Solid-Phase Synthesis of Isocoumarins: A Traceless Halocyclization Approach

Article abstract of DOI:10.1021/jo900281a

(Chemical Equation Presented) A straightforward and traceless solid-phase methodology was developed for the synthesis of isocoumarins. This two-step process involves a Sonogashira cross-coupling reaction between polymer-bound 2-bromobenzoates and terminal

Full text of DOI:10.1021/jo900281a

Solid-Phase Synthesis of Isocoumarins: A Traceless Halocyclization
Approach
Marine Peuchmaur,‡ Vincent Lisowski,*,‡ Ce´line Gandreuil, Ludovic T. Maillard,
Jean Martinez, and Jean-Franc¸ois Hernandez
Institut des Biomole´cules Max-Mousseron, UMR 5247, CNRS, UniVersite´s Montpellier I et II, UFR des
Sciences Pharmaceutiques et Biologiques, 15 AVenue Charles Flahault,
34093 Montpellier Cedex 5, France
Vincent.lisowski@uniV-montp1.fr
ReceiVed February 9, 2009
A straightforward and traceless solid-phase methodology was developed for the synthesis of isocoumarins.
This two-step process involves a Sonogashira cross-coupling reaction between polymer-bound 2-bro-
mobenzoates and terminal alkynes, followed by an electrophile-induced halocyclization of the resulting
2-(alk-1-ynyl)benzoates through activation of the triple bond with the subsequent release of the 3-substituted
4-haloisocoumarins. This polymer-bound parallel synthetic approach allowed us to achieve large diversity
in good to excellent yields and purities.
Introduction
isocarbostyrils, isochromenes, or isoquinolines.6 Accordingly,
numerous synthetic routes to this scaffold have been reported
in the literature.7 Unsubstituted and 3-substituted isocoumarins
were prepared by either (i) ortho-thallation/Pd-catalyzed olefi-
nation of benzoic acids8 or (ii) Pd-catalyzed coupling of
2-halobenzoic acids/esters with alkenes, vinylic stannanes, or
terminal alkynes and subsequent acid- or metal-catalyzed
cyclization.9 Efficient procedures to produce 3,4-disubstituted
isocoumarins were also developed by direct Rh/Cu-catalyzed
oxidative coupling of benzoic acid10 or by Pd-catalyzed annu-
Isocoumarins1 (1H-2-benzopyran-1-one) are an important
class of naturally occurring lactones that exhibit a wide range
of biological activities such as antimicrobial,2 antifungal,3
antiangiogenic4 properties, and enzyme inhibition.5 In addition,
the isocoumarin ring system is a useful synthetic intermediate
for the synthesis of hetero- and carbocyclic compounds including
* To whom correspondence should be addressed. Phone: +33(0)467548653.
Fax: +33(0)467548654.
‡ Both authors contributed equally to this work.
(1) (a) Barry, R. D. Chem. ReV. 1964, 64, 229–260. (b) Hauser, F. M.;
Baghdanov, V. M. J. Org. Chem. 1988, 53, 4676–4681. (c) Mali, R. S.; Babu,
K. N. J. Org. Chem. 1998, 63, 2488–2492, and references therein.
(2) (a) Hussain, M. T.; Rama, N. H.; Malik, A. Indian J. Chem., Sect. B
2001, 40, 372–376. (b) Matsuda, H.; Shimoda, H.; Yoshikawa, M. Bioorg. Med.
Chem. Lett. 1999, 7, 1445–1450. (c) Yoshikawa, M.; Harada, E.; Naitoh, Y.;
Inoue, K.; Matsuda, H.; Shimoda, H.; Yamahara, J.; Murakami, N. Chem. Pharm.
Bull. 1994, 42, 2225–2230.
(3) Nozawa, K.; Yamada, M.; Tsuda, Y.; Kawai, K.; Nakajima, S. Chem.
Pharm. Bull. 1981, 29, 2689–2691.
(4) Lee, J. H.; Park, Y. J.; Kim, H. S.; Hong, Y. S.; Kim, K.-W.; Lee, J. J.
J. Antibiot. 2001, 54, 463–466.
(5) (a) Harper, J. W.; Powers, J. C. J. Am. Chem. Soc. 1984, 106, 7618–
7619. (b) Powers, J. C.; Asgian, J. L.; Ekici, O. D.; James, K. E. Chem. ReV.
2002, 102, 4639–4750. (c) Petit, A.; Bihel, F.; Alve`s da Costa, C.; Pourquie´,
O.; Checler, F.; Kraus, J.-L. Nat. Cell Biol. 2001, 3, 507–511. (d) Heynekamp,
J. J.; Hunsaker, L. A.; Vander Jagt, T. A.; Deck, L. M.; Vander Jagt, D. L.
BMC Chem. Biol. 2006, 6, 1. (e) Heynekamp, J. J.; Hunsaker, L. A.; Vander
Jagt, T. A.; Royer, R. E.; Deck, L. M.; Vander Jagt, D. L. Bioorg. Med. Chem.
2008, 16, 5285–5294.
(7) Napolitano, E. Org. Prep. Proced. Int. 1997, 29, 631–664.
(8) Larock, R. C.; Varaprath, S.; Lau, H. H.; Fellows, C. A. J. Am. Chem.
Soc. 1984, 106, 5274–5284.
(9) (a) Korte, D. E.; Hegedus, L. S.; Wirth, R. K. J. Org. Chem. 1977, 42,
1329–1336. (b) Batu, G.; Stevenson, R. J. Org. Chem. 1980, 45, 1532–1534.
(c) Sakamoto, T.; Annaka, M.; Kondo, Y.; Yamanaka, H. Chem. Pharm. Bull.
1986, 34, 2754–2759. (d) Sakamoto, T.; Kondo, Y.; Yamanaka, H. Heterocycles
1988, 27, 2225–2249. (e) Izumi, T.; Nishimoto, Y.; Kohei, K.; Kasahara, A.
J. Heterocycl. Chem. 1990, 27, 1419–1424. (f) Sakamoto, T.; Kondo, Y.;
Yasuhara, A.; Yamanaka, H. Tetrahedron 1991, 47, 1877–1886. (g) Liao, H.-
Y.; Cheng, C.-H. J. Org. Chem. 1995, 60, 3711–3716. (h) Sashida, H.;
Kawamukai, A. Synthesis 1999, 1145–1148. (i) Bellina, F.; Ciucci, D.; Vergamini,
P.; Rossi, R. Tetrahedron 2000, 56, 2533–2545. (j) Subramanian, V.; Batchu,
V. R.; Barange, D.; Pal, M. J. Org. Chem. 2005, 70, 4778–4783. (k) Cherry, K.;
Parrain, J.-L.; Thibonnet, J.; Ducheˆne, A.; Abarbri, M. J. Org. Chem. 2005, 70,
6669–6675. (l) Uchiyama, M.; Ozawa, H.; Takuma, K.; Matsumoto, Y.;
Yonehara, M.; Hiroya, K.; Sakamoto, T. Org. Lett. 2006, 8, 5517–5520. (m)
Marchal, E.; Uriac, P.; Legouin, B.; Toupet, L.; Van de Weghe, P. Tetrahedron
2007, 63, 9979–9990. (n) Le Bras, G.; Hamze, A.; Messaoudi, S.; Provot, O.;
Le Calvez, P.-B.; Brion, J.-D. Synthesis 2008, 10, 1607–1611. (o) Hellal, M.;
Bourguignon, J.-J.; Bihel, F. J.-J. Tetrahedron Lett. 2008, 49, 62–65.
(10) (a) Ueura, K.; Satoh, T.; Miura, M. Org. Lett. 2007, 9, 1407–1409. (b)
Ueura, K.; Satoh, T.; Miura, M. J. Org. Chem. 2007, 68, 5362–5367.
(6) (a) Jones, J. B.; Pinder, A. R. J. Chem. Soc. 1958, 2612–2618. (b) Minami,
T.; Nishimoto, A.; Nakamura, Y.; Hanaoka, M. Chem. Pharm. Bull. 1994, 42,
1700–1702.
4158 J. Org. Chem. 2009, 74, 4158–4165
10.1021/jo900281a CCC: $40.75 2009 American Chemical Society
Published on Web 05/13/2009

Solid-Phase Synthesis of Isocoumarins
SCHEME 1. General Solid-Phase Synthetic Pathway to Isocoumarins (A) and Possible Byproducts (B, C)a
a Reagents and conditions: (i) (a) Cs2CO3 (0.5 equiv), EtOH, rt, 3 h, (b) bromo-Wang resin, DMF, rt, 12 h; (ii) alkyne (hex-1-yne or phenylacetylene, 10
equiv), PdCl2(PPh3)2 (0.2 equiv), CuI (0.1 equiv), Et3N/CH2Cl2 (5:1), 65°C, 12 h; (iii) see conditions in Table 1.
lation of internal alkynes with 2-halobenzoate esters.11 In this
vein, the main access to 4-haloisocoumarins consists of a
6-endo-dig electrophile-induced halocyclization first described
from 2-(alk-1-ynyl)benzoic acids as starting material and
N-halosuccinimide (NXS) or X2 (X ) I, Br) as electrophiles.12
A drawback of these reactions is their unsatisfactory selectivity
due to the formation of the five-membered ring byproduct via
a 5-exo-dig cyclization. In the past decade, studies were
particularly focused on the iodolactonization process from
2-(alk-1-ynyl)benzoate esters to afford 4-iodoisocoumarins by
a selective 6-endo-dig cyclization using ICl as electrophilic
reagent.12c,13 While the stability of the halonium intermediates
decreases in the order I > Br > Cl, electrophilic bromo- and
chlorocyclizations were also reported.12a,14 As an example, Li
and co-workers15 have recently described the selective synthesis
of 4-bromo- and 4-chloroisocoumarins using CuX2 (X ) Br,
Cl) as the electrophilic reagent with the corresponding HX salt
of dicyclohexylamine to prevent the formation of the nonha-
logenated byproduct.
As part of our ongoing medicinal projects, the 3-substituted
4-haloisocoumarin scaffold has retained our attention. In this
drug discovery context, the synthesis of combinatorial focused
libraries is crucial to explore structure-activity relationships,
and the main key for the success of such an approach is to
develop a parallel solid-phase synthesis. Since isocoumarins are
versatile intermediates in medicinal and organic chemistry, to
the best of our knowledge, no parallel SPOS has been described
yet for the production of highly diverse 3-substituted 4-haloiso-
coumarin libraries. Our aim was to develop such a synthetic
approach to gain both efficiency and diversity around the
isocoumarin scaffold. In this study, we report a new efficient
parallel solid-phase synthesis of 3-substituted 4-haloisocou-
marins based on an electrophile-promoted traceless halocycliza-
tion of supported 2-(alk-1-ynyl)benzoates to simultaneously form
and release the target compounds.16 This route was optimized
to limit the formation of the five-membered ring and the
nonhalogenated byproduct, allowing its application in the
preparation of a highly structurally diverse library.
Results and Discussion
Among previously reported syntheses of 4-haloisocoumarins
in solution, we decided to investigate and adapt a two-step
strategy involving (i) a Sonogashira cross-coupling reaction
between polymer-bound 2-bromobenzoates and terminal alkynes,
followed by (ii) an electrophile-induced halocyclization of the
resulting 2-(alk-1-ynyl)benzoates through activation of the triple
bond with the subsequent release of the isocoumarins (Scheme
1). In addition to the well-known advantages of the solid support,
including the temporary protection of one functional group (i.e.,
the carboxylic group), the traceless cleavage would ensure that
only the desired cyclized product is released from the polymer
with no cleavage of side products arising from an incomplete
Sonogashira reaction. This guarantees a significant enhancement
of the final purity. To validate this strategy, we first focused on
the synthesis of 3-substituted 4-iodo-, 4-bromo-, 4-chloro, and
4-H-isocoumarins 5a-d and 6a-d as model compounds. The
main challenge was to find conditions for the electrocyclization
of the supported 2-(alk-1-ynyl)benzoates that give the desired
(11) Larock, R. C.; Doty, M. J.; Han, X. J. Org. Chem. 1999, 64, 8770–
8779.
(12) (a) Oliver, M. A.; Gandour, R. D. J. Org. Chem. 1984, 49, 558–559.
(b) Nagarajan, A.; Balasubramanian, T. R. Indian J. Chem., Sect. B 1988, 27,
380. (c) Biagetti, M.; Bellina, F.; Carpita, A.; Stabile, P.; Rossi, R. Tetrahedron
2002, 58, 5023–5038.
(13) (a) Yao, T.; Larock, R. C. Tetrahedron Lett. 2002, 43, 7401–7404. (b)
Rossi, R.; Carpita, A.; Bellina, F.; Stabile, P.; Mannina, L. Tetrahedron 2003,
59, 2067–2081. (c) Yao, T.; Larock, R. C. J. Org. Chem. 2003, 68, 5936–5942.
(14) Chin, L.-Y.; Lee, C.-Y.; Lo, Y.-H.; Wu, M.-J. J. Chin. Chem. Soc. 2008,
55, 643–648.
(16) For selected examples of traceless cyclization, see: (a) Ockey, D. E.;
Lane, D. R.; Seeley, J. A.; Schore, N. E. Tetrahedron 2000, 56, 711–717. (b)
Tietze, L. F.; Steinmetz, A. Synlett 1996, 667–668. (c) Mayer, J. P.; Zhang, J.;
Bjergarde, K.; Lenz, D. M.; Gaudino, J. J. Tetrahedron Lett. 1996, 37, 8081–
8084.
(15) Liang, Y.; Xie, Y.-X.; Li, J.-H. Synthesis 2007, 3, 400–406.
J. Org. Chem. Vol. 74, No. 11, 2009 4159

Peuchmaur et al.
RatiosbA/B/C
TABLE 1. Optimization of electrocyclization conditions
R ) -(CH2)3CH3
R ) Ph
Purityb
Conditions and
Yielda
(%)
Purityb
(%)
Ratiosb
A/B/C
Yielda
(%)
Entry
X
reagents
(%)
1
2
3
4
5
6
7
8
I
I
ICl, CH2Cl2, rt
I2, CH2Cl2, rt
quant.
65
n.d.
62
n.d.
n.d.
60
n.d.
71
99
99
96:4:0
99:1:0
n.d.
100:0:0
n.d.
96
93
23
41
19
45
60
47
73
45
55
97
98
48
67
28
40
59
95
98
>99
98
100:0:0
100:0:0
48:0:52
67:0:33
28:0:72
40:0:60
70:0:30
96:0:4
Cl
Cl
Cl
Cl
Br
Br
H
CuCl2, Cy2NH·HCl, (CH2Cl)2, 80 °C
CuCl2, Cy2NH·HCl, (CH2Cl)2, 65 °C
CuCl2, Cy2NH·HCl, (CH2Cl)2, rt
CuCl2, Et3N·HCl, (CH2Cl)2, 65 °C
CuBr2, Cy2NH·HBr, (CH2Cl)2, 65 °C
CuBr2, Cy2NH·HBr, (CH2Cl)2, rt
TFA/CH2Cl2 (1:1 v/v), rt
n.d.
>99
n.d.
n.d.
93
n.d.
89
80
n.d.
96:0:4
n.d.
9
10
11
100:0c
100:0c
n.d.
100:0c
73:27c
97:3c
H
H
ZnCl2, Cy2NH·HCl, (CH2Cl)2, 65 °C
PTSA, (CH2Cl)2, 65 °C
41
n.d.
n.d.
a Based on ascertained loading of resin 2 and determined by NMR analyses using HMDSO as an internal standard (total amount of A, B, and C was
considered), yields are indicated for the combination of Sonogashira reaction plus cyclization. b As determined by reverse-phase HPLC with monitoring
at 214 nm and confirmed by NMR. Purity is based on added amounts of A, B, and C. c For X ) H, compounds A and C are identical.
6-endo-dig products with good yield and high purity. To this
aim, we first explored the electrocyclization conditions of two
model compounds 3 and 4 (Scheme 1 and Table 1).
the quenching of the excess of iodine reagent with an aqueous
sodium thiosulfate solution (0.4 g ·mL-1, 2 mL), followed by a
subsequent filtration of the mixture on a disposable cartridge
packed with diatomaceous earth. Finally, the resulting organic
solutions were evaporated under reduced pressure (see Experi-
mental Section). Whatever the substituent in position 3 (R )
n-butyl or phenyl), the corresponding 4-iodoisocoumarins 5a
and 6a were obtained in good yields (ICl being somewhat more
efficient than I2), high purities (97-99%), and excellent
selectivity (96-100%) with no nonhalogenated byproduct C
(5d or 6d). When R was an alkyl group, the five-membered
ring isomer (B) was detected at a low level and a slightly higher
A/B ratio was observed when I2 (vs ICl) was used (A/B ratio
was 96/4 and 99/1 with ICl and I2 respectively, Table 1, entries
1 and 2). This was confirmed on a large variety of supported
2-(alk-1-ynyl)benzoates (see Tables 2 and 3, compounds 9a,
10a, 12a, 20a, 22a, 24a, 34a, 36a). When R was an aryl group,
a total selectivity was observed due to its R-cation-stabilizing
property. Therefore, the results observed on the solid support
are similar to those previously described in solution.13c Since
ICl generally afforded the desired product with better yields, it
appeared as the reagent of choice. However, when a significant
amount of the five-membered cycle is formed, I2 might be a
useful alternative.
In a first synthetic step, 2-bromobenzoic acid cesium salt was
loaded onto the commercially available, widely used, and acid-
cleavable bromo-Wang resin (4-(bromomethyl)phenoxyben-
zylpolystyrene, 1% divinylbenzene, 1.6 mmol·g-1) in DMF at
room temperature for 12 h (Scheme 1).17 The loading of the
resin was ascertained by treating a measured amount of the resin-
bound 2-bromobenzoate with 50% TFA in DCM and evaluating
the recovery of the starting material (for resin 2, the experimental
loading was measured to be 1.15 mmol ·g-1). Overall yields18
are given depending on this experimental loading of resin 2. In
a second synthetic step, the Sonogashira cross-coupling19
between the polymer-bound arylbromide 2 and 1-hexyne (R )
n-butyl) or phenylacetylene (R ) phenyl) was performed using
PdCl2(PPh3)2 (0.2 equiv) and CuI (0.1 equiv) in Et3N/CH2Cl2
(5:1 v/v) at 65 °C to readily afford the alkynes 3 and 4. To
explore the scope of the lactonization, different electrophiles
(ICl, I2, CuCl2, CuBr2, ZnCl2, PTSA, TFA) in DCM or DCE
have been studied. At this stage, the relative abundance of the
six- (A) and five-membered ring (B) compounds as well as the
nonhalogenated (C) isocoumarin was evaluated by HPLC and
NMR analyses. Results are summarized in Table 1.
Iodolactonization was performed using ICl (1.2 equiv) or I2
(3 equiv) in CH2Cl2 at rt for 5 h (Table 1, entries 1 and 2). At
the end of the reaction, 4-iodoisocoumarins 5a and 6a as a
mixture with the excess of the electrophilic reagent were
recovered by filtering the resin. Since the methodology has to
be useful for parallel synthesis, treatment of the reaction mixture
has to be straightforward for a combinatorial approach. There-
fore, we developed an original purification process that directly
allowed elimination of salts and excess reagents. It consists of
While electrocyclization leading to the 4-chloro- or 4-bro-
moisocoumarins is less documented, a recent study15 has
reported a useful system CuX2/Cy2NH·HX (X ) Cl, Br) for
the selective synthesis of 4-haloisocoumarins. These conditions
were applied to polymer-bound models 3 and 4. At the end of
the reaction, 4-chloro- and 4-bromoisocoumarins along with the
excess of the electrophilic reagent were recovered by draining
the resin. In this case, an efficient treatment consisting of
filtration of the reaction mixture through a pad of silica gel was
applied to remove the excess of reagents and salts followed by
concentration of the solvent. Results were greatly dependent
on the R alkyne substituent. When R ) n-butyl, the use of CuX2
(4 equiv) and Cy2NH·HX (0.1 equiv) in DCE at 65 °C gave
4-chloro- and 4-bromoisocoumarins 5b,c with satisfactory yield
(>60%) and purity (>93%) as well as 6-endo-dig cyclization
selectivity (>96%) (Table 1, entries 4 and 7). However, in the
same conditions, for R ) phenyl, the 4-haloisocoumarins 6b,c
were obtained in a mixture along with around 30% of the 4-H-
isocoumarin 6d (Table 1, entries 4 and 7). Concerning the
bromolactonization, a temperature decrease from 65 °C to rt
led to the desired 4-bromoisocoumarin in 47% yield with less
(17) (a) Wang, S. S. J. Am. Chem. Soc. 1973, 95, 1328–1333. (b) Lu, G.;
Mojsov, S.; Tam, J. P.; Merrifield, R. B. J. Org. Chem. 1981, 46, 3433–3436.
(18) Yields were determined by NMR titration according to Yan, B. Analysis
and Purification Methods in Combinatorial Chemistry; John Wiley & Sons: New
York, 2004. The crude isocoumarin was dissolved in 0.5 mL of CDCl3 containing
hexamethyldisiloxane (2.5µmol/mL) as internal standard. Yield was determined
by quantitative 1H NMR titration of the isocoumarin relative to the standard.
The accuracy of the method was estimated to 5% as confirmed by calibration
with weighted quantities (18, 29, 53 mg) of 4-methoxybenzylalcohol and is
consistent with the literature data.
(19) For reviews, see: (a) Campbell, I. B. In The Sonogashira Cu-Pd-
Catalyzed Alkyne Coupling Reaction. Organocopper Reagents; Tayler, R. T. K.
Ed.; IRL Press: Oxford, UK, 1994; pp 217-235. (b) Sonogashira, K. In
ComprehensiVe Organic Synthesis; Trost, B. M. Ed.; Pergamon: Oxford, UK,
1991; Vol. 3, pp 521-549.
4160 J. Org. Chem. Vol. 74, No. 11, 2009

Solid-Phase Synthesis of Isocoumarins
TABLE 2. Substrate Scope (Variable R Group) for Sonogashira and Electrocyclization Steps
a Based on ascertained loading of resin 2 and determined by NMR analyses using HMDSO as an internal standard (total amount of A, B, and C was
considered), yields are indicated for the combination of Sonogashira reaction plus cyclization. b As determined by reverse-phase HPLC with monitoring
at 214 nm and confirmed by NMR. Purity is based on added amounts of A, B, and C.
than 5% of 6d (A/C ratio 96:4) and an excellent purity (95%)
(Table 1, entry 8). On the other hand, whatever the conditions
(temperature or catalyst) used for chlorolactonization of the
supported 2-(2-phenylethynylbenzoate) 4 (Table 1, entries 3-6),
the selectivity of the reaction was not improved.
purities (up to 99%). Considering the results described above,
the reagent elected for iodolactonization was ICl (when 5-exo-
dig cyclization is detected, I2 is preferred). For bromo- and
chlorolactonization, we recommend the use of the CuX2/
Cy2NH·HX system. In the case of 4-H-isocoumarin, TFA
appeared to be the best reagent.
We also explored the formation of 4-H-isocoumarins by acid-
promoted cyclization of supported 2-(alk-1-ynyl)benzoates 3 and
4. Three main protocols, which have been recently reported in
the literature for the synthesis of such compounds in solution,
were evaluated. First, the Lewis acid, ZnCl2, which has been
reported as a heteroannulation inducing reagent in a two-step
one-pot synthesis of isocoumarin,9g was tested on polymer-
bound derivatives 3 and 4. Only moderate yields (about 40%
for R ) n-Bu and Ph) and incomplete selectivity (for R ) Ph)
were achieved (Table 1, entry 10). In a second protocol
described by Le Bras and co-workers,9n p-toluenesulfonic acid
(Table 1, entry 11) was used to produce the isocoumarin 6d
through a PTSA-catalyzed annulation of the supported diaryl-
alkyne 4 with an excellent 6-endo-dig cyclization selectivity.
Third, we tested the use of TFA9o (Table 1, entry 9). Satisfying
to very good yields and purity and complete selectivity were
obtained, making this protocol the most efficient for the
synthesis of both 3-alkyl- and 3-phenylisocoumarins 5d and 6d.
In parallel to these studies, other solid supports were
evaluated. In particular, the Merrifield resin also gave good
results with slightly lower efficiency in terms of selectivity
(results not shown).
With these optimized conditions in hand (Table 1, entries 1,
2, 4, 7, and 9), the scope of this route was then investigated
toward (i) various substituted terminal alkynes and (ii) a
selection of substituted (het)arylbromides as substrates. All the
compounds of this library were synthesized on a parallel
synthesizer apparatus (Tables 2 and 3).
With respect to the alkyne partner (Table 2, X ) I,
compounds 9a-18a), most of the functional groups studied so
far are compatible with iodolactonization conditions, and good
to excellent yields, purities (above 80%), and A/B ratios were
generally observed. Alkyl groups (9a-13a) and substituted
(het)aryl (14a, 15a, and 18a) were readily accommodated. A
particular case was the anilines 16a and 17a, which were
obtained in low yields. This was due to the formation of extra-
iodinated byproduct, as identified by LC-MS, highlighting the
significant reactivity of the aniline aromatic ring. The presence
of a nitrile, an alcohol, or even a carboxylic acid (11a-13a)
group did not cause any difficulty. Finally, as established during
the study on models 3 and 4 (Table 1, entries 1 and 2), low
selectivity obtained with ICl could be improved using I2 but
this resulted in lower yields.
Thus, we developed an efficient traceless route to prepare
4-X isocoumarins (X ) H, I, Cl, Br) in three steps, without
purification, with moderate to good yields (41-99%) and
In a second set of experiments, keeping R ) n-butyl or
phenyl, the scope of the process was investigated with respect
to the benzene ring, which was substituted with electron-rich
J. Org. Chem. Vol. 74, No. 11, 2009 4161

Peuchmaur et al.
TABLE 3. Substrate Scope (Variation on Aromatic Ring) for Sonogashira and Electrocyclization Steps
4162 J. Org. Chem. Vol. 74, No. 11, 2009

Solid-Phase Synthesis of Isocoumarins
TABLE 3. Continued
a Based on ascertained loading of resin 2 and determined by NMR analyses using HMDSO as an internal standard (total amount of A, B, and C was
considered), yields are indicated for the combination of Sonogashira reaction plus cyclization. b As determined by reverse-phase HPLC with monitoring
at 214 nm and confirmed by NMR. Purity is based on added amounts of A, B, and C. c For X ) H, compounds A and C are identical. d Experiments
on resin loaded at 0.42 instead of 0.98 mmol ·g-1 e Et3N·HCl was used as the catalyst (instead of Cy2NH·HCl).
.
or -poor substituent or replaced by various aromatic moieties
(Table 3). All of the expected 4-X isocoumarins (X ) I, Cl,
Br, H) were generated using our experimental conditions.
Generally, good to excellent selectivity was obtained with a
percentage of the desired isocoumarin often close to 100.
Isocoumarins bearing an electron-donating or electron-with-
drawing substituent at the position 6 or 7 (compounds 19, 20,
27-34) and isosteric pyridine- and thiophene-containing ana-
logues (compounds 39-42) were synthesized in good yields
(Table 3, entries 1-6, 21-39, and 46-52 except for entries
25, 26, and 35). However, when substituents were located at
position 5 or 8 of the scaffold (Table 3, entries 7-20 and
43-45), lower yields were obtained. These results very likely
originated from the bulkiness sensitivity of the Sonogashira
reaction. In order to promote completion of this reaction, the
palladium-catalyzed cross-coupling of 1-hexyne with the resin-
bound 2-bromobenzoate 2 and 2-bromo-3-methylbenzoate ana-
logue was performed under microwave irradiation (150 °C, 2 h).
While the reaction time was significantly shortened in the case
of the unhindered 2-bromobenzoate (2 h vs 16 h under classical
heating), the conversion rate of 2-bromo-3-methylbenzoate was
not improved, even when adding KI to enhance reactivity via
Br/I exchange. To circumvent this issue, we decided to diminish
the resin loading. Going from 0.98 to 0.42 mmol ·g-1 gave a
complete Sonogashira reaction between hex-1-yne or pheny-
lacetylene with the resin-bound 2-bromo-3-methylbenzoate,
leading to 4-H- and 4-haloisocoumarins with good purities and
high yields (Table 3, entries 8, 13, and 14). This result indicated
that the proximity between adjacent sites imposed by the cross-
linked polymer had its own influence on the Sonogashira
reaction: too high loadings, meaning high proximity, might
hinder complete accessibility of the reactive sites. Moreover, a
substituent in position 3 onto the polymer-bound 2-(alk-1-
ynyl)benzoates seemed to disrupt the halonium’s addition on
the triple bond because low selectivity remained in particular
when R ) n-Bu (Table 3, entry 10). Other unexpected different
behaviors have been observed during the electrocyclization
reaction in solution and on solid phase. For instance, the
synthesis of 4-chloro-7-nitroisocoumarins in solution was
already reported (with R ) C8H17).15 The authors described that,
when using CuCl2/Cy2NH·HCl as electrocyclization reagents,
the 4-chloro-7-nitro-3-octylisocoumarin was obtained in 87%
yield as the sole product, while when using CuBr2/Cy2NH·HBr,
the 4-bromo analogue was isolated in 68% yield in combination
J. Org. Chem. Vol. 74, No. 11, 2009 4163

Peuchmaur et al.
SCHEME 2. Proposed Mechanism of Polymer-Bound
Iodolactonization with ICl for Compound 6a
compared to that obtained after the first cycle. In the third cycle,
a more significant difference appeared since the isocoumarin
6a was obtained with 79% yield and 92% purity (Table 4). The
other electrocyclization conditions were also evaluated. The I2-
cleaved resin cannot be loaded again. By using CuCl2, reloading
of the 2-bromobenzoic cesium salt was much less efficient: after
a second cycle, the 4-chloroisocoumarin 6b was obtained in
low yield and low purity. In the case of CuBr2, 2-bromobenzoic
cesium salt was loaded at rt for a second cycle, indicating that
bromo-Wang was effectively regenerated. However, yield and
purity were not as satisfactory as for ICl, and a large drop in
efficiency was obtained during the third cycle (Table 4).
Although the results obtained with copper halides were less
satisfying, halo-Wang resins seemed to be at least partially
generated. This is not in agreement with the mechanism
proposed in solution for such copper reagents that should result
only in the formation of the alcohol-Wang resin.15 Thus, we
established that only resins used for the preparation of iodinated
isocoumarins with ICl could be recycled at least once.
Conclusions
with 29% of the 4-H analogue. However, during the electrophile-
induced annulation on the solid support, we observed opposite
results, with CuBr2/Cy2NH·HBr giving the 6-endo-dig com-
pound with very high selectivity (with R ) C4H9, entries
23-25). This was not the case for the chlorolactonization (64:
36, Table 3, entry 23). Nevertheless, in the latter, higher
selectivity could be achieved by changing the catalyst from
Cy2NH·HCl to Et3N·HCl (Table 3, entry 24). Although no
explanation could be drawn for this behavior, these results
showed again unexpected interference of the solid support,
which could be relieved by limited condition changes.
To conclude this study, we exploited the previously
described13c mechanism of the electrophile-induced cyclization
with ICl to recycle the resin (Scheme 2). According to Scheme
2, nucleophilic attack by the oxygen of the carbonyl group on
the triple bond activated by iodonium coordination would be
followed by the nucleophilic substitution of the chloride anion
on the polymer’s vicinal methylene group. Therefore, a chloro-
Wang resin should be generated and could enter a new synthetic
cycle (Table 4). To test this hypothesis, the resin 4 was reacted
with ICl (1.2 equiv) in CH2Cl2 at rt to afford the 4-iodoisocou-
marin 6a. The recovered resin was washed and dried under
reduced pressure before being loaded again with the 2-bro-
mobenzoic acid cesium salt. Because the chlorinated solid
support was less reactive than the bromo-Wang resin, the loading
needed to be performed at higher temperature (i.e., 65 °C). This
recycled resin was resubmitted to the Sonogashira reaction with
phenylacetylene and electrocyclization, with ICl leading to a
second batch of compound 6a with high purity (97%, as
obtained for the first batch). The yield was slightly lower (93%)
In conclusion, we have developed a straightforward and
traceless electrophile-promoted halocyclization solid-phase meth-
odology that provides access to the isocoumarin scaffold. This
first polymer-bound parallel synthetic approach allowed large
variations at positions 3 and 4 of the isocoumarin scaffold and
on the aromatic moiety. Moreover, further derivatization of the
synthesized 4-haloisocoumarins (X ) I, Cl, Br) should enlarge
the library. Yields for the key reactions (Sonogashira cross-
coupling and electrocyclization) were moderate to excellent
(30% to quantitative). Because we also optimized the purification
process, allowing direct elimination of salts and excess reagents
to obtain desired compounds with high purities (82% of
compounds with purity above 80%), this methodology represents
a valuable improvement for the parallel synthesis of isocou-
marins and analogues. Moreover, we demonstrated that the use
of ICl as electrophile allowed recycling of the resin. In a
fundamental point of view based on results obtained during resin
recycling experiments, we assume that the proposed mechanism
of the electrophile-induced annulation with CuX2/Cy2NH·HX
has to be re-examined. Further studies to understand this
mechanism will be described in due course.
Experimental Section
General SPS Procedure for the Synthesis of Isocoumarins.
To a solution of 2-bromobenzoic acid derivative (4 equiv) in EtOH
(10 mL/mmol) was added portionwise Cs2CO3 (2 equiv). After
stirring for 2 h at rt, the reaction mixture was concentrated and
dried under vacuum. The cesium salt was dissolved in DMF then
added to the preswollen (DMF) bromo-Wang resin (1.6 mmol).
The suspension was stirred for 16 h at rt. After filtration, the
resulting resin 2 was washed with DMF (3 × 3 mL), H2O (2 × 2
mL), DMF (1 × 3 mL), CH2Cl2 (2 × 3 mL), MeOH (2 × 3 mL),
and CH2Cl2 (2 × 3 mL) and dried under vacuum. Experimental
loading of resin was determined by cleavage of 100 mg of the resin
with TFA/CH2Cl2 (1:1, 1 mL) for 1 h at rt. The solution was
collected by filtration, the resin was washed with CH2Cl2 (1 mL),
and the filtrates were mixed and evaporated under reduced pressure.
The loading was determined from the weight of recovered product
(data concerning the loading are available in the Supporting
Information). Resin (200 mg) was loaded into a polypropylene tube
equipped with a polyethylene frit and swelled in CH2Cl2 for 1 h.
The resin was filtered and suspended in a mixture of anhydrous
TABLE 4. Resin Recycling Results When Using ICl and CuBr2 as
Electrophile Reagents
Cycle Xa Electrophile Yieldb (%) Purity (%) Ratios A/B/C
1st
I
I
I
Br
Br
Br
ICl
ICl
ICl
CuBr2
CuBr2
CuBr2
96
93
79
47
31
11
97
97
92
95
71
39
100:0:0
100:0:0
100:0:0
96:0:4
98:0:2
98:0:2
2nd
3rd
1st
2nd
3rd
a Compounds 6a (X ) I) and 6c (X ) Br) were chosen as model
b
experiments. Based on ascertained loading of resin 2 (1.15 mmol ·g-1).
4164 J. Org. Chem. Vol. 74, No. 11, 2009

Solid-Phase Synthesis of Isocoumarins
CH2Cl2 (1 mL) and Et3N (5 mL). CuI (0.1 equiv), Pd(PPh3)2Cl2
(0.2 equiv), and alkyne (10 equiv) were added. The reaction mixture
was heated for 16 h at 65 °C. The resin was drained and washed
successively with CH2Cl2 (2 × 5 mL), H2O/DMF (1:1, 2 × 5 mL),
5% DIEA/DMF (2 × 5 mL), MeOH (1 × 5 mL), DMF (1 × 5
mL), MeOH (2 × 5 mL), and CH2Cl2 (2 × 5 mL). Eventually, a
second cycle of reaction was performed depending on the conver-
sion rate determined by HPLC analysis after cleavage of a sample
of resin by TFA/CH2Cl2 (1:1).
A: Typical Procedure for Iodolactonization: Synthesis of
3-Substituted 4-Iodo-1H-isochromen-1-ones. Polymer-bound 2-(alk-
1-ynyl)benzoate (150 mg) was loaded into a polypropylene tube
equipped with a polyethylene frit and swelled in CH2Cl2 for 1 h.
The resin was filtered and suspended in anhydrous CH2Cl2 (3 mL).
A solution of ICl (1.0 M, 1.2 equiv) in CH2Cl2 or I2 (3 equiv) was
added. The suspension was stirred at rt for 4 h, then the resin was
drained and washed with CH2Cl2 (3 × 3 mL). The filtrates were
mixed and treated with 1 mL of aqueous Na2S2O3 (4.0 g in 10 mL
of water). To remove the aqueous phase, the biphasic mixture was
filtered over a cartridge (5 mL) packed with diatomaceous earth.
The recovered organic phase was evaporated under reduced
pressure, yielding the desired 3-substituted 4-iodo-1H-isochromen-
1-ones.
CDCl3) δ 14.0 (CH3), 22.5 (CH2), 29.2 (CH2), 33.4 (CH2), 101.4
(C), 120.5 (C), 125.9 (CH), 128.7 (CH), 130.0 (CH), 135.6 (CH),
136.6 (C), 156.0 (C), 161.7 (C); IR νmax (neat) 1735, 1624, 1475,
1285, 1102, 1051, 1027, 999, 759, 687 cm-1; HPLC tR 3.24 min;
LC-MS (ESI+) m/z (%) 324 (12), 322 [M + CH3CN + H]+ (12),
283 (88), 281 [M + H]+ (100); HRMS (ESI+) m/z calcd for
C13H14O2Br 281.0177, found 281.0175.
1
4-Chloro-3-phenyl-1H-isochromen-1-one, 6b: H NMR (300
MHz, CDCl3) δ 7.43-7.51 (m, 3H), 7.56-7.62 (m, 1H), 7.81-7.89
(m, 3H), 7.94 (d, J ) 8.0 Hz, 1H), 8.33 (dd, J ) 7.9, 1.2 Hz, 1H);
13C NMR (75 MHz, CDCl3) δ 111.5 (C), 120.7 (C), 124.2 (CH),
128.4 (2 × CH), 129.3 (CH), 129.5 (2 × CH), 130.0 (CH), 130.4
(CH), 131.6 (C), 135.5 (CH), 136.1 (C), 150.6 (C), 161.0 (C); IR
ν
max (neat) 1723, 1558, 1489, 1055, 981, 710, 694 cm-1; HPLC tR
3.02 min; LC-MS (ESI+) m/z (%) 300 (3), 298 [M + CH3CN +
H]+ (10), 259 (32), 257 [M + H]+ (100); HRMS (ESI+) m/z calcd
for C15H10O2Cl 257.0369, found 257.0392.
4-Bromo-3-phenyl-1H-isochromen-1-one, 6c12a: 1H NMR (300
MHz, CDCl3) δ 7.40-7.45 (m, 3H), 7.52 (ddd, J ) 7.9, 7.2, 1.3
Hz, 1H), 7.62-7.66 (m, 2H), 7.76 (ddd, J ) 7.9, 7.2, 1.4 Hz, 1H),
7.81-7.86 (m, 1H), 8.25 (dd, J ) 7.9, 0.9 Hz, 1H); 13C NMR (75
MHz, CDCl3) δ 76.8 (C), 120.6 (C), 128.4 (2 × CH), 129.5 (CH),
130.0 (CH), 130.3 (2 × CH), 130.5 (CH), 131.8 (CH), 135.0 (C),
136.0 (CH), 138.5 (C), 155.1 (C), 161.9 (C); IR νmax (neat) 1737,
1619, 1474, 1231, 1086, 1054, 1020, 960, 758, 688 cm-1 HPLC tR
3.01 min; LC-MS (ESI+) m/z (%) 344 (10), 342 [M + CH3CN +
H]+ (10), 303 (99), 301 [M + H]+ (100); HRMS (ESI+) m/z calcd
for C15H10O2Br 300.9864, found 300.9850.
3-Butyl-4-iodo-1H-isochromen-1-one, 5a: 1H NMR (300 MHz,
CDCl3) δ 0.93 (t, J ) 7.3 Hz, 3H), 1.34-1.48 (m, 2H), 1.62-1.76
(m, 2H), 2.88 (t, J ) 7.9 Hz, 2H), 7.45 (ddd, J ) 8.1, 6.7, 1.8 Hz,
1H), 7.66-7.76 (m, 2H), 8.41 (dd, J ) 8.1, 0.7 Hz, 1H); 13C NMR
(75 MHz, CDCl3) δ 14.1 (CH3), 22.0 (CH2), 29.6 (CH2), 37.4 (CH2),
76.4 (C), 120.3 (C), 128.8 (CH), 129.9 (CH), 130.7 (CH), 134.4
(C), 135.9 (CH), 158.5 (C), 162.2 (C); IR νmax (neat) 1731, 1609,
1473, 1177, 1100, 1053, 1029, 982, 759, 687 cm-1; HPLC tR 3.31
min; LC-MS (ESI+) m/z (%) 329 [M + H]+ (100); HRMS (ESI+)
m/z calcd for C13H14O2I 329.0039, found 329.0019.
C: Typical Procedure for Electrocyclization with TFA:
Synthesis of 3-Substituted 4-H-Isochromen-1-ones. The polymer-
bound 2-(alk-1-ynyl)benzoate (150 mg) was swelled in CH2Cl2 for
1 h, filtered, and suspended in a mixture of CH2Cl2/TFA (v/v 1:1,
4 mL). The suspension was stirred at rt for 2 h, then the resin was
drained and washed with CH2Cl2 (3 × 3 mL). The filtrates were
mixed and evaporated under reduced pressure, yielding the desired
3-substituted 4-H-isochromen-1-one.
4-Iodo-3-phenyl-1H-isochromen-1-one, 6a12c: H NMR (300
1
MHz, CDCl3) δ 7.40-7.45 (m, 3H), 7.52 (ddd, J ) 7.9, 7.2, 1.3
Hz, 1H), 7.62-7.66 (m, 2H), 7.76 (ddd, J ) 7.9, 7.2, 1.4 Hz, 1H),
7.81-7.86 (m, 1H), 8.25 (dd, J ) 7.9, 0.9 Hz, 1H); 13C NMR (75
MHz, CDCl3) δ 76.8 (C), 120.6 (C), 128.4 (2 × CH), 129.5 (CH),
130.0 (CH), 130.3 (2 × CH), 130.5 (CH), 131.8 (CH), 135.0 (C),
136.0 (CH), 138.5 (C), 155.1 (C), 161.9 (C); IR νmax (neat) 1725,
1599, 1471, 1231, 1075, 1029, 946, 758, 687 cm-1; HPLC tR 3.06
min; LC-MS (ESI+) m/z (%) 349 [M + H]+ (100); HRMS (ESI+)
m/z calcd for C15H10O2I 348.9726, found 348.9724.
3-Butyl-1H-isochromen-1-one, 5d9g: 1H NMR (300 MHz,
CDCl3) δ 0.90 (t, J ) 7.4 Hz, 3H), 1.31-1.46 (m, 2H), 1.58-1.75
(m, 2H), 2.48 (t, J ) 7.4 Hz, 2H), 6.20 (s, 1H), 7.30 (d, J ) 7.8
Hz, 1H), 7.39 (dd, J ) 8.0, 7.4 Hz, 1H), 7.62 (dd, J ) 7.8, 7.4 Hz,
1H), 8.20 (d, J ) 8.0 Hz, 1H); 13C NMR (75 MHz, CDCl3) δ 14.0
(CH3), 22.3 (CH2), 29.2 (CH2), 33.4 (CH2), 103.1 (CH), 120.4 (C),
125.2 (CH), 127.7 (CH), 129.7 (CH), 134.9 (CH), 137.9 (C), 158.6
(C), 163.3 (C); IR νmax (neat) 1723, 1655, 1483, 1160, 1045, 1022,
755, 690 cm-1; HPLC tR 2.78 min; LC-MS (ESI+) m/z (%) 244
[M + CH3CN + H]+ (23), 203 [M + H]+ (100); HRMS (ESI+)
m/z calcd for C13H15O2 203.1072, found 203.1052.
B: Typical Procedure for Bromo- or Chlorolactonization:
Synthesis of 3-Substituted 4-Bromo- or 4-Chloro-1H-isochromen-
1-ones. The polymer-bound 2-(alk-1-ynyl)benzoate (150 mg) was
swelled in anhydrous ClCH2CH2Cl for 1 h, filtered, and suspended
in anhydrous ClCH2CH2Cl (3 mL). CuX2 (X ) Cl or Br) (4 equiv)
and Cy2NH·HX or Et3N·HX (0.1 equiv) were added. The suspen-
sion was stirred at 65 °C for 16 h, then the resin was drained and
washed with CH2Cl2 (3 × 3 mL). The filtrates were mixed and
passed through a cake of silica gel. The recovered organic phase
was evaporated under reduced pressure, yielding the desired
3-substituted 4-bromo- or 4-chloro-1H-isochromen-1-one.
3-Butyl-4-chloro-1H-isochromen-1-one, 5b: 1H NMR (300
MHz, CDCl3) δ 0.90 (t, J ) 7.4 Hz, 3H), 1.32-1.46 (m, 2H),
1.62-1.75 (m, 2H), 2.74 (t, J ) 7.4 Hz, 2H), 7.45-7.54 (m, 1H),
7.72-7.80 (m, 2H), 8.25 (dd, J ) 7.9, 0.8 Hz, 1H); 13C NMR (75
MHz, CDCl3) δ 14.0 (CH3), 22.5 (CH2), 29.1 (CH2), 31.3 (CH2),
111.2 (C), 120.6 (C), 123.5 (CH), 128.7 (CH), 130.1 (CH), 135.5
(CH), 135.9 (C), 155.0 (C), 162.2 (C); IR νmax (neat) 1740, 1631,
1478, 1271, 1051, 1027, 761, 688 cm-1; HPLC tR 3.22 min; LC-
MS (ESI+) m/z (%) 280 (3), 278 [M + CH3CN + H]+ (11), 239
(32), 237 [M + H]+ (100); HRMS (ESI+) m/z calcd for C13H14O2Cl
237.0682, found 237.0691.
3-Phenyl-1H-isochromen-1-one, 6d9k 1H NMR (300 MHz,
:
CDCl3) δ 6.86 (s, 1H), 7.27-7.52 (m, 5H), 7.64 (m, 1H), 7.80 (m,
2H), 8.16-8.30 (m, 1H); 13C NMR (75 MHz, CDCl3) δ 102.1 (CH),
120.7 (C), 125.4 (2 × CH), 126.3 (CH), 128.4 (CH), 129.1 (2 ×
CH), 129.8 (CH), 130.2 (CH), 132.1 (C), 135.2 (CH), 137.8 (C),
153.8 (C), 162.6 (C); IR νmax (neat) 1717, 1638, 1482, 1238, 1070,
1028, 749, 681 cm-1; HPLC tR 2.74 min; LC-MS (ESI+) m/z (%)
264 [M + CH3CN + H]+ (7), 223 [M + H]+ (100); HRMS (ESI+)
m/z calcd for C15H11O2 223.0759, found 223.0775.
Acknowledgment. This study and postdoctoral fellowship
of M.P. were supported by grants from the Agence Nationale
de la Recherche (Contract ANR-06-PCVI-0028-02). The authors
thank Pierre Sanchez for performing mass spectrometry analyses.
Supporting Information Available: Experimental details and
1H and 13C NMR spectral data for described compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
4-Bromo-3-butyl-1H-isochromen-1-one, 5c: 1H NMR (300
MHz, CDCl3) δ 0.92 (t, J ) 7.3 Hz, 3H), 1.32-1.45 (m, 2H),
1.63-1.74 (m, 2H), 2.88 (t, J ) 7.6 Hz, 2H), 7.44-7.51 (m, 1H),
7.71-7.79 (m, 2H), 8.22 (d, J ) 7.8 Hz, 1H); 13C NMR (75 MHz,
JO900281A
J. Org. Chem. Vol. 74, No. 11, 2009 4165