N. Y. Kim, C.-H. Cheon / Tetrahedron Letters 55 (2014) 2340–2344
2343
O
N
did not proceed to completion and 5a was obtained as the minor
O
O
O
product along with 3a (Eq. 2). Because DMSO is often used in oxi-
dation reactions,15 we assumed that DMSO might have some ben-
eficial effect on the above aerobic oxidative cyclization reaction.
When the reaction was carried out in the presence of 3 equiv of
DMSO in DMF solution, delightfully, the aerobic oxidative cycliza-
tion went smoothly to afford 5a in quantitative yields. Further-
more, the resulting 5a was directly converted into 9 with POCl3
in the same pot and the subsequent addition of aniline afforded
4-aminoquinazoline 10 in synthetically useful yield over three
steps, even without complete optimization of the one-pot reaction
conditions (Eq. 3).16
NH2
NH2
NH2
NH2
H2O
2
R
R
R
N
H
OH
1
3
6'
6-exo-tet
cyclization
(pathway b)
6-endo-trig
cyclization
(pathway a)
O
O
[O]
NH
NH
N
R
N
H
4
R
5
Aldehyde equivalents could be also subjected to this protocol
(Scheme 3).17 For example, when 3,3-dihydro-2H-pyrone 11, com-
monly used as a 5-hydroxypentanal equivalent, was used in this
transformation in the presence of 10 mol % of para-toluenesulfonic
acid (PTSA), the reaction smoothly proceeded to afford the corre-
sponding quinazolinone 12 in good yield under the standard con-
ditions with aldehydes (Eq. 4). Moreover, the resulting
quinazolinone 12 was directly employed in the Mitsunobu reac-
tion18 to yield mackinazolinone 1319 in synthetically useful yields
in the same pot although all the reaction conditions were not opti-
mized (Eq. 5).
With these results in hand, we attempted to gain information
about the reaction mechanism for this transformation. During opti-
mization, the quinazolinone was found to be formed in the pres-
ence of molecular sieves without any assistance of a nucleophile
(Table 1, entry 6). This result suggested that under these condi-
tions, quinazolinone 5 could be formed by the direct cyclization
of 3 via 6-endo-trig cyclization (pathway a in Scheme 1) although
the yield was low. In addition, it was observed that the choice of
a nucleophile had an influence on the efficiency of this transforma-
tion (Table 1, entries 1–5). These results also supported our work-
ing hypothesis where a nucleophile might be involved in the
cyclization of imine 3 via the 6-exo-tet cyclization of intermediate
6 formed from 3 with the nucleophile (pathway b). Particularly,
since it was observed that the formation of 5 was significantly
accelerated in the absence of molecular sieves and a nucleophile
(Table 1, entry 8), we expected that water could be the nucleophilic
catalyst for the cyclization of 3 into 4, which eventually accelerates
the formation of quinazolinone 5.
Scheme 4. Proposed reaction pathway.
formed from the reaction of 3 with water and the 6-exo-tet cycliza-
tion of intermediate 60 occured along with uncatalyzed 6-endo-trig
cyclization of 3. Since the 6-exo-tet cyclization from 60 was much
faster than the 6-endo-trig cyclization from 3, the yield of quinaz-
olinone was significantly increased under such conditions. This
proposed mechanism also rationalized the significant increase in
the yield of quinazolinone from anthranilamide 1 and aldehyde
2. In the presence of molecular sieves (in the absence of water), ini-
tially formed intermediate 60 was rapidly converted into imine 3,
which could undergo cyclization via only 6-endo-trig cyclization.
However, in the absence of molecular sieves (in the presence of
water), the intermediate 60 readily underwent cyclization through
6-exo-tet cyclization, leading to the desired product in comparable
yields with those from imine 3.
In conclusion, we have developed a highly environmentally be-
nign protocol for the synthesis of 2-substituted and 2,3-disubsti-
tuted quinazolinones from anthranilamides and aldehydes via
aerobic oxidative cyclization in wet DMSO without any additives.
This new protocol features operational simplicity, high atom econ-
omy, and broad substrate scope. The usefulness of this new proto-
col was further demonstrated by the direct application of the
resulting quinazolinones to Vilsmeier and Mitsunobu reactions in
the same pot without their isolations. Further application of this
protocol to total synthesis of biologically important natural prod-
ucts and more detailed mechanistic studies for this transformation
are currently underway in our laboratory.
Based on these results, we proposed a possible reaction mecha-
nism where water would act as a nucleophilic catalyst for this pro-
tocol (Scheme 4). The quinazolinone 5 could be obtained from
imine 3 via either the direct 6-endo-trig cyclization (pathway a)
of 3 or the 6-exo-tet cyclization (pathway b) of intermediate 60
formed by the reaction of 3 with water. Since the formation of
intermediate 60 was intrinsically impossible in the absence of
water, 5 was obtained in low yields through 6-endo-trig cyclization
of 3. However, in the presence of water, intermediate 60 could be
Acknowledgments
This work was partly supported by Basic Science Research Pro-
gram through the National Research Foundation of Korea (NRF)
funded by the Ministry of Science, ICT and Future Planning
(2013R1A1A1008434). C.-H.C. also thanks the Ministry of Educa-
tion (NRF20100020209) for financial support from the NRF fund.
Supplementary data
O
O
OH
NH
PTSA (10 mol%)
open flask
Supplementary data associated with this article can be found, in
NH2
NH2
(4)
(5)
+
O
11
DMSO, 100 o
24 h
C
N
12
51 %
1a
References and notes
11
O
O
OH
NH
O
2176; NaHSO3: (b) López, S. E.; Rosales, M. E.; Urdaneta, N.; Godoy, M. V.;
PTSA (10 mol%)
DIAD
NH2
open flask
N
DMSO, 100 o
24 h
C
PPh3
rt, 12 h
NH2
1a
N
12
N
13
56 % over two-step
not isolated
Scheme 3. Application of aldehyde equivalent 11 and direct further
functionalization.