1 Reagents and conditions: i. neat, 160 °C, 3 h;8 ii. N,N-
diethyltrimethylsilylamine, THF, reflux, 1 h, followed by 4-nitro-
phenyl chloroformate, Et3N, THF, 0 °C, r.t., 3 h; iii. t-BuOK, THF,
reflux, 1.5 h.
Since we were also interested in 6-membered cyclic car-
bamates fused to a pyrimidine ring (2, Figure 1) as scaf-
folds for drug-like molecules, we decided to investigate
whether this new method could also be applied to the con-
struction of the bicyclic scaffold of type 2. To do so, a
similar approach was followed (Scheme 2). In the first
step, uracil was subjected to Mannich reaction conditions
using paraformaldehyde and benzylamine in an ethanol–
water mixture (4:1), affording 5-(benzylaminometh-
Acknowledgment
We would like to thank Cancer Research-UK (grant numbers C309/
A2187 and C107/A3096), the Institute of Cancer Research and
Wellcome Trust for the funding of this work. We are grateful to our
colleagues Lawrence Davies, Ion Niculescu-Duvaz, Esteban
yl)uracil (6) in 65% yield.10 The presence of a consider- Roman and Ian Scanlon (at the ICR) and to Richard Taylor and
able amount of water (ca. 20%) in the solvent mixture was
Adrian Gill (Astex Technology Ltd) for fruitful discussions.
Synlett 2006, No. 10, 1519–1522 © Thieme Stuttgart · New York

LETTER
Facile Route to Procedure Fused Bicyclic Pyrimidine-Carbamate Scaffolds
off-white precipitate. The precipitate was collected by
1521
References and Notes
filtration, washed with H2O (2 × 15 mL) and Et2O (2 × 30
mL) and dried in vacuo, affording the product as an off-
white solid; yield 0.77 g (77%). 1H NMR (250 MHz,
DMSO-d6): d = 4.88 (s, 2 H, CH2), 7.33–7.42 (m, 5 H, ArH),
7.66 (s, 1 H, PyrH), 11.60 (br s, 1 H, OH). 13C NMR (62.9
MHz, DMSO-d6): d = 45.40, 111.98, 123.72, 127.68,
127.79, 127.91, 128.61, 134.56, 151.06, 154.93. HRMS:
m/z calcd for C12H10N3O3 [M + H+]: 244.0722. Found:
244.0726. Anal. Calcd for C12H9N3O3: C, 59.26; H, 3.73; N,
17.28. Found: C, 59.16; H, 4.17; N, 17.64.
(1) For purine kinase inhibitors, see for example: (a) Meijer,
L.; Raymond, E. Acc. Chem. Res. 2003, 36, 417. (b)Laufer,
S. A.; Domeyer, D. M.; Scior, T. R. F.; Albrecht, W.;
Hauser, D. R. J. J. Med. Chem. 2005, 48, 710.
(2) (a) Manning, G.; Whyte, D. B.; Martinez, R.; Hunter, T.;
Sudarsanam, S. Science 2002, 298, 1912. (b) Manning, G.;
Whyte, D. B.; Martinez, R.; Hunter, T.; Sudarsanam, S.
Science 2002, 298, 1933.
(3) Huse, M.; Kuriyan, J. Cell 2002, 109, 275.
(4) For a review on protein kinase inhibitors, see: Bridges, A. J.
Chem. Rev. 2001, 101, 2541.
For the use of p-nitrophenol esters in the synthesis of
carbamates, see also: (a) Nakata, T.; Fukui, M.; Oishi, T.
Tetrahedron Lett. 1988, 29, 2223. (b) Vigroux, A.; Bergon,
M.; Zedde, C. J. Med. Chem. 1995, 38, 3983.
(5) Levitzki, A. Acc. Chem. Res. 2003, 36, 462; this review
includes references to gleevec, iressa and sorafenib.
(6) Brown, D. J. Comprehensive Heterocyclic Chemistry, Vol.
3; Boulton, A. J.; McKillop, A., Eds.; Pergamon Press:
Oxford, 1985, 57.
(7) (a) Wetzel, B.; Woitun, E.; Reuter, W.; Maier, R.; Lechner,
U. Drug. Res. 1985, 35, 343. (b) Wetzel, B.; Woitun, E.;
Reuter, W.; Maier, R.; Lechner, U.; Goeth, H. Eur. Pat.
Appl., EP 4499265, 1982. (c) Maier, R.; Wetzel, B.;
Woitun, E.; Reuter, W.; Lechner, U.; Appel, K.-R. Arzneim.-
Forsch. 1986, 9, 1297.
(10) (a) A modification from a literature procedure was used:
Delia, T. J.; Scovill, J. P.; Munslow, W. D. J. Med. Chem.
1976, 19, 344. (b) Synthesis of 5-(Benzylaminometh-
yl)uracil (6). Paraformaldehyde (2.91 g, 97.0 mmol) and
benzylamine were mixed in EtOH (320 mL) and this mixture
was stirred at r.t. for 10 min. Subsequently, uracil (10.0 g,
88.2 mmol) was added in one portion followed by the
addition of H2O (80 mL) and the resulting mixture was
heated at reflux for 20 h. Next, the filtrate was evaporated to
dryness, leaving a white solid. After washing with acetone (3
× 25 mL) and drying in vacuo, the product was obtained as a
white solid; yield 13.2 g (65%). 1H NMR (250 MHz,
DMSO-d6): d = 3.29 (s, 2 H, CH2), 3.66 (s, 2 H, CH2), 7.17–
7.31 (m, 5 H, ArH), 7.32 (s, 1 H, PyrH). 13C NMR (62.9
MHz, DMSO-d6): d = 44.26, 52.10, 110.61, 126.49, 127.82,
128.07, 138.46, 140.70, 151.26, 164.42. MS (ES+) m/z = 232
[M + H+].
(11) Synthesis of 4-Nitrophenyl Benzyl[(2,4-dioxo-1,2,3,4-
tetrahydropyrimidin-5-yl)methyl]carbamate (7).
5-(Benzylaminomethyl)uracil (2.0 g, 8.7 mmol) was
dissolved in THF (40 mL) and cooled to 0 °C with an ice-
bath. Then, Et3N (1.2 mL, 8.7 mmol) was added followed by
the dropwise addition of a solution of p-nitrophenyl
chloroformate in THF (10 mL). The temperature was
allowed to rise to r.t. and stirring was continued for 20 h at
that temperature. Next, all volatiles were evaporated, leaving
a yellow solid. This solid was washed with H2O (2 × 20 mL),
Et2O (2 × 20 mL), and CH2Cl2 (2 × 20 mL) and dried in
vacuo. Afterwards, THF (15 mL) was added to this solid and
the resulting suspension was stirred at r.t. for 10 min.
Subsequently, Et2O (30 mL) was added and the precipitate
was collected and dried in vacuo, leaving a slightly yellow
solid; yield: 2.82 g (82%). 1H NMR (250 MHz, DMSO-d6,
100 °C): d = 4.24 (s, 2 H, CH2), 4.66 (s, 2 H, CH2), 7.30–7.44
(m, 8 H, ArH + PyrH), 8.24 (d, 3JH,H = 9.0 Hz, 2 H, ArH),
10.48 (br s, 2 H, NH). 13C NMR (62.9 MHz, DMSO-d6, 100
°C): d = 43.53, 50.24, 107.46, 122.77, 125.07, 127.01,
127.31, 128.53, 137.52, 141.24, 144.41, 151.14, 153.12,
156.17, 164.25. HRMS: m/z calcd for C19H17N4O6 [M + H+]
397.1148; found: 397.1145.
(8) For the synthesis of 3, see: Gaulon, C.; Dijkstra, H. P.;
Springer, C. J. Synthesis 2005, 13, 2227.
(9) Synthesis of 4-Nitrophenyl Benzyl(2,4-dioxo-1,2,3,4-
tetrahydropyrimidin-5-yl)carbamate (4).
N,N¢-Diethyltrimethylsilylamine (5.5 mL, 30.32 mmol) was
added to a suspension of 5-(benzylamino)uracil (1.75 g, 7.58
mmol) in THF (120 mL) and the resulting mixture was
heated to reflux for 1 h, affording a clear solution. The
reaction mixture was cooled to ambient temperature and
subsequently all volatiles were removed in vacuo, affording
the disilylated intermediate as a sticky white residue. The
intermediate was dissolved in THF (120 mL) and cooled to
0 °C. Then, Et3N (1.02 mL, 7.58 mmol) was added, followed
by the dropwise addition of p-nitrophenyl chloroformate in
THF (10 mL). The temperature of the resulting mixture was
allowed to rise to r.t. and stirring was continued for 3 h. The
reaction mixture was filtered to remove Et3N salts and H2O
(5 mL) was added to the filtrate to hydrolyse the
trimethylsilyl–oxygen bonds. Next, the THF layer was
concentrated to dryness, leaving a yellow solid. This solid
was redissolved in THF (50 mL) and filtered to remove
traces of insoluble material. The filtrate was concentrated to
ca. 20 mL and upon addition of Et2O (ca. 20 mL), an off-
white solid precipitated. This precipitate was collected by
filtration, washed with Et2O (2 × 10 mL) and dried in vacuo,
affording the product as an off-white solid; yield 2.23 g
(77%). 1H NMR (250 MHz, DMSO-d6): d = 4.50–4.91 (m, 2
H, CH2, different rotamers exist), 7.32–7.50 (m, 8 H, ArH +
PyrH6), 8.29 (d, 3JH,H = 9.0 Hz, 2 H, ArH), 11.01 (br s, 1 H,
NH), 11.42 (br s, 1 H, NH). 13C NMR (62.9 MHz, DMSO-
d6): d = 52.93, 114.39, 115.71, 122.67, 125.11, 126.09,
127.46, 128.27, 128.32, 136.37, 144.52, 150.50, 156.06,
161.17. MS (ES+): m/z = 406.09 [M + Na+].
Synthesis of 3-Benzyl-7-hydroxy-3,4-dihydropyrimi-
do[5,4-e][1,3]oxazin-2-one (8).
Synthesis of 1-Benzyl-5-hydroxyoxazolo[5,4-d]pyrimi-
din-2(1H)-one (5).
4-Nitrophenyl benzyl[(2,4-dioxo-1,2,3,4-tetrahydro-
pyrimidin-5-yl)methyl]carbamate (7, 1.0 g, 2.5 mmol) was
dissolved in THF (100 mL) and t-BuOK (0.57 g, 5.1 mmol)
was added in one portion. The resulting solution was heated
to reflux for 2 h and subsequently cooled to r.t. and
concentrated to dryness. Then, H2O (20 mL) was added to
the yellow solid and this mixture was stirred vigorously for
5 min. This layer was acidified with 1 M HCl aq (ca. 20 mL),
resulting in a white precipitate. The precipitate was
collected, washed with H2O (2 × 10 mL) and Et2O (2 × 15
4-Nitrophenyl benzyl(2,4-dioxo-1,2,3,4-tetrahydropyrimi-
din-5-yl)carbamate (4, 1.58 g, 4.13 mmol) was dissolved in
THF (140 mL) and t-BuOK (0.97 g, 8.26 mmol) was added
in one portion and the resulting mixture was heated to reflux
for 1.5 h. After cooling the reaction mixture to r.t., the
reaction mixture was concentrated to dryness. Then, H2O
(25 mL) was added to the brightly yellow-coloured solid and
this mixture was stirred vigorously for 5 min. Next, this layer
was acidified with 1 M HCl aq (ca. 15 mL), resulting in an
Synlett 2006, No. 10, 1519–1522 © Thieme Stuttgart · New York

1522
H. P. Dijkstra et al.
LETTER
mL) and dried in vacuo, affording the product as a white
solid; yield 0.50 g (84%). 1H NMR (250 MHz, DMSO-d6):
d = 4.15 (s, 2 H, CH2), 4.60 (s, 2 H, CH2), 7.31–7.41 (m,
5 H, ArH), 7.94 (s, 1 H, PyrH), 11.95 (br s, 1 H, OH).
13C NMR (62.9 MHz, DMSO-d6): d = 42.54, 51.65, 95.35,
127.70, 127.77, 128.65, 135.47, 144.67, 148.65, 155.87,
166.03. HRMS: m/z calcd for C13H11N3O3Na [M + Na+]:
280.0698; found: 280.0696. Anal. Calcd for
C13H11N3O3·0.2H2O: C, 59.86; H, 4.41; N, 16.11. Found: C,
60.02; H, 4.37; N, 15.73.
(12) Some examples can be found in the following papers:
(a) Sandosham, J.; Undheim, K. Heterocycles 1994, 37,
501. (b) Sandosham, J.; Undheim, K.; Rise, F. Heterocycles
1993, 35, 235. (c) Sugimoto, O.; Mori, M.; Tanji, K.
Tetrahedron Lett. 1999, 40, 7477.
Synlett 2006, No. 10, 1519–1522 © Thieme Stuttgart · New York