Synthesis of new furocoumarin analogues via cross-coupling reaction of triflate

Article abstract of DOI:10.1016/S0040-4039(02)02476-0

Furocoumarins such as psoralen or angelicin showed important biological activities. We present here the synthesis of new furocoumarin analogues via Suzuki or Sonogashira cross-coupling reaction of triflate.

Full text of DOI:10.1016/S0040-4039(02)02476-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 97–99
Synthesis of new furocoumarin analogues via cross-coupling
reaction of triflate
Ste´phanie Hesse and Gilbert Kirsch*
Laboratoire d’Inge´nierie Mole´culaire et Biochimie Pharmacologique, Faculte´ des Sciences, Ile du Saulcy, 57045 Metz Cedex,
France
Received 9 October 2002; revised 30 October 2002; accepted 4 November 2002
Dedicated to Professor Guy Queguiner for his contribution to heterocyclic chemistry
Abstract—Furocoumarins such as psoralen or angelicin showed important biological activities. We present here the synthesis of
new furocoumarin analogues via Suzuki or Sonogashira cross-coupling reaction of triflate. © 2002 Elsevier Science Ltd. All rights
reserved.
1. Introduction
We have previously described a rapid access to cou-
marin derivatives using Vilsmeier–Haack–Arnold and
Suzuki cross-coupling reactions.9 In order to complete
this series of molecules, we wanted to replace the
phenyl ring by a furan moiety (Scheme 1).
Linear furocoumarins such as psoralen 1 have shown
important biological activities in the PUVA therapy.1
However, undesirable side effects (risks such as muta-
tions or skin cancers) have been noticed2 so that consid-
erable efforts have been expended to develop new
compounds which may prevent them. Indeed, due to
their bifunctional nature (photoactive a-pyrone and
furan sites), psoralens used to form interstrand
crosslinks with DNA which is believed to be mainly
responsible for their toxicity. Several ways have been
developed to permit only monofunctional photobinding
with DNA:
In a first attempt, we wanted to use the same methodol-
ogy as before, including a Vilsmeier–Haack–Arnold
reaction on the 4,5,6,7-tetrahydrobenzo[b]furan-4-one
6. However, Jakobs et al.7 have shown that this b-
chloroacrolein can be obtained with only 34% yield and
is unstable at room temperature, so that we envisaged
another pathway. In order to keep the cyclization possi-
bility, we synthesized triflates on b-ketoesters.
(a) use of angular furocoumarin such as angelicins 2
(their angular structure prevent crosslinking with
DNA)3
(b) blocking the photoactive a-pyrone double bond
by introduction of substituents4 or by annelation of
an additional aromatic ring such as in
pyridopsoralen5 3
(c) enlarging the space between the photoactive dou-
ble bonds of the a-pyrone and furan like in
furonaphtopyrone6 4.
Moreover, various isosters (5) were prepared and
studied7,8 (Fig. 1).
* Corresponding author. Tel.: + (33) 3 87 31 52 95; e-mail: kirsch@
sciences.univ-metz.fr
Figure 1.
0040-4039/03/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)02476-0

98
S. Hesse, G. Kirsch / Tetrahedron Letters 44 (2003) 97–99
Scheme 1.
2. Results and discussion
Scheme 3. Reagents and conditions: (i) o-anisylboronic acid,
PdCl2, PPh3, iPr2NH, toluene, 80°C, 24 h; (ii) BBr3, CH2Cl2.
4,5,6,7-Tetrahydrobenzo[b]furan-4-one 6 was synthe-
sized by the condensation of 1,3-cyclohexanedione with
chloroacetaldehyde10 (Feist–Benary furan synthesis11).
Ketone 6 was converted into b-ketoester 7 by a classical
method12 using sodium hydride and dimethyl carbon-
ate. In a first attempt, b-ketoester 7 was treated with
sodium hydride at 0°C followed by addition of triflic
anhydride to generate the corresponding triflate. After
treatment of the reaction mixture (which do not change
anymore), a NMR spectrum of the crude product indi-
cated that triflate was present in only 50%. The product
was isolated by a column chromatography on silica gel
in only 27% yield. We then tried other conditions (triflic
anhydride in presence of 2,4,6-collidine) which allowed
isolation of the triflate 8 in 55% yield (Scheme 2). We
then performed Suzuki and Sonogashira coupling reac-
tions with triflate 8.
Sonogashira coupling of triflate 8 with phenylacetylene
was performed in presence of 5 mol% PdCl2 and PPh3,
5 mol% of copper iodide and 1.5 equiv. of triethylamine
in THF overnight at 55°C. Coupling product 12 was
isolated after purification in 79% yield (Scheme 4).
Hydrolysis of methyl ester 12 was done under mild
conditions using LiOH 1.8 M in MeOH/H2O at 5°C.
The corresponding 4-phenylethynyl-6,7-dihydrobenzo-
[b]furan-5-carboxylic acid 13 was obtained with 55%
yield.
It is well known that 2-(1-alkynyl)benzoic acids could
easily be transformed either into phtalides or into iso-
coumarins. Bellina and co-workers14 have shown that
catalytic amount of silver nitrate in dry acetone at
room temperature favored 6-endocyclization rather
It is well known that an inorganic base (such as K3PO4)
is recommended for the Suzuki reaction.13 However in
our case, those conditions failed and after many
attempts, we found that only an organic base gave the
coupling product 9: triethylamine gave only a 25%
yield, whereas diisopropylamine gave a 85% yield
(Scheme 3). Cyclization of 9 (demethylation and con-
comitant lactonization) was made using boron tribro-
mide in methylene chloride at room temperature;
compounds 10 and 11 are obtained with 30 and 14%
yield, respectively. Compound 11 was found to be quite
unstable.
Scheme 4. Reagents and conditions: (i) 1.1 equiv. phenyl-
acetylene, 5 mol% PdCl2, 10 mol% PPh3, 5 mol% Cu(I), Et3N,
THF 18 h, 55°C; (ii) LiOH 1.8 M, MeOH/H2O, 5°C, 20 h;
(iii) 20 mol% AgNO3, acetone, in dark, under argon, 18 h, rt.
Scheme 2. Reagents and conditions: (i) NaH, (MeO)2CO,
DME, reflux; (ii) triflic anhydride, 2,4,6-collidine, CH2Cl2.

S. Hesse, G. Kirsch / Tetrahedron Letters 44 (2003) 97–99
99
than 5-exocyclization. So, cyclization of 13 was per-
formed in the presence of 20 mol% of silver nitrate in
dry acetone during 24 h at room temperature, in the
dark and under an argon atmosphere; this process
allowed the synthesis of 14 with 55% yield (cf. Scheme
3).
7.48 (s, 1H2), 7.84 (d, 2H, J=7.9 Hz); lC (CDCl3) 18.90
(CH2), 21.32 (CH2), 106.22 (CH, C9), 110.87 (CH, C1),
117.87 (C9b), 128.79 (CH), 128.93 (CH), 130.62 (CH),
133.13 (C), 134.01 (C), 141.31 (C), 143.32 (CH, C2),
145.22 (C), 157.51 (C), 167.76 (CO2)
In conclusion, we described here new synthetic access to
furocoumarin derivatives via cross-coupling reaction of
triflate ester.
References
1. Knobler, R. M.; Honigsmann, H.; Edelson, R. L. In
Psoralen DNA Photobiology; Gasparro, F. P., Ed.; CRC
Press: Boca Raton, FL, 1988; Vol. II, p. 117.
2. Saffran, W. A. In Psoralen DNA Photobiology; Gasparro,
F. P., Ed.; CRC Press: Boca Raton, FL, 1988; Vol. II, p.
73.
3. Dall’Acqua, F.; Vedaldi, D.; Caffieri, S.; Guiotto, A.;
Bordin, F.; Rodighiero, P. Natl. Cancer Inst. Monogr.
1984, 66, 55.
3. Data for compounds 10, 11 and 14
1H and 13C NMR spectra were recorded on a AC
Bruker 250 MHz spectrometer in CDCl3.
3.1. 3,4-Dioxacyclopenta[c]phenanthren-6-one 10
4. Carlassare, F.; Baccichetti, F.; Guiotto, A.; Rodighiero,
P.; Gia, O.; Capozzi, A.; Pastorine, G.; Bordin, F. J.
Photochem. Photobiol. B: Biol. 1990, 5, 25.
5. Blais, J.; Averbeck, D.; Moron, J.; Bisagni, E.; Vigny, P.
Photochem. Photobiol. 1987, 45, 463.
Colorless solid; mp: 156°C; lH (CDCl3) 7.44–7.45 (m,
2H), 7.53–7.54 (m, H1+H), 7.72 (d, H4, J=8.6 Hz), 7.92
(d, H2, J=2.3 Hz), 8.36 (d, H, J=8.0 Hz), 8.42 (d, H5,
J=8.7 Hz); lC (CDCl3) 107.22 (CH, C1), 113.15 (CH,
C8), 117.14 (C), 117.77 (CH, C4), 118.86 (C), 121.85
(C), 124.41 (CH), 125.12 (CH), 127.04 (CH), 130.01
(C), 130.22 (CH), 146.97 (CH, C2), 151.41 (C7a), 158.76
(C3a), 161.34 (CO2, C6).
6. Adam, W.; Qian, X.; Saha-Mo¨ller, C. R. J. Org. Chem.
1993, 58, 3769.
7. Jakobs, A. E.; Christiaens, L. J. Org. Chem. 1996, 61,
4842.
8. Fossa, P.; Mosti, L.; Menozzi, G.; Marzano, C.; Bacci-
chetti, F.; Bordin, F. Bioorg. Med. Chem. 2002, 10, 743.
9. Hesse, S.; Kirsch, G. Tetrahedron Lett. 2002, 43, 1213.
10. Matsumoto, M.; Watanabe, N. Heterocycles 1984, 22
(10), 2313.
3.2. 4,5-Dihydro-3,4-dioxacyclopenta[c]phenanthren-6-
one 11
Yellow oil; lH (CDCl3) 2.97–2.10 (m, 2H), 3.05–3.08
(m, 2H), 7.00 (d, 1H1, J=2.1 Hz), 7.33–7.40 (m, 2H),
7.50–7.56 (m, 2H), 8.04 (d, 1H11, J=7.9 Hz).
11. Feist, F. Ber. 1902, 35, 1539.
12. Diamanti, J.; Cayen, C.; Bingham, R.; Krapcho, A. P.
Org. Synth. Coll. Vol. 5, 47, 20.
13. Oh-e, T.; Miyaura, N.; Suzuki, A. J. Org. Chem. 1993,
58, 2201.
3.3. 8-Phenyl-4,5-dihydrofuro[3,2-f ]isochromen-6-one 14
14. Bellina, F.; Ciucci, D.; Vergamini, P.; Rossi, R. Tetra-
hedron 2000, 56, 2533.
Yellow oil; lH (CDCl3) 2.84–2.91 (m, 2H), 2.97–3.09
(m, 2H), 6.21 (s, 1H1), 6.70 (s, 1H9), 7.30–7.44 (m, 3H),