A tandem heck-carbocyclization/suzuki-coupling approach to the stereoselective syntheses of asymmetric 3,3-(diarylmethylene)indolinones

Article abstract of DOI:10.1021/jo050016d

(Chemical Equation Presented) An efficient and versatile method for stereoselective synthesis of (E)-3,3-(diarylmethylene)indolinones by a palladium-catalyzed tandem Heck-carbocyclization/Suzuki-coupling sequence is presented. Factors influencing yield and selectivity, namely catalyst, coordinating ligand, and solvent, are detailed.

Full text of DOI:10.1021/jo050016d

A Tandem Heck-Carbocyclization/
Suzuki-Coupling Approach to the
Stereoselective Syntheses of Asymmetric
3,3-(Diarylmethylene)indolinones
Wing S. Cheung, Raymond J. Patch, and
Mark R. Player*
FIGURE 1. Indolinone-based kinase inhibitors.
Drug Discovery, Johnson & Johnson Pharmaceutical
Research and Development, L.L.C., 8 Clarke Drive,
Cranbury, New Jersey 08512
Only a limited number of medicinally relevant 3-(di-
arylmethylenyl)indolinones have been disclosed. These
tend to be symmetrically substituted derivatives such as
3-[bis(4-methoxyphenyl)methylene]-1,3-dihydroindol-2-
one, which are readily accessible by condensation reac-
tions between an oxindole and a symmetrical diaryl
ketone, bis-(4-methoxyphenyl)methanone.5
mplayer@prdus.jnj.com
Received January 4, 2005
Performing this intermolecular condensation with an
unsymmetrical diaryl ketone would be expected to give
rise to a mixture of both geometrical isomer products,
absent overriding directing group effects. The lack of a
general method for the synthesis of unsymmetrically
substituted 3-(diarylmethylenyl)indolinone isomers may
partially explain why these systems have not been more
extensively investigated. Herein, we report the develop-
ment of a versatile stereoselective synthesis of asym-
metric 3-(diarylidenyl)indolinones, 2. Our method in-
volves a two-step tandem Heck-carbocyclization/Suzuki-
coupling process starting from 2-iodoanilines and aryl-
propionic acids.6
An efficient and versatile method for stereoselective syn-
thesis of (E)-3,3-(diarylmethylene)indolinones by a pal-
ladium-catalyzed tandem Heck-carbocyclization/Suzuki-
coupling sequence is presented. Factors influencing yield and
selectivity, namely catalyst, coordinating ligand, and solvent,
are detailed.
To introduce asymmetry at the exocyclic methylene
carbon, we sought a method by which two aryl groups
could be incorporated at this site independent of one
another. Retrosynthetically (Scheme 1) we envisioned a
process wherein an intramolecular Heck-carbocyclization
of an arylpropynamide 5 might favor a single (E)-
vinylpalladium intermediate, 4. Subsequent coupling
with an aryl boronic acid 3 would serve to incorporate
the second aryl substituent in a stereoselective fashion,
provided that this kinetic vinylpalladium species would
resist isomerization to the thermodynamic (Z)-configu-
ration under conditions of the reaction. Arylpropyna-
mides 5 in turn, would be readily accessible from
arylpropynoic acids (6) and 2-iodoanilines (7) (Table 1).
Unsubstituted phenylboronic acid was initially chosen
in order to test the feasibility of this sequence and for
carrying out a cursory probe of coupling conditions. Thus,
phenylpropiolic acid (6) was converted to its acid chloride
The 2-indolinone system has been incorporated in a
significant number of pharmaceutically relevant com-
pounds. Potential utilities for such compounds have been
identified in virtually all of the major therapeutic areas
including oncology, inflammation, CNS, immunology, and
endocrinology. Indolinones bearing a substituted exocy-
clic methylene at the 3-position have gained particular
prominence in the fields of oncology and inflammation,
as this pharmacophore has demonstrated a unique ability
to inhibit kinase activities through a competitive interac-
tion at the ATP binding sites of these enzymes. While
many 3-(arylidenyl)indolinone analogues have been ex-
tensively evaluated for kinase inhibitory activities, these
have by and large been monoaryl-substituted methyli-
denyl analogues, e.g., SU66681 and SU11248 (Figure 1).2
3-(Arylaminomethylenyl)indolinones have also been iden-
tified as potent kinase inhibitors. As exemplified by GW-
491619, these have also largely been trisubstituted
olefinic derivatives.3 One notable exception is the recently
identified Aurora B inhibitor, hesperadin (1), in which
the exocyclic olefin is tetrasubstituted.4
(3) (a) Eberwein, D. J.; Harrington, L.; Griffin, R.; Tadepalli, S.;
Knick, V.; Phillips, K.; Dickerson, S.; Davis, S. Proc. Am. Assoc. Cancer
Res. 2002, 43: Abst 1611. (b) Davis, S. T.; Dickerson, S. H.; Harris, P.
A.; Hunter, R. N.; Kuyper, L. F.; Luzzio, M. J.; Veal, J. M.; Walker, D.
H. US Pat. 6,387,919, 2002.
(4) Hauf, S.; Cole, R. W.; LaTerra, S.; Zimmer, C.; Schnapp, G.;
Walter, R.; Heckel, A.; van Meel, J.; Rieder, C. L.; Peters, J. M. J. Cell
Biol. 2003, 161, 281-94.
(5) Sato, A.; Asao, T.; Hagiwara, Y.; Kitade, M.; Yamazaki, Y. US
Pat. 5,965,600, 1999.
(1) (a) Sun, L.; Liang, C.; Shirazian, S.; Zhou, Y.; Miller, T.; Cui, J.;
Fukuda, J. Y.; Chu, J. Y.; Nematalla, A.; Wang, X.; Chen, H.; Sistla,
A.; Luu, T. C.; Tang, F.; Wei, J.; Tang, C. J. Med. Chem. 2003, 46,
1116-9. (b) Sun, L.; Tran, N.; Liang, C.; Hubbard, S.; Tang, F.; Lipson,
K.; Schreck, R.; Zhou, Y.; McMahon, G.; Tang, C. J. Med. Chem. 2000,
43, 2655-63.
(2) Mendel, D. B.; Laird, A. D.; Xin, X.; Louie, S. G.; Christensen,
J. G.; Li, G.; Schreck, R. E.; Abrams, T. J.; Ngai, T. J.; Lee, L. B.;
Murray, L. J.; Carver, J.; Chan, E.; Moss, K. G.; Haznedar, J. O.;
Sukbuntherng, J.; Blake, R. A.; Sun, L.; Tang, C.; Miller, T.; Shirazian,
S.; McMahon, G.; Cherrington, J. M. Clin. Cancer Res. 2003, 9, 327-
37.
(6) While this paper was in preparation, Yanada et al. published a
complementary method for the syntheses of 3-diarylmethylidenylox-
indoles involving tandem indium carbometalation and palladium-
catalyzed coupling reactions. The applicability of their method with
N-containing heterocycles was not presented. Yanada, R.; Obika, M.;
Takemoto, Y. Org. Lett. 2004, 6, 2825-2828.
10.1021/jo050016d CCC: $30.25 © 2005 American Chemical Society
Published on Web 04/02/2005
J. Org. Chem. 2005, 70, 3741-3744
3741

TABLE 1. Tandem Heck-Carbocyclization/
Suzuki-Coupling: Catalyst Evaluation
TABLE 2. Tandem Reaction: Initial Investigation with
Various Arylboronic Acids
a mixture of palladium acetate and the highly active
carbene ligand 1011 also failed to trigger a room-temper-
ature transformation. Microwave irradiation was still
required to force the reaction to completion, affording the
desired indolinone in 72% yield (entry 3). Interestingly,
Buchwald’s ligand, 2-(dicyclohexylphosphino)biphenyl
11,12 which is usually regarded as a more activating
ligand (entry 4), did not promote the reaction, even under
the microwave conditions.
*Yields are isolated following chromatographic purification.
We next examined the reaction under conditions of
increased palladium catalyst and ligand loading. We were
gratified to find that with 0.2 equiv of Pd(OAc)2 and
carbene ligand 10, a complete reaction could be effected
at room temperature within 24 h (entry 5). A similar
increase in load of catalyst 11 (entry 6) did not result in
a complete reaction in 14 h. Finally, under the base-free
Suzuki coupling reaction conditions of copper(I) thiophene-
2-carboxylate (CuTC) with Pd(PPh3)4,13 the cascade reac-
tion proceeded almost to completion, with only a minor
amount (∼10%) of 8 remaining after 25 h at room
temperature (entry 7).
SCHEME 1
and reacted with 4-chloro-2-iodoaniline (7) to afford the
2-iodoanilide 8 in 94% yield (Table 1). A standard Suzuki
coupling reaction condition was first investigated, utiliz-
ing tetrakis(triphenylphosphine)palladium(0) [Pd(PPh3)4]
as catalyst source and potassium phosphate as base in
THF (entry 1).7 While no reaction took place after 14 h
at room temperature, brief microwave irradiation8 pro-
vided the desired diarylidenyl indolinone product 9 in
78% isolated yield (Table 1).9
Proceeding with the reaction conditions identified in
entry 5 (Table 1), we turned our attention to the regi-
oselectivity of this transformation. The 5-chloro substitu-
ent of the indolinone product (E)-12 (Table 2, entry 1)
was incorporated to assist in the regiochemical assign-
ment of the newly formed exo-alkene. Adjacent phenyl
1
proton H4, readily identifiable in the H NMR spectra
As follow up, a small screening set of different pal-
ladium catalysts was evaluated in order to determine if
the reaction could be effected at ambient temperature.
An air-stable mixture of palladium catalysts, CombiPhos-
Pd610 (entry 2), was ineffective at promoting this cascade
reaction at 25 °C. Under the microwave conditions used
previously, the reaction proceeded to completion, though
not as cleanly as was the case with Pd(PPh3)4. Similarly,
(∼6.4-6.8 ppm; doublet), was appropriately positioned
to interact with protons on the aryl ring positioned anti
to the carbonyl, but not with those of the syn-aryl group.
Thus, the presence of NOE cross-peak(s) between phenyl
proton H4 and an aromatic proton(s) of the newly coupled
aryl ring served to unambiguously assign the olefin
regiochemistry of our asymmetric diarylmethylidenyl-
indolinones.
Four aryl boronic acids were chosen for our initial
investigation of regiochemical selectivity (Table 2). In the
case of 3,5-dichlorophenylboronic acid (entry 1), the
reaction proceeded very slowly at 25 °C for 15 h. Heating
(60 °C, 6 h) was required to force the reaction to
completion. Both regioisomers formed under these condi-
tions in a ratio of 4:1; these were separated by silica gel
(7) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457.
(8) Synthesis of 9. To a solution of 8 (0.060 g, 0.157 mmol) in 2
mL of anhydrous THF were added Pd(PPh3)4 (0.018, 0.0157 mmol),
K3PO4 (0.083 g, 0.393 mmol), and phenylboronic acid (0.029 g, 0.236
mmol). The sealed reaction mixture was irradiated in a microwave
synthesizer (Smith Synthesizer, Personal Chemistry) at 100 °C for 30
min. The reaction mixture was concentrated under reduced pressure,
and the resulting crude product was purified by silica gel chromatog-
raphy [hexanes/EtOAc (4:1)] to afford 9 (78%) as yellow oil.
(9) The structure of this product was confirmed by comparison with
the authentic diphenylmethylidene indolinone, prepared from the
condensation of 5-chlorooxindole and benzophenone with sodium
hydroxide in refluxing toluene.
(11) Andrus, M. B.; Song, C. Org. Lett. 2001, 3, 3761.
(12) Wolf, J. P.; Singer, R. A.; Yang, B. H.; Buchwald, S. L. J. Am.
Chem. Soc. 1999, 121, 9550.
(10) Li, G. Y. J. Org. Chem. 2002, 67, 3643.
(13) Savarin, C.; Liebeskind, L. S. Org. Lett. 2001, 1, 2149.
3742 J. Org. Chem., Vol. 70, No. 9, 2005

TABLE 3. Tandem Reaction with Boronic Acids of
Nitrogen Heterocycles
chromatography and each was analyzed by NOE experi-
ments. A cross-peak was found between indolinone proton
H4 and Ha of the dichlorophenyl ring, thus establishing
an (E)-olefinic geometry of the major product (E)-12. No
such cross-peak signal was observed with minor product
(Z)-12, supporting its (Z)-configurational assignment.
Subjecting 3,5-dimethylisoxazole-4-boronic acid (entry 2)
to the same reaction conditions produced an inseparable
mixture of regioisomeric products, (E)-13 and (Z)-13, in
an approximate ratio of 2:1 (1H NMR determination).
Again, NOE experiments clearly identified the dominant
isomer (E)-13 as having the (E)-configuration based on
the presence of cross-peak between H4 and the isoxazole
methyl groups and an absence of any cross-peak between
H4 and the pendant phenyl protons. Consistent with this
assignment, H4 of minor isomer (Z)-13 exhibited no such
NOE cross-peak with these methyl protons. With 3,4,5-
trifluorobenzeneboronic acid as the coupling partner
(entry 3), heating was again required to force the cascade
reaction to completion (70 °C, 4 h). As with the former
reactions, the higher temperature led to the production
of a pair of regioisomers, (E)-14 and (Z)-14 in a ratio of
∼3:1. Similarly, employing 2-methoxyphenylboronic acid
afforded (after heating) (E)-15 and (Z)-15 in a ratio of
∼2:1 (entry 4).
Thus, under these reaction conditions (Pd(OAc)2, car-
bene ligand 10, and K3PO4 in THF), a complete room-
temperature transformation occurred only with the un-
substituted phenylboronic acid. Electron-deficient, electron-
rich, and heterocyclic aromatic boronic acids each reacted
sluggishly at room temperature; elevating the tempera-
ture resulted in complete conversion to indolinone prod-
ucts, but with modest and variable regioselectivities.
Returning to the conditions investigated in Table 1,
we focused on entry 7 and optimized the catalyst load.
Utilizing 0.2 equiv of Pd(PPh3)4 along with CuTC (2
equiv) and 3,5-dichlorophenylboronic acid (2 equiv) at
room temperature, a complete conversion of 8 to indoli-
none was obtained after 4 h. Furthermore, isomer (E)-
12 was generated (93% isolated yield) as the sole product
of this reaction. With the success of this reaction, we next
examined the applicability of the protocol with a variety
of heteroaryl boronic acids.
A variety of nitrogen-containing heteroaryl boronic
acids were selected as coupling components because of
the prevalence of such heterocyles in biologically impor-
tant compounds. As indicated in Table 3, pyridyl, indole,
and pyrimidinyl boronic acids (entries 1-5) were suc-
cessfully incorporated, providing the corresponding diar-
ylidene indolinone in good yields (80-96%). In each of
these cases, only the (E)-isomer was produced. Phthala-
zine (entry 6), quinoline (entry 7), and pyrrole (entry 8)
boronic acids provided the desired indolinones in good
yields (85-91%) and regioselectivities (E/Z ) 13-25:1).
Reactions with quinoline (entry 9), 2-chloropyridine
(entries 10, 11), and isoxazole (entry 12) boronic acids
provided the corresponding diarylidene indolinone in
yields ranging from 62 to 93%, though in these cases,
regioselectivities were lower (E/Z ) 4-7:1). Contrary to
the results obtained with other pyridyl boronic acids,
pyridine-4-boronic acid (entry 13) failed to undergo the
cascade reaction at room temperature. Mild heating (40
°C) was required to completely consume propynamide 8.
*Reaction carried out at 40 °C.
In this case, a moderate yield (40%) of diarylidene
indolinone regioisomers (E/Z ) 23:1) was obtained. With
3,5-difluoropyridine-4-boronic acid (entry 14), the cascade
reaction failed to afford the desired product, even at an
elevated temperature. Presumably, the two strongly
electronegative ortho fluorine substituents deactivate the
boronic acid, thereby precluding it from participating in
the cross-coupling reaction to any significant extent. To
our knowledge, no Suzuki-type couplings with this bo-
ronic acid have yet been reported in the literature.
To probe the effects of the coordinating ligand on
regioselectivity, we examined six structurally diverse
phosphines14 in the reaction of 2-chloropyridine-5-boronic
acid with Pd2(dba)3.15 All trials with these ligands
resulted in incomplete reactions and reduced regioselec-
tivities. Similarly, an expanded set of palladium sources16
failed to provide an improvement for this process.17
Finally, we investigated the role of solvent effects upon
the reaction rate and regioselectivity (Table 4). Polar
solvents are clearly preferable, providing good selectivi-
ties (MeCN) and/or good rates of conversion (methanol
or DMF). Nonpolar or fluorinated solvents (entries 1, 2,
7, and 8) are unsuitable for this process. Combinations
of polar solvents afforded a practical compromise in
achieving high regioselectivities within short reaction
times. Thus, in MeCN/MeOH (4:1, entry 11) and MeCN/
DMF (4:1, entry 12), the reactions were complete in 1 h
(E/Z ) 17:1) and in 2 h (E/Z ) 14:1), respectively.18
Though the process was not complete after 2 h when
THF/DMF (4:1) was utilized (entry 13), the E-isomer was
generated as the sole product of the reaction. However,
by increasing the number of equivalents of boronic acid
and CuTC from 1.4 to 2.0, a complete transformation
J. Org. Chem, Vol. 70, No. 9, 2005 3743

TABLE 4. Tandem Reaction: Solvent Effects
TABLE 6. Tandem Reaction: Anilide Substituent
Effects
% conv
of 8,
% conv
of 8,
isomeric
ratio (E/Z)
product
product
R2 (yield, %)
isomeric
anilide R1 R2 (yield, %) anilide
R1
entry
solvent
ratio (E/Z) entry
solvent
8
4-Cl
H
I
I
9 (95)
9a (94)
8b
8c
5-MeO
4-Cl
I
9b (92)
9 (56)
1
2
3
4
toluene
CH2Cl2
DME
0
10
70, 11:1
70, 11:1
8
9
CF3CH2OH
DMF
10 MeCN
0
8a
Br
100, 12:1
75, 28:1
acids identified earlier. As indicated in Table 5, these
conditions provided a significant improvement in both
yield and selectivity in each case. Single isomers were
generated with 3-quinoline-3-boronic acid and 2-chloro-
pyridine-4-boronic acid (entries 2 and 3, respectively); in
the case of dimethylisoxazolyl boronic acid (entry 4), a
2-fold improvement in regioselectivity was obtained
compared with the reaction in THF alone.
dioxane
11 MeCN/MeOH 100, 17:1
4:1
5
6
7
Et2O
5
12 MeCN/DMF
100, 14:1
4:1
MeOH
100, 11:1
0
13 THF/DMF
82,
E-isomer
4:1
14
CF3C(OH)-
HCF3
With an optimized process in hand, we investigated
whether substituent effects on the anilide ring would play
a significant role in the success of this sequence (Table
6). The 4-chloro substituent, which had been included to
facilitate stereochemical assignments, might have propi-
tiously activated the system by functioning as a stabilizer
of the intermediate Heck cyclization product. However,
omission of this substituent had no detrimental effect on
the reaction outcome as evidenced by the generation of
oxindole 9a in high yield. Furthermore, incorporating a
5-methoxyl group on the anilide, (8b), which might have
deactivated the 2-position toward palladium insertion,
similarly had no adverse effect on this sequence. Finally,
we questioned whether the methodology would extend
to 2-bromoanilides. In this case (8c), the reaction was
less efficient but nevertheless afforded oxindole product
9 in a moderate yield (56%). Thus, while the reaction
optimally proceeds from the 2-iodoanilide, it can also be
successfully carried out from 2-bromoanilides in cases
where the former is not readily available.
TABLE 5. Tandem Reaction: Optimized Selectivity
occurred within 2 h, providing a highly isomerically
enriched product mixture (E/Z ≈ 27:1; Table 5, entry 1).
Encouraged by these results, we applied these opti-
mized conditions to some of the more problematic boronic
In summary, we have developed and optimized a
tandem Heck-Suzuki coupling reaction for the syntheses
of asymmetric diarylmethylidenyl indolinones. At ambi-
ent temperatures, the process is high yielding, highly
regioselective, and can be accelerated by judicious choice
of solvent mixtures. This method could readily be ex-
tended to a diverse range of coupling partners and, as
such, should allow for the rapid generation of novel
compound libraries for biological evaluations.
(14) Phosphine ligands investigated included the monodentate
ligands: (2-dicycohexyphosphino)biphenyl, (2-di-tert-butylphosphino)-
biphenyl, and 2,8,9-triobutyl-2,5,8,9-tetraaza-1-phosphabicyclo[3.3.3]-
undecane, as well as bidentate ligands: 2-di-tert-butylphosphino-2′-
(N,N-dimethylamino)-biphenyl, 9,9-dimethyl-4,5-bis(diphenylphos-
phino)xanthene (XANTPHOS), and 1,1′-bis(diphenylphosphino)-
ferrocene (dppf).
(15) The general protocol involved stirring 8 (1 equiv), Pd2(dba)3
(0.05 equiv), phosphine ligand (0.1-0.2 equiv), copper 2-thiophenecar-
boxylate CuTC (1.4 equiv), and of 2-chloropyridine-5-boronic acid (1.4
equiv) in THF at ambient temperature for 15 h.
(16) Catalysts examined included palladium(0) catalysts: bis(tri-
tert-butylphosphine)palladium(0), bis(tricyclohexylphosphine)palladium-
(0), and bis[1,2-bis(diphenylphosphino)ethane]palladium (0) and pal-
ladium(II) catalysts: trans-dichlorobis(tri-o-tolylphosphine)palladium
(II), trans-dichlorobis(triphenylphosphine)palladium(II), [1,1′-bis-
(diphenylphosphine)palladium(II) complexed with CH2Cl2, dichlorobis-
(benzonitrile)palladium(II), and dichloro(acetonitrile)palladium(II).
(17) The general procedure involved stirring palladium catalyst (0.1
equiv), CuTC (1.4 equiv), 2-choropyridine-5-boronic acid (1.4 equiv),
and 8 (1 equiv) in anhydrous THF at room temperature for 10 h.
(18) Lesser concentrations of MeOH or DMF resulted in incomplete
reactions. In contrast, excess MeOH or DMF was detrimental to the
regioselectivity of this process.
Acknowledgment. We thank Mr. George Elgar and
Mr. Michael Kolpak for NMR assistance and Mr. Wil-
liam Jones for carrying out HRMS analyses.
Supporting Information Available: Experimental pro-
cedures and 1H, COSY (Table 3, entry 3), and NOE spectra
for E-12, Z-12, E/Z-13, and selected examples from Tables 3
(entries 1-5) and 5 (entries 1-3). This material is available
free of charge via the Internet at http://pubs.acs.org.
JO050016D
3744 J. Org. Chem., Vol. 70, No. 9, 2005