A general strategy for construction of both 2,6-cis- and 2,6-trans-disubstituted tetrahydropyrans: Substrate-controlled asymmetric total synthesis of (+)-scanlonenyne

Article abstract of DOI:10.1002/anie.200705663

(Chemical Equation Presented) A synthetic three-ring circus: The asymmetric total synthesis of (+)-scanlonenyne includes a sequential epimerization and intramolecular hetero-Michael addition for the construction of pyrano-γ-lactones (see scheme; DBU: 1,8-diazabicyclo[5.4.0]undec-7-ene), a highly efficient one-carbon homologation/bromination strategy, and a Weinreb ketone synthesis/cross-metathesis protocol for the elaboration of a sensitive side chain.

Full text of DOI:10.1002/anie.200705663

Communications
DOI: 10.1002/anie.200705663
Natural Product Synthesis
A General Strategy for Construction of Both 2,6-cis- and 2,6-trans-
Disubstituted Tetrahydropyrans: Substrate-Controlled Asymmetric
Total Synthesis of (+)-Scanlonenyne**
Hyunjoo Lee, Kwan Woo Kim, Janghyun Park, Hyoungsu Kim, Sanghee Kim, Deukjoon Kim,*
Xiangqian Hu, Weitao Yang, and Jiyong Hong
Functionalized 2,6-disubstituted tetrahydropyrans constitute
key structural features of a large number of biologically active
natural products.[1,2] We report herein a highly stereoselective
general strategy for the synthesis of both 2,6-cis-disubstituted
(a,a’-cis) and 2,6-trans-disubstituted (a,a’-trans) tetrahydro-
pyrans on the basis of a novel sequential epimerization and
intramolecular hetero-Michael addition of a hydroxybuteno-
lide.[3] To illustrate the potential of this methodology, the first
asymmetric total synthesis of (+)-scanlonenyne (1), a halo-
genated a,a’-trans-tetrahydropyranoid marine natural prod-
uct, has been accomplished in a completely substrate-
controlled fashion.
be incorporated in a stereoselective fashion, and 3) the C9
side chain, which possesses both a sensitive b-alkoxyketone
moiety and a Z-enyne unit.
As shown in Scheme 1, we envisioned that (+)-scanlone-
nyne (1) could be elaborated from bicyclic bromo-g-lactone 2,
which contains all four stereocenters of the natural product,
(+)-Scanlonenyne (1; see Scheme 1) was isolated by
Suzuki and co-workers from the red alga Laurencia obtusa
collected near Scanlonꢀs Island, off the western coast of
Ireland.[4] The constitution and relative stereochemistry of
scanlonenyne were established on the basis of an extensive
NMR spectroscopic study.[4] However, the absolute config-
uration, as well as the biological activity, of the pyranoid
marine natural product could not be investigated due to the
compoundꢀs lability.[4] Scanlonenyne represents the first
halogenated C15 acetogenin with a ketonic functionality at
the C7 position from a Laurencia species. The key structural
features of this unique and unstable natural product that need
to be addressed in a synthesis include 1) the a,a’-trans-
disubstituted tetrahydropyran ring with four stereogenic
centers, 2) the secondary bromine functionality, which must
Scheme 1. Retrosynthetic plan for (+)-scanlonenyne (1). Bn: benzyl;
IHMA: intramolecular hetero-Michael addition; PMB: p-methoxyben-
zyl.
[*] Dr. H. Lee, K. W. Kim, J. Park, Dr. H. Kim, Prof. Dr. S. Kim,
Prof. Dr. D. Kim
by a Weinreb ketone synthesis/cross-metathesis strategy. The
requisite bromo-g-lactone 2 could, in turn, be prepared from
key tricyclic intermediate 3 by a one-carbon homologation of
the C13 side chain and bromination with inversion of
configuration at the C12 position. However, the stereoselec-
tive introduction of the bromine substituent,[5] which is known
to be particularly challenging, and the projected one-carbon
extension[6] would be of considerable concern. Furthermore,
we were intrigued by the possibility of securing the desired
tricyclic a,a’-trans-disubstituted pyranolactone 3 by a sequen-
tial epimerization and intramolecular hetero-Michael addi-
tion of hydroxybutenolide 4, irrespective of its configuration
at the C10 position (see below). Further analysis suggested
that IHMA substrate 4 should be readily accessible from the
Evans syn-aldol adduct 5 of known glycolate oxazolidinone 6
with acrolein.
The Research Institute of Pharmaceutical Sciences
College of Pharmacy
Seoul National University
San 56-1, Shinrim-Dong, Kwanak-Ku, Seoul 151-742 (Korea)
Fax: (+82)2-888-0649
E-mail: deukjoon@snu.ac.kr
Dr. X. Hu, Prof. Dr. W. Yang, Prof. Dr. J. Hong
Department of Chemistry
Duke University
Durham, NC 27708 (USA)
[**] We thank Prof. M. Suzuki (Hokkaido University) for providing the
spectra for (+)-scanlonenyne. We are grateful to Dr. K. J. Shin
(Korea Institute of Science and Technology) for the X-ray crystal
structure determination of 2, ent-3a, and 4b. This work was
supported by the SRC/ERC program of MOST/KOSEF (grant no.:
R11-2007-107-02001-0).
To commence the synthesis, an aldol reaction[7] of the
dibutylboron enolate derived from the readily available
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
4200
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 4200 –4203

Angewandte
Chemie
glycolate oxazolidinone 6[8] with acrolein furnished the
corresponding syn-aldol adduct 5 (62%, 91% based on
recovered starting material (BRSM), d.r. 96:4 determined by
1H NMR analysis; Scheme 2); this enabled the relative and
Scheme 2. Preparation of IHMA substrate 4: a) 1. nBu2BOTf, Et3N,
CH2Cl2, ꢀ78!ꢀ408C, 30 min; 2. acrolein, ꢀ78!08C, 2 h, 62% (91%
BRSM); b) NaBH4, THF/H2O (3:1), room temperature, 2 h; c) PhCH-
(OMe)2, PTSA, CH2Cl2, room temperature, 36 h, 72% (2 steps);
d) PdCl2, CuCl, O2, DMF/H2O (7:1), room temperature, 6 h, 91%;
e) ethyl propiolate, LDA, THF, ꢀ78!ꢀ308C, 2 h, 87%; f) 1. H2,
Lindlar catalyst, EtOAc/pyridine (20:1), room temperature, 1.5 h;
2. silica gel, overnight, 97%; g) DDQ, CH2Cl2/buffer solution (pH 7.0;
9:1), room temperature, 1.5 h, 93%. DDQ=2,3-dichloro-5,6-dicyano-
1,4-benzoquinone, DMF=N,N-dimethylformamide, LDA=lithium di-
isopropylamide, PTSA=p-toluenesulfonic acid, Tf=trifluoromethane-
sulfonyl, THF=tetrahydrofuran.
Scheme 3. IHMA of C12/C13-syn-hydroxybutenolides 4: a) DB U
(2 equiv), CH2Cl2 (0.005m), 348C, 24 h, 76% (87% BRSM).
diaxial substituents at the C10 and C12 positions (as indicated
in Scheme 3). The relative configuration of the newly
generated stereocenters of tricyclic g-lactone 3a was assigned
by NOESY studies and was further verified by single-crystal
X-ray analysis (see the Supporting Information).[13] The X-ray
crystallographic studies revealed that the tetrahydropyran
ring of 3a assumes a twist-boat conformation.
It is worth mentioning at this point that, in contrast to the
results with C12/C13-syn-hydroxybutenolides 4, intramolec-
ular hetero-Michael addition of a 1:1 mixture of C12/C13-
anti-hydroxybutenolides 4’ under comparable conditions
produced a,a’-cis-pyrano-g-lactone 3’b as the sole product
in 88% yield (Scheme 4).[14,16] The preferred chair-like
absolute stereochemistry at the C12/C13 positions of IHMA
substrate 4 to be established. Reductive cleavage of the chiral
auxiliary in 5, followed by protection of the resulting diol 7 as
the benzylidene acetal, provided cyclic acetal 8 in 72% yield
for the 2 steps after isolation. We were pleased to find that a
highly regioselective Wacker reaction[9,10] of monosubstituted
terminal alkene 8 produced the desired aldehyde 9 in
excellent yield (91%).[11] Addition of the lithium anion of
ethyl propiolate to aldehyde 9, followed by semihydrogena-
tion of the resulting acetylenic ester 10 with Lindlar catalyst,
gave rise directly to the desired butenolide 11 (84%, 2 steps).
Oxidative cleavage of the PMB group in 11 by the Yonemitsu
method[12] then afforded key intramolecular hetero-Michael
precursor 4 as a 2.3:1 diastereomeric mixture at the C10
position in 93% yield. The configuration of the major isomer
at the C10 position was determined to be S by X-ray
crystallography (see the Supporting Information).[13]
Having developed a concise synthetic route to the key
substrate for the intramolecular conjugate addition, we were
delighted to find that, upon exposure to 1,8-diazabicyclo-
[5.4.0]undec-7-ene (DBU) in dichloromethane, the 2.3:1
mixture of hydroxybutenolides 4 furnished the desired g-
lactone 3a in 76% yield with 25:1 a,a’-trans:-cis selectivity.
This sequential epimerization and IHMA process, which we
believe proceeds through the intermediacy of b,g-unsaturated
butenolide isomer 4c, establishes the configuration at the C9
and C10 positions in a single operation (Scheme 3).[14,15]
The observed stereochemical outcome can be rationalized
by considering that the internal hetero-Michael addition of
10R isomer 4a via boat-like transition state A proceeds faster
than that of 10S isomer 4b. Chair-like transition state B for 4b
suffers unfavorable nonbonding interactions between the two
Scheme 4. IHMA of C12/C13-anti-hydroxybutenolides 4’: a) DB U
(2 equiv), CH2Cl2 (0.005m), 248C, 24 h, 88%.
Angew. Chem. Int. Ed. 2008, 47, 4200 –4203
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4201

Communications
transition-state geometry B’, which is devoid of the afore-
significant challenge. After extensive experimentation, we
made the unanticipated observation that 14 is a far better
substrate for the bromination than 15 (Scheme 5).[20] Thus,
mentioned destabilizing interactions in B, led to the formation
of a,a’-cis-pyrano-g-lactone 3’b. The NOE interactions
between Ha and Ha’ in 3’b were consistent with a cis
orientation.
With the synthesis of tricyclic a,a’-trans-lactone 3a
accomplished, we focused our attention on its conversion
into key bromo-g-lactone intermediate 2 (Scheme 5). As we
[21]
exposure of 14 to DPPE/Br2 furnished the corresponding
bromide 16 in good yield (77% for 14 versus 17% for 15).
Finally, we were pleased to find that chemoselective hydro-
genolysis of the vinylic dibromide moieties in tribromide 16 in
the presence of the C12 alkyl bromide function, followed by
hydrogenation of the resulting olefin, delivered key bromo-g-
lactone 2 in excellent yield (93%). The structure of 2 was
firmly established by X-ray crystallography.[13]
With key bromo-g-lactone 2 in hand, we proceeded to
address the assembly of the C9 side chain by sequential
installation of the b-alkoxyketone and the Z-enyne moieties
(Scheme 6). It should be emphasized at this point that
Scheme 5. Synthesis of key bromo-g-lactone 2: a) 1. H2, Pd(OH)2,
THF, room temperature, 2 h; 2. hexamethyldisilazane, imidazole, 608C,
2 h; b) 1. CrO3, pyridine, CH2Cl2, room temperature, 1 h; 2. CBr4, Ph3P,
room temperature, 10 min; c) PPTS, MeOH, room temperature,
10 min, 64% from 3a; d) DPPE/Br2, CH2Cl2/toluene (2:1), 708C, 6 h,
77%; e) H2, 10% Pd/C, EtOAc, room temperature, 1 h, 93%.
DPPE=1,2-bis(diphenylphosphino)ethane, PPTS=pyridinium p-tolue-
nesulfonate, TMS=trimethylsilyl.
Scheme 6. Completion of the synthesis of (+)-scanlonenyne (1):
a) MeONHMe·HCl, Me2AlCl, CH2Cl2, room temperature, 50 min, 95%;
=
b) TBSOTf, 2,6-lutidine, CH2Cl2, ꢀ308C, 10 min, 90%; c) H2C CH-
(CH2)2MgBr, THF, room temperature, 40 min, 93%; d) enyne C,
Grubbs catalyst D, benzene, 708C, 6 h, 50% (83% BRSM), Z/E=5:1;
e) TBAF, THF, ꢀ308C, 30 min, 77%. Mes=2,4,6-trimethylphenyl,
TBS=tert-butyldimethylsilyl, TBAF=tetra-n-butylammonium fluoride,
TIPS=triisopropylsilyl.
initially feared, the requisite one-carbon homologation at the
C13 position in the presence of the g-lactone functionality
proved to be problematic. However, we were able to devise
an alternative synthetic sequence for this purpose by taking
advantage of 1) the highly efficient transformation of an
aldehyde function to the corresponding 1,1-dibromoalkene
under the mild conditions of Corey and Fuchs[17] and 2) the
ready hydrogenolysis of a carbon–halogen bond by using a
palladium catalyst. Thus, removal of the benzylidene protect-
ing group in 3a by hydrogenolysis with Pearlmanꢀs catalyst,[18]
followed by protection of the resulting diol with hexamethyl-
disilazane in a one-pot procedure, afforded bis-TMS ether 12.
A novel one-pot chemoselective desilative Collins oxidation/
Corey–Fuchs olefination of the unstable bis-TMS ether 12 led
to 1,1-dibromoalkenyl alcohol 14 after removal of the TMS
group in intermediate 13.[19] Experimentally, alcohol 14 could
be prepared in excellent overall yield from 3a (64%) without
purification of intermediates.
introduction of a ketone function at the C7 position renders
the tetrahydropyran ring susceptible to rupture by a retro-
Michael reaction. With this potential problem in mind,
exposure of bromo-g-lactone 2 to Me2AlCl/MeONHMe·HCl
by using the Shimizu–Nakata modification of the Weinreb
protocol,[22] and subsequent silylation of the resulting g-
hydroxy-N-methoxy-N-methylamide 17 with TBSOTf, fur-
nished TBS-protected Weinreb amide 18 in good yield (86%,
2 steps). Gratifyingly, treatment of Weinreb amide 18 with
butenylmagnesium bromide provided the desired b-alkoxy-
ketone 19 in excellent yield (93%).[23]
To complete the synthesis, we believed that our recently
reported modification of the cross-metathesis protocol of Lee
and co-workers[24] would be ideal for incorporation of the Z-
enyne unit in the presence of the sensitive b-alkoxyketone
moiety. We were pleased to find that, upon exposure to enyne
C and Grubbs catalyst D, b-alkoxyketoalkene 19 delivered
Exposure of 1,1-dibromoalkenyl alcohol 14 to H2/Pd then
gave the requisite 13-ethyl alcohol 15, ready for the crucial
bromination. As anticipated, however, bromination of alco-
hol 15 with inversion of configuration turned out to be a
4202 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 4200 –4203

Angewandte
Chemie
[7] D. A. Evans, J. Bartroli, T. L. Shih, J. Am. Chem. Soc. 1981, 103,
the desired Z enyne 20 as a 5:1 mixture of Z and E isomers in
50% yield. Finally, the removal of both silyl groups in 20 by
treatment with TBAF at low temperature yielded (+)-
scanlonenyne (1) in 77% yield. It is interesting to note that
treatment of enyne 20 with TBAF at room temperature
produced epoxide E as the major product from the afore-
mentioned retro-Michael process. The spectral and optical
rotation data of our synthetic material were in good agree-
ment with those of the natural product. However, closer
examination of the proton and carbon NMR spectra of both
the natural and synthetic materials revealed the presence of a
small amount of inseparable substances, probably a diaste-
reomeric mixture of hemiketals 1’ in equilibrium with the
hydroxyketo form 1.
2127 – 2129.
[8] T. K. Jones, S. G. Mills, R. A. Reamer, D. Askin, R. Desmond,
R. P. Volante, I. Shinkai, J. Am. Chem. Soc. 1989, 111, 1157 –
1159.
[9] For reviews of the Wacker oxidation, see: a) J. Tsuji, Synthesis
1984, 369 – 384; b) J. M. Takacs, X.-T. Jiang, Curr. Org. Chem.
2003, 7, 369 – 396.
[10] For the aldehyde-selective Wacker oxidation of allylic diol
derivatives, see: S.-K. Kang, K.-Y. Jung, J.-U. Chung, E.-Y.
Namkoong, T.-H. Kim, J. Org. Chem. 1995, 60, 4678 – 4679.
[11] It is interesting to note that the Wacker oxidation of the C12/
C13-anti isomer under comparable conditions gave the corre-
sponding ketone as the major regioisomer; see the Supporting
Information.
[12] Y. Oikawa, T. Yochika, O. Yonemitsu, Tetrahedron Lett. 1982,
23, 885 – 888.
In summary, the first asymmetric total synthesis of (+)-
scanlonenyne (1) has been achieved in 18 steps in 9% overall
yield from readily available glycolate oxazolidinone 6 and
acrolein. Our completely substrate-controlled synthesis of
this labile C15 acetogenin features a number of stereo-, regio-,
and chemoselective transformations including 1) a novel
sequential epimerization and intramolecular hetero-Michael
addition of a hydroxybutenolide for construction of both the
a,a’-cis- and a,a’-trans-pyranolactones, 2) a novel highly
efficient one-carbon homologation/bromination strategy,
and 3) a Weinreb ketone synthesis/cross-metathesis protocol
for the elaboration of the sensitive C9 side-chain appendage.
In addition, the present synthesis establishes the absolute
configuration of the natural product.
[13] CCDC-656345 (for 2), CCDC-667564 (for ent-3a), and CCDC-
656346 (for 4b) contain the supplementary crystallographic data
for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.
cam.ac.uk/data_request/cif; see also the Supporting Information.
[14] See the Supporting Information for the optimized molecular
geometries of 3a, 3b, 3’a, and 3’b. The global minimum
conformations (Spartan06, MM2 force field, Monte Carlo
conformational search) were further optimized by density
functional theory (B3LYP) at the 6-31G* level (Gaussian03,
D02 version); P. J. Stephens, F. J. Devlin, C. F. Chabalowski,
M. J. Frisch, J. Phys. Chem. 1994, 98, 11623 – 11627. Note that the
phenyl substituent has been replaced by a methyl substituent to
expedite the B3LYP/6-31G* optimizations. The calculation
showed that the ethylidene derivatives of 3a and 3’b were
more stable than those of 3b and 3’a by 0.6 and 3.7 kcalmolꢀ1
,
respectively.
Received: December 11, 2007
Published online: April 25, 2008
[15] a) Separation of the 10R and 10S isomers and subjection of each
isomer to the IHMA conditions gave roughly the same result.
b) The minor a,a’-cis isomer 3b did not equilibrate to the
corresponding trans isomer 3a under the reaction conditions.
[16] Independently synthesized 3’a equilibrated completely to 3’b
after 80 h under the reaction conditions ( ꢁ 30% after 24 h).
[17] a) F. Ramirez, N. B. Desai, N. McKelvie, J. Am. Chem. Soc. 1962,
84, 1745 – 1747; b) E. J. Corey, P. L. Fuchs, Tetrahedron Lett.
1972, 13, 3769 – 3772.
[18] W. M. Pearlman, Tetrahedron Lett. 1967, 8, 1663 – 1664.
[19] a) R. Ratcliffe, R. Rodehorst, J. Org. Chem. 1970, 35, 4000 –
4002; b) R. Mahrwald, F. Theil, H. Schick, H.-J. Palme, H.
Nowak, G. Weber, S. Schwarz, Synthesis 1987, 1012 – 1013, and
references therein; c) J. Muzart, Synthesis 1993, 11 – 27.
[20] The yields of DPPE/Br2 bromination of C13-ethyl, -vinyl, -(Z)-1-
bromovinyl, -(E)-1-bromovinyl, and -1,1-dibromovinyl groups
were 17, 13, 53, 10, and 77%, respectively.
[21] a) L. Horner, H. Oediger, H. Hoffmann, Justus Liebigs Ann.
Chem. 1959, 626, 26 – 34; b) S. P. Schmidt, D. W. Brooks,
Tetrahedron Lett. 1987, 28, 767 – 768.
[22] a) A. Basha, M. Lipton, S. M. Weinreb, Tetrahedron Lett. 1977,
18, 4171 – 4172; b) T. Shimizu, K. Osako, T. Nakata, Tetrahedron
Lett. 1997, 38, 2685 – 2688, and references therein.
[23] a) S. Nahm, S. M. Weinreb, Tetrahedron Lett. 1981, 22, 3815 –
3818; b) protection of the hydroxy group at the C10 position was
critical for the success of the Weinreb ketone synthesis.
[24] H. Kim, H. Lee, D. Lee, S. Kim, D. Kim, J. Am. Chem. Soc. 2007,
129, 2269 – 2274, and references therein.
Keywords: asymmetric synthesis · bromination ·
Michael addition · natural products · tetrahydropyrans
.
[1] For reviews, see: a) T. L. B. Boivin, Tetrahedron 1987, 43, 3309 –
3362; b) R. D. Norcross, I. Paterson, Chem. Rev. 1995, 95, 2041 –
2114; c) E. J. Kang, E. Lee, Chem. Rev. 2005, 105, 4348 – 4378;
d) P. A. Clarke, S. Santos, Eur. J. Org. Chem. 2006, 2045 – 2053.
[2] a) For a lead reference concerning syntheses of 2,6-disubstituted
tetrahydropyrans, see: N. Kawai, J.-M. Lagrange, M. Ohmi, J.
Uenishi, J. Org. Chem. 2006, 71, 4530 – 4537; b) H. G. Lee, I. L.
Lysenko, J. K. Cha, Angew. Chem. 2007, 119, 3390 – 3392;
Angew. Chem. Int. Ed. 2007, 46, 3326 – 3328.
[3] For an example of an intramolecular hetero-Michael addition to
a hydroxybutenolide for construction of a pyrano-g-lactone, see:
G. A. Kraus, B. Roth, J. Org. Chem. 1978, 43, 4923 – 4924.
[4] M. Suzuki, Y. Takahashi, Y. Matsuo, M. D. Guiry, M. Masuda,
Tetrahedron 1997, 53, 4271 – 4278.
[5] a) H. M. Sheldrake, C. Jamieson, J. W. Burton, Angew. Chem.
2006, 118, 7357 – 7360; Angew. Chem. Int. Ed. 2006, 45, 7199 –
7202, and references therein; b) J. Park, B. Kim, H. Kim, S. Kim,
D. Kim, Angew. Chem. 2007, 119, 4810 – 4812; Angew. Chem. Int.
Ed. 2007, 46, 4726 – 4728.
[6] a) J. W. Burton, J. S. Clark, S. Derrer, T. S. Stork, J. G. Bendall,
A. G. Holmes, J. Am. Chem. Soc. 1997, 119, 7483 – 7498; b) M. T.
Crimmins, E. A. Tabet, J. Am. Chem. Soc. 2000, 122, 5473 – 5476.
Angew. Chem. Int. Ed. 2008, 47, 4200 –4203
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4203