Copper-mediated N-arylation of quinazolinediones

Article abstract of DOI:10.1055/s-0029-1217724

A mild, ligand-free method of arylating 2,4-quinazolinediones using arylboronates in the presence of copper salts is described. The reaction is tolerant of a variety of functional groups and works for arylboronic acid, arylboronic ester, and aryltrifluoro

Full text of DOI:10.1055/s-0029-1217724

LETTER
2253
Copper-Mediated N-Arylation of Quinazolinediones
Copper-Mediated
o
N
-Arylation
l
of
Q
ui
l
nazolin
e
ediones tte S. Guy,a Teyrnon C. Jones*b
a
Department of Chemistry, University of Warwick, Coventry, CV4 7AL, UK
Department of Medicinal Chemistry, AstraZeneca R&D Charnwood, Bakewell Road, Loughborough, Leics, LE11 5RH, UK
Fax +44(1509)644925; E-mail: teyrnon.jones@astrazeneca.com
b
Received 7 May 2009
N-Arylation technology has come far since the discovery
of the Goldberg–Ullmann reaction in 1906.6 Research
driven by the groups of Stephen Buchwald and John
Hartwig among others has delivered important metal-me-
diated methods for heteroatom–aryl bond formation, ini-
tially using palladium catalysts7 and later more
economically viable copper-based systems.8 The aryl do-
nor is generally an aryl halide which can react with a wide
range of nitrogen nucleophiles including amines,9
amides,10 and nitrogen heterocyles.11 It was envisaged
given the wide range of substrate and catalyst combina-
tions described in the literature that similar technology
could be used to effect the N-arylation of quinazolinedi-
ones.
Abstract: A mild, ligand-free method of arylating 2,4-quinazo-
linediones using arylboronates in the presence of copper salts is de-
scribed. The reaction is tolerant of a variety of functional groups and
works for arylboronic acid, arylboronic ester, and aryltrifluorobor-
onate donors. A catalytic variant is also described.
Key words: heterocycle, arylation, copper, cross-coupling, boronic
1-Aryl-2,4-quinazolinediones have long attracted the in-
terest of the pharmaceutical industry because they demon-
strate notable anti-inflammatory activity, which has led to
their development as potential analgesic, antiasthmatic,
and arthritis therapies.1 Several methods for their prepara-
tion have been reported in the literature, including elabo-
ration of uracils2 and nucleophilic aromatic substitution,3
but by far the most common approach is to condense a car-
bonyl equivalent such as phosgene or urea with an appro-
priately functionalised aniline (Scheme 1, disconnection
a).1a,4 All of these methods are afflicted by a lack of gen-
erality and/or poor yield, as well as two bugbears of phar-
maceutical synthesis: route linearity and the use of
potentially mutagenic aniline components. Linear routes
not only lead to lower overall yields during scale up, but
limit diversity-oriented synthesis, while aniline reagents
are often toxic, can persist throughout multistep prepara-
tions, and must be rigorously removed from any potential
drug before in vivo test, with an attendant impact on time
and cost.5
3-Methyl-2,4(1H,3H)-quinazolinedione (1) was chosen
as a model substrate and synthesised in one step via
Willis’ elegant tandem palladium-catalysed urea arylation–
cyclisation (Scheme 2).12
O
O
Me
Me
Me
HN
N
O
+
a
H2N
O
N
O
Br
H
1
Scheme 2 Synthesis of a model quinazolinedione substrate.
Reagents and conditions: a) Cs2CO3 (2 equiv), Pd2(dba)3 (5 mol%),
XantPhos (10 mol%), dioxane, 100 °C, 24 h (61%).
An alternative preparation of 1-aryl-2,4-quinazoline-
diones involving arylation of a quinazolinedione core
(Scheme 1, disconnection b) is therefore desirable. Such
an approach does not demand anilinic reagents, is conver-
gent, and is also amenable to library synthesis since mo-
lecular diversity is incorporated during the last step of the
sequence.
In an attempt to develop a one-pot synthesis of N-arylated
quinazolines based around this chemistry 4-iodotoluene
was added into the reaction broth once the formation of 1
was complete, but unfortunately the palladium which
moderated the original reaction failed to catalyse any sub-
sequent N-arylation. A more comprehensive investigation
was therefore undertaken, treating quinazolinedione 1
with many reported palladium and copper N-arylation
O
O
O
Me
Me
Me
b
X
N
N
N
X
a
H
a
Ar
X
O
NH
Ar
N
Ar
O
N
O
b
H
Scheme 1 Possible approaches to 1-aryl quinazolinediones
SYNLETT 2009, No. 14, pp 2253–2256
x
x
.x
x
.2
0
0
9
Advanced online publication: 03.08.2009
DOI: 10.1055/s-0029-1217724; Art ID: D11409ST
© Georg Thieme Verlag Stuttgart · New York

2254
C. S. Guy, T. C. Jones
LETTER
systems in the presence of 4-iodotoluene as an aryl donor, 2), although using a greater than twofold excess gave no
but in every case no trace of arylated product was evident. further yield improvement. Choice of base or solvent had
a significant impact on conversion: using most common
It was clear that a mechanistically dissimilar arylation us-
inorganic bases resulted in a poor reaction (e.g., entry 5),
ing an alternative aryl donor might give better results.
and of the organic bases investigated triethylamine proved
Alternative aryl sources such as arylboronic acids,13 aryl-
superior to pyridine. Solvent was less crucial, with dichlo-
siloxanes,14 arylbismuths,15 and arylleads16 have all been
romethane, acetonitrile, and toluene (entries 3, 6, and 7,
used to arylate nitrogen nucleophiles, with arylboronic
respectively) giving reasonable conversions, though di-
compounds representing the most attractive alternative
methylformamide was found to be a poor solvent. As in
due to their wide availability and ease of preparation from
related reactions, 4 Å molecular sieves were used to re-
aryl halides.17 Further support for use of this donor as a
move traces of water in the reaction which might cause
possible arylation agent for quinazolinediones was found
degradation of the arylboronic acid reagent and a concom-
in a recent publication describing the arylation of the re-
itant drop in yield (entry 9).20
lated heterocycle uracil with phenylboronic acid.18
One useful feature of the procedure is that high conver-
Consultation of the literature revealed that most common
sions were obtained under very mild ambient conditions.
N-arylation procedures using boronic acid donors are me-
Furthermore, no inert atmosphere was required, making
diated by stoichiometric copper(II) salts under basic, aer-
obic conditions. Pleasingly, when model 1 was treated
with 4-tolylboroxy acid in the presence of copper(II) ace-
the reaction operationally very convenient. The kinetics of
the reaction are also worthy of comment. Similar arylation
procedures commonly suffer from extremely long reac-
tate and triethylamine the desired arylquinazolinedione
tion times of up to three days,13 whereas quinazoline-
was obtained in moderate to good yield (Table 1).
diones were found to react more rapidly. For example, the
reaction in Table 1 entry 3 reached 50% conversion after
only two hours and a synthetically useful isolated yield of
72% after 20 hours.21
Table 1 Copper-Mediated N-Arylation of Quinazolinedione 1
O
Me
B(OH)2
N
O
With optimised conditions in hand the focus of the inves-
tigation shifted to an exploration of reaction scope. Sever-
al commercial arylboronic derivatives possessing a range
of steric, electronic, and chemical properties were reacted
with quinazolinedione 1 under standard conditions, as
shown in Table 2.
Me
N
O
N
+
a
N
H
O
1
2
3
These results demonstrate a wide reaction scope, tolerant
of functionality. Yields seemed unaffected by the elec-
tronic nature of the arylboronic substituent, with electron-
rich (Table 1 entries 1–3) and electron-deficient aryl do-
nors (entries 6–8) giving comparable conversions. The
reaction is, however, sensitive to steric effects, as demon-
strated by the very poor conversion observed when using
o-tolyl arylboronic acid as a reactant (entry 5). Oxidative-
ly labile groups (entry 2) participate well despite an oxi-
dative mechanism mooted for similar reactions,20 as do
acids, bases, and hydrolytically sensitive groups (entries
3, 7, and 8). Using a phenolic arylboronic acid (entry 4)
led to a messy reaction, an unsurprising observation given
the nucleophilicity of phenolic hydroxy groups in similar
arylations.20 The nature of the boron group on the aryl do-
nor was also investigated, with both aryltrifluoroborate
(hitherto unreported as N-arylating amidic nitrogens) and
arylboronate ester donors found to furnish product, signif-
icantly widening the range of commercially available cou-
pling partners.
Entry
Base
Solvent
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
MeCN
toluene
DMF
Conversion (%) b
1c
2d
3
Et3N
59
Et3N
29
Et3N
80 (72)
34
4
pyridine
K2CO3
Et3N
5
<5
6
74
7
Et3N
53
8
Et3N
35
9e
Et3N
CH2Cl2
50
a Unless otherwise stated reactions were performed using boronic acid
(2 equiv), copper(II) acetate (2 equiv), base (2 equiv), and 4 Å MS at
r.t. for 20 h.
b LC-MS conversion based on unreacted quinazolinedione 1, isolated
yield in parentheses.19,
c With 1 equiv Cu(OAc)2.
d With 1 equiv ArB(OH)2.
Lam and others have reported catalytic variants of N-ary-
lation reactions using boronic acids.22 In an attempt to dis-
cover whether arylation of quinazolinediones could also
be rendered catalytic several literature-derived conditions
were applied to model 1. Gratifyingly arylation was found
to occur readily, albeit more slowly than the stoichiomet-
ric version. The best conditions found used a catalytic
e Without molecular sieves.
Copper(II) acetate was found to be the best copper salt
with an excess required for consistently high yields
(Table 1, entries 1 and 3). Unfortunately, a surplus of bo-
ronic acid was also necessary for efficient reaction (entry
Synlett 2009, No. 14, 2253–2256 © Thieme Stuttgart · New York

LETTER
Copper-Mediated N-Arylation of Quinazolinediones
2255
amount of copper(II) acetate in the presence of TEMPO as
co-oxidant (Scheme 3).
Table 2 Scope of the N-Arylation of Quinazolinedione 1
O
Me
B(OR)2
R
N
O
O
Me
Me
+
N
O
B(OH)2
N
N
O
a
N
O
N
H
O
N
R
a
N
H
O
1
Entry
1
Arylboronate
Product
Yield (%)
72
1
2
3
Me
O
Scheme 3 Catalytic arylation of quinazolinedione 1. Reagents and
conditions: a) Cu(OAc)2 (10 mol%), TEMPO (1.1 equiv), Et3N (2
equiv), 4 Å MS, r.t., 48 h (71%).
OH
B
4
OH
2
3
5
6
74
78
OH
B
S
As a final demonstration of utility the new arylation tech-
nology was applied in the synthesis of known pharmaceu-
tical H 27 (13), a known phosphodiesterase inhibitor
(Scheme 4).1a Synthesis of this compound by cyclisation
of methyl-2-bromobenzoate and ethyl urea followed by
N-arylation using 3-trifluoromethyl phenylboronic acid
gave 13 in 76% overall yield over two convergent steps,
compared to the three-step linear literature synthesis.1a
OH
Me
Me
N
OH
OH
B
Me
4
5
7
8
0
OH
OH
HO
B
A mild, ligand-free, and practically convenient method of
synthesising 1-aryl-2,4-quinazolinediones has been de-
veloped. The reaction tolerates a range of functional
groups and furnishes products in good yield even under
aerobic, ambient conditions. Further investigations are
under way to investigate potential chemo- and regioselec-
tivity, and apply the reaction in the synthesis of novel
compounds of biological interest.
<5
OH
B
OH
6
7
8
9
64
73
77
OH
OH
F3C
B
10
11
O
OH
B
HO
OH
OH
Supporting Information for this article is available online at
http://www.thieme-connect.com/ejournals/toc/synlett.
O
Me
B
N
OH
Me
Acknowledgment
9
3
3
40
O
O
B
The authors would like to thank Hitesh Sanganee, Austen Pimm,
Roger Bonnert, Fraser Hunt, Prem Meghani, Theresa Humphries,
and Jeff Stonehouse for valuable discussions and support during
this investigation.
10
92b
BF3K
a Reactions were performed using boronic acid (2 equiv), copper(II)
acetate (2 equiv), base (2 equiv), and 4 Å MS at r.t. for 20 h.
b Reaction was performed in refluxing MeCN for 20 h.
O
N
O
O
Me
N
O
a
b
N
O
13
N
H
O
Br
12
CF3
Scheme 4 H 27 via quinazolinedione arylation. Reagents and conditions: a) N-ethylurea (1 equiv), Cs2CO3 (2 equiv), Pd2(dba)3 (5 mol%),
XantPhos (10 mol%), dioxane, 100 °C, 24 h (77%); b) 3-trifluoromethyl phenylboronic acid (2 equiv), Cu(OAc)2 (2 equiv), Et3N (2 equiv),
4 Å MS, r.t., 16 h (99%).
Synlett 2009, No. 14, 2253–2256 © Thieme Stuttgart · New York

2256
C. S. Guy, T. C. Jones
LETTER
(16) Kim, J.-J.; Park, Y.-D.; Cho, S.-D.; Kim, H.-K.; Chung, H.;
Lee, S.-G.; Falck, J.; Yoon, Y.-J. Tetrahedron Lett. 2004, 45,
8781.
(17) Hall, D. In Boronic Acids: Preparation and Applications in
Organic Synthesis and Medicine; Hall, D., Ed.; John Wiley
and Sons: New York, 2005, 1.
References and Notes
(1) (a) Noda, K.; Nakagawa, A.; Hachiya, T.; Ide, H. DE
2459090, 1975; Chem. Abstr. 1975, 83, 114468.
(b) Wilhelm, R.; Chin, R.; Lipp, R.; Devens, B.; Alvarez, R.
WO 9319068, 1993; Chem. Abstr. 1993, 120, 164213.
(c) Crespo, M. I.; Gracia, J.; Puig, C.; Vega, A.; Bou, J.;
Beleta, J.; Domenech, T.; Ryder, H.; Segarra, V.; Palacios,
J. M. Bioorg. Med. Chem. Lett. 2000, 10, 2661.
(2) Takahashi, M.; Nagaoka, H.; Inoue, K. J. Heterocycl. Chem.
2004, 41, 525.
(3) Tran, T.; Ellsworth, E.; Watson, B.; Sanchez, J.; Showalter,
H.; Rubin, J.; Stier, M.; Yip, J.; Nguyen, D.; Bird, P.; Singh,
R. J. Heterocycl. Chem. 2005, 42, 669.
(4) Baraka, M. Boll. Chim. Farm. 2001, 140, 90.
(5) Bhat, K.; Hayik, S.; Sztandera, L.; Bock, C. QSAR Comb.
Sci. 2005, 24, 831.
(18) Jacobsen, M.; Knudsen, M.; Gothelf, K. J. Org. Chem. 2006,
71, 9183.
(19) A regiochemically controlled synthesis of quinazolinedione
3 was carried out using the route outlined in Scheme 1
(disconnection a). The material so obtained was spectros-
copically identical to that generated by direct arylation of 1,
confirming that the regiochemistry of substitution was
limited to the free nitrogen position and not either of the
available carbonyl oxygens.
(20) Evans, D.; Katz, J.; West, T. Tetrahedron Lett. 1998, 39,
2937.
(6) Goldberg, I. Ber. Dtsch. Chem. Ges. 1906, 39, 1691.
(7) (a) Muci, A. R.; Buchwald, S. Top. Curr. Chem. 2002, 219,
131. (b) Wolfe, J.; Wagaw, S.; Marcoux, J.; Buchwald, S.
Acc. Chem. Res. 1998, 31, 805. (c) Hartwig, J. Angew.
Chem. Int. Ed. 1998, 37, 2046.
(8) (a) Beletskaya, I.; Cheprakov, A. Coord. Chem. Rev. 2004,
248, 2337. (b) Thomas, A.; Ley, S. Angew. Chem. Int. Ed.
2003, 42, 5400.
(9) (a) Goodbrand, H.; Hu, N. J. Org. Chem. 1999, 64, 670.
(b) Patil, N.; Kelkar, A.; Nabi, Z.; Chaudhari, R. Chem.
Commun. 2003, 2460.
(10) Klapars, A.; Antilla, J.; Huang, X.; Buchwald, S. J. Am.
Chem. Soc. 2002, 124, 7421.
(11) (a) Kiyomori, A.; Marcoux, A.; Buchwald, S. Tetrahedron
Lett. 1999, 40, 2657. (b) Antilla, J.; Klapars, A.; Buchwald,
S. J. Am. Chem. Soc. 2002, 124, 11684.
(12) Willis, M.; Snell, R.; Fletcher, A.; Woodward, R. Org. Lett.
2006, 8, 5089.
(13) Chan, D.; Monaco, K.; Wang, R.-P.; Winters, M.
Tetrahedron Lett. 1998, 39, 2933.
(14) Lam, P.; Deudon, K.; Averill, K.; Li, R.; He, M.; DeShong,
P.; Clark, C. J. Am. Chem. Soc. 2000, 122, 7600.
(15) Chan, D. Tetrahedron Lett. 1996, 37, 9013.
(21) Typical Procedure
3-Methylquinazoline-2,4 (1H,3H)-dione (200 mg, 1.14
mmol), 4-methoxyphenylboronic acid (345 mg, 2.27 mmol),
Cu(OAc)2 (412 mg, 2.27 mmol), and Et3N (0.316 mL, 2.27
mmol) were suspended in CH2Cl2 (12 mL) along with 200
mg of activated 4 Å MS. The suspension was allowed to stir
at r.t. for 20 h then partitioned between 1 M HCl (50 mL) and
EtOAc (50 mL). The organic layer was isolated, dried
(MgSO4), and concentrated in vacuo, and the residue was
purified by flash chromatography (15% EtOAc–i-hexane)
to furnish 1-(4-methoxyphenyl)-3-methylquinazoline-2,4-
(1H,3H)-dione (4) as a colourless solid (230 mg, 72%). 1H
NMR (400 MHz, CDCl3): d = 8.1–8.4 (m, 1 H), 7.4–7.6 (m,
1 H), 7.1–7.3 (m, 3 H), 7.0–7.2 (d, 2 H), 6.59 (d, 1 H), 3.89
(s, 3 H), 3.52 (s, 3 H). 13C NMR (100 MHz, CDCl3): d =
162.2, 159.9, 151.2, 141.7, 134.5, 129.9, 129.0, 128.5,
123.0, 115.5, 115.3, 115.0, 55.5, 28.3; MS (ES+): m/z = 281
[M – H]+.
(22) (a) Lam, P.; Vincent, G.; Clark, C.; Deudon, S.; Jadhav, P.
Tetrahedron Lett. 2001, 42, 3415. (b) Collman, J.; Zhong,
M. Org. Lett. 2000, 2, 1233.
Synlett 2009, No. 14, 2253–2256 © Thieme Stuttgart · New York