Efficient preparation of the isoindoline framework via a six component, tandem double A3-coupling and [2+2+2] cycloaddition reaction

Article abstract of DOI:10.1002/adsc.200700500

The preparation of tetrasubstituted isoindolines from three alkyne units, two aldehyde units and a primary amine via three consecutive reactions, two aldehyde-amine-alkyne couplings (A³-couplings) and a final [2+2+2] cycloaddition, in a single

Full text of DOI:10.1002/adsc.200700500

COMMUNICATIONS
DOI: 10.1002/adsc.200700500
Efficient Preparation of the Isoindoline Framework via a S ix
Component, Tandem Double A3-Coupling and [2+2+2]
Cycloaddition Reaction
E. Ryan Bonfielda and Chao-Jun Lia,*
a
Department of Chemistry, McGill University, 801 Sherbrooke Street West, Montreal, Quebec, Canada H3A 2K6
Fax : (+1)-514-398-3797; e-mail: cj.li@mcgill.ca
Received: October 16, 2007; Published online: February 5, 2008
Dedicated to the memory of Prof. Yoshihiko Ito of Kyoto University.
Supporting information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: The preparation of tetrasubstituted isoin-
dolines from three alkyne units, two aldehyde units
and a primary amine via three consecutive reac-
tions, two aldehyde-amine-alkyne couplings (A3-
couplings) and a final [2+2+2]cycloaddition, in a
single synthetic operation, is described. The A3-cou-
plings are catalyzed by copper bromide and the cy-
cloaddition is catalyzed by Wilkinsonꢀs catalyst. It
was found that many catalysts known to be efficient
at the cycloaddition step were not suitable when
this step was part of the tandem reaction sequence.
Wilkinsonꢀs catalyst was found to be unique in its
suitability for the overall domino reaction sequence.
ports of various biological activities of isoindolines
emerged that a comprehensive summary is not possi-
ble in this communication. The most recent reports
show herbicidal activity,[4] dipeptidyl peptidase inhibi-
tion,[5] antagonism of the 5-HT2C receptor,[6] treatment
of cutaneous lupus,[7] metabotropic glutamate recep-
tor potentiator activity,[8] and treatment of cancer,[9]
or abnormal cell growth.[10] Reports of the use of iso-
indolines as key components of pigment compositions
also continue to emerge.[11] Isoindolines have been re-
cently shown to be active ingredients in the natural
commercial drug ant lion.[12] Isoindolines have also
found use as novel spin probes useful for studying
cancer.[13] A number of the active isoindoline deriva-
tives mentioned have 1- and/or 3-oxo substituents
which can be readily prepared by oxidation of the
benzylic positions.[14]
Keywords: alkynylation; copper; cycloaddition re-
action; domino reaction; multicomponent reaction;
rhodium
Besides being efficient and economical, this meth-
odology could be useful for expanding the repertoire
of known useful isoindoline derivatives. The isoindo-
line preparation described herein would be suitable
Tandem/cascade/domino reaction processes have re- for combinatorial chemistry, using a library of various
ceived a good deal of attention over the last decade substitutions on aromatic alkynes and amines. Previ-
due to the efficiency and step-economy inherent in ously reported preparations of isoindolines begin with
the methodology.[1] The hallmark of tandem reactions a relatively complex starting material. Either the aro-
is a considerable increase in molecular complexity re- matic ring is already in place and needs to have ap-
sulting from a single synthetic step.[1] The conversion propriately placed functional groups to construct the
of primary amines, aldehydes, and alkynes to isoindo- dihydropyrrole skeleton of the isoindoline,[15] or a di-
lines in a single synthetic operation is without prece- propargylamine is used as the starting material.[16]
dent. This methodology has potential for application The second example of synthesizing isoindolines by
to the efficient and economical preparation of com- [2+2+2]cycloaddition starting with dipropargyla-
pounds with industrial and pharmaceutical impor- mines has been limited largely to tosylamines.[16] This
tance.
is because the methodology for synthesizing those
amines requires a highly electron-deficient
The ability of an isoindoline to act as a diuretic dipropargylAHCTREUNG
100 times more active than chlorothiazide was report- amine.[17]
ed over four decades ago.[2] The first anti-tumour ac-
tivity for an isoindoline was discovered nearly three
Recently we reported a tandem double direct alk-
AHCTREyUNG nylation of primary followed by secondary imines
decades ago.[3] In the decades to follow so many re- formed in situ (double A3-coupling) to generate di-
370
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 370 – 374

Efficient Preparation of the Isoindoline Framework
COMMUNICATIONS
Table 1. Screening of catalysts for tandem, one-pot double A3-coupling and [2+2+2]cycloaddition using aniline as primary
amine.[a]
Entry CuBr[b] [mol%] [M] [mol%]
T [8C]
Time [h] Solvent
Yield[c]A [%] Yield[c]B [%9
1
2
3
4
5
6
30
None
15
30
30
30
30
30
30
30
15
3
none
RhCl
RhCl
RhCl
RhCl
RhCl
RhCl
40–80
40–80
40–80
60
60
60
8
8
8
17
18
26
8
9
9
Neat
Neat
Neat
Neat
Neat
Neat
Neat
H2O:MeOH, 9:1
H2O:MeOH, 9:1
Toluene
Toluene
Toluene
THF
0
0
81
0
23
3
17
47
5
78
66
73
51
0
N
G
R
N
T
C
39
53
47
29
76
0
0
0
0
0
7[d]
8[f]
9[e]
10[f]
11
12
13
40–80
[Ru
Ru2Cl
RuCl2A
Cp*RuCl
CpCo(CO)2 (6)
Ni(cod)2 (5)/PPh3 (20)
A
(C10H16)2 (2.5)
24–80
57
55
27–100 21
60 21
C
25
19
G
15
T
0
0
[a]
Reaction scale varied from 0.2–0.6 mmol aniline, but always 3 equivs. formaldehyde (37 wt% in H2O) and 5 equivs. phe-
nylacetylene. Neat reactions were run with 9 equivs. phenylacetylene. All reactions were degassed and run under inert at-
mosphere.
[b]
[c]
CuCl, CuI, and CuOTf gave inferior yields.
Yields determined by integrating methylene protons using 400 MHz H NMR and mesitylene as quantitative internal
1
standard.
[d]
[e]
[f]
See Supporting Information for detailed screening of various conditions using Wilkinsonꢀs catalyst.
Catalyst was dichlorobisACHTRE(UGN m-chloro)bis[(1,2,3,6,7,8-h)-2,7-dimethyl-2,6-octadien-1,8-diyl]diruthenium(IV) CAS: [34801-97-3].
Increasing the temperature to 1008C for over 40 h will produce up to a 1:1 ratio of A:B but results in significant degrada-
tion of A even under N2.
propargylamines from a wide variety of aliphatic or increases with decreased loading of Wilkinsonꢀs cata-
aromatic primary amines and aliphatic or aromatic al- lyst, although this also causes lower conversion of di-
kynes.[18] This methodology complements the scope of propargylamine to isoindoline (Table 1, entries 4–6).
dipropargylamines that are readily available. While Ruthenium is known to be an efficient co-catalyst
working on this we envisioned adding an appropriate with copper for A3-couplings[22] and the complexes
catalyst for a [2+2+2]cycloaddition so the dipropar- listed in Table 1, entries 9[23] and 11[24] are known to
gylamines required for the isoindoline synthesis could be excellent catalysts for [2+2+2]cycloadditions. For
be generated in situ and then react immediately to this reason we expected a ruthenium catalyst to be
form the isoindoline in the same pot. Ruthenium, most suitable for the overall tandem reaction se-
cobalt, rhodium, and nickel are the most common quence. The highest expectations were for Cp*RuCl-
transition metals used to catalyze [2+2+2]cycloaddi-
ACHTRE(UNG cod) based on the literature examples that include
tions.[19] Wilkinsonꢀs catalyst is well known for being electron-deficient and relatively electron-rich dipro-
efficient at catalyzing intermolecular [2+2+2]cyclo- pargylamines,[24] yet this catalyst gave no trace of iso-
additions between 1,6-diynes, including dipropargyl- indoline even with a number of variations in reaction
amines, and monoalkynes.[20]
time, temperature, and solvent. As a control the cor-
From Table 1, entries 1 and 2 it is apparent that responding dipropargylamines were isolated and then
copper bromide[21] alone is sufficient to generate the combined with Cp*RuCl
ACHTRE(UNG cod) again under identical
dipropargylamine in good yield under the conditions conditions and isoindoline product was detected in
used, whereas Wilkinsonꢀs catalyst does not catalyze moderate yield by NMR. From this we concluded
the A3-couplings. However, Wilkinsonꢀs catalyst that the catalyst is altered in the early stage of the
seems to interfere with the initial A3-couplings since cascade reaction processes in a way that renders it in-
the overall yield of dipropargylamine and isoindoline effective for the final cycloaddition. The ruthenium
Adv. Synth. Catal. 2008, 350, 370 – 374
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
371

E. Ryan Bonfield[a] and Chao-Jun Li
COMMUNICATIONS
Table 2. Screening of rhodium complexes for tandem, one-
pot double A3-coupling and [2+2+2]cycloaddition using
aniline as primary amine.[a]
Table 3. Tandem, one-pot double A3-coupling and [2+2+
2]cycloaddition for various primary amines.[a]
Entry [Rh]
Time
[h]
Yield[c]
[%]
A
Yield[c]
[%]
B
1
Rh
AgBF4
Rh(PPh3)3Cl//
AgOTf
[Rh(cod)Cl]2/
P(Me)3
Rh(cod)Cl]2/
P(Cy)3
G
23
22
18
26
29
41
6
0
2
14
0
0
Entry
1
Product
Yield[b] [%]
78 (69)
3[b]
4[b]
45
34
0
5[b]
6
7
Rh
Rh
Rh
Rh
A
G
R
29
3
14
21
0
0
21
13
U
2
62 (58)
66 (57)
33 (24)
60 (51)
15 (13)
86 (78)
85 (76)
76 (69)
84 (70)
8
ACHTREUNG(cod)2BF4/PPh3 21
[a]
Reaction scale was 0.4 mmol aniline, with 3 equivs. form-
aldehyde (37 wt% in H2O) and 5 equivs. phenylacetylene
in 0.25 mL solvent.
Paraformaldehyde providing 3 equivs. of formaldehyde
was used in place of the regular aqueous formadehyde
solution and the reaction was carried out under anhy-
drous conditions.
[b]
3
[c]
Yields determined by integrating methylene protons
4
1
using 400 MHz H NMR and mesitylene as quantitative
internal standard.
Isoindoline (see Table 1).
Dipropargylamine (see Table 1).
[d]
[e]
5
triphenylphosphine and p-cymene complexes gave
isoindoline in low yield under the cascade conditions
only if the temperature was raised to over 1008C for
over 24 h and no isoindoline was observed with mod-
erate temperature and time (Table 1, entries 8 and
10). The nickel[25] and cobalt[26] catalysts used are also
quite common and well known for efficient [2+2+
2]cycloadditions(Table 1, entries 12 and 13). Howev-
er, copper-catalyzed A3-couplings did not occur in
their presence so that the potential for efficient cyclo-
additions under the tandem reaction conditions could
not be evaluated.
Variations in counter ion and ligand were then ex-
plored for rhodium complexes (Table 2, entries 1–8).
Electron-rich phosphine ligands rendered the rhodi-
um complex incapable of catalyzing the cycloaddition
reaction (Table 2, entries 3 and 4). Chlorine proved to
be the best counter ion when compared to other com-
monly used counter ions for the cycloaddition step
(Table 2, entries 1 and 2, 5–8) although all were capa-
ble of producing isoindoline under the tandem reac-
tion conditions.
6
7
8
9
10
Clearly the most electron-rich amines gave the
worst results (Table 3, entries 4 and 6). A methyl
372
asc.wiley-vch.de
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 370 – 374

Efficient Preparation of the Isoindoline Framework
COMMUNICATIONS
was filtered open to air through ~1–2 cm of silica with ethyl
acetate. Ethyl acetate was removed with rotary evaporation
followed by high vacuum for 1 h before dissolving in 1 mL
of CDCl3 for NMR analysis of the crude reaction mixture.
Table 3. (Continued)
Entry
Product
Yield[b] [%]
72 (65)
11
Acknowledgements
[a]
Reactions were run neat with 0.56 mmol primary amine,
3 equivs. formaldehyde (135 mL, 37 wt % in H2O) and
9 equivs. phenylacetylene; Reaction mixture was de-
gassed and agitated for 2 h at 408C and 6 h at 808C.
Yields determined by integrating methylene protons
We are grateful to Canada Research Chair (Tier I)founda-
tion (to CJL), NSERC, and CFI for support of our research.
ERB thanks NSERC for a Postgraduate Fellowship.
[b]
1
using 400 MHz H NMR and mesitylene as quantitative
internal standard. Isolated yields after flash chromatogra-
phy shown in parenthesis.
References
[1] For a review see: a) K. C. Nicolaou, D. J. Edmonds,
P. G. Bulger, Angew. Chem. 2006, 118, 7292; Angew.
Chem. Int. Ed. 2006, 45, 7134; b) A. Padwa, Pure Appl.
Chem. 2004, 76, 1933; c) P. J. Parsons, C. S. Penkett,
A. J. Shell, Chem. Rev. 1996, 96, 195; d) R. A. Bunce,
Tetrahedron 1995, 51, 13103.
[2] E. J. Cornish, G. E. Lee, W. R. Wragg, Nature 1963,
197, 1296.
[3] J. P. Lambooy, J. Med. Chem. 1978, 21, 301.
[4] M.-Z. Huang, K.-L. Huang, Y.-G. Ren, M.-X. Lei, L.
Huang, Z.-K. Hou, A.-P. Liu, X.-M. Ou, J. Agric. Food
Chem. 2005, 53, 7908.
[5] a) M.-B. Maes, V. Dubois, I. Brandt, A.-M. Lambeir, P.
Van der Veken, K. Augustyns, J. D. Cheng, X. Chen, S.
Scharpꢂ, I. De Meester, J. Leukocyte Biol. 2007, 81,
1252; b) W.-T. Jiaang, Y.-S. Chen, T. Hsu, S.-H. Wu, C.-
H. Chien, C.-N. Chang, S.-P. Chang, S.-J. Lee, X. Chen,
Bioorg. Med. Chem. Lett. 2005, 15, 687.
[6] D. Hamprecht, F. Micheli, G. Tedesco, A. Checchia, D.
Donati, M. Petrone, S. Terreni, M. Wood, Bioorg. Med.
Chem. Lett. 2007, 17, 428.
group in the meta position provides less inductive
electron donation than either a methyl or a methoxy
group in the para position, (Hammett substituent pa-
rameters of s=À0.06, À0.14, and À0.12, respective-
ly)[27], which could help to explain the difference be-
tween entries 5 and 6 in Table 3. Electron deficiency,
on the other hand, impairs the A3-couplings since
highly electron-deficient amines such as tosylamines
are incapable of undergoing this reaction. Thus, al-
though there appears to be an advantage to electron-
withdrawing substituents (Table 3, entries 1, 7, 8, 10
and 11) there is an upper limit as seen by the de-
crease in yield when p-trifluoromethane or two chloro
substituents are present (Table 3, entries 2 and 3).
When either aliphatic alkynes or amines were used
under the optimal conditions (Table 3) isoindoline
could be detected by mass spectroscopy, but conver-
sion was too low to isolate the product.
[7] J. B. Zeldis, P. E. W. Rohane, P. H. Schafer, US Patent
Appl. US 2006–645319 20061221, 2007; Chem. Abstr.
2007, 147, 110212.
This methodology provides a convenient, efficient,
and economical way to screen new isoindolines which
have enormous potential for industrial application.
Further work to overcome the limitations with ali-
phatic substrates is ongoing.
[8] B. Van Wagenen, R. Ukkiramapandian, J. Clayton, I.
Egle, J. Empfield, M. Isaac, F. Ma , A. Slassi, G. Steel-
man, R. Urbanek, S. Walsh, PCT Int. Appl. WO
2007021309, 2007; Chem. Abstr. 2007, 146, 274221.
[9] a) G. W. Muller, H.-W. Man, PCT Int. Appl. WO
2006025991, 2006; Chem. Abstr. 2006, 144, 369911;
b) C. Murakata, N. Amishiro, T. Atsumi, Y. Yamashita,
T. Takahashi, R. Nakai, H. Tagaya, H. Takahashi, J. Fu-
nahashi, J. Yamamoto, Y. Fukuda, PCT Int. Appl. WO
2006112479, 2006; Chem. Abstr. 2006, 145, 454932;
c) L. A. Funk, M. C. Johnson, P.-P. Kung, J. J. Meng,
J. Z. Zhou, PCT Int. Appl. WO 2006117669, 2006;
Chem. Abstr. 2006, 145, 489014.
[10] G. Chessari, M. S. Congreve, E. Figueroa Navarro, M.
Frederickson, C. Murray, A. J.-A. Woolford, M. G.
Carr, R. Downham, M. A. OꢀBrien, T. R. Phillips, A. J.
Woodhead, PCT Int. Appl. WO 2006109085, 2006;
Chem. Abstr. 2006, 145, 438416.
[11] D. Sasaki, T. Fujiwara, Jpn. Kokai Tokkyo Koho JP
2005304565, 2006; Chem. Abstr. 2007, 146, 472371.
[12] T. Nakatani, E. Nishimura, N. Noda, J. Nat. Med. 2006,
60, 261.
Experimental Section
Typical Procedure for the Tandem Double A3-
Coupling and [2+2+2]Cycloaddition Reaction
Copper bromide and Wilkinsonꢀs catalyst were combined in
a screw cap tube open to air. Phenylacetylene, formaldehyde
solution (37 wt % in water), and primary amine were then
added in that order in open air via syringe. Solid amines
were also added last for consistency. The tube was then
sealed with a cap fitted with a valve connected to a Schlenk
line and after a five minute initial stir the reaction mixture
was degassed by freeze-pump-thaw three times using liquid
nitrogen and then left with an overpressure of nitrogen gas
at 408C for 6 h. After 6 h the temperature was increased to
808C for two additional hours and then the reaction mixture
Adv. Synth. Catal. 2008, 350, 370 – 374
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
373

E. Ryan Bonfield[a] and Chao-Jun Li
COMMUNICATIONS
[13] G.-P. Yan, D. Bischa, S. E. Bottle, Free Radical Biol.
Med. 2007, 43, 111.
[14] A. J. Catino, J. M. Nichols, H. Choi, S. Gottipamula,
[20] a) M. R. Tracey, J. Oppenheimer, R. P. Hsung, J. Org.
Chem. 2006, 71, 8629; b) R. Grigg, R. Scott, P. Steven-
son, J. Chem. Soc., Perkin Trans. 1 1988, 1357; c) R.
Grigg, R. Scott, P. Stevenson, Tetrahedron Lett. 1982,
23, 2691; d) D. D. Young, R. S. Senaiar, A. Deiters,
Chem. Eur. J. 2006, 12, 5563; e) Q. Sun, X. Zhou, K.
Islam, D. J. Kyle, Tetrahedron Lett. 2001, 42, 6495.
[21] a) J. Zhang, C. Wei, C.-J. Li, Tetrahedron Lett. 2002, 43,
5731; b) L. W. Bieber, M. F. da Silva, Tetrahedron Lett.
2004, 45, 8281.
M. P. Doyle, Org. Lett. 2005, 7, 5167.
[15] For example: a) Y. Ju, R. S. Varma, J. Org. Chem. 2006,
71, 135; b) D.-R. Hou, Y.-D. Hsieh, Y.-W. Hsieh, Tetra-
hedron Lett. 2005, 46, 5927; c) T. M. Barnard, G. S.
Vanier, M. J. Collins, Jr., Org. Process Res. Dev. 2006,
10, 1233; d) K.-I. Fujita, Y. Enoki, R. Yamaguchi, Org.
Synth. 2006, 83, 217; e) I. Takahashi, K. Nishiuchi, R.
Miyamoto, M. Hatanaka, H. Uchida, K. Isa, A. Sakush-
ima, S. Hosoi, Lett. Org. Chem. 2005, 2, 40; f) T. P.
Zabawa, D. Kasi, S. R. Chemler, J. Am. Chem. Soc.
2005, 127, 11250.
[16] For example: a) J. Clayden, W. Moran, J. Org. Biomol.
Chem. 2007, 5, 1028; b) P. Novꢃk, R. Pohl, M. Kotora,
M. Hocek, Org. Lett. 2006, 8, 2051; c) K. Tanaka, K.
Takeishi, K. Noguchi, J. Am. Chem. Soc. 2006, 128,
4586; d) H. Kinoshita, H. Shinokubo, K. Oshima, J.
Am. Chem. Soc. 2003, 125, 7784; e) K. Tanaka, G. Nish-
ida, H. Sagae, M. Hirano, Synlett 2007, 1426.
[17] a) M. Nishida, H. Shiga, M. Mori, J. Org. Chem. 1998,
63, 8606; b) A. R. Katritzky, K. Yannakopoulou, P.
Lue, D. Rasala, L. Urogdi, J. Chem. Soc., Perkin Trans.
1 1989, 225; c) A. R. Katritzky, M. S. C. Rao, J. Chem.
Soc., Perkin Trans. 1 1989, 2297; d) R. Gleiter, J. Ritter,
H. Irngartinger, J. Lichtenthaler, Tetrahedron Lett.
1991, 32, 2883.
[22] a) C. Wei, J. T. Mague, C.-J. Li, Proc. Natl. Acad. Sci.
U. S. A. 2004, 101, 5749; b) C. Wei, C.-J. Li, Chem.
Commun. 2002, 268; c) C. Wei, Z. Li, C.-J. Li, Synlett
2004, 1472.
[23] V. Cadierno, S. E. Garcꢅa-Garrido, J. Gimeno, J. Am.
Chem. Soc. 2006, 128, 15094.
[24] a) Y. Yamamoto, K. Hattori, H. Nishiyama, J. Am.
Chem. Soc. 2006, 128, 8336; b) Y. Yamamoto, K. Kin-
para, T. Saigoku, H. Takagishi, S. Okuda, H. Nishiya-
ma, K. Itoh, J. Am. Chem. Soc. 2005, 127, 605; c) Y. Ya-
mamoto, T. Hashimoto, K. Hattori, M. Kikuchi, H.
Nishiyama, Org. Lett. 2006, 8, 3565; d) Y. Yamamoto,
T. Arakawa, R. Ogawa, K. Itoh, J. Am. Chem. Soc.
2003, 125, 12143.
[25] P. Turek, P. Novꢃk, R. Pohl, M. Hocek, M. Kotora, J.
Org. Chem. 2006, 71, 8978.
[26] For a review see: V. Gandon, C. Aubert, M. Malacria,
Chem. Commun. 2006, 2209.
[27] C. D. Ritchie, W. F. Sager, Prog. Phys. Org. Chem.
1964, 2, 323.
[18] E. R. Bonfield, C.-J. Li, Org. Biomol. Chem. 2007, 5,
435.
[19] For a review see: a) P. R. Chopade, J. Louie, Adv.
Synth. Catal. 2006, 348, 2307; b) H.-W. Frꢄhauf, Chem.
Rev. 1997, 97, 523.
374
asc.wiley-vch.de
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 370 – 374