Caesium fluoride-promoted Stille coupling reaction: An efficient synthesis of 9Z-retinoic acid and its analogues using a practical building block

Article abstract of DOI:10.1039/b813760a

A highly efficient and rapid total synthesis of 9Z-retinoic acid was accomplished by caesium fluoride-promoted Stille coupling reaction; using a common building block, 9Z-retinoic acid analogues were also prepared by the same method without isomerisation of the Z-double bond. The Royal Society of Chemistry 2008.

Full text of DOI:10.1039/b813760a

View Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Caesium fluoride-promoted Stille coupling reaction: an efficient synthesis
of 9Z-retinoic acid and its analogues using a practical building blockw
Takashi Okitsu, Kinya Iwatsuka and Akimori Wada*
Received (in Cambridge, UK) 7th August 2008, Accepted 23rd September 2008
First published as an Advance Article on the web 23rd October 2008
DOI: 10.1039/b813760a
A highly efficient and rapid total synthesis of 9Z-retinoic acid
was accomplished by caesium fluoride-promoted Stille coupling
reaction; using a common building block, 9Z-retinoic acid
analogues were also prepared by the same method without
isomerisation of the Z-double bond.
Retinoid X receptors (RXRa, b, g) belong to the nuclear
receptor superfamily of eukaryotic transcription factors.
Fig. 1 9Z-Retinoic acid 1 and its synthetic analogues 2.
RXRs modulate gene transcription through formation of
homodimers (RXR–RXR) or heterodimers including
RXR–RARs (retinoic acid receptors), RXR–LXRs (liver X
receptors), RXR–PPARs (peroxisome proliferator-activated
receptors), RXR–FXR (farnesoid X receptor), RXR–TR
(thyroid receptor), and RXR–VDR (vitamin D receptor).1
9Z-Retinoic acid 1 and its synthetic analogues 2 bind to RXRs
and regulate various biological functions such as cell differ-
entiation, proliferation, and control of embryonic develop-
ment.2 Therefore, the development of an efficient synthetic
route for 1 and its analogues 2 is desirable (Fig. 1).
The structure of 1 consists of a highly conjugated polyene
with a partial Z-configuration of the double bond and a
terminal carboxylic acid. Thus, 1 is a challenging synthetic
target for organic chemists in view of its biological activity and
Scheme 1 Reported synthetic route of 1.
unique structure. To date, several synthetic methods for 1 have
been reported and were based on the Horner–Emmons reac-
8 with stannanyl alcohol 9 (Scheme 2).3c However, this
synthetic methodology was based on a linear synthesis, and
is not suitable for the synthesis of a wider variety of analogues
of 1. Therefore, we sought an easier and more powerful
coupling procedure for the convergent synthesis of 1.
tion, Peterson reaction, Stille coupling reaction, and Suzuki
coupling reaction.3,4 To the best of our knowledge, the short-
est synthesis of 1 was accomplished by de Lera et al. (four
steps from 2,2,6-trimethylcyclohexanone, 48.5% overall yield)
and its elegant convergent synthesis may be applied to
9Z-retinoic acid analogues’ syntheses (Scheme 1).4 However,
a key step of this method includes the transformations of vinyl
iodide 3 to its boronic ester 6 and of organostannane 4 to the
corresponding iodide 7, followed by condensation of both
products by a Suzuki coupling reaction. These troublesome
transformations were essential because the attempted Stille
coupling reaction of vinyl iodide 3 with organostannane 4
failed due to its unreactivity and because of isomerisation of
the Z-double bond of 5. The poor outcome was attributed to
the sterically hindered vinyl iodide 3 and electron-withdrawing
group-substituted organostannane 4. We reported the success-
ful synthesis of 1 by the Stille coupling reaction of vinyl triflate
In this communication, we report an efficient, rapid synth-
esis of 9Z-retinoic acid 1 that includes the Stille coupling
reaction of stannanyl ester 11, which bears an electron-
withdrawing group, as a key reaction (Scheme 2). The advan-
tages of our method are: (1) stannanyl ester 11 can be prepared
from commercially available 2-butyn-1-ol in only three steps in
a large scale procedure; (2) the key Stille coupling reaction
precludes the troublesome transformation by Suzuki coupling
Department of Organic Chemistry for Life Science, Kobe
Pharmaceutical University, 4-19-1, Motoyamakita-machi,
Higashinada-ku, Kobe 658-8558, Japan.
E-mail: a-wada@kobepharma-u.ac.jp; Fax: +81 78 441 7562;
Tel: +81 78 441 7560
w Electronic supplementary information (ESI) available: Experimental
details and NMR spectra. See DOI: 10.1039/b813760a
Scheme 2 Our synthetic route of 1.
ꢀc
This journal is The Royal Society of Chemistry 2008
6330 | Chem. Commun., 2008, 6330–6332

View Online
surprise, the use of caesium fluoride (CsF) as an additive not
only enhanced the reaction rate but also prevented the iso-
merisation of the Z-double bond (entries 3–5). Similar en-
hancement of reaction rate by CsF in Stille coupling is well
known.7b All our trials using other fluoride sources such as
TBAF did not indicate an efficient enhancement. Recently,
Baldwin et al. reported that a combination of CuI and CsF
significantly promoted the Stille coupling,8 but these condi-
tions did not favor the coupling reaction of 8 and 11 (entry 6).
Therefore, the optimised reaction conditions were determined
as in entry 5. Notably, the tin residues from the Stille coupling
reaction mixture were removed completely by column chro-
matography using KF-silica as a stationary phase.9
Scheme 3 Preparation of coupling components 11 and 8.
as mentioned above; (3) the isomerisation of the Z-double
bond does not occur in the reaction; and (4) our originally
developed stannanyl ester 11 can serve as a practical building
block in convergent syntheses leading to useful analogues.
Stannanyl ester 11 was prepared in three steps (Scheme 3).
Treatment of 2-butyn-1-ol with Red-Al followed by Bu3SnCl
afforded Z-alcohol 9 in 93% yield,5 which was transformed to
Z-aldehyde 12 by Dess–Martin oxidation in 84% yield.
Horner–Emmons reaction of 12 with phosphonate in the
presence of 1,3-dimethyl-3,4,5,6-tetrahydro-2(1H)-pyrimidinone
(DMPU) yielded 11 in 93% yield in a highly stereoselective
manner, which was confirmed by various NMR spectra.
The hydrolysis of ester 5 has already been reported by our
group (10% KOH aq., EtOH, 50 1C, 98% yield)3d so we could
accomplish the total synthesis of 9Z-retinoic acid 1 in three
steps from aldehyde 13 in 63.0% overall yield or in four steps
from Z-alcohol 9 in 67.4% overall yield. This 9Z-retinoic acid
synthesis has the fewest number of steps and highest yield of
all reported methods.
The powerful and convergent synthesis of 1 prompted us to
prepare various analogues of 1.10 Thus, Stille coupling reac-
tions of stannanyl ester 11 with other vinyl triflates 14a–j were
tried (Table 2). Nonconjugate vinyl triflate 14a from
(À)-menthone provided coupling product 15a in good yield
by our explored method (Method A). However, vinyl triflate
14b from (+)-camphor did not afford coupling product 15b by
Method A. It is noteworthy that Baldwin’s developed proce-
dure (Method B) was also effective for the coupling reaction of
11. Compound 15b was obtained in good yield under these
conditions, despite the lesser reactivity in the case of 8 and 11
(Table 1, entry 5). Therefore, the results in Table 2 were
presented under the reaction conditions specified by either
Method A or B. O- and N-Heterocyclic triflates 14c–d may
also be applied to our conditions to afford 15c–d in moderate
yields. Bicyclic triflates 14e–g having an aromatic ring, were
not successful with Method A, but were with Method B. The
coupling reactions of vinyl triflates 14h–i from sesquiterpenes
proceeded to obtain 15h–i in satisfactory yields. Finally,
acyclic vinyl triflate 14j derived from (À)-nopol was tried,
Next, another side coupling component, vinyl triflate 8, was
prepared in 73% yield and E : Z = 498 : 2 selection by the
reaction of aldehyde 13 with LDA followed by reaction with
N-phenyl-bis(trifluoromethanesulfonimide) (PhNTf2). The use
of KOt-Bu instead of LDA as a base decreased the observed
stereoselectivity (E : Z = 92 : 8). The E : Z ratios of 8 were
1
determined by H NMR spectra.
With stannanyl ester 11 and triflate 8 are in hand, optimisa-
tion of the Stille coupling reaction was carried out (Table 1).
Our previous coupling conditions [Pd2(dba)3ÁCHCl3 (dba =
dibenzylideneacetone), AsPh3, DMF] yielded 5 in only 12%
after 24 h (entry 1).3c This result is consistent with the previous
finding that the Stille coupling reaction using organostannane
with an electron-withdrawing group suffered from its poor
reactivity and homo-coupling.6 Next, we examined additives
for promoting the Stille coupling (entries 2, 3).7 To our
Table 1 Optimisation of the Stille coupling reaction of 11 and 8a
Entry
Pd source (mol%)
Additive (mol%)
8 : 11 (equiv.)
Time/h
5 (%)
1
2
3
4
5
6
Pd2(dba)3ÁCHCl3 (2)
Pd2(dba)3ÁCHCl3 (2)
Pd2(dba)3ÁCHCl3 (2)
Pd2(dba)3ÁCHCl3 (4)
Pd2(dba)3ÁCHCl3 (4)
Pd(PPh3)4 (10)
AsPh3 (8)
1 : 1.3
1 : 1.3
1 : 1.3
1 : 1.1
1.1 : 1
1 : 1.5
24
24
14
12
1.5
15
12
13
64
74
88
15
AsPh3 (8), LiCl (200)
AsPh3 (8), CsF (200)
AsPh3 (16), CsF (200)
AsPh3 (16), CsF (200)
CuI (20), CsF (200)
a
All reactions were carried out in DMF at 40–45 1C.
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 6330–6332 | 6331

View Online
Table 2 Stille coupling with stannyl ester 11a
In conclusion, we have developed the rapid synthesis of
9Z-retinoic acid 1, and an efficient, convergent synthetic route
to its analogues 15a–j by Stille coupling using the common
stannanyl ester 11. Although electron-withdrawing group-
substituted organostannane is generally known to be unreac-
tive for Stille coupling, we demonstrated the enhancement
effect of the reaction rate by CsF. This research would provide
a new synthetic strategy for not only natural products but also
highly conductive molecules in the field of materials sciences.
Hydrolysis of esters 15a–j to carboxylic acids, tests of their
biological activity, and Stille coupling of 11 with other cou-
pling partners, including aromatic halides and vinyl non-
aflates,11 are ongoing.
This work was supported by a Grant-in-Aid for Encourage-
ment of Young Scientists from the Ministry of Education,
Culture, Sports, Science and Technology, and in part by a
grant from the Second Project for Advanced Research and
Technology by Kobe Pharmaceutical University, a grant from
the Hoh-ansha Foundation.
Notes and references
1 V. Giguere, Endocr. Rev., 1999, 20, 689–725, and references cited
therein.
2 (a) The Retinoids: Biology, Chemistry and Medicine, ed.
M. B. Sporn, A. B. Roberts and D. S. Goodman, Raven, New
York, 2nd edn, 1993; (b) R. A. Heyman, D. J. Mangelsdorf,
J. A. Dyck, R. B. Stein, G. Eichele, R. M. Evans and C. Thaller,
Cell, 1992, 68, 397–406; (c) W. K. Hong and M. B. Sporn, Science,
1997, 278, 1073–1077.
3 (a) Y. L. Bennani, J. Org. Chem., 1996, 61, 3542–3544;
(b) A. Wada, S. Hiraishi, N. Takamura, T. Date, K. Aoe and
M. Ito, J. Org. Chem., 1997, 62, 4343–4348; (c) A. Wada,
Y. Nomoto, K. Tano, E. Yamashita and M. Ito, Chem. Pharm.
Bull., 2000, 48, 1391–1394; (d) A. Wada, K. Fukunaga, M. Ito,
Y. Mizuguchi, K. Nakagawa and T. Okano, Bioorg. Med. Chem.,
2004, 12, 3931–3942.
4 (a) Y. Pazos, B. Iglesias and A. R. de Lera, J. Org. Chem., 2001, 66,
8483–8489; (b) Y. Pazos and A. R. de Lera, Tetrahedron Lett.,
1999, 40, 8287–8290.
5 D. L. Aubele, S. Wan and P. E. Floreancig, Angew. Chem., Int.
Ed., 2005, 44, 3485–3488.
6 E. Piers, P. L. Gladstone, J. G. K. Yee and E. J. McEachern,
Tetrahedron, 1998, 54, 10609–10626.
7 LiCl promotes the oxidative addition and CsF facilitates the
transmetallation in the Stille coupling reaction, see:
(a) W. J. Scott and J. K. Stille, J. Am. Chem. Soc., 1986, 108,
3033–3040; (b) A. G. Martınez, J. O. Barcina, C. A. de Fresno and
´
L. R. Subramanian, Synlett, 1994, 1047–1048.
8 (a) S. P. H. Mee, V. Lee and J. K. Baldwin, Chem.–Eur. J., 2005,
11, 3294–3308; (b) S. P. H. Mee, V. Lee and J. K. Baldwin, Angew.
Chem., Int. Ed., 2004, 43, 1132–1136.
9 D. C. Harrowven and I. L. Guy, Chem. Commun., 2004,
1968–1969.
10 Syntheses of 9Z-retinoic acid analogues, see: (a) A. Wada,
Y. Mizuguchi, H. Miyake, M. Niihara, M. Ito, K. Nakagawa
and T. Okano, Lett. Drug Des. Discovery, 2007, 4, 442–445;
(b) A. Wada, Y. Mizuguchi, M. Shinmen, M. Ito, K. Nakagawa
and T. Okano, Lett. Drug Des. Discovery, 2006, 3, 118–122;
(c) B. H. Lipshutz, G. C. Clososki, W. Chrisman, D. W. Chung,
D. B. Ball and J. Howell, Org. Lett., 2005, 7, 4561–4564;
´
(d) B. Domınguez, Y. Pazos and A. R. de Lera, J. Org. Chem.,
2000, 65, 5917–5925; (e) ref. 3c, 3d and 4.
11 For a review of nonaflate, see: (a) K. Ritter, Synthesis, 1993, 735;
For a recent example of Stille coupling of vinyl nonaflate, see:
(b) A. Wada, Y. Ieki, S. Nakamura and M. Ito, Synthesis, 2005,
1581–1588.
a
Method A: 14 (1.1 equiv.), 11 (1.0 equiv.), Pd2(dba)3ÁCHCl3
(4 mol%), AsPh3 (16 mol%), CsF (2.0 equiv.), DMF, 45 1C. Method
B: 14 (1.0 equiv.), 11 (1.3 equiv.), Pd(PPh3)4 (10 mol%), CuI
(20 mol%), CsF (2.0 equiv.), DMF, 45 1C.
but desired product 15j, being unstable, was obtained in only
low yield. The reactivity of the coupling reaction may depend
on the structure of triflates 14a–j, but no clear tendencies were
observed. Nevertheless, Stille coupling of 11 would directly
offer various 9Z-retinoic acid derivatives 15a–j in a minimum
number of steps.
ꢀc
This journal is The Royal Society of Chemistry 2008
6332 | Chem. Commun., 2008, 6330–6332