1864
A. Srivastava et al. / Tetrahedron Letters 55 (2014) 1863–1867
Cl
MeO
OMe
Ph
O
O
O
O
N
N
N
O
CF3
HN
N
NR
N
Me
O
O
H
Me
Me
N
N
Ph
O
O
H
N
O
O
O
Human ELOVL6
Human ELOVL3
Aldolase Inhibitors
O
(+)-Lennoxamine
(R)-Pazinaclone
Figure 1. Natural product and biologically active compounds that have isoindolin-1-one and dihydroindol-2-one moieties.
ketimine 4 in the presence of Brønsted acid to access this
spirocyclic compound 3 (Scheme 1). During our continuing efforts
toward the development of metal-free based domino protocols as
well as synthesis of N-heterocycles,10 we report a simple, mild,
and robust procedure for the synthesis of 1-aryl/alkyl-substi-
tuted-6,7-dihydrospiro[indole-3,10-isoindole]-2,30,4(1H,5H)-trione
derivatives via a domino dehydration/condensation/cyclization
sequence reaction of cyclic enaminones with 3-ethoxycarbonyl-
3-hydroxyisoindolin-1-one derivative using ( )-CSA as a Brønsted
acid catalyst.
Table 1
Domino reaction of 5,5-dimethyl-3-(phenylamino)cyclohex-2-enone (2aa) with 3-
ethoxycarbonyl-3-hydroxyisoindolin-1-one (1a)a
O
O
O
Catalyst (30 mol%)
Solvent, heating
O
NH
NH
CO2Et
Me
Me
O
PhHN
HO
1a
N
Ph
Me
Me
2aa
3aaa
Entry
Catalyst
Solvent
T (°C)
T (h)
Yieldb (%)
In order to establish the optimal conditions, we tested the mod-
el reaction between 3-ethoxycarbonyl-3-hydroxyisoindolin-1-one
(1a) and 5,5-dimethyl-3-(phenylamino)cyclohex-2-enone (2aa)
under a variety of reaction conditions (solvent, catalyst and tem-
perature). The results are summarized in Table 1. It was observed
that ( )-camphor-10-sulfonic acid was unable to trigger this
reaction at room temperature in CH2Cl2 medium (Table 1, entry
2). However, at 40 °C, after 24 h, a very low yield (19%) of 6,6-di-
methyl-1-phenyl-6,7-dihydrospiro[indole-3,10-isoindole]-
1c
2c
3
4
5
Nil
CH2Cl2
CH2Cl2
CH2Cl2
CHCl3
Toluene
Toluene
Xylene
rt
rt
40
60
24
24
24
24
24
24
24
24
24
24
24
40
24
24
24
30
30
NR
NR
19
37
45
82
79
25
51
79
77
33
72
73
74
69
57
CSA
CSA
CSA
CSA
CSA
CSA
CSA
CSA
CSA
CSA
TCA
pTSA
TFA
TfOH
H2SO4
HCl
60
90
90
90
90
110
120
90
90
90
90
90
90
6
7
8d
9
EtOH
Dioxane
Toluene
Xylene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
10
11
12
13
14
15
16
17
2,30,4(1H,5H)-trione (3aaa) was obtained (entry 3). The product
was fully characterized by its spectroscopic data (1H, 13C NMR
and HRMS). Gratifyingly, these significant results motivated us to
examine this domino reaction in detail. For this catalyst, further
screening of other organic solvents such as toluene, CHCl3, xylene,
dioxane, and EtOH was carried out under heating conditions (en-
tries 4–9). The results clearly demonstrated that at 90 °C, toluene
and xylene provided the highest yields (82% and 79% respectively,
entries 6 and 7) of targeted spiro product 3aaa than those using
other organic solvents (entries 8 and 9). On further increasing
the reaction temperature from 90 °C to 120 °C in xylene, no
improvement in yield (77%) and reaction time was observed (entry
11). Considering the lower boiling point of toluene, it was chosen
as the best solvent for all the further reactions. Next, we examined
several well known Brønsted acid catalysts namely pTSA, trichloro-
acetic acid (TCA), TFA, TfOH, H2SO4, and HCl for this domino pro-
cess. As shown in Table 1, all these catalysts were able to
promote this reaction effectively (except TCA, 33% yield, entry
12), resulting in moderate to good yields of desired compound
3aaa (57–74%, entries 13–17).
A reasonable mechanism for the formation of compound 3aaa
has been proposed as shown in Scheme 2. This domino reaction
is thought to proceed via an iminium ion mode of activation mech-
anism. At first, cyclic ketimine 4a is generated from 1a through a
dehydration process. In the second step, the nitrogen atom of cyclic
ketimine 4a is protonated by CSA to form cyclic iminium ion 5,
which is subsequently attacked by cyclic enaminone 2aa, leading
to intermediate 6. The latter undergoes an imine-enamine tautom-
erism to form condensation product 7, which, finally, is converted
into 3aaa via intramolecular cyclization.
a
Unless otherwise mentioned, all the reactions were carried out with compound
1a (0.2 mmol), 2aa (0.25 mmol) and catalyst (0.06 mmol, 30 mol %) in specified dry
solvent (0.5 mL) and temperature.
b
Yield of isolated product after column chromatography.
NR = no reaction.
Reaction was performed in the sealed tube.
c
d
1-one derivatives 1 under our standard conditions. The results
are summarized in Table 2. As is evident in Table 2, a series of
six-membered cyclic enaminones derived from several aromatic
amines with various substituents on phenyl rings (entries 1–9)
condensed efficiently (74–85%) with substrate 1a, providing a
straightforward way to construct the previously unknown class
of spiro products possessing both isoindolin-1-one and dihydroin-
dol-2-one frameworks. Importantly, N-benzyl substituted cyclic
enaminone (entry 10) also resulted in good yield when substrate
1a was employed. Not only 1a but also 5-halide-substituted 3-eth-
oxycarbonyl-3-hydroxyisoindolin-1-ones (1b–e) too were wit-
nessed to be good enamine acceptors. For example, after 24 h,
high yield (80–85%) of the corresponding products (3baa–3eaa,
entries 11–14) were obtained when 5,5-dimethyl-3-(phenylami-
no)cyclohex-2-enone (2aa) was used. Similarly, substrate 1a
underwent clean reactions of several 3-(arylamino)-cyclohex-2-
enones by this protocol providing the corresponding anticipated
products in good yields (75–78%, entries 15–17). By this synthetic
operation, several functional groups including Me, OMe, Bn, Cl, Br,
With these acceptable results in hand, we examined the scope
of this domino process by using several N-aryl/alky-substituted
cyclic enaminones 2 with 3-ethoxycarbonyl-3-hydroxyisoindolin-