M. Krasavin et al. / Tetrahedron Letters 50 (2009) 5945–5950
5949
Table 2 (continued)
Compound
R1
Cl
R2
H
R3
Isolated yield (%)
4l
54
78
67
*
O
4m
4n
Cl
Cl
H
H
*
*
O
3H-indol-3-ol structure; relieving this strain could be a substantial
driving force for the proposed skeletal rearrangement; (iv) when a
similar reaction was attempted with o-acetylaniline, a complex
mixture of unidentified products was obtained (Scheme 4); this
is possibly due to an alternative pathway existing for the dehydra-
tion of 5; and (v) the present reaction does not work with aromatic
isocyanides (vide infra): according to the mechanism, the lone pair
of the exocyclic nitrogen atom would not be available to partici-
pate in the rearrangement process for intermediates such as 3a0
bearing aromatic R groups.
requiring an initial three-center, two-component Passerini-type
reaction with subsequent skeletal rearrangement of the 3H-in-
dol-3-ol framework has been proposed.
3. Typical procedure
3.1. Synthesis of 3,4-dihydroquinazolin-4-ols 4
A solution of o-aminobenzophenone 1 (5 mmol) and isocyanide
2
(5.5 mmol) in CH2Cl2 (25 mL) was treated with BF3ꢀOEt2
We attempted to probe the mechanism by exposing 3a to
1 equiv of BF3ꢀOEt2 in dichloromethane at room temperature.
Shortly after the addition of the Lewis acid, a precipitate formed
that remained unchanged on stirring the reaction mixture for
72 h. Upon basic aqueous work-up, 3a was fully recovered and
no trace of 4a was detected. These observations are the likely result
of boron trifluoride complexation to the ‘amidine’ moiety of 3a
(and not to the sterically encumbered hydroxy group). This, in
our opinion, does not contradict the proposed tentative mecha-
nism: the skeletal rearrangement of 3a0 is triggered by boron al-
ready complexed to the oxygen atom as a result of the preceding
three-center two-component reaction with the isocyanide and in-
deed, can occur only in the course of the reaction. However, further
studies are needed to provide more evidence for the plausibility of
the proposed mechanism.
We prepared several other 3,4-dihydroquinazolin-4-ols 4 from
commercially available o-aminobenzophenones 1 and isocyanides
2 (Scheme 5, Table 2). Notably, we found that this reaction did not
proceed with aromatic isocyanides, which is consistent with the
proposed reaction mechanism (vide supra). The isolated yields of
4a–n were moderate to good and all the products were character-
ized by 1H and 13C NMR spectroscopy, LC–MS and elemental analy-
ses; X-ray analysis was also performed on representative compound
4e to confirm the generality of the reaction (see Supplementary
data).12
(5.0 mmol) and the resulting mixture was stirred at rt for 8–16 h.
The reaction mixture was washed with 5% aqueous NaOH
(10 mL) and H2O (2 ꢂ 10 mL). The combined aqueous solutions
were back-extracted with CH2Cl2 (3 ꢂ 10 mL). The combined or-
ganic extracts were dried over anhydrous Na2SO4, filtered, and
concentrated in vacuo. The products were purified either by crys-
tallization from MeOH or by chromatography on silica gel using
5% MeOH in CH2Cl2 as eluent.
3.2. 3-(4-Methylbenzyl)-4-phenyl-3,4-dihydroquinazolin-4-ol
(4a)
Beige solid, mp = 145–147 °C; 1H NMR (300 MHz, DMSO-d6) d
7.43 (d, J = 5.2 Hz, 2H), 7.33 (m, 2H), 7.25 (d, J = 5.2 Hz, 2H), 7.15
(m, 1H), 7.03–7.11 (m, 5H), 6.93 (m, 2H), 4.17 (ABq, DmAB = 14.3 Hz,
J = 11.6 Hz, 2H), 3.32 (s, 1H), 2.24 (s, 3H); 13C NMR (75 MHz, DMSO-
d6) d 147.6, 146.0, 140.9, 136.3, 135.2, 128.8, 128.7, 128.2 (2 sig-
nals), 128.0, 127.9, 127.5, 126.4, 124.0, 123.9, 84.3, 48.8, 20.7; LC–
MS (M+H) 329. Anal. Calcd for C22H20N2O: C, 80.46; H, 6.14; N,
8.53. Found: C, 80.51; H, 6.19; N, 8.55.
Acknowledgment
We would like to thank Dr. Alexander Manaev of Chemical
Diversity Research Institute for his help in obtaining X-ray crystal-
lographic data.
3,4-Dihydroquinazolin-4-ols are not well represented in the lit-
erature. They have been isolated and characterized as metabolites
of hypoglycemic 5-phenyl-1,3,4-benzotriazepines.13 The chemical
syntheses of 3,4-dihydroquinazolin-4-ols14,15 are quite elaborate
which may explain the scarcity of the literature reports. We expect
these compounds to serve as bioisosteric replacements for the
CNS-active 3H-indol-3-ols and hope that their rapid synthesis de-
scribed herein will lead to more extensive biological evaluation
of the 3,4-dihydroquinazolin-4-ol scaffold.
Supplementary data
Supplementary data (characterization data for the newly syn-
thesized compounds (3a, 4a–n), X-ray crystallographic files (CIF)
for compounds 4a and 4e) associated with this article can be found,
In conclusion, we have described a novel BF3ꢀOEt2-promoted
reaction of o-aminobenzophenones with aliphatic isocyanides.
Contrary to expectations, hitherto poorly described 3,4-dihydro-
quinazolin-4-ols were formed as products. A tentative mechanism
References and notes
1. (a) Kiselyov, A. S. Tetrahedron Lett. 2005, 46, 4851–4854; (b) Masdeu, C.;
Gomez, E.; Williams, N. A.; Lavilla, R. QSAR Comb. Sci. 2006, 25, 465–473; (c)