Access to isocarbacyclin derivatives via substrate-controlled enolate formation: Total synthesis of 15-deoxy-16-(m-tolyl)-17,18,19,20- tetranorisocarbacyclin

Article abstract of DOI:10.1021/ol0515762

(Chemical Equation Presented) We describe a convergent and flexible synthesis of 15-deoxy-16-(m-tolyl)-17,18,19,20-tetranorisocarbacyclin (15-deoxy-TIC), a simple isocarbacyclin derivative. The synthesis takes advantage of two key step reactions: a regioselective deprotonation of the described ketone under substrate control which is then trapped, as the enol triflate, to generate the C6-C9α endocyclic double bond, followed by an sp²-sp³ Pd-catalyzed cross-coupling reaction (C5-C6) with a suitable primary alkyl Grignard reagent. Introduction of the C13-C14 (E)-double bond in the ω-side chain is performed by the Julia-Kocienski olefination.

Full text of DOI:10.1021/ol0515762

ORGANIC
LETTERS
2005
Vol. 7, No. 23
5115-5118
Access to Isocarbacyclin Derivatives via
Substrate-Controlled Enolate Formation:
Total Synthesis of 15-Deoxy-16-(m-tolyl)-
17,18,19,20-tetranorisocarbacyclin
Neil A. Sheddan and Johann Mulzer*
Institut fu¨r Organische Chemie, Wa¨hringerstrasse 38, A-1090 Wien, Austria
johann.mulzer@uniVie.ac.at
Received July 6, 2005
ABSTRACT
We describe a convergent and flexible synthesis of 15-deoxy-16-(m-tolyl)-17,18,19,20-tetranorisocarbacyclin (15-deoxy-TIC), a simple isocarbacyclin
derivative. The synthesis takes advantage of two key step reactions: a regioselective deprotonation of the described ketone under substrate
control which is then trapped, as the enol triflate, to generate the C6
cross-coupling reaction (C5 C6) with a suitable primary alkyl Grignard reagent. Introduction of the C13
chain is performed by the Julia Kocien˜ski olefination.
−
C9
r
endocyclic double bond, followed by an sp2
−
sp3 Pd-catalyzed
-side
−
−C14 (E)-double bond in the ω
−
In 1996, Takechi and co-workers reported the discovery of
a novel subtype of the prostacyclin (PGI2) receptor,1 one
having different properties compared with those of the known
prostacyclin receptor. Isocarbacyclin (4), a stable analogue
of prostacyclin,2 which is a potent agonist for the known
prostacyclin receptor, possesses high affinity for the novel
subtype. Further investigation led to the design of the central
nervous system (CNS)-specific PGI2 ligandss15-deoxy-16-
(m-tolyl)-17,18,19,20-tetranorisocarbacyclin (15-deoxy-TIC)3
(1) and 15R-TIC4 (2)sthat exhibit neuronal survival-promot-
ing activity. Ever since the discovery of these ligands, the
biofunction of various TIC derivatives in the brain has been
intensively investigated (Figure 1). These ligands have been
successfully used to visualize the specific location of the IP2
receptor for both in vitro and in vivo5 systems by auto-
radiography of rat brain slices and positron emission tomog-
(1) Takechi, H.; Matsumura, K.; Watanabe, Y.; Kato, K.; Noyori, R.;
Suzuki, M.; Watanabe, Y. J. Biol. Chem. 1996, 271, 5901.
(2) Moncada, S.; Gryglewski, R.; Bunting, S.; Vane, J. R. Nature 1976,
263, 663.
(3) Suzuki, M.; Kato, K.; Watanabe, Y.; Satoh, T.; Matsumura, K.;
Watanabe, Y.; Noyori, R. Chem. Commun. 1999, 307.
(4) Suzuki, M.; Kato, K.; Noyori, R.; Watanabe, Y.; Takechi, H.;
Matsumura, K.; Långstro¨m, B.; Watanabe, Y. Angew. Chem., Int. Ed. Engl.
1996, 35, 334.
Figure 1. Isocarbacyclin and selected derivatives.
10.1021/ol0515762 CCC: $30.25
© 2005 American Chemical Society
Published on Web 10/19/2005

raphy (PET)5,6sa powerful noninvasive method for molec-
ular imagingsof a living rhesus monkey,7 using tritium- and
11C-labeled TIC derivatives, respectively. 15-Deoxy-TIC (1)
has shown binding to the CNS-type IP2 receptor (IC50 ) 3
nM) 10 times stronger than its 15R-TIC counterpart. Fur-
thermore, the ratio of the potency of the binding affinity
between 15-deoxy-TIC (1) and 15R-TIC (2) is actually
correlated well with the biological activity. 15-Deoxy-TIC
(1) has also shown an inhibitory effect on apoptosis of
neuronal cells induced by high oxygen (50%) atmosphere
at 10-fold lower concentration than 15R-TIC (2) (IC50 ) 30
and 300 nM), respectively.8
carbacyclin (5) and 3-thiaisocarbacyclin (6)sby appropriate
choice of R-side-chain Grignard reagent. The second key
step in our synthetic stategy is the generation of the five-
ring endocyclic double bond. Previous work has been carried
out on the desymmetrization of such bicyclic systems;13 all,
to the best of our knowledge, have employed the use of a
chiral base. As we were first looking for a cost-efficient
synthesis, available on gram scale, and one that avoided the
use of an expensive chiral base, we envisaged that vinyl
triflate 8a could be synthesized via the regioselective
deprotonation of the corresponding ketone under substrate
control. To these ends we chose the sterically demanding
trityl protecting group on the premise that due to the roof-
shaped structure of ketone 9, it would hinder the approach
of the base at C7, leaving C9R open for deprotonation.
Ketone 9 could then in turn be easily obtained in gram
quantities from the known bicyclic methyl ester 10,14 with a
mimimum of synthetic transformations.
Scheme 1
Scheme 2
Our retrosynthesis of 15-deoxy-TIC (1) (Scheme 1)
features the late-stage construction of the C13-C14 double
bond in the ω-side chain by the Julia-Kocien˜ski olefination.
The attachment of the R-side chain (C1-C5 skeleton) and
construction of the C6-C9R9 double bond has always proven
to be a stumbling block and, as such, has received great
attention over the years.10-12 We decided that a disconnection
at C5-C6 could exploit an sp2-sp3 palladium-catalyzed
cross-coupling approach between vinyl triflate 8a, and
primary alkyl Grignard reagent 12, to install the required
R-side chain. This approach would also allow flexibility for
the synthesis of other isocarbacyclin derivativess3-oxaiso-
Starting from methyl ester 10, after treatment with TBSCl
and reduction with DIBAL-H, primary alcohol 11 was
obtained (Scheme 2). Cleavage of the ketal protecting group
and subsequent tritylation of the primary alcohol, ketone 9
was obtained in 89% yield over four steps.
Scheme 3
(5) Suzuki, M.; Noyori, R.; Långstro¨m, B.; Watanabe, Y. Bull. Chem.
Soc. Jpn. 2000, 73, 1053.
(6) Suzuki, M.; Doi, H.; Hosoya, T.; Långstro¨m, B.; Watanabe, Y. Trends
Anal. Chem. 2004, 23, 595 and references therein.
(7) Watanabe, Y.; Suzuki, M.; Bjo¨rkman, M.; Matsumura, K.; Watanabe,
Y.; Kato, K.; Doi, H.; Onoe, H.; Sihver, S.; Andersson, Y.; Kobayashi, K.;
Inoue, O.; Hazato, A.; Lu, L.; Bergstro¨m, M.; Noyori, R.; Långstro¨m, B.
Abstr. Pap. Neuroimage, Aarhus, May 16-18, 1997; Vol. 5, A1.
(8) Satoh, T.; Ishikawa, Y.; Kataoka, Y.; Cui, Y.; Yanase, H.; Kato, K.;
Watanabe, Y.; Nakadate, K.; Matsumura, K.; Hatanaka, H.; Kataoka, K.;
Noyori, R.; Suzuki, M.; Watanabe, Y. Eur. J. Neurosci. 1999, 11, 3115.
(9) Prostacyclin numbering.
(10) Suzuki, M.; Koyano, H.; Noyori, R. J. Org. Chem. 1987, 52, 5583.
(11) Ogawa, Y.; Shibasaki, M. Tetrahedron Lett. 1984, 25, 1067.
(12) Ishikawa, T.; Ishii, H.; Shimizu, K.; Nakao, H.; Urano, J.; Kudo,
T.; Saito, S. J. Org. Chem. 2004, 69, 8133.
Ketone 9 in THF was treated with KHMDS (2 equiv) at
-108 °C and kept at this temperature for a further 2 h
followed by the addition of PhNTf2 in THF, giving an
inseparable mixture of compounds 8a and 8b (16:1 respec-
tivly from H NMR analysis) (Scheme 3). After recrystal-
lization, needles of triflate 8a were obtained which were then
1
5116
Org. Lett., Vol. 7, No. 23, 2005

subjected to X-ray diffraction, which combined with 2D
NMR studies, unequivocally confirmed the regiochemistry
to be that of the desired compound (Figure 2).
Scheme 4
Figure 2. ORTEP-315 projection (ellipsoids: 50% probability) of
triflate 8a. Selected bond lengths: C1-C8, 1.32 Å; C7-C8, 1.49
Å.
The coupling between enol triflate 8a and alkyl Grignard
reagent 12,16 under Pd-catalysis, in the presence of LiCl, was
carried out following a modified Kuwajima protocol.17 Upon
treatment of enol triflate 8a18 with Pd(PPh3)4 (2 mol %) and
subsequent addition of Grignard reagent 12 (1.1 equiv), a
2:1 mixture of 13 and â-hydride elimination product 13′ was
obtained. In contrast, on treatment with PdCl2(dppf) (2 mol
%) as catalyst, cross-coupled product 13 was obtained as
the only detectable product in 98% yield (Scheme 4).
Chemoselective deprotection of the trityl group proved to
be tricky under standard conditions,19 resulting either in
global deprotection or a 2:1 mixture of the desired primary
alcohol and the diol resulting from additional TBS depro-
tection, respectively. After intensive investigation, we found
that clean deprotection of trityl ether 13 was effected with
Et2AlCl at -50 °C. Swern oxidation of the resulting primary
alcohol gave aldehyde 7 (98% yield over two steps).
Mitsunobu reaction with 1-phenyl-5-mercaptotetrazole (PT-
SH) followed by oxidation with m-CPBA furnished phenyl
tetrazole 16 in 57% yield over four steps (Scheme 5).
Scheme 5
Synthesis of ω-side chain sulfone 16 was accomplished
in four steps from commercially available m-tolyl aldehyde
14. Horner-Wadsworth-Emmons olefination with triethyl
phosphonoacetate yielded the corresponding R,â-unsaturated
ethyl ester, which after complete reduction gave alcohol 15.
Despite the fact that this type of aldehyde has, in a previous
synthesis of this compound,3 been reported to be unstable
under stongly basic conditions, we assumed that this referred
to the Barbier conditions variation of the Julia-Kocien˜ski
olefination. We therefore postulated that by using the
premetalated species we should not encounter any difficulties.
Treatment of sulfone 16 with KHMDS at -60 °C for 45
min, followed by the dropwise addition of aldehyde 7,
delivered E-olefin 17 (Scheme 6). NMR analysis showed
no visible traces of the Z-isomer or epimerization at the C12
position. Deprotection of PMB ether 17 with DDQ yielded
primary alcohol 18, which was subsequently oxidized via
Swern and Pinnick oxidations to carboxylic acid 19. Finally,
deprotection of the TBS ether was effected by TBAF to give
15-deoxy-TIC (1), albeit taking 4 days for the reaction to
complete. Spectral comparison of our synthetic material with
data provided by Professor Masaaki Suzuki matched per-
fectly.20
(13) Vaulont, I.; Gais, H.-J.; Reuter, N.; Schmitz, E.; Ossenkamp, R. K.
L. Eur. J. Org. Chem. 1998, 805, 5.
(14) (a) Weiss, U.; Edwards, J. M. Tetrahedron Lett. 1968, 9, 4885. (b)
Petzoldt, K.; Dahl, H.; Skuballa, W.; Gottwald, M. Liebigs. Ann. Chem.
1990, 1087.
(15) Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565.
(16) Prepared from commerically available 1,5-pentandiol in three steps.
(17) (a) Enev, V.; Ka¨hlig, H.; Mulzer, J. J. Am. Chem. Soc. 2001, 123,
10764 and references cited within. (b) Morihira, K.; Hara, R.; Kawahara,
S.; Nishimori, T.; Nakamura, N.; Kusama, H.; Kuwajima, I. J. Am. Chem.
Soc. 1998, 120, 12980. (c) Hayashi, T.; Konishi, M.; Kobori, Y.; Kumada,
M.; Higuchi, T.; Hirotsu, K. J. Am. Chem. Soc. 1984, 106, 158 and
references cited within.
(18) A mixture of triflates 8a/b (16:1) was used. After three steps and
subsequent column chromatography to yield aldehyde 7, the minor
1
regioisomer could no longer be detected by H NMR.
(19) Greene, T. W.; Wuts, P. G. M. ProtectiVe Groups in Organic
Synthesis; John Wiley & Sons: New York, 1999.
Org. Lett., Vol. 7, No. 23, 2005
5117

Scheme 6
In conclusion, the synthesis of 15-deoxy-TIC (1) was
Acknowledgment. We thank Schering AG, Berlin, for
financial support for N.A.S. and also for the generous gift
of methyl ester 10. We also thank Hanspeter Ka¨hlig and
Lothar Brecker for NMR and Gerald Giester and Vladimir
Arion for crystallography.
performed in 20 stepss13 steps along the longest linear
sequence with a yield of 53%. A novel feature in our
synthetic strategy is the generation of the troublesome C6-
C9R double bond with great ease, via regiospecific ketone
deprotonation under substrate control. Moreoverly, we have
shown that a Pd-catalyzed cross-coupling can provide
practical and easy access to R-side chain incorporation which
could also be applied in the synthesis of other isocarbacylin
analogues such as 3-oxaisocarbacyclin (5) and 3-thiaiso-
carbacyclin (6).
Supporting Information Available: Experimental pro-
cedures and selected NMR data for compounds 7, 8a, 9, 12,
13, 15-19, and 1 and X-ray crystallographic data for 8a
(CIF). This material is available free of charge via the Internet
at http://pubs.acs.org.
(20) We thank Professor Masaaki Suzuki for the private communication
of the 400 MHz 1H and 13C NMR data for 15-deoxy-TIC (1), allowing for
a better comparison than with the originally published 270 MHz data; see
ref 3.
OL0515762
5118
Org. Lett., Vol. 7, No. 23, 2005