Synthesis of undecachlorosulfolipid A: Re-evaluation of the nominal structure

Article abstract of DOI:10.1002/anie.201102521

Halo-giant: The title compound at the left in the scheme was constructed by the synthesis and coupling of two stereochemically challenging fragments. A comparison of the NMR data of the synthetic material and the natural product indicates that the configuration had been misassigned in the latter. PT=5-(1-phenyltetrazolyl).

Full text of DOI:10.1002/anie.201102521

Communications
DOI: 10.1002/anie.201102521
Chlorosulfolipids
Synthesis of Undecachlorosulfolipid A: Re-evaluation of the Nominal
Structure**
Christian Nilewski, Nicholas R. Deprez, Thomas C. Fessard, Dong Bo Li, Roger W. Geisser, and
Erick M. Carreira*
The chlorosulfolipids are an intriguing family of natural
products initially isolated in 1969.[1] They originate from the
phylogenetic class Chrysophyceae that includes golden algae.
These structures are fascinating from a variety of perspectives
and raise questions of relevance in chemistry, biology,
toxicology, and pharmacology. This has led to a remarkable
interest in these entities, as evidenced by numerous publica-
tions.[2–4] It is interesting to note that some of the more
recently isolated chlorosulfolipids have been reported to be
moderately cytotoxic and associated with seafood poisoning.
The chlorosulfolipids are chemically intriguing because of
the complex stereochemical array of secondary chlorides and
alcohols, some of which are O-sulfated. The structures have
provided a forward stage for the development of novel tactics
and strategies en route to their synthesis[5] as well as
spectroscopic studies that enable configurational assign-
ment.[3] The most complicated of the chlorosulfolipids
known to date was isolated in 2002 from the Mediterranean
mussels M. galloprovincialis in which microalgae accumulate,
but knowledge of the organism that produces it remains
elusive.[6] Herein, we document the first total synthesis of
nominal undecachlorosulfolipid A (1), which not only led to
the development of tactics and strategies to access this
previously reported.[2] However, a collection of altogether
different stereochemical patterns is embedded in undeca-
chlorosulfolipid A (1), rendering it substantively more chal-
lenging. These include, for example, the dichloro alcohol
along C17–C19, the congested array at C6–C8, which includes
a vicinal chlorohydrin at C7–C8 juxtaposed to a geminal
dichloride at C6. Finally, any synthesis route would have to
contend with the sheer complexity of nine contiguous
chlorinated or hydroxylated stereogenic centers and a target
structure that is amphipathic, incorporating a charged sulfate
and a palmitoyl ester side chain.
Our retrosynthetic analysis led to the disconnection of 1
into two fragments of comparable complexity, including the
C11–C24 and the C1–C10 fragments. We commence the
disclosure of our work with the preparation of the former
(Scheme 1). Unsaturated ester 2 was prepared in a few steps
Scheme 1. Synthesis of the C11–C24 fragment. a) Et4NCl3, CH2Cl2,
08C, 1.25 h, 66% (d.r. =1.8:1); b) DIBAL-H, toluene, 08C, 30 min,
43% (d.r.=5:1); c) DMAP (cat.), Et3N, AcCl, CH2Cl2, 08C, 10 min,
81%; d) AD-Mix b, MeSO2NH2 (1.05 equiv), tBuOH/H2O=1:1, 74%;
e) DABCO, (F3CSO2)2O, CH2Cl2, À788C to RT, 19 h, 50%; f) CSA
(cat.), MeOH, RT, 2 h, 72%; g) 2,6-lutidine, TBDMSOTf, CH2Cl2,
À788C to À158C, 1.75 h, 91%; h) HF/pyr/THF, 08C to 48C, 5 h, 40%
4, 53% recovered starting material (after four cycles 79% overall yield,
86% brsm); i) DMP, RT, 45 min, 87%; j) 7 (1.3 equiv), KN(SiMe3)2,
THF, 08C, 20 min, then À788C, RCHO (1.0 equiv), 30 min, 08C,
30 min, 55%; k) Et4NCl3, CH2Cl2, À788C, 71% (d.r. =5:1); l) K2CO3,
MeOH, 08C, 10 min, 98%. Bn=benzyl, brsm=based on recovered
starting material, CSA=camphorsulfonic acid, DABCO=1,4-
diazabicyclo[2.2.2]octane, DMAP=4-dimethylaminopyridine,
DMP=Dess–Martin periodinane, pyr=pyridine, TBS=tert-butyl-
dimethylsilyl.
complex polychlorinated lipid, but it also revealed a structural
misassignment of the natural product.
Our interest in this molecule arose from the unique
synthetic challenges that this molecule presents. Examination
of undecachlorosulfolipid A (1) reveals some structural
homology to hexachlorosulfolipid[7] whose synthesis we have
[*] Dr. C. Nilewski, Dr. N. R. Deprez, Dr. T. C. Fessard, Dr. D. B. Li,
Dr. R. W. Geisser, Prof. Dr. E. M. Carreira
ETH Zꢀrich, HCI H335, 8093 Zꢀrich (Switzerland)
E-mail: carreira@org.chem.ethz.ch
from commercially available (S)-1,2,4-butanetriol.[8] Dichlori-
nation[9] of 2 gave the product displaying an anti relationship
between C15 and C17 preferentially (d.r. = 1.8:1; Scheme 1).
The relative configuration was unambiguously secured by J-
based configuration analysis (JBCA)[3,10] following diastereo-
mer separation and acetonide hydrolysis. On preparative
scale it was more convenient to subject the mixture of
diastereomers directly to reduction (DIBAL-H) and separate
Homepage: http://www.carreira.ethz.ch
[**] We thank Louis Bertschi for assistance with HPLC–MS, Dr. Marc-
Olivier Ebert, Rainer Frankenstein, and Philipp Zumbrunnen (ETH
Zꢀrich) and Dr. S. Loss (Bruker Biospin AG, Switzerland) for
assistance with NMR data, and Dr. W. Bernd Schweizer for X-ray
analysis. C.N. thanks the Fonds der Chemischen Industrie (DE) and
Novartis (CH) for support.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201102521.
7940
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 7940 –7943

the resulting allylic alcohols. Acetylation (81% yield),
Sharpless dihydroxylation[11] (74%), and subsequent cyclo-
dehydration (50%)[2,3] furnished the protected epoxy alcohol
acetylide addition to an a,a-dichlorinated aldehyde. Con-
version of the propargylic alcohol into diol 11 commenced
with alkyne semireduction (Red-Al) and subsequent vana-
dium-catalyzed epoxidation. We were unable to identify
conditions for the efficient regioselective opening of the
isolated epoxide to the corresponding chlorohydrin. How-
ever, oxidation of the intermediate secondary epoxyalcohol
to the ketone enabled the subsequent chloride introduction at
C8 with ZrCl4. In order to minimize epimerization at Ca, the
resulting chloroketone was reduced without purification,
which provided diol 11 in 34% yield over the three steps.
The configuration of 11 was confirmed by JBCA combined
with evaluation of the NMR spectra of the two diastereomeric
Mosher ester derivatives at the C9-OH group.[8] Acetonide
protection of the diol (93% yield) was followed by removal of
the TBDPS group (95% yield), Dess–Martin oxidation of the
resulting primary alcohol, and subsequent Still–Gennari
olefination,[17] leading to cis ester 12 in 68% yield. The
intermediate cis allylic alcohol resulting from DIBAL reduc-
tion of 12 was subjected to diastereoselective epoxidation
(d.r. = 9:1)[18] in 92% yield. Subsequently, the chloride at C3
was introduced with TiCl(OiPr)3 (40%).[19] In order to
establish the regiochemical outcome, we treated the major
product of the reaction with NaIO4,[8] which led to an
aldehyde, consistent with the formation of a 1,2-diol in the
preceding step. Formation of the C2,C3 cis epoxide from a
mono-TBS-protected derivative of 13 established the inver-
tive epoxide opening.[8] The synthesis of the C1–C10 fragment
was then completed following acetonide formation, benzyl
ether cleavage, and Mitsunobu displacement of the resulting
primary alcohol with phenyltetrazolylsulfide followed by
oxidation to the corresponding sulfone (40% yield over
four steps).
1
3.[12] The vicinal H,1H coupling constant between H12 and
H13 in 3 (Jvic = 4.0 Hz) is in agreement with a cis epoxide.[13,14]
Intermediate acetonide 3 was then transformed into the
primary alcohol 4 by a sequence involving acetonide cleavage
(72%), silyl protection (91%), and selective deprotection
(HF–pyr, 79% overall yield). Dess–Martin oxidation[15] of 4
led to an aldehyde that was allowed to react with the ylide
derived from 7 to give (Z)-5 in 55% yield. The configuration
of the olefin was deduced from the NMR-spectroscopic
fingerprint of H18 and H19 (Jvic = 11.0 Hz in C6D6). The
synthesis of the C11–C24 fragment was then completed by
treatment of 5 with Et4NCl3 (71%, d.r. = 5:1) followed by
saponification (98%). The stereochemical outcome of the
dichlorination at C18,C19 was confirmed by JBCA as well as
chemical modification, that is, formation of a cis epoxide
between C17 and C18 and stereoselective formation of an
olefin along C18 and C19.[8]
The preparation of the C1–C10 fragment starts with diol 8
which was converted into aldehyde 9 in three steps and 83%
overall yield (Scheme 2). The stereogenic center at C7 was
then envisioned to be installed by enantioselective Zn-
acetylide addition to a dichloro aldehyde.[16] The addition
was found to proceed in 70% yield to give propargylic alcohol
10 in 92% ee as judged by 1H NMR spectroscopic analysis of
the corresponding Mosher ester derivatives. This transforma-
tion represents the first example of an asymmetric Zn-
With the completion of both key fragments, the stage was
set to examine the crucial fragment coupling (Scheme 3).
Dess–Martin oxidation of primary alcohol 6 gave the
corresponding aldehyde that was conveniently used directly
in the next step. Addition of freshly prepared NaHMDS
solution to a cold (À788C) solution of 15 and the epoxyalde-
hyde derived from 6 followed by slow warming to room
temperature gave the coupled product 16 as a 3:1 Z/E
mixture.[20] Chromatographic separation of the diastereomers
proved troublesome; however, epoxide opening with
Scheme 2. Synthesis of the C1–C10 fragment. a) tBuPh2SiCl, imida-
zole, DMF, 08C to RT, 16 h, 96%; b) TEMPO, KBr, NaOCl, CH2Cl2,
08C, 2 h, 93%; c) tBuNH2, NCS, CCl4, 08C to RT, 12 h, then HCl, RT,
2 h, 96%; d) BnOCH2CCH, (À)-N-methylephedrine, Zn(OTf)2, Et3N,
toluene, 24 h, RT, 70% (92% ee); e) NaAlH2(OCH2CH2OMe)2, THF,
À788C to RT, 2 h, 92%; f) VO(acac)2, tBuO2H, CH2Cl2, 08C to RT,
19 h, 62%; g) DMP, CH2Cl2, 08C to RT, 1.5 h, 95%; h) ZrCl4, CH2Cl2,
08C to RT, 20 min; i) NaBH4, MeOH, À788C, 20 min, 36% over two
steps; j) MeCH(OMe)CH2, PPTS, CH2Cl2, 08C to RT, 18 h, 93%;
k) Bu3NF, AcOH, DMF, RT, 24 h, 95%; l) DMP, CH2Cl2, 08C to RT, 1 h;
m) (CF3CH2O)2P(O)CH2CO2Me, KN(SiMe3)2, THF, À788C, 0.5 h, 68%
(2 steps); n) iBu2AlH, THF, À788C, 4 h, 88%; o) Ti(OiPr)4, tBuO2H,
(+)-diethyl l-tartrate, CH2Cl2, À208C, 18 h, 92% (d.r. =9:1); p) TiCl-
(OiPr)3, C6H6, RT, 1 h, 40%; q) CuSO4, TsOH, acetone, RT, 20 h, 83%;
r) Pd/C, H2, EtOAc, RT, 3 h, 94%; s) 1-phenyl-1H-tetrazole-5-thiol,
(iPrO2C)N2, PPh3, THF, 08C to RT, 1 h, 85%; t) mCPBA, 08C to 408C,
22 h, 61%. mCPBA=meta-chloroperbenzoic acid, NCS=N-chlorosuc-
cinimide, PPTS=pyridinium p-toluenesulfonate, TBDPS=tert-butyldi-
phenylsilyl, TEMPO=2,2,6,6-tetramethylpiperidine N-oxyl.
[21]
Ph3PCl2
led to the corresponding chlorohydrins, which
could be separated easily to afford 17 in 64% yield. As a
result of our observation that epoxide openings in related
systems can occur with retention of configuration,[2] the
stereochemical outcome of this transformation was inves-
tigated by subsequent based-induced ring closure, leading to
the starting C12,C13 cis epoxide, supporting the C12–C13 syn
arrangement in 17. Further evidence for the configurational
assignment was secured at a later stage of the synthesis (cf.
18).[8] Subsequent dichlorination of the C11–C10 double bond
using Et4NCl3 provided 18 as the major diastereomer in 70%
yield.[22] The stereochemical outcome of the dichlorination
was established by application of JBCA. In particular,
combined analysis of the 1H NMR, 13C NMR, COSY,
HSQC, PS-HMBC, HSQC-HECADE, and ROESY spectra
of 18 revealed coupling constant patterns along C11–C12 that
Angew. Chem. Int. Ed. 2011, 50, 7940 –7943
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7941

Communications
latter required prolonged stirring (4 h) at room temperature.
Initially, the product was identified by HR ESIMS and
purified by MS-guided HPLC techniques. Subsequent acquis-
1
ition of H NMR as well as COSY and HSQC spectra then
provided spectral data for the synthetic material. Surprisingly,
there were discrepancies in the characteristic region 4–6 ppm,
which includes the protons of the chlorinated and hydroxy-
lated stereogenic centers as well as the olefin protons. This
result compelled us to re-examine the published spectral data
for the natural product.
Careful reconsideration of the data available in the report
of the isolation of 1[6] led to concerns about the assignment of
the configuration at C23, which is based on a temperature-
dependent change of chemical shifts of the neighboring
protons upon esterification of C23-OH with (R)-MTPA
chloride. In the original disclosure of this method by Riguera
et al., methoxy phenylacetic acid (MPA) derivatives were
employed.[24] It is known that the shielding and deshielding
effects of MTPA esters are the result of a complex conforma-
tional equilibrium and typically lead to opposite changes in
the chemical shifts of the neighboring protons compared to
those of MPA derivatives of the same configuration.[25]
Consequently, the implementation of this method without
modification is suspect. In this regard, careful scrutiny of the
data reveals that the original assignment was based on an
erroneous assumption concerning the conformational prefer-
ences of MTPA esters. We thus suggest that the configuration
at C23 may be R instead of S. Some other features of the
described data were intriguing. In this context, we note with
surprise the report that esterification of 1 with Mosherꢀs acid
chloride leads to a product with the free primary hydroxy
group at C1, while the secondary hydroxy group at C2 formed
the corresponding esters. Unless this results from selective
hydrolysis of the less hindered ester during workup (K2CO3,
H2O), such reactivity would stand in stark contrast to what is
typical.[26] Additionally, we noted that analysis of the coupling
constants provided by the authors lead to a conformation
including a syn-pentane interaction between the C19-chloride
and the C17-sulfate groups. Although such a conformation
cannot be dismissed, the assignment does suggest some
caution. Regretfully, the original spectra in electronic form
were not available, which hampers a more detailed re-
examination at this time.
In summary, we have developed a route to the nominal
structure for undecachlorosulfolipid A (1), the most complex
of the chlorosulfolipids isolated to date. As a result of our
studies we note that the configuration of the natural product
is misassigned. In addition to this important result some
salient synthetic features are documented. These features
include: the first example of an alkyne addition to an a,a-
dichloro aldehyde with excellent enantiocontrol, the use of a
Z-selective Julia–Kocienski olefination for the coupling of
two stereochemically complex chlorinated fragments, a strat-
egy for the introduction of the C21–C22 olefin that allows for
regioselective introduction of the requisite double bond by
controlled elimination with the Martin sulfurane reagent.
Given the chemical, biological, and toxicological interest in
these structures, the work we delineate collectively is of
importance for ongoing efforts to understand these fascinat-
Scheme 3. Fragment coupling and completion of the synthesis.
a) DMP, CH2Cl2, 08C to RT, 95%; b) 15 (1.2 equiv), toluene, À788C,
then NaN(SiMe3)2, À788C to RT, 67% (Z/E=3:1); c) Ph3PCl2, CH2Cl2,
08C, 64%; d) Et4NCl3, CH2Cl2, 08C to 48C, 70%; e) H2 (1 atm), Pd/C
(20 mol%), EtOAc, 1.5 h, RT, 79%; f) Martin sulfurane, C6H6, RT, 2 h,
50%; g) HF–pyr, pyr, MeCN, 08C to RT to 408C, 92 h, quant.;
h) palmitoyl chloride, pyr, CH2Cl2, À788C to À408C, then, À788C,
MeOH, À788C to RT; 60%; i) DMF–SO3 (excess), NaSO4, DMF/pyr,
08C to RT to 458C, 5 h 20 min, 60%; j) F3CCO2H/H2O (1:1), 4 h, 08C
to RT.
are, in combination with the observed ROESY correlation
between H10 and H13, typical of a syn arrangement.
The focus was next set on the introduction of the C21–C22
double bond. Hydrogenolytic deprotection of the benzylether
provided 19 in 79% yield and subsequent treatment with
Martin sulfurane[23] gave the E olefin 20 in 50% yield. The
latter transformation is noteworthy because of the chemo-
selectivity in the hydroxyl activation step (C22-OH vs. C13-
OH) as well as the regioselectivity in the subsequent
elimination. The final steps of the route were then examined.
The TBS groups in 20 were cleaved by reaction with HF–
pyridine. Subsequent treatment of the resulting triol with
palmitoyl chloride and pyridine at low temperature allowed
for the selective esterification of the C23-OH group to give 21
in 60% yield over the last two steps. The chemoselectivity in
the esterification step was deduced from the spectroscopic
fingerprint of H23, which experiences a downfield shift from
d = 4.23 to 5.30 ppm in [D6]acetone upon esterification, as
supported by COSYexperiments. Selective mono-sulfation of
C17-OH in 21 was effected with DMF–SO3 in pyridine and
proceeded in 60% yield; however, hydrolysis of the aceto-
nides proved difficult. The dioxolane is rapidly hydrolyzed
within minutes (TFA/H2O 1:1) while the 1,3-dioxane was
surprisingly stable under these conditions. Removal of the
7942 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 7940 –7943

[14] Additionally, epoxide opening of the free epoxyalcohol derived
from 3 using TiCl(OiPr3) provided access to a crystalline diol,
whose configuration could be established by X-ray crystallog-
raphy; see the Supporting Information for details.
ing natural products. Moreover, the route described delin-
eates a roadmap in the form of strategy and tactics for the
synthesis of the most complex chlorosulfolipid. The config-
urational uncertainty provides an additional challenge to
researchers in the field, which will likely be resolved only
through chemical synthesis.[27]
[15] D. B. Dess, J. C. Martin, J. Org. Chem. 1983, 48, 4155 – 4156.
[16] a) D. E. Frantz, R. Fꢁssler, E. M. Carreira, J. Am. Chem. Soc.
1999, 121, 11245 – 12246; b) D. E. Frantz, R. Fꢁssler, E. M.
Carreira, J. Am. Chem. Soc. 2000, 122, 1806 – 1807; c) D. Boyall,
F. Lopez, H. Sasaki, D. E. Frantz, E. M. Carreira, Org. Lett. 2000,
2, 4233 – 4236; d) D. E. Frantz, R. Fꢁssler, C. S. Tomooka, E. M.
Carreira, Acc. Chem. Res. 2000, 33, 373 – 381; e) E. El-Sayed,
N. K. Anand, E. M. Carreira, Org. Lett. 2001, 3, 3017 – 3020;
f) N. K. Anand, E. M. Carreira, J. Am. Chem. Soc. 2001, 123,
9687 – 9688; g) R. Fꢁssler, C. S. Tomooka, D. E. Frantz, E. M.
Carreira, Proc. Natl. Acad. Sci. USA 2004, 101, 5843 – 5845.
[17] W. C. Still, C. Gennari, Tetrahedron Lett. 1983, 24, 4405 – 4408.
[18] a) T. Katsuki, K. B. Sharpless, J. Am. Chem. Soc. 1980, 102,
5974 – 5976; b) R. M. Hanson, K. B. Sharpless, J. Org. Chem.
1986, 51, 1922 – 1925; c) Y. Gao, R. M. Hanson, J. M. Klunder,
S. Y. Ko, H. Masamune, K. B. Sharpless, J. Am. Chem. Soc. 1987,
109, 5765 – 5780.
[19] a) Y. E. Raifeld, A. Nikitenko, B. M. Arshava, Tetrahedron:
Asymmetry 1991, 2, 1083 – 1084; see also: b) M. Caron, K. B.
Sharpless, J. Org. Chem. 1985, 50, 1557 – 1560.
[20] a) P. R. Blakemore, W. J. Cole, P. J. Kocienski, A. Morley, Synlett
1998, 26 – 28. For a recent review of the Julia – Kocienski
coupling, see: b) C. Aꢂssa, Eur. J. Org. Chem. 2009, 1831 – 1844.
[21] D. Dꢃaz, T. Martꢃn, V. S. Martꢃn, J. Org. Chem. 2001, 66, 7231 –
7233.
[22] The long reaction times required for this transformation (up to
17.5 h) are most likely to be a result of the steric and electronic
deactivation of the C10–C11 double bond. This finding retro-
spectively justified our assumption that a selective dichlorination
of the C10–C11 double bond would be troublesome or even
impossible when the C22–C21 double bond was already present,
since the C22–C21 double bond can be anticipated to be more
electron-rich and less sterically hindered.
[23] a) J. C. Martin, R. J. Arhart, J. Am. Chem. Soc. 1971, 93, 2339 –
2341; b) J. C. Martin, R. J. Arhart, J. Am. Chem. Soc. 1971, 93,
2341 – 2342; c) J. C. Martin, R. J. Arhart, J. Am. Chem. Soc. 1971,
93, 4327 – 4329; d) R. J. Arhart, J. C. Martin, J. Am. Chem. Soc.
1972, 94, 5003 – 5010.
Received: April 12, 2011
Published online: July 8, 2011
Keywords: asymmetric synthesis · chlorination ·
.
chlorosulfolipids · natural products · total synthesis
[1] J. Elovson, P. R. Vagelos, Proc. Natl. Acad. Sci. USA 1969, 62,
957 – 963.
[2] C. Nilewski, R. W. Geisser, E. M. Carreira, Nature 2009, 457,
573 – 576.
[3] C. Nilewski, R. W. Geisser, M.-O. Ebert, E. M. Carreira, J. Am.
Chem. Soc. 2009, 131, 15866 – 15876.
[4] a) G. M. Shibuya, J. S. Kanady, C. D. Vanderwal, J. Am. Chem.
Soc. 2008, 130, 12514 – 12518; b) J. S. Kanady, J. D. Nguyen, J. W.
Ziller, C. D. Vanderwal, J. Org. Chem. 2009, 74, 2175 – 2178;
c) D. K. Bedke, C. Vanderwal, Nat. Prod. Rep. 2011, 28, 15 – 25,
and references therein.
[5] The following total syntheses of chlorosulfolipids were achieved
after the disclosure of the total synthesis of hexachlorosulfolipid
(see Ref. [2]): a) D. K. Bedke, G. M. Shibuya, A. Pereira, W. H.
Gerwick, T. H. Haines, C. D. Vanderwal, J. Am. Chem. Soc. 2009,
131, 7570 – 7572; b) D. K. Bedke, G. M. Shibuya, A. R. Pereira,
W. H. Gerwick, C. D. Vanderwal, J. Am. Chem. Soc. 2010, 132,
2542 – 2543; c) T. Yoshimitsu, N. Fukumoto, R. Nakatani, N.
Kojima, T. Tanaka, J. Org. Chem. 2010, 75, 5425 – 5437; d) T.
Yoshimitsu, R. Nakatani, A. Kobayashi, T. Tanaka, Org. Lett.
2011, 13, 908 – 911; e) T. Umezawa, M. Shibata, K. Kaneko, T.
Okino, F. Matsuda, Org. Lett. 2011, 13, 904 – 907.
[6] P. Ciminiello, C. DellꢀAversano, E. Fattorusso, M. Forino, S.
Magno, M. Di Rosa, A. Ianaro, R. Poletti, J. Am. Chem. Soc.
2002, 124, 13114 – 13120.
[7] P. Ciminiello, E. Fattorusso, M. Forino, M. Di Rosa, A. J. Ianaro,
J. Org. Chem. 2001, 66, 578 – 582.
[24] S. K. Latypov, J. M. Seco, E. Quinoa, R. Riguera, J. Am. Chem.
Soc. 1998, 120, 877 – 882.
[8] See the Supporting Information for details.
[9] T. Schlama, K. Gabriel, V. Gouverneur, C. Mioskowski, Angew.
Chem. 1997, 109, 2440 – 2442; Angew. Chem. Int. Ed. Engl. 1997,
36, 2342 – 2344.
[10] N. Matsumori, D. Kaneno, M. Murata, H. Nakamura, K.
Tachibana, J. Org. Chem. 1999, 64, 866 – 876.
[11] a) S. G. Hentges, K. B. Sharpless, J. Am. Chem. Soc. 1980, 102,
4263 – 4265; b) E. N. Jacobsen, I. Marko, W. S. Mungall, G.
Schroder, K. B. Sharpless, J. Am. Chem. Soc. 1988, 110, 1968 –
1970.
[12] For the configurational assignment after bishydroxylation, see
the Supporting Information.
[25] a) J. A. Dale, H. S. Mosher, J. Am. Chem. Soc. 1973, 95, 512 –
519; b) S. K. Latypov, J. M. Seco, E. QuiÇoꢄ, R. Riguera, J. Org.
Chem. 1996, 61, 8569 – 8577; c) J. M. Seco, E. QuiÇoꢄ, R.
Riguera, Chem. Rev. 2004, 104, 17 – 118, and references therein.
[26] For the selective acetylation of primary alcohols, see: a) K.
Ishihara, H. Kurihara, H. Yamamoto, J. Org. Chem. 1993, 58,
3791 – 3793. For other examples of acylations of the primary
alcohol moiety in 1,2-diols, see: b) K. C. Nicolaou, C. A. Veale,
S. E. Webber, H. Katerinopoulos, J. Am. Chem. Soc. 1985, 107,
7515 – 7518; c) S. Hanessian, Y. Yang, S. Giroux, V. Mascitti,
Jianguo Ma, F. Raeppel, J. Am. Chem. Soc. 2003, 125, 13784 –
13792.
[13] a) J. J. Riehl, J. M. Lehn, F. Hemmert, Bull. Soc. Chim. Fr. 1963,
224; b) J. M. Lehn, J. J. Riehl, Mol. Phys. 1964, 8, 33 – 38; c) T. J.
Batterham, NMR Spectra of Simple Heterocycles, Wiley-Inter-
science, New York, USA, 1973, pp. 365 – 369.
[27] For a review on structurally misassigned natural products, see:
K. C. Nicolaou, S. A. Snyder, Angew. Chem. 2005, 117, 1036 –
1069; Angew. Chem. Int. Ed. 2005, 44, 1012 – 1044.
Angew. Chem. Int. Ed. 2011, 50, 7940 –7943
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7943