Total synthesis of (+)-Oocydin A: Application of the suzuki-miyaura cross-coupling of 1,1-Dichloro-1-alkenes with 9-alkyl 9-BBN

Article abstract of DOI:10.1002/anie.200800585

(Chemical Equation Presented) Two sensitive fragments were coupled in the mild title reaction to create the Zchlorovinyl functionality of the target macrolide oocydin A (1; see scheme). Another highlight in the total synthesis of 1 was an efficient stereoselective Pd⁰-catalyzed cyclization to form the highly substituted tetrahydrofuran ring. Bz = benzoyl, TBS = tert-butyldimethylsilyl, MPM = 4-methoxyphenylmethyl.

Full text of DOI:10.1002/anie.200800585

Communications
DOI: 10.1002/anie.200800585
Natural Product Synthesis
Total Synthesis of (+)-Oocydin A: Application of the Suzuki–Miyaura
Cross-Coupling of 1,1-Dichloro-1-alkenes with 9-Alkyl 9-BBN**
Emmanuel Roulland*
In 1999, the chlorinated macrolide (+)-oocydin A (1;
Scheme 1) was extracted from the bacterium Serratia mar-
cescens, which grows as an epiphyte on a Venezuelan aquatic
considers that symbiotic bacteria may biosynthesize this
metabolite in sponges.
From a structural point of view, these molecules display a
number of salient motifs in their complex framework, such as
a tetrahydrofuran ring fused with a macrocyclic lactone and a
Z chlorovinyl functionality. The latter functionality aroused
our interest. We believed that it might be possible to
synthesize (+)-oocydin A (1) through a palladium-mediated
cross-coupling of a 9-alkyl 9-BBN reagent with a 1,1-dichloro-
ꢀ
1-alkene to create the C8 C9 bond by the selective substi-
tution of the most accessible trans chloride atom (Scheme 2).
Scheme 1. The structures of (+)-oocydin A (1) and some related
natural products.
plant.[1] In the same year, (ꢀ)-haterumalide NA[2a] was
isolated from the sponge Ircinia sp. during the screening of
extracts of marine organisms collected near Okinawa island
and was presented as a diastereomer of 1. The same substance
was isolated from the soil bacterium Serratia plymuthica in
2001.[2b] In 2005, Sato et al. isolated (+)-FR177391[3] from
Serratia liquefaciens and determined its structure unambigu-
ously by X-ray crystallographic analysis of the corresponding
propylamide. The 1H and 13C NMR spectroscopic data of
(+)-oocydin A (1), (ꢀ)-haterumalide NA, and FR177391
appear to be strictly identical, which indicates that they are
one and the same compound despite their different optical
rotations. (+)-Oocydin A (1) is a representative member of a
unique class of structurally complex natural products with
similar skeletons (Scheme 1, 2–5). The substances that belong
to the oocydin A family were extracted from various sources,
such as bacteria, sponges, and ascidians[4] (Didemnidae sp. and
Lissoclinum sp.). It is intriguing that a structure as complex as
1 has been isolated from such a variety of sources, unless one
Scheme 2. Retrosynthetic analysis of (+)-oocydin A (1). 9-BBN=
9-borabicyclo[3.3.1]nonane, Bz=benzoyl, MPM=4-methoxyphenyl-
methyl, TBS=tert-butyldimethylsilyl.
As reactions between two coupling partners of this type had
not been described in the literature, we investigated this
approach and established conditions for their efficient cross-
coupling, confirming for this reaction that the use of
bisphosphines with a large bite angle as the palladium ligands
is instrumental.[5] We describe herein the first application of
this methodology to total synthesis. The target, (+)-oocy-
din A (1), not only provides a synthetic challenge, but has
intrinsic value as a compound with cytotoxic and phytopa-
thogenic properties.[1–3]
.
[*] Dr. E. Roulland
Several research groups strove to remove the ambiguities
that remained in terms of the structure of 1 and finally
demonstrated that (+)-oocydin A (1) and (ꢀ)-haterumali-
de NA were one and the same molecule. Only Hoye and
Wang completed the total synthesis of 1; however, they made
no mention of the optical rotation of their product.[6a] Both
Kigoshi et al.[6b] and Gu and Snider[6c] described the synthesis
of the methyl ester of ent-oocydin.
Institut de Chimie des Substances Naturelles, CNRS
Avenue de la Terrasse, 91198 Gif-sur-Yvette (France)
Fax: (+33)1-6907-7247
E-mail: emmanuel.roulland@icsn.cnrs-gif.fr
Homepage: http://www.icsn.cnrs-gif.fr
[**] Financial support from the Institut de Chimie des Substances
Naturelles, CNRS is acknowledged. 9-BBN=9-borabicyclo-
[3.3.1]nonane.
Our retrosynthetic analysis (Scheme 2) based on the
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
0
ꢀ
formation of the C8 C9 bond by a Pd -catalyzed cross-
3762
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 3762 –3765

Angewandte
Chemie
Table 1: Optimization of the synthesis of 13a from 12 (cf. Scheme 2).[a]
coupling reaction required the synthesis of the tetrahydro-
furan derivative 6a and the 1,1-dichloro-1-alkene subunit 7.
The tetrahydrofuran substructural motif of 6a prompted us to
explore a Pd0-catalyzed cyclization through the postulated
p-allyl palladium intermediate A. In this way, we expected to
be able to install diastereoselectively the required vinyl
functionality at C11. This vinyl group would then be hydro-
borated to give the desired 9-alkyl 9-BBN nucleophilic cross-
coupling partner. The synthesis of the 1,1-dichloro-1-alkene
subunit 7 was based on a challenging ring-closing metathesis
(RCM) reaction to provide the a,b-unsaturated lactone 15
Entry
Phosphine[b]
t [h]
T [8C]
Yield [%]
13a/13b
1
2
3
4
5
6
7
8
PPh3
PPh3
PPh3
48
20
40
66
66
50
40
40
50
55
45
40
54[c]
98
100
n.r.[c]
100
93
90:10
88:12
87:13
3.75
0.67
18
2.5
0.5
20
24
20
0.67
4
P(o-Tol)3
P(o-Tol)Ph2
P(2-furyl)3
AsPh3
P(p-CF3C6H4)3
S,S ligand[e]
R,R ligand[e]
P(p-OMePh)3
92:8
83:17
90:10
90[d]
n.r.[d]
60[c]
100
99
9
91:6
40:60
96:4
10
ꢀ
11[f]
(Scheme 4). Thus, we would control the geometry of the C4
C5 double bond. We envisioned macrolactone ring closure
under Yamaguchi conditions[7] as described by Gu and
Snider,[6c] and planned to end our synthesis as reported by
Kigoshi et al.[6b] with the Nozaki–Hiyama–Kishi (NHK)
coupling[8] of fragment 8.
[a] Reactions were carried out in THF. [b] Palladium source: [Pd2(dba)3]
(2.5 mol%); added monophosphine: 10 mol% or bisphosphine:
5 mol%. [c] Unchanged compound 12 was recovered. [d] Partial degra-
dation was observed. [e] Trost ligand: 1,2-diaminocyclohexane-N,N’-
bis(2’-diphenylphosphanylbenzoyl). [f] Results of a trial reaction on an
11-mmol scale with [Pd2(dba)3] (1 mol%).
Our straightforward synthesis of fragment 6a started from
the d-tartrate ester 9 (Scheme 3), which constitutes the sole
method have been reported.[12] It must be added that the
occurrence of acetate 12 as an E/Z mixture has no conse-
quence for the diastereomeric outcome of the reaction. In
acetonitrile, the rate of the reaction remained slow at 208C,
and palladium black was formed. In THF, higher temper-
atures led to much quicker reactions with 13a,b formed in
quantitative yield but with lower diastereoselectivity (Table 1,
entries 3 and 5). The presence of chloride anions is known to
accelerate the rate of the inversion of configuration of the
p-allyl palladium intermediate.[13] However, in our case, there
was no improvement in the 13a/13b ratio upon the addition of
Bu4NCl; we only observed faster palladium-black formation.
The encumbered phosphine P(o-Tol)3 or electron-deficient
P(p-CF3C6H4)3 did not promote any reaction even in THF at
reflux (Table 1, entries 4 and 8). Moreover, neither trifuryl-
phosphane nor AsPh3 led to any improvement in the
diastereomeric ratio of the product (Table 1, entries 6 and
7). Assays with the Trost chiral bidentate phosphines were
more interesting, as we observed mismatching effects: The S,S
ligand gave good diastereoselectivity for the desired product
13a, albeit with a very low reaction rate and incomplete
conversion (Table 1, entry 9), whereas the use of the R,R
ligand led to a fast and complete reaction with the opposite
diastereoselectivity (Table 1, entry 10). Finally, we discovered
that P(p-OMeC6H4)3 with [Pd2(dba)3] in THF at 408C gave
the product with the best diastereomeric ratio and in
quantitative yield (Table 1, entry 11). The two diastereomers
13a and 13b were not readily separable even by preparative
HPLC; however, the corresponding benzoyl esters 6a and 6b
were separated by simple flash chromatography. Thus, the
targeted tetrahydrofuran 6a was obtained through a method
that compares favorably with those used by other research
groups for the total synthesis of 1.
Scheme 3. Synthesis of subunit 6a: a) LiAlH4, Et2O, 99%; b) MPMCl,
NaOH, PhH, reflux, 73%; c) I2, PPh3, imidazole, toluene, reflux, 90%;
d) vinylmagnesium chloride, CuI, THF, HMPA, ꢀ508C, 85%; e) HCl,
H2O, EtOH, ethylene glycol, 08C, 97%; f) second-generation Grubbs
catalyst, allyl acetate, CH2Cl2, reflux, 77%; g) [Pd2(dba)3],
P(p-OMeC6H4)3, THF, 408C, 99%, 13a/13b 96:4; h) PhCOCl, DMAP,
pyridine. dba=dibenzylidene acetone, DMAP=4-dimethylaminopyri-
dine, HMPA=hexamethylphosphoramide.
source of chirality in our strategy. Thus, 9 was reduced to a
diol and monoprotected as a 4-methoxybenzyl ether. After its
transformation into the iodide derivative 10, we performed a
copper-catalyzed substitution with vinylmagnesium chloride
and obtained, after deprotection, the allylic compound 11. A
cross-metathesis reaction[9] catalyzed by the second-genera-
tion Grubbs catalyst[10] between 11 and allyl acetate furnished
compound 12 in good yield as an E/Z mixture (ca. 95:5).
With compound 12 in hand, we investigated the formation
of the tetrahydrofuran ring (Table 1). In a first trial, we
treated the acetate 12 with a catalytic amount of [Pd2(dba)3]
and PPh3 in THF at 208C. Good diastereoselectivity was
observed in favor of the expected product (13a/13b 90:10),[11]
albeit with a poor combined yield of 54% (Table 1, entry 1).
This diastereoselectivity was unexpected, as only a few cases
of the synthesis of substituted tetrahydrofurans by this
We then focused our efforts on the synthesis of the C1–C8
fragment 7 (Scheme 4). As Hoye and Wang[6c] had demon-
strated previously that the configuration at C3 can be
established readily by the reduction of the ketone 21
(Scheme 5), we chose to perform a non-asymmetric synthesis
of 7. We started from alcohol 14,[14] which was synthesized in
Angew. Chem. Int. Ed. 2008, 47, 3762 –3765ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3763

Communications
Scheme 4. Synthesis of subunit 7: a) Methacryloyl chloride, iPr2NEt,
CH2Cl2, ꢀ788C!RT, 99%; b) second-generation Grubbs catalyst,
CH2Cl2, reflux under Ar, c=0.01m, 84%; c) NaBH4, CeCl3, MeOH;
d) TBSCl, imidazole, DMF; e) Ac2O, pyridine, 94% (3 steps); f) PPTS,
MeOH, reflux, 96%; g) Swern oxidation; h) EtOAc, LDA, THF, 93%
(2 steps); i) TBSCl, imidazole, DMF, 100%; j) SmI2, THF, 97%.
DMF=N,N-dimethylformamide, LDA=lithium diisopropylamide,
PPTS=pyridinium 4-toluenesulfonate.
high yield by the treatment of a preformed allylindium
reagent with chloral hydrate in DMF. We prepared the
methacrylic ester of alcohol 14 in good yield by coupling it
with methacryloyl chloride at low temperature. RCM then
gave the lactone 15. The RCM reaction has to be conducted in
a dilute medium (0.01m) in freshly distilled CH2Cl2 (distilled
over CaH2 under argon) to avoid dimerization of the starting
material. Toluene has been reported to be a better solvent for
the RCM of substrates that bear an electron-poor and/or
hindered double bond;[15] however, no improvement was
noticed in our case. Inspired by a recent article,[16] we also
attempted a direct synthesis of lactone 15 by condensing
chloral with tigloyl chloride through a hetero-Diels–Alder
process, but lactone 15 was obtained in poor yield. We
reduced the lactone 15 to a diol to provide alcohol 16 after a
selective protection–deprotection sequence. Alcohol 16 was
then oxidized to an aldehyde under Swern conditions,[17] and
the product was submitted directly to an aldol condensation
with EtOAc to give compound 17 in good yield after TBS
protection. Finally, the 1,1-dichloro-1-alkene functionality
was generated by the treatment of 17 with SmI2 (2 equiv) in
THF[18] to afford the targeted subunit 7.
With compounds 6a and 7 in hand, the setting for studying
the key step of this synthesis was established (Scheme 5).
Unfortunately, the use of our previously described optimal
conditions[5] ([Pd2(dba)3], xantphos, KF, and K3PO4 in THF at
reflux) led to compound 18 in only 34% yield along with
degradation of the starting material. We therefore reinvesti-
gated our methodology and eventually found that the use of
dpephos (another large-bite-angle bisphosphine) in place of
xantphos in the absence of KF led to an effective cross-
coupling of 6a and 7 to give 18 in a much improved yield of
87%. At that stage our strategy had been fully validated, and
we were able to complete the synthesis of (+)-oocydin A (1).
The diester 18 was saponified to give the seco acid 19,
which underwent macrolactonization under Yamaguchi con-
ditions in acceptable yield. We also tried the Shiina macro-
lactonization method,[19] but the yield was not improved. The
removal of the TBS protecting group produced alcohols 20a
and 20b, which were separated readily.[6c] The unwanted
isomer 20b was recycled into 20a through an oxidation–
Scheme 5. Completion of the total synthesis of 1: a) 6a, 9-BBN, THF,
then [Pd2(dba)3] (6 mol%), dpephos (13 mol%), K3PO4, THF, reflux,
44 h, 87%; b) KOH, THF, MeOH, H2O, 84%; c) trichlorobenzoyl
chloride, Et3N, THF, then DMAP, toluene, 55%; d) TBAF, THF, 20a:
34%, 20b: 54%; e) DMP, CH2Cl2, 100%; f) NaBH4, CeCl3, MeOH,
99%; g) Ac2O, pyridine, 95%; h) DDQ, H2O, CH2Cl2, 95%; i) DMP,
CH2Cl2, j) 8, CrCl2, NiCl2, DMSO, 24a: 43%, 24b: 4% (2 steps);
k) Et3SiH, TFA, CH2Cl2, 08C, 71%. DDQ=2,3-dichloro-5,6-dicyano-1,4-
benzoquinone, DMP=Dess–Martin periodinane, DMSO=dimethyl
sulfoxide, TBAF=tetrabutylammonium fluoride, TFA=trifluoroacetic
acid.
reduction sequence during which the corresponding ketone
21 was reduced stereoselectively into 20a under Luche
conditions in EtOH. The acetylation of alcohol 20a and
removal of the 4-methoxybenzyl protecting group led to the
primary alcohol 22, which was oxidized to the fragile aldehyde
23. Fragment 8[6c] was synthesized in three steps from
homopropargylic alcohol by a zirconium-catalyzed carboalu-
mination reaction.[20]
The condensation of fragment 8 with aldehyde 23 under
the NHK conditions furnished alcohol 24a (43% yield from
alcohol 22) along with a trace amount of 24b. Nevertheless, a
special workup with sodium serinate to sequester the
chromium cations was necessary to obtain 24a in acceptable
yield.[21] To improve this step, we investigated a catalytic
[22]
version of the NHK coupling;
however, the expected
alcohol 24 was not formed under these conditions. In a final
step, deprotection of the masked carboxylic acid[23] in 24a
supplied the target compound (+)-oocydin A (1), the chem-
ical data of which were similar to those of the naturally
occurring
compound
([a]2D3 = + 11.8 degcm3 gꢀ1 dmꢀ1
MeOH);
lit.:[1] [a]2D3 = + 18.2 deg
(c=0.79 g/100 cm3,
cm3 gꢀ1 dmꢀ1).
In conclusion, we have completed the total synthesis of
(+)-oocydin A (1) in 18 steps from alcohol 14 and in 6.1%
overall yield. The subunit 6a was synthesized in eight steps in
39% overall yield by an efficient and stereoselective Pd0-
catalyzed cyclization to form the tetrahydrofuran ring. The
subunit 7 was obtained in 68% yield over 10 steps. The key
3764
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 3762 –3765

Angewandte
Chemie
[8] K. Takai, K. Kimura, T. Kuroda, T. Hiyama, H. Nozaki,
step of this total synthesis, a new application of the Suzuki–
Miyaura cross-coupling, constitutes the first implementation
of methodology that we had described previously.[5] This mild
reaction enabled the cross-coupling of the sensitive fragments
6a and 7 in 87% yield on a 4.9-mmol scale.
Tetrahedron Lett. 1983, 24, 5281 – 5284.
[9] A. M. Chatterjee, T.-L. Choi, D. P. Sanders, R. H. Grubbs, J. Am.
Chem. Soc. 2003, 125, 11360 – 11370.
[10] M. Scholl, S. Ding, C. W. Lee, R. H. Grubbs, Org. Lett. 1999, 1,
953 – 956.
[11] The relative configuration was determined by NOESY experi-
ments.
Received: February 5, 2008
Published online: April 10, 2008
[12] a) S. A. Stanton, S. W. Felman, C. S. Parkhurst, S. A. Godleski, J.
Am. Chem. Soc. 1983, 105, 1964 – 1969; b) G. Stork, J. M. Poirier,
J. Am. Chem. Soc. 1983, 105, 1073 – 1074; c) L. Jiang, S. D.
Burke, Org. Lett. 2002, 4, 3411 – 3414; d) M. J. Zacuto, J. L.
Leighton, Org. Lett. 2005, 7, 5525 – 5527.
[13] B. M. Trost, F. D. Toste, J. Am. Chem. Soc. 1999, 121, 4544 – 4554.
[14] S. Araki, H. Ito, Y. Butsugan, J. Org. Chem. 1988, 53, 1833 – 1835.
The Barbier-type reaction described in the literature did not
work in our hands.
ꢀ
Keywords: C C coupling · natural products · palladium ·
ring-closing metathesis · total synthesis
.
[1] G. Strobel, J. Li, F. Sugarawa, H. Koshino, J. Harper, W. M. Hess,
Microbiology 1999, 145, 3557 – 3564.
[2] a) N. Takada, H. Sato, K. Suenaga, H. Arimoto, K. Yamada, K.
Ueda, D. Uemura, Tetrahedron Lett. 1999, 40, 6309 – 6312; b) C.
Thaning, J. C. Welch, J. J. Borowicz, R. Hedman, B. Gerhardson,
Soil Biol. Chem. 2001, 33, 1817 – 1826.
[15] A. Fꢀrstner, O. R. Thiel, L. Ackermann, S. P. Nolan, H.-J.
Schanz, J. Org. Chem. 2000, 65, 2204 – 2207.
[16] P. S. Tiseni, R. Peters, Angew. Chem. 2007, 119, 5419 – 5422;
Angew. Chem. Int. Ed. 2007, 46, 5325 – 5328.
[3] H. Sato, H. Nakajima, T. Fujita, S. Takase, S. Yoshimura, T.
Kinoshita, H. Terano, J. Antibiot. 2005, 58, 634 – 639.
[4] a) T. Teruya, K. Suenaga, S. Maruyama, M. Kurotaki, H. Kigoshi,
Tetrahedron 2005, 61, 6561 – 6567; b) T. Teruya, H. Shimogawa,
K. Suenaga, H. Kigoshi, Chem. Lett. 2004, 33, 1184 – 1185; c) K.
Ueda, Y. Hu, Tetrahedron Lett. 1999, 40, 6305 – 6308.
[5] F. Liron, C. Fosse, A. Pernolet, E. Roulland, J. Org. Chem. 2007,
72, 2220 – 2223.
[17] K. Omura, D. Swern, Tetrahedron 1978, 34, 1651 – 1660.
[18] J. Li, X. Xu, Y. Zhang, Tetrahedron Lett. 2003, 44, 9349 – 9351.
[19] I. Shiina, M. Kubota, H. Oshiumi, M. Hashizume, J. Org. Chem.
2004, 69, 1822 – 1830.
[20] C. L. Rand, D. E. Van Horn, M. W. Moore, E.-i. Negishi, J. Org.
Chem. 1981, 46, 4093 – 4096.
[21] D. P. Stamos, C. Sheng, S. S. Chen, Y. Kishi, Tetrahedron Lett.
1997, 38, 6355 – 6358.
[6] a) T. R. Hoye, J. Wang, J. Am. Chem. Soc. 2005, 127, 6950 – 6951;
b) H. Kigoshi, M. Kita, S. Ogawa, M. Itoh, D. Uemura, Org. Lett.
2003, 5, 957 – 960; c) Y. Gu, B. B. Snider, Org. Lett. 2003, 5,
4385 – 4388.
[22] a) K. Namba, Y. Kishi, J. Am. Chem. Soc. 2005, 127, 15382 –
15383; b) K. Namba, S. Cui, J. Wang, Y. Kishi, Org. Lett. 2005, 7,
5417 – 5419; c) K. Namba, J. Wang, S. Cui, Y. Kishi, Org. Lett.
2005, 7, 5421 – 5424.
[7] J. Inanaga, K. Hirata, H. Saeki, T. Katsuki, M. Yamaguchi, Bull.
Chem. Soc. Jpn. 1979, 52, 1989 – 1993.
[23] D. A. Pearson, M. Blanchette, M. L. Baker, C. A. Guindon,
Tetrahedron Lett. 1989, 30, 2739 – 2742.
Angew. Chem. Int. Ed. 2008, 47, 3762 –3765ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3765